Handbook of Cell Biosensors 3030232166, 9783030232160

Citation preview

Gérald Thouand Editor-in-Chief Shimshon Belkin Sylvia Daunert · Paul Freemont Julie Hermans · Isao Karube Sylvain Martel · Elisa Michelini Aldo Roda  Section Editors

Handbook of Cell Biosensors

Handbook of Cell Biosensors

Ge´rald Thouand Editor-in-Chief

Handbook of Cell Biosensors With 230 Figures and 41 Tables

Editor-in-Chief Gérald Thouand CNRS, GEPEA, UMR 6144 Université de Nantes La Roche-sur-Yon, France

ISBN 978-3-030-23216-0 ISBN 978-3-030-23217-7 (eBook) ISBN 978-3-030-23218-4 (print and electronic bundle) https://doi.org/10.1007/978-3-030-23217-7 © Springer Nature Switzerland AG 2022 All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG. The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Preface

A cell biosensor is, above all, a hybrid system combining biology, electronics, materials, and digital science. It is a measuring instrument that marries a biological recognition element/bioelement (here a cell) with an electronic component and has useful applications in many economic sectors. When I started working on the subject of microbial biosensors in my laboratory, more than 20 years ago, I was faced with an exciting subject because it was highly interdisciplinary, combining biology, physics, engineering, and electronics, but also economics. All of these facets were new to me and it took a long time to gain a broader view of the subject. Of course, there were books on some specific topics or examples, and there were also excellent reviews, but there were no documents linking the major scientific aspects of biosensors to the economics, business, or pedagogy specific to this discipline. The aim of this handbook on cell biosensors is to provide the best current knowledge for researchers in the process of changing their field of work, for those working in industry and for students who want to learn more about the subject. This handbook is intended to be easy to use. It contains three parts coordinated by some of the best researchers in their disciplines, covering all the relevant topics, including those related to economics, standardization, and the relationship between university and industry, all of which must be taken into account when dealing with this scientific field. Another advantage of this handbook is its “living” character, as the articles will be updated as regularly as possible. – Part 1 – Fundamentals and Genetics for Cell Biosensors Applications, coordinated by Shimshon Belkin, and Paul Freemont – Part 2 – Transducers, Materials, and Systems, coordinated by Sylvia Daunert, Sylvain Martel, Elisa Michelini, and Aldo Roda – Part 3 – Cell Biosensors Applications, coordinated by Sylvia Daunert, Julie Hermans, Isao Karube, and Gérald Thouand Each part contains summaries of different topics written by leading specialists, which should stimulate interest for more in-depth reading. The introduction provides a review of the 3126 articles published since 1965 and offers an “ideal bibliography” of 506 references (10% most-cited articles). v

vi

Preface

I would like to thank all the authors for their excellent contributions, the Springer Nature editing team for their valuable support and patience throughout the preparation of this handbook, and, finally, the section editors for their efforts and help in bringing this book to fruition through the years over which it was conceived. I wish you, dear reader, as much pleasure in reading this book as I had in its conception. Roche-sur-Yon, France October 2021

Gérald Thouand Editor-in-Chief

Contents

Part I 1

Introduction

.......................................

1

Introduction to Cell Biosensors Through 55 Years of Scientific Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Gérald Thouand

3

Part II Fundamentals and Genetics for Cell Biosensors Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

3

Stress Response-Based Whole-Cell Biosensor Development: Sentinels, Serendipity, and Circuitry . . . . . . . . . . . . . . . . . . . . . . . Robert A. LaRossa Engineering Autobioluminescent Eukaryotic Cells as Tools for Environmental and Biomedical Surveillance . . . . . . . . . . . . . . . . . Tingting Xu, Dan Close, Ghufran Ud Din, Gary Sayler, and Steven Ripp

4

Biosensors of the Well-being of Cell Cultures . . . . . . . . . . . . . . . . . Karen Marie Polizzi

5

Systematic Design of a Quorum Sensing-Based Biosensor for the Detection of Metal Ions in Escherichia coli . . . . . . . . . . . . . . . . Bor-Sen Chen

41

43

57

71

89

6

Riboswitches as Sensor Entities . . . . . . . . . . . . . . . . . . . . . . . . . . . Svetlana Harbaugh, Michael Goodson, Yaroslav Chushak, Jorge Chávez, and Nancy Kelley-Loughnane

111

7

Integration of Sensor Cells into Hardware Platforms Rajendra P. Shukla, Avia Lavon, and Hadar Ben-Yoav

..........

141

8

Whole-Cell-Based Fiber-Optic Biosensors . . . . . . . . . . . . . . . . . . . Boris Veltman and Evgeni Eltzov

163

9

Optical Approaches to Visualization of Cellular Activity . . . . . . . . Mei-Yi Lu and Ji-Yen Cheng

189 vii

viii

10

11

Contents

Digital and Analogue Approaches to Whole-Cell Sensor Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Luna Rizik, Litovco Phyana, and Daniel Ramez

205

Engineering of Sensory Proteins with New Ligand-Binding Capacities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Diogo Tavares, Vitali Maffenbeier, and Jan Roelof van der Meer

223

12

Cell-Free Biosensors: Synthetic Biology Without Borders . . . . . . . Aidan Tinafar, Yu Zhou, Fan Hong, Kirstie L. Swingle, Anli A. Tang, Alexander A. Green, and Keith Pardee

243

13

Engineering Prokaryote Synthetic Biology Biosensors . . . . . . . . . . Xinyi Wan, Trevor Y. H. Ho, and Baojun Wang

283

14

Cell-Free Synthetic Biology Biosensors . . . . . . . . . . . . . . . . . . . . . . David K. Karig, Allison Reno, Lauren Elizabeth Franklin, and Andrea C. Timm

319

15

Genetic Circuit Design Principles . . . . . . . . . . . . . . . . . . . . . . . . . . Jing Wui Yeoh, Salvador Gomez-Carretero, Wai Kit David Chee, Ai Ying Teh, and Chueh Loo Poh

339

16

Fundamental Building Blocks of Whole-Cell Biosensor Design . . . Ke Yan Wen, Jack W. Rutter, Chris P. Barnes, and Linda Dekker

383

17

Mammalian Cell-Based Biosensors . . . . . . . . . . . . . . . . . . . . . . . . . Karl-Heinz Feller

407

18

Mammalian Synbio Sensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Fabiana Tedeschi and Velia Siciliano

435

19

Environmental Biosensors: A Microbiological View . . . . . . . . . . . Juan-Carlos Gutiérrez, Francisco Amaro, Silvia Díaz, and Ana Martín-González

455

Part III

Transducers, Materials, and Systems

.................

477

20

Live Cell Immobilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Antonia Lopreside, Maria Maddalena Calabretta, Laura Montali, Aldo Roda, and Elisa Michelini

479

21

Sol-Gel Process, Structure, and Properties . . . . . . . . . . . . . . . . . . . Thibaud Coradin

497

22

Acoustic Transducer and Its Applications in Biosensors . . . . . . . . Junyu Zhang, Qian Wu, Xi Zhang, Hao Wan, and Ping Wang

517

23

Acoustic Biosensors for Cell Research . . . . . . . . . . . . . . . . . . . . . . Samar Damiati

537

Contents

ix

24

Electrodes for Cell Sensors Interfacing . . . . . . . . . . . . . . . . . . . . . Hadar Ben-Yoav, Heftsi Ragones, Richa Pandey, Giorgia Fiaschi, and Yosi Shacham-Diamand

25

New Materials for the Construction of Electrochemical Cell-Based Biosensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Andreea Cernat, Bianca Ciui, Luminita Fritea, Mihaela Tertis, and Cecilia Cristea

26

Biosensors and Bioelectronics on Smartphone . . . . . . . . . . . . . . . . Shuang Li, Daizong Ji, Gang Xu, Jinglong Liu, Yanli Lu, Sze Shin Low, and Qingjun Liu

Part IV

Biosensor, Market, and Innovation . . . . . . . . . . . . . . . . . . .

569

601

627

657

27

Business Models for Biosensors in the Food Industry . . . . . . . . . . Rosa Caiazza and Barbara Bigliardi

659

28

Biosensors: Ethical, Regulatory, and Legal Issues . . . . . . . . . . . . . Mohammad Ershadul Karim

679

29

University-Industry Relationships for the Development and Commercialization of Biosensors . . . . . . . . . . . . . . . . . . . . . . . . . . Christina G. Siontorou

Part V 30

Applications of Cells Biosensors . . . . . . . . . . . . . . . . . . . . . .

International Organization of Standards for Measurement Validation: Food Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Michèle Lees

31

Nutrient Detection with Whole-Cell Biosensors . . . . . . . . . . . . . . . Yan-Zhai Wang, Joseph Kirubaharan Christopher, Yang-Chun Yong, and Dan-Dan Zhai

32

Luminescent Microbial Bioassays and Microalgal Biosensors as Tools for Environmental Toxicity Evaluation . . . . . . . . . . . . . . Jara Hurtado-Gallego, Gerardo Pulido-Reyes, Miguel González-Pleiter, and Francisca Fernández-Piñas

33

34

Detection and Effects of Metal and Organometallic Compounds with Microbial Bioluminescence and Raman Spectroscopy . . . . . . Sulivan Jouanneau, Ali Assaf, Marie-José Durand, and Gérald Thouand Microbial Biosensors for the Detection of Organic Pollutants . . . . Benjamin Shemer and Shimshon Belkin

707

723

725 747

767

825

851

x

Contents

35

Microbial Fuel Cells, Concept, and Applications . . . . . . . . . . . . . . Carlo Santoro, Mike Brown, Iwona Gajda, John Greenman, Oluwatosin Obata, Maria José Salar García, Pavlina Theodosiou, Alexis Walter, Jonathan Winfield, Jiseon You, and Ioannis Ieropoulos

875

36

Organic Matter BOD Biosensor Monitoring . . . . . . . . . . . . . . . . . Akihito Nakanishi, Wataru Yoshida, and Isao Karube

911

37

Cell-Based Biosensor for Rapid Screening of Pathogens and Toxins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Celina To, Pratik Banerjee, and Arun K. Bhunia

929

38

Application of Bacterial Whole-Cell Biosensors in Health . . . . . . . Yizhi Song, Cordelia P. N. Rampley, Xiaoyu Chen, Fawen Du, Ian P. Thompson, and Wei E. Huang

945

39

Smartphone-Based Cell Detection . . . . . . . . . . . . . . . . . . . . . . . . . Maria Maddalena Calabretta, Laura Montali, Antonia Lopreside, Aldo Roda, and Elisa Michelini

963

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

979

About the Editor-in-Chief

Gérald Thouand received his PhD in Microbiology from the University of Nancy, France, in 1993. He is tenured Professor in microbiology at the University of Nantes and Deputy for research innovation at the French Ministry of Research (DRRT Pays de La Loire) from 2008 to 2018 and now Expert for the County Pays de la Loire for innovation. In 2008 he was Auditor at the Institute of Higher Studies for Science and Technology (IHEST, Paris). His research interest includes environmental monitoring of biodegradation and biotechnology using microbial biosensors. He is mainly involved in the development of biosensors for chemical pollutants detection and pathogenic bacteria. He is President of the International Society for Bioluminescence and Chemiluminescence, editor for Environmental Science and Pollution Research (ESPR, Springer), and editor for Sensors and was associate editor for Frontiers in Microbiology.

xi

Section Editors

Shimshon Belkin Institute of Life Sciences The Hebrew University of Jerusalem Jerusalem, Israel

Sylvia Daunert University of Miami Miami, FL, USA

xiii

xiv

Section Editors

Paul Freemont Imperial College London London, UK

Julie Hermans Université catholique de Louvain Mons, Belgium

Isao Karube School of Bioscience and Biotechnology Tokyo University of Technology Tokyo, Japan

Section Editors

xv

Sylvain Martel Ecole Polytechnique de Montreal Montréal, QC, Canada

Elisa Michelini Department of Chemistry “G. Ciamician” University of Bologna Bologna, Italy INBB, Istituto Nazionale di Biostrutture e Biosistemi Rome, Italy Health Sciences and Technologies-Interdepartmental Center for Industrial Research (HST-ICIR) University of Bologna Bologna, Italy

Aldo Roda Department of Chemistry “G. Ciamician” University of Bologna Bologna, Italy INBB, Istituto Nazionale di Biostrutture e Biosistemi Rome, Italy

Contributors

Francisco Amaro Dpto. Genética, Fisiología y Microbiología. Facultad de Biología, Universidad Complutense (UCM), Madrid, Spain Ali Assaf CNRS, GEPEA, UMR 6144, Université de Nantes, La Roche-sur-Yon, France Pratik Banerjee Division of Epidemiology, Biostatistics, and Environmental Health, The University of Memphis, Memphis, TN, USA Chris P. Barnes Division of Biosciences, University College London, London, UK Shimshon Belkin Institute of Life Sciences, The Hebrew University of Jerusalem, Jerusalem, Israel Hadar Ben-Yoav Nanobioelectronics Laboratory (NBEL), Department of Biomedical Engineering, Ben-Gurion University of the Negev, Beer-Sheva, Israel Arun K. Bhunia Molecular Food Microbiology Laboratory, Department of Food Science, Purdue University, West Lafayette, IN, USA Department of Comparative Pathobiology, Purdue University, West Lafayette, IN, USA Purdue Institute of Inflammation, Immunology and Infectious Disease, Purdue University, West Lafayette, IN, USA Barbara Bigliardi Department of Engineering and Architecture, University of Parma, Parma, Italy Mike Brown Bristol BioEnergy Centre, Bristol Robotics Laboratory, University of the West of England, Bristol, UK Rosa Caiazza Parthenope University, Naples, Italy Maria Maddalena Calabretta Department of Chemistry “G. Ciamician”, University of Bologna, Bologna, Italy Andreea Cernat Department of Analytical Chemistry, Faculty of Pharmacy, University of Medicine and Pharmacy Iuliu Hatieganu Cluj-Napoca, Cluj-Napoca, Romania xvii

xviii

Contributors

Jorge Chávez 711th Human Performance Wing, Air Force Research Laboratory, Dayton, OH, USA Wai Kit David Chee Department of Biomedical Engineering, Faculty of Engineering, National University of Singapore, Singapore, Singapore NUS Synthetic Biology for Clinical and Technological Innovation (SynCTI), Centre for Life Sciences, National University of Singapore, Singapore, Singapore Bor-Sen Chen Department of Electric Engineering, National Tsing Hua University, Hsinchu, Taiwan Xiaoyu Chen Department of Engineering Science, University of Oxford, Oxford, UK Ji-Yen Cheng Research Center for Applied Sciences, Academia Sinica, Taipei, Taiwan Department of Mechanical and Mechatronic Engineering, National Taiwan Ocean University, Keelung, Taiwan Institute of Biophotonics, National Yang-Ming University, Taipei, Taiwan College of Engineering, Chang Gung University, Taoyuan, Taiwan Joseph Kirubaharan Christopher Biofuels Institute, School of the Environment, Jiangsu University, Jiangsu Province, China Yaroslav Chushak Henry M. Jackson Foundation for the Advancement of Military Medicine, Dayton, OH, USA Bianca Ciui Department of Analytical Chemistry, Faculty of Pharmacy, University of Medicine and Pharmacy Iuliu Hatieganu Cluj-Napoca, Cluj-Napoca, Romania Dan Close 490 BioTech Inc., Knoxville, TN, USA Thibaud Coradin Chimie de la Matière Condensée de Paris, Sorbonne Université, CNRS, Collège de France, Paris, France Cecilia Cristea Department of Analytical Chemistry, Faculty of Pharmacy, University of Medicine and Pharmacy Iuliu Hatieganu Cluj-Napoca, Cluj-Napoca, Romania Samar Damiati Department of Biochemistry, Faculty of Science, King Abdulaziz University (KAU), Jeddah, Saudi Arabia Institute for Synthetic Bioarchitectures, Department of Nanobiotechnology, University of Natural Resources and Life Sciences (BOKU), Vienna, Austria Division of Nanobiotechnology, Department of Protein Science, Science for Life Laboratory, School of Engineering Sciences in Chemistry, Biotechnology and Health, KTH Royal Institute of Technology, Stockholm, Sweden Linda Dekker Division of Biosciences, University College London, London, UK

Contributors

xix

Silvia Díaz Dpto. Genética, Fisiología y Microbiología. Facultad de Biología, Universidad Complutense (UCM), Madrid, Spain Fawen Du Hesen Biotech Co., Ltd., Shanghai, China Marie-José Durand CNRS, GEPEA, UMR 6144, Université de Nantes, La Rochesur-Yon, France Evgeni Eltzov Department of Postharvest Science, The Volcani Center, Agricultural Research Organization, Bet Dagan, Israel Karl-Heinz Feller Ernst-Abbe-University Jena, Jena, Germany Francisca Fernández-Piñas Department of Biology, Faculty of Science, Universidad Autónoma de Madrid, Madrid, Spain Giorgia Fiaschi Department of Physical Electronics/Department of Materials Science and Engineering, School of EE/Tel Aviv University, Tel Aviv, Israel Lauren Elizabeth Franklin Department of Bioengineering, Clemson University, Clemson, SC, USA Luminita Fritea Department of Preclinical Disciplines, Faculty of Medicine and Pharmacy, University of Oradea, Oradea, Romania Iwona Gajda Bristol BioEnergy Centre, Bristol Robotics Laboratory, University of the West of England, Bristol, UK Maria José Salar García Bristol BioEnergy Centre, Bristol Robotics Laboratory, University of the West of England, Bristol, UK Salvador Gomez-Carretero Department of Biochemistry, Yong Loo Lin School of Medicine, National University of Singapore, Singapore, Singapore NUS Synthetic Biology for Clinical and Technological Innovation (SynCTI), Centre for Life Sciences, National University of Singapore, Singapore, Singapore Miguel González-Pleiter Department of Biology, Faculty of Science, Universidad Autónoma de Madrid, Madrid, Spain Michael Goodson 711th Human Performance Wing, Air Force Research Laboratory, Dayton, OH, USA Alexander A. Green Biodesign Center for Molecular Design and Biomimetics, The Biodesign Institute and the School of Molecular Sciences, Arizona State University, Tempe, AZ, USA John Greenman Bristol BioEnergy Centre, Bristol Robotics Laboratory, University of the West of England, Bristol, UK Juan-Carlos Gutiérrez Dpto. Genética, Fisiología y Microbiología. Facultad de Biología, Universidad Complutense (UCM), Madrid, Spain

xx

Contributors

Svetlana Harbaugh 711th Human Performance Wing, Air Force Research Laboratory, Dayton, OH, USA Trevor Y. H. Ho School of Biological Sciences, University of Edinburgh, Edinburgh, UK Centre for Synthetic and Systems Biology, University of Edinburgh, Edinburgh, UK Fan Hong Biodesign Center for Molecular Design and Biomimetics, The Biodesign Institute and the School of Molecular Sciences, Arizona State University, Tempe, AZ, USA Wei E. Huang Department of Engineering Science, University of Oxford, Oxford, UK Jara Hurtado-Gallego Department of Biology, Faculty of Science, Universidad Autónoma de Madrid, Madrid, Spain Ioannis Ieropoulos Bristol BioEnergy Centre, Bristol Robotics Laboratory, University of the West of England, Bristol, UK Daizong Ji Biosensor National Special Laboratory, Key Laboratory for Biomedical Engineering of Education Ministry, Department of Biomedical Engineering, Zhejiang University, Hangzhou, P. R. China Sulivan Jouanneau CNRS, GEPEA, UMR 6144, Université de Nantes, La Rochesur-Yon, France David K. Karig Department of Bioengineering, Clemson University, Clemson, SC, USA Research and Exploratory Development Department, Johns Hopkins University Applied Physics Laboratory, Laurel, MD, USA Mohammad Ershadul Karim Faculty of Law, University of Malaya, Kuala Lumpur, Malaysia Bangladesh Supreme Court, Dhaka, Bangladesh Isao Karube School of Bioscience and Biotechnology, Tokyo University of Technology, Tokyo, Japan Nancy Kelley-Loughnane 711th Human Performance Wing, Air Force Research Laboratory, Dayton, OH, USA Robert A. LaRossa Red Jay Consulting, Chadds Ford, PA, USA Avia Lavon Department of Biomedical Engineering, Ben-Gurion University of the Negev, Beer-Sheva, Israel Michèle Lees Food Integrity Consultancy, Sucé-sur-Erdre, France

Contributors

xxi

Shuang Li Biosensor National Special Laboratory, Key Laboratory for Biomedical Engineering of Education Ministry, Department of Biomedical Engineering, Zhejiang University, Hangzhou, P. R. China Jinglong Liu Biosensor National Special Laboratory, Key Laboratory for Biomedical Engineering of Education Ministry, Department of Biomedical Engineering, Zhejiang University, Hangzhou, P. R. China Qingjun Liu Biosensor National Special Laboratory, Key Laboratory for Biomedical Engineering of Education Ministry, Department of Biomedical Engineering, Zhejiang University, Hangzhou, P. R. China Antonia Lopreside Department of Chemistry “G. Ciamician”, University of Bologna, Bologna, Italy Sze Shin Low Biosensor National Special Laboratory, Key Laboratory for Biomedical Engineering of Education Ministry, Department of Biomedical Engineering, Zhejiang University, Hangzhou, P. R. China Mei-Yi Lu Research Center for Applied Sciences, Academia Sinica, Taipei, Taiwan Yanli Lu Biosensor National Special Laboratory, Key Laboratory for Biomedical Engineering of Education Ministry, Department of Biomedical Engineering, Zhejiang University, Hangzhou, P. R. China Vitali Maffenbeier Department of Fundamental Microbiology, Faculty of Biology and Medicine, University of Lausanne, Lausanne, Switzerland Ana Martín-González Dpto. Genética, Fisiología y Microbiología. Facultad de Biología, Universidad Complutense (UCM), Madrid, Spain Elisa Michelini Department of Chemistry “G. Ciamician”, University of Bologna, Bologna, Italy INBB, Istituto Nazionale di Biostrutture e Biosistemi, Rome, Italy Health Sciences and Technologies-Interdepartmental Center for Industrial Research (HST-ICIR), University of Bologna, Bologna, Italy Laura Montali Department of Chemistry “G. Ciamician”, University of Bologna, Bologna, Italy Akihito Nakanishi School of Bioscience and Biotechnology, Tokyo University of Technology, Tokyo, Japan Oluwatosin Obata Bristol BioEnergy Centre, Bristol Robotics Laboratory, University of the West of England, Bristol, UK Richa Pandey Department of Physical Electronics/Department of Materials Science and Engineering, School of EE/Tel Aviv University, Tel Aviv, Israel Keith Pardee Leslie Dan Faculty of Pharmacy, University of Toronto, Toronto, ON, Canada

xxii

Contributors

Litovco Phyana Faculty of Biomedical Engineering Technion, Israel Institute of Technology, Haifa, Israel Chueh Loo Poh Department of Biomedical Engineering, Faculty of Engineering, National University of Singapore, Singapore, Singapore NUS Synthetic Biology for Clinical and Technological Innovation (SynCTI), Centre for Life Sciences, National University of Singapore, Singapore, Singapore Karen Marie Polizzi Department of Chemical Engineering, Imperial College Centre for Synthetic Biology, London, UK Gerardo Pulido-Reyes Department of Biology, Faculty of Science, Universidad Autónoma de Madrid, Madrid, Spain Heftsi Ragones School of Chemistry, Faculty of Exact Sciences, Tel Aviv University, Tel Aviv, Israel Daniel Ramez Faculty of Biomedical Engineering Technion, Israel Institute of Technology, Haifa, Israel Cordelia P. N. Rampley Department of Engineering Science, University of Oxford, Oxford, UK Allison Reno Department of Bioengineering, Clemson University, Clemson, SC, USA Steven Ripp The Center for Environmental Biotechnology and Department of Microbiology, The University of Tennessee, Knoxville, TN, USA 490 BioTech Inc., Knoxville, TN, USA Luna Rizik Faculty of Biomedical Engineering Technion, Israel Institute of Technology, Haifa, Israel Aldo Roda Department of Chemistry “G. Ciamician”, University of Bologna, Bologna, Italy INBB, Istituto Nazionale di Biostrutture e Biosistemi, Rome, Italy Jack W. Rutter Division of Biosciences, University College London, London, UK Carlo Santoro Bristol BioEnergy Centre, Bristol Robotics Laboratory, University of the West of England, Bristol, UK Gary Sayler 490 BioTech Inc., Knoxville, TN, USA Yosi Shacham-Diamand Department of Physical Electronics/Department of Materials Science and Engineering, School of EE/Tel Aviv University, Tel Aviv, Israel Benjamin Shemer Institute of Life Sciences, The Hebrew University of Jerusalem, Jerusalem, Israel Rajendra P. Shukla Nanobioelectronics Laboratory (NBEL), Department of Biomedical Engineering, Ben-Gurion University of the Negev, Beer-Sheva, Israel

Contributors

xxiii

Velia Siciliano Synthetic and Systems Biology Lab for Biomedicine, Istituto Italiano di Tecnologia, Naples, Italy Christina G. Siontorou Laboratory of Simulation of Industrial Processes, Department of Industrial Management and Technology, School of Maritime and Industry, University of Piraeus, Piraeus, Greece Yizhi Song Department of Engineering Science, University of Oxford, Oxford, UK Kirstie L. Swingle Biodesign Center for Molecular Design and Biomimetics, The Biodesign Institute and the School of Molecular Sciences, Arizona State University, Tempe, AZ, USA Anli A. Tang Biodesign Center for Molecular Design and Biomimetics, The Biodesign Institute and the School of Molecular Sciences, Arizona State University, Tempe, AZ, USA Diogo Tavares Department of Fundamental Microbiology, Faculty of Biology and Medicine, University of Lausanne, Lausanne, Switzerland Fabiana Tedeschi Synthetic and Systems Biology Lab for Biomedicine, Istituto Italiano di Tecnologia, Naples, Italy Ai Ying Teh Department of Biomedical Engineering, Faculty of Engineering, National University of Singapore, Singapore, Singapore NUS Synthetic Biology for Clinical and Technological Innovation (SynCTI), Centre for Life Sciences, National University of Singapore, Singapore, Singapore Mihaela Tertis Department of Analytical Chemistry, Faculty of Pharmacy, University of Medicine and Pharmacy Iuliu Hatieganu Cluj-Napoca, Cluj-Napoca, Romania Pavlina Theodosiou Bristol BioEnergy Centre, Bristol Robotics Laboratory, University of the West of England, Bristol, UK Ian P. Thompson Department of Engineering Science, University of Oxford, Oxford, UK Gérald Thouand CNRS, GEPEA, UMR 6144, Université de Nantes, La Rochesur-Yon, France Andrea C. Timm Research and Exploratory Development Department, Johns Hopkins University Applied Physics Laboratory, Laurel, MD, USA Aidan Tinafar Leslie Dan Faculty of Pharmacy, University of Toronto, Toronto, ON, Canada Celina To Molecular Food Microbiology Laboratory, Department of Food Science, Purdue University, West Lafayette, IN, USA Hygiena, Camarillo, CA, USA

xxiv

Contributors

Ghufran Ud Din The Department of Microbiology, Quaid-i-Azam University, Islamabad, Pakistan Jan Roelof van der Meer Department of Fundamental Microbiology, Faculty of Biology and Medicine, University of Lausanne, Lausanne, Switzerland Boris Veltman Institute of Biochemistry, Food Science and Nutrition, Robert H. Smith Faculty of Agriculture, Food and Environment, Hebrew University of Jerusalem, Rehovot, Israel Department of Postharvest Science, The Volcani Center, Agricultural Research Organization, Bet Dagan, Israel Alexis Walter Bristol BioEnergy Centre, Bristol Robotics Laboratory, University of the West of England, Bristol, UK Hao Wan Biosensor National Special Laboratory, Key Lab for Biomedical Engineering of Ministry of Education, Department of Biomedical Engineering, Zhejiang University, Hangzhou, China Xinyi Wan School of Biological Sciences, University of Edinburgh, Edinburgh, UK Centre for Synthetic and Systems Biology, University of Edinburgh, Edinburgh, UK Baojun Wang School of Biological Sciences, University of Edinburgh, Edinburgh, UK Centre for Synthetic and Systems Biology, University of Edinburgh, Edinburgh, UK Ping Wang Biosensor National Special Laboratory, Key Lab for Biomedical Engineering of Ministry of Education, Department of Biomedical Engineering, Zhejiang University, Hangzhou, China Yan-Zhai Wang Biofuels Institute, School of the Environment, Jiangsu University, Jiangsu Province, China Ke Yan Wen Division of Biosciences, University College London, London, UK Jonathan Winfield Bristol BioEnergy Centre, Bristol Robotics Laboratory, University of the West of England, Bristol, UK Qian Wu Biosensor National Special Laboratory, Key Lab for Biomedical Engineering of Ministry of Education, Department of Biomedical Engineering, Zhejiang University, Hangzhou, China Gang Xu Biosensor National Special Laboratory, Key Laboratory for Biomedical Engineering of Education Ministry, Department of Biomedical Engineering, Zhejiang University, Hangzhou, P. R. China Tingting Xu The Center for Environmental Biotechnology and Department of Microbiology, The University of Tennessee, Knoxville, TN, USA Jing Wui Yeoh Department of Biochemistry, Yong Loo Lin School of Medicine, National University of Singapore, Singapore, Singapore

Contributors

xxv

NUS Synthetic Biology for Clinical and Technological Innovation (SynCTI), Centre for Life Sciences, National University of Singapore, Singapore, Singapore Yang-Chun Yong Biofuels Institute, School of the Environment, Jiangsu University, Jiangsu Province, China Wataru Yoshida School of Bioscience and Biotechnology, Tokyo University of Technology, Tokyo, Japan Jiseon You Bristol BioEnergy Centre, Bristol Robotics Laboratory, University of the West of England, Bristol, UK Dan-Dan Zhai Biofuels Institute, School of the Environment, Jiangsu University, Jiangsu Province, China College of Biological Engineering, Henan University of Technology, Zhengzhou, China Junyu Zhang Biosensor National Special Laboratory, Key Lab for Biomedical Engineering of Ministry of Education, Department of Biomedical Engineering, Zhejiang University, Hangzhou, China Xi Zhang Biosensor National Special Laboratory, Key Lab for Biomedical Engineering of Ministry of Education, Department of Biomedical Engineering, Zhejiang University, Hangzhou, China Yu Zhou Biodesign Center for Molecular Design and Biomimetics, The Biodesign Institute and the School of Molecular Sciences, Arizona State University, Tempe, AZ, USA

Part I Introduction

1

Introduction to Cell Biosensors Through 55 Years of Scientific Production Ge´rald Thouand

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 What Is a Cell Biosensor? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Evolution of the Research Topics from 1965 to 2020 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 The “Ideal” Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Conclusions and Future Directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

Abstract

Cell biosensors, whose concept was born 55 years ago, have never ceased to be enriched by the scientific ambitions and ingenuity of researchers, and societal needs in terms of applications leading to a production of 3126 articles. But, if you want to discover a new research theme or if you are a student studying for a Master or PhD degree, where should you start reading? This introduction summarizes the literature in a straightforward manner in just a few pages, with the main objective of allowing nonexperts (students, researchers changing their field, and industry professionals) to quickly get an overview of the field. First, it covers the description of a cell biosensor, then a study of the evolution of the research themes over the last 55 years, and finally an “ideal bibliography” of 506 articles from among the most cited of the last 55 years. Keywords

Biosensor · Bacteria · Bacteriophage · Microalgae · Fungi · Animal cell · Bioreporter · Transducer · Immobilization · Detection · Food · Environment · Health

G. Thouand (*) CNRS, GEPEA, UMR 6144, Université de Nantes, La Roche-sur-Yon, France e-mail: [email protected] © Springer Nature Switzerland AG 2022 G. Thouand (ed.), Handbook of Cell Biosensors, https://doi.org/10.1007/978-3-030-23217-7_194

3

4

G. Thouand

Introduction The very first observations of cells, like those made of microorganisms by Antonie Van Leeuwenhoek and Louis Joblot in the seventeenth century, paved the way to what we now know as modern microbiology. It has been estimated that 50% of the earth’s biomass consists of microbes – parasites, diatoms, microalgae, fungi, bacteria, and viruses – and, of these, viruses (living parasitic organisms that thrive off the cells they infect) have reached an astronomical number (1031). The use of microbes in modern biotechnology goes back more than a hundred years, but cell biosensors are more recent. According to the main bibliographic databases, the idea of associating a cell with a sensor to identify a phenomenon dates from Serat’s work in the 1960s (Serat et al. 1965). It seems to me to have been very ahead of its time, since the authors published a way to measure the toxicity of air pollution with bioluminescent bacteria. Once the concept had been accepted, the same authors published a paper on a real transportable device 2 years later (Serat et al. 1967). Detecting the toxicity of airborne chemical pollution remains an ongoing research challenge 50 years later! Introducing as broad a topic as cell biosensors is complicated because everyone wants to find their favorite theme. This handbook lets the experts of each of the three different sections introduce each one. My aim here is to summarize the literature in a straightforward manner in just a few pages, with the main objective of allowing nonexperts (students, researchers changing their field, and industry professionals) to quickly get an overview of the field. Therefore, in the following parts of this introduction, I propose the following: first, the description of a cell biosensor; then, a study of the evolution of the research themes over the last 55 years; and finally, an “ideal bibliography” of 506 articles from among the most cited of the last 55 years.

What Is a Cell Biosensor? A cell biosensor is, above all, a hybrid system associating biology, electronics, materials, and digital science. It is a measuring instrument with multiple applications that combines a biological recognition element (here a cell usually called bioelement) with an electronic part (Eltzov and Marks 2011). Figure 1 summarizes the concept, which can be broken down into five parts, forming a coherent and complementary series from the cell to be used as the bioelement to the finished biosensor application. The cell part is the key to the biosensor. It comes into contact with the analyte to be measured (metals, antibiotics, organic pollutants, organic mixtures, pathogens, etc.) and reacts with it, leading to a signal (electron transfer, bioluminescence, heat, etc.). Six families of cells are described in the literature. The main one is the bacterial family, used since 1965, for which the knowledge on microbial diversity is constantly evolving (Hug et al. 2016; Thompson et al. 2017). Next come fungal cells and microalgae, then animal and plant cells, and over the past 10 years, bacterial

1

Introduction to Cell Biosensors Through 55 Years of Scientific Production

5

Fig. 1 Illustration of the five parts comprising the concept of cell biosensors

viruses (bacteriophages), the last of which are used for the detection of pathogenic bacteria in healthcare, food, and the environment. A virus is certainly not considered as an autonomous cell, but as belonging globally to the world of microorganisms and playing an essential role in the regulation of microbial ecosystems (Clokie et al. 2011; Al-Shayeb et al. 2020). The choice of one family of cells over another will depend not only on the detection capabilities, the robustness of the cell during use, and its storage capacity,

6

G. Thouand

but also on the ease of its production. Section “Introduction” (Fundamentals and Genetics for Cell Biosensor Applications – coordinated by Shimshon Belkin and Paul Freemont) covers this important aspect. The choice of transducer depends entirely on the signal produced by the cells. For example, an optical transducer, such as a photomultiplier, could be used if the microorganism produces light, or an electrochemical sensor could be used if an electric current is modified or produced by the cell as a result of redox reactions, etc. (see section “What Is a Cell Biosensor?” – Transducers, Materials and Systems – coordinated by Sylvia Daunert, Sylvain Martel, Elisa Michelini, and Aldo Roda, for more details). To ensure contact between the cell and the transducer, immobilization is sometimes necessary. The immobilization matrix is an inert material, nontoxic material for the cells. Its role is to confine them close to the transducer in an aqueous environment, as limiting their movement will improve the signal transfer to the transducer (see section “What Is a Cell Biosensor?” – Transducers, Materials and Systems – for more details). System integration is an important issue when we want to transfer the biosensor to an application (environment, food, or health). System integration must take into account the constraints of the field application. This is a question of integrating the biological part, the transducer, its electronics, and the environment necessary for the survival of the cells (at least the temperature and oxygen controls) into a compact system ready to be used or transported. Systems engineering associated with biology requires specific skills to successfully develop operational biosensors (see section “What Is a Cell Biosensor?” – Transducers, Materials and Systems – and section “Evolution of the Research Topics from 1965 to 2020” – Applications of Biosensors – coordinated by Sylvia Daunert, Julie Hermans, Isao Karube, and Gérald Thouand, for more details).

Evolution of the Research Topics from 1965 to 2020 The evolution of the research topics associated with cell biosensors is part of the history of this scientific field. It is not only a sign of the perception of the researchers of the generations who built the themes and published, but also a reflection of the technologies and scientific knowledge of the time. The 55 years of the “cell biosensor” theme are marked by 3126 publications listed in the databases (Fig. 2). This scientific production was divided into four main periods (1965–1990, 1991– 2000, 2001–2010, and 2011–2020) and then analyzed using a bibliometric tool that identifies recurrent keywords forming the most important themes covered. Articles are only listed if the authors adopted the usual keywords of the themes covered by “cell biosensor.” In the first period (Fig. 2a), research was initiated essentially around microorganisms as bioelements (the cyanobacterium Synechococcus, for example) and tackled the nascent subject of recombinant bacteria (that has subsequently flourished). Indeed, the early 1980s were a turning point for microbial biosensors

B C

A D

2001 - 2010

1991 - 2000

966 publications

274 publications

Threshold : 7

Threshold : 4

Introduction to Cell Biosensors Through 55 Years of Scientific Production

Fig. 2 The main topics associated with cell biosensors. Bibliometric overview using the Scopus databank (keywords: biosensors AND (microbial OR Bioreporter OR whole-cell OR Bacterial) and VOSviewer software for analysis (free access, https://www.vosviewer.com/)). The threshold is the value applied to

1844 publications Threshold : 11

2011 - 2020

Threshold : 1

42 publications

1965 - 1990

1 7

8

G. Thouand

because Belas et al. (1982) and Engebrecht et al. (1983) cloned bioluminescence genes from Vibrio harveyi and Aliivibrio fischeri, respectively, into Escherichia coli bacteria, which then became bioluminescent. The major application of biosensors for measuring biochemical oxygen demand (BOD) in combination with amperometric transducers was already well represented and continues to be 55 years later. The second period, during which there were 274 publications (Fig. 2b), saw a continuation of the “historical” themes (research into new cell immobilization matrices, BOD biosensors) and the emergence of another strongly represented theme between 1991 and 2000: the application of bioluminescent bacteria to the detection of heavy metals and effects of pollutants (toxicity). The trend was confirmed in the following decade (third period, 966 publications, Fig. 2c) with an explosion of themes and, therefore, of keywords. A threshold of seven had to be imposed (i.e., the keyword had to be cited at least seven times to be included) to avoid saturation of the graph. This was a period of confirmation and hyperproduction on the themes of bioluminescent bioassays and biosensors, this time seeking to apply these techniques to all kinds of targets (pesticides, mercury, etc.) and to explain their bioavailability or biodegradation. Optical sensors (fluorescence and bioluminescence measurements) are being deployed and becoming more generalized in biology laboratories. Three bioelements in particular were used by researchers: not only Escherichia coli (the model most frequently used in publications involving recombinant bacteria), but also Pseudomonas putida and Gluconobacter oxydans. The topic of microbial fuel cells gained in interest, reflected by its significant presence among the publications. In the current period (Fig. 2d), biosensor cell research is flourishing, with 1844 publications. The historical themes (BOD, fuel cells, etc.) continue, but work in the current decade is also following the times, particularly the progress in genetics and molecular biology. Themes related to synthetic biology, transcription factors, and metabolic engineering, to modify, improve, or understand the bioelements associated with biosensors, are also appearing in significant numbers. Bacteriophages, as new bioelements for cell biosensors, are reemerging in force after timid attempts in earlier decades, opening the way for the detection of pathogenic bacteria.

The “Ideal” Bibliography If you want to discover a new research theme or if you are a student studying for a Master or PhD degree, where should you start reading? Should you attempt to read the entire bibliography produced since the emergence of the research theme or ä Fig. 2 (continued) select the number of times a keyword has been used and is used to limit the saturation effect when too many insignificant keywords are taken into account. A threshold of 1 means that no threshold has been set, and so all published keywords will be taken into account in the figure. A threshold of 11 means that the keyword must have been cited at least 11 times by different publications to be taken into account

1

Introduction to Cell Biosensors Through 55 Years of Scientific Production

9

instead make specific choices? The scientific questions we ask ourselves can often already be found in the literature, although this is sometimes in older publications. So, how can we identify these articles? Ideally, one should read all 3126 articles, but, with research programs becoming increasingly shorter, there is rarely enough time. Is there, therefore, an “ideal bibliography” to gain a complete and historical view of a field without leaving out the articles that have most influenced it? In this last chapter, I propose to consider the top 10% of most cited articles published on cell biosensors since 1965. These papers have left their mark on the field because so many researchers have seized upon them and cited them. The selection was made according to the bibliometric standards proposed by specialists (Bornmann 2014). This does not detract from the other articles, which are perhaps just as exciting and of equally high scientific quality, but simply less visible. The result is 506 scientific articles and reviews published since 1965. These are referenced at the end of this introduction, accompanied by their DOIs for easier searching. Figure 3 illustrates this production according to the four main themes: bioelements, immobilization, transducer-engineering, and field applications. Each point represents a reference (x-axis) associated with its number of citations (y-axis) according to the four periods of scientific production (1965–2020) identified by colors. A single reference can be found under several themes. The articles of the current period (2011–2020) have obviously been cited less than those of the previous periods (150 citations maximum), and the oldest period has apparently been totally forgotten, since excellent, founder articles do not have more than 100 citations 50 years after their publication. Figure 3 shows, on the one hand, that publications have been mainly oriented toward bioelements and field applications (mostly environmental) with continuous production. D0 Souza’s review (2001, N 125, 469 citations) on microbial biosensors is the most frequently cited. On the other hand, publications on transducers associated with the engineering needed for the bioelement integration, as well as those on immobilization, have shown discontinuous production. This can be explained by the fact that innovations in immobilization materials and transducers can cover the needs of the two main themes several times over. The theme of materials for immobilization, which is the least represented, is covered by 53 articles, including the article by Avnir et al. (2006, N 85, 632 citations). This latter paper inspired many scientists and not only in the field of biosensors, as biomaterials cover a wide range of cross-cutting application themes. The same is true of the article by Gil et al. (2003, N 137, 795 citations) dealing with the transducer part associated with a microbial fuel cell.

Conclusions and Future Directions Cell biosensors, whose concept was born 55 years ago, have never ceased to be enriched by the scientific ambitions and ingenuity of researchers, and societal needs in terms of applications. With several thousand articles and reviews, this crossdisciplinary theme, which draws on multiple disciplines, has seen continual innovation and progression. It is now entering a phase of scientific maturity that enables it

10

G. Thouand BIOELEMENT

IMMOBILISATION 800

1965-1990

700

1991-2000

600

2001-2010

500

2011-2020

400

300 200 100

0 0

20

40

60

80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400 420 440 460 480 500

TRANSDUCER-ENGINEERING 800

1965-1990

700

1991-2000

600

2001-2010

500

2011-2020

400

300 200 100

Number of citations of the publication

0 0

20

40

60

80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400 420 440 460 480 500

FIELD APPLICATIONS 800

1965-1990

700

1991-2000

600

2001-2010

500

2011-2020

400

300 200 100

0 0

20

40

60

80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400 420 440 460 480 500

Number of the Top 10% cited publication

Fig. 3 “The ideal bibliography” A representation of the 10% most cited articles during the period 1965–2020 in relation to the four main categories involving cell biosensors. Each point corresponds to a citation associated with a number on the x-axis and its number of citations on the y-axis. Each number corresponds to a bibliographic reference at the end of the document and its DOI, by which it may be easily found in databases

1

Introduction to Cell Biosensors Through 55 Years of Scientific Production

11

to respond to needs for field applications. New challenges, including the training of future technicians, engineers, and researchers, are emerging. Indeed, in a transdisciplinary scientific field combining so much specific knowledge, the need for pedagogy and transversal teaching is keenly felt. This handbook on cell biosensors aims to modestly contribute to this ambition.

References Al-Shayeb B, Sachdeva R, Chen LX et al (2020) Clades of huge phages from across Earth’s ecosystems. Nature 578:425–431 Belas R, Mileham A, Cohn D, Hilmen M, Simon M, Silverman M (1982) Bacterial bioluminescence: isolation and expression of the luciferase genes from Vibrio harveyi. Science 218:791–793 Bornmann L (2014) How are excellent (highly cited) papers defined in bibliometrics? A quantitative analysis of the literature. Res Eval 23(2):166 Clokie MRJ, Millard AD, Letarov AV, Heaphy S (2011) Phages in nature. Bacteriophage 1(1):31–45 Eltzov E, Marks RS (2011) Whole-cell aquatic biosensors. Anal Bioanal Chem 400:895–913 Engebrecht J, Nealson K, Silverman M (1983) Bacterial bioluminescence: isolation and genetic analysis of functions from Vibrio fischeri. Cell 32:773–781 Hug L, Baker B, Anantharaman K et al (2016) A new view of the tree of life. Nat Microbiol 1:16048 Serat WF, Budinger FE, Mueller PK (1965) Evaluation of biological effects of air pollutants by use of luminescent bacteria. J Bacteriol 90(3):832–833 Serat WF, Budinger FE, Mueller PK (1967) Toxicity evaluation of air pollutants by use of luminescent bacteria. Atmos Environ Pergamont Press 1:21–32 Thompson L, Sanders J, McDonald D et al (2017) A communal catalogue reveals Earth’s multiscale microbial diversity. Nature 551:457–463

References Related to Fig. 3 10% most cited papers (research and review) from 1965 to 2020 on the topic of cell biosensors. References are classified in alphabetical order. Numbers correspond to positions in Fig. 3.

1965–1990 1. Corcoran CA, Rechnitz GA (1985) Cell-based biosensors. Trends Biotechnol 3(4):92–96. https://doi.org/10.1016/0167-7799(85)90091-5 2. Deshpande MV, Hall EAH (1990) An electrochemically grown polymer as an immobilisation matrix for whole cells: application in an amperometric dopamine sensor. Biosens Bioelectron 5 (6):431–448. https://doi.org/10.1016/0956-5663(90)80033-A 3. Galindo E, Bautista D, García JL, Quintero R (1990) Microbial sensor for penicillins using a recombinant strain of Escherichia coli. Enzym Microb Technol 12(9):642–646. https://doi.org/ 10.1016/0141-0229(90)90001-7 4. Karube I, Suzuki S (1984) Amperometric and potentiometric determinations with immobilized enzyme and microorganisms. Ion-Selective Electrode Rev 6(1):15–58. https://doi.org/10.1016/ b978-0-08-033201-7.50006-2 5. Karube I, Tamiya E (1987) Biosensors for environmental control. Pure Appl Chem 59(4):545– 554. https://doi.org/10.1351/pac198759040545

12

G. Thouand

6. Karube I, Sode K, Tamiya E, Yokoyama K (1989) Microbial bod sensor utilizing thermophilic bacteria. Anal Lett 22(4):791–801. https://doi.org/10.1080/00032718908051367 7. Kreysa G, Sell D, Kraemer P (1990) Bioelectrochemical fuel cells. Berichte Bunsengesellschaft/Phys Chem Chem Phys 94(9):1042–1045. https://doi.org/10.1002/bbpc. 19900940933 8. Okada T, Karube I, Suzuki S (1983) NO2 sensor which uses immobilized nitrite oxidizing bacteria. Biotechnol Bioeng 25(6):1641–1651. https://doi.org/10.1002/bit.260250616 9. Rawson DM, Willmer AJ, Cardosi MF (1987) The development of whole cell biosensors for online screening of herbicide pollution of surface waters. Toxic Assess 2(3):325–340. https://doi. org/10.1002/tox.2540020308 10. Rawson DM, Willmer AJ, Turner APP (1989) Whole-cell biosensors for environmental monitoring. Biosensors 4(5):299–311. https://doi.org/10.1016/0265-928X(89)80011-2 11. Riedel K, Scheller F (1987) Inhibitor-treated microbial sensor for the selective determination of glutamic acid. Analyst 112(3):341–342. https://doi.org/10.1039/an9871200341 12. Riedel K, Lange K, Stein H, Kühn M, Ott P, Scheller F (1990) A microbial sensor for BOD. Water Res 24(7):883–887. https://doi.org/10.1016/0043-1354(90)90138-V 13. Riedel K, Renneberg R, Wollenberger U, Kaiser G, Scheller FW (1989) Microbial sensors: fundamentals and application for process control. J Chem Technol Biotechnol 44(2):85–106. https://doi.org/10.1002/jctb.280440202 14. Scheller FW, Schubert F, Renneberg R, Müller H, Jänchen M, Weise H (1985) Biosensors: trends and commercialization. Biosensors 1(2):135–160. https://doi.org/10.1016/0265-928X (85)80001-8 15. Serat WF, Budinger FE, Mueller PK (1965) Evaluation of biological effects of air pollutants by use of luminescent bacteria. J Bacteriol 90(3):832–833. https://doi.org/10.1128/JB.90.3.832833.1965 16. Trevan MD, Mak AL (1988) Immobilized algae and their potential for use as biocatalysts. Trends Biotechnol 6(3):68–73. https://doi.org/10.1016/0167-7799(88)90094-7

1991–2000 17. Belkin S, Smulski DR, Dadon S, Vollmer AC, Van Dyk TK, Larossa RA (1997) A panel of stress-responsive luminous bacteria for the detection of selected classes of toxicants. Water Res 31(12):3009–3016. https://doi.org/10.1016/S0043-1354(97)00169-3 18. Belkin S, Smulski DR, Vollmer AC, Van Dyk TK, Larossa RA (1996) Oxidative stress detection with Escherichia coli harboring a katG’::Lux fusion. Appl Environ Microbiol 62(7):2252– 2256. https://doi.org/10.1128/aem.62.7.2252-2256.1996 19. Billard P, DuBow MS (1998) Bioluminescence-based assays for detection and characterization of bacteria and chemicals in clinical laboratories. Clin Biochem 31(1):1–14. https://doi.org/10. 1016/S0009-9120(97)00136-7 20. Biran I, Babai R, Levcov K, Rishpon J, Ron EZ (2000) Online and in situ monitoring of environmental pollutants: electrochemical biosensing of cadmium. Environ Microbiol 2 (3):285–290. https://doi.org/10.1046/j.1462-2920.2000.00103.x 21. Bontidean I, Lloyd JR, Hobman JL, Wilson JR, Csöregi E, Mattiasson B, Brown NL (2000) Bacterial metal-resistance proteins and their use in biosensors for the detection of bioavailable heavy metals. J Inorg Biochem 79(1–4):225–229. https://doi.org/10.1016/S01620134(99)00234-2 22. Bousse L (1996) Whole cell biosensors. Sensors Actuators B Chem 34(1–3):270–275. https:// doi.org/10.1016/S0925-4005(96)01906-5 23. Byfield MP, Abuknesha RA (1994) Biochemical aspects of biosensors. Biosens Bioelectron 9 (4–5):373–399. https://doi.org/10.1016/0956-5663(94)80038-3

1

Introduction to Cell Biosensors Through 55 Years of Scientific Production

13

24. Chan C, Lehmann M, Chan K, Chan P, Chan C, Gruendig B, . . . Renneberg R (2000) Designing an amperometric thick-film microbial BOD sensor. Biosens Bioelectron 15(7–8):343–353. https://doi.org/10.1016/S0956-5663(00)00090-7 25. Cheol Gil G, Mitchell RJ, Tai Chang S, Bock Gu M (2000) A biosensor for the detection of gas toxicity using a recombinant bioluminescent bacterium. Biosens Bioelectron 15(1–2):23–30. https://doi.org/10.1016/S0956-5663(99)00074-3 26. Collard JM, Corbisier P, Diels L, Dong Q, Jeanthon C, Mergeay M, . . . Wuertz S (1994) Plasmids for heavy metal resistance in Alcaligenes eutrophus CH34: mechanisms and applications. FEMS Microbiol Rev 14(4):405–414. https://doi.org/10.1111/j.1574-6976.1994. tb00115.x 27. Daunert S, Barrett G, Feliciano JS, Shetty RS, Shrestha S, Smith-Spencer W (2000) Genetically engineered whole-cell sensing systems: coupling biological recognition with reporter genes. Chem Rev 100(7):2705–2738. https://doi.org/10.1021/cr990115p 28. Elasri MO, Miller RV (1999) Study of the response of a biofilm bacterial community to UV radiation. Appl Environ Microbiol 65(5):2025–2031. https://doi.org/10.1128/aem.65.5.20252031.1999 29. Errampalli D, Leung K, Cassidy MB, Kostrzynska M, Blears M, Lee H, Trevors JT (1999) Applications of the green fluorescent protein as a molecular marker in environmental microorganisms. J Microbiol Methods 35(3):187–199. https://doi.org/10.1016/S0167-7012 (99)00024-X 30. Fernandes PB (1998) Technological advances in high-throughput screening. Curr Opin Chem Biol 2(5):597–603. https://doi.org/10.1016/S1367-5931(98)80089-6 31. Gonchar MV, Maidan MM, Moroz OM, Woodward JR, Sibirny AA (1998) Microbial O2- and H2O2-electrode sensors for alcohol assays based on the use of permeabilized mutant yeast cells as the sensitive bioelements. Biosens Bioelectron 13(9):945–952. https://doi.org/10.1016/ S0956-5663(98)00034-7 32. Hansen LH, Sørensen SJ (2000) Detection and quantification of tetracyclines by whole cell biosensors. FEMS Microbiol Lett 190(2):273–278. https://doi.org/10.1016/S0378-1097(00) 00347-5 33. Hansen LH, Sørensen SJ (2000) Versatile biosensor vectors for detection and quantification of mercury. FEMS Microbiol Lett 193(1):123–127. https://doi.org/10.1016/S0378-1097(00) 00462-6 34. Hazama A, Hayashi S, Okada Y (1998) Cell surface measurements of ATP release from single pancreatic β cells using a novel biosensor technique. Pflugers Archiv Eur J Physiol 437(1):31– 35. https://doi.org/10.1007/s004240050742 35. Heitzer A, Malachowsky K, Thonnard JE, Bienkowski PR, White DC, Sayler GS (1994) Optical biosensor for environmental on-line monitoring of naphthalene and salicylate bioavailability with an immobilized bioluminescent catabolic reporter bacterium. Appl Environ Microbiol 60(5):1487–1494. https://doi.org/10.1128/aem.60.5.1487-1494.1994 36. Højberg O, Schnider U, Winteler HV, Sørensen J, Haas D (1999) Oxygen-sensing reporter strain of Pseudomonas fluorescens for monitoring the distribution of low-oxygen habitats in soil. Appl Environ Microbiol 65(9):4085–4093. https://doi.org/10.1128/aem.65.9.4085-4093. 1999 37. Hollis RP, Killham K, Glover LA (2000) Design and application of a biosensor for monitoring toxicity of compounds to eukaryotes. Appl Environ Microbiol 66(4):1676–1679. https://doi. org/10.1128/AEM.66.4.1676-1679.2000 38. Hyun C, Tamiya E, Takeuchi T, Karube I, Inoue N (1993) A novel BOD sensor based on bacterial luminescence. Biotechnol Bioeng 41(11):1107–1111. https://doi.org/10.1002/bit. 260411114 39. Jaeger CH III, Lindow SE, Miller W, Clark E, Firestone MK (1999) Mapping of sugar and amino acid availability in soil around roots with bacterial sensors of sucrose and tryptophan. Appl Environ Microbiol 65(6):2685–2690. https://doi.org/10.1128/aem.65.6.2685-2690.1999

14

G. Thouand

40. Joyner DC, Lindow SE (2000) Heterogeneity of iron bioavailability on plants assessed with a whole-cell GFP-based bacterial biosensor. Microbiology 146(10):2435–2445. https://doi.org/ 10.1099/00221287-146-10-2435 41. Kim HJ, Hyun MS, Chang IS, Kim BH (1999) A microbial fuel cell type lactate biosensor using a metal-reducing bacterium, Shewanella putrefaciens. J Microbiol Biotechnol 9(3):365–367. www.scopus.com 42. Köhler S, Belkin S, Schmid RD (2000) Reporter gene bioassays in environmental analysis. Fresenius J Anal Chem 366(6–7):769–779. https://doi.org/10.1007/s002160051571 43. Kong IC, Bitton G, Koopman B, Jung KH (1995) Heavy metal toxicity testing in environmental samples. Rev Environ Contam Toxicol 142:119–147. https://doi.org/10.1007/978-1-46124252-9_5 44. Lehmann M, Riedel K, Adler K, Kunze G (2000) Amperometric measurement of copper ions with a deputy substrate using a novel Saccharomyces cerevisiae sensor. Biosens Bioelectron 15 (3–4):211–219. https://doi.org/10.1016/S0956-5663(00)00060-9 45. Lin J, Brown CW (1997) Sol-gel glass as a matrix for chemical and biochemical sensing. TrAC – Trends Anal Chem 16(4):200–211. https://doi.org/10.1016/S0165-9936(97)00021-6 46. Liu J, Björnsson L, Mattiasson B (2000) Immobilised activated sludge based biosensor for biochemical oxygen demand measurement. Biosens Bioelectron 14(12):883–893. https://doi. org/10.1016/S0956-5663(99)00064-0 47. Loper JE, Lindow SE (1994) A biological sensor for iron available to bacteria in their habitats on plant surfaces. Appl Environ Microbiol 60(6):1934–1941. https://doi.org/10.1128/aem.60.6. 1934-1941.1994 48. Marty J, Garcia D, Rouillon R (1995) Biosensors: Potential in pesticide detection. Trends Anal Chem 14(7):329–333. https://doi.org/10.1016/0165-9936(95)97060-E 49. Mulchandani A, Kaneva I, Chen W (1998) Biosensor for direct determination of organophosphate nerve agents using recombinant Escherichia coli with surface-expressed organophosphorus hydrolase. 2. Fiber-optic microbial biosensor. Anal Chem 70(23):5042–5046. https://doi. org/10.1021/ac980643l 50. Mulchandani A, Mulchandani P, Kaneva I, Chen W (1998) Biosensor for direct determination of organophosphate nerve agents using recombinant Escherichia coli with surface-expressed organophosphorus hydrolase. 1. Potentiometric microbial electrode. Anal Chem 70(19):4140– 4145. https://doi.org/10.1021/ac9805201 51. Nomura Y, Ikebukuro K, Yokoyama K, Takeuchi T, Arikawa Y, Ohno S, Karube I (1998) Application of a linear alkylbenzene sulfonate biosensor to river water monitoring. Biosens Bioelectron 13(9):1047–1053. https://doi.org/10.1016/S0956-5663(97)00077-8 52. Nomura Y, Ikebukuro K, Yokoyama K, Takeuchi T, Karube I, Arikawa Y, Ohno S (1994) A novel microbial sensor for anionic surfactant determination. Anal Lett 27(15):3095–3108. https://doi.org/10.1080/00032719408000313 53. Paddle BM (1996) Biosensors for chemical and biological agents of defence interest. Biosens Bioelectron 11(11):1079–1113. https://doi.org/10.1016/0956-5663(96)82333-5 54. Peitzsch N, Eberz G, Nies DH (1998) Alcaligenes eutrophus as a bacterial chromate sensor. Appl Environ Microbiol 64(2):453–458. https://doi.org/10.1128/aem.64.2.453-458.1998 55. Peter J, Hutter W, Stöllnberger W, Hampel W (1996) Detection of chlorinated and brominated hydrocarbons by an ion sensitive whole cell biosensor. Biosens Bioelectron 11(12):1215–1219. https://doi.org/10.1016/0956-5663(96)88086-9 56. Preininger C, Klimant I, Wolfbeis OS (1994) Optical fiber sensor for biological oxygen demand. Anal Chem 66(11):1841–1846. https://doi.org/10.1021/ac00083a011 57. Preston S, Coad N, Townend J, Killham K, Paton GI (2000) Biosensing the acute toxicity of metal interactions: are they additive, synergistic, or antagonistic? Environ Toxicol Chem 19 (3):775–780. https://doi.org/10.1002/etc.5620190332 58. Ptitsyn LR, Horneck G, Komova O, Kozubek S, Krasavin EA, Bonev M, Rettberg P (1997) A biosensor for environmental genotoxin screening based on an SOS lux assay in

1

Introduction to Cell Biosensors Through 55 Years of Scientific Production

15

recombinantEscherichia coli cells. Appl Environ Microbiol 63(11):4377–4384. https://doi.org/ 10.1128/aem.63.11.4377-4384.1997 59. Rainina EI, Efremenco EN, Varfolomeyev SD, Simonian AL, Wild JR (1996) The development of a new biosensor based on recombinant E. coli for the direct detection of organophosphorus neurotoxins. Biosens Bioelectron 11(10):991–1000. https://doi.org/10.1016/0956-5663(96) 87658-5 60. Ramanathan S, Ensor M, Daunert S (1997) Bacterial biosensors for monitoring toxic metals. Trends Biotechnol 15(12):500–506. https://doi.org/10.1016/S0167-7799(97)01120-7 61. Rasmussen LD, Sørensen SJ, Turner RR, Barkay T (2000) Application of a mer-lux biosensor for estimating bioavailable mercury in soil. Soil Biol Biochem 32(5):639–646. https://doi.org/ 10.1016/S0038-0717(99)00190-X 62. Ripp S, Nivens DE, Ahn Y, Werner C, Jarrell IV J, Easter JP, . . . Sayler GS (2000) Controlled field release of a bioluminescent genetically engineered microorganism for bioremediation process monitoring and control. Environ Sci Technol 34(5):846–853. https://doi.org/10.1021/ es9908319 63. Schmidt A, Standfuß-Gabisch C, Bilitewski U (1996) Microbial biosensor for free fatty acids using an oxygen electrode based on thick film technology. Biosens Bioelectron 11(11):1139– 1145. https://doi.org/10.1016/0956-5663(96)82336-0 64. Simpson ML, Sayler GS, Applegate BM, Ripp S, Nivens DE, Paulus MJ, Jellison GE Jr (1998) Bioluminescent-bioreporter integrated circuits form novel whole-cell biosensors. Trends Biotechnol 16(8):332–338. https://doi.org/10.1016/S0167-7799(98)01199-8 65. Sinclair GM, Paton GI, Meharg AA, Killham K (1999) Lux-biosensor assessment of pH effects on microbial sorption and toxicity of chlorophenols. FEMS Microbiol Lett 174(2):273–278. https://doi.org/10.1016/S0378-1097(99)00157-3 66. Sousa S, Duffy C, Weitz H, Glover LA, Bär E, Henkler R, Killham K (1998) Use of a luxmodified bacterial biosensor to identify constraints to bioremediation of BTEX-contaminated sites. Environ Toxicol Chem 17(6):1039–1045. https://doi.org/10.1897/1551-5028(1998) 0172.3.CO;2 67. Sticher P, Jaspers MCM, Stemmler K, Harms H, Zehnder AJB, Van Der Meer JR (1997) Development and characterization of a whole-cell bioluminescent sensor for bioavailable middle-chain alkanes in contaminated groundwater samples. Appl Environ Microbiol 63 (10):4053–4060. https://doi.org/10.1128/aem.63.10.4053-4060.1997 68. Švitel J, Čurilla O, Tkáč J (1998) Microbial cell-based biosensor for sensing glucose, sucrose or lactose. Biotechnol Appl Biochem 27(2):153–158. www.scopus.com 69. Tan TC, Li F, Neoh KG, Lee YK (1992) Microbial membrane-modified dissolved oxygen probe for rapid biochemical oxygen demand measurement. Sensors Actuators: B Chem 8(2):167–172. https://doi.org/10.1016/0925-4005(92)80175-W 70. Tauriainen S, Karp M, Chang W, Virta M (1998) Luminescent bacterial sensor for cadmium and lead. Biosens Bioelectron 13(9):931–938. https://doi.org/10.1016/S0956-5663 (98)00027-X 71. Tauriainen S, Virta M, Chang W, Karp M (1999) Measurement of firefly luciferase reporter gene activity from cells and lysates using Escherichia coli arsenite and mercury sensors. Anal Biochem 272(2):191–198. https://doi.org/10.1006/abio.1999.4193 72. Tescione L, Belfort G (1993) Construction and evaluation of a metal ion biosensor. Biotechnol Bioeng 42(8):945–952. https://doi.org/10.1002/bit.260420805 73. Turpeinen R, Salminen J, Kairesalo T (2000) Mobility and bioavailability of lead in contaminated boreal forest soil. Environ Sci Technol 34(24):5152–5156. https://doi.org/10.1021/ es001200d 74. Van Dyk TK, Majarian WR, Konstantinov KB, Young RM, Dhurjati PS, LaRossa RA (1994) Rapid and sensitive pollutant detection by induction of heat shock gene- bioluminescence gene fusions. Appl Environ Microbiol 60(5):1414–1420. https://doi.org/10.1128/aem.60.5.14141420.1994

16

G. Thouand

75. Vollmer AC, Kwakye S, Halpern M, Everbach EC (1998) Bacterial stress responses to 1megahertz pulsed ultrasound in the presence of microbubbles. Appl Environ Microbiol 64 (10):3927–3931. https://doi.org/10.1128/aem.64.10.3927-3931.1998 76. Vulkan R, Zhao F, Barbosa-Jefferson V, Preston S, Paton GI, Tipping E, McGrath SP (2000) Copper speciation and impacts on bacterial biosensors in the pore water of copper-contaminated soils. Environ Sci Technol 34(24):5115–5121. https://doi.org/10.1021/es0000910 77. Willardson BM, Wilkins JF, Rand TA, Schupp JM, Hill KK, Keim P, Jackson PJ (1998) Development and testing of a bacterial biosensor for toluene-based environmental contaminants. Appl Environ Microbiol 64(3):1006–1012. https://doi.org/10.1128/aem.64.3.1006-1012.1998 78. Wise AA, Kuske CR (2000) Generation of novel bacterial regulatory proteins that detect priority pollutant phenols. Appl Environ Microbiol 66(1):163–169. https://doi.org/10.1128/ AEM.66.1.163-169.2000 79. Ziegler C, Göpel W (1998) Biosensor development. Curr Opin Chem Biol 2(5):585–591. https://doi.org/10.1016/S1367-5931(98)80087-2

2001–2010 80. Akyilmaz E, Dinçkaya E (2005) An amperometric microbial biosensor development based on Candida tropicalis yeast cells for sensitive determination of ethanol. Biosens Bioelectron 20 (7):1263–1269. https://doi.org/10.1016/j.bios.2004.04.010 81. Akyilmaz E, Erdoǧan A, Öztürk R, Yaşa I (2007) Sensitive determination of L-lysine with a new amperometric microbial biosensor based on Saccharomyces cerevisiae yeast cells. Biosens Bioelectron 22(6):1055–1060. https://doi.org/10.1016/j.bios.2006.04.023 82. Akyilmaz E, Yaşa I, Dinçkaya E (2006) Whole cell immobilized amperometric biosensor based on Saccharomyces cerevisiae for selective determination of vitamin B1 (thiamine). Anal Biochem 354(1):78–84. https://doi.org/10.1016/j.ab.2006.04.019 83. Alvarez GS, Foglia ML, Copello GJ, Desimone MF, Diaz LE (2009) Effect of various parameters on viability and growth of bacteria immobilized in sol-gel-derived silica matrices. Appl Microbiol Biotechnol 82(4):639–646. https://doi.org/10.1007/s00253-008-1783-9 84. Aravanis AM, DeBusschere BD, Chruscinski AJ, Gilchrist KH, Kobilka BK, Kovacs GTA (2001) A genetically engineered cell-based biosensor for functional classification of agents. Biosens Bioelectron 16(7–8):571–577. https://doi.org/10.1016/S0956-5663(01)00171-3 85. Avnir D, Lev O, Livage J (2006) Recent bio-applications of sol-gel materials. J Mater Chem 16 (11):1013–1030. https://doi.org/10.1039/b512706h 86. Axtell CA, Beattie GA (2002) Construction and characterization of a proU-gfp transcriptional fusion that measures water availability in a microbial habitat. Appl Environ Microbiol 68 (9):4604–4612. https://doi.org/10.1128/AEM.68.9.4604-4612.2002 87. Balasubramanian S, Sorokulova IB, Vodyanoy VJ, Simonian AL (2007) Lytic phage as a specific and selective probe for detection of Staphylococcus aureus – a surface plasmon resonance spectroscopic study. Biosens Bioelectron 22(6):948–955. https://doi.org/10.1016/ j.bios.2006.04.003 88. Banerjee P, Bhunia AK (2010) Cell-based biosensor for rapid screening of pathogens and toxins. Biosens Bioelectron 26(1):99–106. https://doi.org/10.1016/j.bios.2010.05.020 89. Banerjee P, Lenz D, Robinson JP, Rickus JL, Bhunia AK (2008) A novel and simple cell-based detection system with a collagen-encapsulated B-lymphocyte cell line as a biosensor for rapid detection of pathogens and toxins. Lab Investig 88(2):196–206. https://doi.org/10.1038/ labinvest.3700703 90. Banik RM, Mayank PR, Upadhyay SN (2008) Microbial biosensor based on whole cell of Pseudomonas sp. for online measurement of p-nitrophenol. Sensors Actuators B Chem 131 (1):295–300. https://doi.org/10.1016/j.snb.2007.11.022 91. Baronian KHR (2004) The use of yeast and moulds as sensing elements in biosensors. Biosens Bioelectron 19(9):953–962. https://doi.org/10.1016/j.bios.2003.09.010

1

Introduction to Cell Biosensors Through 55 Years of Scientific Production

17

92. Barron JA, Rosen R, Jones-Meehan J, Spargo BJ, Belkin S, Ringeisen BR (2004) Biological laser printing of genetically modified Escherichia coli for biosensor applications. Biosens Bioelectron 20(2):246–252. https://doi.org/10.1016/j.bios.2004.01.011 93. Bechor O, Smulski DR, Van Dyk TK, LaRossa RA, Belkin S (2002) Recombinant microorganisms as environmental biosensors: pollutants detection by Escherichia coli bearing fabA0 :: Lux fusions. J Biotechnol 94(1):125–132. https://doi.org/10.1016/S0168-1656(01)00423-0 94. Belkin S (2003) Microbial whole-cell sensing systems of environmental pollutants. Curr Opin Microbiol 6(3):206–212. https://doi.org/10.1016/S1369-5274(03)00059-6 95. Ben-Yoav H, Biran A, Pedahzur R, Belkin S, Buchinger S, Reifferscheid G, ShachamDiamand Y (2009) A whole cell electrochemical biosensor for water genotoxicity bio-detection. Electrochim Acta 54(25):6113–6118. https://doi.org/10.1016/j.electacta.2009.01.061 96. Bhattacharyya J, Read D, Amos S, Dooley S, Killham K, Paton GI (2005) Biosensor-based diagnostics of contaminated groundwater: assessment and remediation strategy. Environ Pollut 134(3):485–492. https://doi.org/10.1016/j.envpol.2004.09.002 97. Biran I, Rissin DM, Ron EZ, Walt DR (2003) Optical imaging fiber-based live bacterial cell array biosensor. Anal Biochem 315(1):106–113. https://doi.org/10.1016/S0003-2697(02) 00700-5 98. Bjerketorp J, Håkansson S, Belkin S, Jansson JK (2006) Advances in preservation methods: keeping biosensor microorganisms alive and active. Curr Opin Biotechnol 17(1):43–49. https://doi.org/10.1016/j.copbio.2005.12.005 99. Boldt TS, Sørensen J, Karlson U, Molin S, Ramos C (2004) Combined use of different gfp reporters for monitoring single-cell activities of a genetically modified PCB degrader in the rhizosphere of alfalfa. FEMS Microbiol Ecol 48(2):139–148. https://doi.org/10.1016/j.femsec. 2004.01.002 100. Bolton EK, Sayler GS, Nivens DE, Rochelle JM, Ripp S, Simpson ML (2002) Integrated CMOS photodetectors and signal processing for very low-level chemical sensing with the bioluminescent bioreporter integrated circuit. Sensors Actuators B Chem 85(1–2):179–185. https://doi.org/10.1016/S0925-4005(02)00106-5 101. Bontidean I, Mortari A, Leth S, Brown NL, Karlson U, Larsen MM, . . . Csöregi E (2004) Biosensors for detection of mercury in contaminated soils. Environ Pollut 131(2):255–262. https://doi.org/10.1016/j.envpol.2004.02.019 102. Branchini BR, Ablamsky DM, Davis AL, Southworth TL, Butler B, Fan F, . . . Pule MA (2010) Red-emitting luciferases for bioluminescence reporter and imaging applications. Anal Biochem 396(2):290–297. https://doi.org/10.1016/j.ab.2009.09.009 103. Brandt KK, Holm PE, Nybroe O (2008) Evidence for bioavailable copper-dissolved organic matter complexes and transiently increased copper bioavailability in manure-amended soils as determined by bioluminescent bacterial biosensors. Environ Sci Technol 42(8):3102–3108. https://doi.org/10.1021/es071916+ 104. Bundy JG, Paton GI, Campbell CD (2004) Combined microbial community level and single species biosensor responses to monitor recovery of oil polluted soil. Soil Biol Biochem 36 (7):1149–1159. https://doi.org/10.1016/j.soilbio.2004.02.025 105. Burmølle M, Hansen LH, Oregaard G, Sørensen SJ (2003) Presence of N-acyl homoserine lactones in soil detected by a whole-cell biosensor and flow cytometry. Microb Ecol 45 (3):226–236. https://doi.org/10.1007/s00248-002-2028-6 106. Casavant NC, Thompson D, Beattie GA, Phillips GJ, Halverson LJ (2003) Use of a sitespecific recombination-based biosensor for detecting bioavailable toluene and related compounds on roots. Environ Microbiol 5(4):238–249. https://doi.org/10.1046/j.1462-2920.2003. 00420.x 107. Castellarnau M, Zine N, Bausells J, Madrid C, Juárez A, Samitier J, Errachid A (2007) Integrated cell positioning and cell-based ISFET biosensors. Sensors Actuators B Chem 120 (2):615–620. https://doi.org/10.1016/j.snb.2006.01.057 108. Chang IS, Jang JK, Gil GC, Kim M, Kim HJ, Cho BW, Kim BH (2004) Continuous determination of biochemical oxygen demand using microbial fuel cell type biosensor. Biosens Bioelectron 19(6):607–613. https://doi.org/10.1016/S0956-5663(03)00272-0

18

G. Thouand

109. Chang IS, Moon H, Jang JK, Kim BH (2005) Improvement of a microbial fuel cell performance as a BOD sensor using respiratory inhibitors. Biosens Bioelectron 20(9):1856–1859. https://doi.org/10.1016/j.bios.2004.06.003 110. Chen P, He C (2004) A general strategy to convert the MerR family proteins into highly sensitive and selective fluorescent biosensors for metal ions. J Am Chem Soc 126(3):728–729. https://doi.org/10.1021/ja0383975 111. Chen T, Small DA, McDermott MK, Bentley WE, Payne GF (2003) Enzymatic methods for in situ cell entrapment and cell release. Biomacromolecules 4(6):1558–1563. https://doi.org/10. 1021/bm034145k 112. Cheng Vollmer A, Van Dyk TK (2004) Stress responsive bacteria: biosensors as environmental monitors. https://doi.org/10.1016/S0065-2911(04)49003-1. www.scopus.com 113. Chouteau C, Dzyadevych S, Chovelon J, Durrieu C (2004) Development of novel conductometric biosensors based on immobilised whole cell Chlorella vulgaris microalgae. Biosens Bioelectron 19(9):1089–1096. https://doi.org/10.1016/j.bios.2003.10.012 114. Chu Y, Hsu C, Soma PK, Lo YM (2009) Immobilization of bioluminescent Escherichia coli cells using natural and artificial fibers treated with polyethyleneimine. Bioresour Technol 100 (13):3167–3174. https://doi.org/10.1016/j.biortech.2009.01.072 115. Close DM, Ripp S, Sayler GS (2009) Reporter proteins in whole-cell optical bioreporter detection systems, biosensor integrations, and biosensing applications. Sensors 9(11):9147– 9174. https://doi.org/10.3390/s91109147 116. Cohen T, Starosvetsky J, Cheruti U, Armon R (2010) Whole cell imprinting in sol-gel thin films for bacterial recognition in liquids: macromolecular fingerprinting. Int J Mol Sci 11 (4):1236–1252. https://doi.org/10.3390/ijms11041236 117. Darwent MJ, Paterson E, McDonald AJS, Tomos AD (2003) Biosensor reporting of root exudation from Hordeum vulgare in relation to shoot nitrate concentration. J Exp Bot 54 (381):325–334. https://doi.org/10.1093/jxb/erg017 118. Date A, Pasini P, Daunert S (2007) Construction of spores for portable bacterial whole-cell biosensing systems. Anal Chem 79(24):9391–9397. https://doi.org/10.1021/ac701606g 119. DeAngelis KM, Ji P, Firestone MK, Lindow SE (2005) Two novel bacterial biosensors for detection of nitrate availability in the rhizosphere. Appl Environ Microbiol 71(12):8537–8547. https://doi.org/10.1128/AEM.71.12.8537-8547.2005 120. Defraia CT, Schmelz EA, Mou Z (2008) A rapid biosensor-based method for quantification of free and glucose-conjugated salicylic acid. Plant Methods 4(1). https://doi.org/10.1186/1746-4811-4-28 121. Di Lorenzo M, Curtis TP, Head IM, Scott K (2009) A single-chamber microbial fuel cell as a biosensor for wastewaters. Water Res 43(13):3145–3154. https://doi.org/10.1016/j.watres. 2009.01.005 122. Diesel E, Schreiber M, Van Der Meer JR (2009) Development of bacteria-based bioassays for arsenic detection in natural waters. Anal Bioanal Chem 394(3):687–693. https://doi.org/10. 1007/s00216-009-2785-x 123. Ding L, Du D, Zhang X, Ju H (2008) Trends in cell-based electrochemical biosensors. Curr Med Chem 15(30):3160–3170. https://doi.org/10.2174/092986708786848514 124. Drepper T, Eggert T, Circolone F, Heck A, Krauß U, Guterl J, . . . Jaeger K (2007) Reporter proteins for in vivo fluorescence without oxygen. Nat Biotechnol 25(4):443–445. https://doi. org/10.1038/nbt1293 125. D’Souza SF (2001) Microbial biosensors. Biosens Bioelectron 16(6):337–353. https://doi.org/ 10.1016/S0956-5663(01)00125-7 126. Eltzov E, Marks RS, Voost S, Wullings BA, Heringa MB (2009) Flow-through real time bacterial biosensor for toxic compounds in water. Sensors Actuators B Chem 142(1):11–18. https://doi.org/10.1016/j.snb.2009.08.024 127. Everhart JL, McNear D Jr, Peltier E, van der Lelie D, Chaney RL, Sparks DL (2006) Assessing nickel bioavailability in smelter-contaminated soils. Sci Total Environ 367(2–3):732–744. https://doi.org/10.1016/j.scitotenv.2005.12.029

1

Introduction to Cell Biosensors Through 55 Years of Scientific Production

19

128. Fakhrullin RF, Shlykova LV, Zamaleeva AI, Nurgaliev DK, Osin YN, García-Alonso J, Paunov VN (2010) Interfacing living unicellular algae cells with biocompatible polyelectrolyte-stabilised magnetic nanoparticles. Macromol Biosci 10(10):1257–1264. https://doi.org/ 10.1002/mabi.201000161 129. Fang Y (2006) Label-free cell-based assays with optical biosensors in drug discovery. Assay Drug Dev Technol 4(5):583–595. https://doi.org/10.1089/adt.2006.4.583 130. Farré M, Barceló D (2003) Toxicity testing of wastewater and sewage sludge by biosensors, bioassays and chemical analysis. TrAC – Trends Anal Chem 22(5):299–310. https://doi.org/ 10.1016/S0165-9936(03)00504-1 131. Farré M, Pasini O, Carmen Alonso M, Castillo M, Barceló D (2001) Toxicity assessment of organic pollution in wastewaters using a bacterial biosensor. Anal Chim Acta 426(2):155–165. https://doi.org/10.1016/S0003-2670(00)00826-6 132. Fletcher MP, Diggle SP, Crusz SA, Chhabra SR, Cámara M, Williams P (2007) A dual biosensor for 2-alkyl-4-quinolone quorum-sensing signal molecules. Environ Microbiol 9 (11):2683–2693. https://doi.org/10.1111/j.1462-2920.2007.01380.x 133. Flynn HC, Meharg AA, Bowyer PK, Paton GI (2003) Antimony bioavailability in mine soils. Environ Pollut 124(1):93–100. https://doi.org/10.1016/S0269-7491(02)00411-6 134. Fujimoto H, Wakabayashi M, Yamashiro H, Maeda I, Isoda K, Kondoh M, . . . Yagi K (2006) Whole-cell arsenite biosensor using photosynthetic bacterium Rhodovulum sulfidophilum: Rhodovulum sulfidophilum as an arsenite biosensor. Appl Microbiol Biotechnol 73(2):332– 338. https://doi.org/10.1007/s00253-006-0483-6 135. Gammoudi I, Tarbague H, Othmane A, Moynet D, Rebière D, Kalfat R, Dejous C (2010) Love-wave bacteria-based sensor for the detection of heavy metal toxicity in liquid medium. Biosens Bioelectron 26(4):1723–1726. https://doi.org/10.1016/j.bios.2010.07.118 136. Garmendia J, De Las Heras A, Galvão TC, De Lorenzo V (2008) Tracing explosives in soil with transcriptional regulators of pseudomonas putida evolved for responding to nitrotoluenes. Microb Biotechnol 1(3):236–246. https://doi.org/10.1111/j.1751-7915.2008.00027.x 137. Gil G, Chang I, Kim BH, Kim M, Jang J, Park HS, Kim HJ (2003) Operational parameters affecting the performance of a mediator-less microbial fuel cell. Biosens Bioelectron 18 (4):327–334. https://doi.org/10.1016/S0956-5663(02)00110-0 138. Girotti S, Ferri EN, Fumo MG, Maiolini E (2008) Monitoring of environmental pollutants by bioluminescent bacteria. Anal Chim Acta 608(1):2–29. https://doi.org/10.1016/j.aca.2007. 12.008 139. Gou N, Onnis-Hayden A, Gu AZ (2010) Mechanistic toxicity assessment of nanomaterials by whole-cell-array stress genes expression analysis. Environ Sci Technol 44(15):5964–5970. https://doi.org/10.1021/es100679f 140. Gu MB, Mitchell RJ, Kim BC (2004) Whole-cell-based biosensors for environmental biomonitoring and application. Adv Biochem Eng Biotechnol 87:269–305. https://doi.org/ 10.1007/b13533 141. Gupta RK, Patterson SS, Ripp S, Simpson ML, Sayler GS (2003) Expression of the photorhabdus luminescence lux genes (luxA, B, C, D, and E) in Saccharomyces cerevisiae. FEMS Yeast Res 4(3):305–313. https://doi.org/10.1016/S1567-1356(03)00174-0 142. Hahn T, Tag K, Riedel K, Uhlig S, Baronian K, Gellissen G, Kunze G (2006) A novel estrogen sensor based on recombinant Arxula adeninivorans cells. Biosens Bioelectron 21(11):2078– 2085. https://doi.org/10.1016/j.bios.2005.10.019 143. Hakkila K, Green T, Leskinen P, Ivask A, Marks R, Virta M (2004) Detection of bioavailable heavy metals in EILATox-oregon samples using whole-cell luminescent bacterial sensors in suspension or immobilized onto fibre-optic tips. J Appl Toxicol 24(5):333–342. https://doi. org/10.1002/jat.1020 144. Hakkila K, Maksimow M, Karp M, Virta M (2002) Reporter genes lucFF, luxCDABE, gfp, and dsred have different characteristics in whole-cell bacterial sensors. Anal Biochem 301 (2):235–242. https://doi.org/10.1006/abio.2001.5517

20

G. Thouand

145. Han T, Sasaki S, Yano K, Ikebukuro K, Kitayama A, Nagamune T, Karube I (2002) Flow injection microbial trichloroethylene sensor. Talanta 57(2):271–276. https://doi.org/10.1016/ S0039-9140(02)00027-9 146. Hansen LH, Ferrari B, Sørensen AH, Veal D, Sørensen SJ (2001) Detection of oxytetracycline production by Streptomyces rimosus in soil microcosms by combining whole-cell biosensors and flow cytometry. Appl Environ Microbiol 67(1):239–244. https://doi.org/10.1128/AEM. 67.1.239-244.2001 147. Hansen LH, Sørensen SJ (2001) The use of whole-cell biosensors to detect and quantify compounds or conditions affecting biological systems. Microb Ecol 42(4):483–494. https:// doi.org/10.1007/s00248-001-0025-9 148. Hayashi S, Hazama A, Dutta AK, Sabirov RZ, Okada Y (2004) Detecting ATP release by a biosensor method. Sci STKE: Signal Transduct Knowl Environ 2004(258). https://doi.org/10. 1126/stke.2582004pl14 149. Hearn MTW, Acosta D (2001) Applications of novel affinity cassette methods: use of peptide fusion handles for the purification of recombinant proteins. J Mol Recognit 14(6):323–369. https://doi.org/10.1002/jmr.555 150. Heijerick DG, Janssen CR, Karlèn C, Odnevall Wallinder I, Leygraf C (2002) Bioavailability of zinc in runoff water from roofing materials. Chemosphere 47(10):1073–1080. https://doi. org/10.1016/S0045-6535(02)00014-0 151. Held M, Schuhmann W, Jahreis K, Schmidt H (2002) Microbial biosensor array with transport mutants of Escherichia coli K12 for the simultaneous determination of mono-and disaccharides. Biosens Bioelectron 17(11–12):1089–1094. https://doi.org/10.1016/S0956-5663(02) 00103-3 152. Hillson NJ, Hu P, Andersen GL, Shapiro L (2007) Caulobacter crescentus as a whole-cell uranium biosensor. Appl Environ Microbiol 73(23):7615–7621. https://doi.org/10.1128/AEM. 01566-07 153. Hondroulis E, Liu C, Li C (2010) Whole cell based electrical impedance sensing approach for a rapid nanotoxicity assay. Nanotechnology 21(31). https://doi.org/10.1088/0957-4484/21/31/ 315103 154. Huang S, Yang H, Lakshmanan RS, Johnson ML, Wan J, Chen I, . . . Chin BA (2009) Sequential detection of Salmonella typhimurium and Bacillus anthracis spores using magnetoelastic biosensors. Biosens Bioelectron 24(6):1730–1736. https://doi.org/10.1016/j. bios.2008.09.006 155. Huang WE, Huang L, Preston GM, Naylor M, Carr JP, Li Y, . . . Wang H (2006) Quantitative in situ assay of salicylic acid in tobacco leaves using a genetically modified biosensor strain of Acinetobacter sp. ADP1. Plant J 46(6):1073–1083. https://doi.org/10.1111/j.1365-313X.2006. 02758.x 156. Hynninen A, Tõnismann K, Virta M (2010) Improving the sensitivity of bacterial bioreporters for heavy metals. Bioeng Bugs 1(2):132–138. https://doi.org/10.4161/bbug.1.2.10902 157. Ivask A, François M, Kahru A, Dubourguier H, Virta M, Douay F (2004) Recombinant luminescent bacterial sensors for the measurement of bioavailability of cadmium and lead in soils polluted by metal smelters. Chemosphere 55(2):147–156. https://doi.org/10.1016/j. chemosphere.2003.10.064 158. Ivask A, Green T, Polyak B, Mor A, Kahru A, Virta M, Marks R (2007) Fibre-optic bacterial biosensors and their application for the analysis of bioavailable hg and as in soils and sediments from aznalcollar mining area in Spain. Biosens Bioelectron 22(7):1396–1402. https://doi.org/10.1016/j.bios.2006.06.019 159. Ivsak A, Rõlova T, Kahru A (2009) A suite of recombinant luminescent bacterial strains for the quantification of bioavailable heavy metals and toxicity testing. BMC Biotechnol 9. https:// doi.org/10.1186/1472-6750-9-41 160. Jansson JK (2003) Marker and reporter genes: illuminating tools for environmental microbiologists. Curr Opin Microbiol 6(3):310–316. https://doi.org/10.1016/S1369-5274(03)00057-2

1

Introduction to Cell Biosensors Through 55 Years of Scientific Production

21

161. Jha SK, Kanungo M, Nath A, D’Souza SF (2009) Entrapment of live microbial cells in electropolymerized polyaniline and their use as urea biosensor. Biosens Bioelectron 24 (8):2637–2642. https://doi.org/10.1016/j.bios.2009.01.024 162. Jia J, Tang M, Chen X, Qi L, Dong S (2003) Co-immobilized microbial biosensor for BOD estimation based on sol-gel derived composite material. Biosens Bioelectron 18(8):1023– 1029. https://doi.org/10.1016/S0956-5663(02)00225-7 163. Jiang Y, Xiao L, Zhao L, Chen X, Wang X, Wong K (2006) Optical biosensor for the determination of BOD in seawater. Talanta 70(1 Special issue):97–103. https://doi.org/10. 1016/j.talanta.2005.11.046 164. Jin HL, Mitchell RJ, Byoung CK, Cullen DC, Man BG (2005) A cell array biosensor for environmental toxicity analysis. Biosens Bioelectron 21(3):500–507. https://doi.org/10.1016/ j.bios.2004.12.015 165. Kahru A, Dubourguier H, Blinova I, Ivask A, Kasemets K (2008) Biotests and biosensors for ecotoxicology of metal oxide nanoparticles: a minireview. Sensors 8(8):5153–5170. https:// doi.org/10.3390/s8085153 166. Katrlík J, Voštiar I, Šefčovičová J, Tkáč J, Mastihuba V, Valach M, . . . Gemeiner P (2007) A novel microbial biosensor based on cells of Gluconobacter oxydans for the selective determination of 1,3-propanediol in the presence of glycerol and its application to bioprocess monitoring. Anal Bioanal Chem 388(1):287–295. https://doi.org/10.1007/s00216-007-1211-5 167. Kim BC, Gu MB (2003) A bioluminescent sensor for high throughput toxicity classification. Biosens Bioelectron 18(8):1015–1021. https://doi.org/10.1016/S0956-5663(02)00220-8 168. Kim BH, Chang IS, Gil GC, Park HS, Kim HJ (2003) Novel BOD (biological oxygen demand) sensor using mediator-less microbial fuel cell. Biotechnol Lett 25(7):541–545. https://doi.org/ 10.1023/A:1022891231369 169. Kuang Y, Biran I, Walt DR (2004) Living bacterial cell array for genotoxin monitoring. Anal Chem 76(10):2902–2909. https://doi.org/10.1021/ac0354589 170. Kumar J, D’Souza SF (2010) An optical microbial biosensor for detection of methyl parathion using Sphingomonas sp. immobilized on microplate as a reusable biocomponent. Biosens Bioelectron 26(4):1292–1296. https://doi.org/10.1016/j.bios.2010.07.016 171. Kumar J, Jha SK, D’Souza SF (2006) Optical microbial biosensor for detection of methyl parathion pesticide using Flavobacterium sp. whole cells adsorbed on glass fiber filters as disposable biocomponent. Biosens Bioelectron 21(11):2100–2105. https://doi.org/10.1016/j. bios.2005.10.012 172. Kumari A, Pasini P, Daunert S (2008) Detection of bacterial quorum sensing N-acyl homoserine lactones in clinical samples. Anal Bioanal Chem 391(5):1619–1627. https://doi.org/10. 1007/s00216-008-2002-3 173. Kumari A, Pasini P, Deo SK, Flomenhoft D, Shashidhar H, Daunert S (2006) Biosensing systems for the detection of bacterial quorum signaling molecules. Anal Chem 78(22):7603– 7609. https://doi.org/10.1021/ac061421n 174. Kumlanghan A, Kanatharana P, Asawatreratanakul P, Mattiasson B, Thavarungkul P (2008) Microbial BOD sensor for monitoring treatment of wastewater from a rubber latex industry. Enzym Microb Technol 42(6):483–491. https://doi.org/10.1016/j.enzmictec.2008. 01.012 175. Kumlanghan A, Liu J, Thavarungkul P, Kanatharana P, Mattiasson B (2007) Microbial fuel cell-based biosensor for fast analysis of biodegradable organic matter. Biosens Bioelectron 22 (12):2939–2944. https://doi.org/10.1016/j.bios.2006.12.014 176. Kwok N, Dong S, Lo W, Wong K (2005) An optical biosensor for multi-sample determination of biochemical oxygen demand (BOD). Sensors Actuators B Chem 110(2):289–298. https:// doi.org/10.1016/j.snb.2005.02.007 177. Lee JH, Gu MB (2005) An integrated mini biosensor system for continuous water toxicity monitoring. Biosens Bioelectron 20(9):1744–1749. https://doi.org/10.1016/j.bios.2004. 06.036

22

G. Thouand

178. Lee JH, Youn CH, Kim BC, Gu MB (2007) An oxidative stress-specific bacterial cell array chip for toxicity analysis. Biosens Bioelectron 22(9–10):2223–2229. https://doi.org/10.1016/j. bios.2006.10.038 179. Lee SA, Choi Y, Jung S, Kim S (2002) Effect of initial carbon sources on the electrochemical detection of glucose by gluconobacter oxydans. Bioelectrochemistry 57(2):173–178. https:// doi.org/10.1016/S1567-5394(02)00115-9 180. Leedjärv A, Ivask A, Virta M, Kahru A (2006) Analysis of bioavailable phenols from natural samples by recombinant luminescent bacterial sensors. Chemosphere 64(11):1910–1919. https://doi.org/10.1016/j.chemosphere.2006.01.026 181. Lei Y, Chen W, Mulchandani A (2006) Microbial biosensors. Anal Chim Acta 568(1–2):200– 210. https://doi.org/10.1016/j.aca.2005.11.065 182. Lei Y, Mulchandani P, Wang J, Chen W, Mulchandani A (2005) Highly sensitive and selective amperometric microbial biosensor for direct determination of p-nitrophenyl-substituted organophosphate nerve agents. Environ Sci Technol 39(22):8853–8857. https://doi.org/10.1021/ es050720b 183. Li S, Li Y, Chen H, Horikawa S, Shen W, Simonian A, Chin BA (2010) Direct detection of Salmonella typhimurium on fresh produce using phage-based magnetoelastic biosensors. Biosens Bioelectron 26(4):1313–1319. https://doi.org/10.1016/j.bios.2010.07.029 184. Li Y, Li F, Ho C, Liao VH (2008) Construction and comparison of fluorescence and bioluminescence bacterial biosensors for the detection of bioavailable toluene and related compounds. Environ Pollut 152(1):123–129. https://doi.org/10.1016/j.envpol.2007.05.002 185. Liao VH, Chien M, Tseng Y, Ou K (2006) Assessment of heavy metal bioavailability in contaminated sediments and soils using green fluorescent protein-based bacterial biosensors. Environ Pollut 142(1):17–23. https://doi.org/10.1016/j.envpol.2005.09.021 186. Liao VH, Ou K (2005) Development and testing of a green fluorescent protein-based bacterial biosensor for measuring bioavailable arsenic in contaminated groundwater samples. Environ Toxicol Chem 24(7):1624–1631. https://doi.org/10.1897/04-500R.1 187. Livage J, Coradin T, Roux C (2001) Encapsulation of biomolecules in silica gels. J Phys Condens Matter 13(33):R673–R691. https://doi.org/10.1088/0953-8984/13/33/202 188. Lorenzelli L, Margesin B, Martinoia S, Tedesco MT, Valle M (2003) Bioelectrochemical signal monitoring of in-vitro cultured cells by means of an automated microsystem based on solid state sensor-array. Biosens Bioelectron 18(5–6):621–626. https://doi.org/10.1016/ S0956-5663(03)00040-X 189. Martín-González A, Díaz S, Borniquel S, Gallego A, Gutiérrez JC (2006) Cytotoxicity and bioaccumulation of heavy metals by ciliated protozoa isolated from urban wastewater treatment plants. Res Microbiol 157(2):108–118. https://doi.org/10.1016/j.resmic.2005.06.005 190. Marx KA, Zhou T, Montrone A, McIntosh D, Braunhut SJ (2007) A comparative study of the cytoskeleton binding drugs nocodazole and taxol with a mammalian cell quartz crystal microbalance biosensor: different dynamic responses and energy dissipation effects. Anal Biochem 361(1):77–92. https://doi.org/10.1016/j.ab.2006.09.023 191. Matsui N, Kaya T, Nagamine K, Yasukawa T, Shiku H, Matsue T (2006) Electrochemical mutagen screening using microbial chip. Biosens Bioelectron 21(7):1202–1209. https://doi. org/10.1016/j.bios.2005.05.004 192. May KML, Wang Y, Bachas LG, Anderson KW (2004) Development of a whole-cell-based biosensor for detecting histamine as a model toxin. Anal Chem 76(14):4156–4161. https://doi. org/10.1021/ac049810+ 193. Mejri MB, Baccar H, Baldrich E, Del Campo FJ, Helali S, Ktari T, . . . Abdelghani A (2010) Impedance biosensing using phages for bacteria detection: generation of dual signals as the clue for in-chip assay confirmation. Biosens Bioelectron 26(4):1261–1267. https://doi.org/10. 1016/j.bios.2010.06.054 194. Miller WG, Brandl MT, Quiñones B, Lindow SE (2001) Biological sensor for sucrose availability: relative sensitivities of various reporter genes. Appl Environ Microbiol 67 (3):1308–1317. https://doi.org/10.1128/AEM.67.3.1308-1317.2001

1

Introduction to Cell Biosensors Through 55 Years of Scientific Production

23

195. Mitchell RJ, Gu MB (2004) An Escherichia coli biosensor capable of detecting both genotoxic and oxidative damage. Appl Microbiol Biotechnol 64(1):46–52. https://doi.org/10.1007/ s00253-003-1418-0 196. Moon H, Chang IS, Jang JK, Kim KS, Lee J, Lovitt RW, Kim BH (2005) On-line monitoring of low biochemical oxygen demand through continuous operation of a mediator-less microbial fuel cell. J Microbiol Biotechnol 15(1):192–196. www.scopus.com 197. Moreno-Garrido I (2008) Microalgae immobilization: current techniques and uses. Bioresour Technol 99(10):3949–3964. https://doi.org/10.1016/j.biortech.2007.05.040 198. Mosier-Boss PA, Lieberman SH, Andrews JM, Rohwer FL, Wegley LE, Breitbart M (2003) Use of fluorescently labeled phage in the detection and identification of bacterial species. Appl Spectrosc 57(9):1138–1144. https://doi.org/10.1366/00037020360696008 199. Mulchandani P, Chen W, Mulchandani A (2006) Microbial biosensor for direct determination of nitrophenyl-substituted organophosphate nerve agents using genetically engineered Moraxella sp. Anal Chim Acta 568(1–2):217–221. https://doi.org/10.1016/j.aca.2005.11.063 200. Mulchandani P, Chen W, Mulchandani A, Wang J, Chen L (2001) Amperometric microbial biosensor for direct determination of organophosphate pesticides using recombinant microorganism with surface expressed organophosphorus hydrolase. Biosens Bioelectron 16(7– 8):433–437. https://doi.org/10.1016/S0956-5663(01)00157-9 201. Mulchandani P, Hangarter CM, Lei Y, Chen W, Mulchandani A (2005) Amperometric microbial biosensor for p-nitrophenol using Moraxella sp.-modified carbon paste electrode. Biosens Bioelectron 21(3):523–527. https://doi.org/10.1016/j.bios.2004.11.011 202. Nakamura H, Suzuki K, Ishikuro H, Kinoshita S, Koizumi R, Okuma S, . . . Karube I (2007) A new BOD estimation method employing a double-mediator system by ferricyanide and menadione using the eukaryote Saccharomyces cerevisiae. Talanta 72(1):210–216. https:// doi.org/10.1016/j.talanta.2006.10.019 203. Nanduri V, Balasubramanian S, Sista S, Vodyanoy VJ, Simonian AL (2007) Highly sensitive phage-based biosensor for the detection of β-galactosidase. Anal Chim Acta 589(2):166–172. https://doi.org/10.1016/j.aca.2007.02.071 204. Nanduri V, Bhunia AK, Tu S, Paoli GC, Brewster JD (2007) SPR biosensor for the detection of L. monocytogenes using phage-displayed antibody. Biosens Bioelectron 23(2):248–252. https://doi.org/10.1016/j.bios.2007.04.007 205. Nassif N, Bouvet O, Rager MN, Roux C, Coradin T, Livage J (2002) Living bacteria in silica gels. Nat Mater 1(1):42–44. https://doi.org/10.1038/nmat709 206. Natarajan A, Molnar P, Sieverdes K, Jamshidi A, Hickman JJ (2006) Microelectrode array recordings of cardiac action potentials as a high throughput method to evaluate pesticide toxicity. Toxicol In Vitro 20(3):375–381. https://doi.org/10.1016/j.tiv.2005.08.014 207. Nguyen-Ngoc H, Tran-Minh C (2007) Fluorescent biosensor using whole cells in an inorganic translucent matrix. Anal Chim Acta 583(1):161–165. https://doi.org/10.1016/j.aca.2006.10. 005 208. Nielsen M, Larsen LH, Jetten MSM, Revsbech NP (2004) Bacterium-based NO2- biosensor for environmental applications. Appl Environ Microbiol 70(11):6551–6558. https://doi.org/ 10.1128/AEM.70.11.6551-6558.2004 209. Norman A, Hansen LH, Sørensen SJ (2005) Construction of a ColD cda promoter-based SOSgreen fluorescent protein whole-cell biosensor with higher sensitivity toward genotoxic compounds than constructs based on recA, umuDC, or sulA promoters. Appl Environ Microbiol 71(5):2338–2346. https://doi.org/10.1128/AEM.71.5.2338-2346.2005 210. Odaci D, Kiralp Kayahan S, Timur S, Toppare L (2008) Use of a thiophene-based conducting polymer in microbial biosensing. Electrochim Acta 53(12):4104–4108. https://doi.org/10. 1016/j.electacta.2007.12.065 211. Olsen EV, Sorokulova IB, Petrenko VA, Chen I, Barbaree JM, Vodyanoy VJ (2006) Affinityselected filamentous bacteriophage as a probe for acoustic wave biodetectors of Salmonella typhimurium. Biosens Bioelectron 21(8):1434–1442. https://doi.org/10.1016/j.bios.2005.06. 004

24

G. Thouand

212. Ong KG, Wang J, Singh RS, Bachas LG, Grimes CA (2001) Monitoring of bacteria growth using a wireless, remote query resonant-circuit sensor: application to environmental sensing. Biosens Bioelectron 16(4–5):305–312. https://doi.org/10.1016/S0956-5663(01)00131-2 213. Paggi RA, Martone CB, Fuqua C, De Castro RE (2003) Detection of quorum sensing signals in the haloalkaliphilic archaeon Natronococcus occultus. FEMS Microbiol Lett 221(1):49–52. https://doi.org/10.1016/S0378-1097(03)00174-5 214. Paitan Y, Biran I, Shechter N, Biran D, Rishpon J, Ron EZ (2004) Monitoring aromatic hydrocarbons by whole cell electrochemical biosensors. Anal Biochem 335(2):175–183. https://doi.org/10.1016/j.ab.2004.08.032 215. Patil S, Harnisch F, Schröder U (2010) Toxicity response of electroactive microbial biofilms – a decisive feature for potential biosensor and power source applications. ChemPhysChem 11 (13):2834–2837. https://doi.org/10.1002/cphc.201000218 216. Paton GI, Viventsova E, Kumpene J, Wilson MJ, Weitz HJ, Dawson JJC (2006) An ecotoxicity assessment of contaminated forest soils from the kola peninsula. Sci Total Environ 355(1– 3):106–117. https://doi.org/10.1016/j.scitotenv.2005.04.036 217. Pellegrini GE, Carpico G, Coni E (2004) Electrochemical sensor for the detection and presumptive identification of quinolone and tetracycline residues in milk. Anal Chim Acta 520(1–2):13–18. https://doi.org/10.1016/j.aca.2004.04.052 218. Pellinen T, Bylund G, Virta M, Niemi A, Karp M (2002) Detection of traces of tetracyclines from fish with a bioluminescent sensor strain incorporating bacterial luciferase reporter genes. J Agric Food Chem 50(17):4812–4815. https://doi.org/10.1021/jf020402l 219. Petrenko VA (2008) Landscape phage as a molecular recognition interface for detection devices. Microelectron J 39(2):202–207. https://doi.org/10.1016/j.mejo.2006.11.007 220. Petrenko VA, Vodyanoy VJ (2003) Phage display for detection of biological threat agents. J Microbiol Methods 53(2):253–262. https://doi.org/10.1016/S0167-7012(03)00029-0 221. Pfleger BF, Pitera DJ, Newman JD, Martin VJJ, Keasling JD (2007) Microbial sensors for small molecules: development of a mevalonate biosensor. Metab Eng 9(1):30–38. https://doi. org/10.1016/j.ymben.2006.08.002 222. Philp JC, Balmand S, Hajto E, Bailey MJ, Wiles S, Whiteley AS, . . . Dunbar SA (2003) Whole cell immobilised biosensors for toxicity assessment of a wastewater treatment plant treating phenolics-containing waste. Anal Chim Acta 487(1):61–74. https://doi.org/10.1016/S00032670(03)00358-1 223. Phoenix P, Keane A, Patel A, Bergeron H, Ghoshal S, Lau PCK (2003) Characterization of a new solvent-responsive gene locus in Pseudomonas putida F1 and its functionalization as a versatile biosensor. Environ Microbiol 5(12):1309–1327. https://doi.org/10.1111/j.1462-2920. 2003.00426.x 224. Popovtzer R, Neufeld T, Biran D, Ron EZ, Rishpon J, Shacham-Diamand Y (2005) Novel integrated electrochemical nano-biochip for toxicity detection in water. Nano Lett 5(6):1023– 1027. https://doi.org/10.1021/nl0503227 225. Power M, Hosticka B, Black E, Daitch C, Norris P (2001) Aerogels as biosensors: viral particle detection by bacteria immobilized on large pore aerogel. J Non-Cryst Solids 285(1– 3):303–308. https://doi.org/10.1016/S0022-3093(01)00471-9 226. Premkumar JR, Rosen R, Belkin S, Lev O (2002) Sol-gel luminescence biosensors: encapsulation of recombinant E. coli reporters in thick silicate films. Anal Chim Acta 462(1):11–23. https://doi.org/10.1016/S0003-2670(02)00301-X 227. Rastogi S, Kumar A, Mehra NK, Makhijani SD, Manoharan A, Gangal V, Kumar R (2003) Development and characterization of a novel immobilized microbial membrane for rapid determination of biochemical oxygen demand load in industrial waste-waters. Biosens Bioelectron 18(1):23–29. https://doi.org/10.1016/S0956-5663(02)00108-2 228. Ren D, Zuo R, Barrios AFG, Bedzyk LA, Eldridge GR, Pasmore ME, Wood TK (2005) Differential gene expression for investigation of Escherichia coli biofilm inhibition by plant extract ursolic acid. Appl Environ Microbiol 71(7):4022–4034. https://doi.org/10.1128/AEM. 71.7.4022-4034.2005

1

Introduction to Cell Biosensors Through 55 Years of Scientific Production

25

229. Ren S, Frymier PD (2005) Toxicity of metals and organic chemicals evaluated with bioluminescence assays. Chemosphere 58(5):543–550. https://doi.org/10.1016/j.chemosphere.2004. 07.005 230. Rensing C, Maier RM (2003) Issues underlying use of biosensors to measure metal bioavailability. Ecotoxicol Environ Saf 56(1):140–147. https://doi.org/10.1016/S0147-6513(03) 00057-5 231. Riether KB, Dollard M, Billard P (2001) Assessment of heavy metal bioavailability using Escherichia coli zntAp::Lux and copAp::Lux-based biosensors. Appl Microbiol Biotechnol 57(5–6):712–716. https://doi.org/10.1007/s00253-001-0852-0 232. Ripp S, Jegier P, Johnson CM, Brigati JR, Sayler GS (2008) Bacteriophage-amplified bioluminescent sensing of Escherichia coli O157:H7. Anal Bioanal Chem 391(2):507–514. https:// doi.org/10.1007/s00216-007-1812-z 233. Roberto FF, Barnes JM, Bruhn DF (2002) Evaluation of a GFP reporter gene construct for environmental arsenic detection. Talanta 58(1):181–188. https://doi.org/10.1016/S0039-9140 (02)00266-7 234. Roda A, Guardigli M, Michelini E, Mirasoli M (2009) Bioluminescence in analytical chemistry and in vivo imaging. TrAC – Trends Anal Chem 28(3):307–322. https://doi.org/10.1016/ j.trac.2008.11.015 235. Roda A, Pasini P, Mirasoli M, Michelini E, Guardigli M (2004) Biotechnological applications of bioluminescence and chemiluminescence. Trends Biotechnol 22(6):295–303. https://doi. org/10.1016/j.tibtech.2004.03.011 236. Ron EZ (2007) Biosensing environmental pollution. Curr Opin Biotechnol 18(3):252–256. https://doi.org/10.1016/j.copbio.2007.05.005 237. Sakaguchi T, Morioka Y, Yamasaki M, Iwanaga J, Beppu K, Maeda H, . . . Tamiya E (2007) Rapid and onsite BOD sensing system using luminous bacterial cells-immobilized chip. Biosens Bioelectron 22(7):1345–1350. https://doi.org/10.1016/j.bios.2006.06.008 238. Salins LLE, Deo SK, Daunert S (2004) Phosphate binding protein as the biorecognition element in a biosensor for phosphate. Sensors Actuators B Chem 97(1):81–89. https://doi. org/10.1016/j.snb.2003.07.019 239. Sanders CA, Rodriguez M Jr, Greenbaum E (2001) Stand-off tissue-based biosensors for the detection of chemical warfare agents using photosynthetic fluorescence induction. Biosens Bioelectron 16(7–8):439–446. https://doi.org/10.1016/S0956-5663(01)00158-0 240. Sanseverino J, Eldridge ML, Layton AC, Easter JP, Yarbrough J, Schultz TW, Sayler GS (2009) Screening of potentially hormonally active chemicals using bioluminescent yeast bioreporters. Toxicol Sci 107(1):122–134. https://doi.org/10.1093/toxsci/kfn229 241. Scott CW, Peters MF (2010) Label-free whole-cell assays: expanding the scope of GPCR screening. Drug Discov Today 15(17–18):704–716. https://doi.org/10.1016/j.drudis.2010.06. 008 242. Shapiro E, Lu C, Baneyx F (2005) A set of multicolored Photinus pyralis luciferase mutants for in vivo bioluminescence applications. Protein Eng Des Sel 18(12):581–587. https://doi. org/10.1093/protein/gzi066 243. Shetty RS, Deo SK, Liu Y, Daunert S (2004) Fluorescence-based sensing system for copper using genetically engineered living yeast cells. Biotechnol Bioeng 88(5):664–670. https://doi. org/10.1002/bit.20331 244. Shi H, Wen Su W (2001) Display of green fluorescent protein on Escherichia coli cell surface. Enzym Microb Technol 28(1):25–34. https://doi.org/10.1016/S0141-0229(00)00281-7 245. Shitanda I, Takamatsu S, Watanabe K, Itagaki M (2009) Amperometric screen-printed algal biosensor with flow injection analysis system for detection of environmental toxic compounds. Electrochim Acta 54(21):4933–4936. https://doi.org/10.1016/j.electacta.2009.04.005 246. Simpson ML, Sayler GS, Fleming JT, Applegate B (2001) Whole-cell biocomputing. Trends Biotechnol 19(8):317–323. https://doi.org/10.1016/S0167-7799(01)01691-2 247. Simpson ML, Sayler GS, Patterson G, Nivens DE, Bolton EK, Rochelle JM, . . . Guillorn MA (2001) Integrated CMOS microluminometer for low-level luminescence sensing in the

26

G. Thouand

bioluminescent bioreporter integrated circuit. Sensors Actuators B Chem 72(2):134–140. https://doi.org/10.1016/S0925-4005(00)00641-9 248. Singh A, Arya SK, Glass N, Hanifi-Moghaddam P, Naidoo R, Szymanski CM, . . . Evoy S (2010) Bacteriophage tailspike proteins as molecular probes for sensitive and selective bacterial detection. Biosens Bioelectron 26(1):131–138. https://doi.org/10.1016/j.bios.2010.05.024 249. Singh A, Glass N, Tolba M, Brovko L, Griffiths M, Evoy S (2009) Immobilization of bacteriophages on gold surfaces for the specific capture of pathogens. Biosens Bioelectron 24(12):3645–3651. https://doi.org/10.1016/j.bios.2009.05.028 250. Smirnova IA, Dian C, Leonard GA, McSweeney S, Birse D, Brzezinski P (2004) Development of a bacterial biosensor for nitrotoluenes: the crystal structure of the transcriptional regulator DntR. J Mol Biol 340(3):405–418. https://doi.org/10.1016/j.jmb.2004.04.071 251. Song Y, Li G, Thornton SF, Thompson IP, Banwart SA, Lerner DN, Huang WE (2009) Optimization of bacterial whole cell bioreporters for toxicity assay of environmental samples. Environ Sci Technol 43(20):7931–7938. https://doi.org/10.1021/es901349r 252. Sørensen SJ, Burmølle M, Hansen LH (2006) Making bio-sense of toxicity: new developments in whole-cell biosensors. Curr Opin Biotechnol 17(1):11–16. https://doi.org/10.1016/j. copbio.2005.12.007 253. Stein NE, Hamelers HVM, Buisman CNJ (2010) Stabilizing the baseline current of a microbial fuel cell-based biosensor through overpotential control under non-toxic conditions. Bioelectrochemistry 78(1):87–91. https://doi.org/10.1016/j.bioelechem.2009.09.009 254. Stiner L, Halverson LJ (2002) Development and characterization of a green fluorescent protein-based bacterial biosensor for bioavailable toluene and related compounds. Appl Environ Microbiol 68(4):1962–1971. https://doi.org/10.1128/AEM.68.4.1962-1971.2002 255. Stocker J, Balluch D, Gsell M, Harms H, Feliciano J, Daunert S, . . . Van der Meer JR (2003) Development of a set of simple bacterial biosensors for quantitative and rapid measurements of arsenite and arsenate in potable water. Environ Sci Technol 37(20):4743–4750. https://doi.org/ 10.1021/es034258b 256. Strachan G, Preston S, Maciel H, Porter AJR, Paton GI (2001) Use of bacterial biosensors to interpret the toxicity and mixture toxicity of herbicides in freshwater. Water Res 35(14):3490– 3495. https://doi.org/10.1016/S0043-1354(01)00065-3 257. Struss A, Pasini P, Ensor CM, Raut N, Daunert S (2010) Paper strip whole cell biosensors: a portable test for the semiquantitative detection of bacterial quorum signaling molecules. Anal Chem 82(11):4457–4463. https://doi.org/10.1021/ac100231a 258. Tag K, Riedel K, Bauer H, Hanke G, Baronian KHR, Kunze G (2007) Amperometric detection of Cu2+ by yeast biosensors using flow injection analysis (FIA). Sensors Actuators B Chem 122(2):403–409. https://doi.org/10.1016/j.snb.2006.06.007 259. Tani H, Maehana K, Kamidate T (2004) Chip-based bioassay using bacterial sensor strains immobilized in three-dimensional microfluidic network. Anal Chem 76(22):6693–6697. https://doi.org/10.1021/ac049401d 260. Taranova L, Semenchuk I, Manolov T, Iliasov P, Reshetilov A (2002) Bacteria-degraders as the base of an amperometric biosensor for detection of anionic surfactants. Biosens Bioelectron 17(8):635–640. https://doi.org/10.1016/S0956-5663(01)00307-4 261. Tecon R, Van Der Meer JR (2008) Bacterial biosensors for measuring availability of environmental pollutants. Sensors 8(7):4062–4080. https://doi.org/10.3390/s8074062 262. Temple TN, Stockwell VO, Loper JE, Johnson KB (2004) Bioavailability of iron to Pseudomonas fluorescens strain A506 on flowers of pear and apple. Phytopathology 94(12):1286– 1294. https://doi.org/10.1094/PHYTO.2004.94.12.1286 263. Tibazarwa C, Corbisier P, Mench M, Bossus A, Solda P, Mergeay M, . . . Van der Lelie D (2001) A microbial biosensor to predict bioavailable nickel in soil and its transfer to plants. Environ Pollut 113(1):19–26. https://doi.org/10.1016/S0269-7491(00)00177-9 264. Timur S, Anik U, Odaci D, Gorton L (2007) Development of a microbial biosensor based on carbon nanotube (CNT) modified electrodes. Electrochem Commun 9(7):1810–1815. https:// doi.org/10.1016/j.elecom.2007.04.012

1

Introduction to Cell Biosensors Through 55 Years of Scientific Production

27

265. Timur S, Haghighi B, Tkac J, Pazarlioǧlu N, Telefoncu A, Gorton L (2007b) Electrical wiring of Pseudomonas putida and Pseudomonas fluorescens with osmium redox polymers. Bioelectrochemistry 71(1):38–45. https://doi.org/10.1016/j.bioelechem.2006.08.001 266. Timur S, Pazarlioǧlu N, Pilloton R, Telefoncu A (2003) Detection of phenolic compounds by thick film sensors based on Pseudomonas putida. Talanta 61(2):87–93. https://doi.org/10. 1016/S0039-9140(03)00237-6 267. Tizzard AC, Bergsma JH, Lloyd-Jones G (2006) A resazurin-based biosensor for organic pollutants. Biosens Bioelectron 22(5):759–763. https://doi.org/10.1016/j.bios.2006.01.011 268. Tkac J, Vostiar I, Gorton L, Gemeiner P, Sturdik E (2003) Improved selectivity of microbial biosensor using membrane coating. Application to the analysis of ethanol during fermentation. Biosens Bioelectron 18(9):1125–1134. https://doi.org/10.1016/S0956-5663(02) 00244-0 269. Tom-Petersen A, Hosbond C, Nybroe O (2001) Identification of copper-induced genes in Pseudomonas fluorescens and use of a reporter strain to monitor bioavailable copper in soil. FEMS Microbiol Ecol 38(1):59–67. https://doi.org/10.1016/S0168-6496(01)00174-X 270. Tront JM, Fortner JD, Plötze M, Hughes JB, Puzrin AM (2008) Microbial fuel cell biosensor for in situ assessment of microbial activity. Biosens Bioelectron 24(4):586–590. https://doi. org/10.1016/j.bios.2008.06.006 271. Trosok SP, Driscoll BT, Luong JHT (2001) Mediated microbial biosensor using a novel yeast strain for wastewater BOD measurement. Appl Microbiol Biotechnol 56(3–4):550–554. https://doi.org/10.1007/s002530100674 272. Tuncagil S, Odaci D, Yildiz E, Timur S, Toppare L (2009) Design of a microbial sensor using conducting polymer of 4-(2,5-di(thiophen-2-yl)-1H-pyrrole-1-l) benzenamine. Sensors Actuators B Chem 137(1):42–47. https://doi.org/10.1016/j.snb.2008.10.067 273. Turner K, Xu S, Pasini P, Deo S, Bachas L, Daunert S (2007) Hydroxylated polychlorinated biphenyl detection based on a genetically engineered bioluminescent whole-cell sensing system. Anal Chem 79(15):5740–5745. https://doi.org/10.1021/ac0705162 274. Urban A, Eckermann S, Fast B, Metzger S, Gehling M, Ziegelbauer K, . . . Freiberg C (2007) Novel whole-cell antibiotic biosensors for compound discovery. Appl Environ Microbiol 73 (20):6436–6443. https://doi.org/10.1128/AEM.00586-07 275. Van Der Meer JR, Belkin S (2010) Where microbiology meets microengineering: design and applications of reporter bacteria. Nat Rev Microbiol 8(7):511–522. https://doi.org/10.1038/ nrmicro2392 276. Védrine C, Leclerc J, Durrieu C, Tran-Minh C (2003) Optical whole-cell biosensor using Chlorella vulgaris designed for monitoring herbicides. Biosens Bioelectron 18(4):457–463. https://doi.org/10.1016/S0956-5663(02)00157-4 277. Verma N, Singh M (2003) A disposable microbial based biosensor for quality control in milk. Biosens Bioelectron 18(10):1219–1224. https://doi.org/10.1016/S0956-5663(03)00085-X 278. Vijayaraghavan R, Islam SK, Zhang M, Ripp S, Caylor S, Bull ND, . . . Sayler GS (2007) A bioreporter bioluminescent integrated circuit for very low-level chemical sensing in both gas and liquid environments. Sensors Actuators B Chem 123(2):922–928. https://doi.org/10.1016/ j.snb.2006.10.064 279. Virolainen NE, Pikkemaat MG, Elferink JWA, Karp MT (2008) Rapid detection of tetracyclines and their 4-epimer derivatives from poultry meat with bioluminescent biosensor bacteria. J Agric Food Chem 56(23):11065–11070. https://doi.org/10.1021/jf801797z 280. Wanekaya AK, Chen W, Mulchandani A (2008) Recent biosensing developments in environmental security. J Environ Monit 10(6):703–712. https://doi.org/10.1039/b806830p 281. Wang H, Wang XJ, Zhao JF, Chen L (2008) Toxicity assessment of heavy metals and organic compounds using CellSense biosensor with E. coli. Chin Chem Lett 19(2):211–214. https:// doi.org/10.1016/j.cclet.2007.10.053 282. Weitz HJ, Campbell CD, Killham K (2002) Development of a novel, bioluminescence-based, fungal bioassay for toxicity testing. Environ Microbiol 4(7):422–429. https://doi.org/10.1046/ j.1462-2920.2002.00315.x

28

G. Thouand

283. Werlen C, Jaspers MCM, Van Der Meer JR (2004) Measurement of biologically available naphthalene in gas and aqueous phases by use of a pseudomonas putida biosensor. Appl Environ Microbiol 70(1):43–51. https://doi.org/10.1128/AEM.70.1.43-51.2004 284. Xu T, Petridou S, Lee EH, Roth EA, Vyavahare NR, Hickman JJ, Boland T (2004) Construction of high-density bacterial colony arrays and patterns by the ink-jet method. Biotechnol Bioeng 85(1):29–33. https://doi.org/10.1002/bit.10768 285. Yagur-Kroll S, Bilic B, Belkin S (2010) Strategies for enhancing bioluminescent bacterial sensor performance by promoter region manipulation. Microb Biotechnol 3(3):300–310. https://doi.org/10.1111/j.1751-7915.2009.00149.x 286. Yemini M, Levi Y, Yagil E, Rishpon J (2007) Specific electrochemical phage sensing for Bacillus cereus and Mycobacterium smegmatis. Bioelectrochemistry 70(1):180–184. https:// doi.org/10.1016/j.bioelechem.2006.03.014 287. Yi C, Li C, Ji S, Yang M (2006) Microfluidics technology for manipulation and analysis of biological cells. Anal Chim Acta 560(1–2):1–23. https://doi.org/10.1016/j.aca.2005.12.037 288. Yu D, Volponi J, Chhabra S, Brinker CJ, Mulchandani A, Singh AK (2005) Aqueous sol-gel encapsulation of genetically engineered Moraxella spp. cells for the detection of organophosphates. Biosens Bioelectron 20(7):1433–1437. https://doi.org/10.1016/j.bios.2004.04.022

2011–2020 289. Abrevaya XC, Sacco NJ, Bonetto MC, Hilding-Ohlsson A, Cortón E (2015) Analytical applications of microbial fuel cells. Part I: Biochemical oxygen demand. Biosens Bioelectron 63:580–590. https://doi.org/10.1016/j.bios.2014.04.034 290. Abrevaya XC, Sacco NJ, Bonetto MC, Hilding-Ohlsson A, Cortón E (2015) Analytical applications of microbial fuel cells. Part II: Toxicity, microbial activity and quantification, single analyte detection and other uses. Biosens Bioelectron 63:591–601. https://doi.org/10. 1016/j.bios.2014.04.053 291. Alcaine SD, Law K, Ho S, Kinchla AJ, Sela DA, Nugen SR (2016) Bioengineering bacteriophages to enhance the sensitivity of phage amplification-based paper fluidic detection of bacteria. Biosens Bioelectron 82:14–19. https://doi.org/10.1016/j.bios.2016.03.047 292. Alcaine SD, Tilton L, Serrano MAC, Wang M, Vachet RW, Nugen SR (2015) Phage-proteasepeptide: a novel trifecta enabling multiplex detection of viable bacterial pathogens. Appl Microbiol Biotechnol 99(19):8177–8185. https://doi.org/10.1007/s00253-015-6867-8 293. Amaro F, Turkewitz AP, Martín-González A, Gutiérrez, J. (2011) Whole-cell biosensors for detection of heavy metal ions in environmental samples based on metallothionein promoters from Tetrahymena thermophila. Microb Biotechnol 4(4):513–522. https://doi.org/10.1111/j. 1751-7915.2011.00252.x 294. Anu Prathap MU, Chaurasia AK, Sawant SN, Apte SK (2012) Polyaniline-based highly sensitive microbial biosensor for selective detection of lindane. Anal Chem 84(15):6672– 6678. https://doi.org/10.1021/ac301077d 295. Arlyapov VA, Yudina NY, Asulyan LD, Alferov SV, Alferov VA, Reshetilov AN (2013) BOD biosensor based on the yeast Debaryomyces hansenii immobilized in poly(vinyl alcohol) modified by N-vinylpyrrolidone. Enzym Microb Technol 53(4):257–262. https://doi.org/10. 1016/j.enzmictec.2013.05.004 296. Aslan S, Anik Ü (2016) Microbial glucose biosensors based on glassy carbon paste electrodes modified with Gluconobacter oxydans and graphene oxide or graphene-platinum hybrid nanoparticles. Microchim Acta 183(1):73–81. https://doi.org/10.1007/s00604-015-1590-9 297. Axelrod T, Eltzov E, Marks RS (2016) Bioluminescent bioreporter pad biosensor for monitoring water toxicity. Talanta 149:290–297. https://doi.org/10.1016/j.talanta.2015.11.067

1

Introduction to Cell Biosensors Through 55 Years of Scientific Production

29

298. Ayyaru S, Dharmalingam S (2014) Enhanced response of microbial fuel cell using sulfonated poly ether ether ketone membrane as a biochemical oxygen demand sensor. Anal Chim Acta 818:15–22. https://doi.org/10.1016/j.aca.2014.01.059 299. Bárdy P, Pantůček R, Benešík M, Doškař J (2016) Genetically modified bacteriophages in applied microbiology. J Appl Microbiol 121(3):618–633. https://doi.org/10.1111/jam.13207 300. Bayram E, Akyilmaz E (2016) Development of a new microbial biosensor based on conductive polymer/multiwalled carbon nanotube and its application to paracetamol determination. Sensors Actuators B Chem 233:409–418. https://doi.org/10.1016/j.snb.2016.04.029 301. Behzadian F, Barjeste H, Hosseinkhani S, Zarei AR (2011) Construction and characterization of Escherichia coli whole-cell biosensors for toluene and related compounds. Curr Microbiol 62(2):690–696. https://doi.org/10.1007/s00284-010-9764-5 302. Ben-Yoav H, Melamed S, Freeman A, Shacham-Diamand Y, Belkin S (2011) Whole-cell biochips for bio-sensing: integration of live cells and inanimate surfaces. Crit Rev Biotechnol 31(4):337–353. https://doi.org/10.3109/07388551.2010.532767 303. Bereza-Malcolm L, Aracic S, Franks AE (2016) Development and application of a synthetically-derived lead biosensor construct for use in gram-negative bacteria. Sensors (Switzerland) 16(12). https://doi.org/10.3390/s16122174 304. Bereza-Malcolm LT, Mann G, Franks AE (2015) Environmental sensing of heavy metals through whole cell microbial biosensors: a synthetic biology approach. ACS Synth Biol 4 (5):535–546. https://doi.org/10.1021/sb500286r 305. Bertels F, Merker H, Kost C (2012) Design and characterization of auxotrophy-based amino acid biosensors. PLoS One 7(7). https://doi.org/10.1371/journal.pone.0041349 306. Bhardwaj N, Bhardwaj SK, Mehta J, Mohanta GC, Deep A (2016) Bacteriophage immobilized graphene electrodes for impedimetric sensing of bacteria (Staphylococcus arlettae). Anal Biochem 505:18–25. https://doi.org/10.1016/j.ab.2016.04.008 307. Bhardwaj N, Bhardwaj SK, Mehta J, Nayak MK, Deep A (2016) Bacteriophage conjugated IRMOF-3 as a novel opto-sensor for: S. arlettae. New J Chem 40(9):8068–8073. https://doi. org/10.1039/c6nj00899b 308. Biran A, Ben Yoav H, Yagur-Kroll S, Pedahzur R, Buchinger S, Shacham-Diamand Y, . . . Belkin S (2011) Microbial genotoxicity bioreporters based on sulA activation. Anal Bioanal Chem 400(9):3013–3024. https://doi.org/10.1007/s00216-011-5007-2 309. Bownik A (2017) Daphnia swimming behaviour as a biomarker in toxicity assessment: a review. Sci Total Environ 601–602:194–205. https://doi.org/10.1016/j.scitotenv.2017.05.199 310. Branco R, Cristóvão A, Morais PV (2013) Highly sensitive, highly specific whole-cell bioreporters for the detection of chromate in environmental samples. PLoS One 8(1). https:// doi.org/10.1371/journal.pone.0054005 311. Brzozowska E, Śmietana M, Koba M, Górska S, Pawlik K, Gamian A, Bock WJ (2015) Recognition of bacterial lipopolysaccharide using bacteriophage-adhesin-coated long-period gratings. Biosens Bioelectron 67:93–99. https://doi.org/10.1016/j.bios.2014.07.027 312. Buffi N, Merulla D, Beutier J, Barbaud F, Beggah S, Van Lintel H, . . . Roelof Van Der Meer J (2011) Development of a microfluidics biosensor for agarose-bead immobilized Escherichia coli bioreporter cells for arsenite detection in aqueous samples. Lab Chip 11(14):2369–2377. https://doi.org/10.1039/c1lc20274j 313. Cerminati S, Soncini FC, Checa SK (2011) Selective detection of gold using genetically engineered bacterial reporters. Biotechnol Bioeng 108(11):2553–2560. https://doi.org/10. 1002/bit.23213 314. Cevenini L, Calabretta MM, Tarantino G, Michelini E, Roda A (2016) Smartphone-interfaced 3D printed toxicity biosensor integrating bioluminescent “sentinel cells”. Sensors Actuators B Chem 225:249–257. https://doi.org/10.1016/j.snb.2015.11.017 315. Chang H, Voyvodic PL, Zúñiga A, Bonnet J (2017) Microbially derived biosensors for diagnosis, monitoring and epidemiology. Microb Biotechnol 10(5):1031–1035. https://doi. org/10.1111/1751-7915.12791

30

G. Thouand

316. Charrier T, Chapeau C, Bendria L, Picart P, Daniel P, Thouand G (2011) A multi-channel bioluminescent bacterial biosensor for the on-line detection of metals and toxicity. Part II: Technical development and proof of concept of the biosensor. Anal Bioanal Chem 400 (4):1061–1070. https://doi.org/10.1007/s00216-010-4354-8 317. Charrier T, Durand M, Jouanneau S, Dion M, Pernetti M, Poncelet D, Thouand G (2011) A multi-channel bioluminescent bacterial biosensor for the on-line detection of metals and toxicity. Part I: Design and optimization of bioluminescent bacterial strains. Anal Bioanal Chem 400(4):1051–1060. https://doi.org/10.1007/s00216-010-4353-9 318. Chen J, Sun S, Li C, Zhu Y, Rosen BP (2014) Biosensor for organoarsenical herbicides and growth promoters. Environ Sci Technol 48(2):1141–1147. https://doi.org/10.1021/es4038319 319. Chen Z, Niu Y, Zhao S, Khan A, Ling Z, Chen Y, . . . Li X (2016) A novel biosensor for pnitrophenol based on an aerobic anode microbial fuel cell. Biosens Bioelectron 85:860–868. https://doi.org/10.1016/j.bios.2016.06.007 320. Chong H, Ching CB (2016) Development of colorimetric-based whole-cell biosensor for organophosphorus compounds by engineering transcription regulator DmpR. ACS Synth Biol 5(11):1290–1298. https://doi.org/10.1021/acssynbio.6b00061 321. Chouler J, Bentley I, Vaz F, O'Fee A, Cameron PJ, Di Lorenzo M (2017) Exploring the use of cost-effective membrane materials for microbial fuel cell based sensors. Electrochim Acta 231:319–326. https://doi.org/10.1016/j.electacta.2017.01.195 322. Chouler J, Cruz-Izquierdo Á, Rengaraj S, Scott JL, Di Lorenzo M (2018) A screen-printed paper microbial fuel cell biosensor for detection of toxic compounds in water. Biosens Bioelectron 102:49–56. https://doi.org/10.1016/j.bios.2017.11.018 323. Chouler J, Di Lorenzo M (2015) Water quality monitoring in developing countries; can microbial fuel cells be the answer? Biosensors 5(3):450–470. https://doi.org/10.3390/ bios5030450 324. Close D, Xu T, Smartt A, Rogers A, Crossley R, Price S, . . . Sayler G (2012) The evolution of the bacterial luciferase gene cassette (lux) as a real-time bioreporter. Sensors 12(1):732–752. https://doi.org/10.3390/s120100732 325. Corbella C, Hartl M, Fernandez-gatell M, Puigagut J (2019) MFC-based biosensor for domestic wastewater COD assessment in constructed wetlands. Sci Total Environ 660:218– 226. https://doi.org/10.1016/j.scitotenv.2018.12.347 326. Dávila D, Esquivel JP, Sabaté N, Mas J (2011) Silicon-based microfabricated microbial fuel cell toxicity sensor. Biosens Bioelectron 26(5):2426–2430. https://doi.org/10.1016/j.bios. 2010.10.025 327. Di Lorenzo M, Thomson AR, Schneider K, Cameron PJ, Ieropoulos I (2014) A small-scale aircathode microbial fuel cell for on-line monitoring of water quality. Biosens Bioelectron 62:182–188. https://doi.org/10.1016/j.bios.2014.06.050 328. Do MH, Ngo HH, Guo W, Chang SW, Nguyen DD, Liu Y, . . . Kumar M (2020) Microbial fuel cell-based biosensor for online monitoring wastewater quality: a critical review. Sci Total Environ 712. https://doi.org/10.1016/j.scitotenv.2019.135612 329. Dopson M, Ni G, Sleutels THJA (2016) Possibilities for extremophilic microorganisms in microbial electrochemical systems. FEMS Microbiol Rev 40(2):164–181. https://doi.org/10. 1093/femsre/fuv044 330. Ejeian F, Etedali P, Mansouri-Tehrani H, Soozanipour A, Low Z, Asadnia M, . . . Razmjou A (2018) Biosensors for wastewater monitoring: a review. Biosens Bioelectron 118:66–79. https://doi.org/10.1016/j.bios.2018.07.019 331. Elad T, Almog R, Yagur-Kroll S, Levkov K, Melamed S, Shacham-Diamand Y, Belkin S (2011) Online monitoring of water toxicity by use of bioluminescent reporter bacterial biochips. Environ Sci Technol 45(19):8536–8542. https://doi.org/10.1021/es202465c 332. ElMekawy A, Hegab HM, Pant D, Saint CP (2018) Bio-analytical applications of microbial fuel cell–based biosensors for onsite water quality monitoring. J Appl Microbiol 124(1):302– 313. https://doi.org/10.1111/jam.13631

1

Introduction to Cell Biosensors Through 55 Years of Scientific Production

31

333. Eltzov E, Marks RS (2011) Whole-cell aquatic biosensors. Anal Bioanal Chem 400(4):895– 913. https://doi.org/10.1007/s00216-010-4084-y 334. Eltzov E, Pavluchkov V, Burstin M, Marks RS (2011) Creation of a fiber optic based biosensor for air toxicity monitoring. Sensors Actuators B Chem 155(2):859–867. https://doi.org/10. 1016/j.snb.2011.01.062 335. Farooq U, Yang Q, Ullah MW, Wang S (2018) Bacterial biosensing: recent advances in phagebased bioassays and biosensors. Biosens Bioelectron 118:204–216. https://doi.org/10.1016/j. bios.2018.07.058 336. Feng Y, Barr W, Harper WF (2013) Neural network processing of microbial fuel cell signals for the identification of chemicals present in water. J Environ Manag 120:84–92. https://doi. org/10.1016/j.jenvman.2013.01.018 337. Feng Y, Kayode O, Harper WF Jr (2013) Using microbial fuel cell output metrics and nonlinear modeling techniques for smart biosensing. Sci Total Environ 449:223–228. https:// doi.org/10.1016/j.scitotenv.2013.01.004 338. Fernandes E, Martins VC, Nóbrega C, Carvalho CM, Cardoso FA, Cardoso S, . . . Azeredo J (2014) A bacteriophage detection tool for viability assessment of Salmonella cells. Biosens Bioelectron 52:239–246. https://doi.org/10.1016/j.bios.2013.08.053 339. Fernández M, Morel B, Ramos JL, Krell T (2016) Paralogous regulators ArsR1 and ArsR2 of Pseudomonas putida KT2440 as a basis for arsenic biosensor development. Appl Environ Microbiol 82(14):4133–4144. https://doi.org/10.1128/AEM.00606-16 340. Gao G, Fang D, Yu Y, Wu L, Wang Y, Zhi J (2017) A double-mediator based whole cell electrochemical biosensor for acute biotoxicity assessment of wastewater. Talanta 167:208– 216. https://doi.org/10.1016/j.talanta.2017.01.081 341. Gao G, Qian J, Fang D, Yu Y, Zhi J (2016) Development of a mediated whole cell-based electrochemical biosensor for joint toxicity assessment of multi-pollutants using a mixed microbial consortium. Anal Chim Acta 924:21–28. https://doi.org/10.1016/j.aca.2016.04.011 342. Gireesh-Babu P, Chaudhari A (2012) Development of a broad-spectrum fluorescent heavy metal bacterial biosensor. Mol Biol Rep 39(12):11225–11229. https://doi.org/10.1007/s11033012-2033-x 343. Golitsch F, Bücking C, Gescher J (2013) Proof of principle for an engineered microbial biosensor based on Shewanella oneidensis outer membrane protein complexes. Biosens Bioelectron 47:285–291. https://doi.org/10.1016/j.bios.2013.03.010 344. Grattieri M, Minteer SD (2018) Self-powered biosensors. ACS Sensors 3(1):44–53. https:// doi.org/10.1021/acssensors.7b00818 345. Gui Q, Lawson T, Shan S, Yan L, Liu Y (2017) The application of whole cell-based biosensors for use in environmental analysis and in medical diagnostics. Sensors (Switzerland) 17(7). https://doi.org/10.3390/s17071623 346. Gupta N, Renugopalakrishnan V, Liepmann D, Paulmurugan R, Malhotra BD (2019) Cellbased biosensors: recent trends, challenges and future perspectives. Biosens Bioelectron 141. https://doi.org/10.1016/j.bios.2019.111435 347. Gupta S, Saxena M, Saini N, Mahmooduzzafar, Kumar R, Kumar A (2012) An effective strategy for a whole-cell biosensor based on putative effector interaction site of the regulatory dmpr protein. PLoS One 7(8). https://doi.org/10.1371/journal.pone.0043527 348. Gutiérrez JC, Amaro F, Martín-González A (2015) Heavy metal whole-cell biosensors using eukaryotic microorganisms: an updated critical review. Front Microbiol 6. https://doi.org/10. 3389/fmicb.2015.00048 349. Hassan SHA, Van Ginkel SW, Hussein MAM, Abskharon R, Oh, S. (2016) Toxicity assessment using different bioassays and microbial biosensors. Environ Int 92–93:106–118. https:// doi.org/10.1016/j.envint.2016.03.003 350. He W, Yuan S, Zhong W, Siddikee MA, Dai C (2016) Application of genetically engineered microbial whole-cell biosensors for combined chemosensing. Appl Microbiol Biotechnol 100 (3):1109–1119. https://doi.org/10.1007/s00253-015-7160-6

32

G. Thouand

351. Hillger JM, Schoop J, Boomsma DI, Eline Slagboom P, IJzerman AP, Heitman LH (2015) Whole-cell biosensor for label-free detection of GPCR-mediated drug responses in personal cell lines. Biosens Bioelectron 74:233–242. https://doi.org/10.1016/j.bios.2015.06. 031 352. Hnaien M, Lagarde F, Bausells J, Errachid A, Jaffrezic-Renault N (2011) A new bacterial biosensor for trichloroethylene detection based on a three-dimensional carbon nanotubes bioarchitecture. Anal Bioanal Chem 400(4):1083–1092. https://doi.org/10.1007/s00216-0104336-x 353. Hosseinidoust Z, Olsson ALJ, Tufenkji N (2014) Going viral: designing bioactive surfaces with bacteriophage. Colloids Surf B: Biointerfaces 124:2–16. https://doi.org/10.1016/j. colsurfb.2014.05.036 354. Hsieh M, Cheng C, Liu M, Chung Y (2015) Effects of operating parameters on measurements of biochemical oxygen demand using a mediatorless microbial fuel cell biosensor. Sensors (Switzerland) 16(1). https://doi.org/10.3390/s16010035 355. Hsieh M, Chung Y (2014) Measurement of biochemical oxygen demand from different wastewater samples using a mediator-less microbial fuel cell biosensor. Environ Technol (UK) 35(17):2204–2211. https://doi.org/10.1080/09593330.2014.898700 356. Huang C, Wei C, Liao VHC (2015) A low cost color-based bacterial biosensor for measuring arsenic in groundwater. Chemosphere 141:44–49. https://doi.org/10.1016/j.chemosphere. 2015.06.011 357. Hung YP, Yellen G (2014) Live-cell imaging of cytosolic NADH-NAD+ redox state using a genetically encoded fluorescent biosensor. https://doi.org/10.1007/978-1-62703-622-1_7. www.scopus.com 358. Husu I, Rodio G, Touloupakis E, Lambreva MD, Buonasera K, Litescu SC, . . . Rea G (2013) Insights into photo-electrochemical sensing of herbicides driven by Chlamydomonas reinhardtii cells. Sensors Actuators B Chem 185:321–330. https://doi.org/10.1016/j.snb. 2013.05.013 359. Idil N, Hedström M, Denizli A, Mattiasson B (2017) Whole cell based microcontact imprinted capacitive biosensor for the detection of Escherichia coli. Biosens Bioelectron 87:807–815. https://doi.org/10.1016/j.bios.2016.08.096 360. Jasim A, Ullah MW, Shi Z, Lin X, Yang G (2017) Fabrication of bacterial cellulose/polyaniline/single-walled carbon nanotubes membrane for potential application as biosensor. Carbohydr Polym 163:62–69. https://doi.org/10.1016/j.carbpol.2017.01.056 361. Jia H, Yang G, Wang J, Ngo HH, Guo W, Zhang H, Zhang X (2016) Performance of a microbial fuel cell-based biosensor for online monitoring in an integrated system combining microbial fuel cell and upflow anaerobic sludge bed reactor. Bioresour Technol 218:286–293. https://doi.org/10.1016/j.biortech.2016.06.064 362. Jiang Y, Liang P, Huang X, Ren ZJ (2018) A novel microbial fuel cell sensor with a gas diffusion biocathode sensing element for water and air quality monitoring. Chemosphere 203:21–25. https://doi.org/10.1016/j.chemosphere.2018.03.169 363. Jiang Y, Liang P, Liu P, Bian Y, Miao B, Sun X, . . . Huang X (2016) Enhancing signal output and avoiding BOD/toxicity combined shock interference by operating a microbial fuel cell sensor with an optimized background concentration of organic matter. Int J Mol Sci 17(9). https://doi.org/10.3390/ijms17091392 364. Jiang Y, Liang P, Liu P, Wang D, Miao B, Huang X (2017) A novel microbial fuel cell sensor with biocathode sensing element. Biosens Bioelectron 94:344–350. https://doi.org/10.1016/j. bios.2017.02.052 365. Jiang Y, Liang P, Zhang C, Bian Y, Yang X, Huang X, Girguis PR (2015) Enhancing the response of microbial fuel cell based toxicity sensors to cu(II) with the applying of flowthrough electrodes and controlled anode potentials. Bioresour Technol 190:367–372. https:// doi.org/10.1016/j.biortech.2015.04.127 366. Joshi N, Ngwenya BT, Butler IB, French CE (2015) Use of bioreporters and deletion mutants reveals ionic silver and ROS to be equally important in silver nanotoxicity. J Hazard Mater 287:51–58. https://doi.org/10.1016/j.jhazmat.2014.12.066

1

Introduction to Cell Biosensors Through 55 Years of Scientific Production

33

367. Jouanneau S, Durand MJ, Thouand G (2012) Online detection of metals in environmental samples: comparing two concepts of bioluminescent bacterial biosensors. Environ Sci Technol 46(21):11979–11987. https://doi.org/10.1021/es3024918 368. Jouanneau S, Recoules L, Durand MJ, Boukabache A, Picot V, Primault Y, . . . Thouand G (2014) Methods for assessing biochemical oxygen demand (BOD): a review. Water Res 49 (1):62–82. https://doi.org/10.1016/j.watres.2013.10.066 369. Käkinen A, Bondarenko O, Ivask A, Kahru A (2011) The effect of composition of different ecotoxicological test media on free and bioavailable copper from CuSO4 and CuO nanoparticles: comparative evidence from a Cu-selective electrode and a Cu-biosensor. Sensors 11 (11):10502–10521. https://doi.org/10.3390/s111110502 370. Kang Y, Lee W, Kim S, Jang G, Kim B, Yoon Y (2018) Enhancing the copper-sensing capability of Escherichia coli-based whole-cell bioreporters by genetic engineering. Appl Microbiol Biotechnol 102(3):1513–1521. https://doi.org/10.1007/s00253-017-8677-7 371. Kaur A, Ibrahim S, Pickett CJ, Michie IS, Dinsdale RM, Guwy AJ, Premier GC (2014) Anode modification to improve the performance of a microbial fuel cell volatile fatty acid biosensor. Sensors Actuators B Chem 201:266–273. https://doi.org/10.1016/j.snb.2014.04.062 372. Kaur A, Kim JR, Michie I, Dinsdale RM, Guwy AJ, Premier GC (2013) Microbial fuel cell type biosensor for specific volatile fatty acids using acclimated bacterial communities. Biosens Bioelectron 47:50–55. https://doi.org/10.1016/j.bios.2013.02.033 373. Kaur H, Kumar R, Babu JN, Mittal S (2015) Advances in arsenic biosensor development – a comprehensive review. Biosens Bioelectron 63:533–545. https://doi.org/10.1016/j.bios.2014. 08.003 374. Kim HJ, Jeong H, Lee SJ (2018) Synthetic biology for microbial heavy metal biosensors. Anal Bioanal Chem 410(4):1191–1203. https://doi.org/10.1007/s00216-017-0751-6 375. Kim HJ, Lim JW, Jeong H, Lee S, Lee D, Kim T, Lee SJ (2016) Development of a highly specific and sensitive cadmium and lead microbial biosensor using synthetic CadC-T7 genetic circuitry. Biosens Bioelectron 79:701–708. https://doi.org/10.1016/j.bios.2015.12.101 376. Kim M, Lim JW, Kim HJ, Lee SK, Lee SJ, Kim T (2015) Chemostat-like microfluidic platform for highly sensitive detection of heavy metal ions using microbial biosensors. Biosens Bioelectron 65:257–264. https://doi.org/10.1016/j.bios.2014.10.028 377. Kim SK, Kim SH, Subhadra B, Woo S, Rha E, Kim S, . . . Lee S (2018) A genetically encoded biosensor for monitoring isoprene production in engineered Escherichia coli. ACS Synth Biol 7(10):2379–2390. https://doi.org/10.1021/acssynbio.8b00164 378. Kim T, Lee D, Choi J (2017) Live cell biosensing platforms using graphene-based hybrid nanomaterials. Biosens Bioelectron 94:485–499. https://doi.org/10.1016/j.bios.2017.03.032 379. Knecht LD, Pasini P, Daunert S (2011) Bacterial spores as platforms for bioanalytical and biomedical applications. Anal Bioanal Chem 400(4):977–989. https://doi.org/10.1007/ s00216-011-4835-4 380. Kokkali V, van Delft W (2014) Overview of commercially available bioassays for assessing chemical toxicity in aqueous samples. TrAC – Trends Anal Chem 61:133–155. https://doi.org/ 10.1016/j.trac.2014.08.001 381. Kretzschmar J, Rosa LFM, Zosel J, Mertig M, Liebetrau J, Harnisch F (2016) A microbial biosensor platform for inline quantification of acetate in anaerobic digestion: potential and challenges. Chem Eng Technol 39(4):637–642. https://doi.org/10.1002/ceat.201500406 382. Kumar J, D’Souza SF (2011) Immobilization of microbial cells on inner epidermis of onion bulb scale for biosensor application. Biosens Bioelectron 26(11):4399–4404. https://doi.org/ 10.1016/j.bios.2011.04.049 383. Kumar J, D’Souza SF (2011) Microbial biosensor for detection of methyl parathion using screen printed carbon electrode and cyclic voltammetry. Biosens Bioelectron 26(11):4289– 4293. https://doi.org/10.1016/j.bios.2011.04.027 384. Kuncova G, Pazlarova J, Hlavata A, Ripp S, Sayler GS (2011) Bioluminescent bioreporter Pseudomonas putida TVA8 as a detector of water pollution. Operational conditions and selectivity of free cells sensor. Ecol Indic 11(3):882–887. https://doi.org/10.1016/j.ecolind. 2010.12.001

34

G. Thouand

385. Lagarde F, Jaffrezic-Renault N (2011) Cell-based electrochemical biosensors for water quality assessment. Anal Bioanal Chem 400(4):947–964. https://doi.org/10.1007/s00216-011-4816-7 386. Lee J, Song J, Hwang MP, Lee KH (2013) Nanoscale bacteriophage biosensors beyond phage display. Int J Nanomedicine 8:3917–3925. https://doi.org/10.2147/IJN.S51894 387. Li L, Liang J, Hong W, Zhao Y, Sun S, Yang X, . . . Chen S (2015) Evolved bacterial biosensor for arsenite detection in environmental water. Environ Sci Technol 49(10):6149–6155. https:// doi.org/10.1021/acs.est.5b00832 388. Li P, Müller M, Chang MW, Frettlöh M, Schönherr H (2017) Encapsulation of autoinducer sensing reporter bacteria in reinforced alginate-based microbeads. ACS Appl Mater Interfaces 9(27):22321–22331. https://doi.org/10.1021/acsami.7b07166 389. Li S, Si T, Wang M, Zhao H (2015) Development of a synthetic malonyl-CoA sensor in Saccharomyces cerevisiae for intracellular metabolite monitoring and genetic screening. ACS Synth Biol 4(12):1308–1315. https://doi.org/10.1021/acssynbio.5b00069 390. Liang B, Lang Q, Tang X, Liu A (2013a) Simultaneously improving stability and specificity of cell surface displayed glucose dehydrogenase mutants to construct whole-cell biocatalyst for glucose biosensor application. Bioresour Technol 147:492–498. https://doi.org/10.1016/j. biortech.2013.08.088 391. Liang B, Li L, Tang X, Lang Q, Wang H, Li F, . . . Liu A (2013b) Microbial surface display of glucose dehydrogenase for amperometric glucose biosensor. Biosens Bioelectron 45(1):9–24. https://doi.org/10.1016/j.bios.2013.01.050 392. Libis V, Delépine B, Faulon J (2016) Sensing new chemicals with bacterial transcription factors. Curr Opin Microbiol 33:105–112. https://doi.org/10.1016/j.mib.2016.07.006 393. Lim JW, Ha D, Lee J, Lee SK, Kim T (2015) Review of micro/nanotechnologies for microbial biosensors. Front Bioeng Biotechnol 3. https://doi.org/10.3389/fbioe.2015.00061 394. Liu B, Lei Y, Li B (2014) A batch-mode cube microbial fuel cell based “shock” biosensor for wastewater quality monitoring. Biosens Bioelectron 62:308–314. https://doi.org/10.1016/j. bios.2014.06.051 395. Liu C, Ma C, Yu D, Jia J, Liu L, Zhang B, Dong S (2011) Immobilized multi-species based biosensor for rapid biochemical oxygen demand measurement. Biosens Bioelectron 26 (5):2074–2079. https://doi.org/10.1016/j.bios.2010.09.004 396. Liu C, Yong D, Yu D, Dong S (2011) Cell-based biosensor for measurement of phenol and nitrophenols toxicity. Talanta 84(3):766–770. https://doi.org/10.1016/j.talanta.2011.02.006 397. Liu P, Huang Q, Chen W (2012) Construction and application of a zinc-specific biosensor for assessing the immobilization and bioavailability of zinc in different soils. Environ Pollut 164:66–72. https://doi.org/10.1016/j.envpol.2012.01.023 398. Liu R, Yang C, Xu Y, Xu P, Jiang H, Qiao C (2013) Development of a whole-cell biocatalyst/ biosensor by display of multiple heterologous proteins on the Escherichia coli cell surface for the detoxification and detection of organophosphates. J Agric Food Chem 61(32):7810–7816. https://doi.org/10.1021/jf402999b 399. Liu Z, Liu J, Zhang S, Xing X, Su Z (2011) Microbial fuel cell based biosensor for in situ monitoring of anaerobic digestion process. Bioresour Technol 102(22):10221–10229. https:// doi.org/10.1016/j.biortech.2011.08.053 400. Maderova L, Watson M, Paton GI (2011) Bioavailability and toxicity of copper in soils: integrating chemical approaches with responses of microbial biosensors. Soil Biol Biochem 43 (6):1162–1168. https://doi.org/10.1016/j.soilbio.2011.02.004 401. Mahr R, Frunzke J (2016) Transcription factor-based biosensors in biotechnology: current state and future prospects. Appl Microbiol Biotechnol 100(1):79–90. https://doi.org/10.1007/ s00253-015-7090-3 402. Mahr R, von Boeselager RF, Wiechert J, Frunzke J (2016) Screening of an Escherichia coli promoter library for a phenylalanine biosensor. Appl Microbiol Biotechnol 100(15):6739– 6753. https://doi.org/10.1007/s00253-016-7575-8 403. Mallevre F, Fernandes TF, Aspray TJ (2014) Silver, zinc oxide and titanium dioxide nanoparticle ecotoxicity to bioluminescent Pseudomonas putida in laboratory medium and artificial

1

Introduction to Cell Biosensors Through 55 Years of Scientific Production

35

wastewater. Environ Pollut (Barking, Essex: 1987) 195:218–225. https://doi.org/10.1016/j. envpol.2014.09.002 404. Mehta J, Bhardwaj SK, Bhardwaj N, Paul AK, Kumar P, Kim K, Deep A (2016) Progress in the biosensing techniques for trace-level heavy metals. Biotechnol Adv 34(1):47–60. https:// doi.org/10.1016/j.biotechadv.2015.12.001 405. Melamed S, Elad T, Belkin S (2012) Microbial sensor cell arrays. Curr Opin Biotechnol 23 (1):2–8. https://doi.org/10.1016/j.copbio.2011.11.024 406. Merulla D, Buffi N, Beggah S, Truffer F, Geiser M, Renaud P, Van der Meer JR (2013) Bioreporters and biosensors for arsenic detection. Biotechnological solutions for a world-wide pollution problem. Curr Opin Biotechnol 24(3):534–541. https://doi.org/10.1016/j.copbio. 2012.09.002 407. Michelini E, Cevenini L, Calabretta MM, Spinozzi S, Camborata C, Roda A (2013) Field-deployable whole-cell bioluminescent biosensors: so near and yet so far optical nanosensing in cells. Anal Bioanal Chem 405(19):6155–6163. https://doi.org/10.1007/s00216013-7043-6 408. Michelini E, Roda A (2012) Staying alive: new perspectives on cell immobilization for biosensing purposes. Anal Bioanal Chem 402(5):1785–1797. https://doi.org/10.1007/ s00216-011-5364-x 409. Mollon LC, Norton GJ, Trakal L, Moreno-Jimenez E, Elouali FZ, Hough RL, Beesley L (2016) Mobility and toxicity of heavy metal(loid)s arising from contaminated wood ash application to a pasture grassland soil. Environ Pollut 218:419–427. https://doi.org/10.1016/ j.envpol.2016.07.021 410. Mulchandani A, Rajesh (2011) Microbial biosensors for organophosphate pesticides. Appl Biochem Biotechnol 165(2):687–699. https://doi.org/10.1007/s12010-011-9288-x 411. Mustafi N, Grünberger A, Kohlheyer D, Bott M, Frunzke J (2012) The development and application of a single-cell biosensor for the detection of l-methionine and branched-chain amino acids. Metab Eng 14(4):449–457. https://doi.org/10.1016/j.ymben.2012.02.002 412. Nitisoravut R, Regmi R (2017) Plant microbial fuel cells: a promising biosystems engineering. Renew Sust Energ Rev 76:81–89. https://doi.org/10.1016/j.rser.2017.03.064 413. Olaniran AO, Hiralal L, Pillay B (2011) Whole-cell bacterial biosensors for rapid and effective monitoring of heavy metals and inorganic pollutants in wastewater. J Environ Monit 13 (10):2914–2920. https://doi.org/10.1039/c1em10032g 414. Olsson ALJ, Wargenau A, Tufenkji N (2016) Optimizing bacteriophage surface densities for bacterial capture and sensing in quartz crystal microbalance with dissipation monitoring. ACS Appl Mater Interfaces 8(22):13698–13706. https://doi.org/10.1021/acsami.6b02227 415. Park M, Tsai S, Chen W (2013) Microbial biosensors: engineered microorganisms as the sensing machinery. Sensors (Switzerland) 13(5):5777–5795. https://doi.org/10.3390/ s130505777 416. Pasco NF, Weld RJ, Hay JM, Gooneratne R (2011) Development and applications of whole cell biosensors for ecotoxicity testing. Anal Bioanal Chem 400(4):931–945. https://doi.org/10. 1007/s00216-011-4663-6 417. Pasternak G, Greenman J, Ieropoulos I (2017) Self-powered, autonomous biological oxygen demand biosensor for online water quality monitoring. Sensors Actuators B Chem 244:815– 822. https://doi.org/10.1016/j.snb.2017.01.019 418. Peixoto L, Min B, Martins G, Brito AG, Kroff P, Parpot P, . . . Nogueira R (2011) In situ microbial fuel cell-based biosensor for organic carbon. Bioelectrochemistry 81(2):99–103. https://doi.org/10.1016/j.bioelechem.2011.02.002 419. Pini F, East AK, Appia-Ayme C, Tomek J, Karunakaran R, Mendoza-Suárez M, . . . Poole PS (2017) Bacterial biosensors for in vivo spatiotemporal mapping of root secretion. Plant Physiol 174(3):1289–1306. https://doi.org/10.1104/pp.16.01302 420. Ponamoreva ON, Kamanina OA, Alferov VA, Machulin AV, Rogova TV, Arlyapov VA, . . . Ivanova EP (2015) Yeast-based self-organized hybrid bio-silica sol-gels for the design of biosensors. Biosens Bioelectron 67:321–326. https://doi.org/10.1016/j.bios.2014.08.045

36

G. Thouand

421. Ponomareva ON, Arlyapov VA, Alferov VA, Reshetilov AN (2011) Microbial biosensors for detection of biological oxygen demand (a review). Appl Biochem Microbiol 47(1):1–11. https://doi.org/10.1134/S0003683811010108 422. Priyadarshi H, Alam A, Gireesh-Babu P, Das R, Kishore P, Kumar S, Chaudhari A (2012) A GFP-based bacterial biosensor with chromosomally integrated sensing cassette for quantitative detection of hg(II) in environment. J Environ Sci 24(5):963–968. https://doi.org/10.1016/ S1001-0742(11)60820-6 423. Quek SB, Cheng L, Cord-Ruwisch R (2015) Microbial fuel cell biosensor for rapid assessment of assimilable organic carbon under marine conditions. Water Res 77:64–71. https://doi.org/ 10.1016/j.watres.2015.03.012 424. Raud M, Tenno T, Jõgi E, Kikas T (2012) Comparative study of semi-specific Aeromonas hydrophila and universal Pseudomonas fluorescens biosensors for BOD measurements in meat industry wastewaters. Enzym Microb Technol 50(4–5):221–226. https://doi.org/10. 1016/j.enzmictec.2012.01.003 425. Ravikumar S, Baylon MG, Park SJ, Choi J (2017) Engineered microbial biosensors based on bacterial two-component systems as synthetic biotechnology platforms in bioremediation and biorefinery. Microb Cell Factories 16(1). https://doi.org/10.1186/s12934-017-0675-z 426. Ravikumar S, Ganesh I, Yoo I, Hong SH (2012) Construction of a bacterial biosensor for zinc and copper and its application to the development of multifunctional heavy metal adsorption bacteria. Process Biochem 47(5):758–765. https://doi.org/10.1016/j.procbio.2012.02.007 427. Richter Ł, Janczuk-Richter M, Niedziółka-Jönsson J, Paczesny J, Hołyst R (2018) Recent advances in bacteriophage-based methods for bacteria detection. Drug Discov Today 23 (2):448–455. https://doi.org/10.1016/j.drudis.2017.11.007 428. Robinson GM, Tonks KM, Thorn RMS, Reynolds DM (2011) Application of bacterial bioluminescence to assess the efficacy of fast-acting biocides. Antimicrob Agents Chemother 55(11):5214–5219. https://doi.org/10.1128/AAC.00489-11 429. Roda A, Cevenini L, Borg S, Michelini E, Calabretta MM, Schüler D (2013) Bioengineered bioluminescent magnetotactic bacteria as a powerful tool for chip-based whole-cell biosensors. Lab Chip 13(24):4881–4889. https://doi.org/10.1039/c3lc50868d 430. Roda A, Cevenini L, Michelini E, Branchini BR (2011) A portable bioluminescence engineered cell-based biosensor for on-site applications. Biosens Bioelectron 26(8):3647– 3653. https://doi.org/10.1016/j.bios.2011.02.022 431. Roda A, Roda B, Cevenini L, Michelini E, Mezzanotte L, Reschiglian P, . . . Virta M (2011) Analytical strategies for improving the robustness and reproducibility of bioluminescent microbial bioreporters. Anal Bioanal Chem 401(1):201–211. https://doi.org/10.1007/s00216011-5091-3 432. Roggo C, van der Meer JR (2017) Miniaturized and integrated whole cell living bacterial sensors in field applicable autonomous devices. Curr Opin Biotechnol 45:24–33. https://doi. org/10.1016/j.copbio.2016.11.023 433. Saltepe B, Kehribar EŞ, Su Yirmibeşoǧlu SS, Şafak Şeker UO (2018) Cellular biosensors with engineered genetic circuits. ACS Sensors 3(1):13–26. https://doi.org/10.1021/acssensors. 7b00728 434. Scott D, Dikici E, Ensor M, Daunert S (2011) Bioluminescence and its impact on bioanalysis. https://doi.org/10.1146/annurev-anchem-061010-113855. www.scopus.com 435. Sharma S, Chatterjee S, Datta S, Prasad R, Dubey D, Prasad RK, Vairale MG (2017) Bacteriophages and its applications: an overview. Folia Microbiol 62(1):17–55. https://doi. org/10.1007/s12223-016-0471-x 436. Shen Y, Wang M, Chang IS, Ng HY (2013) Effect of shear rate on the response of microbial fuel cell toxicity sensor to cu(II). Bioresour Technol 136:707–710. https://doi.org/10.1016/j. biortech.2013.02.069 437. Shen YJ, Lefebvre O, Tan Z, Ng HY (2012) Microbial fuel-cell-based toxicity sensor for fast monitoring of acidic toxicity. Water Sci Technol 65(7):1223–1228. https://doi.org/10.2166/ wst.2012.957

1

Introduction to Cell Biosensors Through 55 Years of Scientific Production

37

438. Shing WL, Heng LY, Surif S (2013) Performance of a cyanobacteria whole cell-based fluorescence biosensor for heavy metal and pesticide detection. Sensors (Switzerland) 13 (5):6394–6404. https://doi.org/10.3390/s130506394 439. Si R, Yang Y, Yu Y, Han S, Zhang C, Sun D, . . . Yong Y (2016) Wiring bacterial electron flow for sensitive whole-cell amperometric detection of riboflavin. Anal Chem 88(22):11222– 11228. https://doi.org/10.1021/acs.analchem.6b03538 440. Si R, Zhai D, Liao Z, Gao L, Yong Y (2015) A whole-cell electrochemical biosensing system based on bacterial inward electron flow for fumarate quantification. Biosens Bioelectron 68:34–40. https://doi.org/10.1016/j.bios.2014.12.035 441. Siedler S, Khatri NK, Zsohár A, Kjærbølling I, Vogt M, Hammar P, . . . Joensson HN (2017) Development of a bacterial biosensor for rapid screening of yeast p-coumaric acid production. ACS Synth Biol 6(10):1860–1869. https://doi.org/10.1021/acssynbio.7b00009 442. Siedler S, Schendzielorz G, Binder S, Eggeling L, Bringer S, Bott M (2014) SoxR as a singlecell biosensor for NADPH-consuming enzymes in Escherichia coli. ACS Synth Biol 3(1):41– 47. https://doi.org/10.1021/sb400110j 443. Siedler S, Stahlhut SG, Malla S, Maury J, Neves AR (2014) Novel biosensors based on flavonoid-responsive transcriptional regulators introduced into Escherichia coli. Metab Eng 21:2–8. https://doi.org/10.1016/j.ymben.2013.10.011 444. Singh J, Mittal SK (2012) Chlorella sp. based biosensor for selective determination of mercury in presence of silver ions. Sensors Actuators B Chem 165(1):48–52. https://doi.org/10.1016/j. snb.2012.02.009 445. Skjoedt ML, Snoek T, Kildegaard KR, Arsovska D, Eichenberger M, Goedecke TJ, . . . Keasling JD (2016) Engineering prokaryotic transcriptional activators as metabolite biosensors in yeast. Nat Chem Biol 12(11):951–958. https://doi.org/10.1038/nchembio.2177 446. Smartt AE, Xu T, Jegier P, Carswell JJ, Blount SA, Sayler GS, Ripp S (2012) Pathogen detection using engineered bacteriophages. Anal Bioanal Chem 402(10):3127–3146. https:// doi.org/10.1007/s00216-011-5555-5 447. Smith MB, Rocha AM, Smillie CS, Olesen SW, Paradis C, Wu L, . . . Hazen TC (2015) Natural bacterial communities serve as quantitative geochemical biosensors. MBio 6(3):1–13. https:// doi.org/10.1128/mBio.00326-15 448. Steager EB, Sakar MS, Kim DH, Kumar V, Pappas GJ, Kim MJ (2011) Electrokinetic and optical control of bacterial microrobots. J Micromech Microeng 21(3). https://doi.org/10.1088/ 0960-1317/21/3/035001 449. Stein NE, Hamelers HMV, Van Straten G, Keesman KJ (2012) On-line detection of toxic components using a microbial fuel cell-based biosensor. J Process Control 22(9):1755–1761. https://doi.org/10.1016/j.jprocont.2012.07.009 450. Stein NE, Hamelers HVM, Buisman CNJ (2012) Influence of membrane type, current and potential on the response to chemical toxicants of a microbial fuel cell based biosensor. Sensors Actuators B Chem 163(1):1–7. https://doi.org/10.1016/j.snb.2011.10.060 451. Stein NE, Hamelers HVM, Buisman CNJ (2012) The effect of different control mechanisms on the sensitivity and recovery time of a microbial fuel cell based biosensor. Sensors Actuators B Chem 171–172:816–821. https://doi.org/10.1016/j.snb.2012.05.076 452. Stein NE, Hamelers HVM, van Straten G, Keesman KJ (2012) Effect of toxic components on microbial fuel cell-polarization curves and estimation of the type of toxic inhibition. Biosensors 2(3):255–268. https://doi.org/10.3390/bios2030255 453. Stein NE, Keesman KJ, Hamelers HVM, van Straten G (2011) Kinetic models for detection of toxicity in a microbial fuel cell based biosensor. Biosens Bioelectron 26(7):3115–3120. https:// doi.org/10.1016/j.bios.2010.11.049 454. Su L, Jia W, Hou C, Lei Y (2011) Microbial biosensors: a review. Biosens Bioelectron 26 (5):1788–1799. https://doi.org/10.1016/j.bios.2010.09.005 455. Sun J, Kingori GP, Si R, Zhai D, Liao Z, Sun D, . . . Yong Y (2015) Microbial fuel cell-based biosensors for environmental monitoring: a review. Water Sci Technol 71(6):801–809. https:// doi.org/10.2166/wst.2015.035

38

G. Thouand

456. Tang X, Zhang T, Liang B, Han D, Zeng L, Zheng C, . . . Liu A (2014) Sensitive electrochemical microbial biosensor for p-nitrophenylorganophosphates based on electrode modified with cell surface-displayed organophosphorus hydrolase and ordered mesopore carbons. Biosens Bioelectron 60:137–142. https://doi.org/10.1016/j.bios.2014.04.001 457. Tao H, Peng Z, Li P, Yu T, Su J (2013) Optimizing cadmium and mercury specificity of CadRbased E. coli biosensors by redesign of CadR. Biotechnol Lett 35(8):1253–1258. https://doi. org/10.1007/s10529-013-1216-4 458. Tawil N, Sacher E, Mandeville R, Meunier M (2014) Bacteriophages: biosensing tools for multi-drug resistant pathogens. Analyst 139(6):1224–1236. https://doi.org/10.1039/ c3an01989f 459. Tawil N, Sacher E, Mandeville R, Meunier M (2012) Surface plasmon resonance detection of E. coli and methicillin-resistant S. aureus using bacteriophages. Biosens Bioelectron 37(1):24– 29. https://doi.org/10.1016/j.bios.2012.04.048 460. Teo WS, Chang MW (2015) Bacterial XylRs and synthetic promoters function as genetically encoded xylose biosensors in Saccharomyces cerevisiae. Biotechnol J 10(2):315–322. https:// doi.org/10.1002/biot.201400159 461. Tlili C, Sokullu E, Safavieh M, Tolba M, Ahmed MU, Zourob M (2013) Bacteria screening, viability, and confirmation assays using bacteriophage-impedimetric/loop-mediated isothermal amplification dual-response biosensors. Anal Chem 85(10):4893–4901. https://doi.org/10. 1021/ac302699x 462. Torkzadeh-Mahani M, Ataei F, Nikkhah M, Hosseinkhani S (2012) Design and development of a whole-cell luminescent biosensor for detection of early-stage of apoptosis. Biosens Bioelectron 38(1):362–368. https://doi.org/10.1016/j.bios.2012.06.034 463. Truffer F, Buffi N, Merulla D, Beggah S, Van Lintel H, Renaud P, . . . Geiser M (2014) Compact portable biosensor for arsenic detection in aqueous samples with Escherichia coli bioreporter cells. Rev Sci Instrum 85(1). https://doi.org/10.1063/1.4863333 464. Tsai H, Tsai Y, Yagur-Kroll S, Palevsky N, Belkin S, Cheng J (2015) Water pollutant monitoring by a whole cell array through lens-free detection on CCD. Lab Chip 15 (6):1472–1480. https://doi.org/10.1039/c4lc01189a 465. Tschirhart T, Zhou XY, Ueda H, Tsao C, Kim E, Payne GF, Bentley WE (2016) Electrochemical measurement of the β-galactosidase reporter from live cells: a comparison to the miller assay. ACS Synth Biol 5(1):28–35. https://doi.org/10.1021/acssynbio.5b00073 466. Tucci M, Grattieri M, Schievano A, Cristiani P, Minteer SD (2019) Microbial amperometric biosensor for online herbicide detection: photocurrent inhibition of Anabaena variabilis. Electrochim Acta 302:102–108. https://doi.org/10.1016/j.electacta.2019.02.007 467. van Wijk SJL, Fiskin E, Putyrski M, Pampaloni F, Hou J, Wild P, . . . Dikic I (2012) Fluorescence-based sensors to monitor localization and functions of linear and K63-linked ubiquitin chains in cells. Mol Cell 47(5):797–809. https://doi.org/10.1016/j.molcel.2012.06.017 468. Velasquez-Orta SB, Werner D, Varia JC, Mgana S (2017) Microbial fuel cells for inexpensive continuous in-situ monitoring of groundwater quality. Water Res 117:9–17. https://doi.org/10. 1016/j.watres.2017.03.040 469. Vinay M, Franche N, Grégori G, Fantino J, Pouillot F, Ansaldi M (2015) Phage-based fluorescent biosensor prototypes to specifically detect enteric bacteria such as E. coli and Salmonella enterica typhimurium. PLoS One 10(7). https://doi.org/10.1371/journal.pone.0131466 470. Wang B, Barahona M, Buck M (2013) A modular cell-based biosensor using engineered genetic logic circuits to detect and integrate multiple environmental signals. Biosens Bioelectron 40(1):368–376. https://doi.org/10.1016/j.bios.2012.08.011 471. Wang H, Lang Q, Li L, Liang B, Tang X, Kong L, . . . Liu A (2013) Yeast surface displaying glucose oxidase as whole-cell biocatalyst: construction, characterization, and its electrochemical glucose sensing application. Anal Chem 85(12):6107–6112. https://doi.org/10.1021/ ac400979r 472. Wang X, Ahmed NB, Alvarez GS, Tuttolomondo MV, Hélary C, Desimone MF, Coradin T (2015) Sol-gel encapsulation of biomolecules and cells for medicinal applications. Curr Top Med Chem 15(3):223–244. https://doi.org/10.2174/1568026614666141229112734

1

Introduction to Cell Biosensors Through 55 Years of Scientific Production

39

473. Wang X, Gao N, Zhou Q (2013) Concentration responses of toxicity sensor with Shewanella oneidensis MR-1 growing in bioelectrochemical systems. Biosens Bioelectron 43(1):264–267. https://doi.org/10.1016/j.bios.2012.12.029 474. Wang X, Liu M, Wang X, Wu Z, Yang L, Xia S, . . . Zhao J (2013) P-benzoquinone-mediated amperometric biosensor developed with Psychrobacter sp. for toxicity testing of heavy metals. Biosens Bioelectron 41(1):557–562. https://doi.org/10.1016/j.bios.2012.09.020 475. Wang X, Lu X, Chen J (2014) Development of biosensor technologies for analysis of environmental contaminants. Trends Environ Anal Chem 2:25–32. https://doi.org/10.1016/j. teac.2014.04.001 476. Watstein DM, McNerney MP, Styczynski MP (2015) Precise metabolic engineering of carotenoid biosynthesis in Escherichia coli towards a low-cost biosensor. Metab Eng 31:171–180. https://doi.org/10.1016/j.ymben.2015.06.007 477. Watstein DM, Styczynski MP (2018) Development of a pigment-based whole-cell zinc biosensor for human serum. ACS Synth Biol 7(1):267–275. https://doi.org/10.1021/ acssynbio.7b00292 478. Weaver AA, Halweg S, Joyce M, Lieberman M, Goodson HV (2015) Incorporating yeast biosensors into paper-based analytical tools for pharmaceutical analysis. Anal Bioanal Chem 407(2):615–619. https://doi.org/10.1007/s00216-014-8280-z 479. Webster DP, TerAvest MA, Doud DFR, Chakravorty A, Holmes EC, Radens CM, . . . Angenent LT (2014) An arsenic-specific biosensor with genetically engineered Shewanella oneidensis in a bioelectrochemical system. Biosens Bioelectron 62:320–324. https://doi.org/ 10.1016/j.bios.2014.07.003 480. Wei T, Zhang C, Xu X, Hanna M, Zhang X, Wang Y, . . . Xiao W (2013) Construction and evaluation of two biosensors based on yeast transcriptional response to genotoxic chemicals. Biosens Bioelectron 44(1):138–145. https://doi.org/10.1016/j.bios.2013.01.029 481. Wei W, Liu X, Sun P, Wang X, Zhu H, Hong M, . . . Zhao J (2014) Simple whole-cell biodetection and bioremediation of heavy metals based on an engineered lead-specific operon. Environ Sci Technol 48(6):3363–3371. https://doi.org/10.1021/es4046567 482. Wen G, Wen X, Shuang S, Choi MMF (2014) Whole-cell biosensor for determination of methanol. Sensors Actuators B Chem 201:586–591. https://doi.org/10.1016/j.snb.2014. 04.107 483. Woolston BM, Roth T, Kohale I, Liu DR, Stephanopoulos G (2018) Development of a formaldehyde biosensor with application to synthetic methylotrophy. Biotechnol Bioeng 115 (1):206–215. https://doi.org/10.1002/bit.26455 484. Woutersen M, Belkin S, Brouwer B, Van Wezel AP, Heringa MB (2011) Are luminescent bacteria suitable for online detection and monitoring of toxic compounds in drinking water and its sources? Anal Bioanal Chem 400(4):915–929. https://doi.org/10.1007/s00216-010-4372-6 485. Xiao Y, De Araujo C, Sze CC, Stuckey DC (2015) Toxicity measurement in biological wastewater treatment processes: a review. J Hazard Mater 286:15–29. https://doi.org/10. 1016/j.jhazmat.2014.12.033 486. Xu X, Ying Y (2011) Microbial biosensors for environmental monitoring and food analysis. Food Rev Intl 27(3):300–329. https://doi.org/10.1080/87559129.2011.563393 487. Xu Z, Liu B, Dong Q, Lei Y, Li Y, Ren J, . . . Li B (2015) Flat microliter membrane-based microbial fuel cell as “on-line sticker sensor” for self-supported in situ monitoring of wastewater shocks. Bioresour Technol 197:244–251. https://doi.org/10.1016/j.biortech.2015.08. 081 488. Xu Z, Liu Y, Williams I, Li Y, Qian F, Zhang H et al (2016) Disposable self-support paperbased multi-anode microbial fuel cell (PMMFC) integrated with power management system (PMS) as the real time “shock” biosensor for wastewater. Biosens Bioelectron 85:232–239. https://doi.org/10.1016/j.bios.2016.05.018 489. Yagur-Kroll S, Amiel E, Rosen R, Belkin S (2015) Detection of 2,4-dinitrotoluene and 2,4,6trinitrotoluene by an Escherichia coli bioreporter: performance enhancement by directed evolution. Appl Microbiol Biotechnol 99(17):7177–7188. https://doi.org/10.1007/s00253015-6607-0

40

G. Thouand

490. Yagur-Kroll S, Belkin S (2011) Upgrading bioluminescent bacterial bioreporter performance by splitting the lux operon. Anal Bioanal Chem 400(4):1071–1082. https://doi.org/10.1007/ s00216-010-4266-7 491. Yagur-Kroll S, Lalush C, Rosen R, Bachar N, Moskovitz Y, Belkin S (2014) Escherichia coli bioreporters for the detection of 2,4-dinitrotoluene and 2,4,6-trinitrotoluene. Appl Microbiol Biotechnol 98(2):885–895. https://doi.org/10.1007/s00253-013-4888-8 492. Yagur-Kroll S, Schreuder E, Ingham CJ, Heideman R, Rosen R, Belkin S (2015) A miniature porous aluminum oxide-based flow-cell for online water quality monitoring using bacterial sensor cells. Biosens Bioelectron 64:625–632. https://doi.org/10.1016/j.bios.2014.09.076 493. Yang H, Zhou M, Liu M, Yang W, Gu T (2015) Microbial fuel cells for biosensor applications. Biotechnol Lett 37(12):2357–2364. https://doi.org/10.1007/s10529-015-1929-7 494. Yang W, Wei X, Fraiwan A, Coogan CG, Lee H, Choi S (2016) Fast and sensitive water quality assessment: a μl-scale microbial fuel cell-based biosensor integrated with an air-bubble trap and electrochemical sensing functionality. Sensors Actuators B Chem 226:191–195. https:// doi.org/10.1016/j.snb.2015.12.002 495. Yeom S, Kim M, Kwon KK, Fu Y, Rha E, Park S, Lee S (2018) A synthetic microbial biosensor for high-throughput screening of lactam biocatalysts. Nat Commun 9(1). https://doi. org/10.1038/s41467-018-07488-0 496. Yilmaz E, Majidi D, Ozgur E, Denizli A (2015) Whole cell imprinting based Escherichia coli sensors: a study for SPR and QCM. Sensors Actuators B Chem 209:714–721. https://doi.org/ 10.1016/j.snb.2014.12.032 497. You J, Walter XA, Greenman J, Melhuish C, Ieropoulos I (2015) Stability and reliability of anodic biofilms under different feedstock conditions: towards microbial fuel cell sensors. Sens Bio-Sens Res 6:43–50. https://doi.org/10.1016/j.sbsr.2015.11.007 498. Yu D, Bai L, Zhai J, Wang Y, Dong S (2017) Toxicity detection in water containing heavy metal ions with a self-powered microbial fuel cell-based biosensor. Talanta 168:210–216. https://doi.org/10.1016/j.talanta.2017.03.048 499. Yuan L, Lin W, Tan L, Zheng K, Huang W (2013) Lighting up carbon monoxide: fluorescent probes for monitoring CO in living cells. Angew Chem – Int Ed 52(6):1628–1630. https://doi. org/10.1002/anie.201208346 500. Zamaleeva AI, Sharipova IR, Shamagsumova RV, Ivanov AN, Evtugyn GA, Ishmuchametova DG, Fakhrullin RF (2011) A whole-cell amperometric herbicide biosensor based on magnetically functionalised microalgae and screen-printed electrodes. Anal Methods 3(3):509–513. https://doi.org/10.1039/c0ay00627k 501. Zhai J, Yong D, Li J, Dong S (2013) A novel colorimetric biosensor for monitoring and detecting acute toxicity in water. Analyst 138(2):702–707. https://doi.org/10.1039/ c2an36160d 502. Zhang X, Wiseman S, Yu H, Liu H, Giesy JP, Hecker M (2011) Assessing the toxicity of naphthenic acids using a microbial genome wide live cell reporter array system. Environ Sci Technol 45(5):1984–1991. https://doi.org/10.1021/es1032579 503. Zhang Y, Angelidaki I (2012) A simple and rapid method for monitoring dissolved oxygen in water with a submersible microbial fuel cell (SBMFC). Biosens Bioelectron 38(1):189–194. https://doi.org/10.1016/j.bios.2012.05.032 504. Zhang Y, Angelidaki I (2011) Submersible microbial fuel cell sensor for monitoring microbial activity and BOD in groundwater: focusing on impact of anodic biofilm on sensor applicability. Biotechnol Bioeng 108(10):2339–2347. https://doi.org/10.1002/bit.23204 505. Zhou M (2015) Recent progress on the development of biofuel cells for self-powered electrochemical biosensing and logic biosensing: a review. Electroanalysis 27(8):1786– 1810. https://doi.org/10.1002/elan.201500173 506. Zhou Y, Marar A, Kner P, Ramasamy RP (2017) Charge-directed immobilization of bacteriophage on nanostructured electrode for whole-cell electrochemical biosensors. Anal Chem 89 (11):5734–5741. https://doi.org/10.1021/acs.analchem.6b03751

Part II Fundamentals and Genetics for Cell Biosensors Applications

2

Stress Response-Based Whole-Cell Biosensor Development: Sentinels, Serendipity, and Circuitry Robert A. LaRossa

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Foundation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Precedents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Baltimore . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . New Haven . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Palo Alto . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Wilmington, DE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Other Technologies Were also Advancing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Serendipity Intervened in the Form of the Sulfonylurea Herbicides . . . . . . . . . . . . . . . . . . . . . . . . The Heat Shock Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Environmental Biotechnology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conceptual Protection of Wastewater Treatment Plant Bacterial Flora . . . . . . . . . . . . . . . . . . . . . . Expanding the Paradigm at DuPont . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Screening for Stress-Responsive Promoters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . An Ordered Array of E. coli Promoter::lux Whole-Cell Biosensors . . . . . . . . . . . . . . . . . . . . . . . . . Comparison of Whole-Cell Biosensor and Nucleic Acid Hybridization Measures of Gene Expression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Coda . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

44 44 45 45 45 46 46 47 47 48 48 49 51 53 53 53 54 55

Abstract

By 1970, perhaps the two outstanding examples of global control of gene expression had been described in Escherichia coli, namely, stringent control of ribosome biogenesis and activation of alternative catabolic operons upon glucose exhaustion. In the ensuing decades, a large number of analogous global regulatory circuits were discovered using ever more sophisticated biochemical and genetic techniques. These global regulatory systems included those activated

R. A. LaRossa (*) Red Jay Consulting, Chadds Ford, PA, USA © Springer Nature Switzerland AG 2022 G. Thouand (ed.), Handbook of Cell Biosensors, https://doi.org/10.1007/978-3-030-23217-7_113

43

44

R. A. LaRossa

by stresses, be they chemical, physical, biological, or nutritional. Such stress responses serve an adaptive function allowing cells to readjust in a battle to thrive. Thus these responses, often transcriptional, generally are triggered at sublethal levels by still metabolically active cells. In the early 1990s, facile monitoring of stress response induction using easily measured reporter gene products was exploited as early warning systems for a wide range of environmental and toxicological applications. The development of this whole-cell biosensor concept is recounted within the context of interacting societal, economic, and fundamental scientific concerns as well as prior research efforts. Keywords

Bioluminescence · DuPont · Global regulatory circuits · Histidine school · Microarrays · Pleiotropy · Promoter::lux fusions · Starvation responses · Stress responses · Transcriptomics · Translation

Introduction All organisms, including prokaryotes, have evolved over long periods to fill specific ecological niches. Not only must they grow and divide but they must adapt to a changing environment. Control of protein activity is a near-instantaneous response, while gene expression changes take place over a longer time frame of several minutes. Advances now allow unprecedented measures of transcript abundance and usage with great accuracy. Unfortunately, turnaround time for such assays is long, preventing the use of many of these technologies in monitoring/remediation applications. Thus the development of whole-cell biosensors that report on cellular responses in real time was a “Holy Grail.” A reminiscence of this undertaking is captured here with an emphasis upon events leading to the synthesis of a whole-cell biosensor concept.

Foundation A productive approach to solving biological problems can be traced back to studies of Euclidean geometry. Unlike algebra which is quite linear, geometric proofs needed to be devised by starting at both the front and back end; only then does the entire path come into focus. That training turned out to be critical in organic chemistry; while degradation is rather linear, synthesis can utilize many routes from the starting materials to the final product. Molecular biological breakthroughs also tend to require novelty. When stuck, thoughts would be melded, looking for new relationships, or dogma inverted. From such mental exercises, new avenues emerged that often were useful. It is hoped that this retrospective look, colored by such a problem-solving perspective, illustrates a powerful means by which to tackle biological questions.

2

Stress Response-Based Whole-Cell Biosensor Development: Sentinels. . .

45

Precedents Classroom and laboratory preparation was crucial in garnering the requisite insights needed for contributions to the whole-cell biosensor approach.

Baltimore Training in the Biology Department of Johns Hopkins in the early 1970s was a unique undergraduate experience. After general biology, courses were taught in a proscribed sequence: cell biology, biochemistry, genetics, and developmental biology. There were exceptional professors such as Steve Roth, Saul Roseman, and Phil Hartman who whetted novices’ appetites. Undergraduates were left to their own devices until they entered research laboratories where they apprenticed in the biological arts. There they were treated as full lab members, attending research group meetings and the weekly Departmental Seminar where speakers visiting from other institutions each presented a different, fascinating story. The Hartman Lab was focused upon biochemical genetics and molecular biology. Hartman’s interests were manifold but integrative including biosynthesis, biochemistry, genetics, toxicology, and the environment. Mutagens, together with genetic selections using metabolic inhibitors, had opened many new avenues to explore. Using inhibitors targeted toward different proteins as a screening device, it was determined that some regulatory mutations were quite specific (e.g., the lac operon repressor, lacI), while histidine biosynthetic inhibitors yielded both specific mutations such as hisS effecting only histidine biosynthetic enzymes and pleiotropic mutations, for example, hisT, altering the levels of many biosynthetic proteins involved in disparate metabolic pathways. Such pleiotropic mutations are not a repudiation of the Beadle-Tatum “one gene-one polypeptide” hypothesis; rather they emphasize the interlocking nature of carrying out 1000s of reactions in the small space of a cell. Laboratory work under the guidance of a postdoctoral fellow, Takashi Kasai, was biochemical in nature providing protein, DNA, and RNA reagents for Kasai’s investigation of genetic regulation of the histidine operon (Kasai 1974). Importantly, Roseman and Hartman were collaborating on genetic dissection of the phosphotransferase (pts) system that emphasized interaction between it and cAMP-mediated control of non-glucose carbon source utilization in Salmonella. In addition, Ames and Hartman were actively developing means of identifying mutagens and carcinogens by reversion of his mutations. One key advance in carcinogen detection was to alter the bacterial cell surface to allow hydrophobic molecules to physically interact with the bacterial genome (Ames et al. 1973).

New Haven In the mid-1970s, the Department of Molecular Biophysics and Biochemistry (MB&B) at Yale, headed by Fred Richards, provided a complimentary set of

46

R. A. LaRossa

experiences. First-year rotations involved working in labs studying DNA replication and repair (Dean Rupp), chemotaxis (Bob Macnab), and translation (Dieter Söoll). With Gabriel Vogeli, Brooks Low, and Dieter, second-site suppressors of a thermolabile leucyl-tRNA synthetase (leuS) mutant of E. coli were biochemically, physiologically, and genetically characterized. The playbook for this work (LaRossa 1977) was templated upon the his operon studies of Hartman and Ames. Mutations were combined to test hypotheses, visual phenotypes were used, and metabolic inhibitors were exploited. From Bruce Ames, Dieter had obtained copies of index cards that recorded the effects of a wide range of inhibitors upon Salmonella; these cards served as a most useful guide. Using radioactive tracers for in vivo studies emphasized the need for looking at biological responses as a function of time. Moreover, the impact of macromolecular three-dimensional structure (the primary focus of MB&B) upon function was highly emphasized. Reading ongoing work emanating from the histidine school (Hartman, John Roth, Ames, and many others) and Baltimore (the pleiotropy of certain mutations in the pts sugar transport system) broadened horizons. Crucially important was the description of the stringent response as both a negative regulator of ribosome synthesis and a stimulator of amino acid biosynthesis (Stephens et al. 1975). Similarities between the stringent response and catabolite activation system were becoming more apparent. Such hierarchal global regulatory mechanisms would prove to be quite common, present in both prokaryotes and eukaryotes.

Palo Alto The Stanford Biochemistry Department, circa 1977, was a wonder, perhaps the leader of the recombinant DNA revolution. Greatly influenced both by these emerging technologies and efforts to use transposons as genetic tools, the genetic arsenal available to address intriguing biological questions in a wide variety of organisms was expanded greatly. Dale Kaiser’s lab was using genetics to understand intercellular coordination. Kaiser’s associates, Doug Berg and Jerry Kuner, had developed a phage vehicle that would inject DNA into the target bacterium, Myxococcus xanthus, but its genome would not be replicated. Thus the only way to form a drug-resistant bacterial colony was for the transposon to move from the phage to the bacterial genome; it did so in a random manner. Thus a library of independent transposition events could be isolated each at a unique location in the bacterial genome. Cotransduction of drug resistance and the ability to form fruiting bodies into defective (non-fruiting) mutants could be scored, providing a means of mapping non-communicating mutations (LaRossa et al. 1983).

Wilmington, DE In 1980 transposon genetics was integrated into DuPont’s Central Research and Development Department in the hope of developing bacteria that could make large quantities of commodity chemicals. At that time, the DuPont Experimental Station,

2

Stress Response-Based Whole-Cell Biosensor Development: Sentinels. . .

47

the Roche Institute, and the Bell Labs were outstanding places to pursue the combination of applied and fundamental research.

Other Technologies Were also Advancing At Harvard Medical School, a transposon was constructed that would randomly insert both drug resistance and a promoter-less lac operon. This promoter probe transposon would be most useful for defining the extent of global regulatory circuits (Casadaban and Cohen 1979). In a complementary manner, a biochemical approach was developed. The total protein content of a bacterial culture was reproducibly separated by charge in the first dimension and by size in the second dimension. The Neidhardt lab at the University of Michigan used this separation technique to determine protein patterns of responses (stimulons) to various stimuli and response patterns (regulons) to regulatory mutations, be they global or pathway specific (VanBogelen et al. 1996). Finally the cellular mononucleotides were separated by two-dimensional thin-layer chromatography providing a complete cellular inventory of these important metabolic and signaling molecules (Bochner and Ames 1982).

Serendipity Intervened in the Form of the Sulfonylurea Herbicides Knowledge of global regulatory circuitry, cell surface biology, and amino acid biosynthesis was decisive in determining the enzymatic target, acetolactate synthase, of this most potent class of herbicides produced by the DuPont Company. This class of herbicides inhibits bacteria grown on a minimal medium but not on a rich broth. Minimal medium supplementation with the branched chain amino acids restored growth indicating that this biosynthetic pathway was targeted. Other nutrients also overcame these herbicides’ inhibition suggesting both the identity of the targeted enzyme and that a 2-ketoacid imbalance resulted from the herbicide treatment. It was found that both global regulatory mutations and pleiotropic mutations allowing accumulation of many hydrophobic molecules within the cell interior were much more sensitive to the sulfonylurea herbicides on minimal medium than the parental strain. It was demonstrated that the parental, herbicide-sensitive strain had a herbicide-sensitive acetolactate synthase strain, while resistant Salmonella mutants had a herbicide-resistant acetolactate synthase (LaRossa and Schloss 1984). Later work in eukaryotes, including yeast, tobacco, and Arabidopsis, yielded analogous results, showing that bacteria could be excellent model organisms for determining how certain herbicides work. Importantly, using transposon genetics, many herbicide hypersensitive mutants were isolated that together suggested that these herbicides were potent because of the resultant imbalance among 2-ketoacids. This provides a rationale for the observation that only a single enzyme of the branched chain amino acid biosynthetic pathway was a worthwhile herbicide target (LaRossa and Van Dyk 1987). Thus the importance of bacterial models, their pleiotropic mutations, and prokaryotic global regulatory mechanisms was evident. Moreover, Ames’ and Hartman’s long-standing interest in metabolic inhibitors had become important to

48

R. A. LaRossa

DuPont; a great deal of future efforts would be consumed by striving to both understand chemical-cellular interactions that are metabolic or regulatory in nature and to isolate mutants that can thrive under conditions that halt growth of the parental strain. This focus led to a comprehensive review of genetic selections (LaRossa 1996) in the second edition of the E. coli/Salmonella “Bible.” As an adjunct, heat-sensitive mutations in ilvG encoding a polypeptide of the herbicide’s target enzyme, acetolactate synthase, were isolated.

The Heat Shock Response In the late 1980s, colleagues at DuPont had found that overproduction of the E. coli GroESL complex allowed a foreign bacterial RUBISCO protein to achieve an active confirmation. To test if this was a general phenomenon, GroESL overproduction was introduced into two heat-sensitive ilvG mutations. Activity was restored; groESL was acting as a multicopy suppressor of the ilvG heat sensitivity (LaRossa and Van Dyk 1991). To generalize this suppression, thermolabile his operon mutations were exploited; Kohno and Roth (1979) had demonstrated that a large number of his operon heat-sensitive mutations were osmotically remediable. Many of these thermolabile his mutations were also suppressed by multicopy groESL (LaRossa and Van Dyk 1991). GroESL is now known to be a protein folding machine that is overproduced in response to heat shock. Thus this work had come to another global regulatory circuit, the heat shock response.

Environmental Biotechnology In the early 1990s, DuPont placed great emphasis upon being environmental stewards, attempting to insure that chemicals would not pollute the environment. The effort took many forms; one was discussions between engineers and biologists. Particularly productive was an interaction between Central Research and Development molecular biologists and wastewater treatment plant engineers. Wastewater treatment plants are an open ecosystem in which endogenous bacterial flora catabolize organic chemicals in an aqueous effluent from a chemical production facility. If the flora within the wastewater treatment plant were no longer metabolically active, the wastewater stream leaving the plant would contaminate the downstream waterways that it entered. Thus maintaining health of the wastewater treatment plant flora was of paramount importance both ecologically and economically. If the wastewater treatment plant could not do its job, the entire upstream chemical production had to be curtailed with the resultant economic losses associated with downtime, recovery of the bacterial flora in the wastewater plant, and reinitiation of otherwise continuous chemical processes in the production facility. At that time, about 1 of more than 20 wastewater treatment facilities within the DuPont Company failed each year resulting in a productivity loss (equivalent to an income deficit) that, if the failure went undetected, was compounded by associated environmental damage, often of adjacent rivers as well as groundwater.

2

Stress Response-Based Whole-Cell Biosensor Development: Sentinels. . .

49

State of the Art There were several tests that were in use at this time. Among them were the Ames test, Microtox, and Lux-based metal biosensors. In addition a variety of stress responses were being defined. The Ames Test (Ames et al. 1973) This was designed to detect DNA damage by quantitatively monitoring reversion of histidine-requiring Salmonella mutants to strains that regained the ability to grow in the absence of histidine. One problem was that many carcinogens that were mutagenic to animal cells were not mutagenic to Salmonella initially used in the Ames test. Inability of the chemical to access the bacterial genome was a possible explanation for this discrepancy. Genetic alteration of the cell surface allowed many of these otherwise “inert” carcinogens to cause mutations indicating that the wildtype surface indeed provided a protective barrier. Microtox (Blaise et al. 1994) This was a radically different test that indirectly measured loss of cellular viability of a bioluminescent bacterium. Bacterial bioluminescence requires energy in the form of ATP and reducing power in the form of NADPH and FMH2. Thus loss of bioluminescence is correlated with loss of viability. Measurement of light production is a powerful assay that has a dynamic range of more than seven orders of magnitude using instrumentation available in the 1990s. Metal-Detecting Biosensors Several metallo-controlled regulatory circuits were known. As an adjunct to analytical chemistry, metal-regulated promoter elements for these regulatory circuits were fused to a reporter gene, and induction was monitored as a function of dosage. In this way bioavailability of distinct metals could be assessed with a rather specific assay (e.g., Condee and Summers 1992). This fusion produced a reporter that needed to be supplied with an exogenous substrate, narrowing its utility. Stress Response Induction A general phenomenon was observed, in that application of a low level of a particular stress (the adaptation) would allow cells to survive a higher dose of that particular stress (the challenge), but not other stresses, that was sufficient to kill naive cells. This acquired tolerance to the challenge required new protein synthesis during the time of adaptation (Storz and Zheng 2000).

Conceptual Protection of Wastewater Treatment Plant Bacterial Flora Discussion with the wastewater treatment engineers was relatively straightforward. It was asked if the engineer would dilute the influent from the chemical facility 10- to

50

R. A. LaRossa

100-fold if that would protect the resident bacterial flora in the wastewater treatment plant. With an affirmative answer, the following concept was formulated. It was predicated upon the thought that stress responses are triggered at sublethal doses in an attempt to alter the cells’ protein composition so that the cell can adapt and hence overcome an otherwise lethal challenge. The stress response should occur at concentrations far less than the lethal dosage (see Fig. 1). It was envisioned that a number of bacterial sentinels, each capable of producing a common signal (e.g., increased light production) in response to distinct, defined stress responses, could be

Fig. 1 The Reporter System. (a) A stress-activated control region (also referred to as a promoter) is fused to the luxCDABE genes whose gene products are necessary and sufficient to impart bioluminescence on an aerobically grown bacterium. Mutations, such as a null tolC allele of E.coli, allowing chemicals to better accumulate in the tester strain, are desirable. (b) The reporter bacterium senses a stress and generates one or several signals. If metabolism is compromised by a toxic level of the stressor, respiration is stopped, requisite cofactors for bioluminescence are no longer available, and light output is diminished. If the stress activates the promoter, then more Lux proteins are made resulting in an increased rate of bioluminescent output. (c) Idealized kinetic response of a promoter::lux fusion to a low level of the stressor to which the organism can mount an adaptive response (purple) and a higher stress level (red) that can be either bacteriostatic or bactericidal. The control response is indicated by the black line. High-level stress results in a rapid decrease in light output. The response to the lower, inducing level of the stressor is much slower since time is required, after the stressor or its derivative signal is sensed, for the lux operon transcript to accumulate to a higher level followed by translation into the five Lux proteins. (d) Idealized dose responses, at a prescribed time after administration of stress, of a specific promoter::lux fusion to two stressors: stressor 1 (purple) activates the promoter region, while stressor 2 (red) does not alter expression from the fused promoter. The untreated output of the two strains is illustrated by the horizontal black line. Note that for stressor 1, the concentration elevating lux expression is much less than the level that compromises metabolism shutting off bioluminescence

2

Stress Response-Based Whole-Cell Biosensor Development: Sentinels. . .

51

assembled. Together this ensemble, a set of distinct scouts, would provide an early warning system for many stresses that could compromise cellular health. This set of E. coli strains carrying stress-responsive promoters fused to the luxCDABE operon was constructed in the mid-1990s when the sequence of the E. coli genome was fragmentary. LuxAB is the heterodimeric, oxygen-requiring enzyme luciferase that produces light when supplied with an aldehyde substrate and FMNH2. The other three polypeptides form a complex. LuxD is a thioesterase that liberates a fatty acid that had been appended to acyl carrier protein, and LuxE forms a fatty acyl-AMP adduct at the expense of energy in the form of ATP. The acyl-AMP adduct is reduced to the aldehyde consuming NADPH by LuxC. It is apparent that light production requires energy in the form of ATP and reductants in the forms of NADPH and FMNH2.

Concept Validation The first stress promoter::lux fusions, described in 1994 (Van Dyk et al. 1994), took advantage of the heat shock response, a conserved response in all forms of life. It is now known that this is a response to un- or mis-folded proteins present within the cytoplasm. Moreover, not only heat but many chemicals, including alcohols, trigger this response. As expected, low levels of alcohols would increase lux expression. As the dosage increased, the light production ceased, a Microtox-like indication that metabolism was compromised at the high dosage. Importantly, it was known that this regulatory circuitry was dependent upon the heat shock-specific sigma factor, RpoH, which is an alternative subunit of RNA polymerase. Deletion of rpoH in the heat shock promoter::lux fusion strain extinguished the increase in light production at low doses but not the decrease in basal level light emission. Thus a whole-cell biosensor of unfolded protein in the cytoplasm was developed and shown to be capable of detecting a wide range of pollutants (Van Dyk et al. 1994).

Expanding the Paradigm at DuPont The range of stress-responsive promoter::lux fusions was rapidly expanded to include the peroxide (OxyR-controlled) regulon member katG, the superoxide (SoxR, SoxS-controlled) regulon member micF, SOS (RecA, LexA-controlled) regulon members recA and uvrA, adaptive (Ada-controlled) regulon member alkA, universal (FadR, IHF, Crp, and RelA modulate this response) stress gene uspA, and general (RpoS-controlled) regulon member xthA of E. coli. In addition metabolic promoters are also regulated by global regulators and constitute several starvationstress responses. Thus fabA, his, lac, phoA, and glnA promoter::lux fusions were assembled to monitor responses to membrane perturbation, amino acid starvation, catabolite activation, phosphate starvation, and nitrogen starvation. PCR amplification of upstream regulatory regions was the starting point for construction of these operon fusions; such constructions required that sequences of the genomic region of interest had been accomplished, something that was far from a certainty before the complete genomic sequence of E. coli was available. In all cases genetic blocks extinguishing the regulatory signal prevented the cognate response of increased light production driven by the known chemical inducer of the specific regulatory circuit

52

R. A. LaRossa

(LaRossa and Van Dyk 2000). To expand the range of detection, tolC derivatives of the fusions were made that lack the ability to export toxic compounds (Van Dyk et al. 1994).

Parallel Efforts The construction of promoter-reporter gene fusions has a long history. A major advance was the construction of a Mu phage lacZ promoter probe (Casadaban and Cohen 1979). This phage inserts randomly into the E. coli chromosome generating a strain library of phage inserts. The resultant strains can be screened for increased lac expression under a stressful condition such as DNA damage or phosphate starvation. Indeed, such genetic experiments were key to defining the extents of the SOS (Kenyon and Walker 1980) and phosphate starvation regulons (Wanner 1996). In addition, the Ames Lab was interested in global regulatory circuitry and invested in defining broad stress responses (notably the OxyR regulon (Christman et al. 1985)) and the alarmone concept (Stephens et al. 1975). These studies produced valuable insights complimentary to the heat shock work; moreover the Ames group proved to be an incubator for friendly competitors/colleagues. These included the Ames postdoctoral fellows Spencer Farr and Barry Bochner who founded Xenometrix and Biolog, respectively. Xenometrix produced a set of stress promoter-lac fusions analogous to the DuPont promoter::lux fusions, while Biolog used respiratory reduction of dyes as a measure of metabolic capacity. Moreover, the influence of Ames on his Berkeley faculty colleague, Jasper Rine, is reflected by Rine’s entrepreneurial use of yeast promoter-gfp fusions to study eukaryotic cellular responses to chemical insults (Dimster-Denk et al. 1999). Advantages of lux-Based Whole-Cell Biosensors This luxCDABE class of reporters is quite attractive. Among their advantages are: (a) The assay does not require cell breakage or permeabilization; rather the readout is an in vivo measure. (b) Substrate, cofactors, and buffers are not required simplifying experimentation and lowering cost. (c) Commercial microplate luminometers have a linear range of 107 to 108. (d) Assays can be performed in a 96-well microtiter-based format with readings at an interval of 2 or 3 min without sacrifice or removal of culture aliquots. Thus time course experiments are readily performed. (e) This microtiter-based format allows twofold dilutions covering a chemical range of either approximately 100 (in the 8-well dimension)- or 1000-fold (in the 12well dimension). Such a broad range of challenges with a single chemical allows visualization of both toxic (lights off) and stress induction (lights on). (f) Again, the toxic (lights off) response is much more rapid than the stress induction (lights on) which requires sensing of a chemical concentration, a change in gene expression, and an accumulation of gene products before increased light production is observed. (g) These stress-responsive promoter-operon fusions were a feat of engineering; presuming an answer, could a whole-cell biosensor be built to meet that goal,

2

Stress Response-Based Whole-Cell Biosensor Development: Sentinels. . .

53

e.g., detection of intracellular unfolded protein? This engineering approach could be described as one that shows chutzpah, a Yiddish term for nerve, guts, or boldness.

Screening for Stress-Responsive Promoters After success, it was time for humility. Still not fully understanding why acetolactate synthase was a great target for herbicides, it was worthwhile to screen for gene expression changes associated with the administration of sub-inhibitory levels of a sulfonylurea herbicide to E. coli (Van Dyk et al. 1998). It was hypothesized that this treatment would elevate expression of ppGpp (stringent response)-activated genes involved in amino acid synthesis. A library of random pieces of the E. coli genome was inserted into a lux promoter probe vector and transformed into a herbicidesensitive E. coli mutant; 19 of 8000+ transformants screened showed the hoped-for result, namely, that the herbicidal application induced increased light production from the set of 19. Led by Scott Tingey and Antoni Rafalski, the DuPont Agricultural Biotechnology Group had developed a high-throughput sequencing service to characterize plant cDNA clones; they were happy to contribute this skill to these efforts. The combined effort produced a surprising result; 12 of these fused, herbicide-induced promoters were controlled by the RpoS-mediated general stress response (Van Dyk et al. 1998). Could it be that 2-ketoacid and derivative acyl-CoA imbalances that were previously identified as an important part of the response to sulfonylurea herbicides (LaRossa and Van Dyk 1987) were triggering RpoSmediated growth arrest?

An Ordered Array of E. coli Promoter::lux Whole-Cell Biosensors Once the complete DNA sequence of the E. coli K-12 genome was determined, it was decided that sequencing of each E.coli chromosomal region upstream of the lux operon within the 8000+ member promoter library was appropriate. The Agricultural Biotechnology colleagues again concurred; the sequenced library was culled yielding a non-redundant set of 714 unique fusion constructs that was arrayed, stored, and exploited. Bioinformatic analyses suggested that this subset reported upon about 30% of the promoters present in the E. coli genome (Van Dyk et al. 2001a). The arrayed subset was used in screens for promoters responsive to specific chemicals of interest (humble screening) as well as being cherry-picked for hypothesis-driven research (e.g., returning to the engineering approach).

Comparison of Whole-Cell Biosensor and Nucleic Acid Hybridization Measures of Gene Expression Yan Wei spearheaded the development of nucleic acid hybridization (DNA microarray) methods to perform transcriptomic studies of E. coli in collaboration with the

54

R. A. LaRossa

Agricultural Biotechnology colleagues at DuPont (Wei et al. 2001). Having two experimental approaches (operon fusions and nucleic acid hybridization) for global gene expression profiling in a single laboratory allowed cross-platform comparisons. It was found that the lux fusions could detect much lower level expression than the hybridization experiments; that is, background noise was not the complicating issue for the lux fusions that it was for the microarray studies (Van Dyk et al. 2001b). In addition, signal saturation, a known limitation in hybridization experiments, was not problematic within the context of bioluminescent readout due to the aforementioned huge dynamic range of commercial luminometers (Van Dyk et al. 2001b). Hybridization studies have been supplanted by massive sequencing technologies such as RNAseq that obviate much of both the noise and saturation problems (Croucher and Thompson 2010). The significant amount of downstream operations (such as sampling, molecular biological workup, quantitation of transcript abundance by sequencing, and computational analysis) associated with the various transcriptomic technologies indicates an important, complementary, and ongoing role for whole-cell biosensors in the molecular toxicology and environmental arenas.

Coda The requirements placed upon a molecular biologist in an industrial setting are such that long periods on a single program are rare. Compensating for such potentially frustrating changes is the knowledge that the interaction between chemicals and cells is essential to many corporate successes and for protection of the biosphere. Later fundamental work on missense suppression (Ruan et al. 2008) and applied efforts on biofuel/industrial chemical production (an original reason for bacterial molecular biology at DuPont) utilized inhibitors, classical genetics, evolution, transcriptomics, and stress-responsive biosensors to define both challenges faced by cells in unusual situations and approaches by which more robust “microbial cell factories” can be produced. Was it serendipity that academic interests could be so congruent with an industrial advances? Corporate fermentation production targets, like herbicides or mutagens before them, elicit responses that can be interpreted in terms of biochemical knowledge and global regulatory circuitry. Thus the work described here, one initial step in the development of the field described in this compendium, has had a much greater impact than envisioned at its outset. Finally scientific giants such as Hartman, Ames, Söll, and Kaiser provided a wealth of knowledge and experiences upon which to build. Each, in their own way, blazed a unique path though their approaches displayed some commonalities. They made outstanding contributions to their fields, provided great training to their students, and were quite willing to flaunt conventional wisdom. As Frank Sinatra sang, they did it their way(s). It is hoped that you see a reflection of their science and commitment within the work described here; namely, “riffs” often enunciated by these four individuals are found within the whole-cell biosensor concept. Providing a unique interface between DuPont needs and fundamental scientific advances, it is

2

Stress Response-Based Whole-Cell Biosensor Development: Sentinels. . .

55

hoped that the DuPont Central Research and Development team adequately served corporate stakeholders, the broad scientific community, and the world.

References Ames BN, Lee FD, Durston WE (1973) An improved bacterial test system for the detection and classification of mutagens and carcinogens. Proc Natl Acad Sci USA 70:782–786 Blaise C, Forghani R, Legault R et al (1994) A bacterial toxicity assay performed with microplates, microluminometry and Microtox reagent. Biotechniques 16:932–937 Bochner BR, Ames BN (1982) Complete analysis of cellular nucleotides by two-dimensional thin layer chromatography. J Biol Chem 257:9759–9769 Casadaban MJ, Cohen SN (1979) Lactose genes fused to exogenous promoters in one step using a Mu-lac bacteriophage: in vivo probe for transcriptional control sequences. Proc Natl Acad Sci USA 76:4530–4533 Christman MF, Morgan RW, Jacobson FS et al (1985) Positive control of a regulon for defenses against oxidative stress and some heat-shock proteins in Salmonella typhimurium. Cell 41:753–762 Condee CW, Summers AO (1992) A mer-lux transcriptional fusion for real-time examination of in vivo gene expression kinetics and promoter response to altered superhelicity. J Bacteriol 174:8094–8101 Croucher NJ, Thompson NR (2010) Studying bacterial transcriptomes using RNA-seq. Curr Opin Microbiol 13:619–624 Dimster-Denk D, Rine J, Phillips J et al (1999) Comprehensive evaluation of isoprenoid biosynthesis regulation in Saccharomyces cerevisiae utilizing the Genome Reporter Matrix. J Lipid Res 40:850–860 Kasai T (1974) Regulation of the expression of the histidine operon in Salmonella typhimurium. Nature 249:523–527 Kenyon CJ, Walker GC (1980) DNA-damaging agents stimulate gene expression at specific loci in Escherichia coli. Proc Natl Acad Sci USA 77:2819–2823 Kohno T, Roth J (1979) Electrolyte effects on the activity of mutant enzymes in vivo and in vitro. Biochemistry 18:1386–1392 LaRossa RA (1977) The regulation of leucyl-tRNA synthetase biosynthesis in Escherichia coli K12. Dissertation, Yale University LaRossa RA (1996) Mutant selections linking physiology, inhibitors and genotypes. In: Neidhardt FC (ed) Escherichia coli and Salmonella typhimurium: cellular and molecular biology, 2nd edn. ASM Press, Washington, DC, pp 2527–2587 LaRossa RA, Schloss JV (1984) The sulfonylurea herbicide sulfometuron methyl is an extremely potent and specific inhibitor of acetolactate synthase in Salmonella typhimurium. J Biol Chem 259:8753–8757 LaRossa RA, Van Dyk TK (1987) Metabolic mayhem caused by 2-ketoacid imbalances. Bioessays 7:125–130 LaRossa RA, Van Dyk TK (1991) Physiological roles of the dnaK and groE stress proteins: catalysts of protein folding or macromolecular sponges? Mol Microbiol 5:529–534 LaRossa RA, Van Dyk TK (2000) Applications of stress responses for environmental monitoring and molecular toxicology. In: Storz G, Hengge-Aronis R (eds) Bacterial stress responses. ASM Press, Washington, DC, pp 453–468 LaRossa R, Kuner J, Hagen D et al (1983) Developmental cell interactions of Myxococcus xanthus: analysis of mutants. J Bacteriol 153:1394–1404 Ruan B, Palioura S, Sabina J et al (2008) Quality control despite mistranslation caused by an ambiguous genetic code. Proc Natl Acad Sci USA 105:16502–16507

56

R. A. LaRossa

Stephens JC, Artz SW, Ames BN (1975) Guanosine 50 -diphosphate 30 -diphosphate (ppGpp): positive effector for histidine operon transcription and general signal for amino-acid deficiency. Proc Natl Acad Sci USA 72:4389–4393 Storz G, Zheng M (2000) Oxidative stress. In: Storz G, Hengge-Aronis R (eds) Bacterial stress responses. ASM Press, Washington, DC, pp 453–468 Van Dyk TK, Majarian WR, Konstantinov KB et al (1994) Rapid and sensitive pollutant detection by induction of heat shock-bioluminescence gene fusions. Appl Environ Microbiol 60:1414–1420 Van Dyk TK, Ayers BL, Morgan RW et al (1998) Constricted flux through the branched-chain amino acid biosynthetic enzyme acetolactate synthase triggers elevated expression of genes regulated by rpoS and internal acidification. J Bacteriol 180:785–792 Van Dyk TK, Gonye GE, Reeve MJG et al (2001a) Genome-wide expression profiling with luxCDABE gene fusions. In: Case J, Herring P, Robison B et al (eds) Proceedings of the 11th international symposium on bioluminescence and chemiluminescence. World Scientific, Singapore, pp 461–464 Van Dyk TK, Wei Y, Hanafey MK et al (2001b) A genomic approach to gene fusion technology. Proc Natl Acad Sci USA 98:2555–2560 VanBogelen RA, Abshire KZ, Pertsemlidis A et al (1996) Gene-protein database of Escherichia coli K-12, Edition 6. In: Neidhardt FC (ed) Escherichia coli and Salmonella typhimurium: cellular and molecular biology, 2nd edn. ASM Press, Washington, DC, pp 2067–2117 Wanner BL (1996) Phosphorous assimilation and control of the phosphate regulon. In: Neidhardt FC (ed) Escherichia coli and Salmonella typhimurium: cellular and molecular biology, 2nd edn. ASM Press, Washington, DC, pp 1357–1381 Wei Y, Lee J-M, Richmond C et al (2001) High-density microarray mediated gene expression profiling of Escherichia coli. J Bacteriol 183:545–556

3

Engineering Autobioluminescent Eukaryotic Cells as Tools for Environmental and Biomedical Surveillance Tingting Xu, Dan Close, Ghufran Ud Din, Gary Sayler, and Steven Ripp

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Whole-Cell Bioluminescent Bioreporters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Transition from Bioluminescence to Autobioluminescence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Environmental Surveillance Using Autobioluminescent Yeast Bioreporters . . . . . . . . . . . . . . . . . . . . Biomedical Surveillance Using Autobioluminescent Human Cell Lines . . . . . . . . . . . . . . . . . . . . . . . Conclusions and Future Directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

58 59 61 62 66 67 68

T. Xu The Center for Environmental Biotechnology and Department of Microbiology, The University of Tennessee, Knoxville, TN, USA e-mail: [email protected] D. Close · G. Sayler 490 BioTech Inc., Knoxville, TN, USA e-mail: [email protected]; [email protected] G. Ud Din The Department of Microbiology, Quaid-i-Azam University, Islamabad, Pakistan e-mail: [email protected] S. Ripp (*) The Center for Environmental Biotechnology and Department of Microbiology, The University of Tennessee, Knoxville, TN, USA 490 BioTech Inc., Knoxville, TN, USA e-mail: [email protected] © Springer Nature Switzerland AG 2022 G. Thouand (ed.), Handbook of Cell Biosensors, https://doi.org/10.1007/978-3-030-23217-7_117

57

58

T. Xu et al.

Abstract

Cells engineered to express bioluminescence generate photons through a biochemical reaction catalyzed by a luciferase enzyme and a luciferin substrate. Conventional bioluminescent imaging approaches require that the luciferin substrate be added to the cell externally since it cannot be generated intracellularly. Thus, these cells remain “dark,” only emitting their bioluminescent signal in concert with the extracellular addition of the chemical substrate. Autobioluminescence represents an alternative signaling strategy whereby a synthetic bacterial luciferase, optimized for expression in eukaryotic cells, is used to sustain continuous photon emission without the necessary external addition of a substrate to drive reaction kinetics. This substrate-independent generation of light thereby allows autobioluminescent cells to be imaged at any longitudinal time scale desired to generate real-time metabolic data under both highthroughput in vitro and small animal in vivo experimental set-ups for applications ranging from environmental monitoring to preclinical biomedical imaging. Keywords

Autobioluminescence · Bacterial luciferase · Bioluminescent imaging · In vivo imaging · Luciferin · Lux · Optical imaging

Introduction In nature, various bacteria, fungi, protozoa, dinoflagellates, and higher-order marine and terrestrial organisms have evolved with the unique ability to emit bioluminescent light. This “glow-in-the-dark” trait, most often recognized in the firefly, enables species to communicate with each other or with their environment or provides for an effective self-defense mechanism. As is typical in scientific research and development, nature oftentimes provides the impetus for new tools and technologies, and bioluminescence followed this path to become a powerful imaging technique whose genetic architecture has been modified and manipulated to function in a variety of prokaryotic and eukaryotic cell types to inform upon environmental cues and cell physiology. Whole-cell bioreporter organisms engineered to emit bioluminescence have a distinct advantage in that the environmental background in which measurements are obtained is generally free from any interfering bioluminescent “noise,” thus delivering a superior signal-to-noise ratio (compared to fluorescent imaging) that allows for precise discrimination of bioluminescent signaling events. This level of sensitivity has proven especially useful in preclinical biomedical imaging applications where bioluminescent cells or tissues can be visualized directly within a living animal model, which subsequently launched a new research discipline referred to as in vivo bioluminescent imaging (BLI). The toolset central to in vivo BLI applications continues to expand as this technology becomes more empowering in its ability to decipher disease pathology and contribute toward novel

3

Engineering Autobioluminescent Eukaryotic Cells as Tools for. . .

59

therapeutic interventions. Autobioluminescence, defined as the self-generation of bioluminescence from a cell, represents one such contribution that advantageously enables cells to emit bioluminescence autonomously with no experimental intervention required for the external addition of substrates to drive bioluminescent reaction kinetics. This differs substantially from mainstream in vivo BLI techniques that rely upon genetics borrowed from the North American firefly, Photinus pyralis, and its luciferase protein FLuc, which requires the external addition of a luciferin substrate to generate bioluminescence at each desired time point. For in vivo BLI, this involves injection of the luciferin substrate into the animal prior to each scheduled bioluminescent measurement. For in vitro applications, such as microtiter plate-based assays, this involves the addition of the luciferin substrate to each microtiter plate well prior to each measurement. The ability to achieve imaging data that is longitudinal and/or continuous in time therefore is challenging under these constraints. Autobioluminescent reporter cells may provide a complementary strategy towards achieving true longitudinal imaging under both in vivo and in vitro experimental designs. The historical application of autobioluminescence from its bacterial foundations to its synthetic expression in lower eukaryotic and mammalian cells will be discussed in this chapter to provide the reader with a fundamental knowledgebase of autobioluminescent whole-cell bioreporter systems.

Whole-Cell Bioluminescent Bioreporters Autobioluminescence was being expressed in whole-cell bacterial bioreporters long before “autobioluminescence” became part of the bioimaging vocabulary. Using the bacterial luciferase lux operon derived from Vibrio, Aliivibrio, Photorhabdus, and Photobacterium bacteria, researchers bioengineered a vast number of whole-cell bacterial bioreporters for applications primarily focused on environmental monitoring and bioremediation process monitoring and control (Xu et al. 2013). With environmentally isolated bacteria naturally harboring genetic architectures capable of expressing catabolic pathways for the processing of contaminant chemicals, it was only a means of hijacking these genetic operons and reconfiguring them to instead generate bioluminescence, thereby creating a bioreporter organism capable of emitting light upon exposure to specific environmental contaminants or classes of environmental contaminants. The bacterial luciferase operon, consisting of the luxC, luxD, luxA, luxB, and luxE genes, provided everything the cell needed to self-generate bioluminescence in the presence of molecular oxygen and FMNH2 reactants scavenged from normal cellular metabolism (Fig. 1a). Bioreporters were also designed that contained only the luxA and luxB genes, which provided for the luciferase component of the bioluminescent reaction but without the ability to regenerate the long-chain aldehyde reactant normally realized by the luxC, luxD, and luxE gene products. These luxAB bioreporters therefore required external supplementation of aldehyde substrate to generate bioluminescence. Firefly luciferase was similarly integrated into bacterial bioreporters with the caveat that the bacteria likewise required supplemental

60

T. Xu et al. Bacterial luciferase

a

FMNH2 + O2 + RCHO

b

ATP + D-luciferin + O2

FMN + RCOOH + H2O + light (490 nm)

Firefly luciferase

CO2 + AMP + PPi + oxyluciferin + light (562 nm)

Fig. 1 (a) The chemical reaction for bacterial-based lux bioluminescence emission involves the production of blue-green light at a peak wavelength of 490 nm through the oxidation of a long chain fatty aldehyde in the presence of oxygen and reduced riboflavin phosphate with subsequent regeneration of the aldehyde substrate. (b) The chemical reaction for bioluminescence emission in the firefly (luc) system involves the oxidation of reduced luciferin in the presence of ATP and oxygen to generate CO2, AMP, PPi, oxyluciferin, and yellow-green light at a peak wavelength of 562 nm.

addition of the luciferin substrate to generate bioluminescence, albeit with the distinct advantage that the resulting quantum yield far surpassed that of any other characterized bioluminescent system (Widder and Falls 2014) (Fig. 1b). Besides bioluminescence, numerous whole-cell bioreporters were similarly designed using other reporter gene systems with, for example, fluorescent (i.e., green fluorescent protein) or colorimetric (i.e., β-galactosidase) signaling outputs (Xu et al. 2013). To understand the fundamentals of a whole-cell bioreporter and its application as an environmental monitor, we can use as an example a lux-based bacterial bioreporter engineered to detect mercury, referred to as ARL1, which was created within an Escherichia coli host cell (Dahl et al. 2011; Xu et al. 2015b). The occurrence of mercury in the environment derives from natural sources or is associated with human/anthropogenic activities like the combustion of fossil fuel and industrial waste. The release of mercury to subsurface and aquatic bodies can ultimately enter and biomagnify within the food chain and result in severe health issues, thus making its environmental detection critically important. To construct a whole-cell bioluminescent bioreporter for mercury, the innate ability of certain bacteria to resist and biotransform mercury compounds was exploited. These bacteria harbor a genetic operon referred to as mer, whose series of genes shape the protein machinery that drives mercury resistance and biotransformation (Barkay et al. 2003). The activation of the mer operon falls under the regulatory control of merR. For assembly of the ARL1 bioreporter, a 505 base pair sequence of this merR promotor/operator region was excised and ligated in front of the luxCDABE gene cassette from Photorhabdus luminescens and transformed into E. coli, thereby creating a bioreporter whose response to mercury is now directly linked to the selfgeneration of bioluminescent light. The intensity of bioluminescence generated by the E. coli ARL1 bioreporter upon exposure to a mercury-contaminated environmental sample correlates to the approximate concentration of mercury elements within the sample, and therefore serves as an easy-to-perform and rapid assay for determining contaminant presence. The bioreporter also reports on contaminant bioavailability; that is, the overall effect of the contaminant on a living system, which informs upon actual detrimental biological effects of a contaminant at particular exposure levels.

3

Engineering Autobioluminescent Eukaryotic Cells as Tools for. . .

61

The Transition from Bioluminescence to Autobioluminescence The high signal-to-noise ratio of bioluminescence relative to fluorescence rapidly established it as a preferred method for optical imaging (Close et al. 2011). Initially the firefly luciferase system, which with its single gene architecture and externally supplied chemical luciferin substrate, was the primary reporter gene employed for this purpose. This basic approach allowed investigators to perform high resolution imaging across a variety of species ranging from bacteria to small animal models (Thorne et al. 2010). In contrast, the autobioluminescent bacterial luciferase system was not as popular because of its relatively complex five gene genetic makeup and ability to only be expressed in prokaryotic hosts. Nonetheless, the lux cassette’s ability to autonomously modulate its output signal provided a significant advantage over the luciferin-dependent activation kinetics of the firefly luciferase system and lead to a series of improvements that have overcome these initial hurdles and popularized its usage. The advantages of the autobioluminescent approach were realized shortly after its first successful demonstration in E. coli when Engebrecht et al. showed that a cloned lux cassette could be placed under the control of a promoter of interest to observe transcriptional activation without necessitating external perturbation of the system (Engebrecht et al. 1983, 1985). Using this approach, autobioluminescence allowed for continuous, noninvasive monitoring while avoiding the potential complications associated with exogenous luciferin supplementation that were required by traditional bioluminescent reporters. This advantage, which coincided with an increased ease-of-use and decreased performance cost via removal of the luciferin addition step, proved to be popular among investigators and further drove adoption of autobioluminescence in place of bioluminescence for tasks such as tracking bacterial colonization of new niches (de Weger et al. 1991; Shaw and Kado 1986), environmental monitoring (King et al. 1990; Ripp et al. 2000), visualization of DNA damage (Belkin et al. 1997; Vollmer et al. 1997), and the evaluation of antibiotic efficacy (Francis et al. 2000). However, while autobioluminescence rapidly grew in popularity as an optical imaging approach within prokaryotic hosts, it was not similarly embraced within the eukaryotic community due to early reports that it was incapable of functioning within the compartmentalized physiological structure of eukaryotic cells (Sambrook and Russell 2001) and difficulties in transitioning the prokaryotic genetic structure to function with eukaryotic transcriptional machinery (Almashanu et al. 1990; Kirchner et al. 1989; Olsson et al. 1988). It was not until almost 20 years after the first autobioluminescent E. coli that the system was successfully modified to overcome these hurdles and the first autobioluminescent yeast strain (i.e., Saccharomyces cerevisiae) was developed (Gupta et al. 2003). This demonstration bolstered the prevalence of autobioluminescence by allowing its continuous monitoring, non-interaction, low cost, and improved ease-of-use attributes to be utilized for applications that could not be achieved using prokaryotic hosts, such as endocrine disruptor compound screening (Sanseverino et al. 2005). The key findings

62

T. Xu et al.

enabling this transition were the use of lux cassette genes from the terrestrial bacterium P. luminescens, whose gene products displayed increased thermostability at relevant eukaryotic temperatures, advances in synthetic gene circuit assembly that used internal ribosomal entry sites (IRESs) to express multiple lux genes from the same promoter within a eukaryotic host, and the inclusion of a complementary oxidoreductase gene ( frp) that shifted the cytosolic FMN:FMNH2 balance further towards the reductive orientation to better mimic the reductive environment of prokaryotic hosts (Close et al. 2012). The final step in popularizing autobioluminescence came when the lux cassette was further modified to function within mammalian cellular hosts. This process, which initially required the introduction of multiple plasmids containing human codon-optimized cassette genes (Close et al. 2010), was simplified to place all of the required genetic components under the control of a single promoter using viral 2A linker sequences that allowed the full cassette to be deployed similarly to the popular single gene firefly luciferase reporter construct (Xu et al. 2014) (Fig. 2). With these modifications, autobioluminescence was finally capable of performing all of the same functions as bioluminescence in all of the same organisms, allowing investigators to leverage its unique advantages regardless of their system of study.

Environmental Surveillance Using Autobioluminescent Yeast Bioreporters Bioluminescent whole-cell bioreporters expressing the lux operon provide a costeffective, rapid, and high-throughput means for assessing the bioavailability of environmental contaminants. This role has been mainly fulfilled by a large inventory of lux-based bioluminescent bacterial bioreporters that have been developed and applied to monitor various environmental conditions over the past two decades (Xu et al. 2013). Nevertheless, due to the growing needs for the evaluation of bioavailability and toxicological effects of environmental pollutants in humans and animals, there has been a pressing shift to the use of eukaryotic cells for this purpose. Yeast, as a single-cell eukaryote, stand out as ideal hosts for bioreporter development for the purpose of environmental surveillance. Yeast cells have the advantages of rapid growth and easy maintenance like their bacterial counterparts and ample genetic manipulation tools for reporter construction. Most importantly, their eukaryotic genetic and biochemical background opens the opportunity to monitor the interaction between environmental contaminants and eukaryote-specific cellular pathways. Consequently, unlike bacterial bioreporters that are commonly designed to detect the presence of contaminants of a given chemical structure, yeast-based environmental bioreporters focus on detecting specific toxicological activities that can be predictive of human and animal health impacts. The most common use of yeast bioreporters so far has been for detecting compounds capable of disrupting the vertebrate endocrine system. Shortly after the successful demonstration of lux expression in S. cerevisiae (Gupta et al. 2003), the first ever analyte-specific autobioluminescent yeast whole-cell bioreporter,

Engineering Autobioluminescent Eukaryotic Cells as Tools for. . .

Fig. 2 (a) To achieve the autobioluminescent phenotype under eukaryotic expression, the genes of the Photorhabdus luminescens bacterial lux operon (luxCDABE) and the Vibrio campbellii frp oxidoreductase gene were each human codon optimized and co-assembled into an open reading frame using viral 2A linker sequences to enable expression from a single user-defined promoter element. (b) Mammalian cells expressing the autobioluminescent phenotype can be imaged in vitro in high-throughput microtiter plate assays or in vivo in live animal models

3 63

64

T. Xu et al.

BLYES, was developed for rapid detection of estrogenic compounds (Sanseverino et al. 2005). This was achieved by engineering the human estrogen receptor signaling pathway in S. cerevisiae and coupling estrogen receptor-mediated transcriptional activation with lux expression. Specifically, the lux genes were split into two plasmids. The first plasmid harbored the luxA and luxB luciferase genes under the control of cis-regulating estrogen responsive elements (EREs), whereas the second plasmid conferred constitutive expression of the substrate-processing luxC, luxD, luxE, and frp genes. Exposure to estrogen or estrogenic compounds activated the estrogen receptor, which in turn activated ERE-regulated transcription of the luciferase genes. With the presence of constitutively expressed substrate-processing genes, upregulation of the luciferase genes ultimately resulted in dose-dependent induction of light output. The BLYES bioreporter has been shown to detect as low as 45 pM of the natural estrogen 17β-estradiol and produce a rapid response within 3–4 h, as compared to the 3 day assay time for its colorimetric (lacZ) counterpart YES reporter system (Sanseverino et al. 2005). In addition to natural and synthetic estrogens, the BLYES bioreporter also detects a variety of estrogenic compounds, including but not limited to pharmaceutical compounds, personal care products, industrial compounds, herbicides, fungicides, and plant-derived compounds (Sanseverino et al. 2009). An autobioluminescent yeast bioreporter for androgenic compounds, BLYAS, was developed using a similar strategy with human androgen receptor and androgen response elements, and has been demonstrated to report the presence of androgenic compounds in a dose-dependent manner (Eldridge et al. 2007; Sanseverino et al. 2009). Their easeof-use, low cost, and rapid response have made the BLYES and BLYAS bioreporters valuable tools for monitoring these emerging contaminants in the environment. For example, BLYES has been applied to evaluate the estrogenic potential of wastewater effluent (Wang et al. 2015), to survey the estrogenicity of surface and drinking water in the USA (Conley et al. 2017) and Brazil (Bergamasco et al. 2011), and to evaluate the estrogenic contribution of bisphenol analogs in sewage sludge in China (Ruan et al. 2015). Developed during the early stages of transitioning the lux cassette into eukaryotic cells, the first generation autobioluminescent yeast bioreporters (i.e., BLYES and BLYAS) relied upon the strategy of splitting the complete lux cassette into two expression vectors and using multiple promoters to regulate gene expression (Eldridge et al. 2007; Gupta et al. 2003; Sanseverino et al. 2005). This approach, albeit effective, was not ideal for reporter development due to inefficient reporter gene delivery and regulation. Recent efforts of integrating the lux cassette in mammalian cells have overcome these limitations and resulted in a streamlined expression strategy to arrange the full complement of lux genes under the regulation of a single promoter using viral 2A linkers for straightforward gene delivery and precise regulation (Xu et al. 2014). Taking advantage of this improvement in the mammalian cell work, a new autobioluminescent yeast bioreporter, S. cerevisiae BLYAhS, was recently developed for the detection of dioxin and dioxin-like compounds (Xu et al. 2018). BLYAhS cells express the requisite signaling proteins of

Engineering Autobioluminescent Eukaryotic Cells as Tools for. . .

3

a 1

10 µM TCDD 2

3

4

5 µM TCDD 5

6

7

1 µM TCDD 8

9

65

10

500 nM TCDD 11

12

A 100 nM TCDD

50 nM TCDD

10 nM TCDD

5 nM TCDD

1 nM TCDD

500 pM TCDD

100 pM TCDD

Vehicle control

B

C

25000

Relative light unit/second

b

20000

Well A1 (10 µM TCDD)

15000 10000 5000

Well C10 (Vehicle control) 0 0:00:00

2:00:00

4:00:00

6:00:00

8:00:00 10:00:00 12:00:00

Time

Fig. 3 An example of using the autobioluminescent yeast bioreporter in a high-throughput microtiter plate assay in a plate reader instrument. (a) The autobioluminescent yeast dioxin bioreporter BLYAhS cells are plated in a 96-well microtiter plate and treated with varying concentrations of 2,3,7,8-tetrachlorodibenzodioxin (TCDD). Bioluminescent output is measured every 10 min for 12 h. The curves represent the kinetic data throughout the 12-h exposure in each well. (b) A side-by-side comparison of light readout between the 10 μM TCDD treatment (Well A1) and the vehicle control (Well C10) shows the near-continuous data acquisition while the reporter cells are exposed

the human aryl hydrocarbon receptor pathway and the 2A-linked, yeast codonoptimized lux cassette under the control of xenobiotic response elements. Unlike in BLYES and BLYAS where only the luxAB luciferase genes are regulated, exposure of BLYAhS to dioxin or dioxin-like compounds activates transcription of the full lux cassette. The more coordinated control of lux reporter expression using the single promoter strategy leads to lower background signal in the absence of target analyte. BLYAhS has been demonstrated to detect the presence of 2,3,7,8-tetrachlorodibenzodioxin (TCDD) at >500 pM within 4 h as well as a variety of dioxin-like compounds (Xu et al. 2018) (Fig. 3). Although not as sensitive as other mammalian cell-based bioreporters, the BLYAhS bioreporter provides superior response time and significant cost-savings and is a potential candidate for highthroughput Tier I screening of environmental samples.

66

T. Xu et al.

Biomedical Surveillance Using Autobioluminescent Human Cell Lines Autobioluminescence has only recently become routinely applied towards biomedical surveillance in the time since the lux cassette was adapted to function within mammalian cellular hosts. However, even before this time, it found limited use as a facile method for tracking bacterial colonization in small animal models. In one example of this approach, Contag et al. used autobioluminescent strains of Salmonella typhimurium with differing levels of virulence to characterize the factors that lead to progressive versus persistent or abortive infections (Contag et al. 1995). Using this approach, it was possible, for the first time, to track infections in real-time and monitor the in vivo efficacy of antibiotic treatment. This allowed pathogenesis research to move beyond ex vivo assays and cell-culture correlates to function within whole animal model systems that were capable of producing significantly more informative data. As autobioluminescence transitioned to function in mammalian cell models, its utility further expanded. At a basic level, the autonomous signal generation of the autobioluminescent system provided a way for investigators to track a single subpopulation of cells within a mixed culture and assay their metabolic activity in realtime. This ability is exemplified in a 2015 study, where E. coli O157:H7 infection was modeled using an autobioluminescent human cell culture system (Xu et al. 2015a). The background autofluorescence of the system precluded the use of fluorescent imaging techniques for population-level observation, while traditional bioluminescent methods were hindered by the release of activating metabolites from the co-cultured bacterial cells during the requisite cell lysis step prior to exogenous substrate addition. By using a constitutively autobioluminescent human cell model, the investigators were able to selectively track the metabolic activity and survivability of this subpopulation within the mixed consortia and evaluate how their survival dynamics responded to the timing of bacterial clearance. Similarly, because the autobioluminescent output of human cell models expressing the lux cassette genes self-modulates in response to metabolic activity dynamics (Close et al. 2010), physically separated populations of co-cultured cells can be grown under the “organism-on-chip” format to monitor how their interactions influence their physiology. This process can be similarly performed using traditional bioluminescent imaging approaches but is handicapped in this application because the need for sample destruction concurrent with exogenous substrate addition necessitates foreknowledge of the timing at which these interactions will manifest (Nouri et al. 2015). Using autobioluminescence, however, each of the interacting cell types under study can be simultaneously monitored to ascertain the presence, timing, magnitude, and duration of these effects in a single assay (Xu et al. 2017). In addition to these types of procedures, which take advantage of the continuous, self-modulated signal output of autobioluminescent systems, autobioluminescence can also be applied for biomedical surveillance of compound bioavailability and gene activation responsiveness similarly to how it was initially employed in

3

Engineering Autobioluminescent Eukaryotic Cells as Tools for. . .

67

prokaryotic systems. In this application, the lux cassette genes are placed under the control of a promoter that is naturally activated in response to the presence of a compound or is part of a transcriptional activity cascade representative of a specific metabolic process. This provides a novel method for identifying the timing of transcriptional responses while concurrently determining human cell compound bioavailability. Under these conditions, it is possible to make more informed predictions about the toxicity of compounds and to develop a better understanding of what those compounds effects are following exposure. Interestingly, there are recent indications that autobioluminescent systems may be capable of providing additional information along with these traditional data, as it has also been shown that autobioluminescent activity dynamics correlate with the activation of common phase I, and in some cases, phase II detoxification pathway enzyme biomarkers in the HepG2 liver and T-47D breast cancer cell lines (Xu et al. 2017). Given the additional expenses that are incurred due to the requisite use of human cell line and small animal model systems for biomedical surveillance applications, this wide utility of autobioluminescence, its streamlined usage conditions, and its ability to track samples for extended time periods when working with novel or poorly understood activation events all provide significant advantages relative to traditional bioluminescent approaches. It is therefore likely that the use of autobioluminescence for biomedical surveillance will continue to grow as the system is further employed to address new applications.

Conclusions and Future Directions The application of bioluminescence as a signaling endpoint in prokaryotic and eukaryotic bioreporter cells has evolved as a robust and well-vetted technology that is used worldwide among a diversity of scientific disciplines that range from environmental monitoring to biomedical imaging. With autobioluminescent genetic architectures now contributing to the inventory of bioluminescent whole-cell bioreporter developmental strategies, especially among mammalian cell hosts, investigators have exciting new possibilities to explore in vivo biology under performance metrics that are truly longitudinal with real-time data acquisition outputs. For in vitro assays, autobioluminescence provides new opportunities to achieve higher throughput endpoints with near unlimited continuous data downloads. The enriched informational knowledgebase obtained through the application of autobioluminescent bioreporter cells will contribute substantially to our understanding of human disease status and its therapeutic management and assist in elucidating the myriad environmental factors that impinge upon and influence human health. As state-of-the-art precision medicine, theranostics, and organ/ human-on-chip diagnostic platforms advance toward real-world application scenarios, autobioluminescence, in conjunction with an ever-expanding technological toolbox of complementary bioimaging methods, will be key to our ability to visualize and comprehend the dynamic physiology of the cell.

68

T. Xu et al.

Acknowledgments The authors acknowledge research funding support by the US National Institutes of Health, National Institute of Environmental Health Sciences, under award numbers NIEHS-1R15ES023979-01, NIEHS-1R43ES022567-01, and NIEHS-2R44ES022567-02, the US Department of Agriculture under award number 2015-33610-23598, the University of Tennessee Institute for a Secure and Sustainable Environment (ISSE), and the International Research Support Initiative Program of the Higher Education Commission of Pakistan.

References Almashanu S, Musafia B, Hadar R, Suissa M, Kuhn J (1990) Fusion of luxA and luxB and its expression in Escherichia coli, Saccharomyces cerevisiae and Drosophila melanogaster. J Biolumin Chemilumin 5:89–97 Barkay T, Miller SM, Summers AO (2003) Bacterial mercury resistance from atoms to ecosystems. FEMS Microbiol Rev 27:355–384. https://doi.org/10.1016/s0168-6445(03)00046-9 Belkin S, Smulski DR, Dadon S, Vollmer AC, Van Dyk TK, LaRossa RA (1997) A panel of stressresponsive luminous bacteria for the detection of selected classes of toxicants. Water Res 31:3009–3016 Bergamasco AM et al (2011) Bioluminescent yeast estrogen assay (BLYES) as a sensitive tool to monitor surface and drinking water for estrogenicity. J Environ Monit 13:3288–3293 Close DM, Patterson SS, Ripp S, Baek SJ, Sanseverino J, Sayler GS (2010) Autonomous bioluminescent expression of the bacterial luciferase gene cassette (lux) in a mammalian cell line. PLoS One 5:e12441. https://doi.org/10.1371/journal.pone.0012441 Close DM, Hahn RE, Patterson SS, Baek SJ, Ripp SA, Sayler GS (2011) Comparison of human optimized bacterial luciferase, firefly luciferase, and green fluorescent protein for continuous imaging of cell culture and animal models. J Biomed Opt 16:047003. https://doi.org/10.1117/ 1.3564910 Close D et al (2012) The evolution of the bacterial luciferase gene cassette (lux) as a real-time bioreporter. Sensors 12:732–752. https://doi.org/10.3390/s120100732 Conley JM et al (2017) Occurrence and in vitro bioactivity of estrogen, androgen, and glucocorticoid compounds in a nationwide screen of United States stream waters. Environ Sci Technol. https://doi.org/10.1021/acs.est.6b06515 Contag CH, Contag PR, Mullins JI, Spilman SD, Stevenson DK, Benaron DA (1995) Photonic detection of bacterial pathogens in living hosts. Mol Microbiol 18:593–603 Dahl AL, Sanseverino J, Gaillard J-F (2011) Bacterial bioreporter detects mercury in the presence of excess EDTA. Environ Chem 8:552–560. https://doi.org/10.1071/en11043 de Weger LA, Dunbar P, Mahafee WF, Lugtenberg BJJ, Sayler GS (1991) Use of bioluminescence markers to detect Pseudomonas spp. in the rhizosphere. Appl Environ Microbiol 57:3641–3644 Eldridge ML, Sanseverino J, Layton AC, Easter JP, Schultz TW, Sayler GS (2007) Saccharomyces cerevisiae BLYAS, a new bioluminescent bioreporter for detection of androgenic compounds. Appl Environ Microbiol 73:6012–6018 Engebrecht J, Nealson K, Silverman M (1983) Bacterial bioluminescence: Isolation and genetic analysis of functions from Vibrio fischeri. Cell 32:773–781 Engebrecht J, Simon M, Silverman M (1985) Measuring gene expression with light. Science 227:1345–1347. https://doi.org/10.1126/science.2983423 Francis KP, Joh D, Bellinger-Kawahara C, Hawkinson MJ, Purchio TF, Contag PR (2000) Monitoring bioluminescent Staphylococcus aureus infections in mice using a novel luxABCDE construct. Infect Immun 68:3594–3600 Gupta RK, Patterson SS, Ripp S, Simpson ML, Sayler GS (2003) Expression of the Photorhabdus luminescens lux genes (luxA, B, C, D, and E) in Saccharomyces cerevisiae. FEMS Yeast Res 4:305–313

3

Engineering Autobioluminescent Eukaryotic Cells as Tools for. . .

69

King JMH et al (1990) Rapid, sensitive bioluminescence reporter technology for naphthalene exposure and biodegradation. Science 249:778–781 Kirchner G, Roberts JL, Gustafson GD, Ingolia TD (1989) Active bacterial luciferase from a fused gene expression of a Vibrio harveyi luxAB translational fusion in bacteria, yeast and plant cells. Gene 81:349–354 Nouri FS, Wang X, Hatefi A (2015) Genetically engineered theranostic mesenchymal stem cells for the evaluation of the anticancer efficacy of enzyme/prodrug systems. J Control Release 200:179–187. https://doi.org/10.1016/j.jconrel.2015.01.003 Olsson O, Koncz C, Szalay AA (1988) The use of the luxA gene of the bacterial luciferase operon as a reporter gene. Mol Gen Genet 215:1–9 Ripp S et al (2000) Controlled field release of a bioluminescent genetically engineered microorganism for bioremediation process monitoring and control. Environ Sci Technol 34:846–853. https://doi.org/10.1021/es9908319 Ruan T, Liang D, Song SJ, Song MY, Wang HL, Jiang GB (2015) Evaluation of the in vitro estrogenicity of emerging bisphenol analogs and their respective estrogenic contributions in municipal sewage sludge in China. Chemosphere 124:150–155. https://doi.org/10.1016/j. chemosphere.2014.12.017 Sambrook J, Russell DW (2001) Molecular cloning: a laboratory manual, 3rd edn. Cold Spring Harbor Laboratory Press, Cold Spring Harbor Sanseverino J et al (2005) Use of Saccharomyces cerevisiae BLYES expressing bacterial bioluminescence for rapid, sensitive detection of estrogenic compounds. Appl Environ Microbiol 71:4455–4460 Sanseverino J, Eldridge ML, Layton AC, Easter JP, Yarbrough J, Schultz TW, Sayler GS (2009) Screening of potentially hormonally active chemicals using bioluminescent yeast bioreporters. Toxicol Sci 107:122–134. https://doi.org/10.1093/toxsci/kfn229 Shaw JJ, Kado CI (1986) Development of a Vibrio bioluminescence gene set to monitor phytopathogenic bacteria during the ongoing disease process in a nondisruptive manner. Biotechnology 4:560–564. https://doi.org/10.1038/nbt0686-560 Thorne N, Inglese J, Auldl DS (2010) Illuminating insights into firefly luciferase and other bioluminescent reporters used in chemical biology. Chem Biol 17:646–657. https://doi.org/ 10.1016/j.chembiol.2010.05.012 Vollmer AC, Belkin S, Smulski DR, van Dyk TK, Larossa RA (1997) Detection of DNA damage by use of Escherichia coli carrying recA::lux, uvrA::lux, or alkA::lux reporter plasmids. Appl Environ Microbiol 63:2566–2571 Wang J, Eldridge M, Menn F-M, Dykes T, Sayler G (2015) Standardized application of yeast bioluminescent reporters as endocrine disruptor screen for comparative analysis of wastewater effluents from membrane bioreactor and traditional activated sludge. Ecotoxicology 24:2088–2099. https://doi.org/10.1007/s10646-015-1556-z Widder EA, Falls B (2014) Review of bioluminescence for engineers and scientists in biophotonics. IEEE J Sel Top Quantum Electron 20. https://doi.org/10.1109/jstqe.2013.2284434 Xu T, Close DM, Sayler GS, Ripp S (2013) Genetically modified whole-cell bioreporters for environmental assessment. Ecol Indic 28:125–141. https://doi.org/10.1016/j. ecolind.2012.01.020 Xu T, Ripp S, Sayler G, Close D (2014) Expression of a humanized viral 2A-mediated lux operon efficiently generates autonomous bioluminescence in human cells. PLoS One 9:e96347. https:// doi.org/10.1371/journal.pone.0096347 Xu T, Marr E, Lam H, Ripp S, Sayler G, Close D (2015a) Real-time toxicity and metabolic activity tracking of human cells exposed to Escherichia coli O157:H7 in a mixed consortia. Ecotoxicology 24:2133–2140. https://doi.org/10.1007/s10646-015-1552-3 Xu X, Oliff K, Xu T, Ripp S, Sayler G, Zhuang J (2015b) Microbial availability of mercury: Effective detection and organic ligand effect using a whole-cell bioluminescent bioreporter. Ecotoxicology 24:2200–2206. https://doi.org/10.1007/s10646-015-1553-2

70

T. Xu et al.

Xu T, Conway M, Frank A, Ripp S, Sayler G, Close D (2017) Co-cultured continuously bioluminescent human cells as bioreporters for the detection of prodrug therapeutic impact preand post-metabolism. Sensors 17:e2827. https://doi.org/10.3390/s17122827 Xu T, Young A, Marr E, Sayler G, Ripp S, Close D (2018) A rapid and reagent-free bioassay for the detection of dioxin-like compounds and other aryl hydrocarbon receptor (AhR) agonists using autobioluminescent yeast. Anal Bioanal Chem 410:1247–1256. https://doi.org/10.1007/ s00216-017-0780-1

4

Biosensors of the Well-being of Cell Cultures Karen Marie Polizzi

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Current Paradigms for Measurement in Bioproduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sensor Architectures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cell Stress Signaling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I Am Well Generally . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Stress from the External Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Stress from Internal Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Biosensors of Metabolites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions and Future Directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

72 74 74 77 77 78 80 81 83 84

Abstract

Genetically-encoded biosensors offer advantages over traditional analytical technologies in biological manufacturing. Various biosensor architectures can be employed based on activation of transcription, translation, or folding of a reporter. Such biosensors can be used in small volume, high throughput culture devices to provide a more complete picture of cellular physiology. In addition, because of their simple read out, genetically encoded biosensors can be employed for monitoring during production and offer the opportunity for increased measurement density. This chapter reviews current applications of biosensors to measure cellular health and well-being including designs for detecting the activation of cell stress responses as well as sensing of critical metabolites. Many of these designs have been demonstrated to be relevant to bioprocess engineering or cell line selection, but to date very few have been adopted by industry. In future, ensuring that biosensors function robustly under industrial conditions is K. M. Polizzi (*) Department of Chemical Engineering, Imperial College Centre for Synthetic Biology, London, UK e-mail: [email protected] © Springer Nature Switzerland AG 2022 G. Thouand (ed.), Handbook of Cell Biosensors, https://doi.org/10.1007/978-3-030-23217-7_119

71

72

K. M. Polizzi

a priority, as is designing systems that are compatible with industrial workflows. In the long term, engineering systems that move beyond sensing, to include the ability to adjust the environment of the culture based on signals, will allow biotechnology to fully capitalize on the advantages that genetically encoded biosensors can bring. Keywords

Bioprocess engineering · Bioreactor · Biosynthesis · Metabolism · Cell stress response · Recombinant protein production · Metabolic engineering

Introduction Modern biotechnology harnesses the power of biological systems to do complex chemical reactions like the synthesis of polypeptides, nucleic acids, and chiral molecules by using living organisms as “cell factories.” Examples include the production of medicines such as antibody therapies and vaccines (Aw et al. 2018; Sou et al. 2015), food supplements (Bermejo et al. 2018), and even monomers for the synthesis of plastics (Yang et al. 2018a). Developing a new biological production process involves a series of steps. First, the host organisms with the correct biosynthetic capabilities are identified from natural sources or, more frequently, genetic engineering is used to create augmented organisms that can make the product of interest in high yield (Fig. 1a). Next, the optimal conditions for maximum production are identified, usually through empirical screening of medium formulation, feeding, and cultivation strategies (Fig. 1b). Iterations between genetic engineering and bioprocess development are common. Regardless of the product that is produced, several key parameters are of interest that can be thought of as indicators of overall cellular “well-being.” One such

a

b V1 V5

1

V2 2

V3 V4 V3 3

- - ++ + + +++

+ ++ -+++ +

Fig. 1 Bioprocess development workflow. (a) Screening is used to identify potential hosts that can synthesize the product of interest in high yield. This may involve screening different types of cells or genetic modifications of the same base strain. (b) Once a strain has been chosen, empirical screening of process conditions is used to maximize yield. There may be iterations between strain identification and process development

4

Biosensors of the Well-being of Cell Cultures

73

indicator is the availability of key nutrients (e.g., the primary carbon and nitrogen sources, product-specific building blocks); deficiencies in which will lead to a reduced rate of cell growth and/or product formation. Therefore, changes in medium formulation, feeding strategies, or cell line engineering to increase the efficiency of metabolism may be indicated. Similarly, the accumulation of waste products (e.g., products of overflow metabolism, ammonia) can lead to harmful effects on cell growth or productivity and so mitigation strategies may need to be developed. Finally, natural signaling pathways that can indicate cellular stress due to oxidative damage, mechanical forces, high protein folding load, or osmolarity may be activated under some conditions. Understanding and avoiding these responses can lead to cultures with more stable production and less evolutionary drive toward gene silencing. Particularly in early stages of process development, small volume, high throughput culture systems are used to limit the volumes under cultivation to maximize the number of cell lines and conditions that can be screened for a fixed cost. However, the small culture volumes typically limit the type of analyses conducted to those that are noninvasive or require very small sample sizes, which provides incomplete information on the overall health of cells in individual conditions. Understanding the link between bioprocessing conditions and cellular physiology is becoming increasingly important as regulators in the pharmaceutical and biopharmaceutical industries have suggested that a “quality-by-design” approach should be adopted (Yu et al. 2014). Increasingly, genetically encoded biosensors have been proposed as attractive option for real-time, high throughput data generation, because they remove the need for sampling. However, the potential utility of biosensors extends beyond high throughput screening. Even once a manufacturing process has been developed, cells are typically monitored during production to ensure product quality and batch-to-batch reproducibility (Pais et al. 2014; Zhao et al. 2015). Currently, most of these analyses are off-line, which introduces a delay between sampling and any corrective actions that need to be applied to the culture. On the other hand, genetically encoded biosensors can be used online and noninvasively. Moreover, the simple readouts of such biosensors would enable increased density of measurements and reduce analysis times. Therefore, there is significant scope for using biosensors as analytical technology during production. The aim of this chapter is to provide an overview of the application of biosensors to measure cellular “well-being” in biological manufacturing, with a specific focus on genetically-encoded biosensors. First, the standard measurement practices in industry, which are currently based largely on other types of analyses, are reviewed. Next, a broad overview of the different architectures of genetically-encoded biosensors is described. Then, different examples of previously reported sensors for important indicators of cellular well-being are discussed including examples from both cellular stress detection and the measurement of critical metabolites. Finally, the broad trends in the field are summarized with a view to identifying future directions in research to develop new biosensors relevant to bioprocess development and monitoring.

74

K. M. Polizzi

Current Paradigms for Measurement in Bioproduction At both small and large scale, physical properties such as temperature, pH, and dissolved oxygen content are measured using electrochemical or optical sensors either as large probes within a bioreactor or small spot-based technology in high throughput culture devices (Demuth et al. 2016). At larger scales, e.g., in shake flasks or bioreactors, key metabolites can be measured from spent medium samples using automated biosensor devices (Dzyadevych et al. 2008). A sample is aspirated into the device and a combination of amperometric, potentiometric, and photometric sensors are used to simultaneously measure 2–16 analytes (depending on the device). The analytes measured are critical nutrients, waste metabolites, and dissolved gases, such as glucose, glutamine, lactate, ammonia, and pO2, which are known to be important factors in cell growth and productivity. The main advantage of such analyzers is the ability to obtain multiple measurements from a single sample. However, the sample volumes required are large (typically 0.25–1 mL), which means they cannot be applied to the high throughput screening phase of process development. Moreover, to obtain information on intracellular metabolites, cellular metabolism must be quenched, the cells subjected to lysis and then the samples analyzed, meaning that cells are destroyed during the analysis. Nonetheless, the development of new electrochemical and optical biosensors is still a very active area of research (recently reviewed in Dekker and Polizzi (2017), Ma et al. (2018), McLamore and Porterfield (2011) Yang et al. (2018b)). In terms of cell stress responses, measurements are typically limited to assessing the total cell number and the percentage of viable cells using automated cell counting devices along with staining (Chen et al. 2017). Newer methodologies have been developed, which can differentiate between live, dead, and apoptotic cells, potentially providing early warning about inadequate conditions (Merck Millipore 2018). Various additional analyses can be conducted using flow cytometry in combination with fluorescent dyes or antibodies to probe-specific traits. Automated inline sampling and flow cytometry has been developed (Brognaux et al. 2013a; Delvigne and Goffin 2014; Kuystermans et al. 2012, 2016), although the analysis usually still results in destruction of the sample.

Sensor Architectures A biosensor refers to a device that uses a biological recognition element to detect an analyte of interest (Goers et al. 2013). Interactions between the analyte and the recognition element are transformed into a readable signal through a transducer element. There are many different varieties of biosensor architectures. They vary by the type of recognition element (e.g., proteins or nucleic acids) and whether transduction elements are biological or nonbiological. In some types, such as wholecell biosensors, the recognition and transduction elements are entirely biological, whereas other types of designs couple the biological recognition to a physical or chemical transducer. This chapter focuses on genetically-encoded biosensors, which

4

Biosensors of the Well-being of Cell Cultures

75

Signaling cascade Promoter

Reporter

Transcription

Protein

mRNA

DNA

Signal

+

Promoter

Reporter Protein

mRNA

DNA

Signal

+

Translation Protein

Signal

Protein (Constitutive)

Signal

mRNA (Constitutive) +

Post-translation

+

Protein (Constitutive)

Signal

Fig. 2 Different types of biosensor architectures for genetically encoded biosensors. Interactions between the analyte of interest can activate transcription, translation, or posttranslational mechanisms in the cell that lead to changes in the signal of a reporter protein. (From top): Transcription-based biosensors can rely on a membrane bound receptor protein, which interacts with the analyte leading to the activation of a signaling cascade. The result is the production of the mRNA for the reporter, which then undergoes translation and folding to produce the signal. Alternatively, the analyte can interact with a repressor or activator protein and change its affinity for DNA, leading to transcription, translation, and then folding. (Middle): Translation-based biosensors rely on constitutively expressed mRNA whose translation is influenced by the presence of the analyte. The result is the production of the reporter protein and then signal production. (Bottom): Posttranslational mechanisms rely on reporter proteins that are constitutively produced reporter proteins whose signal is dependent on the presence of the analyte

are genetic circuits housed in engineered cells or cell extracts that contain the DNA instructions for a mechanism to sense the analyte of interest and link its detection to the expression of a reporter such as a fluorescent protein or enzyme that catalyzes a fluorescent, colorimetric, or luminescent output (Fig. 2). Genetically-encoded biosensors often exploit natural macromolecules that cells use to sense their surrounding environment and adjust gene expression to maintain homeostasis under changing conditions. This includes components from natural signaling pathways, protein transcriptional repressors and activators, or riboswitches. Engineered proteins and nucleic acids can also be developed into sensing circuits that are orthogonal to the native cellular machinery in an effort to reduce crosstalk and sensor interference. Overall, the interaction of the analyte with the recognition element may initiate transcription or translation of the reporter protein or the output of a constitutively expressed protein may be influenced via a posttranslational mechanism (e.g. Förster Resonance Energy Transfer). With the advent of synthetic biology methodology, new biosensors can now be developed in a systematic and rational way (reviewed in Bradley and Wang (2015), De Paepe et al. (2017), Liu et al. (2015a), Scognamiglio et al. (2015), Shi et al. (2018), Slomovic et al. (2015)).

76

K. M. Polizzi

Biosensors based on the initiation of transcription usually rely on protein recognition elements such as transmembrane receptors (for analytes that cannot cross the cell membrane) or repressors/activators (for analytes that can cross the cell membrane) (Goers et al. 2013). For transmembrane receptors, interaction of the analyte and the extracellular protein recognition element leads to conformational changes that result in activation of an intracellular signaling cascade, usually mediated by phosphorylation of a transcription factor. For repressors and activators, the analyte interacts directly with the transcription factor, leading to a conformational change that results in a different affinity for a DNA binding element. Thus, the transcription factor will either release from a promoter thereby relieving steric hindrance or bind and recruit RNA polymerase. In all cases, the presence of the analyte results in the expression of genes downstream of a regulated promoter. Therefore, a basic biosensor can be built from a copy of this regulated promoter fused to the gene for a reporter protein. In practice, designs are usually more sophisticated and may include co-expression of the receptor or transcription factor in a finely tuned quantity or building a synthetic promoter from DNA binding elements juxtaposed around a minimal nonregulated promoter. Regulated promoters are easily identified from transcriptomics studies and therefore, at present, transcription-based biosensors are the most ubiquitous type of genetically encoded biosensor. A different type of biosensor design relies on regulated translation of a constitutively produced mRNA encoding the reporter protein. Here, nucleic acid is used as the detection element and interactions between the analyte and the RNA influence the translation or stability of the RNA, affecting the amount of reporter protein that is made. This mechanism occurs naturally in some metabolic pathways in cells (Tapsin et al. 2018), but artificial RNA molecules that can bind an analyte of interest can also be selected via in vitro evolution (Ellington and Szostak 1990; Tuerk and Gold 1990). Using these in a biosensor involves determining where to place the detection element to control the translation of the mRNA. The performance of the biosensor can also be affected by the overall production level of the mRNA using the constitutive promoter. Finally, biosensors can be built from constitutively expressed proteins whose signal output changes in response to the presence of the analyte of interest. The most common design that relies on constitutive expression of proteins whose output signal changes in response to the presence of the analyte is based on Forster Resonance Energy Transfer (FRET). Here, a fusion protein is constructed from two fluorescent proteins connected by a ligand binding domain such that when the analyte is bound the distance between the fluorophores changes, leading to a difference in the fluorescence spectrum obtained. Thus, the approximate amount of analyte can be determined by measuring how much FRET is occurring. There are also additional designs possible, which involve a single fluorescent protein whose spectral properties change in the presence of the analyte such as the redox or pH-sensitive fluorescent proteins (Hanson et al. 2004; Miesenbock et al. 1998). One of the key advantages of genetically-encoded biosensors is that, when expressed within the cell producing the product of interest, they can provide realtime, noninvasive, and potentially nondestructive information about intracellular

4

Biosensors of the Well-being of Cell Cultures

77

processes. In particular, when fluorescent proteins are used as reporters, the output signal can easily cross the cell membrane without the need for sampling. Therefore, genetically-encoded biosensors can also be used in small volume, high throughput culture devices such as those used in process development. This allows a better understanding of the effects of process conditions and/or genetic modifications on the production capabilities of the cells, which in turn can be used to develop new strategies to increase productivity.

Cell Stress Signaling Biosensors can also be used to understand the effect of bioprocessing conditions on cellular stress responses and to monitor cells during production to prevent the occurrence of stress. Many signaling pathways exist to detect and counteract perturbations to cellular homeostasis that, if left unchecked, would lead to cell damage or death. These include mechanisms to cope with changes to the external environment (extrinsic stress) such as oxygen depletion, mechanical stress, and fluctuations in pH or temperature, as well as mechanisms to cope with the overloading of cellular machinery (intrinsic stress) from high protein folding load, oxidative damage, etc. Activation of stress response not only indicates conditions that can be potentially damaging to cells, it also can divert cellular resources away from growth and product formation toward coping mechanisms to deal with stress, leading to undesired decreases in bioprocess performance. Overall, biosensors of cellular stress can be useful for establishing the limits of various operating parameters beyond which cells cannot maintain high productivity.

I Am Well Generally Stress leads to global changes in cellular physiology that can negatively impact transcription and translation of all proteins. Therefore, one elegant strategy for detecting cellular well-being is to measure the expression of a constitutively expressed reporter protein. If the environment is suitable, the expression of the reporter should remain constant. However, if the expression decreases, this indicates that the current environment has a negative impact on the cell and suggests that process conditions need to be altered to restore homeostasis. Biosensors of global well-being are simple to apply and tolerated by most cell types (Carlquist et al. 2012; Randers-Eichhorn et al. 1997; Zhang and Yang 2011). However, they are relatively nonspecific, and it can be difficult to determine which environmental conditions need to be modified in order to restore cellular well-being. An interesting alternative application of a biosensor of global well-being is to use it to probe the effect of genetic modifications on cells to find alternative designs that have less metabolic burden (Ceroni et al. 2015). Traditionally, these effects are screened by analyzing the effect of different genetic modifications on growth rate. However, a biosensor of well-being can provide additional

78

K. M. Polizzi

information and also be used for high throughput screening of alternative designs using the fluorescent output. Choosing a design with lower metabolic burden should decrease the evolutionary pressure toward silencing the genetic modification in longterm culture. A related strategy is to develop biosensors to detect apoptosis or programmed cell death. These biosensors are specifically useful for mammalian cell cultures, which can undergo apoptosis due to a number of factors including nutrient starvation, hypoxia, shear stress, pH gradients, and high osmolarity (Krampe and Al-Rubeai 2010). Most apoptosis biosensors developed to date detect one of the enzymatic activities in the cascade toward cell death, e.g., caspase activity (Buschhaus et al. 2018; Chiang and Truong 2005; Zhang et al. 2013) or the activity of kinases such as JNK (Fosbrink et al. 2010) or ROCK (Li et al. 2017a). Because they are nonspecific, apoptosis biosensors can be thought of as indicating that cells are experienceing generally poor conditions but cannot pinpoint what should be changed in order to rectify the situation.

Stress from the External Environment Cells are affected by their external environment and must adjust their behavior to survive and maintain/restore homeostasis under changing conditions. Therefore, cellular machinery that responds to detrimental culture conditions in the external environment also exists (Fig. 3). This includes changes in pH, temperature, pressure or mechanical stress, and low oxygen availability, among other responses. Every organism has an optimal range of operating temperature, outside of which additional signaling pathways are activated to counteract detrimental effects to protein folding and other essential physiological processes (the heat- and cold-shock responses to deal with the effects of temperatures that are too high or too low, respectively) (Lim and Gross 2011; Sonna et al. 2002). These responses are primarily transcriptional in nature and the promoters driving the expression of genes involved in these responses can be used as genetically encoded biosensors to report when the cell has experienced a temperature outside of the normal operating range. Similarly, intracellular pH is maintained within a narrow range to prevent detrimental effects on protein structure even when the extracellular pH varies. The mechanisms for sensing and maintaining this can also be exploited for the development of biosensors. In fact, in one example, researchers made a biosensor that responded to a combination of temperature and pH (Hoynes-O’Connor et al. 2017). Although electrical pH and temperature probes are established technology that can report on these phenomena in a reliable manner across all scales, it is interesting to note that genetically-encoded biosensors can be made for virtually any property of interest. In addition, these gene circuits serve as a basis for more sophisticated designs such as the development of self-regulating processes (see “Conclusions and Future Directions” below) or biosensor designs that are more accurate because they combine more than one input signal to develop the output.

4

Biosensors of the Well-being of Cell Cultures

79

Temperature

pH Mechanical Stress

Hypoxia

Protein Folding Load Temperature

pH Oxidative Damage Mechanical Stress

Hypoxia

Protein Folding Load Oxidative Damage

Fig. 3 Different stresses that can affect cells in a bioreactor. Cells can experience stress from both external and internal factors. External factors include fluctuations in pH or temperature, mechanical stress from impellers, or a lack of access to oxygen (hypoxia). Internal factors include overload of the protein folding capacity or the production of reactive oxygen species, leading to oxidative damage

Within a bioreactor, cells can experience mechanical stress from air sparging and impellers (stirrers) operating at high speed to keep a well-mixed environment. Mammalian cells are particularly subject to damage because, unlike microbes, they lack a rigid cell wall for support and therefore are more prone to deformation under flow fields. High levels of mechanical stress are also experienced by some natural cell types in vivo including airway and aortic endothelial cells, muscles cells, and bone, which leads to changes in their morphology and secretion of proteins (Lim et al. 2006). Therefore, promoters that control the response of natural systems to mechanical deformation can be used as the basis for the development of mechanical stress biosensors (Seefried et al. 2010). Alternatively, by comparing a larger number of mammalian promoters that regulate gene expression in response to mechanical strain, researchers were able to identify the sequence of small DNA binding elements called “shear stress response elements” where the transcription factors controlling the stress response actually bind (Silberman et al. 2009). These can be used to develop synthetic promoters that may have less crosstalk with the other components of the cell but can still serve as the basis of mechanical stress biosensors. A completely different approach is to measure the strength of the forces directly, e.g., using FRET biosensor (Cost et al. 2015), or to measure the rate of GFP leakage from

80

K. M. Polizzi

cells harboring a biosensor of global well-being as a measure of membrane integrity (Brognaux et al. 2013b; Carlquist et al. 2012). Except for fermentation processes to produce small molecules like ethanol, most bioprocesses are run under aerobic conditions. However, providing sufficient oxygen to rapidly dividing cells can be a challenge due to the low solubility of gases in water-based media. This can be a particular problem for high density cultures or when using cells with high oxygen demand such as the methylotrophic yeast, Pichia pastoris. The detection of hypoxia can indicate it is necessary to change bioprocessing conditions, e.g., increase sparging rates of gas, change medium formulation, or use different cultivation strategies to reduce oxygen demand. As with previous stress signaling examples, the majority of such biosensors are derived from promoters that control the natural hypoxic response in cells either in their natural state or as synthetic promoters (Cilek et al. 2011; Garcia et al. 2009; Liu et al. 2005; Passoth et al. 2003). It is also important to note that most of the commonly used fluorescent protein reporters require oxygen to produce their fluorescence signal and so a different reporter protein such as those with fluorescence derived from flavin binding may be required (Tielker et al. 2009). Alternatively, the dependence of fluorescent proteins on oxygen for chromophore maturation has been exploited in FRET-based oxygen sensors (Potzkei et al. 2012). In one very interesting example, a transcription-based hypoxia biosensor was used to analyze the effect of mixing on E.coli physiology in a bioreactor. Two different bioreactor set ups with different numbers of impellers but the same concentration of dissolved oxygen in the bulk phase showed different profiles of induction of the hypoxia biosensor. This suggested the presence of micro or even nanoscale zones of low oxygen concentration that were negatively affecting the physiology of a subpopulation of cells that could not be detected with standard bioreactor probes, illustrating the power of genetically encoded biosensors to detect complex phenomena (Garcia et al. 2009).

Stress from Internal Processes Cells also possess a number of signaling pathways to deal with internal overload of critical biochemical pathways. Among these, the two most relevant to biological manufacturing are the detection of overload in the protein folding pathway and the oxidative damage from metabolism, although the production of specific products may have additional effects (e.g., production of hydrophobic products may damage membrane integrity and lead to additional harmful effects on cells (Cao et al. 2017)). Recombinant protein production in eukaryotic hosts most often relies on secretion of the product of interest to the culture medium where it is easier to purify due to the presence of fewer host cell protein contaminants. Proteins are co-translationally inserted into the endoplasmic reticulum (ER), where they undergo folding and disulfide bond formation, as well as glycosylation, if needed. High levels of overexpression of a given protein can overwhelm the protein folding machinery in

4

Biosensors of the Well-being of Cell Cultures

81

the ER, leading to the accumulation of unfolded or misfolded proteins (Ma and Hendershot 2001). Eukaryotic cells have evolved a dedicated stress response to deal with excess protein folding demand; the unfolded protein response (UPR), which upregulates the expression of chaperones, increases the rate of extraction and degradation of misfolded proteins and if homeostasis cannot be restored, activates apoptosis. Detection and prevention of UPR activation under manufacturing conditions is necessary to avoid proteolytic degradation of the product and prevent cell death. As with other stress response biosensors, work on developing UPR biosensors has exploited the natural transcriptional responses of cells (Du et al. 2013; Kober et al. 2012). First, natural promoters involved in the response were explored to identify sensitive and specific indicators of UPR (as opposed to other types of stress) (Kober et al. 2012). The resulting biosensor showed induction levels that correlated with the amount of recombinant protein produced by the cells under study, meaning that it could be used as a screening tool to identify high-producing cell lines. Subsequent work created a synthetic promoter that combined signals from multiple branches of the UPR pathway to use as a biosensor (Du et al. 2013). In this work, different cell lines and different process conditions lead to different induction kinetics of the biosensor. Crucially, this work was able to highlight the effect of osmolarity of the culture medium as an important variable in preventing UPR activation. Oxidative stress can occur as a byproduct of normal metabolic reactions, particularly under high energy demand conditions. High protein folding load can also contribute to oxidative damage. The addition of antioxidants can partially mitigate oxidative stress. Thus, biosensor to detect the activation of stress response due to oxidation can be useful for developing medium formulation. Transcriptionbased biosensors of oxidative stress have been developed for a variety of organisms ranging from bacteria (Kotova et al. 2010) to yeast (Jayaraman et al. 2005) to mammalian cells (Hendriks et al. 2012). Alternatively, redox-sensitive GFP derivatives have been developed that can measure the redox potential directly in whole cells or specific subcellular compartments of interest (Arias-Barreiro et al. 2010; Hanson et al. 2004; Merksamer et al. 2008). Biosensors that are specific indicators of hydrogen peroxide production have also been developed (Belousov et al. 2006; Ermakova et al. 2014).

Biosensors of Metabolites Healthy metabolism is a prerequisite for robust cell growth and product yield. A conserved set of metabolites, related to central metabolism, is often of interest for most production systems, regardless of the type of cell or the product produced. These include the primary source of carbon (often glucose or another sugar), amino acids as nitrogen sources and the building blocks for protein expression, and the accumulation waste metabolites such as lactate (Fig. 4). Biosensors can also be developed for important intermediates in the synthesis of the product of interest that are known to be critical for the yield obtained (e.g., shikimic acid in the case of

82

K. M. Polizzi Glucose

Amino Acids

ATP

Amino Acids

Pyruvate Ac-CoA

Amino Acids

Oxaloacetate

Lactate Acetate

Citrate

Malate Fumarate

ATP TCA Cycle

Isocitrate α-ketoglutarate

Succinate

Amino Acids

Succinyl-CoA

Fig. 4 A set of conserved metabolites is often important regardless of which product is being manufactured. Central metabolism in cells includes the metabolism of glucose and amino acids, leading to the production of ATP. Waste metabolites such as lactate may also be produced. Dashed arrows represent multiple steps. Most reactions in metabolism are reversible

aromatic amino acids (Li et al. 2017b)). Biosensors can also be developed for direct detection of the product of interest when this is a small molecule (Jang et al. 2017; Rogers and Church 2016), allowing a direct, high throughput measurement of yield during screening. In contrast to the cell stress examples discussed above where transcription-based biosensors are by far the most common architecture, biosensors of small molecule metabolites are more varied and include designs that operate at the transcriptional, translational, and posttranslational level. In particular, if accurate quantification of metabolite concentrations is desired, e.g., to inform the development of kinetic models of the biochemical processes, FRET-based biosensors are often chosen because their inherently ratiometric format gives higher accuracy measurements. A wide variety of FRET biosensors for metabolites have been reported to date including designs to sense glucose, various amino acids, intermediates of the TCA cycle, and ATP (reviewed in Constantinou and Polizzi (2013)). Many of the papers reporting new biosensor designs use fluorescence microscopy to conduct the FRET measurements. However, both flow cytometry (Banning et al. 2010) and fluorescence plate readers (Behjousiar et al. 2012) can be used, provided that the sensors have sufficient signal-to-noise ratio. A general lack of nutrients can be detected via transcription-based biosensors using promoters from starvation responses. This strategy has been commonly employed in Escherichia coli using different promoters (Brognaux et al. 2013a, b; Delvigne et al. 2009, 2011; Delvigne and Goffin 2014; Pokhilko 2017) to explore the

4

Biosensors of the Well-being of Cell Cultures

83

response of cells to changes in bioprocess conditions in both bioreactors and shake flasks. Transcription-based biosensors have also been developed to detect various amino acids (Skjoedt et al. 2016), phenylalanine (Mahr et al. 2016), valine (Mahr et al. 2015), and threonine (Liu et al. 2015b), which are often used to identify metabolic engineering strategies that lead to overproduction of that amino acid as the target product. However, they could equally be used to sense the intracellular concentrations of these amino acids during production of another molecule of interest. Finally, with respect to measuring the accumulation of waste products, biosensors have been developed to detect both lactate (Goers et al. 2017) and ammonia (Xiao et al. 2017), both of which have been shown to decrease the growth and productivity of mammalian cells due to alterations in the pH of the medium. Lactate is also a precursor for the production of some bioplastics, so this sensor could also be deployed as a measure of product formation in systems designed for its biosynthesis.

Conclusions and Future Directions Currently, a number of diverse genetically encoded biosensors have been developed for measuring traits associated with cellular well-being. Some have already been demonstrated to be useful for the optimization of bioprocesses, for example, by screening for highly productive cell lines (Kober et al. 2012; Rogers and Church 2016), understanding the impact of mixing on aeration (Garcia et al. 2009), or quantifying the concentration of critical metabolites under different conditions (Behjousiar et al. 2012; Goers et al. 2017). To date, most sensors have been applied as research tools and the concepts have been demonstrated in academia, but not yet adopted by industry. This is, to some extent, expected given the pace at which new technologies are translated from research into practice. To help ease this transition, new biosensors need to be developed and tested with industrial end users in mind. In some cases, this will require further exploration of their function under industrial conditions, e.g., using relevant cell lines and media, testing at small and large scale, and examining differences in performance in inhomogeneous environments. For other cases, it will be necessary that the biosensors can be integrated into the industrial workflow, e.g., signals measured using flow cytometry or plate readers by operators during the normal course of sampling. Widespread adoption of inline or automated flow cytometers will be an important part of this process. In addition, developing strategies that enable multiplexed measurements of multiple properties from a single sample (analogous to the current analyzers used in industry) will help encourage adoption of new biosensors. One multiplexing concept that has emerged in the literature recently is to create a mixed population of cells, each housing a biosensor for a particular analyte of interest that can be read independently. Different biosensor populations are then marked using the expression of a second indicator protein or through the use of a fluorescent dye (Doucette et al. 2016; Varma et al. 2017) so that the individual

84

K. M. Polizzi

biosensor responses can be tracked. The measurement in this case must be done by flow cytometry in order to separate the signals from different populations. There is also a drive toward devising modular architectures which will enable the rapid generation of biosensors for new targets with a minimum amount of trialand-error. Currently, the development cycle for a new biosensor can be lengthy as initial prototypes may not have sufficient signal-to-noise ratio to enable accurate measurements. Thus, the genetic network may be modified and retested in a number of iterations before the biosensor design is finalized. Developing generic strategies that are more likely to yield functional prototypes in the first iteration is of interest. For example, reusing well-characterized components that are known to have reliable performance (Feng et al. 2015), insulating parts from one another (Rugbjerg et al. 2016), or exploiting recurring structural features that are independent of the recognition elements used (Zheng et al. 2014) will all facilitate the rapid redesign of existing high-performance biosensors for new targets. Finally, a future horizon is to develop systems that do not just sense and report the well-being of cultures to human operators, but rather take steps to change bioprocessing conditions to improve conditions directly based on biosensor signals. Initial steps, where fluorescence outputs from GFP expression have been linked with a computational algorithm for process control, have already been reported (Menolascina et al. 2014; Milias-Argeitis et al. 2011). The next steps will be to build biological systems capable of coupling sensing and output actions without an in silico step in between.

References Arias-Barreiro CR et al (2010) A bacterial biosensor for oxidative stress using the constitutively expressed redox-sensitive protein roGFP2. Sensors (Basel, Switzerland) 10:6290–6306. https://doi.org/10.3390/s100706290 Aw R, McKay PF, Shattock RJ, Polizzi KM (2018) A systematic analysis of the expression of the anti-HIV VRC01 antibody in Pichia pastoris through signal peptide optimization. Protein Expr Purif 149:43–50. https://doi.org/10.1016/j.pep.2018.03.013 Banning C et al (2010) A flow cytometry-based FRET assay to identify and analyse protein-protein interactions in living cells. PLoS One 5:e9344. https://doi.org/10.1371/journal.pone.0009344 Behjousiar A, Kontoravdi C, Polizzi KM (2012) In situ monitoring of intracellular glucose and glutamine in CHO cell culture. PLoS One 7. https://doi.org/10.1371/journal.pone.0034512 Belousov VV, Fradkov AF, Lukyanov KA, Staroverov DB, Shakhbazov KS, Terskikh AV, Lukyanov S (2006) Genetically encoded fluorescent indicator for intracellular hydrogen peroxide. Nat Methods 3:281 Bermejo E et al (2018) Production of lutein, and polyunsaturated fatty acids by the acidophilic eukaryotic microalga Coccomyxa onubensis under abiotic stress by salt or ultraviolet light. J Biosci Bioeng 125:669–675. https://doi.org/10.1016/j.jbiosc.2017.12.025 Bradley RW, Wang B (2015) Designer cell signal processing circuits for biotechnology. New Biotechnol 32:635–643 Brognaux A, Han SS, Sorensen SJ, Lebeau F, Thonart P, Delvigne F (2013a) A low-cost, multiplexable, automated flow cytometry procedure for the characterization of microbial stress dynamics in bioreactors. Microb Cell Factories 12. https://doi.org/10.1186/1475-2859-12-100

4

Biosensors of the Well-being of Cell Cultures

85

Brognaux A, Thonart P, Delvigne F, Neubauer P, Twizere JC, Francis F, Gorret N (2013b) Direct and indirect use of GFP whole cell biosensors for the assessment of bioprocess performances: design of milliliter scale-down bioreactors. Biotechnol Prog 29:48–59 Buschhaus JM, Humphries B, Luker KE, Luker GD (2018) A Caspase-3 reporter for fluorescence lifetime imaging of single-cell apoptosis. Cell 7:57 Cao H et al (2017) Systems-level understanding of ethanol-induced stresses and adaptation in E. coli. Sci Rep 7:44150 Carlquist M et al (2012) Physiological heterogeneities in microbial populations and implications for physical stress tolerance. Microb Cell Factories 11. https://doi.org/10.1186/14752859-11-94 Ceroni F, Algar R, Stan G-B, Ellis T (2015) Quantifying cellular capacity identifies gene expression designs with reduced burden. Nat Methods 12:415 Chen A, Leith M, Tu R, Tahim G, Sudra A, Bhargava S (2017) Effects of diluents on cell culture viability measured by automated cell counter. PLoS One 12:e0173375 Chiang JJH, Truong K (2005) Using co-cultures expressing fluorescence resonance energy transfer based protein biosensors to simultaneously image caspase-3 and Ca2+ signaling. Biotechnol Lett 27:1219–1227 Cilek MZ et al (2011) AHR, a novel acute hypoxia-response sequence, drives reporter gene expression under hypoxia in vitro and in vivo. Cell Biol Int 35:1–8 Constantinou A, Polizzi KM (2013) Opportunities for bioprocess monitoring using FRET biosensors. Biochem Soc Trans 41:1146–1151 Cost AL, Ringer P, Chrostek-Grashoff A, Grashoff C (2015) How to measure molecular forces in cells: a guide to evaluating genetically-encoded FRET-based tension sensors. Cell Mol Bioeng 8:96–105 De Paepe B, Peters G, Coussement P, Maertens J, De Mey M (2017) Tailor-made transcriptional biosensors for optimizing microbial cell factories. J Ind Microbiol Biotechnol 44:623–645 Dekker L, Polizzi KM (2017) Sense and sensitivity in bioprocessing – detecting cellular metabolites with biosensors. Curr Opin Chem Biol 40:31–36 Delvigne F, Goffin P (2014) Microbial heterogeneity affects bioprocess robustness: dynamic singlecell analysis contributes to understanding of microbial populations. Biotechnol J 9:61–72 Delvigne F, Boxus M, Ingels S, Thonart P (2009) Bioreactor mixing efficiency modulates the activity of a prpoS::GFP reporter gene in E. coli. Microb Cell Factories 8:15 Delvigne F et al (2011) Characterization of the response of GFP microbial biosensors sensitive to substrate limitation in scale-down bioreactors. Biochem Eng J 55:131–139 Demuth C, Varonier J, Jossen V, Eibl R, Eibl D (2016) Novel probes for pH and dissolved oxygen measurements in cultivations from millilitre to benchtop scale. Appl Microbiol Biotechnol 100:3853–3863 Doucette J, Zhao Z, Geyer RJ, Barra MM, Balunas MJ, Zweifach A (2016) Flow cytometry enables multiplexed measurements of genetically encoded intramolecular FRET sensors suitable for screening. J Biomol Screen 21:535–547 Du Z et al (2013) Non-invasive UPR monitoring system and its applications in CHO production cultures. Biotechnol Bioeng 110:2184–2194 Dzyadevych SV, Arkhypova VN, Soldatkin AP, El’skaya AV, Martelet C, Jaffrezic-Renault N (2008) Amperometric enzyme biosensors: past, present and future. IRBM 29:171–180 Ellington AD, Szostak JW (1990) In vitro selection of RNA molecules that bind specific ligands. Nature 346:818 Ermakova YG et al (2014) Red fluorescent genetically encoded indicator for intracellular hydrogen peroxide. Nat Commun 5:5222 Feng J et al (2015) A general strategy to construct small molecule biosensors in eukaryotes. elife 4: e10606 Fosbrink M, Aye-Han NN, Cheong R, Levchenko A, Zhang J (2010) Visualization of JNK activity dynamics with a genetically encoded fluorescent biosensor. Proc Natl Acad Sci U S A 107:5459–5464

86

K. M. Polizzi

Garcia JR, Cha HJ, Rao G, Marten MR, Bentley WE (2009) Microbial nar-GFP cell sensors reveal oxygen limitations in highly agitated and aerated laboratory-scale fermentors. Microb Cell Factories 8:15 Goers L, Kylilis N, Tomazou M, Wen KY, Freemont P, Polizzi K (2013) Engineering microbial biosensors. In: Harwood C, Wipat A (eds) Microbial synthetic biology. Methods in microbiology, vol 40. Academic, Oxford, pp 119–156 Goers L, Ainsworth C, Goey CH, Kontoravdi C, Freemont PS, Polizzi KM (2017) Whole-cell Escherichia coli lactate biosensor for monitoring mammalian cell cultures during biopharmaceutical production. Biotechnol Bioeng 114:1290–1300 Hanson GT, Aggeler R, Oglesbee D, Cannon M, Capaldi RA, Tsien RY, Remington SJ (2004) Investigating mitochondrial redox potential with redox-sensitive green fluorescent protein indicators. J Biol Chem 279:13044–13053 Hendriks G et al (2012) The ToxTracker assay: novel GFP reporter systems that provide mechanistic insight into the genotoxic properties of chemicals. Toxicol Sci 125:285–298 Hoynes-O’Connor A, Shopera T, Hinman K, Creamer JP, Moon TS (2017) Enabling complex genetic circuits to respond to extrinsic environmental signals. Biotechnol Bioeng 114:1626–1631 Jang S, Jang S, Xiu Y, Kang TJ, Lee S-H, Koffas MAG, Jung GY (2017) Development of artificial riboswitches for monitoring of naringenin in vivo. ACS Synth Biol 6:2077–2085 Jayaraman M, Radhika V, Bamne MN, Ramos R, Briggs R, Dhanasekaran DN (2005) Engineered Saccharomyces cerevisiae strain BioS-OS1/2, for the detection of oxidative stress. Biotechnol Prog 21:1373–1379 Kober L, Zehe C, Bode J (2012) Development of a novel ER stress based selection system for the isolation of highly productive clones. Biotechnol Bioeng 109:2599–2611 Kotova VY, Manukhov IV, Zavilgelskii GB (2010) Lux-biosensors for detection of SOS-response, heat shock, and oxidative stress. Appl Biochem Microbiol 46:781–788 Krampe B, Al-Rubeai M (2010) Cell death in mammalian cell culture: molecular mechanisms and cell line engineering strategies. Cytotechnology 62:175–188 Kuystermans D, Mohd A, Al-Rubeai M (2012) Automated flow cytometry for monitoring CHO cell cultures. Methods 56:358–365 Kuystermans D, Avesh M, Al-Rubeai M (2016) Online flow cytometry for monitoring apoptosis in mammalian cell cultures as an application for process analytical technology. Cytotechnology 68:399–408 Li CJ, Imanishi A, Komatsu N, Terai K, Amano M, Kaibuchi K, Matsuda M (2017a) A FRET biosensor for ROCK based on a consensus substrate sequence identified by KISS technology. Cell Struct Funct 42:1–13 Li H, Liang C, Chen W, Jin J-M, Tang S-Y, Tao Y (2017b) Monitoring in vivo metabolic flux with a designed whole-cell metabolite biosensor of shikimic acid. Biosens Bioelectron 98:457–465 Lim B, Gross CA (2011) Cellular response to heat shock and cold shock. In: Bacterial stress responses, 2nd edn. American Society of Microbiology, Washington, DC Lim CT, Zhou EH, Quek ST (2006) Mechanical models for living cells – a review. J Biomech 39:195–216 Liu JY, Qu RJ, Ogura M, Shibata T, Harada H, Hiraoka M (2005) Real-time imaging of hypoxiainducible factor-1 activity in tumor xenografts. J Radiat Res 46:93–102 Liu D, Evans T, Zhang FZ (2015a) Applications and advances of metabolite biosensors for metabolic engineering. Metab Eng 31:35–43 Liu YN et al (2015b) Developing a high-throughput screening method for threonine overproduction based on an artificial promoter. Microb Cell Factories 14:121 Ma YJ, Hendershot LM (2001) The unfolding tale of the unfolded protein response. Cell 107:827–830 Ma TF, Chen YP, Guo JS, Wang W, Fang F (2018) Cellular analysis and detection using surface plasmon resonance imaging. TrAC Trends Anal Chem 103:102–109

4

Biosensors of the Well-being of Cell Cultures

87

Mahr R, Gatgens C, Gatgens J, Polen T, Kalinowski J, Frunzke J (2015) Biosensor-driven adaptive laboratory evolution of L-valine production in Corynebacterium glutamicum. Metab Eng 32:184–194 Mahr R, von Boeselager RF, Wiechert J, Frunzke J (2016) Screening of an Escherichia coli promoter library for a phenylalanine biosensor. Appl Microbiol Biotechnol 100:6739–6753 McLamore ES, Porterfield DM (2011) Non-invasive tools for measuring metabolism and biophysical analyte transport: self-referencing physiological sensing. Chem Soc Rev 40:5308–5320 Menolascina F et al (2014) In-vivo real-time control of protein expression from endogenous and synthetic gene networks. PLoS Comput Biol 10:e1003625–e1003625 Merksamer PI, Trusina A, Papa FR (2008) Real-time redox measurements during endoplasmic reticulum stress reveal interlinked protein folding functions. Cell 135:933–947 Miesenbock G, De Angelis DA, Rothman JE (1998) Visualizing secretion and synaptic transmission with pH-sensitive green fluorescent proteins. Nature 394:192–195 Milias-Argeitis A et al (2011) In silico feedback for in vivo regulation of a gene expression circuit. Nat Biotechnol 29:1114–1116 Millipore M (2018) 21CFR Part 11 Compliance Guidelines for Guava® ViaCount® Reagent Assay. https://www.merckmillipore.com/GB/en/life-science-research/cell-analysis-flow-cytometry/guavaeasycyte-flow-cytometers/Flow-Cytometry-Applications/bioprocess/Yo.b.qB.dwoAAAFE5k1zAv 6g,nav?ReferrerURL=https%3A%2F%2Fwww.google.co.uk%2F&bd=1 Pais DAM, Carrondo MJT, Alves PM, Teixeira AP (2014) Towards real-time monitoring of therapeutic protein quality in mammalian cell processes. Curr Opin Biotechnol 30:161–167 Passoth V, Cohn M, Schafer B, Hahn-Hagerdal B, Klinner U (2003) Analysis of the hypoxiainduced ADH2 promoter of the respiratory yeast Pichia stipitis reveals a new mechanism for sensing of oxygen limitation in yeast. Yeast 20:39–51 Pokhilko A (2017) Monitoring of nutrient limitation in growing E. coli: a mathematical model of a ppGpp-based biosensor. BMC Syst Biol 11:106 Potzkei J, Kunze M, Drepper T, Gensch T, Jaeger K-E, Büchs J (2012) Real-time determination of intracellular oxygen in bacteria using a genetically encoded FRET-based biosensor. BMC Biol 10:28 Randers-Eichhorn L, Albano CR, Sipior J, Bentley WE, Rao G (1997) Online green fluorescent protein sensor with LED excitation. Biotechnol Bioeng 55:921–926 Rogers JK, Church GM (2016) Genetically encoded sensors enable real-time observation of metabolite production. Proc Natl Acad Sci U S A 113:2388–2393 Rugbjerg P, Gence HJ, Jensen K, Sarup-Lytzen K, Sommer MOA (2016) Molecular buffers permit sensitivity tuning and inversion of riboswitch signals. ACS Synth Biol 5:632–638 Scognamiglio V, Antonacci A, Lambreva MD, Litescu SC, Rea G (2015) Synthetic biology and biomimetic chemistry as converging technologies fostering a new generation of smart biosensors. Biosens Bioelectron 74:1076–1086 Seefried L et al (2010) A small scale cell culture system to analyze mechanobiology using reporter gene constructs and polyurethan dishes. Eur Cells Mater 20:344–355 Shi S, Ang EL, Zhao H (2018) In vivo biosensors: mechanisms, development, and applications. J Ind Microbiol Biotechnol 45:491–516 Silberman M, Barac YD, Yahav H, Wolfovitz E, Einav S, Resnick N, Binah O (2009) Shear stressinduced transcriptional regulation via hybrid promoters as a potential tool for promoting angiogenesis. Angiogenesis 12:231–242 Skjoedt ML et al (2016) Engineering prokaryotic transcriptional activators as metabolite biosensors in yeast. Nat Chem Biol 12:951 Slomovic S, Pardee K, Collins JJ (2015) Synthetic biology devices for in vitro and in vivo diagnostics. Proc Natl Acad Sci U S A 112:14429–14435 Sonna LA, Fujita J, Gaffin SL, Lilly CM (2002) Invited review: effects of heat and cold stress on mammalian gene expression. J Appl Physiol 92:1725–1742

88

K. M. Polizzi

Sou SN, Sellick C, Lee K, Mason A, Kyriakopoulos S, Polizzi KM, Kontoravdi C (2015) How does mild hypothermia affect monoclonal antibody glycosylation? Biotechnol Bioeng 112:1165–1176 Tapsin S et al (2018) Genome-wide identification of natural RNA aptamers in prokaryotes and eukaryotes. Nat Commun 9:1289 Tielker D, Eichhof I, Jaeger KE, Ernst JF (2009) Flavin mononucleotide-based fluorescent protein as an oxygen-independent reporter in Candida albicans and Saccharomyces cerevisiae. Eukaryot Cell 8:913–915 Tuerk C, Gold L (1990) Systematic evolution of ligands by exponential enrichment: RNA ligands to bacteriophage T4 DNA polymerase. Science 249:505 Varma S, Fendyur A, Box A, Voldman J (2017) Multiplexed cell-based sensors for assessing the impact of engineered systems and methods on cell health. Anal Chem 89:4663–4670 Xiao Y, Jiang W, Zhang F (2017) Developing a genetically encoded, cross-species biosensor for detecting ammonium and regulating biosynthesis of cyanophycin. ACS Synth Biol 6:1807–1815 Yang SJ, Yu HL, You YZ, Li XL, Jiang JX (2018a) Effective lactic acid production from waste paper using Streptococcus thermophilus at low enzyme loading assisted by Gleditsia saponin. Carbohydr Polym 200:122–127 Yang T et al (2018b) A review of ratiometric electrochemical sensors: from design schemes to future prospects. Sensors Actuators B Chem 274:501–516 Yu LX, Amidon G, Khan MA, Hoag SW, Polli J, Raju GK, Woodcock J (2014) Understanding pharmaceutical quality by design. AAPS J 16:771–783 Zhang XD, Yang ST (2011) An online, non-invasive fluorescence probe for immobilized cell culture process development. Process Biochem 46:2030–2035 Zhang J et al (2013) Visualization of caspase-3-like activity in cells using a genetically encoded fluorescent biosensor activated by protein cleavage. Nat Commun 4:2157 Zhao L, Fu HY, Zhou WC, Hu WS (2015) Advances in process monitoring tools for cell culture bioprocesses. Eng Life Sci 15:459–468 Zheng H, Bi J, Krendel M, Loh SN (2014) Converting a binding protein into a biosensing conformational switch using protein fragment exchange. Biochemistry 53:5505–5514

5

Systematic Design of a Quorum SensingBased Biosensor for the Detection of Metal Ions in Escherichia coli Bor-Sen Chen

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Basic Principles of Quorum Sensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Design Example of QS-Based Metal Ion Biosensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Construction of QS-Based Metal Ion Biosensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Mathematical Model of QS-Based Metal Ion Biosensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Design Specifications for QS-Based Metal Ion Biosensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Design Procedure for the QS-Based Metal Ion Biosensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Experimental Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Component Libraries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

90 91 95 96 97 101 101 102 106 107 107 108

Abstract

With the recent industrial expansion, heavy metals and other pollutants have increasingly contaminated our living surroundings. The non-degradability of heavy metals may lead to accumulation in food chains, and the resulting toxicity could cause damage in organisms. Hence, detection techniques have gradually received attention. In this study, a quorum sensing (QS)-based amplifier is introduced to improve the detection performance of metal ion biosensing. The design utilizes diffusible signal molecules, which freely pass through the cell membrane into the environment to communicate with others. Bacteria cooperate via the cell-cell communication process, thereby displaying synchronous behavior, even if only a minority of the cells detect the metal ion. In order to facilitate the design, the ability of the engineered biosensor to detect metal ions is described in a steady-state model. The design can be constructed according to user-oriented specifications by selecting adequate components from B.-S. Chen (*) Department of Electric Engineering, National Tsing Hua University, Hsinchu, Taiwan e-mail: [email protected] © Springer Nature Switzerland AG 2022 G. Thouand (ed.), Handbook of Cell Biosensors, https://doi.org/10.1007/978-3-030-23217-7_120

89

90

B.-S. Chen

corresponding libraries, with the help of a genetic algorithm (GA)-based design method. The experimental results validate enhanced efficiency and detection performance of the quorum sensing-based biosensor of metal ions.

Introduction Metal ions play important roles in biological metabolic pathways and affect cellular processes through a wide variety of reactions. Organisms require metal ions, which are involved in processes, such as bacterial respiration, electron transport, and peroxide reduction (Achtman and Suerbaum 2001). The uptake and efflux systems in biology regulate ion homeostasis to prevent a lack or excess of metal ions. With the recent industrial expansion, there has been either an excess of metal ions in wastewater (Ngah and Hanafiah 2008; Rezvani-Boroujeni et al. 2015; Singha and Guleria 2014; Teodosiu et al. 2014; Mulligan et al. 2001) or the intentional or careless discharge of other contaminants (Migaszewski and Galuszka 2015) into the natural environment. Metal ions can be classified based on their effects and toxicity on cells into essential ions, trace ions, and heavy metals. The stronger redox ability of heavy metals allows them to compete with essential ions and trace ions in redox actions, leading to superoxide or hydroxyl radicals (Hartwig 1995). Superoxide and hydroxyl radicals may damage lipids and proteins (Davies 2005; Raha and Robinson 2000), disrupt DNA oxidation (Hartwig 1995), and even cause cell death (Hartwig 1995). Furthermore, heavy metals are difficult to remove via metabolism and can accumulate in organs, causing permanent damage. Hence, detection techniques have gradually received great attention. In this chapter, we introduce a synthetic genetic circuit that can enhance the detection performance of metal ion biosensors and has the potential to be applied for environmental detection and environmental bioremediation. Recent progress in research on synthetic biology (Chang et al. 2013; Danino et al. 2010; Khalil and Collins 2010; Silva-Rocha and de Lorenzo 2014; Sohka et al. 2009; Stricker et al. 2008; Wang et al. 2011; Cameron et al. 2014; Kotula et al. 2014; Teo et al. 2015) has provided an alternative means for metal ion detection aided by specific promoter elements such as derived from R. metallidurans (Taghavi et al. 1997; Hobman et al. 2012) or found in E. coli (Munson et al. 2000). In order to improve the detection performance, a metal ion-induced promoter connects to bacterial quorum sensing (QS) system. Quorum sensing is a mechanism which regulates gene expression for many functions through cell-cell communication (Brint and Ohman 1995; Darch et al. 2012; Miller and Bassler 2001; Passador et al. 1993; Pearson et al. 1995). Bacteria produce diffusible signal molecules, generally known as autoinducers (AIs) (Kaplan and Greenberg 1985; Dunlap and Kuo 1992; Fuqua et al. 1994; Ruby and Mcfallngai 1992; Ruby and Nealson 1976), and release them into the environment to communicate with others. Through this process, bacteria cooperate and thereby display synchronous behavior. The most extensively studied quorum sensing system is the autoinduction of luminescence in the marine bacterium, Aliivibrio fischeri, which is symbiotic with squids and some

5

Systematic Design of a Quorum Sensing-Based Biosensor for the. . .

91

marine fishes (Ruby and Mcfallngai 1992; Ruby and Nealson 1976; Nealson and Hastings 1979). The luminescence here is produced by the operon acquired from V. fischeri, of which the LuxR protein is the transcriptional activator of luminescence and the LuxI protein synthesizes the specific N-acylated homoserine lactone (AHL) (Engebrecht and Silverman 1984; Hanzelka and Greenberg 1996; Val and Cronan 1998). After the LuxR protein binds AHL to form a complex, the complex can bind the target promoter sequence and activate downstream gene transcription, even if only some of the bacteria can detect the metal ions. Thus, it can be employed to enhance the detection performance by signal amplification. To make the design of the QS-based metal ion biosensor easier, we introduce a mathematical model to describe the dynamic and steady-state regulatory behavior, which is related to transcriptional and translational processes (Chen and Wang 2014). A promoter allows RNA-polymerase molecules to latch onto a DNA strand and initialize transcription of a downstream gene into mRNA, and a RBS allows ribosome to bind and translate mRNA. A promoter combined with a RBS thus in this study is viewed as a component, of which characterizations can be identified with a reporter protein by measuring the fluorescence intensity. One can therefore use kinetic parameters as component libraries to design the circuit for QS-based metal ion detection. The design can be constructed according to user-oriented specifications by selecting adequate promoter-RBS components within a feasible range of metal ion concentrations. A long computation time is required in general when the component libraries become large. Hence, a genetic algorithm-based design method proposed here provides a simple and useful tool that saves time in evaluating and selecting components. So in summary, this study provides a systematic methodology for developments in next-generation synthetic biology, from component library construction to gene circuit assembly. When the libraries are more complete, more precise detection can be achieved. The purposes of this chapter are fivefold: (a) The basic principles of quorum sensing are introduced as the basic background of quorum sensing-based biosensor. (b) A QS-based amplifier is introduced into a biosensor to enhance the ability for metal ion detection. (c) Based on the promoter-RBS kinetic strengths, we establish three kinds of component libraries. (d) According to user-oriented specifications, the quorum sensing (QS)-based metal ion biosensor can be constructed by selecting adequate promoter-RBS components from the corresponding libraries to achieve a desired ability for metal ion detection. (e) The proposed methodology could provide synthetic biologists with a useful tool to design metal ion biosensors.

The Basic Principles of Quorum Sensing Before we discuss the systematic design of a quorum sensing-based biosensor for detection of metal ion in E. coli, the basic principles of quorum sensing involved in using this phenomenon are described as the basis for biosensor design and construction in the following. Quorum sensing was discovered and described in two luminous marine bacterial species over four decades, i.e., Aliivibrio fischeri

92

B.-S. Chen

and Vibrio harveyi (Achtman and Suerbaum 2001; Ruby 1996). In both species, the enzymes responsible for light production are encoded by the luciferase structural operon lux CDABE (Achtman and Suerbaum 2001; Miller and Bassler 2001; Ruby 1996), and light emission was determined to occur only at high cell population density in response to the accumulation of secreted autoinducer signaling molecules (Ruby 1996). Only a few other cases of bacterial regulation of gene expression in response to cell-cell signaling were known, for example, antibiotic production by Streptomyces spp., conjugation in Enterococcus faecalis, and fruiting body development in Myxococcus xanthus which were also recognized to be controlled by intercellular signaling. Therefore, these bacterial communication systems were considered anomalous because bacteria as a whole were not believed to use cellcell communication in general. However, the exchange of chemical signals between cells and organisms could be assumed to be a trait highly characteristic of eukaryotes (Ruby 1996). In the recent explosion of advances in the field of cell-cell communication in bacteria, many or most bacteria have been shown to communicate using secreted chemical molecules to coordinate the behaviors of the group. Furthermore, a vast assortment of different classes of chemical signals have been employed that individual species of bacteria use more than one chemical signal and/or more than one type of signal to communicate. Therefore complex hierarchical regulatory circuits have evolved to integrate and process the sensory information, and the signals can be used to differentiate between species in consortia. Obviously, the ability to communicate both within and between species is critical for bacterial survival and interaction in natural habitats (Miller and Bassler 2001; Ruby 1996). In the past decades quorum sensing circuits have been identified in many species of Gram-negative bacteria. The quorum sensing circuits identified in Gram-negative bacteria resemble the canonical quorum sensing circuit of the symbiotic bacterium V. fischeri. Specifically, these Gram-negative bacterial quorum sensing circuits contain homologues of two V. fischeri regulatory proteins, i.e., LuxI and LuxR, at least. The LuxI-like proteins respond for the biosynthesis of a specific acylated homoserine lactone signaling molecule (HSL) known as an autoinducer. The autoinducer concentration will increase with the increase of cell population density. The LuxR-like protein could bind cognate HSL autoinducers that cause achieved a critical threshold concentration, and the LuxR-autoinducer complexes could also activate target gene transcription (Miller and Bassler 2001). Therefore, using this quorum sensing mechanism, Gram-negative bacteria could efficiently couple gene expression to fluctuations in cell population density. Among these species of bacteria that could mediate quorum sensing by means of a LuxI/LuxR-type circuit, the V. fischeri, Pseudomonas aeruginosa, Agrobacterium tumefaciens, and Erwinia carotovora systems are the most understood (Miller and Bassler 2001). The most intensely studied quorum sensing system is the bioluminescent marine bacterium V. fischeri. This bacterium lives in symbiotic association with a number of eukaryotic hosts. In each case the host has developed a specialized light organ that is inhabited by a pure culture of a specific strain of V. fischeri at very high cell density. In these symbiotic associations, scientists have found that the eukaryotic host supplies V. fischeri with a nutrient-rich environment in which to live. The role of

5

Systematic Design of a Quorum Sensing-Based Biosensor for the. . .

93

V. fischeri could provide the host with light, and then each eukaryotic host uses the light which is provided by the bacteria for a specific purpose. For example, in the squid Euprymna scolopes and V. fischeri association, the squid has evolved into an antipredation strategy in which it counters and illuminates itself using the light from V. fischeri. In contrast, the fish Monocentris japonica could use the light produced by V. fischeri to attract a mate. Other uses of the V. fischeri light, such as warding off predators and attracting prey, have been documented (Miller and Bassler 2001). In general, V. fischeri culture grows to extremely high cell densities, reaching 1011 cells per ml (Ruby 1996). As V. fischeri culture grows, it could produce and release an autoinducer hormone into the extracellular environment, and the hormone is trapped inside the light organ with the bacteria. Detection of autoinducer by V. fischeri could elicit a signaling cascade that can cumulate in the emission of light. Thus, the quorum sensing system of V. fischeri has evolved to specifically enable the bacteria to produce light only under conditions in which there is a positive selective advantage for the light (Miller and Bassler 2001). As mentioned above, the luciferase enzymes required for light production in V. fischeri are encoded by luxCDABE, which exists as part of the luxICDABE operon. Two regulatory proteins called LuxI and LuxR comprise the quorum sensing apparatus. As shown in Fig. 1, LuxI is the autoinducer synthase enzyme, and it acts in the production of an HSL, N-(3-oxohexanoyl)-homoserine lactone (Ruby 1996). LuxR functions both to bind the autoinducer and to activate transcription of the luxICDABE operon as shown in the quorum sensing system of V. fischeri in Fig. 1. Specifically, at low cell densities, the luxICDABE operon is transcribed at a low basal level. Therefore, a low level of autoinducer is produced via luxI, and because the genes encoding luciferase are located directly downstream of the luxI gene, only a low level of light is produced. The HSL autoinducer is freely diffusible across the cell membrane, so the concentration of autoinducer in the extracellular environment is the same as the intracellular concentration of the autoinducer (Miller and Bassler 2001). As the V. fischeri culture grows, autoinducer accumulates to a threshold level that is sufficient for detection and binding by the cytoplasmic LuxR protein. In this situation, the interaction of LuxR with the autoinducer unmasks the LuxR DNAbinding domain, allowing LuxR to bind the luxICDABE promoter and activate its transcription. This action results in an exponential increase in both autoinducer production and light emission. The LuxR-HSL complex also acts to negatively regulate the expression of LuxR. The negative feedback loop is a compensatory mechanism that decreases luxICDABE expression in response to the positive feedback circuit (Miller and Bassler 2001). Gram-positive bacteria also regulate a variety of processes in response to increasing cell population density. However, in contrast to Gram-negative bacteria, which use HSL autoinducer, Gram-positive bacteria employ secreted peptides as autoinducer for quorum sensing. In general, the peptide is secreted via a dedicated ATP-binding cassette (ABC) transporter. Again, in contrast to the widespread use of LuxR-type proteins as autoinducer sensors by Gram-negative bacteria, Gram-positive bacteria use two-component adaptive response proteins for detection of the autoinducers. The signaling mechanism is a phosphorylation/

94

B.-S. Chen

Fig. 1 The Aliivibrio fischeri LuxI/LuxR quorum sensing circuit. There are five luciferase structure genes (luxICDABE) and two regulatory genes (LuxR and LuxI) required for quorum sensingcontrolled light emission in V. fischeri. The genes are arranged in two adjacent but divergently transcribed units. LuxR is transcribed to the left, and the luxICDABE operon is transcribed to the right. The LuxI protein is responsible for synthesis of HSL autoinducer N-(3-oxohexanoyl)homoserine lactone. As the cell population density increases, the concentration of the autoinducer increases both intra- and extracellularly. At a critical autoinducer concentration, the LuxR protein binds the autoinducer. The LuxR-autoinducer complex binds at the luxICDABE promoter and activates transcription of this operon. This action results in an exponential increase in autoinducer synthesis via the increase in transcription of LuxR and an exponential increase in light production via the increase in transcription of luxICDABE. The LuxR-autoinducer complex also binds at the LuxR promoter. This negative action compensates for the positive action at luxICDABE promoter (Miller and Bassler 2001; Ruby 1996)

dephosphorylation cascade. A general model for quorum sensing in Gram-positive bacteria is shown in Fig. 2. In brief, secreted peptide autoinducers increase in concentration as a function of the cell population density. Two-component sensor kinases are the detectors for the secreted peptide signals. Interaction with the peptide ligand initiates a series of phosphoryl events that culminate in the phosphorylation of cognate response regulator protein. Phosphorylation of the response regulator activates it, allowing it to bind DNA and later the transcription of the quorum sensingcontrolled target genes. Several Gram-positive quorum sensing systems have been extensively studied, for example, the model systems controlling competence in Streptococcus pneumoniae, competence and sporulation in Bacillus subtilis, and virulence in Straphylococcus aureus (Miller and Bassler 2001) After we learn how quorum sensing facilitates species-specific and interspecies cell-cell communication, how quorum sensing allows populations to act synergistically, and how quorum sensing can be used to conquer competitors, we could learn about an assortment of the signals that are employed by bacteria and about the biosynthesis of these signals as well as how the information encoded in these chemical signals is processed and transduced to control gene expression. These

5

Systematic Design of a Quorum Sensing-Based Biosensor for the. . .

95

ABC transporter H

P

peptide autoinducer signal precusrsor protein

sensor kinase

D

phosphorglation cascade

Response regulator

bacterial cell peptide signal precursor locus

target genes

Fig. 2 A general model for peptide-mediated quorum sensing in Gram-positive bacteria. In Grampositive bacteria, a peptide signal precursor locus is translated into a precursor protein that is cleaved to produce the processed peptide autoinducer signal. Generally, the peptide signal is transported out of the cell via an ABC transporter. When the extracellular concentration of the peptide signal accumulates to the minimal stimulatory level, a histidine sensor kinase protein of a two-component signaling system detects it. The sensor kinase autophosphorylates on a conserved histidine (H) residue, and subsequently, the phosphoryl group is transferred to a cognate response regulator protein. The response regulator is phosphorylated on a conserved aspartate residue (D). The phosphorylated response regulator activates the transcription of target genes (Miller and Bassler 2001; Ruby 1996)

basic principles of quorum sensing will be employed for systematic design of a quorum sensing-based biosensor for the metal ion in E.coli in the following sections.

Design Example of QS-Based Metal Ion Biosensor In the design example, the materials and methods are given as follows: In biosensor reagents, all restriction enzymes and DNA ligation kit are purchased from New England Biolabs. The chemicals used here are from Sigma-Aldrich. The oligonucleotides are from Integrated DNA Technologies. In bacteria strains and medium, Escherichia coli strain DH5α cells from Yeastern Biotech (ECOS) are used for the construction of the procedure and for the fluorescence measurement experiments. An M9 working medium (34 g/L Na2HPO4, 15 g/L K2HPO4, 2.5 g/L NaCl, 5 g/L NH4Cl, 2 g/L casamino acids, 2 μM vitamin B1, 2 mM MgSO4, 0.2% glucose) with proper antibiotics is used for E. coli cultivation at 37
C and 200 r.p.m. In the plasmid construction, the DNA parts used in this study are selected from BioBrick or synthesized by MDBio Biotech Co. Ltd. All DNA parts are assembled

96

B.-S. Chen

into the backbone pSB3K3 (20–30 copies per cell), via the BioBrick standard assembly method. In the fluorescence measurement, the cells containing the genetic circuit are inoculated in the M9 working medium with 50 mg/mL kanamycin. The culture is incubated for approximately 14–16 h, after which it is diluted 250-fold and further incubated until the OD600 reaches 0.1. Different concentrations of CuSO4 are then added to the medium. The fluorescence intensity of the culture is measured subsequently by the microplate reader (BioTek, Synergy H1, GFP settings are 490/ 510 nm for excitation and emission).

Construction of QS-Based Metal Ion Biosensor Recent research on synthetic biology offers an alternative means for metal ion detection using specific promoter elements derived from microorganisms. For example, PcusC and PpcoE are copper-inducible promoters regulated by CusRS, which are a part of the E. coli chromosome system and system, respectively (Bondarczuk and Piotrowska-Seget 2013). The system (cusCFBA) found in E. coli is a tetrapartite system endowing resistance to copper ion and is involved in periplasmic copper detoxification (Franke et al. 2003). The system (pcoABCDRSE) is discovered in bacteria that survive in copper-rich environments; plasmid-encoded operons in this system are known to be responsible for resistance to copper toxicity (Rouch and Brown 1997; Lee et al. 2002; Tetaz and Luke 1983). Another example of a metal ionregulated promoter is the lead resistance promoter (Taghavi et al. 1997). The promoter is acquired from Ralstonia metallidurans strain CH34 isolated from lead-contaminated soil, containing at least seven determinants encoding resistances to toxic heavy metals (Taghavi et al. 1997; Hobman et al. 2012). Combining functions involved in uptake, efflux, and accumulation of Pb(II) ion, the lead resistance operon (pbrABCDRT) is regulated by PbrR, which belongs to the MerR family of metal regulatory proteins (Borremans et al. 2001; Brown et al. 2003). To improve the detection performance, a metal ion-induced promoter-RBS Mi is connected to the LuxI protein-coding sequence (in Fig. 3). This allows the LuxI protein to be translated by the activation of the metal ion-induced promoter. The luxI component derived from the lux operon in V. fischeri is involved in the production of the LuxI protein, which catalyzes S-adenosyl methionine and acyl-acyl carrier protein into a specific AHL (referred to as HSL) as signal molecules (Hanzelka and Greenberg 1996; Val and Cronan 1998). HSLs may be classified into several types according to acyl group length (C4–C18). Here, the type of HSL synthesized by LuxI protein is C6-HSL. Because the lux operon is an exogenous DNA sequence for Escherichia coli (E. coli), it is required to supply LuxR protein for the activation of promoter Plux. The LuxR protein-coding sequence is connected to a constitutive promoter-RBS component Ci. When sufficient amounts of the LuxR protein is produced using a constitutive promoter-RBS component with the presence of C6HSL, C6-HSL can then bind LuxR protein to form the LuxR complex. The complex targets the cognate QS-dependent promoter-RBS component Ak and activates the transcription of the green fluorescent protein (GFP) coding sequence, even if only

5

Systematic Design of a Quorum Sensing-Based Biosensor for the. . .

97

Cell 1

Metal ion Cell N xR

xl

G xG

RBS Metal ion-induced promoter-RBS Mi

luxl

luxR

RBS Constitutive promoter-RBS Cj

RBS

gfp

QS-dependent promoter-RBS Ak

QS-based amplifier Cell i

Fig. 3 The QS-based metal ion biosensor. A metal ion-induced promoter-RBS connects with a quorum sensing system found in the marine bacterium, V. fischeri. Via the transmission of signal molecules, the output signal will be produced significantly by the activation of the QS-dependent promoter

some of the bacteria can detect the metal ions. This mechanism is helpful to enhance the detection ability by signal amplification.

Mathematical Model of QS-Based Metal Ion Biosensor To construct the dynamic model, promoter-RBS regulation must be introduced. The promoter-RBS regulation function is first defined by P(Pu, Pl, TF, I), in which Pu and Pl denote the maximum and minimum promoter-RBS strengths, respectively, TF denotes transcription factor concentration, and I is inducer concentration. This function describes the biochemical aspect of the transcription and translation process. The dynamic model of the metal ion biosensor with QS-based amplifier in Fig. 3 is thus described as follows. Note that the QS-based metal ion biosensor is divided into three stages. The first stage involves a metal ion-induced promoter-RBS Mi for producing autoinducer synthase LuxI, which catalyzes S-adenosyl methionine and acyl-acyl carrier protein into a specific AHL. The dynamics is described in the following equation (Alon 2007):
 x_E ðt Þ ¼ PM Pu,i , Pl,i , xS , I M  ðd þ rE Þ  xE ðt Þ

(1)

98

B.-S. Chen

in which
 P M P u ,i , P l , i , x S , I M ¼ P u ,i þ

xSI ðxS , I M Þ ¼

P u ,i  P l ,i  nSI 1 þ xSI ðKxSSI, I M Þ

x S  KM 1þ IM

where xE denotes the concentration of autoinducer synthase. IM is the concentration of metal ion; xS is the total concentration of metal ion-dependent regulatory protein. xSI denotes the complex of xI and IM. i denotes the ith metal ion-induced promoterRBS component in Table 1. PM(Pu, i, Pl, i, xS, IM) denotes the promoter-RBS regulation activity of the metal ion-induced promoter-RBS components. Pu, i and Pl, i denote the maximum and minimum promoter-RBS strengths of the ith metal ion-induced promoter-RBS component in Tables 1-3 in Appendix. rE denotes the degradation rate for autoinducter synthase. d is the dilution rate due to cell growth. KSI and nSI denote the binding affinity and binding cooperativity between the complex xSI and the corresponding promoter-RBS component, respectively. KM is the dissociation rate between the metal ion IM and the metal regulatory protein xS. Then, the dynamics for the catalysis of S-adenosyl methionine and acyl-acyl carrier protein into a specific autoinducer is given by x_I ðt Þ ¼ axE ðt Þ  ðd þ rI Þ  xI ðt Þ

(2)

where xI is the concentration of autoinducer, a is autoinducer synthesis rate, and rI denotes the degradation rate for autoinducter. The second stage involves a constitutive promoter-RBS part Ci for producing protein, LuxR. The change in concentration of regulatory protein LuxR with respect to time is given by (Alon 2007)
 x_R ðt Þ ¼ PC Pu,j ,0,0,0  ðd þ rR Þ  xR ðt Þ

(3)

and
 PC Pu,j ,0,0,0 ¼ Pu,j where xR denotes the concentration of transcriptional activator protein, LuxR. j is the jth constitutive promoter-RBS component in Table 2. PC(Pu, j, 0, 0, 0) is the regulation activity of the constitutive promoter-RBS components. Pu, j is the promoterRBS strength of the jth constitutive promoter-RBS component in Table 2. rR denotes the degradation rate for transcriptional activator protein. The third stage contains a QS-dependent promoter-RBS part for driving the expression of the immature reporter protein. When complex of xR and xI

5

Systematic Design of a Quorum Sensing-Based Biosensor for the. . .

99

accumulates, it activates the QS-dependent promoter and enhances the expression of reporter gene. Therefore, the dynamics is governed by the following equation (Alon 2007):
 x_G ðt Þ ¼ PA Pu,k , Pl,k , xR , xI  ðd þ rG Þ  xG ðt Þ

(4)

and
 P A P u ,k , P l ,k , x R , x I ¼ P u ,k þ

xRI ðxR , xI Þ ¼

Pu,k  Pl,k  nRI 1 þ xRI ðKxRRI, xI Þ

xR   KI 1þ xI

where xG is the concentration of immature reporter protein and represents the complex of xR and xI. k is the kth QS-dependent promoter-RBS component in Table 3. PA(Pu, k, Pl, k, xR, xI) is the activity of the QS-dependent promoter-RBS components. Pu, k and Pl, k are the maximum and minimum promoter-RBS strengths of the kth QS-dependent promoter-RBS component in Table 3. rG is the degradation rate for immature reporter protein. KRI and nRI are the binding affinity and binding cooperativity between the complex xRI and the promoter-RBS part, respectively. KI is the dissociation rate between the autoinducer xI and the transcriptional activator protein xR. Finally, the dynamics for the maturation of xG into the mature reporter protein G is given by G_ ðt Þ ¼ mxG  ðd þ rÞGðt Þ

(5)

where m is the maturation rate for the reporter protein and r is the degradation rate for mature reporter protein. The dynamic model for the QS-based metal ion biosensor is then transformed into steady-state model by assuming that the dynamics in (1), (2), (3), (4), and (5) are equal to zero. The steady-state model of the QS-based metal ion biosensor with promoter-RBS parts Mi, Cj, and Ak being selected from the corresponding libraries in Supplementary Appendix A, respectively, is derived as follows:
 8 P M P u ,i , P l ,i , x S , I M a > > xISS ¼  > > > d þ rI ðd þ rE
Þ > > > > P P ,0,0,0 C u , j > < xRSS ¼ x_R ðt Þ ¼ ðd þ rR Þ
> > P P , P A u , k l , k , xR , xI > > xGSS ¼ > > > d þ rG > > > : Gss ¼ m xG dþr

(6)

100

B.-S. Chen

where xISS, xRSS, xGSS and Gss are the steady-state concentrations of autoinducter, transcriptional activator protein, and immature reporter protein, as well as steadystate mature reporter protein, respectively. The steady-state expression of the QS-based metal ion biosensor can be obtained from the steady-state model in (6) with the regulation functions PM(Pu, i, Pl, i, xS, IM), PC(Pu, j, 0, 0, 0), as well as PA(Pu, k, Pl, k, xR, xI), respectively. In general, a synthetic genetic circuit in vivo also suffers from extrinsic environmental cellular noise, such as transmitted noise from upstream and global noise affecting all cells (Chen and Wang 2006; Soltani et al. 2015; Kyung et al. 2015). Thus, the equation in (6) should be modified as follows:
 8 P M P u ,i , P l ,i , x S , I M a > > xISS ¼  þ v1 > > > d þ ð d þ r Þ E
>  rI > > > > < xRSS ¼ x_R ðt Þ ¼ PC Pu,j ,0,0,0 þ v2 ðd þ rR Þ
> > PA Pu,k , Pl,k , xR , xI > > xGSS ¼ þ v3 > > > d þ rG > > m > : Gss ¼ xG þ v4 dþr

(7)

where the Gaussian random noise vp, p = 1, 2, 3, with a zero mean and variance of σ 2p, denotes cellular noise for both the transcriptional and translational gene expression processes at the steady state. v4 denotes cellular noise in mature protein expression at the steady state. Besides, biological components are inherently uncertain in a molecular biological system (Chen and Wang 2006). For example, the intrinsic kinetic parameters of the components including the processes of transcription and translation, the degradation rates of regulatory proteins, the dilution rates of the cells, and the maturation rate for the reporter proteins are all stochastically uncertain in vivo, as a result of gene expression noise from biochemical processes, thermal fluctuations, DNA mutation, and evolution. These perturbations are defined as follows: Pu,i ! Pu,i þ ΔPu,i n1 ðt Þ,Pl,i ! Pl,i þ ΔPl,i n2 ðt Þ, Pu,j ! Pu,j þ ΔPu,j n3 ðt Þ, Pu,k ! Pu,k þ ΔPu,k n4 ðt Þ,Pl,k ! Pl,k þ ΔPl,k n5 ðt Þ, rE ! rE þ ΔrE n6 ðt Þ, rR ! rR þ ΔrR n7 ðt Þ, rG ! rG þ ΔrG n8 ðt Þ, m ! m þ Δmn9 ðt Þ, r ! r þ Δrn10 ðt Þ, d ! d þ Δdn11 ðt Þ

(8)

where ΔPu, i, ΔPl, i, ΔPu, j, ΔPu, k, ΔPl, k, ΔrE, ΔrR, ΔrG, Δm, Δr, and Δd are the standard deviations of the corresponding stochastic parameters and nq(t), q = 1, 2, . . . , 11 denote Gaussian noise, which have zero mean and unit variance,

5

Systematic Design of a Quorum Sensing-Based Biosensor for the. . .

101

and account for random fluctuation sources. If the kinetic parameters in the steady-state model in (7) are replaced by the parameter perturbations shown in for robust design of the gene circuit, then the QS-based metal ion biosensor can tolerate these fluctuations in vivo, i.e., a design that accounts for (8) should be able to tolerate the parameter fluctuations.

Design Specifications for QS-Based Metal Ion Biosensor The purpose of our design is to construct a QS-based metal ion biosensor by selecting a set of suitable components from the corresponding libraries to achieve optimal tracking of a desired I/O response within a feasible range of metal ion concentrations. To achieve this, the following design specifications are needed: • Desired I/O response Gref(IM) of the synthetic genetic QS-based metal ion biosensor • A feasible range of metal ion concentrations • Standard derivations of cellular disturbances and parameter fluctuations to be tolerated in vivo • A cost function between the desired reference steady-state fluorescence intensity Gref and the steady-state fluorescence intensity Gss in within a specified range of metal ion concentrations is given to be minimized as follows: ð J ðS Þ ¼ E




2 Gss ðS, I M Þ  Gref ðI M Þ dI M

(9)

where S denotes the set of promoter-RBS components Mi, Cj, and Ak to be selected from the corresponding libraries in Appendix. If the cost function in is minimized by choosing the most appropriate set of components under design specifications, the fluorescence intensity of an engineered metal ion biosensor will be as close as possible to the specified steady-state fluorescence intensity under parameter fluctuations and environmental noise optimally. Although the cost function J(S) can be minimized by traditional conventional search methods, it will require long computation times and several trial-and-error experimentations when component libraries become large. Thus, a more effective and efficient genetic algorithm (GA)-based searching method is proposed here to save time in evaluating and selecting adequate promoter-RBS components from corresponding libraries in Appendix.

Design Procedure for the QS-Based Metal Ion Biosensor The design procedure for a QS-based metal ion biosensor is summarized as follows: 1. Provide user-defined design specifications Gref(IM) for the quorum sensing-based metal ion biosensor.

102

B.-S. Chen

2. Select an initial set S of promoter-RBS components from the corresponding libraries. 3. Calculate the cost value J(S) in for S. The relationship between the fitness value F (S) and the cost value J(S) is inversely proportional to (10) F ðS Þ ¼

1 : J ðS Þ

(10)

The fitness value plays a key role in natural selection for selecting a set S of promoter-RBS components to maximize the fitness value F(S) or equivalently to minimize the cost function J(S) in (9). For example max F ðS Þ ¼ S

1 : min J ðS Þ

(11)

S

4. Create an offspring set S, using GA operators such as reproduction, crossover, and mutation (Chen et al. 1995). (a) Make copies of possible solutions, on basis of their fitness. (b) Swap values between two possible solutions. (c) Randomly alter the value in a possible solution. 5. Calculate the cost value of the new set S obtained by natural selection. Stop when the design goal is achieved or an acceptable solution is obtained due to a limited number of components available from the libraries. Otherwise, create the next generation and return to step 3.

Experimental Results For the convenience of description and explanation, as shown in Fig. 3, a QS-based metal ion biosensor is assembled by selecting a set of components from component libraries in Appendix, namely, a metal ion-induced promoter-RBS component Mi, a constitutive promoter-RBS component Cj, and a QS-dependent promoter-RBS component Ak. The design specifications are as follows: • The desired fluorescence intensity to different metal ion concentrations is described as follows: Gref ðI M Þ ¼ 65 þ

5000  1 2 : 1 þ 10I M

(12)

• Input operation range: 103 to 100 μM (National Technical Information Service 1980). • The standard deviations of parameter fluctuations that are supposed to be tolerated in vivo are given by

5

Systematic Design of a Quorum Sensing-Based Biosensor for the. . .

103

ΔPu,i ¼ 0:05Pu,i ,ΔPl,i ¼ 0:05Pl,i ,ΔPu,j ¼ 0:05Pu,j , ΔPu,k ¼ 0:05Pu,k ,ΔPl,k ¼ 0:05Pl,k ,ΔrE ¼ 0:05rE , ΔrR ¼ 0:05rR ,ΔrG ¼ 0:05rG ,Δr ¼ 0:05r, Δm ¼ 0:05m,Δd ¼ 0:05d

(13)

• and the external environmental noise vp, p = 1, 2, 3, for transcription and translation processes, and noise v4, for mature reporter protein expression, are all Gaussian, with zero mean and unit variance. • The cost function is given by 10 ð0

J ðS Þ ¼ E 10




2 Gss ðS, I M Þ  Gref ðI M Þ dI M

(14)

3

Fig. 4 Results for the QS-based metal ion biosensor. By minimizing the cost function in for the metal ion biosensor in Fig. 1, an adequate set S = (M1, C6, A3) is selected from the corresponding libraries in Tables 1-3 in Appendix. The green points are the experimental results (mean of three trials) by S = (M1, C6, A3). The gray solid line is the desired I/O response in (12)

Fluorescence Intensity (a.u.)

In order to efficiently solve the constrained optimal matching design problem of the metal ion biosensor, a GA-based library search method is employed to search a set S from corresponding libraries in Appendix to minimize the cost function (14). The adequate promoter-RBS components from the corresponding libraries in Appendix are found to be M1, C6, and A3. The desired response is shown in Fig. 2, with the fluorescence intensity values taken from Fig. 3 under different Cu(II) ion concentrations. Clearly, the metal ion biosensor can be as close as possible to the specified I/O response despite the parameter fluctuations and environmental noise. Besides, the detection performance is better than that without quorum sensing-based amplifier in Fig. 4 (see “Appendix” for mathematical details) at the same copper ion concentrations (see Figs. 5 and 6). In particular, the QS-based metal ion biosensor increases the sensitivity to Cu(II) ion by roughly one order of magnitude and the dynamic range by roughly fourfold. It confirms the quorum sensing-based metal ion biosensor is useful for enhancing the detection ability (Figs. 7 and 8).

6000 5000 4000 3000 2000 1000 0 10–3

10–2

10–1

Cu(ll) (µM)

100

104

B.-S. Chen

Cu(ll) = 10

–3

Cu(ll) = 10–2 µM

µM 6000 Fluorescence Intensity (a.u.)

Fluorescence Intensity (a.u.)

6000 5000 4000 3000 2000 1000

5000 4000 3000 2000 1000 0

0 0

50 100 150 200 250 300 350 400 450 500

0

Time (min) –1

Cu(ll) = 5×10

Time (min) Cu(ll) = 10–1 µM

µM 6000 Fluorescence Intensity (a.u.)

Fluorescence Intensity (a.u.)

6000 5000 4000 3000 2000 1000 0

0

5000 4000 3000 2000 1000 0

50 100 150 200 250 300 350 400 450 500

0

Time (min)

50 100 150 200 250 300 350 400 450 500 Time (min) Cu(ll) = 100 µM

Cu(ll) = 5×10–1 µM 6000 Fluorescence Intensity (a.u.)

6000 Fluorescence Intensity (a.u.)

50 100 150 200 250 300 350 400 450 500

5000 4000 3000 2000 1000 0

5000 4000 3000 2000 1000 0

0

50 100 150 200 250 300 350 400 450 500 Time (min)

0

50 100 150 200 250 300 350 400 450 500 Time (min)

Fig. 5 Time profiles for the QS-based metal ion biosensor under different concentrations of Cu(II) ion, with measurements taken every 15 min. The green points are the experimental results by S = (M1, C6, A3)

5

Systematic Design of a Quorum Sensing-Based Biosensor for the. . .

Fig. 6 Copper ion biosensor without quorum sensingbased amplifier, using metal ion-induced promoter-RBS component M1 selected from the library in Table 1 in Appendix

Cell 1 Cu(ll)

Cell N RBS

Metal ion-induced promoter-RBS Cell i

Fig. 7 Results for the Cu (ii) ion biosensor without QS-based amplifier. The green points are the experimental results (mean of three trials) by M1 from the library in Table 1 in Appendix

Fig. 8 Comparison of Cu (II) ion biosensors with (top) and without (bottom) QS-based amplifier. The Cu (II) concentrations from left to right are 0, 103, 102, 5  102, 101, 5  101 and 1 μM

105

gfp

106

B.-S. Chen Cell 1

Cu(ll)

Cell N C6-HSL

RBS Metal ion-induced promoter-RBS

luxi

RBS

luxR

Constitutive promoter-RBS

RBS

OmpC CBP

QS-dependent promoter-RBS

QS-based amplifier

Cell i

Fig. 9 A copper sensing and adsorbing system in E. coli

Conclusion In this chapter, a QS-based biosensor is constructed to improve the performance of metal ion detection. Based on four design specifications, the QS-based biosensor can be constructed by selecting adequate promoter-RBS components in combination with a feasible range of metal ion concentrations. The proposed GA-based searching method could provide synthetic biologists with a useful tool to save time in selecting adequate promoter-RBS components to meet a specified detection performance. The experimental results confirm that the quorum sensing-based biosensor can efficiently enhance the performance in metal ion detection. In the future, the mechanism of metal ion detection can be used for environmental bioremediation (Wang and Chen 2009). Bioremediation, which utilizes the ability of biosorption (e.g., protein) to bind the ions using biomass or biopolymers, is a promising approach for environmental treatment. Proteins for biosorption are extensively studied for their specific binding regions and are engineered as biosorbent because of their higher adsorption capacity, low cost, and reusability (Wang and Chen 2009; He and Chen 2014). Research indicates that biosorbents contain a variety of functional sites including carboxyl, phosphate, amino groups, etc., and biology materials, like bacteria, yeast, and fungi, have gained attention for their removal and recovery abilities (Wang and Chen 2009; He and Chen 2014). Ravikumar et al. used a

5

Systematic Design of a Quorum Sensing-Based Biosensor for the. . .

107

recombinant strain based on the specific interaction of Cu(II) with metal-binding peptide, which displayed selective adsorption of Cu(II) from aqueous solutions (Ravikumar et al. 2012). Thus, the QS-based biosensor can be combined with the Cu(II) biosorption system to promote bioremediation (see Fig. 9).

Appendix The steady-state model of the metal ion biosensor without QS-based amplifier is as follows:
 8 > < xGSS ¼ PM Pu,i , Pl,i , xS , I M d þ rG > : Gss ¼ m xG dþr

(15)

where xGSS and Gss are the steady-state concentrations of immature and mature reporter proteins, respectively.

Component Libraries Table 1 Metal ion-induced promoter-RBS component library in E. coli strain DH5α Index M1

Component PcusC-B0034

Pu 100.630

Pl 4.000

M2

PpcoE-B0034

98.237

1.903

K SI ¼ 81 nM K M ¼ 0:3 μM nSI ¼ 1 K SI ¼ 154 nM K M ¼ 0:3 μM nSI ¼ 1

Table 2 Constitutive promoter-RBS component library in E. coli strain DH5α Index C1 C2 C3 C4 C5 C6 C7 C8 C9

Component J23101-B0031 J23101-B0032 J23101-B0034 J23105-B0031 J23105-B0032 J23105-B0034 J23106-B0031 J23106-B0032 J23106-B0034

Pu 48.203 22.263 145.947 5.178 3.449 14.342 8.580 5.813 28.874

108

B.-S. Chen

Table 3 QS-dependent promoter-RBS component library in E. coli strain DH5α Index A1 A2 A3

Component R0062-B0031 R0062-B0032 R0062-B0034

Pu 362.902 238.132 848.678

Pl 1.100 1.000 1.740

KRI = 95 nM KI = 3.8 nM nRI = 1

References Achtman M, Suerbaum S (2001) Helicobacter pylori: molecular and cellular biology. Horizon Scientific, Wymondham Alon U (2007) An introduction to systems biology: Design principles of biological circuits. Chapman & Hall/CRC Press, Boca Raton National Technical Information Service (1980) Ambient water quality criteria for copper. National Technical Information Service, Washington, DC/Springfield Bondarczuk K, Piotrowska-Seget Z (2013) Molecular basis of active copper resistance mechanisms in gram-negative bacteria. Cell Biol Toxicol 29:397–405 Borremans B, Hobman JL, Provoost A, Brown NL, van Der Lelie D (2001) Cloning and functional analysis of the PBR lead resistance determinant of Ralstonia metallidurans CH34. J Bacteriol 183:5651–5658 Brint JM, Ohman DE (1995) Synthesis of multiple exoproducts in Pseudomonas aeruginosa is under the control of RhlR-RhlI, another set of regulators in strain PAO1 with homology to the autoinducer-responsive LuxR-LuxI family. J Bacteriol 177:7155–7163 Brown NL, Stoyanov JV, Kidd SP, Hobman JL (2003) The MerR family of transcriptional regulators. FEMS Microbiol Rev 27:145–163 Cameron DE, Bashor CJ, Collins JJ (2014) A brief history of synthetic biology. Nat Rev Microbiol 12:381–390 Chang YC, Lin CL, Jennawasin T (2013) Design of synthetic genetic oscillators using evolutionary optimization. Evol Bioinform 9:137 Chen BS, Wang YC (2006) On the attenuation and amplification of molecular noise in genetic regulatory networks. BMC Bioinform 7:52 Chen B-S, Wang Y-C (2014) Synthetic gene network: modeling, analysis, robust design methods. CRC Press, Boca Raton Chen BS, Cheng YM, Lee CH (1995) A genetic approach to mixed H-2/H-infinity optimal pid control. IEEE Control Syst Mag 15:51–60 Danino T, Mondragon-Palomino O, Tsimring L, Hasty J (2010) A synchronized quorum of genetic clocks. Nature 463:326–330 Darch SE, West SA, Winzer K, Diggle SP (2012) Density-dependent fitness benefits in quorumsensing bacterial populations. Proc Natl Acad Sci U S A 109:8259–8263 Davies MJ (2005) The oxidative environment and protein damage. Biochim Biophys Acta-Protein Proteom 1703:93–109 Dunlap PV, Kuo A (1992) Cell density-dependent modulation of the Vibrio-fischeri luminescence system in the absence of autoinducer and luxr protein. J Bacteriol 174:2440–2448 Engebrecht J, Silverman M (1984) Identification of genes and gene-products necessary for bacterial bioluminescence. Proc Natl Acad Sci USA Biol Sci 81:4154–4158 Franke S, Grass G, Rensing C, Nies DH (2003) Molecular analysis of the copper-transporting efflux system CusCFBA of Escherichia coli. J Bacteriol 185:3804–3812 Fuqua WC, Winans SC, Greenberg EP (1994) Quorum sensing in bacteria – the luxr-luxi family of cell density-responsive transcriptional regulators. J Bacteriol 176:269–275 Hanzelka BL, Greenberg EP (1996) Quorum sensing in Vibrio fischeri: evidence that S-adenosylmethionine is the amino acid substrate for autoinducer synthesis. J Bacteriol 178:5291–5294

5

Systematic Design of a Quorum Sensing-Based Biosensor for the. . .

109

Hartwig A (1995) Current aspects in metal genotoxicity. BioMetals 8:3–11 He J, Chen JP (2014) A comprehensive review on biosorption of heavy metals by algal biomass: materials, performances, chemistry, modeling simulation tools. Bioresour Technol 160:67–78 Hobman JL, Julian DJ, Brown NL (2012) Cysteine coordination of Pb(II) is involved in the PbrRdependent activation of the lead-resistance promoter, PpbrA, from Cupriavidus metallidurans CH34. BMC Microbiol 12:109 Kaplan HB, Greenberg EP (1985) Diffusion of autoinducer is involved in regulation of the Vibrio-fischeri luminescence system. J Bacteriol 163:1210–1214 Khalil AS, Collins JJ (2010) Synthetic biology: applications come of age. Nat Rev Genet 11:367–379 Kotula JW, Kerns SJ, Shaket LA, Siraj L, Collins JJ, Way JC et al (2014) Programmable bacteria detect and record an environmental signal in the mammalian gut. Proc Natl Acad Sci U S A 111:4838–4843 Kyung Hyuk K, Kiri C, Bartley B, Sauro HM (2015) Controlling E. coli gene expression noise. IEEE Trans Biomed Circuits Syst 9(4):497–504 Lee SM, Grass G, Rensing C, Barrett SR, Yates CJD, Stoyanov JV et al (2002) The PCO proteins are involved in periplasmic copper handling in Escherichia coli. Biochem Biophys Res Commun 295:616–620 Migaszewski ZM, Galuszka A (2015) The characteristics, occurrence, geochemical behavior of rare earth elements in the environment: a review. Crit Rev Environ Sci Technol 45:429–471 Miller MB, Bassler BL (2001) Quorum sensing in bacteria. Annu Rev Microbiol 55:165–199 Mulligan CN, Yong RN, Gibbs BF (2001) Remediation technologies for metal-contaminated soils and groundwater: an evaluation. Eng Geol 60:193–207 Munson GP, Lam DL, Outten FW, O’Halloran TV (2000) Identification of a copper-responsive twocomponent system on the chromosome of Escherichia coli K-12. J Bacteriol 182:5864–5871 Nealson KH, Hastings JW (1979) Bacterial bioluminescence: its control and ecological significance. Microbiol Rev 43:496–518 Ngah WSW, Hanafiah MAKM (2008) Removal of heavy metal ions from wastewater by chemically modified plant wastes as adsorbents: a review. Bioresour Technol 99:3935–3948 Passador L, Cook JM, Gambello MJ, Rust L, Iglewski BH (1993) Expression of Pseudomonas aeruginosa virulence genes requires cell-to-cell communication. Science 260:1127–1130 Pearson JP, Passador L, Iglewski BH, Greenberg EP (1995) A second N-acylhomoserine lactone signal produced by Pseudomonas aeruginosa. Proc Natl Acad Sci U S A 92:1490–1494 Raha S, Robinson BH (2000) Mitochondria, oxygen free radicals, disease and ageing. Trends Biochem Sci 25:502–508 Ravikumar S, Ganesh I, Yoo IK, Hong SH (2012) Construction of a bacterial biosensor for zinc and copper and its application to the development of multifunctional heavy metal adsorption bacteria. Process Biochem 47:758–765 Rezvani-Boroujeni A, Javanbakht M, Karimi M, Shahrjerdi C, Akbari-adergani B (2015) Immoblization of thiol-functionalized nanosilica on the surface of poly(ether sulfone) membranes for the removal of heavy-metal ions from industrial wastewater samples. Ind Eng Chem Res 54:502–513 Rouch DA, Brown NL (1997) Copper-inducible transcriptional regulation at two promoters in the Escherichia coli copper resistance determinant pco. Microbiology 143:1191–1202 Ruby EG (1996) Lesson from a cooperative, bacteria-animal association: The Vibrio fischeriEuprymna scolopes light organ symbiosis. Annu Rev Microbiol 50:591–624 Ruby EG, Mcfallngai MJ (1992) A squid that glows in the night – development of an animalbacterial mutualism. J Bacteriol 174:4865–4870 Ruby EG, Nealson KH (1976) Symbiotic association of Photobacterium fischeri with the marine luminous fish Monocentris japonica; a model of symbiosis based on bacterial studies. Biol Bull 151:574–586 Silva-Rocha R, de Lorenzo V (2014) Engineering multicellular logic in bacteria with metabolic wires. ACS Synth Biol 3:204–209

110

B.-S. Chen

Singha AS, Guleria A (2014) Chemical modification of cellulosic biopolymer and its use in removal of heavy metal ions from wastewater. Int J Biol Macromol 67:409–417 Sohka T, Heins RA, Phelan RM, Greisler JM, Townsend CA, Ostermeier M (2009) An externally tunable bacterial band-pass filter. Proc Natl Acad Sci USA 106:10135–10140 Soltani M, Vargas-Garcia CA, Singh A (2015) Conditional moment closure schemes for studying stochastic dynamics of genetic circuits. IEEE Trans Biomed Circuits Syst 9(4):518–526 Stricker J, Cookson S, Bennett MR, Mather WH, Tsimring LS, Hasty J (2008) A fast, robust and tunable synthetic gene oscillator. Nature 456:516–U39 Taghavi S, Mergeay M, Nies D, Vanderlelie D (1997) Alcaligenes eutrophus as a model system for bacterial interactions with heavy metals in the environment. Res Microbiol 148:536–551 Teo JJY, Sung Sik W, Sarpeshkar R (2015) Synthetic biology: a unifying view and review using analog circuits. IEEE Trans Biomed Circuits Syst 9(4):453–474 Teodosiu C, Wenkert R, Tofan L, Paduraru C (2014) Advances in preconcentration/removal of environmentally relevant heavy metal ions from water and wastewater by sorbents based on polyurethane foam. Rev Chem Eng 30:403–420 Tetaz TJ, Luke RK (1983) Plasmid-controlled resistance to copper in Escherichia coli. J Bacteriol 154(Jun):1263–1268 Val DL, Cronan JE (1998) In vivo evidence that S-adenosylmethionine and fatty acid synthesis intermediates are the substrates for the luxi family of autoinducer synthases. J Bacteriol 180:2644–2651 Wang J, Chen C (2009) Biosorbents for heavy metals removal and their future. Biotechnol Adv 27:195–226 Wang BJ, Kitney RI, Joly N, Buck M (2011) Engineering modular and orthogonal genetic logic gates for robust digital-like synthetic biology. Nat Commun 2:508

6

Riboswitches as Sensor Entities Svetlana Harbaugh, Michael Goodson, Yaroslav Chushak, Jorge Chávez, and Nancy Kelley-Loughnane

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Development of Riboswitches Responsive to Small Molecule Analytes . . . . . . . . . . . . . . . . . . . . . . Aptamer Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Riboswitch Selection In Vivo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Reengineering Natural Riboswitches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Riboswitch Selection In Vitro . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Rational and Computational Design of Synthetic Riboswitches . . . . . . . . . . . . . . . . . . . . . . . . . . . Riboswitch Detection Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Monitoring Functionality of Synthetic Riboswitches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Coupling Synthetic Riboswitches with New Reporter Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . Riboswitch Optimization toward Sensing Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Selectivity, but at a Cost . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Biological Circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Riboswitch Circuitry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Corollary Advantages of Signal Amplification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tandem Riboswitches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Applications of Synthetic Riboswitches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

112 115 115 118 121 122 122 125 126 127 129 129 129 130 130 131 132 134 135

S. Harbaugh (*) · M. Goodson · J. Chávez · N. Kelley-Loughnane 711th Human Performance Wing, Air Force Research Laboratory, Dayton, OH, USA e-mail: [email protected]; [email protected]; [email protected]; [email protected] Y. Chushak Henry M. Jackson Foundation for the Advancement of Military Medicine, Dayton, OH, USA e-mail: [email protected] © This is a U.S. Government work and not under copyright protection in the US; foreign copyright protection may apply 2022 G. Thouand (ed.), Handbook of Cell Biosensors, https://doi.org/10.1007/978-3-030-23217-7_121

111

112

S. Harbaugh et al.

Abstract

Riboswitches are regulatory noncoding RNAs, predominantly located in the 50 untranslated region of mRNA, that can serve as molecular switches able to regulate the level of gene expression. This occurs through the conformational changes caused by binding to a specific metabolite. Riboswitches contain two structural domains: an aptamer domain that senses and binds to a metabolite and an expression platform that undergoes a conformational change in response to aptamer-ligand binding resulting in regulation of expression of downstream gene. In addition to natural riboswitches found in living organisms, a variety of synthetic riboswitches that respond to nonnatural small molecules have been developed. Synthetic riboswitches can be engineered to regulate expression of any gene in response to any nonnatural molecule that is capable of being bound by RNA and is not toxic to cells. This feature demonstrates a strong possibility for RNA switches to serve as sensor entities for design and development of cellbased biosensors with a variety of different applications. This chapter gives an overview of riboswitch selection techniques, describes reporter systems for monitoring riboswitch activation and approaches for riboswitch tuning and performance optimization in order to fulfill biosensor requirements, and discusses riboswitch applications as sensor entities. Keywords

Synthetic riboswitch · Aptamer · Gene regulation · SELEX · In vivo selection · Biosensor

Introduction Living systems have developed the ability to control their gene expression patterns in response to changes in the extracellular and intracellular environment. While the majority of regulatory mechanisms involve protein action, genetic regulation by noncoding RNAs, or riboswitches, is widespread in bacteria (Mandal and Breaker 2004; Nudler and Mironov 2004; Winkler and Breaker 2005; Serganov and Nudler 2013) and is also found in some eukaryotes (Cheah et al. 2007; Wachter 2010; Serganov and Nudler 2013). Riboswitches are predominantly located in the 50 untranslated region (50 -UTR) of mRNA; however, they can be present in the 30 UTR in some eukaryotic mRNA. Similar to regulatory proteins, riboswitches act as molecular sensors able to regulate the level of gene expression through the conformational changes caused by binding to a specific ligand. Structurally, riboswitches are composed of two functional domains: an aptamer domain that senses and binds to a ligand and an expression platform that controls the expression of a downstream gene by changing its conformation in response to the ligand-induced changes in the aptamer domain (Tucker and Breaker 2005; Nudler 2006; Edwards and Batey 2010). Riboswitches can up- or downregulate gene

Riboswitches as Sensor Entities

Fig. 1 Examples of riboswitch mechanisms. Bacterial riboswitches upregulate or downregulate gene expression operating at level of transcription, translation, mRNA degradation, or splicing. Aptamer is colored green. Terminator hairpin is colored red. RNAse cleavage site is colored grey. (a) Ligand (L) binding to the aptamer causes formation of a terminator hairpin leading to transcription termination. (b) In the absence of ligand, an intrinsic terminator stops transcription. Ligand binding to the aptamer induces a conformational shift that forms an antiterminator, enabling expression of downstream gene. (c) Ligand binding to the aptamer causes an alternative structure to form, blocking the ribosome binding site (RBS) and preventing initiation of translation. (d) The RBS is blocked in the absence of ligand. Upon ligand binding to the aptamer, the RNA undergoes a conformational shift, revealing the RBS and enabling translation. (e) Upon ligand binding to the aptamer, the riboswitch adopts a conformation that exposes RNase cleavage site leading to mRNA degradation. (f) Ligand binding to the aptamer leads to mRNA self-cleavage and splicing that brings together two halves of the RBS, and the resulting mRNA is efficiently translated

6 113

114

S. Harbaugh et al.

expression employing a variety of regulatory mechanisms (Fig. 1). The mechanism of riboswitch action is based on the existence of two thermodynamically stable conformations (“off” and “on” states), separated by the energy barrier which inhibits spontaneous switching. In general, when the concentration of a recognized molecule reaches the binding threshold value, the interaction with the aptamer domain takes place. This promotes the switching to the alternative conformation by stabilization of the intermediate and final states (Edwards and Batey 2010; Machtel et al. 2016). The most frequently riboswitch-mediated changes in gene expression occur either transcriptionally or translationally. The expression platform for a riboswitch that acts during transcription typically involves the ligand-dependent formation of an intrinsic terminator or antiterminator structure (Blouin et al. 2009; Peselis and Serganov 2014; Topp and Gallivan 2010; Lemay et al. 2011; Gong et al. 2017b); Hallberg et al. 2017). In contrast, riboswitches that operate at a translational level most often function by masking or releasing the Shine-Dalgarno (SD) sequence (also known as the ribosome binding site; RBS) in a ligand-dependent fashion. When the SD sequence is released, the ribosome can bind to the mRNA and permit translation; masking the SD sequence represses translation (Blouin et al. 2009; Peselis and Serganov 2014; Topp and Gallivan 2010; Lemay et al. 2011; Hallberg et al. 2017; Antunes et al. 2018). To date, numerous examples of naturally occurring riboswitches responding to a variety of ligands (enzyme cofactors, nucleotide precursors, amino acids, and metal ions) as well as changes in temperature and pH-value have been discovered (Coppins et al. 2007; Barrick and Breaker 2007; Roth and Breaker 2009; Serganov and Nudler 2013; Peselis and Serganov 2014; Hallberg et al. 2017; Pham et al. 2017). The size and structure of riboswitch ligands vary significantly, an indication that riboswitches could be used to detect a wide range of targets. In fact, some natural riboswitches found applications in enzyme and strain engineering, in controlling gene expression and cellular physiology, and in real-time imaging of cellular metabolites and signals (Topp and Gallivan 2010; Machtel et al. 2016; Hallberg et al. 2017). This apparent versatility of riboswitches in nature is being exploited by researchers in order to develop synthetic riboswitches that regulate gene expression in response to a desired target molecule. The advantage of such engineered riboswitches is that they offer a way to control gene expression via nonnatural molecules, for example, drugs or explosives. Moreover, synthetic riboswitches are of particular interest and demand since the majority of natural riboswitches respond to essential metabolites and, therefore, their use as sensing elements can be compromised by fluctuations in intracellular metabolite concentrations, and the exogenous addition of these compounds could negatively impact normal cellular function. This chapter gives an overview of riboswitch selection techniques, describes reporter systems for monitoring riboswitch activation and approaches for riboswitch tuning and performance optimization in order to fulfill biosensor requirements, and discusses riboswitch applications as sensor entities.

6

Riboswitches as Sensor Entities

115

Development of Riboswitches Responsive to Small Molecule Analytes Different methods have been applied to develop synthetic riboswitches that respond to a variety of different target analytes (Wittmann and Suess 2012; Groher and Suess 2014; Berens and Suess 2015; Gong et al. 2017a; Etzel and Mörl 2017; Findeiß et al. 2017). Similar to their natural counterparts, synthetic riboswitches are composed of the same structural domains and can be designed to downregulate or upregulate gene expression at transcriptional or translational levels. To create new riboswitches, some researchers follow a bottom-up approach, in which individual riboswitch building blocks are designed de novo and assembled into functional regulators. Another approach to generate synthetic riboswitches is reengineering natural riboswitches by altering the specificity of the aptamer domain or modifying the expression platform. Both approaches require time consuming in vivo screening with analysis of large number of riboswitch clones before a functional riboswitch can be obtained. In order to speed up the selection process, researchers employ new strategies based on rational and computational design for synthetic riboswitches.

Aptamer Selection Typically, synthetic riboswitch development starts with selection of an analyte sensing domain, the aptamer. Synthetic riboswitches can be designed to respond to different targets by integrating different aptamers as recognition elements. Suitable aptamers must bind a molecule of interest (that, in most cases, does not resemble a natural cell compound) with high affinity and specificity, and undergo sufficient changes in structure and/or stability upon an analyte binding. Such aptamers can be generated using a method known as Systematic Evolution of Ligands by Exponential Enrichment (SELEX) (Ellington and Szostak 1990; Tuerk and Gold 1990) depicted in Fig. 2. In a typical SELEX experiment, the target molecules are usually immobilized on a solid surface to enable an efficient separation of target-binding and nonbinding oligonucleotides. The procedure begins with 1013–1015 unique sequences from a random oligonucleotide pool that compete to bind a target. This library consists of sequences designed with two PCR primer regions flanking a random region of, typically, 30–50 nucleotides. After incubation with the immobilized target, nonbinding oligonucleotides are removed by washing, whereas bound oligonucleotides are specifically eluted with free target. The binding sequences are reverse transcribed, amplified, and subjected to the next round of selection. The process is repeated in cyclical fashion until the final pool is enriched for sequences binding to the target. Typically, after the first round, researchers may institute a counter-selection step in which sequences binding to a control (such as the support matrix alone or a molecule similar in structure to the target) are removed from solution, and those that do not bind to the control are retained for future SELEX

Fig. 2 Example of selection strategy to generate synthetic riboswitches functional in E. coli cells using a combination of in vitro and in vivo selections. Riboswitch development begins with aptamer selection. RNA aptamers are isolated from large pools of randomized sequences using an in vitro selection technique called SELEX. After incubation with the immobilized target, nonbinding RNA oligonucleotides are removed by washing whereas bound oligonucleotides are specifically eluted with free target. The binding RNA oligonucleotide sequences are reverse transcribed, amplified, and subjected to the next round of selection. The process is repeated in cyclical fashion until the final pool is enriched for sequences binding to the target with high affinity. To select for synthetic riboswitches, a library of aptamers is cloned upstream of a randomized sequence (expression platform) in the 50 -untranslated region (UTR) of a reporter gene. The created riboswitch plasmid library is transformed in E. coli cells and screened without and in the presence of target analyte. Colonies of E. coli cells showing an increase in the reporter gene expression (increase in the fluorescence intensity or fluorescence color change are shown as examples) are selected, riboswitch containing plasmids are isolated and sequenced

116 S. Harbaugh et al.

6

Riboswitches as Sensor Entities

117

rounds. Gradually increasing the stringency during the selection cycles can lead to aptamers that bind a desired target with high affinity. A small portion of the pool obtained in the last selection round is sequenced and the binders are identified. While SELEX has been tremendously successful in attaining high affinity aptamers for a variety of targets (Jenison et al. 1994; Berens et al. 2001; Weigand et al. 2008; Ehrentreich-Förster et al. 2008; Carothers et al. 2010; Filonov et al. 2014; Han et al. 2014; Xu et al. 2016; Xiu et al. 2017; Groher et al. 2018), it has some disadvantages. First of all, the standard SELEX process tends to be time consuming and uses a large amount of reagents making aptamer selection costly. Also, immobilization of the target may change its original conformation; steric hindrance of the immobilized target may block the binding site, and nonspecific binding to a solid substrate may lead to the selection of nonspecific sequences. To overcome these disadvantages, several modified SELEX methods have been set up. Combinations of traditional SELEX with a microfluidic system (Hybarger et al. 2006), and later integration of the magnetic bead-based SELEX with microfluidic technology and a continuousflow magnetic activated chip-based separation device (Lou et al. 2009), resulted in a more efficient, rapid, and automatic aptamer selection system. To reduce time and reagent consumption, Szeto et al. developed a modified version of SELEX called RNA Aptamer Isolation via Dual-cycles (RAPID) (Szeto et al. 2013). This method incorporates nonamplification cycles in which the eluted RNA is purified and used in a second binding cycle without prior amplification of the material. To compare the techniques, RAPID was performed in parallel with traditional SELEX using the same targets, and many of the same top aptamer candidates (~10% identical aptamers in the top 10,000 enriched sequences) emerged. Both selections using RAPID and traditional SELEX started with a library of 5  1015 unique sequences, yet the RAPID selection took approximately one third the time as the SELEX process and significantly reduced the amount of reagents used (Szeto et al. 2013). To perform a multiplexed RNA aptamer selection, the same group developed a microcolumn-based device, MEDUSA (Microplate-based Enrichment Device Used for the Selection of Aptamers) (Szeto and Craighead 2014). This device is designed around a 96-well microplate format, consisting of 96 microcolumns packed with a resin. Various targets are individually immobilized on a resin in each column and the nucleic acid library is pumped through the columns in serial and/or parallel mode. Combination of the multi-well format MEDUSA device with the RAPID approach allowed simultaneous multiplexed aptamer selection to 19 different targets to be performed, that significantly reduced the time and reagents needed for selection (Reinholt et al. 2016). Another modified version of SELEX is Structure-Switching or Capture SELEX (Morse 2007). This technique is the opposite of above mentioned SELEX methods with regards to target immobilization, and relies on a conformational change in the aptamer sequence upon target binding, coined as “structure-switching.” Rather than immobilizing the target to the solid support matrix of choice, an oligonucleotide pool is tethered and the target is free in solution. To tether the pool to the matrix (often magnetic spherical polymer beads), a short oligonucleotide strand complementary to the 50 -region of the RNA library is conjugated to the matrix. The library is

118

S. Harbaugh et al.

hybridized to the short oligonucleotide and the pool-labeled beads are exposed to the target. Only the sequences that change structure conformation upon binding to the target will effectively release themselves from the immobilized short oligonucleotide. Sequences found in solution following incubation with the target are sequestered, converted to cDNA, and amplified to produce the transcription template for the next round of selection. Capture SELEX is ideal for molecules that are difficult to immobilize, such as small molecules, and does not require modification of the analyte. Modifying a small molecule can block groups that are critical for binding, which can affect the selectivity of identified aptamers. Although some limitations are still present, the considerable contributions to improving SELEX allow for better selections, thus yielding high affinity binders.

Riboswitch Selection In Vivo The first synthetic riboswitches were generated using a simple and straight forward approach in which in vitro-selected small molecule-binding aptamers were inserted into the 50 -UTR of mRNAs, without being fused to an expression platform domain. Werstuck and Green inserted aptamers specific to Hoechst dye 33258, and the closely related drug H33342, into the 50 -UTR of a mammalian β-galactosidase gene (Werstuck and Green 1998). When bound to their ligands, the aptamers formed a stable structure that resulted in blocking either the scanning ribosomal small subunit or the ribosome-mRNA interaction leading to down-regulation of gene expression in vivo. Later, Harvey et al. (2002) demonstrated that insertion of theophylline or biotin binding aptamers into the 50 -UTR of a eukaryotic chloramphenicol acetyltransferase gene led to inhibition of mRNA translation in the presence of the corresponding ligands. It was found that small molecule ligand–RNA interactions were sufficiently stable and prevented 80S ribosome assembly on the mRNA template. Similarly, Suess et al. (2003) and Hanson et al. (2003) introduced aptamers selected against tetracycline that behaved as highly efficient ribosomeblocking elements into the 50 -UTRs of reporter genes in yeast. To demonstrate that synthetic riboswitches can be used to perform genetic screens and selections for the presence of small molecules in Escherichia coli, Desai and Gallivan subcloned the theophylline aptamer sequence at a location five base pairs upstream of the ribosome binding site (RBS) of the β-galactosidase reporter gene (Desai and Gallivan 2004). The aptamer insertion led to a theophylline-dependent upregulation of β-galactosidase gene expression. The riboswitch performance was further optimized by increasing the distance between the aptamer and RBS to eight nucleotides, resulting in a ten-fold increase in β-galactosidase expression in the presence of theophylline compared to the system without the analyte. It was determined that the created riboswitch activated gene expression at translational level. The theophylline aptamer was also used to engineer a riboswitch for regulation of pre-mRNA splicing in HeLa nuclear extracts (Kim et al. 2005). Insertion of this aptamer into the 30 splice site region of a model pre-mRNA enabled its splicing to be repressed by theophylline addition. Weigand and Suess (2007) applied the synthetic tetracycline binding

6

Riboswitches as Sensor Entities

119

riboswitch to establish a gene expression system for conditional tetracycline-dependent control of pre-mRNA splicing in yeast. Efficient regulation was obtained when the aptamer was inserted close to the 50 splice site (SS) with the consensus sequence of the SS located within the aptamer stem. Structural probing indicated limited spontaneous cleavage within this stem in the absence of the ligand. Addition of tetracycline led to tightening of the stem and the whole aptamer structure which prevented recognition of the 50 SS (Weigand and Suess 2007). Although the above mentioned riboswitch constructs demonstrated regulation of gene expression in response to specific analytes, they actually did not correspond to the structural composition of natural riboswitches (aptamer-expression platform), and as a result they did not completely repress gene expression and could be further optimized. To achieve more sufficient gene regulation with synthetic riboswitches, more sophisticated design and selection strategies engaging in vivo screening for riboswitch functionality were developed. Combination of in vitro and in vivo selection is a useful strategy to identify riboswitches with desired functionalities (Fig. 2). It should be noted that, traditionally, SELEX experiments have been designed to find the tightest-binding RNAs. However, using a single aptamer with high affinity to a specific analyte in a riboswitch selection does not always result in functional riboswitches (Suess et al. 2003). To control gene expression, riboswitches must not only bind the ligand, they must also undergo a conformational change on a physiologically relevant timescale. The enriched library of aptamers incorporated into the riboswitch architecture will help to reduce the sequence search space by offering a higher percentage of potential analyte binders. To generate functional riboswitches, an expression platform is generally added to an in vitro-selected aptamer or a library of aptamers. An expression platform is usually a fully randomized region of 12–30 bases introduced between the aptamer and a region of 5–7 constant bases located immediately before the start codon. The introduction of a completely randomized nucleotide region allows an in vivo selection to identify the functional expression platform, and optimization of the strength of the ribosome-binding site to achieve a high level of reporter protein expression. The resultant aptamer-expression platform library is cloned upstream of a reporter gene to generate a signal output for selection (Fig. 2). A variety of methods for screening and selection of robust-performing synthetic riboswitches from RNA libraries in different cell types have been reported in the literature (Lynch et al. 2007; Topp and Gallivan 2008a; Wieland and Hartig 2008; Weigand et al. 2008; Lynch and Gallivan 2009). Using a screening method based on green fluorescent protein expression, Suess and coworkers screened libraries of up to 50,000 members and isolated riboswitches that reduce gene expression 7.5-fold in the presence of the antibiotic neomycin in Saccharomyces cerevisiae yeast cells (Weigand et al. 2008). Gallivan and coworkers developed a high-throughput screen (Lynch et al. 2007) for theophylline-dependent riboswitches in E. coli cells. In their screening method, the theophylline binding aptamer, followed by randomized sequence of nucleotides, was placed upstream of the β-galactosidase reporter gene. The riboswitch libraries of up to 65,000 members were screened without and in the presence of theophylline using a β-galactosidase assay on solid media and in cellular

120

S. Harbaugh et al.

lysates. Using this screening assay, they identified a new riboswitch that could activate the expression of β-galactosidase by 36-fold in the presence of theophylline in E. coli cells. The same screening strategy also allowed for the identification of a theophylline-dependent riboswitch that could repress the expression of β-galactosidase by 27-fold in the presence of analyte (Topp and Gallivan 2008b). The highthroughput screen was applied to develop a series of synthetic theophylline-sensitive riboswitches that functioned as genetic control elements in a diverse set of gramnegative and gram-positive bacteria (Topp et al. 2010). An alternative method for identification of functional riboswitches is a high-throughput selection, which is based on ligand-induced changes in cell motility. Utilizing cheZ as a reporter gene, responsible for migration of E. coli cells on semisolid media, Gallivan and coworkers performed positive and negative selections for isolation of new theophylline responsive riboswitches from libraries of more than 105 E. coli clones (Topp and Gallivan 2008a). Another example of utilizing an in vitro-selected aptamer to generate a synthetic riboswitch is the development of 2,4-dinitrotoluene (DNT)-responsive riboswitch in E. coli cells (Davidson et al. 2013). A riboswitch library was constructed by incorporation of 30 degenerate bases between an in vitro-selected 2,4,6-trinitrotoluene (TNT) aptamer (Ehrentreich-Förster et al. 2008) and the RBS. Screening was performed by placing the riboswitch library upstream of the Tobacco Etch Virus (TEV) protease coding sequence in one plasmid; a second plasmid encoded a FRETbased construct linked with a peptide containing the TEV protease cleavage site. Although the aptamer was selected for efficient TNT binding, it turned out to also bind DNT. Addition of DNT to bacterial cell culture activated the riboswitch, initiating translation of TEV protease. The produced protease cleaved the linker in the FRET-based fusion protein, causing a change in fluorescence. The DNT-responsive riboswitch exhibited a 10-fold increase in fluorescence in the presence of 0.5 mM DNT compared to the system without target. Using an in vitro-selected ciprofloxacin binding aptamer, Suess and coworkers developed a ciprofloxacin-responsive riboswitch by next-generation sequencing (NGS)-guided cellular screening (Groher et al. 2018). The application of NGS allows the collection of detailed information for the individual selection rounds. Thus, it was possible to choose selection rounds that showed a certain degree of enrichment, yet maintained maximum diversity. This approach allows for a substantial acceleration of the transition between in vitro and in vivo, while simultaneously reducing screening efforts. Advances in the development of fluorescence-activated cell sorting (FACS) facilitates application of this technique for quick and efficient screening of large (~108 members) libraries of riboswitch-like sequences to identify those with desired activity. Because FACS can readily distinguish small differences in fluorescence emission intensity, it was successfully used for discovering synthetic riboswitches that display low background levels of gene expression in the absence of a ligand and robust increases in gene expression in its presence (Fowler et al. 2008; Lynch and Gallivan 2009; Ghazi et al. 2014). Using FACS assay, Lynch and

6

Riboswitches as Sensor Entities

121

Gallivan identified a theophylline synthetic riboswitch that activates protein translation to 96-fold in E. coli.

Reengineering Natural Riboswitches An alternative strategy to generate synthetic riboswitches is to reengineer natural riboswitches. Despite advances in the development of SELEX methods for aptamer selection, high-throughput genetic screens, and selection strategies for creation of novel riboswitches with desired gene regulatory functions, only a very limited range of in vitro-selected aptamers have been successfully exploited in artificial riboswitch applications (Berens and Suess 2015; Findeiß et al. 2017; Hallberg et al. 2017). A variety of existing natural riboswitches offers a huge possibility to use these regulatory elements for biosensing applications. However, as was mentioned above, the major problem with using natural riboswitches is that their ligands are typically metabolites ordinarily present at varying levels in the cell, and, therefore, any activation due to binding the target molecule would result in unwanted background activity. To overcome this problem, Micklefield and coworkers have reengineered natural riboswitches to bind nonnatural synthetic small molecules, while the original natural target molecules were no longer recognized (Dixon et al. 2010; Robinson et al. 2014; Wu et al. 2015). Using randomization by site-directed mutagenesis with genetic selection, they reprogrammed two natural riboswitches, an adenine-sensing add A-riboswitch from Vibrio vulnificus and a queuosine precursor-responsive PreQ1 class I riboswitch from Bacillus subtilis, to respond to synthetic triazinebased and pyrimidopyrimidine ligands and diamino-faced analogs of preQ1, respectively. A majority of naturally found riboswitches downregulate gene expression upon metabolite binding, probably because of their roles in negative feedback regulation within the metabolic pathways. However, for sensing applications, riboswitch activation is more desirable to produce a positive signal when the aptamer binds the target. A reengineering approach was applied to reverse a translational expression platform of a natural thiamine pyrophosphate (TPP)-responsive riboswitch which originally downregulates gene expression in E. coli cells (Nomura and Yokobayashi 2007). To identify riboswitches that upregulate gene expression upon TPP binding, a dual selection strategy based on antibiotic resistance was developed. A randomized region of up to 30 nucleotides was inserted between the TPP aptamer and the RBS, and the resultant riboswitch library was placed upstream of the tetracycline resistance gene tetA. In a dual-selection strategy, E. coli cells harboring functional riboswitches were able to grow in medium containing tetracycline and TPP but couldn’t survive in the presence of Ni+2 and without TPP. The dual selection strategy permitted the analysis of large riboswitch libraries and allowed sufficient elimination of false positives. Besides the reprogramming of a natural expression platform for the generation of upregulating riboswitches, the reciprocal redesign of upregulating riboswitches to downregulating was also successful (Muranaka et al. 2009a).

122

S. Harbaugh et al.

Riboswitch Selection In Vitro A promising variation of a SELEX-type approach for riboswitch selection in vitro with nonimmobilized small molecule target was performed by Martini et al. (2015). The researchers utilized a library version of a synthetic riboswitch for thiamine pyrophosphate (TPP). This library was modified from the original riboswitch, such that four regions responsible for riboswitch activity were randomized, and the RNA pool and TPP were free in solution. Upon binding of a sequence to the target, a change in conformation occured, exposing an area for a short double stranded biotinlabeled DNA reporter with a complementary sequence to bind. Upon binding to the RNA sequence, a strand on the reporter was displaced. Sequences which have bound to the reporter were isolated with streptavidin beads whereas those that did not change conformation couldn’t bind the biotin-labeled reporter and thus were discarded. Using this approach, only three rounds were required to obtain a number of functional TPP-responsive riboswitches.

Rational and Computational Design of Synthetic Riboswitches To rationally design new synthetic riboswitches, it is crucial to understand the fundamental mechanisms of riboswitch functionality. As was mentioned above, there are two main classes of riboswitches: transcriptional riboswitches that regulate production of full-length mRNA by controlling the formation of transcription terminators and translational riboswitches that alter protein expression by changing the initiation of mRNA translation upon the ligand binding. Furthermore, increasing the ligand concentration can increase protein synthesis (where the riboswitch upregulates gene expression and performs as an ON switch) or decrease it (the riboswitch downregulates gene expression and performs as an OFF switch). For biosensing applications the ON switches are preferred. Although, transcriptional and translational riboswitches function by different specific mechanisms, they have the same global mechanism in which the mRNA molecule can fold reversibly into two distinct conformations that are associated with a different level of protein expression and different ligand-binding affinities. Mathematically, riboswitches can be described by a simple three-state population-shift model where binding of the target ligand shifts a preexisting equilibrium between the OFF and ON conformations (Lynch et al. 2007; Vallée-Bélisle et al. 2009; Beisel and Smolke 2009). In case of the transcriptional ON switch (Fig. 3), mRNA folds into a stable conformation where the aptamer domain forms a terminator hairpin with the complementary expression platform, thus leading to a termination of transcription (“off” state). This conformation also has a low ligand-binding affinity K2 = 0. The ON conformation, in which terminator is disrupted, has a high mRNA translational rate KP and ligand-binding affinity K2, but it is unstable without the presence of ligand. The equilibrium between these two conformations is characterized by the equilibrium constant K1. In the presence of the ligand, conformation “on” +L becomes stable leading to a large increase in the level of protein expression. Therefore, the

6

Riboswitches as Sensor Entities

123

ligand thermodynamically shifts the equilibrium from conformation “off” to conformation “on.” Translational riboswitch functions in a similar way (Fig. 4). In the absence of a ligand, the aptamer extensively binds with the expression platform that includes the RBS, thus making it inaccessible for the ribosome and preventing protein expression. Addition of the ligand stabilizes the “on” conformation in which the RBS is unpaired and available for ribosome binding leading to a large increase in the level of protein expression. It was shown (Vallée-Bélisle et al. 2009) that the performance of biomolecular switches can be modulated by changing the equilibrium constant K1. For example, changing the equilibrium constant K1 toward a nonbinding, nonactive “off” conformation ensures a large signal change. However, it also reduces the affinity since the riboswitch now must overcome more significant conformational changes. On the other hand, changing the equilibrium constant toward the active “on” conformation

Fig. 3 Schematic illustration of the ligand-activated transcriptional riboswitch. The aptamer domain is shown in green and expression platform is shown in red. Conformation “off” is a stable confirmation that terminate transcription. This conformation can spontaneously switch into the “on” state which is unstable without the presence of a ligand L. Addition of ligand stabilizes the conformation “on,” leading to a large increase in the level of protein expression

Fig. 4 Schematic illustration of the ligand-activated translational riboswitch. The aptamer domain is shown in green, expression platform is shown in red as in Fig. 1, yellow is a ribosome binding site (RBS) and blue is a gene start codon AUG. In the absence of the ligand, the aptamer extensively binds with the expression platform that includes RBS thus making it inaccessible for ribosome and preventing protein expression. Addition of the ligand stabilizes the “on” conformation, in which the RBS is unpaired and available for ribosome binding leading to a large increase in the level of protein expression

124

S. Harbaugh et al.

increases the background signal without the presence of the ligand. Therefore, the optimal performance of the riboswitch is achieved at the intermediate values of K1. Folding of mRNA molecules and switching between the conformations plays a key role in the functioning of riboswitches. The three-dimensional structure of RNA molecules is dominated by the formation of their secondary structure via the WatsonCrick and GU base pairs. There are several software tools that predict optimal and suboptimal secondary structure of RNA molecules based on their sequences such as Mfold (Zuker 2003), ViennaRNA (Lorenz et al. 2011), NUPACK (Zadeh et al. 2011), and RNAStructure (Reuter and Mathews 2010). However, readers should be aware that different software sometimes can produce different results due to different algorithms used, as well as using different thermodynamic parameters for RNA folding. For example, Mfold and NUPACK use thermodynamic parameters derived in Mathews et al. (1999), while the ViennaRNA package uses revised parameters that take into account experimental results (Mathews et al. 2004). Software tools that predict folding of RNA molecules, together with the knowledge of mechanisms that govern riboswitch functions, make it possible to develop computational approaches for the design of synthetic riboswitches. Wachsmuth et al. (2012) used the ViennaRNA package to generate and select short spacer sequences, which can fold into functional terminators with a sequence complementary to the 30 -part of the theophylline aptamer, to design theophyllinedependent transcriptional ON switches. The stability of the terminator hairpin was considered as the main factor in the selection of designed riboswitches for experimental verification. Three of their designed constructs were functional in vivo and regulated the expression of the reporter, β-galactosidase gene. Further sequence optimization of the best performing riboswitch resulted in a 6.5-fold activation of gene expression in the presence of theophylline (Wachsmuth et al. 2012). Further investigation found that riboswitch functionality is strongly affected not only by the stability of the terminator hairpin, but also by the folding pathway and the existence of potential folding traps (Wachsmuth et al. 2015). The revised in silico strategy was applied to design ligand-dependent riboswitches for tetracycline and streptomycin aptamers (Domin et al. 2017). The resulting tetracycline riboswitches were functional in vivo and showed a fold change up to 3.4. However, none of the four tested candidates for the streptomycin sensing riboswitches showed a clear switching behavior. Further investigations of the streptomycin aptamer by in-line probing showed that the actual probed secondary structure is different from the predicted lowest energy secondary structure (Domin et al. 2017). The conclusion was that further improvement to the design strategy is required to include aptamer structures that are not the minimal free energy structures as predicted by ViennaRNA software. In a different approach, the Batey group developed transcriptional riboswitches in a modular “mix-and-match” fashion (Ceres et al. 2013). From a set of natural riboswitches, they derived expression platforms that can be uncoupled structurally from their aptamer domain. These expression platforms were then combined with a variety of natural and synthetic aptamers. Two sets of modular transcriptional ON

6

Riboswitches as Sensor Entities

125

switches were developed: one based on the expression platform of pbuE riboswitch from B. subtilis and another one based on the expression platform of metH riboswitch from Dechloromonas aromatica. Initially, only one of the aptamers connected to an expression platform showed switching functionality in vitro. A series of alterations in the wild-type expression platforms were required to optimize the performance of designed riboswitches (Ceres et al. 2013). Rational design of translational riboswitches functional in bacteria was presented by Suess et al. (2004). In B. subtilis, a translational control element was created by combining the theophylline aptamer with a helical communication module for which a ligand-dependent one-nucleotide slipping mechanism has been proposed. This structural element was inserted close to the RBS so that its nonbound conformation interfered with ribosome accessibility. Binding of theophylline then induced a structural transition in the helix leading to a one nucleotide shift which moved the inhibitory element away by exactly the critical distance to allow ribosome binding. Synthetic riboswitches acting at translational level were also designed for internal ribosomal entry sites (IRES) in an eukaryotic cell-free translation system (wheat germ extract) (Ogawa 2011). In these riboswitches, the specific aptamer is coupled with short sequences competing for base pairing with essential IRES parts (antiIRES as well as anti-anti-IRES sequences) and a folding-modulating element (MS). Translation of a reporter gene was promoted only in the presence of a specific ligand due to rearrangement and refolding of the IRES and anti-IRES sequences. When the order of individual parts of the riboswitch was changed, the system was reprogrammed into functional OFF switches, where ligand interaction suppressed IRES function (Ogawa 2012). To design synthetic riboswitches that regulate translation initiation in bacteria, Salis and coworkers developed the riboswitch calculator for automated physicsbased design of synthetic riboswitches (Espah Borujeni et al. 2015). Using statistical thermodynamics, it calculates the energy ΔGtotal for mRNA molecules in the “off” and “on” states. This energy can be converted into the translation initiation rate r = exp (βΔGtotal). The riboswitch activation ratio can be then calculated as AR = rON/rOFF = exp [β(ΔGON  ΔGOFF)]. To design new riboswitches, the developed algorithm searches about 1036 sequences to identify sequences that maximize a selected objective function. This approach was applied to design 62 synthetic riboswitches for six different aptamers that were tested in vitro and/or in vivo. The authors demonstrated the versatility of their design method as the performances of 55% of the tested riboswitches were correctly modeled.

Riboswitch Detection Systems A readable detection of an output signal that is either suppressed or produced due to riboswitch-mediated gene regulation is an important factor which not only reflects the synthetic riboswitch functionality but also determines further riboswitch applications.

126

S. Harbaugh et al.

Monitoring Functionality of Synthetic Riboswitches A variety of different reporter systems were applied to monitor the functionality of synthetic riboswitches during and after the selection process. Most frequently used reporter systems are based on expression of green fluorescent protein (GFP) or its variants in prokaryotic and eukaryotic cells (Fowler et al. 2008; Weigand et al. 2008; Lynch and Gallivan 2009; Dixon et al. 2010). The relative fluorescence of produced GFP or variants can be directly measured for quantitative analysis of riboswitch activation. Another very common reporter system is based on bacterial lacZ gene, encoding for β-galactosidase (Lynch et al. 2007). The β-galactosidase enzymatic system allows a qualitative evaluation of riboswitch clones that relies on color-based screening of individual colonies and quantitative analysis of riboswitch functionality by measuring enzymatic activity in cellular extracts using the chromogenic substrate, o-nitrophenyl-β-D-galactopyranoside (ONPG) (Lynch et al. 2007). In some cases, enzymatic reporters coupled with colorimetric or fluorescent detection systems can be more advantageous for monitoring riboswitch activation, since only a small amount of enzyme is required to catalyze an enzyme-specific reaction and to achieve a measurable output signal. For example, Harbaugh et al. (2009) have developed a fluorescence resonance energy transfer (FRET)-based enzymatic assay for monitoring riboswitch activation in E. coli cells. This assay is based on the expression of two genes, tobacco etch virus (TEV) protease and its optical engineered protein substrate, within the same cell. The protein substrate is a FRET construct composed of a donor, an enhanced green fluorescent protein (eGFP), and an acceptor, a nonfluorescent mutant of yellow fluorescent protein called resonance energy-accepting chromoprotein (REACh), connected with a peptide linker containing a TEV protease cleavage site. Use of this type of FRET pair eliminates acceptor fluorescence, and therefore little to no fluorescence is observed prior to cleavage. Using the TEV protease-based reporter system coupled with a theophylline synthetic riboswitch, the authors were able to observe a detectable increase in fluorescence intensity as early as 30 min post analyte addition. Moreover, the superiority of TEV protease–FRET substrate system over direct coupling of the riboswitch with fluorescent protein in terms of sensitivity was demonstrated. When the eGFP gene was placed downstream of theophylline riboswitch, only a very modest increase (~1.4-fold) in fluorescence intensity of cells in response to analyte was observed. In contrast, riboswitch activation of TEV protease gene expression followed by cleavage of FRET protein resulted in an 11.3-fold increase in fluorescence (Harbaugh et al. 2009). Another example of an enzymatic reporter, that was used to monitor synthetic riboswitch activation and could amplify the output of the riboswitch and improve its efficiency, is a T7 polymerase-based system. To boost riboswitch functionality in plastids and to develop a generally applicable tool for high-level transgene expression under riboswitch control, Emadpour et al. (2015) placed the T7 RNA polymerase under the control of theophylline-responsive riboswitch. The transgene of interest (which, for the purpose of establishing the system, was gfp encoding the

6

Riboswitches as Sensor Entities

127

green fluorescent protein) was driven by the T7 promoter. Thus, treatment of plants with theophylline resulted in induction of T7 RNA polymerase expression at relatively low levels, which however was sufficient to trigger strong transcriptional activation of the gfp transgene (Emadpour et al. 2015). Some reporter systems are highly efficient for riboswitch selection but cannot be directly used to quantitatively monitor riboswitch performance. Using reporter systems based on genes conferring resistance against antibiotics allows a growthdependent analysis of riboswitch-mediated regulation. As was mentioned above, the possibility to use the tetracycline-resistance mediating tetA gene for positive as well as negative selection renders this reporter a very versatile selection system (Nomura and Yokobayashi 2007). However, the lack of a quantitative readout of the expression level of tetA necessitated a design of the TetA-GFPuv translational fusion to enable quantitative fluorescence-based screening of the selected riboswitch clones (Muranaka et al. 2009a, b)). A different reporter system, based on cheZ gene that controls E. coli motility, also can be used to perform both positive and negative selections resulting in riboswitches with down- and upregulating functions. However, similar to tetA this system lacks a quantitative readout and was fused with lacZ gene to estimate the efficiency of selected riboswitches (Topp and Gallivan 2008a).

Coupling Synthetic Riboswitches with New Reporter Systems Although the reporter systems utilized for riboswitch selection can be used as readout assays for riboswitch-based biosensing, in some cases coupling previously selected synthetic riboswitches with new reporters are highly desirable. Moreover, to be truly useful as sensor entities, riboswitches should be compatible with a variety of different reporters, i.e., they should be modular “plug and play” platforms. However, this is not always the case. For many riboswitches, selective binding of a small molecule of interest and conformational change in RNA secondary structure is influenced not only by its own sequence, but also by the surrounding genetic context including the proximal open reading frame (ORF) under the control of the riboswitch. Thus, substituting the original ORF with a new one on a “start codon for start codon” basis can affect the ability of the riboswitch to regulate gene expression in response to a given ligand (Caron et al. 2012; Folliard et al. 2017). To overcome this lack of modularity, many studies have created fusions comprised of a desired riboswitch, the first few hundred base pairs of its working ORF, and a new reporter gene (Dixon et al. 2012). A short N-terminal peptide fusion can also be coupled with a reporter gene utilized for riboswitch selection (Sinha et al. 2011). In this case, the incorporation of a short peptide fusion ensures that the sequence immediately 30 to the expression platform is constant throughout the selection and the riboswitch functionality will be preserved even if the original reporter gene will be replaced with a new one. However, these approaches can fail in some circumstances since the inclusion of 50 fusions can alter the new reporter gene’s functionality (Folliard et al. 2017).

128

S. Harbaugh et al.

To solve the problem associated with 50 fusions, Folliard et al. (2017) have designed and introduced a novel genetic element called the ribo-attenuator (Fig. 5). The ribo-attenuator is placed after 150 base pairs of a riboswitch’s original ORF. It consists of a hairpin containing a ribosome binding site (RBS) on the downstream portion of a stem in order to silence translation independent of the upstream riboswitch activity. This is followed by a negative one base pair shifted transcriptionally fused stop and start codon (TAATG). The passage of ribosomes recruited by the riboswitch opens up the introduced hairpin, before the ribosome dissociates at the proximal in-frame stop codon (TAA) in the junction between the original ORF and the ORF containing the gene of interest. Additional ribosomes can then assemble at the ribo-attenuator RBS and initiate translation at the first start codon of the introduced gene of interest (Fig. 5). Therefore, instead of directly controlling the translation of a gene, the riboswitch controls the translational initiation from the downstream attenuator RBS. Thus, the developed ribo-attenuator system enabled riboswitch controlled expression of a new gene of interest without the inclusion of a 50 fusion (Folliard et al. 2017). Another way to preserve the riboswitch functionality and expand the applicability of synthetic riboswitches is to use recombinase-based reporter system. Recently, Harbaugh et al. have developed a dual-color detection system comprising of the E. coli FimE recombinase controlled by a synthetic riboswitch and an invertible DNA

Fig. 5 Riboswitch activation and ribo-attenuator context. A ribosome binding site (RBS, yellow) is sequestered within a riboswitch preventing ribosome recruitment. A ribo-attenuator adds a second RBS, sequestered away by a local hairpin. Binding of a ligand (L) causes a conformational change exposing the riboswitch RBS. The hairpin can be opened by a ribosome travelling from the riboswitch RBS, exposing the attenuator RBS and allowing translation of the gene of interest (Folliard et al. 2017)

6

Riboswitches as Sensor Entities

129

segment containing a constitutively active promoter placed between two fluorescent protein genes (Harbaugh et al. 2017). Utilizing the recombinase-based system as a means of riboswitch selection negates the requirement of an insertion of a 50 fusion when a different reporter is necessary because a new reporter can be placed on either side of invertible promoter, keeping the riboswitch-FimE construct intact. The T7 polymerase reporter system mentioned above also can be advantageous for riboswitch selection and further riboswitch applications since it can initiate expression of any reporter gene.

Riboswitch Optimization toward Sensing Requirements Selectivity, but at a Cost As it was discussed, riboswitches offer the capability to design and develop specific sensors to potentially any ligand, but often require sensitive, laboratory-based equipment to detect their reporter output. Once a specific riboswitch has been optimized, it is often constrained to the context it was optimized for. For example, altering riboswitch output by switching out the RBS sequence with one that would increase reporter gene expression is likely to disrupt the function of the riboswitch, through changes in its secondary and tertiary structure. Similarly, if a riboswitch is used to control expression of a reporter gene that it was not designed and optimized for, the conformation of the riboswitch may be altered, rendering it unswitchable. Further, synthetic riboswitches are often optimized within a cellular chassis that has been developed for enhanced transcript or protein expression (Lynch and Gallivan 2009; Davidson et al. 2013). These cell types do not generally survive well outside of controlled laboratory conditions. Overcoming these hurdles would enhance the utility of riboswitches in fieldable sensors.

Biological Circuits Recent advances in synthetic biology have allowed different genetic “components” to be connected into biological circuits via signaling molecules. The signaling molecules that connect these components are often those involved in bacterial quorum sensing (Brenner et al. 2007; Tabor et al. 2009; Tamsir et al. 2011). These molecules freely cross cell membranes and, when their concentration surpasses a threshold level, they activate transcription of genes that have the cognate quorum sensing promoter. In nature, this process allows appropriate genes to be activated only in a high cell density environment (Pesci et al. 1997). In engineered biological circuits, using quorum molecules that are orthologous to the cellular chassis, this system allows the output from an upstream component to be biologically “wired” to the input of a downstream component, generating genetic logic gates and forming genetic programs within and between cells (Brenner et al. 2007; Tabor et al. 2009; Tamsir et al. 2011; Friedland et al. 2009; Lou et al. 2010; Moon et al. 2012).

130

S. Harbaugh et al.

Similarly, this method has also been used to produce a signal amplification circuit that increases the dynamic range of responses from promoters to a specific input (Karig and Weiss 2005).

Riboswitch Circuitry Wiring riboswitches into biological circuits using quorum signaling molecules overcomes many of the hurdles discussed earlier in section “Selectivity, but at a Cost.” By separating the riboswitch-induced quorum signaling molecule production and the quorum sensing promoter into different cell types, Goodson et al. (2017) demonstrated tunable riboswitch signal amplification even in a chassis that was not optimized for enhanced protein expression. The riboswitch specifically controlled quorum sensing molecule production in one cell type. Meanwhile, the reporter cell type contained a quorum sensing promoter upstream of a RBS and reporter gene. This architecture resulted from the design-build-test paradigm of synthetic biology, since initial designs of amplification circuits as positive feedback loops on a single plasmid resulted in nonspecific activation because of localized buildup of the quorum signal (Goodson et al. 2015). This was also evident when the riboswitch and reporter were separated into two plasmids but were located in same cell. Separating the riboswitch and reporter into different cell types insulated the function of each plasmid (Tamsir et al. 2011), thus reducing the likelihood of false positives. This arrangement also negated the sometimes “leaky” nature of riboswitches since degradation tags could be used to reduce the production of quorum sensing signal producing proteins, even when the riboswitch is “off,” thus keeping the signaling molecule concentration below threshold levels until riboswitch activation. Concomitantly, this had the effect of converting the analog nature of the riboswitch to a digital output, dependent upon the activation threshold concentration of the quorum sensing signal being used.

Corollary Advantages of Signal Amplification Using the amplification circuit increases riboswitch sensitivity and response time compared to designs where the riboswitch is upstream of the reporter gene (Goodson et al. 2017), since only a small increase in signaling molecule production can activate the quorum sensing promoter upstream of the reporter gene, the strength of which can be modified by optimizing RBS strength and rate of reporter degradation. This increases the utility of riboswitches that have low activation ratios (i.e., a small difference in gene expression between the inactivated and activated states), and, similarly, it can aid in riboswitch screening and selection by ensuring even weak riboswitches are identified. Indeed, identifying riboswitches in this way facilitates their inclusion into existing sensing systems that contain reporter cell types that respond to the signaling molecule produced by the riboswitch, and separating the reporter from the riboswitch

6

Riboswitches as Sensor Entities

131

enables reporter genes to be exchanged without the associated context-related disruption of riboswitch function. Thus separating the riboswitch and the reporter, but biologically wiring them together, increases the modularity of the system and streamlines the inclusion of riboswitches into fieldable sensors for practical applications.

Tandem Riboswitches Another promising approach for improving riboswitch response to its analyte is to combine several repeats of the riboswitch construct 50 of the gene to be regulated. Tandem riboswitches arranged in different configurations have been found in nature (Breaker 2012; Roßmanith and Narberhaus 2017). For example, a riboswitch for the amino acid glycine from Vibrio cholerae carries two aptamers and only a single expression platform (Mandal et al. 2004). These aptamers function cooperatively, such that glycine binding by one aptamer increases the affinity for glycine binding to the adjacent aptamer. Other natural tandem architectures consist of two complete riboswitch repeats, as observed, for instance for some TPP riboswitches, where each element acts independently of the other (Rodionov et al. 2004; Welz and Breaker 2007). Similar to their natural analogues, synthetic riboswitches can be toggled in series that, in some cases, helps to reduce background activity and increase the dynamic range. Using a tetracycline binding aptamer to regulate translation initiation in yeast, Kötter et al. (2009) demonstrated that addition of a second or third copy of the aptamer presented in the 50 -UTR of a reporter gene strongly increases the dynamic range of regulation. Muranaka and Yokobayashi (2010) have developed a system with two individual riboswitches by inserting previously engineered TPP riboswitches into the 50 -UTR of a GFP-encoding mRNA in E. coli cells. In this tandem, the first riboswitch represented a transcriptional OFF switch, where TPP binding terminates mRNA transcription. The second riboswitch, placed downstream of the first element, was a translational ON switch, where ligand binding initiates mRNA translation. In the designed system, transcription termination could be achieved only at high thiamine concentrations resulting in a repression of a reporter gene expression. Low thiamine concentrations were not sufficient to activate the transcriptional OFF switch and terminate mRNA transcription, however the translational ON switch couldn’t be activated blocking mRNA translation. Only an intermediate amount of thiamine resulted in GFP production, as both riboswitches were not completely repressed at such concentrations. Hence, this tandem riboswitches functioned as a chemical band-pass filter circuit allowing gene expression only in certain concentration range of the ligand, where the OFF switch acts as a low-pass filter and the ON switch as a high-pass filter (Muranaka and Yokobayashi 2010). The Mörl group (Wachsmuth et al. 2015) inserted two or three copies of theophylline-responsive identical transcriptional ON switches into the 50 -UTR of a βgalactosidase reporter gene. While a single-copy riboswitch had activation ratios

132

S. Harbaugh et al.

of 3–6.5-fold, the serial repeats showed increased response ratios from 17-fold (tandem) to 23-fold (tridem). The authors suggested that a possible reason for this improved ligand-dependent response is the presence of several terminator elements within one transcriptional unit. The enhanced transcription termination leads to a very low background activity in the riboswitch-mediated OFF state, resulting in a dramatically increased difference in reporter gene expression without and in the presence of the ligand. The described tandem and tridem riboswitch arrangements, consisting of aptamers that recognize the same ligand, allow a better activation ratio in gene regulation. The combination of riboswitches interacting with different target molecules, however, leads to construction of Boolean logics allowing possible detection of multiple analytes. Using the previously described tetA based dual genetic selection method, Yokobayashi and coworkers combined the theophylline aptamer with reengineered or natural TPP riboswitches (Sharma et al. 2008). Two libraries were generated by incorporation of 20 randomized nucleotides between the theophylline aptamer and TPP riboswitch. Combining the theophylline aptamer and the reengineered TPP-responsive translational ON switch resulted in selection of several AND gates, activating gene expression in the presence of both theophylline and TPP. On the other hand, coupling the theophylline riboswitch with the natural TPP sensing translational OFF switch allowed identification of several NAND gates, repressing gene expression only when both theophylline and TPP are present. Thus, the riboswitch regulatory mechanism and the sequence of the linker connecting two sensing domains can determine the type of created logic gates. Other logic gates switches were designed on the basis of transcriptional riboswitches. Mörl and coworkers constructed AND gates consisting of theophylline and tetracycline riboswitches. When the theophylline switch was placed upstream of the tetracycline switch only the addition of both ligands induced gene expression by 10.4 fold as expected for a logic AND gate (Domin et al. 2017). The described examples of tandem constructs clearly show that it is possible to generate Boolean logic gates from synthetic translational and transcriptional riboswitch elements.

Applications of Synthetic Riboswitches As was described above, a number of small molecule analytes sensing riboswitches were developed in the past couple of decades. Like their natural counterparts, synthetic riboswitches have the ability to regulate gene expression in response to levels of a specific small molecule in a concentration-dependent manner and, as a result, can be valuable for a range of applications. To date, synthetic riboswitches responsive to nonnatural small molecules have been used for regulation of gene expression, to control cellular behavior, and to optimize small molecule production in metabolic engineering. Riboswitch-based regulation of protein production has been demonstrated in a wide range of gram-positive and gram-negative bacterial species. Topp et al.

6

Riboswitches as Sensor Entities

133

(2010) have developed a set of five theophylline synthetic riboswitches that enable inducible gene expression in eight diverse bacterial species, including organisms that currently have few or no tools, which have been used to titrate gene expression in the laboratory. Developed riboswitches showed 25-fold increase in gene expression in all species tested and a greater-than-50-fold increase in two human pathogens, Acinetobacter baumannii and Streptococcus pyogenes. These riboswitches can be useful tools to enable studies on the mechanisms of A. baumannii and S. pyogenes pathogenesis that have been hindered by the inability to induce gene expression in conditional knockouts in these organisms (Topp et al. 2010). Reynoso et al. (2012) have applied theophylline-sensitive synthetic riboswitches to induce protein expression in the intracellular pathogen Francisella tularensis. It was demonstrated that riboswitches can be used to functionally control a bacterial gene that is critical to the ability of a pathogen to cause disease during intracellular infection. Since this system can be adapted to diverse bacteria, riboswitches will likely facilitate the in-depth study of the virulence mechanisms of numerous difficult-to-study intracellular pathogens, as well as future emerging pathogens (Reynoso et al. 2012). Seeliger et al. (2012) have applied a synthetic riboswitch-based system for translational control of gene expression in the human pathogen Mycobacterium tuberculosis. Although the optimized system resulted in only an 8.2-fold activation, it robustly regulated protein production in a macrophage infection model (Seeliger et al. 2012). Similar systems of theophylline-responsive riboswitches have been demonstrated in streptomyces and cyanobacterial species (Nakahira et al. 2013; Rudolph et al. 2013; Ma et al. 2014; Ohbayashi et al. 2016). In other studies, theophylline riboswitch-mediated regulation has been applied to control genes in viral replication and in plant plastids (Wang and White 2007; Verhounig et al. 2010). Coupling a theophylline-binding aptamer with a viral regulatory RNA element (RE) resulted in a theophylline-dependent induction of viral replication. Analysis of this engineered viral genome revealed that this RE, located in the 50 -UTR, specifically mediates efficient accumulation of plus-strands of the virus genome. Therefore, in addition to allowing for modulation of virus reproduction, the RE riboswitch system also provided insight into RE function (Wang and White 2007). Verhounig et al. (2010) have identified a theophyllineresponsive synthetic riboswitch that functions as an efficient translational regulator of gene expression in plastids. This riboswitch provides a novel tool for plastid genome engineering that facilitates the tightly regulated inducible expression of chloroplast genes and transgenes and provides opportunities for plastid biotechnology (Verhounig et al. 2010). Another application of riboswitch-mediated regulation is to generate conditional knockouts of essential genes for basic biological studies. Using a tetracyclinecontrolled expression system, a conditional knockdown system for five essential genes in S. cerevisiae was created by varying the promoter sequence to change the expression strength (Kötter et al. 2009). Additionally, riboswitches can enable the selective expression of dominant negative mutations, as shown with theophyllineresponsive csrA in E. coli to demonstrate its role in autoaggregation and cell cycle

134

S. Harbaugh et al.

control (Jin et al. 2009). Both studies suggest that synthetic riboswitches may find broad use in investigating genetics of microorganisms. The ability of riboswitches to control gene expression has been applied to reprogram bacterial behaviors such as cell motility. Gallivan and coworkers have reprogrammed E. coli chemotaxis system by placing a key chemotaxis signaling protein (CheZ) under the control of theophylline-sensitive riboswitch (Topp and Gallivan 2007; Mishler et al. 2010). Reprogrammed cells could migrate up gradients if this analyte and autonomously localize to regions of high theophylline concentration, which is a behavior that cannot be accomplished by the natural E. coli chemotaxis system. In a different study, Micklefield and coworkers coupled cheZ gene with a synthetic riboswitch responsive to pyrimido[4,5-d]pyrimidine-2,4diamine (PPDA) (Robinson et al. 2014). E. coli cells harboring a riboswitch-cheZ construct exhibited dose-dependent cell migration in response to PPDA. The ability to modulate bacterial motility in response to arbitrary chemical signals can provide new tools for bioremediation and drug delivery. Another riboswitch application is a control of cell morphology. PPDA-sensitive transcriptional OFF switch was used to chemically control a production of an actin homologue, MreB protein, in B. subtilis (Robinson et al. 2014). MreB has a critical role in morphogenesis in rod-shaped bacterial species and is an important new target for the development of antimicrobial agents. Dose-dependent regulation of antimicrobial targets, such as mreB, can find use in sensitive and specific antimicrobial screening systems or mechanism-of-action studies. Riboswitch reporters can also be used to monitor and optimize in vivo production of small molecules. For instance, Xiu et al. (2017) have developed RNA riboswitchbased biosensor for identification of naringenin over producing E. coli strains. Naringenin is a key flavonoid precursor, and its production is in a particular interest since it is being researched as a potential treatment for Alzheimer’s disease. Riboswitch-mediated regulation of gene expression in response to explosive compounds can be applied for cell-based environmental biosensing. KelleyLoughnane and coworkers have demonstrated that 2,4-dinitrotoluene (DNT)- and 2,4,6-trinitrotoluene (TNT)-sensitive synthetic riboswitches activate expression of downstream reporter genes at concentrations of DNT and TNT that are in a range of previously reported environmental concentrations of these compounds found in contaminated water and soil (Davidson et al. 2013; Harbaugh et al. 2017). Although application of riboswitches as sensor entities is still limited due to a low number of selected functional switches responsive to small molecule analytes, the development of new approaches for fast and modular riboswitch design and discovery will allow to expand their utilization in a nearest future.

Conclusions Synthetic riboswitches can be engineered to regulate expression of any gene in response to any nonnatural molecule that is capable of being bound by RNA and is nontoxic to cells. This feature demonstrates a strong possibility for RNA switches

6

Riboswitches as Sensor Entities

135

to serve as sensor entities for design and development of cell-based biosensors with a variety of different applications. Riboswitches have passed the proof-of-concept state and, due to the current mechanistic understanding of their function, now are being developed to address specific applications. The advantage of using riboswitches is that they do not require protein cofactors and give access to directly alter protein expression at an early stage, which saves resources and is potentially faster than regulation by proteins such as transcription factors, which need to be produced on demand. They consist solely of RNA and therefore they are easy to implement, because they involve the transfer of only a single genetic control element into an organism. Finally, they control gene expression in dose dependent manner resulting in the ability not only to sense a desired target but also to determine its concentration. Synthetic tandem riboswitches demonstrate that multiunit riboswitches can achieve complex functions similar to those of the more elaborate circuits consisting of multiple genes and regulatory proteins. Despite their useful features, synthetic riboswitches still remain outnumbered by other genetic regulatory elements (such as promoters and transcriptional factors) in the field of biosensors, mainly due to the low number of available aptamers that can be converted into functional switches. Another reason for limited riboswitch applications is their incompatibility with a variety of different reporters and low sensitivity. Advances in the development of aptamer and riboswitch selection and screening techniques, improvements in rational structure-based designs and computational modeling, and progress in development of sophisticated approaches for riboswitch tuning and performance optimization will guide the design of increased number of better riboswitches to harness the full regulatory and sensing potential of these RNA switches in sensing devices.

References Antunes D, Jorge NAN, Caffarena ER, Passetti F (2018) Using RNA sequence and structure for the prediction of riboswitch aptamer: a comprehensive review of available software and tools. Front Genet 8:231 Barrick JE, Breaker RR (2007) The distributions, mechanisms, and structures of metabolite-binding riboswitches. Genome Biol 8:R239 Beisel CL, Smolke CD (2009) Design principles for riboswitch function. PLoS Comput Biol 5(4): e1000363 Berens C, Suess B (2015) Riboswitch engineering – making the all-important second and third steps. Curr Opin Biotechnol 31:10–15 Berens C, Thain A, Schroeder R (2001) A tetracycline-binding RNA aptamer. Bioorg Med Chem 9(10):2549–2556 Blouin S, Mulhbacher J, Penedo JC, Lafontaine DA (2009) Riboswitches: ancient and promising genetic regulators. Chembiochem 10:400–416 Breaker RR (2012) Riboswitches and the RNAworld. Cold Spring Harbor Perspect Biol 4:a003566 Brenner K, Karig DK, Weiss R, Arnold FH (2007) Engineered bidirectional communication mediates a consensus in a microbial biofilm consortium. Proc Natl Acad Sci U S A 104:17300–17304 Caron M-P, Bastet L, Lussier A et al (2012) Dual-acting riboswitch control of translation initiation and mRNA decay. Proc Natl Acad Sci U S A 109:3444–3453

136

S. Harbaugh et al.

Carothers JM, Goler JA, Kapoor Y et al (2010) Selecting RNA aptamers for synthetic biology: investigating magnesium dependence and predicting binding affinity. Nucleic Acids Res 38:2736–2747 Ceres P, Trausch JJ, Batey RT (2013) Engineering modular ‘ON’ RNA switches using biological components. Nucleic Acids Res 41:10449–10461 Cheah MT, Wachter A, Sudarsan N, Breaker RR (2007) Control of alternative RNA splicing and gene expression by eukaryotic riboswitches. Nature 447:497–500 Coppins RL, Hall KB, Groisman EA (2007) The intricate world of riboswitches. Curr Opin Chem Biol 10:176–181 Davidson ME, Harbaugh SV, Chushak YG et al (2013) Development of 2,4-dinitrotoluene-responsive synthetic riboswitch in E. coli cells. ACS Chem Biol 8:234–241 Desai SK, Gallivan JP (2004) Genetic screens and selections for small molecules based on a synthetic riboswitch that activates protein translation. J Am Chem Soc 126:13247–13254 Dixon N, Duncan JN, Geerlings T et al (2010) Reengineering orthogonally selective riboswitches. Proc Natl Acad Sci U S A 107:2830–2835 Dixon N, Robinson CJ, Geerling T et al (2012) Orthogonal riboswitches for tuneable coexpression in bacteria. Angew Chem Int Ed 51:3620–3624 Domin G, Findeiß S, Wachsmuth M et al (2017) Applicability of a computational design approach for synthetic riboswitches. Nucleic Acids Res 45:4108–4119 Edwards AL, Batey RT (2010) Riboswitches: a common RNA regulatory element. Nat Educ 3:9 Ehrentreich-Förster E, Orgel D, Krause-Griep A et al (2008) Biosensor-based on-site explosives detection using aptamers as recognition elements. Anal Bioanal Chem 391:1793–1800 Ellington AD, Szostak JW (1990) In vitro selection of RNA molecules that bind specific ligands. Nature 346:818–822 Emadpour M, Karcher D, Bock R (2015) Boosting riboswitch efficiency by RNA amplification. Nucleic Acids Res 43:e66 Espah Borujeni A, Mishler DM, Wang J et al (2015) Automated physics-based design of synthetic riboswitches from diverse RNA aptamers. Nucleic Acids Res 44:1–13 Etzel M, Mörl M (2017) Synthetic riboswitches: from plug and pray toward plug and play. Biochemistry 56:1181–1198 Filonov GS, Moon JD, Svensen N, Jaffrey SR (2014) Broccoli: rapid selection of an RNA mimic of green fluorescent protein by fluorescence-based selection and directed evolution. J Am Chem Soc 136:16299–16308 Findeiß S, Etzel M, Will S, Mörl M, Stadler PE (2017) Design of artificial riboswitches as biosensors. Sensors 17:1990 Folliard T, Mertins B, Steel H et al (2017) Ribo-attenuators: novel elements for reliable and modular riboswitch engineering. Sci Rep 7:4599–4610 Fowler CC, Brown ED, Li Y (2008) A FACS-based approach to engineering artificial riboswitches. Chembiochem 9:1906–1911 Friedland AE, Lu TK, Wang X et al (2009) Synthetic gene networks that count. Science 324:1199–1202 Ghazi Z, Fowler CC, Li Y (2014) Artificial riboswitch selection: a FACS-based approach. Methods Mol Biol 1111:57–75 Gong S, Wang Y, Wang Z, Zhang W (2017a) Computational methods for modeling aptamers and designing riboswitches. Int J Mol Sci 18:2442 Gong S, Wang Y, Wang Z, Zhang W (2017b) Co-transcriptional folding and regulation mechanisms of riboswitches. Molecules 22:1169–1183 Goodson MS, Harbaugh SV, Chushak YG, Kelley-Loughnane N (2015) Integrating and amplifying signal from riboswitch biosensors. Methods Enzymol 550:73–91 Goodson MS, Bennett AC, Jennewine BR et al (2017) Amplifying riboswitch signal output using cellular wiring. ACS Synth Biol 18:1440–1444 Groher F, Suess B (2014) Synthetic riboswitches – a tool comes of age. Biochim Biophys Acta 1839:964–973

6

Riboswitches as Sensor Entities

137

Groher F, Bofill-Bosch C, Schneider C et al (2018) Riboswitching with ciprofloxacin – development and characterization of a novel RNA regulator. Nucleic Acids Res 46:2121–2132 Hallberg ZF, Su Y, Kitto RZ, Hammond MC (2017) Engineering and in vivo applications of riboswitches. Annu Rev Biochem 86:515–539 Han SR, Yu J, Lee S-W (2014) In vitro selection of RNA aptamers that selectively bind danofloxacin. Biochem Biophys Res Commun 448:397–402 Hanson S, Berthelot K, Fink B et al (2003) Tetracycline-aptamer-mediated translational regulation in yeast. Mol Microbiol 49:1627–1637 Harbaugh S, Kelley-Loughnane N, Davidson M et al (2009) FRET-based optical assay for monitoring riboswitch activation. Biomacromolecules 10:1055–1060 Harbaugh SV, Goodson MS, Dillon K et al (2017) Riboswitch-based reversible dual color sensor. ACS Synth Biol 6:766–781 Harvey I, Garneau P, Pelletier J (2002) Inhibition of translation by RNA-small molecule interactions. RNA 8:452–463 Hybarger Z, Bynum J, Robert F. Williams, James J. Valdes, James P. Chambers (2006) A microfluidic SELEX prototype. Analytical and Bioanalytical Chemistry 384 (1):191–198 Jenison RD, Gill SC, Pardi A, Polisky B (1994) High-resolution molecular discrimination by RNA. Science 263:1425–1429 Jin Y, Watt RM, Danchin A, Huang JD (2009) Use of a riboswitch-controlled conditional hypomorphic mutation to uncover a role for the essential csrA gene in bacterial autoaggregation. J Biol Chem 284:28738–28745 Karig D, Weiss R (2005) Signal-amplifying genetic circuit enables in vivo observation of weak promoter activation in the Rhl quorum sensing system. Biotechnol Bioeng 89:709–718 Kim D-S, Gusti V, Pillai SG, Gaur RK (2005) An artificial riboswitch for controlling pre-mRNA splicing. RNA 11:1667–1677 Kötter P, Weigand JE, Meyer B et al (2009) A fast and efficient translational control system for conditional expression of yeast genes. Nucleic Acids Res 37:e120 Lemay J-F, Desnoyers G, Blouin S et al (2011) Comparative study between transcriptionally- and translationally-acting adenine riboswitches reveals key differences in riboswitch regulatory mechanisms. PLoS Genet 7:e1001278 Lorenz R, Bernhart SH, Zu Siederdissen CH et al (2011) ViennaRNA Package 2.0. Algorithms Mol Biol 6:26 Lou X, Qian J, Xiao Y et al (2009) Micromagnetic selection of aptamers in microfluidic channels. Proc Natl Acad Sci U S A 106:2989–2994 Lou C, Liu X, Ni M et al (2010) Synthesizing a novel genetic sequential logic circuit: a push-on push-off switch. Mol Syst Biol 6:350 Lynch SA, Gallivan JP (2009) A flow cytometry-based screen for synthetic riboswitches. Nucleic Acids Res 37:184–192 Lynch SA, Desai S, Sajja HK, Gallivan JP (2007) A high-throughput screen for synthetic riboswitches reveals mechanistic insights into their function. Chem Biol 14:173–184 Ma AT, Schmidt CM, Golden JW (2014) Regulation of gene expression in diverse cyanobacterial species by using theophylline-responsive riboswitches. Appl Environ Microbiol 80:6704–6713 Machtel P, Bąkowska-Żywicka K, Żywicki M (2016) Emerging applications of riboswitches – from antibacterial targets to molecular tools. J Appl Genet 57:531–541 Mandal M, Breaker RR (2004) Gene regulation by riboswitches. Nat Rev Mol Cell Biol 5:451–463 Mandal M, Lee M, Barrick JE et al (2004) A glycine-dependent riboswitch that uses cooperative binding to control gene expression. Science 306:275–279 Martini L, Meyer AJ, Ellefson JW et al (2015) In vitro selection for small-molecule-trigged strand displacement and riboswitch activity. ACS Synth Biol 4:1144–1150 Mathews DH, Sabina J, Zuker M, Turner DH (1999) Expanded sequence dependence of thermodynamic parameters improves prediction of RNA secondary structure1. J Mol Biol 288:911–940

138

S. Harbaugh et al.

Mathews DH, Disney MD, Childs JL et al (2004) Incorporating chemical modification constraints into a dynamic programming algorithm for prediction of RNA secondary structure. Proc Natl Acad Sci U S A 101:7287–7292 Mishler DM, Topp S, Reynoso GM, Gallivan JP (2010) Engineering bacteria to recognize and follow small molecules. Curr Opin Biotechnol 21:653–656 Moon TS, Lou C, Tamsir A et al (2012) Genetic programs constructed from layered logic gates in single cells. Nature 491:249–253 Morse DP (2007) Direct selection of RNA beacon aptamers. Biochem Biophys Res Commun 359:94–101 Muranaka N, Yokobayashi Y (2010) A synthetic riboswitch with chemical band-pass response. Chem Commun 46:6825–6827 Muranaka N, Abe K, Yokobayashi Y (2009a) Mechanism-guided library design and dual genetic selection of synthetic OFF riboswitches. Chembiochem 10:2375–2384 Muranaka N, Sharma V, Nomura Y, Yokobayashi Y (2009b) An efficient platform for genetic selection and screening of gene switches in Escherichia coli. Nucleic Acids Res 37:e39 Nakahira Y, Ogawa A, Asano H et al (2013) Theophylline-dependent riboswitch as a novel genetic tool for strict regulation of protein expression in cyanobacterium Synecbococcus elongatus PCC 7942. Plant Cell Physiol 54:1724–1735 Nomura Y, Yokobayashi Y (2007) Reengineering a natural riboswitch by dual genetic selection. J Am Chem Soc 129:13814–13815 Nudler E (2006) Flipping riboswitches. Cell 126:19–22 Nudler E, Mironov AS (2004) The riboswitch control of bacterial metabolism. Trends Biochem Sci 29:11–17 Ogawa A (2011) Rational design of artificial riboswitches based on ligand-dependent modulation of internal ribosome entry in wheat germ extract and their applications as label-free biosensors. RNA 17:478–488 Ogawa A (2012) Rational construction of eukaryotic OFF-riboswitches that downregulate internal ribosome entry site-mediated translation in response to their ligands. Bioorg Med Chem Lett 22:1639–1642 Ohbayashi R, Akai H, Yoshikawa H et al (2016) A tightly inducible riboswitch system in Synecbocystis sp. PCC 6803. J Gen Appl Microbiol 159:154–159 Pesci EC, Pearson JP, Seed PC, Iglewski BH (1997) Regulation of las and rhl quorum sensing in Pseudomonas aeruginosa. J Bacteriol 179:3127–3132 Peselis A, Serganov A (2014) Themes and variations in riboswitch structure and function. Biochim Biophys Acta 1839:908–918 Pham HL, Wong A, Chua N et al (2017) Engineering a riboswitch-based genetic platform for the self-directed evolution of acid-tolerant phenotypes. Nat Commun 8:411 Reinholt SJ, Ozer A, Lis JT, Craighead HG (2016) Highly multiplexed RNA aptamer selection using a microplate-based microcolumn device. Sci Rep 6:29771 Reuter JS, Mathews DH (2010) RNAstructure: software for RNA secondary structure prediction and analysis. BMC Bioinformatics 11:129 Reynoso CMK, Miller MA, Bina JE et al (2012) Riboswitches for intracellular study of genes involved in Francisella pathogenesis. mBio 3:e00253–e00212 Robinson CJ, Vincent HA, Wu M-C et al (2014) Modular riboswitch toolsets for synthetic genetic control in diverse bacterial species. J Am Chem Soc 136:10615–10624 Rodionov DA, Dubchak I, Arkin A et al (2004) Reconstruction of regulatory and metabolic pathways in metal-reducing delta-proteobacteria. Genome Biol 5:R90 Roßmanith J, Narberhaus F (2017) Modular arrangement of regulatory RNA elements. RNA Biol 14:287–292 Roth A, Breaker RR (2009) The structural and functional diversity of metabolite-binding riboswitches. Annu Rev Biochem 78:305–334 Rudolph MM, Vockenhuber MP, Suess B (2013) Synthetic riboswitches for the conditional control of gene expression in Streptomyces coelicolor. Microbiology 159:1416–1422 Seeliger JC, Topp S, Sogi KM et al (2012) A riboswitch-based inducible gene expression system for mycobacteria. PLoS One 7:e29266

6

Riboswitches as Sensor Entities

139

Serganov A, Nudler E (2013) A decade of riboswitches. Cell 152:17–24 Sharma V, Nomura Y, Yokobayashi Y (2008) Engineering complex riboswitch regulation by dual genetic selection. J Am Chem Soc 130:16310–16315 Sinha J, Topp S, Gallivan JP (2011) From SELEX to cell: dual selections for synthetic riboswitches. Methods Enzymol 497:207–220 Suess B, Hanson S, Berens C et al (2003) Conditional gene expression by controlling translation with tetracycline-binding aptamers. Nucleic Acids Res 31:1853–1858 Suess B, Fink B, Berens C et al (2004) A theophylline responsive riboswitch based on helix slipping controls gene expression in vivo. Nucleic Acids Res 32:1610–1614 Szeto K, Craighead HG (2014) Devices and approaches for generating specific high-affinity nucleic acid aptamers. Appl Phys Rev 11:2727–2732 Szeto K, Latulippe DR, Ozer A et al (2013) RAPID-SELEX for RNA aptamers. PLoS One 8:e82667 Tabor JJ, Salis HM, Simpson ZB et al (2009) A synthetic genetic edge detection program. Cell 137:1272–1281 Tamsir A, Tabor JJ, Voigt CA (2011) Robust multicellular computing using genetically encoded NOR gates and chemical ‘wires’. Nature 469:212–215 Topp S, Gallivan JP (2007) Guiding bacteria with small molecules and RNA. J Am Chem Soc 129:6807–6811 Topp S, Gallivan JP (2008a) Random walks to synthetic riboswitches – a high-throughput selection based on cell motility. Chembiochem 9:210–213 Topp S, Gallivan JP (2008b) Riboswitches in unexpected places – a synthetic riboswitch in a protein coding region. RNA 14:2498–2503 Topp S, Gallivan JP (2010) Emerging applications of riboswitches in chemical biology. ACS Chem Biol 5:139–148 Topp S, Reynoso CMK, Seeliger JC et al (2010) Synthetic riboswitches that induce gene expression in diverse bacterial species. Appl Environ Microbiol 76:7881–7884 Tucker BJ, Breaker RR (2005) Riboswitches as versatile gene control elements. Curr Opin Struct Biol 15:342–348 Tuerk C, Gold L (1990) Systematic evolution of ligands by exponential enrichment: RNA ligands to bacteriophage T4 DNA polymerase. Science 249:505–510 Vallée-Bélisle A, Ricci F, Plaxco KW (2009) Thermodynamic basis for the optimization of bindinginduced biomolecular switches and structure-switching biosensors. Proc Natl Acad Sci U S A 106:13802–13807 Verhounig A, Karcher D, Bock R (2010) Inducible gene expression from the plastid genome by a synthetic riboswitch. Proc Natl Acad Sci U S A 107:6204–6209 Wachsmuth M, Findeiß S, Weissheimer N et al (2012) De novo design of a synthetic riboswitch that regulates transcription termination. Nucleic Acids Res 41(4):2541–2551 Wachsmuth M, Domin G, Lorenz R et al (2015) Design criteria for synthetic riboswitches acting on transcription. RNA Biol 12(2):221–231 Wachter A (2010) Riboswitch-mediated control of gene expression in eukaryotes. RNA Biol 7:67–76 Wang S, White KA (2007) Riboswitching on RNA virus replication. Proc Natl Acad Sci U S A 104:10406–10411 Weigand JE, Suess B (2007) Tetracycline-aptamer-controlled regulation of pre-mRNA splicing in yeast. Nucleic Acids Res 35:4179–4185 Weigand JE, Sanchez M, Gunnesch E-B et al (2008) Screening for engineered neomycin riboswitches that control translation initiation. RNA 14:89–97 Welz R, Breaker RR (2007) Ligand binding and gene control characteristics of tandem riboswitches in Bacillus anthracis. RNA 13:573–582 Werstuck G, Green MR (1998) Controlling gene expression in living cells through small moleculeRNA interactions. Science 282:296–298 Wieland M, Hartig JS (2008) Artificial riboswitches: synthetic mRNA-based regulators of gene expression. Chembiochem 9:1873–1878 Winkler WC, Breaker RR (2005) Regulation of bacterial gene expression by riboswitches. Annu Rev Microbiol 59:487–517

140

S. Harbaugh et al.

Wittmann A, Suess B (2012) Engineered riboswitches: expanding researchers’ toolbox with synthetic RNA regulators. FEBS Lett 586:2076–2083 Wu M-C, Lowe PT, Robinson CJ et al (2015) Rational re-engineering of transcriptional silencing PreQ1 riboswitch. J Am Chem Soc 137:9015–9021 Xiu Y, Jang S, Jones JA et al (2017) Naringenin-responsive riboswitch-based fluorescent biosensor module for Escherichia coli co-cultures. Biotechnol Bioeng 114:2235–2244 Xu J, Carrocci TJ, Hoskins AA (2016) Evolution and characterization of abenzylguanine-binding RNA aptamer. Chem Commun 52:549–552 Zadeh JN, Steenberg CD, Bois JS et al (2011) NUPACK: analysis and design of nucleic acid systems. J Comput Chem 32:170–173 Zuker M (2003) Mfold web server for nucleic acid folding and hybridization prediction. Nucleic Acids Res 31:3406–3415

7

Integration of Sensor Cells into Hardware Platforms Rajendra P. Shukla, Avia Lavon, and Hadar Ben-Yoav

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Hardware Platforms for Sensor Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Platform Substrate Material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Platform Fabrication Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Platform Architecture and Geometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sensor Cells Immobilization onto Hardware Platforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sensor Cells Immobilization Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Immobilization Method Biocompatibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sensor Cells Bio-response Transduction to Bioelectronic Signals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Electrochemical Transducers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Optical Transducers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Mechanical Transducers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sensor Cells-Integrated Platform Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sensor Cell Biological Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Bioelectronic Interface and Physicochemical Transducer Responses . . . . . . . . . . . . . . . . . . . . . . Conclusions and Future Directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

142 142 143 144 145 146 147 148 149 150 152 154 155 155 157 157 158

Abstract

Cell biosensors have shown great potential for their use in various analytical applications, such as environmental monitoring and biomedical diagnostics. The main aspect of this field is the integration of live sensor cells into miniaturized hardware platforms that enable portability of the integrated system and improve its detection performance. Here, the authors review the (1) various hardware R. P. Shukla · H. Ben-Yoav (*) Nanobioelectronics Laboratory (NBEL), Department of Biomedical Engineering, Ben-Gurion University of the Negev, Beer-Sheva, Israel e-mail: [email protected] A. Lavon Department of Biomedical Engineering, Ben-Gurion University of the Negev, Beer-Sheva, Israel © Springer Nature Switzerland AG 2022 G. Thouand (ed.), Handbook of Cell Biosensors, https://doi.org/10.1007/978-3-030-23217-7_122

141

142

R. P. Shukla et al.

platform types that are used for cell biosensors; (2) sensor cell immobilization strategies onto these platforms; (3) transduction mechanisms of the biological response to a bioelectronic signal; and (4) modeling of the sensor cell-hardware platform bioelectronic interface. Keywords

Integrated cell-based biosensors · Bio-microelectromechanical systems · Biochip · Biomedical diagnostics · Cell immobilization · Bioelectronics · Physicochemical transducers · Microfabrication · Bio-interface

Introduction Successful interfacing of sensor cells into hardware platforms has established a new and exciting era that results in the development of hybrid devices and systems, merging both organic and inorganic materials. These hybrid devices have a multidisciplinary nature that converges biological, physical, and chemical disciplines into one system that are continuously interacting with each other, generating unique characteristics that can be used in variety of applications (Ben-Yoav et al. 2011; Elad et al. 2008; Gui et al. 2017; Kim et al. 2018; Lim et al. 2018; ShachamDiamand et al. 2010). As the integration of sensor cells with hardware platforms is crucial for the successful performance of the hybrid system, in this chapter the authors review the major integration components in the system, summarize recent advances in this field, and stress the importance of understanding the bioelectronic interface properties. The authors list hardware platforms that are used for sensor cell integration, immobilization strategies of the cells, commonly used physicochemical transducers that translate the biological response to a measurable signal, and theoretical models of the bioelectronic interface that are developed to improve the understanding and future engineering of optimal sensor cells-integrated hardware platforms. Finally, the authors highlight what the authors believe are the current challenges and future directions in the field.

Hardware Platforms for Sensor Cells Rigid and soft materials are commonly used as substrates for sensor cells hardware platforms. The substrate’s material type, surface properties, architecture, and geometry have a crucial factor in the interaction of the sensor cells with the platform – their immobilization and bioelectronic signal transduction mechanisms. Precision fabrication tools with micro- and nano-range resolution are utilized to control the substrate properties with low contamination rate of micron-size particles of similar size as the biological cells (Fig. 1). Nowadays, the research community mostly uses a small pool of materials as sensor cells platforms that, despite the vast amount of knowledge on how to properly utilize these materials and tailor them for specific

7

Integration of Sensor Cells into Hardware Platforms

143

Fig. 1 Scheme of the platform substrate properties that can affect sensor cells immobilization and biological response

applications, limit the capabilities of bio-integrated hardware platforms. In this section, the authors present an overview of the basic principles of hardware platforms (substrate material, architecture, and geometry) applicable for cell biosensors and their advantages and limitations in interfacing sensor cells.

Platform Substrate Material Most commonly used platform substrate materials for sensor cells are silicon, glass, metals, metal oxides, and plastics. Among these materials, silicon is the most commonly used substrate material due to the vast amount of experience and methods available due to its common use in the semiconductors industry. Silicon is preprocessed to be available in the form of a “wafer” – a disk-like and thin (thickness of 0.25–1 mm) substrate with a typical diameter of 75–300 mm. Silicon is very attractive due to its unique electrical and mechanical properties that enable microfabricating various microstructures for numerous micromachining applications. (Krotz et al. 1982) Moreover, the surface roughness of the silicon substrate can be controlled by polishing methods, enabling unique surface characteristics that are attractive to biological molecules, such as cells and proteins. Despite the attractive electrical and mechanical characteristics provided by silicon, several fundamental limitations are present. Silicon substrates are not transparent to light, thus preventing the use of commonly used biological characterization methods such as transmission microscopy and fluorescent plate readers. Furthermore, the high cost of silicon may limit its utilization for large area (>1 cm2) disposable platforms. Other semiconductors (e.g., GaAs) are more brittle and expensive than silicon – characteristics that dramatically limit their use as substrate materials for sensor cells platforms (Thrush et al. 2005). Another substrate material that is commonly used for sensor cells platforms is glass. Despite its lower compatibility with microfabrication methods compared to silicon, glass provides some unique features; chief among them is its optical transparency. Glass substrates are available in two main forms: fused silica and borosilicate. Fused silica glass is pure amorphous silicon dioxide (SiO2) that can withstand high temperatures, has a wide range of optical transparency (down to 1800 Å wavelengths), and radiates very low autofluorescence. On the other hand,

144

R. P. Shukla et al.

borosilicate glass cannot be exposed to high temperatures that is required for some microfabrication methods, such as thin-film deposition, radiates high autofluorescence that masks fluorescence-based sensor cells, but can be easily bonded with silicon substrates forming integrated platforms. Soft polymeric substrates provide attractive alternative to silicon and glass due to their ease-of-fabrication, low cost, mechanical flexibility, and biocompatibility. Among the variety of studied soft substrates, most commonly used in the last decade are polydimethylsiloxane (PDMS) and polyimide (Khan et al. 2015; Liu et al. 2017; Saem et al. 2017). Polydimethylsiloxane substrates are manufactured using a microfabricated mold, resulting in a transparent PDMS molding permeable to air, yet, radiating high autofluorescence and comprising poor mechanical stability due to its low Young modulus and the large thermal expansion coefficient. Another challenge of using PDMS as the sensor cells substrate material is its hydrophobic nature that may cause difficulties filling closed platform structures (such as microfluidic channels) with aqueous solutions and may increase platform fouling due to increased absorbance of hydrophobic compounds. Various approaches have been explored to increase the hydrophilicity of PDMS by chemically modifying its surface (Ni et al. 2009), yet, as it affects also its biocompatibility to the integrated sensor cells, there is a need to investigate other materials and approaches for the unique integration with biological cells.

Platform Fabrication Approach Microfabrication technology is used to manufacture the hardware platforms and to curve specific features in the platform’s substrate that are utilized for the interfacing with the sensor cells, such as to improve immobilization or to enable transduction of the biological response. The platform can be microfabricated by using either “top-down” (carved out of the substrate bulk material) approach or “bottom-up” (gradually formed by an addition of material to the substrate’s surface) approach, or the combination of both approaches (Bustillo et al. 1998; Kovacs et al. 1998). Five basic technologies can be employed to microfabricate the platform: 1. Lithography – photons or particles are used to pattern features in a layer of a photo- (“photoresist”) or a particle- (“electron beam resist”) sensitive polymer, respectively. The patterning is performed by means of light irradiation for a photoresist-coated substrate or electron beam irradiation for an electron beam resist-coated substrate. As features within the size range of microns down to nanometers, organic contaminants may form on the substrate surface due to the multiple process steps involved and can damage the integrity of the pattern features. 2. Thin-film growth/deposition – a physical or chemical reaction-driven process that forms several atoms to a few nanometers thick films of various metals or inorganic materials. As part of the deposition process, high temperatures are

7

Integration of Sensor Cells into Hardware Platforms

145

applied that significantly affect the surface properties and enable the modifications with the films. 3. Etching – a selective removal of material from the microfabricated area (either the deposited thin films or the bulk substrate), defining patterns of features. The material removal can be done by either wet (solution-assisted) or dry (gas-phase chemistry) etching processes. Both wet and dry etching can lead to isotropic and anisotropic process profiles that can be utilized for biological cell trapping in particular locations on the platform substrate. 4. Substrate bonding – separate microfabricated substrates are directly attached together to form an integrated device combining their microfabricated properties. For example, glass-capping substrate can be bonded onto a silicon structure forming an optically accessible integrated micro-system. 5. Surface and near surface modification – implantation of ions or chemical modification (SiO2 wettability modification using self-assembled silane compounds) of the surface of the substrate that alter the functional properties of the substrate (such as electrical and biophysical attraction). Other than the conventional microfabrication technologies listed above, new methods have been developed specifically for biological applications. Whitesides and Xia (Qin et al. 2010; Xia and Whitesides 1998) have reported the “soft lithography” technology; they utilized microstructured surfaces as a mold for an elastomeric molding (PDMS) that following curing in an oven, patterns are generated with feature sizes ranging from 30 nm to 100 μm. This method is simple and relatively inexpensive since the PDMS molding can be repeatedly used from the microfabricated master mold. Soft lithography, which is “solvent free” method, is advantageous over conventional lithography for biologists since many polymeric materials, routinely employed in the biological laboratory, are damaged by the organic solvents used in conventional lithography. The soft lithography method is later developed into five advanced technologies: microcontact printing (μCP), replica molding (REM), microtransfer molding (μTM), micromolding in capillaries (MIMIC), and solvent-assisted micromolding (SAMIM).

Platform Architecture and Geometry Micro-features in the hardware platform affect the integration of biological cells. Specifically, the architecture and geometry properties of these micro-features have dominant effect on the adherence of the biological cell to the platform and its resulted biological response (Chuang et al. 2006; Khademhosseini et al. 2004; Li et al. 2000; Liu et al. 2010; Ng et al. 2017; Zhu et al. 2012). Therefore, scientists have been investigating various microfabricated structures with unique micro-features as sensor cell platforms. For example, researchers have used a 150 μm diameter microholes to facilitate cell adherence (Fig. 2) (Deng et al. 2014; Zeck and Fromherz 2001). Other structures have been also investigated to guide cell adhesion, such as micropyramids, microchannels, and microwells (Peterson et al. 2005;

146

R. P. Shukla et al.

Fig. 2 Neuron silicon chip. (a) Electronmicrograph of two-way contact with picket fence made of polyimide. Stimulator wings (St) and transistor (S, source; D, drain; G, gate) are marked (Scale bar = 20 μm.). (b) Electronmicrograph, after fixation, of a neuron (Scale bar = 20 μm.). (c) Micrograph of neuronal cell bodies (dark blobs) in picket fences on a circle of two-way contacts connected by neurites (bright threads) (Scale bar = 100 μm) (Zeck and Fromherz 2001)

Yoetz-Kopelman et al. 2016). The shape and size of the microstructure platform controls the cell mobility and response. While simple microstructures can be easily fabricated, more complex architectures are harder to control and may currently limit the potential of utilizing microstructures for precise spatial control of sensor cells integration and biological response.

Sensor Cells Immobilization onto Hardware Platforms Immobilization of sensor cells onto substrate platforms has a major role on the functionality of the cells, hence specific design considerations need to be taken into account when strategizing the appropriate method. For example, an immobilization matrix that limits the flow of nutrients to the cells and the removal of waste products will affect the functionality of the cells (Drachuk et al. 2017; Durrieu et al. 2016; Ghiglione et al. 2016). Therefore, creative solutions have been developed to improve cells response, which sometimes involve more than one classical immobilization method (Efremenko et al. 2016). Over the years, various sensor cell immobilization methods have been developed and reviewed (Partovinia and Rasekh 2018; Sekoai et al. 2018). In this section, the authors present an overview of the chemical and physical cell immobilization strategies and the main considerations related to biocompatibility effects (Fig. 3).

7

Integration of Sensor Cells into Hardware Platforms

147

Fig. 3 Scheme of the sensor cells immobilization onto hardware platforms

Sensor Cells Immobilization Method The immobilization methods can be categorized according to the physical or chemical mechanism governing the interactions between the sensor cells and the platform substrate.

Covalent Attachment Covalent bonds are generated between the functional groups on the cell’s membrane (such as NH2–, –OH, –SH) and reactive groups (inorganic or organic matrix) present at the platform surface. Usually, chemical functionalization or activation of the platform’s surface area is performed prior to cells’ immobilization. Cross-Linking Organic bonding molecules are utilized as cross-linkers (such as aldehydes and mainly glutaraldehyde) between the sensor cells and the platform surface. For example, glutaraldehyde was used to cross-link between Bacillus sp. sensor cells and the conductive polymer polyaniline-carboxylated multiwalled carbon nanotubes composite for the detection of paracetamol (Bayram and Akyilmaz 2016). Affinity Binding The binding between the sensor cells and the platform substrate are based on functional interactions. Most common affinity-based immobilization is the avidin (sometimes streptavidin)-biotin binding. Electrostatic, Van der Waals, Hydrophobic, and Ionic Interactions Weak interactions without electron sharing happen between the sensor cells and the platform substrate and they are based on their opposite charge (or partial charge). These weak interactions do not allow the sensor cell to be properly adsorbed to the substrate. For example, Fe3O4 nanoparticles were used to functionalize gold electrodes to facilitate the adhesion of Proteus mirabilis cells (Braham et al. 2015). Another work modified graphene oxide substrate with poly-L-lysine to assist E. coli cells adhesion using electrostatic interactions between the cells and the positively charged substrate (Romero-Vargas Castrillon et al. 2015).

148

R. P. Shukla et al.

Entrapment Sensor cells can be confined within semipermeable materials, such as fibers, membranes, or polymeric matrices. The entrapment of the cells using a polymeric matrix is commonly used among the sensor cells biochip community as it can be easily performed and maintains the activity of the cells. For that purpose, sensor cells, such as E. coli, and polymers are mixed before being polymerized chemically or physically depending on the nature of the polymer (e.g., k-carrageenan, agar, alginate, gelatin, an agar/gelatin mixture, chitosan, and polyvinyl alcohol gel, polyurethanepolycarbomyl sulfonate, polyacrylamide, polyvinyl alcohol, and sol-gel) (Belkin 2003; Ben-Yoav et al. 2011; Chouteau et al. 2005; Eltzov and Marks 2011; Ferro et al. 2012; Gutiérrez et al. 2011, 2015; Verma et al. 2016; Van Der Meer and Belkin 2010). The deposition of the cell-polymer bio-composite onto the platform substrate can be induced by either a chemical or an electronic trigger. For electronic triggering example, alginate solution containing suspended CaCO3 particles is mixed with sensor cells and can be deposited on electrodes by applying electrical current that generates electrolysis conditions, changes the pH near the electrode, and electrodeposits calcium alginate hydrogel (Fig. 4) (Cheng et al. 2012). Moreover, in the case of E. coli, polyvinyl alcohol gel has showed the highest apparent bioactivity and the longest storage time (40 days) in pH 7.0 phosphate-buffered saline solution (Liu et al. 2018). Another example are Anabaena torulosa cyanobacteria sensor cells, which are used for the detection of heavy metals, 2,4-dichlorophenoxyacetate, and chlorpyrifos, and are entrapped in a cellulose membrane through filtration method. The entrapped cells-matrix bio-composite are dried and fixed into a cylindrical well, which is designed to be attached to an optical probe (Wong et al. 2013). Recently, it has been reported that eggshell membrane can be used to immobilize sensor cells (Wen et al. 2014). As eggshell membrane is mainly composed of biological molecules and protein fibers, together with its net-veined structure and gas-permeable hydrophilic properties, it provides an excellent biological microenvironment for the cell to survive and maintain its enzymatic activity. Mammalian cells can be immobilized by standard cell seeding and adhesion methods. A common practice is pretreating the seeding surface with different adhesion factors (e.g., collagen, laminin, poly-L-ornithine) based on cell type and membrane properties (Liu et al. 2007).

Immobilization Method Biocompatibility The cell’s viability is maintained by different parameters of its close environment, such as solution temperature, pH and ionic strength, nutrients concentrations, dynamics, substrate architecture and chemistry, and the vulnerability of the cells to mechanical stress (Bonde et al. 2013; Perullini et al. 2007). An important factor to be considered during sensor cells immobilization is the effect of the immobilization method and its material on the viability of the cells. Using entrapment

7

Integration of Sensor Cells into Hardware Platforms

149

Fig. 4 Multilayer cell assembly on individual electrodes. (a–b) Schematic diagram (a) and fluorescence micrograph (b) of a multilayer structure of, sequentially, an electrodeposited alginate gel with a layer of pure gel sandwiched by two layers of gel with entrapped E. coli cells expressing GFP. (c–d) Schematic diagram (c) and fluorescence micrograph (d) of a multilayer structure of sequentially electrodeposited alginate gels with a layer of gel with entrapped E. coli cells expressing GFP sandwiched by two layers of pure alginate gel. (e–f) Schematic diagram (e) and fluorescence (f) micrograph of a multilayer structure of cell populations with a co-deposition sequence of RFP
, GFP
, and BFP-expressing E. coli cells. (Cheng et al. 2012)

immobilization method with polymeric matrix offers improved viability of the sensor cells. Recent work has demonstrated using sol-gel matrices to encapsulate cells and to maintain their possible division (Blocki et al. 2017; Wang et al. 2015). Maintenance of the possibility of the cells to divide within the encapsulation matrices expends the possible applications of cell biosensors to monitor external effects on cell growth (Hassan et al. 2016; Si et al. 2016).

Sensor Cells Bio-response Transduction to Bioelectronic Signals A crucial part of integrating sensor cells into hardware platforms is the ability to effectively transduce the biological response of the cells to a measurable output signal, which is not necessarily an electrical one. Numerous physicochemical transducers have been developed for cell biosensors, including electrochemical, optical, mechanical, magnetic, and thermometric transducers. In this section, the authors will review the most commonly used transducers for cell biosensors, i.e., electrochemical, optical, and mechanical transducers (Fig. 5).

150

R. P. Shukla et al.

Fig. 5 Scheme of physicochemical transducers that convert the biological response to a measurable bioelectronic signal

Electrochemical Transducers Electrochemical transducers are widely used among the platforms for cell sensors and are applied in numerous applications, such as medical diagnostic, environmental monitoring, and food health assessment. The physical size, geometry, and electrode material of electrochemical transducers are governed by their application. Electrochemical transducers are highly sensitive, provide rapid response, and can be designed to be selective toward the analyte in question (Han et al. 2013). As opposed to other physicochemical transducers, electrochemical transducers don’t suffer from an inherent sensitivity limitation and single molecules can be measured. Furthermore, the modular nature of electrochemical transducers enables their easy miniaturization and incorporation into portable micro-systems hardware, offering less expensive read-out electronics. In this section, electrochemical transducers are categorized and reviewed according to the transducer electrochemical signal, i.e., potentiometric (measures potential), amperometric (records current), and impedometric (analyzes impedance) detection (Lafleur et al. 2016).

Potentiometric Transducers Potentiometry transducers are based on measuring the electric potential difference between the working and the reference electrodes, while the potential difference is dependent on molecules (redox-active, ions, etc.) in the vicinity of the working electrode. Commonly used potentiometric transducers include a gas-sensing electrode or an ion-selective electrode, such as pH, ammonium, or chloride. The sensitivity and selectivity of potentiometric transducers can be significantly improved by modifying the electrodes with selective films, acting as a filtering barrier, and screening interfering molecules. However, a highly stable and accurate reference electrode is always required and challenging to maintain, which may potentially limit the application of potentiometry in cell biosensors. Amperometric Transducers Amperometric transducer are based on the changes in the applied potential between the working electrode and the reference electrode, while the generated electrochemical current can be recorded and correlated with the concentration of target compounds. Cell biosensors utilize amperometric transducers which are used to measure the current generated by the oxidation or reduction of electro-active species secreted

7

Integration of Sensor Cells into Hardware Platforms

151

by the sensor cells in response to external stimuli. Amperometric transducers are simple, sensitive, and can be easily miniaturized. However, while their specificity can be depended on the applied electric potential, in cases of high potential, other redox-active species may overlap and mask with the signal of the analyte, leading to inaccuracies in results (Higson 2012; Kim et al. 2006) – a particularly important challenge in the presence of biological media, which may contain a wealth of potential interferents. Many examples of amperometric cell biosensors are available in literature (Ionescu et al. 2006; Kanyong et al. 2016; Zhang et al. 2014). For example, amperometric transducers can be used to measure oxidation current generated by enzymatic products that are produced in the presence of the analyte (Fig. 6) (BenYoav et al. 2009a). Furthermore, sensor bacteria can be immobilized on a surface of an electrode and metabolize the analyte. While the bacteria degraded organic molecules, electro-active species were consumed or produced by the bacteria, generating a measurable electrochemical signal that can be quantified and correlated to the levels of the target analyte (Chen and Shamsi 2017; Liu and Mattiasson 2002; Su et al. 2011).

a

b

Ag/AgCl Reference electrode Au Working electrode Au Counter electrode Micro-chamber wall

c

Induced bacteria, + pAPP Induced bacteria, No pAPP Non-induced bacteria, + pAPP

i / nA

6

3 Addition of pAPP 0 0

600 t/s

1200

Fig. 6 (a) Silicon-based microchip comprises four differentially sized electrochemical microchambers. (b) Inside view of a single three-electrode electrochemical micro-chamber. (c) Chronoamperometric results of bacterial cells following 1 h of incubation in the presence and the absence of the toxicant Nalidixic acid (Ben-Yoav et al. 2009a)

152

R. P. Shukla et al.

Impedometric Transducers Impedometric (or conductometric) transducers measure impedance changes in the solution due to the production or consumption of ions (for measuring changes in the bulk solution) and electro-active (for measuring changes at the electrode–electrolyte interface) species (Bahadir and Sezgintürk 2016; De La Rica et al. 2009; Tran et al. 2016). Impedometric transducers are highly sensitive to small changes of the sample’s chemical composition; they are suitable for miniaturization (reference electrode is not required) and large-scale production using inexpensive technology of thin-film electrodes. Transducers are not light sensitive, the driving voltage can be sufficiently low to reduce power consumption, and a large spectrum of compounds of different natures can be identified on the basis of various reactions and mechanisms (Jaffrezic-Renault and Dzyadevych 2008). However, as impedometric transducers are very sensitive to any ions that change the conductivity of the solution, their selectivity is relatively poor (Hassan et al. 2016). Despite that limitation, impedometric transducers are used in a variety of cell biosensors. Changes of the ionic composition in the solution bulk can be induced by sensor cells in response to an analyte. For example, the known metabolic and enzymatic activity of microorganisms (Rodriguez-Mozaz et al. 2006) can be used to detect the response of microbial cell sensors (Lawrence and Moores 1972; Su et al. 2011). The microalgae Chlorella vulgaris was used as a cell biosensor by measurement of the inhibition of microalgae’s alkaline phosphatase activity in presence of Cd(II) ions (Berezhetskyy et al. 2007). Alternatively, sensor cells can induce changes at the electrode–electrolyte interface that can be measured by scanning the interface response to dielectric spectroscopy (also known as electrochemical impedance spectroscopy [EIS]). Electrochemical impedance spectroscopy is a sensitive and label-free electrochemical transduction mechanism that responds to changes in the physicochemical properties of the interface, such as the charge transfer resistance, double layer capacitance, and ionic diffusion several nanometers near the electrode. Therefore, EIS is used to measure changes in the electro-active species concentrations produced or consumed by a sensor cell induced in the presence of an analyte. These transducers can also be utilized to detect analyte–bioreceptor interaction induced by the sensor cells, which cause changes in the double layer capacitance and the charge transfer resistance.

Optical Transducers Optical transducers detect changes of the optical properties of the sample, which can be due to binding events between a labeling molecule and the cell sensor, production or consumption of photon-emitting labeling molecule, etc. The physical nature of these optical changes can be in the form of absorption, emission, transmittance, scattering, reflection, and refraction. While optical transducers are present in various mechanisms and are commonly reviewed (Dippel et al. 2018; Højris et al. 2016; Jiang et al. 2016; La Rosa et al. 2015), the most common optical transducers are based on colorimetric, chemiluminescent, fluorescent, and bioluminescent detection and will be reviewed in this section.

7

Integration of Sensor Cells into Hardware Platforms

153

Colorimetric Transducers Colorimetric transducers detect color changes that can be measured by a naked eye, smart phone camera, or scanner (An et al. 2016; Hu et al. 2017; Kim et al. 2017). The detected optical signal is then processed and amplified using digital signal processing tools to extract the relevant information about the sample. Colorimetric transducers are commonly used in sensor cells that respond in a form of an enzymatic reaction and are considered as the easiest and less expensive method for such goal. The color products generated by the sensor cell can be monitored using a spectrophotometer detector, while an increase of the reaction product is directly linked to the substrate concentration. While optical transducers are very selective by using specific labeling molecules, one of their major limitations is the relatively long time (hours) required for the enzymatic reaction in the sensor cell to reach threshold level that will generate a measurable signal. Moreover, to increase the intensity of the optical signal, preparation steps are done to collect and concentrate the cells in the sample. Chemiluminescent Transducers Similar to colorimetric transducers, chemiluminescent transducers transform chemical energy into light that can be measured using a luminometer. Platforms for chemiluminescent transducers are commercially available and provide various properties for sensitivity, duration of light emission, and compatibility with other reporter systems. Fluorescent Transducers Fluorescent transducers are based on the emission of light from a fluorophore molecule following its excitation with a stimulus light. The response of the sensor cell to an analyte is usually related to the fluorescence intensity. While the sensor cell response can be based on enzymatic reaction, one of the most common fluorescent transducers is the production of fluorescent proteins – chief among them is the Green Fluorescent Protein (GFP). The advantage of these proteins is the possibility for a quantitative online monitoring of promoter activity in viable cells (e.g., growing liquid cultures in microtiter plates) without the need of harvesting cells and performing an assay (Kremers et al. 2011). However, a major limitation of fluorescent proteins is the autofluorescence of the cells due to basal activity in the cell. Bioluminescent Transducers Bioluminescent transducers are based on proteins that are produced in the sensor cell and emit photons without the need for excitation (such as in florescence). These transducers offer large dynamic range and high sensitivity due to the absence of background signals generated in an optical system that require excitation. For example, a chamber for sensor cells is designed in a half sphere shape and coated with aluminum film to provide a reflective interface to allow maximizing the flux of photons generated from sensor cells in response to toxicity (Fig. 7) (Ben-Yoav et al. 2009b).

154

R. P. Shukla et al.

Micro-reaction chambers

a

b

Photo-detector Micro-fluidic channel

c

Photo-detector diameter - 2rd

Micro-reaction chamber

Rays emitted from bacteria in fixation rod

Bacterial fixation rod diameter - 2ra

Fig. 7 Modeling of a toxicity analysis biochip: (a) The layout of the biochip; (b) the manufactured biochip; and (c) simulation of the bioluminescence rays distribution in the toxicity analysis biochip (Ben-Yoav et al. 2009b)

Mechanical Transducers Mechanical transducers can detect mechanical changes and are highly sensitive to mass variations. Most common mechanical transducers are based on physical mechanisms such as surface acoustic wave (SAW) and quartz crystal microbalance (QCM) (Marx et al. 2001). Typically, a SAW transducer detects variation in the velocity of a surface acoustic wave due to binding events that are related to the sensor cells using interdigitated transducers on a ceramic resonator or quartz crystalbased piezoelectric substrate (Mamishev et al. 2004). On the other hand, QCM monitors changes in the resonating frequency of a piezoelectric quartz crystal transducer. An external electrical potential to a piezoelectric material produces internal mechanical stress, which oscillates quartz crystal at its resonance frequency. Attributed to the effect of mass changes (Kaspar et al. 2000; Muramatsu et al. 1986), sensor cell’s response, for example, a produced bio-product, can be detected due to the binding at the solid–liquid interface (Olsen et al. 2003). A QCM was utilized to transduce the biological response of living endothelial cells into piezoelectric signal (Marx et al. 2001).

7

Integration of Sensor Cells into Hardware Platforms

155

Sensor Cells-Integrated Platform Modeling Modeling of the biological, physical, and chemical reactions, happening at the interface between the sensor cells and the hardware platform, is an important step in understanding the dominant parameters that affect the integrated bioelectronic system. Unravelment of these parameters is a powerful tool that enables simulating the sensor cell biological response and maximizing the output bioelectronic signal. Developing a model requires basic understanding of the mechanisms involved in the biological cell (such as the induction rate of the reporter gene and replication rate of the sensor cell) at the interface between the cell and the transducer (such as the mass transfer rate of the reporting molecule or the absorption of the light, which is generated by the cells, by the immobilization matrix) and by the physicochemical transducer (such as the electron transfer rate in an electrochemical reaction). The model describes these mechanisms using a series of state variables and equations, and their solution provides spatial and temporal information about the integrated system (Fig. 8). While models can be developed using either mathematical calculation (theoretical models) or numerical solutions (using software packages), in this section, the authors will focus on theoretical models of sensor cells-integrated platforms.

Sensor Cell Biological Response Modeling the biological response of the sensor cell requires taking into account the inducer interaction with the cell’s membrane and promoter, transcription rate of the reporting genes from DNA to RNA, translation rate of the RNA to reporting proteins or enzymes, biochemical kinetics of the enzymes, secretion rate of the reporting proteins or the enzymatic products outside of the cell, and replication rate of the cells. While most of these biological reactions can be described by similar equations, the choice of the reporting mechanism (such as protein or enzyme, generation of photons or enzymatic products, etc.) varies between different integrated systems. For

Fig. 8 Scheme of the main biological, physical, and chemical mechanisms of the integrated sensor cell-transducer system that govern the bioelectronic signal

156

R. P. Shukla et al.

example, a reporting mechanism that involves photoluminescence requires the assumption that the intensity of the light output signal is directly proportional to the concentration of the fluorescent proteins, such as GFP. Another parameter that affects the concentration of the fluorescent proteins is their generation rate by the sensor cell. The generation rate is dependent on the type and concentration of the inducer, promotor gene, reporter gene, and ambient conditions. Shacham-Diamand and colleagues modeled the response of E. coli bacterial sensor cells that were genetically engineered to produce the reporting protein GFP in the presence of the toxicant inducer nalidixic acid (Shacham-Diamand et al. 2011). The model describes the relationship between the GFP generation rate and the promoter concentration that triggers the GFP reporter gene’s expressing (Eq. 1). dC GFP ¼ αPr dt

(1)

Where CGFP is the concentration of the expressed GFP, α is the generation rate constant, and Pr is the promoter concentration. The promoter concentration is dependent on the bacterial density and the nalidixic acid concentration. Therefore, using a first-order approximation, the promoter induction rate can be described as in Eq. 2: dPr ¼ μG0 C Tox dt

(2)

Where μ is the induction rate constant, G0 is the bacterial density, and CTox is the concentration of the nalidixic acid toxicant concentration. Other work has modeled the biological response of a cell sensor that is designed to be integrated with an electrochemical transducer (Ben-Yoav et al. 2013). The reporting mechanism is based on an enzyme that its product is an electrochemicallyactive species that can be measured on a surface of an electrode. In this model, the promoter concentration is modeled using the following equation (Eq. 3): Pr ¼ μG0 C Tox

(3)

Where Pr is the promoter’s product concentration, μ is the promoter’s production constant, G0 is the bacterial density, and CTox is the concentration of the toxicant inducer. The enzyme production rate is dependent on the promoter concentration and is modeled using the following equation (Eq. 4): dE T ¼ α∙Pr dt

(4)

Where ET is the total enzyme concentration and α is the enzyme production rate constant. By combining Eqs. 3 and 4, the total enzyme concentration is described as (Eq. 5): dET ¼ k 0 C Tox dt

(5)

7

Integration of Sensor Cells into Hardware Platforms

157

Where k0 is a rate constant that comprised of the product of α, μ, and G0.

Bioelectronic Interface and Physicochemical Transducer Responses The response mechanism of a photoluminescent transducer is modeled based on the intensity of the light generated by the sensor cells (Eq. 6) (Shacham-Diamand et al. 2011). Y ≜I out ¼ ɳ C GFP I ex

(6)

Where Y is the output signal, Iout is the output light intensity, CGFP is the GFP concentration, and Iex is the excitation light intensity. The output signal generated by an electrochemical transducer can be modeled as electrons transferred between the electro-active product generated by the sensor cells and the electrode. The electro-active product diffuses through the interface between the cell and the electrode, and the total amount of electrons transferred due to the electrochemical reaction can be modeled (Ben-Yoav et al. 2013). Solving this system using the diffusion equation can be done under the assumption that diffusion is the dominant transport mechanism, resulting in the total electrochemical charge transferred in the sensor cell-integrated system (Eq. 7): rffiffiffiffiffiffi Dp Qðt Þ ¼ nFA π

5

3

8Bt 2 4Ct 2 þ 15 3

! (7)

Where B ¼ k 0 C Tox C¼

k2k1S k 1 S þ k
1 þ k 2

k2k1S ½E0 þ k 0 C Tox ðt 1
t 0 Þ k 1 S þ k
1 þ k 2

Where Q is the generated electrochemical charge, Dp is the product diffusion coefficient, E0 is the native enzyme concentration present in the sensor cell at the addition time of the toxicant inducer (basal concentration in the bacteria), t0 is the addition time of the toxicant inducer, t1 is the addition time of the electrochemical substrate, n is the stoichiometric number of electrons involved in an electrode reaction, and F is Faraday constant.

Conclusions and Future Directions The integration of live sensor cells into diverse microfabricated hardware platforms has been described in a wide range of approaches and applications with the goal to highlight key design guidelines. The platform substrate’s material, architecture, and

158

R. P. Shukla et al.

geometry are key parameters in the integration of cells with hardware platforms. Despite the numerous reported substrate materials, polymeric substrates have gained much interest due to their fabrication simplicity, low cost, and improved biocompatibility. However, these substrate materials suffer from limited robustness that should be improved to fit the performance shown by their silicon and glass counterparts. Another important parameter that is discussed in this chapter is the biological cell’s immobilization strategy. The observed viability of the immobilized cells remains to be a challenging task in this field. In addition to the required biocompatibility of the immobilization method, the effect on nutrients accessibility and cell replication need to be monitored as they directly affect the cell’s biological response (such as decreased enzymatic activity) to external induction. Reproducible immobilization of the cells is another important challenge that requires developing more accurate deposition methods. This chapter also reviews the various mechanisms involved in the transduction of the cell’s biological response to a quantifiable output signal. Signal characteristics, such as sensitivity, stability, and reproducibility, significantly impact the transduction efficiency and are affected by biological, chemical, and physical mechanisms at the bioelectronic interface. Furthermore, the influence of environmental parameters on the interface should be closely monitored in order to decrease potential interferences and to optimize the performance of the transducer. The integration of microfluidic networks provides a beneficial addition to cell biosensors due to the improved control of various environmental effects (such as temperature), use of low volume samples, and the additive effect of laminar flow that can be used to improve analyte and reporting molecules transport rate. Understanding the governing physicochemical mechanisms in the sensor cell-integrated platform enables controlling the output signal of the hybrid system. Therefore, the output signal can be modeled to identify dominant dynamics and reaction kinetics that can be used to optimize the performance of the biosensor. Theoretical models have been reviewed in this chapter with the emphasis on optical and electrochemical transducer-based platforms. As current models include basic assumptions in their development, further improvement of these models can be considered to provide a comprehensible analysis of different physical and chemical aspects in the system and their effect on the generated bioelectronic signal.

References An R, Zhang D, Chen Y, Cui YZ (2016) A “turn-on” fluorescent and colorimetric sensor for selective detection of Cu2+ in aqueous media and living cells. Sens Actuators B Chem 222:48–54 Bahadir EB, Sezgintürk MK (2016) A review on impedimetric biosensors. Artif Cells Nanomed Biotechnol 44(1):248–262 Bayram E, Akyilmaz E (2016) Development of a new microbial biosensor based on conductive polymer/multiwalled carbon nanotube and its application to paracetamol determination. Sens Actuators B Chem 233:409–418 Belkin S (2003) Microbial whole-cell sensing systems of environmental pollutants. Curr Opin Microbiol 6(3):206–212

7

Integration of Sensor Cells into Hardware Platforms

159

Ben-Yoav H, Biran A, Pedahzur R, Belkin S, Buchinger S, Reifferscheid G, Shacham-Diamand Y (2009a) A whole cell electrochemical biosensor for water genotoxicity bio-detection. Electrochim Acta 54(25):6113–6118 Ben-Yoav H, Elad T, Shlomovits O, Belkin S, Shacham-Diamand Y (2009b) Optical modeling of bioluminescence in whole cell biosensors. Biosens Bioelectron 24(7):1969–1973 Ben-Yoav H, Melamed S, Freeman A, Shacham-Diamand Y, Belkin S (2011) Whole-cell biochips for bio-sensing: integration of live cells and inanimate surfaces. Crit Rev Biotechnol 31(4):337–353 Ben-Yoav H, Biran A, Sternheim M, Belkin S, Freeman A, Shacham-Diamand Y (2013) Functional modeling of electrochemical whole-cell biosensors. Sens Actuators B Chem 181:479–485 Berezhetskyy A, Durrieu C, Nguyen-Ngoc H, Chovelon J-M, Dzyadevych S, Tran-Minh C (2007) Conductometric biosensor based on whole-cell microalgae for assessment of heavy metals in wastewater. Biopolym Cell 23(6):511–518 Blocki A, Löwenberg C, Jiang Y, Kratz K, Neffe AT, Jung F, Lendlein A (2017) Response of encapsulated cells to a gelatin matrix with varied bulk and microenvironmental elastic properties. Polym Adv Technol 28(10):1245–1251 Bonde S, Berthing T, Madsen MH, Andersen TK, Buch-Månson N, Guo L, Li X, Badique F, Anselme K, Nygård J, Martinez KL (2013) Tuning InAs nanowire density for hek293 cell viability, adhesion, and morphology: perspectives for nanowire-based biosensors. ACS Appl Mater Interfaces 5(21):10510–10519 Braham Y, Barhoumi H, Maaref A, Bakhrouf A, Grauby-Heywang C, Cohen-Bouhacina T, Jaffrezic-Renault N (2015) Characterization of urea biosensor based on the immobilization of bacteria proteus mirabilis in interaction with iron oxide nanoparticles on gold electrode. J Biosens Bioelectron 6(1):1000160 Bustillo J, Howe R, Muller R (1998) Surface micromachining for microelectromechanical systems. Proc IEEE 86(8):1552–1573 Chen S, Shamsi MH (2017) Biosensors-on-chip: a topical review. J Micromech Microeng 27 (8):083001 Cheng Y, Tsao CY, Wu HC, Luo X, Terrell JL, Betz J, Payne GF, Bentley WE, Rubloff GW (2012) Electroaddressing functionalized polysaccharides as model biofilms for interrogating cell signaling. Adv Funct Mater 22(3):519–528 Chouteau C, Dzyadevych S, Durrieu C, Chovelon JM (2005) A bi-enzymatic whole cell conductometric biosensor for heavy metal ions and pesticides detection in water samples. Biosens Bioelectron 21(2):273–281 Chuang CH, Lee YC, Huang HS, Hsiao FB, Hsu YM, Wang KH, Ding L, Hsu HJ (2006) 3D microstructures array single-cell-based dep chip for studying apoptosis of U937 & A431 cells. In: Proceedings of 1st IEEE international conference on nano micro engineered and molecular systems (1st IEEE-NEMS), pp 1207–1210 De La Rica R, Thompson S, Baldi A, Fernandez-Sanchez C, Drain CM, Matsui H (2009) Label-free cancer cell detection with impedimetric transducers. Anal Chem 81(24):10167–10171 Deng Y, Yi P, Peng L, Lai X, Lin Z (2014) Experimental investigation on the large-area fabrication of micro-pyramid arrays by roll-to-roll hot embossing on PVC film. J Micromech Microeng 24(4):045023 Dippel AB, Anderson WA, Evans RS, Deutsch S, Hammond MC (2018) Chemiluminescent biosensors for detection of second messenger cyclic di-GMP. ACS Chem Biol 13:1872. https://doi. org/10.1021/acschembio.7b01019. (in print) Drachuk I, Harbaugh S, Geryak R, Kaplan DL, Tsukruk VV, Kelley-Loughnane N (2017) Immobilization of recombinant E. coli cells in a bacterial cellulose-silk composite matrix to preserve biological function. ACS Biomater Sci Eng 3(10):2278–2292 Durrieu C, Ferro Y, Perullini M, Gosset A, Jobbágy M, Bilmes SA (2016) Feasibility of using a translucid inorganic hydrogel to build a biosensor using immobilized algal cells. Environ Sci Pollut Res 23(1):9–13 Efremenko EN, Maslova OV, Kholstov AV, Senko OV, Ismailov AD (2016) Biosensitive element in the form of immobilized luminescent photobacteria for detecting ecotoxicants in aqueous flowthrough systems. Luminescence 31:1283–1289

160

R. P. Shukla et al.

Elad T, Lee JH, Belkin S, Gu MB (2008) Microbial whole-cell arrays. Microb Biotechnol 1(2): 137–148 Eltzov E, Marks RS (2011) Whole-cell aquatic biosensors. Anal Bioanal Chem 400(4):895–913 Ferro Y, Perullini M, Jobbagy M, Bilmes SA, Durrieu C (2012) Development of a biosensor for environmental monitoring based on microalgae immobilized in silica hydrogels. Sensors 12(12):16879–16891 Ghiglione J-F, Martin-Laurent F, Pesce S (2016) Microbial ecotoxicology: an emerging discipline facing contemporary environmental threats. Environ Sci Pollut Res 23(5):3981–3983 Gui Q, Lawson T, Shan S, Yan L, Liu Y (2017) The application of whole cell-based biosensors for use in environmental analysis and in medical diagnostics. Sensors-Basel 17(7):1623 Gutiérrez JC, Amaro F, Díaz S, De Francisco P, Cubas LL, Martín-González A (2011) Ciliate metallothioneins: unique microbial eukaryotic heavy-metal-binder molecules. J Biol Inorg Chem 16(7):1025–1034 Gutiérrez JC, Amaro F, Martín-González A (2015) Heavy metal whole-cell biosensors using eukaryotic microorganisms: An updated critical review. Front Microbiol 6:1–8 Han KN, Li CA, Seong GH (2013) Microfluidic chips for immunoassays. Annu Rev Anal Chem 6(1):119–141 Hassan SHA, Van Ginkel SW, Hussein MM, Abskharon R, Oh SE (2016) Toxicity assessment using different bioassays and microbial biosensors. Environ Int 92–93:106–118 Higson S (2012) Biosensors for medical applications. Woodhead Publishing, Sawston, Cambridge, UK Højris B, Christensen SCB, Albrechtsen HJ, Smith C, Dahlqvist M (2016) A novel, optical, on-line bacteria sensor for monitoring drinking water quality. Sci Rep 6:1–10 Hu Y, Zhao F, Hu S, Dong Y, Li D, Su Z (2017) A novel turn-on colorimetric and fluorescent sensor for Fe3+ and its application in living cells. J Photochem Photobiol A Chem 332(3):351–356 Ionescu RE, Abu-Rabeah K, Cosnier S, Durrieu C, Chovelon JM, Marks RS (2006) Amperometric algal chlorella vulgaris cell biosensors based on alginate and polypyrrole-alginate gels. Electroanalysis 18(11):1041–1046 Jaffrezic-Renault N, Dzyadevych S (2008) Conductometric microbiosensors for environmental monitoring. Sensors-Basel 8(4):2569 Jiang H, Jiang D, Shao J, Sun X, Wang J (2016) High-throughput living cell-based optical biosensor for detection of bacterial lipopolysaccharide (LPS) using a red fluorescent protein reporter system. Sci Rep 6:1–12 Kanyong P, Hughes G, Pemberton RM, Jackson SK, Hart JP (2016) Amperometric screen-printed galactose biosensor for cell toxicity applications. Anal Lett 49(2):236–244 Kaspar M, Stadler H, Weiß T, Ziegler C (2000) Thickness shear mode resonators (“mass-sensitive devices”) in bioanalysis. Fresenius J Anal Chem 366(6):602–610 Khademhosseini A, Yeh J, Jon S, Eng G, Suh KY, Burdick JA, Langer R (2004) Molded polyethylene glycol microstructures for capturing cells within microfluidic channels. Lab Chip 4(5):425–425 Khan S, Lorenzelli L, Dahiya RS (2015) Technologies for printing sensors and electronics over large flexible substrates: a review. IEEE Sensors J 15(6):3164–3185 Kim HJ, Bennetto HP, Halablab MA, Choi C, Yoon S (2006) Performance of an electrochemical sensor with different types of liposomal mediators for the detection of hemolytic bacteria. Sens Actuators B Chem 119(1):143–149 Kim SD, Koo Y, Yun Y (2017) A smartphone-based automatic measurement method for colorimetric pH detection using a color adaptation algorithm. Sensors 17(7):1604 Kim HJ, Jeong H, Lee SJ (2018) Synthetic biology for microbial heavy metal biosensors. Anal Bioanal Chem 410(4):1191–1203 Kovacs GTA, Maluf NI, Petersen KE (1998) Bulk micromachining of silicon. Proc IEEE 86(8): 1536–1551 Kremers G-J, Gilbert SG, Cranfill PJ, Davidson MW, Piston DW (2011) Fluorescent proteins at a glance. J Cell Sci 124(2):157–160

7

Integration of Sensor Cells into Hardware Platforms

161

Krotz G, Legner W, Wapner C, Moller H, Sonntag H, Muller G (1982) Silicon as a mechanical material. In: Proceedings of the international solid-state sensors and actuators conference – transducers ’95, vol 2, no 5, pp 186–189 La Rosa SL, Solheim M, Diep DB, Nes IF, Brede DA (2015) Bioluminescence based biosensors for quantitative detection of enterococcal peptide – pheromone activity reveal inter-strain telesensing in vivo during polymicrobial systemic infection. Sci Rep 5:8339–8339 Lafleur JP, Jönsson A, Senkbeil S, Kutter JP (2016) Recent advances in lab-on-a-chip for biosensing applications. Biosens Bioelectron 76:213–233 Lawrence AJ, Moores GR (1972) Conductimetry in enzyme studies. Eur J Biochem 24(3):538–546 Li M, Glawe JD, Green H, Mills DK, Mcshane MJ, Gale BK (2000) Effect of high-aspect-ratio microstructures on cell growth and attachment. In: Proceedings of the 1st annual international IEEE-EMBS special topic conference on microtechnologies in medicine and biology, pp 531–536 Lim HG, Jang S, Jang S, Seo SW, Jung GY (2018) Design and optimization of genetically encoded biosensors for high-throughput screening of chemicals. Curr Opin Biotechnol 54:18–25 Liu J, Mattiasson B (2002) Microbial bod sensors for wastewater analysis. Water Res 36(15): 3786–3802 Liu Q, Cai H, Xu Y, Xiao L, Yang M, Wang P (2007) Detection of heavy metal toxicity using cardiac cell-based biosensor. Biosens Bioelectron 22(12):3224–3229 Liu L, Luo C, Ni X, Wang L, Yamauchi K, Nomura SM, Nakatsuji N, Chen Y (2010) A microchannel-well system for culture and differentiation of embryonic stem cells on different types of substrate. Biomed Microdevices 12(3):505–511 Liu YL, Qin Y, Jin ZH, Hu XB, Chen MM, Liu R, Amatore C, Huang WH (2017) A stretchable electrochemical sensor for inducing and monitoring cell mechanotransduction in real time. Angew Chem Int Ed 56(32):9454–9458 Liu C, Xu Y, Han X, Chang X (2018) The fabrication and the use of immobilized cells as test organisms in a ferricyanide-based toxicity biosensor. Environ Toxicol Chem 37(2):329–335 Mamishev AV, Sundara-Rajan K, Yang F, Du Y, Zahn M (2004) Interdigital sensors and transducers. Proc IEEE 92(5):808–844 Marx KA, Zhou T, Montrone A, Schulze H, Braunhut SJ (2001) A quartz crystal microbalance cell biosensor: detection of microtubule alterations in living cells at nm nocodazole concentrations. Biosens Bioelectron 16(9–12):773–782 Muramatsu H, Kajiwara K, Tamiya E, Karube I (1986) Piezoelectric immuno sensor for the detection of candida albicans microbes. Anal Chim Acta 188:257–261 Ng K, Gao B, Yong KW, Li Y, Shi M, Zhao X, Li Z, Zhang XH, Pingguan-Murphy B, Yang H, Xu F (2017) Paper-based cell culture platform and its emerging biomedical applications. Mater Today 20(1):32–44 Ni M, Tong WH, Choudhury D, Rahim NA, Iliescu C, Yu H (2009) Cell culture on mems platforms: a review. Int J Mol Sci 10(12):5411–5441 Olsen EV, Pathirana ST, Samoylov AM, Barbaree JM, Chin BA, Neely WC, Vodyanoy V (2003) Specific and selective biosensor for salmonella and its detection in the environment. J Microbiol Methods 53(2):273–285 Partovinia A, Rasekh B (2018) Review of the immobilized microbial cell systems for bioremediation of petroleum hydrocarbons polluted environments. Crit Rev Environ Sci Technol 48(1):1–38 Perullini M, Rivero MM, Jobbágy M, Mentaberry A, Bilmes SA (2007) Plant cell proliferation inside an inorganic host. J Biotechnol 127(3):542–548 Peterson SL, Mcdonald A, Gourley PL, Sasaki DY (2005) Poly(dimethylsiloxane) thin films as biocompatible coatings for microfluidic devices: cell culture and flow studies with glial cells. J Biomed Mater Res A 72(1):10–18 Qin D, Xia Y, Whitesides GM (2010) Soft lithography for micro-and nanoscale patterning. Nat Protoc 5(3):491

162

R. P. Shukla et al.

Rodriguez-Mozaz S, Lopez De Alda MJ, Barceló D (2006) Biosensors as useful tools for environmental analysis and monitoring. Anal Bioanal Chem 386(4):1025–1041 Romero-Vargas Castrillon S, Perreault FO, De Faria AF, Elimelech M (2015) Interaction of graphene oxide with bacterial cell membranes: insights from force spectroscopy. Environ Sci Technol Lett 2(4):112–117 Saem S, Zhu Y, Luu H, Moran-Mirabal J (2017) Bench-top fabrication of an all-pdms microfluidic electrochemical cell sensor integrating micro/nanostructured electrodes. Sensors 17(4):732 Sekoai PT, Awosusi AA, Yoro KO, Singo M, Oloye O, Ayeni AO, Bodunrin M, Daramola MO (2018) Microbial cell immobilization in biohydrogen production: a short overview. Crit Rev Biotechnol 38(2):157–171 Shacham-Diamand Y, Belkin S, Rishpon J, Elad T, Melamed S, Biran A, Yagur-Kroll S, Almog R, Daniel R, Ben-Yoav H, Rabner A, Vernick S, Elman N, Popovtzer R (2010) Optical and electrical interfacing technologies for living cell bio-chips. Curr Pharm Biotechnol 11(4): 376–383 Shacham-Diamand Y, Almog R, Daniel R, Rabner A, Popovtzer R (2011) Nano-bio interfacing with living cell biochips. In: Sattler KD (ed) Handbook of nanophysics: nanomedicine and nanorobotics. CRC Press, Boca Raton Si R-W, Yang Y, Yu Y-Y, Han S, Zhang C-L, Sun D-Z, Zhai D-D, Liu X, Yong Y-C (2016) Wiring bacterial electron flow for sensitive whole-cell amperometric detection of riboflavin. Anal Chem 88(22):11222–11228 Su L, Jia W, Hou C, Lei Y (2011) Microbial biosensors: a review. Biosens Bioelectron 26(5):1788–1799 Thrush E, Levi O, Cook LJ, Deich J, Kurtz A, Smith SJ, Moerner WE, Harris JS (2005) Monolithically integrated semiconductor fluorescence sensor for microfluidic applications. Sens Actuators B Chem 105(2):393–399 Tran TB, Son SJ, Min J (2016) Nanomaterials in label-free impedimetric biosensor: current process and future perspectives. Biochip J 10(4):318–330 Van Der Meer JR, Belkin S (2010) Where microbiology meets microengineering: design and applications of reporter bacteria. Nat Rev Microbiol 8(7):511–522 Verma N, Sharma R, Kumar S (2016) Advancement towards microfluidic approach to develop economical disposable optical biosensor for lead detection. Austin J Biosens Bioelectron 2(2):1021 Wang X, Ben Ahmed N, S Alvarez G, V Tuttolomondo M, Hélary C, F Desimone M, Coradin T (2015) Sol-gel encapsulation of biomolecules and cells for medicinal applications. Curr Top Med Chem 15(3):223–244 Wen G, Wen X, Shuang S, Choi MMF (2014) Whole-cell biosensor for determination of methanol. Sens Actuators B Chem 201:586–591 Wong LS, Lee YH, Surif S (2013) Whole cell biosensor using anabaena torulosa with optical transduction for environmental toxicity evaluation. J Sens 2013:1 Xia YN, Whitesides GM (1998) Soft lithography. Annu Rev Mater Sci 37(5):551–575 Yoetz-Kopelman T, Dror Y, Shacham-Diamand Y, Freeman A (2016) “Cells-on-beads”: a novel immobilization approach for the construction of whole-cell amperometric biosensors. Sens Actuators B Chem 232:758–764 Zeck G, Fromherz P (2001) Noninvasive neuroelectronic interfacing with synaptically connected snail neurons immobilized on a semiconductor chip. PNAS 98(18):10457–10462 Zhang Y, Bai X, Wang X, Shiu KK, Zhu Y, Jiang H (2014) Highly sensitive graphene-pt nanocomposites amperometric biosensor and its application in living cell H2O2 detection. Anal Chem 86(19):9459–9465 Zhu M, Baffou G, Meyerbröker N, Polleux J (2012) Micropatterning thermoplasmonic gold nanoarrays to manipulate cell adhesion. ACS Nano 6(8):7227–7233

8

Whole-Cell-Based Fiber-Optic Biosensors Boris Veltman and Evgeni Eltzov

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Whole-Cell-Based Fiber-Optic Biosensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Biosensors as Analytical Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Fiber-Optic Transducers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Whole-Cell Bioreporters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Immobilization Approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Biosensors for Toxicity Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Determining Chemicals in Water . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Determining Chemicals in Soil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Determining Chemicals in Air . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions and Future Directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

164 166 166 168 169 172 175 175 180 181 181 182

Abstract

In the last decade, various whole-cell-based fiber-optic biosensors have been described. All of those reports have shown great potential for their use in agriculture, healthcare, and environmental fields. Genetically programmed to produce a measurable signal (e.g., light emission, changes in color or current), such bioreporters, coupled with fiber-optic transducers, create fast, simple, and easy measurement tools for the determination of contaminant presence, bioavailability, and relative toxicity in tested samples. In this chapter, the B. Veltman Institute of Biochemistry, Food Science and Nutrition, Robert H. Smith Faculty of Agriculture, Food and Environment, Hebrew University of Jerusalem, Rehovot, Israel Department of Postharvest Science, The Volcani Center, Agricultural Research Organization, Bet Dagan, Israel E. Eltzov (*) Department of Postharvest Science, The Volcani Center, Agricultural Research Organization, Bet Dagan, Israel e-mail: [email protected] © Springer Nature Switzerland AG 2022 G. Thouand (ed.), Handbook of Cell Biosensors, https://doi.org/10.1007/978-3-030-23217-7_126

163

164

B. Veltman and E. Eltzov

advances and drawbacks of whole-cell fiber-optic applications are reviewed, and the critical challenges that need to be overcome to implement these applications in the real world are discussed. Keywords

Biosensor · Fiber-optic · Whole cell · Bioreporter · Bioluminescence

Introduction From the first whole-cell-based applications proposed by Gary Sayler’s group in 1990 (Harms et al. 2006), many whole-cell-based biosensors and bioassays have been developed and used in various areas, such as agriculture, environment, and healthcare. In general, conventional analytical approaches (e.g., high-pressure liquid chromatography, gas chromatography–mass spectroscopy, enzyme-linked immunosorbent assay) are more sensitive than biosensors and allow precise detection and determination of the target analytes (Eltzov and Marks 2011). However, their high cost, the requirement for skilled personnel and a well-equipped laboratory, and lack of real-time monitoring capabilities make these approaches unfavorable for many applications. Biosensors, on the other hand, together with operation simplicity, low price, and portability, not only provide sensitive and rapid detection of the target analytes but also do so continuously and in real time. A typical biosensor (Fig. 1) has three parts: a recognition element (e.g., antibodies, DNA, enzymes, microorganisms), a transducer (e.g., amperometric (Ding et al. 2008), potentiometric (Bobacka et al. 2008), voltammetric (Alpat et al. 2008), conductimetric (Lei et al. 2006), optical (Eltzov and Marks 2010), colorimetric (Azevedo et al. 2005)), and an interface (immobilization methods). The principles of measurement by biosensors are based on changes in a biorecognition element during the interaction with a specific analyte and the monitoring of such changes with transducers. Biosensors based on antibodies (Liu et al. 2011; Wingren and Borrebaeck 2006), DNA (Ehrenreich 2006; Petersen and Kawasaki 2007), enzymes (Ansari and Husain 2012), and other proteins (Kricka et al. 2006; Yang et al. 2011) impart a superior level of specificity and sensitivity. However, their applicability is limited by some shortcomings, such as the high cost of macromolecules (purified enzymes, proteins, and DNA), limited detection capability, and the short useable lifetime of the identifying molecules (Gui et al. 2017). In contrast to molecular-based bioreporters, whole-cell-based applications have numerous benefits, e.g., low cost, simple maintenance of growth and viability, reagent-free reactions, no need for pre- or posttreatment steps, low susceptibility to biological contamination, the ability to use large and homogeneous populations, and robustness to a variety of physical and chemical environments. However, the most important benefit of whole-cell bioreporters is that they can be molecularly engineered to provide measurable responses (colorimetric, electrochemical, fluorescent, or luminescent) in a dosedependent manner. This is achieved by the fusion of a suitable reporter system to

8

Whole-Cell-Based Fiber-Optic Biosensors

165

Fig. 1 Schematic representation of a fiber-optic-based biosensor

a selective promoter (Gui et al. 2017). Furthermore, a combination of strains that are sensitive to different stresses or analytes allows the creation of specific “fingerprints” that will show not only the presence of the toxicants in the sample but also predict their biological effects (Jouanneau et al. 2011). The second part of the biosensor is an interface that links the bioreporter recognition element to a given transducer. Traditional methods for the immobilization of microorganisms include adsorption, encapsulation, entrapment, covalent binding, and crosslinking (Belkin 2003; Bjerketorp et al. 2006; D’Souza 2001a, b; Ding et al. 2008). Even though immobilization strategy usually depends on the chosen bioreporter, some matrix parameters (e.g., durability and stability, their effect on cell viability, specificity and sensitivity, approximation of the fixed cells to the transducer and its effect on diffusion rates)

166

B. Veltman and E. Eltzov

always have to be considered (Collings and Caruso 1997). The transducer transforms the specific bioreporter signals to measurable values, and optical techniques (e.g., fiber optics) are most commonly used in whole-cell-based applications. The key to the success of such sensors is based on their ability to tackle difficult measurement situations where the use of conventional sensors is not suitable. Despite the fact that whole-cell-based approaches allow simple, real-time, and multiple analyte determination in complex media, these technologies are still not ready for commercial use due to the need for highly sensitive devices (e.g., photomultiplier tube (PMT), colorimeter) to detect the cells’ responses and problems with long-term preservation (shelf life). In this chapter, the latest advances in whole-cell-based fiber-optic biosensors are reviewed. Recent research on new bioreporters and technologies for patterning microorganisms on fiber-optic surfaces with various immobilization polymers will be discussed, along with studies of related applications. Also presented is a summary of the potential challenges and future prospects of the practical applications of whole-cell-based fiber-optic biosensors.

Whole-Cell-Based Fiber-Optic Biosensors Biosensors as Analytical Tools In the literature, a biosensor is described as a self-contained, integrated receptor –transducer device that can provide selective quantitative or semi-quantitative analytical information and that uses a biological recognition element (bioreceptor) and transducer in intimate contact (Thevenot et al. 2001). The optimal biosensor not only responds to low chemical concentrations but also discriminates target analytes among species according to the immobilized biorecognition elements. The typical biosensor has three major parts: the biorecognition component, the interface (immobilization techniques), and the transducing element (Eltzov et al. 2008). In the first part, bioreporter recognition elements (e.g., whole cells, nucleic acids, immunological molecules (natural or synthetic) and enzymes) are selectively interacted with a target analyte(s) to ensure sensor selectivity (Fig. 1). This chapter will concentrate on biosensors using whole-cell organisms. The transducing element transforms the specific bioreporter’s signals to measurable values. In general, there are four major groups of transducing elements: 1. Electrochemical: Determines electrical changes produced during a biological reaction 2. Optical: Measures bacterial, chemical, and enzymatic light reactions 3. Mass-sensitive: Monitors extremely low changes in mass induced by binding of the analyte to recognition elements 4. Thermal: Detects heat production during a biological reaction

8

Whole-Cell-Based Fiber-Optic Biosensors

167

Electrochemical transducers are widely used in many biosensor applications due to their high sensitivity, compatibility with modern micromanufacturing technologies, portability, disposability, and minimal power requirement (Lei et al. 2006; Wang 2002). The measuring processes are based on the electrical changes produced by whole-cell bioreporter activities (production or consumption of electroactive species) which can be monitored using conductimetric, amperometric, or potentiometric methodologies. In the amperometric-based application, the current generated by the oxidation or reduction of a species at the surface of the electrode (operated at a fixed potential with respect to a reference electrode) is correlated to the biological activity of immobilized cells (Ben-Yoav et al. 2009). Conventional potentiometricbased approaches are based on the changes in potential resulting from ion accumulation or depletion. The signal is correlated to the concentration of the target analyte and is measured with potentiometric transducers as the difference between a working and reference electrode (Simonian et al. 1998; Tran 1993). In whole-cell biosensors, such changes are generated by microbe layers that are immobilized above ion-selective (e.g., pH, ammonium, or chloride) or gas-sensing (e.g., pCO2 and pNH3) electrodes. An additional extremely sensitive electrical whole-cell-based approach measures changes in solution conductance due to microorganism reactions (Lei et al. 2006). Also common is transducer technology based on optical principles, e.g., ultraviolet (UV)–visible absorption, bioluminescence, chemiluminescence, fluorescence, phosphorescence, reflectance, scattering, and changes in refractive index produced by interaction of the receptor with the target analyte (Collings and Caruso 1997; Sergeyeva et al. 2000). The main advantage of this technique is its potential for highthroughput screening, enabling the monitoring of multiple analytes simultaneously (Brogan and Walt 2005). Bioluminescence and fluorescence are the two most widely used technologies in whole-cell-based optical applications. The fluorescent technique is based on differences between excitation and emission wavelengths of fluorescent substances. It can be divided into “in vivo” and “in vitro” types: in the “in vitro” approach, microorganisms change the environment surrounding them and in the “in vivo” approach, reporter molecules (e.g., green fluorescent protein (GFP)) are produced inside bioreporters without any external addition. The bioluminescent sensing technique is based on luminescence emitted by genetically modified microorganisms which respond to the target analyte in a dose-dependent manner. In bioluminescent microbial-based biosensors, lux (the gene encoding luciferase in cells) is the most commonly used reporter system (Marks et al. 2007). The colorimetric sensing technique used with whole-cell biosensors involves the conversion of a chromogen substrate into a colored compound by the metabolic activity of the microbial sensing element (Su et al. 2011). The simplicity and low cost of colorimetric biosensors have led to the development many whole-cell applications (Yagi 2007). The interface in a typical biosensor links the biorecognition elements to a given transducer (Fig. 1). Choosing the optimum strategy for whole-cell immobilization on the transducer surface is a necessary and critical step in the design of biosensors. These techniques not only place the biological layer in proximity to the transducer

168

B. Veltman and E. Eltzov

but they also maintain its activity and sensitivity after immobilization. During sensor operations, the immobilization matrix must remain stable and durable, and provide (in some cases) the possibility for biomaterial reuse (Collings and Caruso 1997; D’Souza 2001a). In the following sections, immobilization methodologies and bioreporters which have been used in fiber-optic applications are further described.

Fiber-Optic Transducers Most fiber-optic biosensor applications are conducted in the visible and near-infrared spectral range (from 340 nm to 2 μm). Suitable types of fibers (e.g., single-mode, multi-mode, etc.) are made mainly from SiO2 (Pospisilova et al. 2015) and commonly range from 80 μm to several millimeters in diameter. A typical optical fiber has two parts: core (refractive index n1) and cladding (refractive index n2) (Fig. 1). For light propagation through the fiber, the refractive index of the core must be larger than that of the cladding (i.e., n1 > n2). Internal reflection and propagation through the fiber occur only when the light strikes the core, and the angle of incidence is larger than the critical angle, defined by Snell’s law (θc = sin – 1[n2/n1]) (Marazuela and Moreno-Bondi 2002). During reflection, a small portion of light penetrates the reflecting medium by a fraction of the wavelength. The intensity of this electromagnetic field (termed evanescent wave) decays exponentially with distance, starting at the interface and extending into the medium with the lower refractive index (Marazuela and Moreno-Bondi 2002). Although this principle has been widely applied in the design of immunosensors, it is not used in whole-cell-based applications. In general, fiber-optic transducers can be used in combination with absorption (bio/chemo/fluoro)-luminescence, Raman, and surface plasmon resonance optical techniques, while the whole-cell bioreporters are immobilized directly on, or close to the optical fiber. Despite some drawbacks (e.g., the requirement for proper light isolation due to light interference, inability to simultaneously monitor more than one analyte with the same transducer, and inability to detect the target analyte without additional biomarkers), incorporation of an optical fiber into a biochemical sensor confers a number of advantages: 1. The transducer can monitor the light emitted from bioreporters without their intimate contact with the optical fiber, enabling a wider range of noninvasive configurations. 2. Together with geometric convenience and flexibility, fibers can be used to transmit light over long distances. 3. A light guide is not only nonelectrical and therefore free from signal interference and atmospheric disturbances, but it can also carry more information than an electrical wire. The optical fiber can transmit multiple optical signals simultaneously, thereby offering multiplexing capabilities for sensing (Eltzov and Marks 2010). Furthermore, temperature-dependence of the fiber is lower than that of electrodes.

8

Whole-Cell-Based Fiber-Optic Biosensors

169

4. The possibility of miniaturization not only reduces the cost of the fiber optics system but also minimizes sample volumes. Smaller volumes result in saving on reagent costs, enable convenient portability and storage, and provide access to difficult areas. 5. The optical fiber’s mechanical properties (e.g., robust and high tensile strength) allow its use in harsh environments. 6. In addition to the high-efficiency coupling of the blue–green region, which is ideal for bioluminescence, SiO2-based optical fibers can be chemically modified when required. 7. Optrode systems are polyvalent, so they can be easily adapted from one wholecell reporter system to another (Eltzov and Marks 2010). Fiber-optic performance can also be improved with different optimization procedures. For example, the sensitivity of a whole-cell sensor for benzene, toluene, xylene, and ethylbenzene was increased by immobilizing bioreporters on tapered quartz optical fibers. Fiber tapering enables increasing photon-detection efficiency by increasing the number of light sources on its widened end (Kuncová et al. 2016). Light-coupling efficiency can also be optimized by surface modifications, which provide easier adsorption of bioreporter cells at the fiber’s end (Zajíc et al. 2016). Figure 2 summarizes advantages and disadvantages of the fiber-optic biosensors.

Whole-Cell Bioreporters Whole living cells have been used as biorecognition elements in biosensors for many years. Despite longer response times and poorer selectivity compared to enzyme biosensors, whole cells have several advantages: 1. Living cells contain a large number of enzymes for use in the biosensor as compared to the use of pure enzymes. 2. Whole cells are more tolerant to changes in pH and temperature than purified enzymes.

Fig. 2 Advantages and disadvantages of fiber-optic biosensors

170

B. Veltman and E. Eltzov

3. New strains of microorganisms (e.g., bacteria, fungi, yeast, etc.) with unique properties are continually being discovered and isolated from natural sources. 4. Whole cells usually contain all of the enzymes, cofactors, and substrates for the detection of analytes and can therefore be used for real-time monitoring. 5. Whole-cell bioreporters are inexpensive and can be continuously regenerated by regrowing. However, the main advantage of whole cells is that they can be genetically programmed, by recombinant DNA techniques, to produce a measurable response (e.g., light emission, changes in color or current) to specific pollutants or stresses. To achieve this, a reporter gene encoding a protein or enzyme is fused to an inducible promoter taken from prokaryotic (bacteria) or eukaryotic (fungi, algae, animal) cellular-adaptation mechanisms against chemical, physical, or biological stresses. During the triggering process, a regulatory protein recognizes the particular chemical (or physical) stress, and together with response mechanisms activates the promoter, causing the reporter gene to be expressed and to begin synthesis of the reporter proteins. Thus, whole-cell bioreporters not only detect the target analytes in the sample but also determine their biological effects. In many cases, chemicals with dissimilar structures have the same toxic effect; in such cases, the determination of the toxic effect is more important than the identification of the compound (Botana 2014; Marks et al. 2007). There are two main assay types in whole-cell-based optical biosensor systems: constitutive (“light off”) and inducible (“light on”) (Fig. 3). The “light off” system is based on reduction of the naturally high basal expression of the reporter gene due to cell damage. The main problem here is that a reduced signal may be caused by a problem with the bacteria and not the specific pollutant of interest (Eltzov and Marks 2011). In the “light on” scheme, the signal (e.g., changes in color, light, current) is generated by the cells with exposure to the specific trigger (Fig. 3). Such triggers activate the reporter genes that are transcribed and translated to a measurable signal. Thus, understanding the genetic processes behind such responses will enable the selection of specific promoters and creation of bioreporters for specific chemicals or overall toxicity (e.g., genotoxicity or cytotoxicity). Furthermore, the dose dependency (i.e., the intensity of the cell’s response indicates the chemical concentration) of the cell reaction in the “light on” approach makes it more attractive for use in environmental sensing. There are many different reporter systems (e.g., fluorescence with the GFP gene and colorimetric measurements relying on the β-galactosidase (β-gal) gene lacZ), but bioluminescence is still the most widely used approach in the biosensor field (Abbas et al. 2018; Eltzov and Marks 2011; Eltzov et al. 2012; Sharifian et al. 2018). In the bioluminescent approach, the light is derived from bacterial (lux) or firefly (luc) genes’ transcription to the enzyme luciferase, which generates excited-state molecules that, when returning to the ground state, emit light. The quantum yield (i.e., efficiency conversion of chemical energy to light during enzymatic catalysis) of the firefly-based enzyme is much higher than that of the bacterial enzyme (90% compared to 5–10%, respectively (McElroy and Seliger 1963; Meighen 1988)), but

8

Whole-Cell-Based Fiber-Optic Biosensors

171

Fig. 3 Schematic representation of the bacterium-based bioluminescence mechanisms and two different measurement approaches: “light on” where light production is induced by the toxicants and “light off” where light strength is reduced by the toxicants

it requires an externally added substrate (luciferin) for the reaction, which not only complicates the monitoring process but also increases the final cost of the device (Eltzov and Marks 2010). Bacterial luciferases, on the other hand, express light without any external additions; substrate and enzyme production are controlled genetically by a five-gene operon consisting of the luxA, luxB, luxC, luxD, and luxE genes (luxCDABE) (Fig. 3). The luxC, luxD, and luxE gene products supply and regenerate the long-chain aldehyde, and the luxA and luxB (luxAB) gene products form a heterodimeric luciferase. GFP cloned from the jellyfish Aequorea victoria has also been used as a reporter system in whole-cell-based biosensors. Even though the GFP sensor does not require substrate addition, the requirement for an external light source to activate its fluorescent output makes GFP-based applications much more expensive and technologically complicated than lux systems. On the other hand, GFP-based sensors enable the development of systems based on cells with different light emissions,

172

B. Veltman and E. Eltzov

which allows the creation of multisensing platforms with a different color for each pollutant (Hever and Belkin 2006). Another reporter system is based on colorimetric changes controlled by lacZ genes, which encode the β-gal enzyme. The enzymatic reaction changes the color by hydrolyzing the substrate β-gal disaccharides into monosaccharides. In this case, the most commonly used substrate is O-nitrophenyl β-D-galactopyranoside (Yagi 2007), but others have also been used (Biran et al. 1999; Jain and Magrath 1991).

Immobilization Approaches Whole-cell fiber-optic sensors rely on microorganisms to generate a specific signal in response to chemicals or environmental conditions. For more efficient light coupling, the bioreporter should be immobilized close to the transducer. Moreover, the ideal immobilization technique not only fixes cells on the fiber optic but should also be functional at ambient temperature, biocompatible, enable complete retention of the cells, and allow the flow of nutrients, oxygen, and analytes through the matrix. In theory, encapsulated cells should remain viable but not growing (Michelini and Roda 2012). Figure 4 summarizes different immobilization methodologies. In general, microorganism-encapsulation approaches can be divided into chemical and physical concepts (Malan 1992). In the chemical approach, covalent and crosslinkage binding are used to immobilize whole cells on the transducer surface. The microorganisms are attached directly to the transducer surface by stable covalent bonds with functional groups on the cell wall (e.g., amine, carboxylic, sulfhydryl) (Arica et al. 1993; Marks et al. 2007) or by crosslinking with glutaraldehyde (Kumar and D’Souza 2010; Švitel et al. 1998). Even though this methodology allows the creation of a stable bioreporter layer, it is less attractive for biosensors with microorganisms as bioreporters due to decreasing biological activity after exposure to harmful chemicals and harsh reaction conditions. In the physical-based immobilization approaches (e.g., adsorption, encapsulation, and entrapment), both natural (agar, agarose, alginate, and carrageenan) and synthetic (polyacrylamide, polystyrene, and polyurethane) polymers have been used to fix cells. Hydrogels are networks of water-insoluble polymer chains produced in response to various triggers, e.g., ions, heat, light, or other chemicals that will also act as electrophiles. Easily controlled diffusion, high water content, pliability, and biocompatibility make these matrices the preferred choice in many biosensor applications (Elisseeff et al. 2000; Polyak et al. 2000b, 2001, 2004; Urban and Weiss 2010). As in the case of covalent immobilization, the immobilization process should be chosen carefully, due to possible damage effects (e.g., excessive heat or UV light) or a problem with diffusion resistance. A very common class of hydrogel immobilization matrices is alginates (Tombs and Harding 1998). In the presence of divalent ions (e.g., calcium (Polyak et al. 2000b), strontium (Heitzer et al. 1994)), alginates can spontaneously form gels in a single-step process. In addition to the high porosity provided by the open lattice structure and gentle environment (Polyak et al. 2004), alginates may be chemically modified to increase their immobilization properties.

Sol-Gels

(Human cells, Hepatocytes) [Yang, J., et al. 2002]

Alginates

Nonmodified

Photo polymers

Modified

(Bacteria, E.coli) [Polyak, B., et al. 2004]

Biotin-Alginate

(Mammalian cells, Myoblasts), [Rowley, J.A., et al. 1999]

Peptide-Alginate

(Bacteria, Pseudomonas), [Heitzer, A., et al. 1994]

Strontium-Alginate

Calcium-Alginate

(Bacteria, E.coli ) [Polyak, B., et al. 2000]

(Human cells, 3T3 fibroblasts) [Bryant, S.J., et al. 2000]

2-hydroxy-1-[4-(2hydroxyethoxy)phenyl]2-methyl-1-propanone

(Bacteria, E.coli ) [Premkumar, J.R., et al. 2001]

Anti E.coli antibody

Galactose moieties

Hydrogel

Whole-cell biosensors

Other

Whole-Cell-Based Fiber-Optic Biosensors

Fig. 4 Examples of the different immobilization methodologies

(Bacteria, E.coli) [Bettaieb, F., et al. 2007]

Agarose

Agarose

(Human cells, Fibrobla sts) [Desimone, M.F., et al. 2010]

Silica-Collagen

(Fungi, Trichosporon cutaneum) [Liu, L., et al. 2009]

Silica-PVA

(Bacteria, E.coli) [Premkumar, J.R., et al 2001]

Sodium-Silicate

(Humman cells, Hep G2) [Haigh-Flórez, D., et al. 2014]

Silica-PDMS-TEOS

8 173

174

B. Veltman and E. Eltzov

Nevertheless, despite several successful chemical modifications – such as coupling with short peptides (Rowley et al. 1999), galactose moieties (Yang et al. 2002), poly (ethylene glycol) diamines (Eiselt et al. 1999), and biotin (Polyak et al. 2004) – alginates are not the optimal choice for the whole-cell-based biosensors. Instability in calcium-poor solutions, deterioration in the presence of phosphate and other calcium chelators, low deformation resistance, and biodegradability have led to the search for alternative encapsulation methods. Another immobilization strategy is based on the incorporation of bioreporter organisms in sol-gels. Sol-gels are hybrid organic/inorganic compounds that bridge between glasses and polymers (Livage 1997). Due to their higher rigidity, thermal and structural stability, and transparency, sol-gel films are widely recognized as suitable tools for optical biosensors. A large variety of living cells, including bacteria, fungi, and animal cells, can be encapsulated using this technology which also extends viability rates (Desimone et al. 2010; Kataoka et al. 2005; Liu et al. 2009; Premkumar et al. 2001b). Sol-gel membranes have also been used in fiber-optics-based applications (Dai et al. 2004; Lin et al. 2006; PenaVazquez et al. 2009; Verma et al. 2012). To increase cell viability by eliminating alcohol release during polycondensation processes, some researchers have proposed using aqueous instead of alkoxide precursors for sol-gel formation (Chen et al. 2004). Cell viability can also be prolonged by adding glycerol to the immobilization process (Nassif et al. 2003), or by decreasing formation temperatures (Yu et al. 2005). Entrapment-based immobilization strategies consist of fixing cells in the membranes through different mechanical processes, such as filtration through cellulose membranes (Futra et al. 2015; Wong et al. 2013) or a quartz microfiber filter (Vedrine et al. 2003). There are many other immobilization strategies for optic applications. For example, bacteria can be fixed directly on the end of the optical fiber in wells produced by chemical etching, with each microwell being occupied by only one cell (Biran et al. 2003; Kuang et al. 2004). Several successful reports have been published on bacteria encapsulated in agarose (Alkorta et al. 2006; Bettaieb et al. 2007). The bioreporters can also be fixed in photopolymer-based matrices, where monomer crosslinking is formatted by visible or UV light (Bryant et al. 2000; Koh et al. 2002). However, the harsh chemical initiators required to facilitate membrane polymerization may damage the microorganisms during the immobilization process and reduce sensor efficiency. Another approach suggests immobilizing bacteria with an antibodymodified surface (Premkumar et al. 2001a). Antibody immobilization on various surfaces is very versatile and well developed. Thus, this approach paves the way for cell incorporation on or in virtually any substrate (Premkumar et al. 2001a). Another encapsulation technology, artificial spores, is based on bioinspired layer-by-layer (LbL) biosilification of individual cells (Yang et al. 2013). This immobilization approach mimics bacterial endospore structure by encapsulating individual cells in thin and tough shells in a cytocompatible way. The LbL coating process enables surface-functionality, which is useful in color identification and site-selective immobilization of encapsulated cells into optical devices (Yang et al. 2013). The LbL coating process has been studied for the deposition of

8

Whole-Cell-Based Fiber-Optic Biosensors

175

nanomaterials, such as magnetic nanoparticles that facilitate cell separation by a magnetic field and functional coating of optical fibers (Korposh et al. 2013).

Biosensors for Toxicity Monitoring In this section, recent advances in whole-cell fiber-optic bioassays for the detection of target toxicants in the air, soil, and water are described (Table 1). Today, the main factor limiting the application of bioreporters is the fact that they are genetically modified, and their use outside authorized laboratories is strictly regulated in developed countries such as those of the EU and the USA (EU 2009).

Determining Chemicals in Water A safe drinking-water supply is crucial to human life, and drinking water should certainly not carry any risk for humans (WHO 2011). Therefore, there is a significant need for biosensors that will allow continuous and rapid water-toxicity measurements without any additional pretreatment. Heavy metals, which are released into environmental water sources (e.g., surface, ground and sea) through natural processes and anthropogenic activities, pose a significant health risk and their toxicity is an increasing problem with respect to ecological, evolutionary, nutritional, and environmental aspects (Jaishankar et al. 2014). Different whole-cell-based fiber-optic technologies have been proposed for the detection of heavy metals in water. For example, a fluorescence-based fiber-optic toxicity biosensor based on Escherichia coli genetically modified with GFP was developed to evaluate the toxicity of common heavy metals (Futra et al. 2015). The biosensor response was stable for at least 5 weeks, and the detection limits for Cu(II), Cd(II), Pb(II), Zn(II), Cr(VI), Co(II), Ni(II), Ag(I), and Fe(III) with the proposed application were 0.04, 0.32, 0.46, 2.80, 100, 250, 400, 720, and 2600 μg/L, respectively (Futra et al. 2015). Another GFP-based approach used E. coli strain DH5α (pNV12) to detect different combinations of Pb(II), Cd(II), and Zn(II) ions in milk samples (Kumar et al. 2017). The working principle of the biosensor was based on the expression of a GFP gene under the control of the cad promoter and measurements of the bioreporters’ fluorescence responses with a fiber-optic spectrofluorometer. A good linear range for cadmium ion concentrations of 10–50 μg/L was obtained with a detection limit of 10 μg/L (Kumar et al. 2017). In another application, differences in the fluorescence emission (before and after exposure) of Anabaena torulosa allowed quantitative and qualitative detection of heavy metals (Cu, Pb, and Cd), and pesticides (2,4-dichlorophenoxyacetate (2,4-D), chlorpyrifos) (Wong et al. 2013). In this case, the bioreporter was entrapped on a cellulose membrane through filtration, dried and fixed in a cylindrical well attached to an optical fiber. The biosensor showed good sensitivity, with the lowest limits of detection for Cu, Pb, Cd, 2,4-D and chlorpyrifos determined at 1.195, 0.100, 0.027, 0.025, and 0.025 μg/L, respectively. This device was also tested with different combinations of toxicants, resulting

E. coli strain DH5α (pNV12)

Anabaena torulosa

Chlorella vulgaris, Euglena gracilis, Anabaena flosaquae

Liquid

Liquid

Liquid

Liquid

Liquid

Liquid

Liquid

Liquid

Soil

Soil Air

Cu, Pb, Cd

2,4dichlorophenoxyacetae Cd

Herbicide

Methyl parathion

L-asparagine

Mitomycin C

Cu, Ni, and Zn

Hg, As Chloroform

E. coli (MC1061) E. coli (TV1061)

Rh. leguminosarum bv. trifolii

Escherichia coli (DPD2794)

Staphylococcus sp.

Flavobacterium

Dictiosphaerium chlorelloides

Chlorella vulgaris

Anabaena torulosa

Bioreporter E. coli (DH5α)

Sample matrix Liquid

Target analyte Cu, Cd, Pb, Zn, Cr, Co, Ni, Ag, Fe Cd

Luciferase Luciferase

Luciferase

Chlorophyll a Chlorophyll a pnitrophenol L-aspartic acid Luciferase

Photosystem II Photosystem II Chlorophyll a

GFP

Reporter system GFP

Table 1 Fiber-optic-based applications for monitoring water, air and soil toxicity

2.6, 18 μg/L 10 fg/mL

0.1, 0.1 and 0.5 μg/L

0.1 mg/L

1 nM

0.3 μM

0.5 μg/L

10 6 M 50 μM 1.25 μM 0.1 μg/mL

0.025 (μg/L)

1.195, 0.100, 0.027 (μg/L)

Sensitivity 0.04, 0.32, 0.46, 2.80, 100, 250, 400, 720, and 2600 (μg/L) 10 (μg/L)

(Naessens et al. 2000) (Haigh-Flórez et al. 2014) (Verma et al. 2012) (Verma et al. 2012) (Woutersen et al. 2017) (Renella and Giagnoni 2016) (Ivask et al. 2007) (Eltzov et al. 2011, 2015a)

(Ben Ahmed et al. 2018)

(Kumar et al. 2017) (Wong et al. 2013)

References (Futra et al. 2015)

176 B. Veltman and E. Eltzov

8

Whole-Cell-Based Fiber-Optic Biosensors

177

in predominantly antagonistic responses (Kumar et al. 2017). Another application took advantage of the effect of heavy metals on chlorophyll a fluorescence intensity of the microalgae Anabaena flosaquae, Chlorella vulgaris, and Euglena gracilis. In the first step, the effects of Cd(II), Pb(II), and anthracene were studied on yeast, while complementary ATP-metry measurements demonstrated a direct relationship between the optical response and pollutant toxicity, in a cell- and dose-dependent manner. Then, cells were encapsulated in silicate–colloidal silica nanocomposite matrices and exposed to heavy metals (Ben Ahmed et al. 2018). For on-site analysis of arsenite and zinc in complex matrices, for example, serum and freshwater samples, a sensing system based on a microfluidic compact disc with immobilized whole-cell spores was proposed (Michelini et al. 2013). The flow device was coupled with a fiber-optic-based detection system and fluids were mixed by centrifugal pumping. Results were comparable to those obtained with the same assay performed in a conventional microplate format, but the application required relatively long incubation times (150 min), and the spores needed to be properly activated to recover their sensing capability, likely representing the main disadvantage of such systems (Michelini et al. 2013). Other toxicant groups (e.g., pesticides, herbicides, and fungicides) reach the ground or surface water by leaching or runoff following normal agricultural practices, or by accidental (or deliberate) spills. In the latter case, large quantities can enter the surface water, leading to high concentrations (van Leeuwen 2000). Because of their toxicity, these chemicals are considered to be a particular health hazard for the general population and should therefore be continuously monitored in the water. Different whole-cell-based biosensors have been developed for these issues. For example, herbicides were detected by kinetic changes in chlorophyll fluorescence in Chlorella vulgaris cells (Naessens et al. 2000). Cells were immobilized on removable membranes and placed in front of the tip of an optical fiber bundle. The proposed device measured the concentration of a toxic chemical in the form of a single drop or dissolved in a continuous flow. The detection of 0.1 μg/mL of a single herbicide, as required by EC legislation for drinking water, was possible with this algal biosensor, especially for atrazine, simazine, and diuron (Naessens et al. 2000). In another application, herbicides were monitored with microalgae encapsulated in a matrix of sodium silicate and glycerol. The matrix with the cells was placed at the tip of a bifurcated fiber-optic cable in a flow-through cell, and chlorophyll fluorescence increased due to photosystem destruction by the herbicide. The biosensor worked at concentrations of 0.5 μg/L to 10 mg/L of herbicide with a partially reversible response (Pena-Vazquez et al. 2009). The same authors used microalgae to construct an fiber-optic device for monitoring copper in reservoirs and water supplies (Peña-Vázquez et al. 2010). Specificity to pesticides may also be achieved by manufacturing a fiber-optic dual-head device containing both analyte-sensitive and analyte-resistant microalgal species (HaighFlórez et al. 2014). Dictyosphaerium chlorelloides was immobilized in porous silicon films and presence of the target analyte was monitored through changes in O2 concentration, where herbicide decreased O2 production in the analyte-sensitive immobilized strain without affecting the analyte-resistant population’s response.

178

B. Veltman and E. Eltzov

The biosensing device provided in situ measurements of the herbicide concentration every 180 min with a 12 μg/L limit of detection and a 50–800 μg/L working range. The main disadvantage of this approach is lack of specificity: any other pollutant might also decrease O2 production of both strains (Haigh-Flórez et al. 2014). For detection of methyl parathion (an organophosphate pesticide and insecticide), Flavobacterium was trapped in a glass-fiber filter and used as a bioreporter in a fiberoptic system. The measuring process was based on methyl parathion’s hydrolysis with organophosphorus hydrolase enzyme into the detectable product p-nitrophenol. The disposable, simple, and cost-effective biosensor enabled sensitive (0.3 μM) methyl parathion determination in a single step with high reproducibility and uniformity (Verma et al. 2012). As already noted, the main advantage of the whole-cell-based bioreporters is their ability not only to detect chemicals in water but also to determine their toxic biological effect. In many cases, no toxicological information is available for the chemicals. Therefore, whole-cell-based applications can be used as tools for the assessment of the potential toxicity of unidentified substances to avoid delays in their risk assessment. For example, due to the potential carcinogenicity of acrylamide, formed by Maillard reaction from free asparagine and a few other amino acids, there is a demand for L-asparagine monitoring in food (Verma et al. 2012). Coliform bacteria isolated from a wastewater sample from Fortis Multispecialty Hospital, Mohali, India, and immobilized with an indicator in sol-gel onto circular plastic discs adjusted to a an optical fiber were used for L-asparagine monitoring in different food samples. The measurement process was based on changes in the absorption spectrum caused by L-asparagine hydrolysis with the bacterial enzyme asparaginase. The novelty of that study lay in its describing the first-ever application of a fiber-optic approach to analyze asparagine, with minimization of the sample volume to 5 μL, a response time of 7 min and a detection limit of 1 nM (Verma et al. 2012). Another device was developed for real-time monitoring of 1,2-dichloroethane (DCA) in aqueous solutions (Campbell et al. 2006). DCA is one of the most widely used halogenated organics, and it has been released into the environment and contaminates groundwater at many hazardouswaste sites. In this application, Xanthobacter autotrophicus strain GJ10 was immobilized in calcium alginate on the tip of a fiber-optic fluoresceinamine-based pH optode. Together with a fast response time (8–10 min) and measurement reproducibility (SE 1-byte capacity. Nat Methods 11(12):1261–1266. https://doi. org/10.1038/nmeth.3147

Engineering of Sensory Proteins with New Ligand-Binding Capacities

11

Diogo Tavares, Vitali Maffenbeier, and Jan Roelof van der Meer

Contents Introduction: What Are Bioreporters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sensor Elements for Bioreporters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Strategies for Obtaining New Sensory Proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Mutagenesis and Selection of Mutant Transcription Activators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Periplasmic Binding Proteins and Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Bacterial Bioreporters Based on Methylaccepting Chemotaxis Proteins . . . . . . . . . . . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

224 225 227 229 230 233 236 236

Abstract

Biosensors based on whole cell living bacteria or their isolated components mostly use the capacity of natural protein classes to “sense,” in other words to interact with, chemical ligands. Frequently deployed sensor protein classes consist of transcription regulators, two-component sensory proteins, methylaccepting chemoreceptors and periplasmic binding proteins. All classes have been linked to some form of transducer and sensor output upon binding of the ligand, such as by de novo synthesis of reporter proteins, through Föster resonance energy transfer, from cell accumulation, from chemotaxis, or in dyeencapsulating competitive liposome assays. Despite the partially successful deployment of such biosensors, one of their current limitations is the covered spectrum of chemical ligands, which in most cases reflects the cognate ligand of the used sensory protein. Here we will summarize some of the past and recent efforts to obtain new ligand-binding specificities in these sensory protein classes. Most strategies have followed combinations of random mutagenesis, selection and screening methodologies, and protein-structure guided D. Tavares · V. Maffenbeier · J. R. van der Meer (*) Department of Fundamental Microbiology, Faculty of Biology and Medicine, University of Lausanne, Lausanne, Switzerland e-mail: [email protected] © Springer Nature Switzerland AG 2022 G. Thouand (ed.), Handbook of Cell Biosensors, https://doi.org/10.1007/978-3-030-23217-7_129

223

224

D. Tavares et al.

predictions. They have led to some success of obtaining new specificities, but which not unexpectedly lay relatively close to the original ligand. Computational structure-function predictions have become more important to reduce the mutagenesis efforts to smaller library sizes, but their success has been limited so far to those sensory proteins that do not undergo major conformational changes upon ligand-binding. In particular for the class of periplasmic binding proteins, new ligand-binding specificities have been notoriously difficult to predict, likely as a result of the current limitations in dynamic structure predictions. It is to be expected that future advances in computational algorithms will facilitate this process, which will make it more straightforward to obtain sensory proteins with targeted ligand-binding properties, enabling plug-and-play biosensor design. Keywords

Bioreporters · Transcription regulators · Periplasmic binding proteins · Methylaccepting chemotaxis proteins · Protein-structure guided predictions · Random mutagenesis · Selection · Screening

Introduction: What Are Bioreporters Bioreporters are engineered living microorganisms, which contain a synthetic genetic circuit that enables the cell to sense one or more input signals or conditions, execute the instructions imposed by the circuit and produce one or more outputs (Fig. 1) (van der Meer and Belkin 2010; Xu et al. 2013; Way et al. 2014). Traditionally, outputs of bioreporters consist of de novo synthesized (reporter) protein(s) formed in response to the input signal, yielding some sort of proportional and quantifiable relation between input and output in the context of the assay (Daunert et al. 2000; van der Meer et al. 2004). For ease of quantification, the choice of deployed reporter output has frequently fallen on light-emitting or colorproducing enzymes or autofluorescent reporter proteins, whose (spectral) activity can be easily and/or noninvasively measured (Daunert et al. 2000). Bioreporter assays are relatively simple and consist in essence of an incubation of the bioreporter cells with a liquid sample for a defined duration, after which the reporter output is recorded (van der Meer 2006; van der Meer 2010). The ease of designing a bioreporter genetic circuit, the quantifiable aspect, and the assay simplicity have attracted both scientific interest and popularity as potential alternative for environmental, medical, or food analytics (Paton et al. 2009; Siegfried et al. 2012; Merulla et al. 2013; Xu et al. 2013; Courbet et al. 2015). Numerous bioreporters have thus been produced and tested with different applications in mind (Table 1). Some have been quite successful in terms of (correctly) quantifying inputs, although eventually very few have been tested sufficiently rigorously for robustness and on real samples (Trang et al. 2005; Harms et al. 2006; Siegfried et al. 2012; Xu et al. 2013) (Table 1). Engineering and application aspects of bioreporters have been reviewed elsewhere and are not but briefly rehearsed here (van der Meer 2010). Instead, the main focus of

11

Engineering of Sensory Proteins with New Ligand-Binding Capacities

225

Fig. 1 Conceptual idea of bioreporter cells. Two scenarios are depicted, where the bioreporter senses the ligand inside (a) or outside (b) the cell and, as a consequence, trigger de novo expression of a reporter gene. This leads to formation of reporter protein in the cell, which is measured in the bioreporter assay after a defined duration of the incubation with the sample. (c) Schematic simplified instructions to be engineered on a synthetic reporter gene circuit. Colored arrows: coding regions of the sensing/regulatory protein and reporter protein; colored ellipses: ribosome binding sites; colored hexagons: transcription terminators; 90 -angle arrows: promoter sequences; colored square: operator site for the interaction of the sensor/regulatory protein with the DNA that leads to transmission of the sensed signal to altered reporter gene expression

the underlying review is the question of the sensing elements that are central for the bioreporter circuitry.

Sensor Elements for Bioreporters The sensing elements that enable the bioreporter cell to recognize an input signal are either protein or RNA (e.g., aptamers) (Bazin et al. 2017). Sensor proteins and aptamers form specific three-dimensional domains/configurations for interaction with ligands. Ligand-binding effectuates an intramolecular conformational change, which allosterically may effect a different domain in the protein, for example, one that is involved in DNA binding or yielding a chemical modification of specific amino acid residues on the protein. In other cases, the sensory proteins transmit ligand binding intermolecularly in a chain of phosphorylation events to one or more separate protein(s). In particular, transcription regulators (TRs) have been frequently deployed as sensory switches or as synthetic building blocks for hybrid switches in bioreporter circuits. TRs combine sensing and transducing domains in a single protein (although they are active mostly in multimeric forms) (Tropel and van der

RbsB of E. coli LivK (LBP) of E. coli MalE of E. coli TBP of E. coli Tsr, Tar of E. coli

Hybrids to Tar

PBP PBP PBP PBP MCP

MCP

E. coli

E. coli E. coli E. coli Cell free E. coli

E. coli E. coli E. coli

E. coli

E. coli

Host chassis Escherichia coli E. coli

gfp FRET FRET Liposomes Cell accumulation FRET

luxCDABE luxCDABE luxCDABE

luxAB

luxAB

luxAB

Reporter system lucFF

Nitrite, nitrate, L-malate

Bi et al. 2016

50 nM 400 nM 4 mM 0.5 nM 10 μM ~0.3 μM

1 nM 45 nM

Hg2+ Tetracyclines 2,4-dinitrotoluene and 2,4,6-trinitrotolene D-Ribose L-Leu Maltose Thiamine Serine, aspartate

Turner et al. 2007; Lewis et al. 2009 Sticher et al. 1997

Reference Willardson et al. 1998

Trang et al. 2005; Baumann and van der Meer 2007 Selifonova et al. 1993 Korpela et al. 1998 Yagur-Kroll et al. 2014; Belkin et al. 2017 Reimer et al. 2014 Ko et al. 2017 Iijima and Hohsaka 2009 Edwards et al. 2016 Roggo et al. 2018

5 nM

10 nM

C6-C10 alkanes Arsenite

0.4 μM

Detection sensitivity 40 μM

Ligand molecules Benzene, Toluene, and xylene Hydroxylated biphenyls

TR transcription regulator, PBP periplasmic binding protein, MCP methylaccepting chemotaxis protein

MerR of E. coli TetR of E. coli YqJF of E. coli

Sensor protein XylR of Pseudomonas putida HbpR of Pseudomonas nitroreducens AlkS of Pseudomonas oleovorans ArsR of E. coli

TR TR TR

TR

TR

TR

Sensor protein class TR

Table 1 Selected examples of periplasmic binding proteins, transcription activators, and methyaccepting chemotaxis proteins used in bioreporter systems

226 D. Tavares et al.

11

Engineering of Sensory Proteins with New Ligand-Binding Capacities

227

Meer 2004). Ligand binding leads to altered DNA-binding properties of the TR on an “operator” site, which directly influences transcription efficiency or frequency from the promoters controlled by the TR (Browning and Busby 2004). The coupling of sensing to “actuation” in a single protein leading to the change in occupancy of a promoter has been one of the most useful properties for bioreporter constructions, since the promoter can be used to control expression of the reporter gene. The wide repertoire of regulatory proteins in microorganisms has been exploited intensively for the construction of bioreporters, capitalizing on the native ligands recognized by the individual TRs (van der Meer 2010). Various other protein families carry sensing domains useful for bioreporting purposes but do not directly interact themselves with operator sites in or nearby promoters. Examples include periplasmic binding proteins (PBPs), which have high affinity for specific ligands that they scavenge for the cell, frequently in order to present those to import channels (Berntsson et al. 2010). PBPs can also interact with methyl-accepting chemotaxis proteins (MCPs) (Quiocho and Ledvina 1996), a wide class of proteins that enable bacterial cells to recognize gradients of chemical compounds and either become attracted or repelled (Matilla and Krell 2017). Further sensory proteins operate in the so-called bacterial two-component signaling pathways (Stock et al. 2000). Ligand-binding to the sensory protein results in autophosphorylation, which is transmitted to a response regulator protein that controls gene expression or specific physiological reactions. Sensory proteins of two-component signaling pathways may span across the cytoplasmic membrane or be found within the cytoplasm itself. Finally, antibodies or specific designed ligand-binding proteins have been deployed as sensory proteins for bioreporter developments (Bazin et al. 2017; Chang et al. 2018). Ligand perception by aptamers influences the efficiency of translation of the reporter gene rather than its transcription (Bazin et al. 2017).

Strategies for Obtaining New Sensory Proteins The design of bioreporter bacteria has so far mostly exploited natural compounds and their cognate sensory proteins (TRs, PBPs, MCPs, etc.) (Ivask et al. 2009; Paton et al. 2009; Hynninen and Virta 2010; Tecon et al. 2010; Siegfried et al. 2012; Xu et al. 2014; Yoon et al. 2016). This has yielded a number of very sensitive and potentially applicable bioreporters, but the current weak part in bioreporter deployment is their limited range of detectable compounds, which may not align with the environmental, industrial or medical interest in the analysis (Chang et al. 2018). Characterization of promoter activation as a consequence of ligand exposure by global gene expression analysis may in some cases reveal novel signaling chains or TRs exploitable as sensory elements in bioreporters (Belkin et al. 2017; Shemer et al. 2018). One of the alternatives to obtain sensory proteins that recognize nonnatural ligands is to alter or extend the ligand-binding capacities of existing sensory proteins through mutagenesis. Different strategies have been deployed for this, which usually consist of some form of DNA gene shuffling, or directed

228

D. Tavares et al.

mutagenesis procedure on the full gene coding for the sensory protein, followed by rigorous selection and or screening of promising mutants (Vogne et al. 2010). Alternatively or as a complement to this strategy, computational approaches have been proposed to better predict the amino acid residues or protein areas to target and to limit the mutagenesis effort to restricted or site-specific libraries (Tinberg et al. 2013). In the following section, we will briefly expose in general terms a number of current mutagenesis and selection or screening strategies, after which we will focus more specifically on actual obtained sensory mutants in the TR, PBP, and MCP families, respectively. Most sensory domains of TRs, PBPs, or MCPs are too large to produce completely randomized mutant gene libraries that would encode all possible variants. And if this were possible, the screening of these variants would not be realistically achievable. Practically speaking, random mutant libraries span sizes from 104 (clone libraries) to 1012 (e.g., phage or ribosome display) mutants. Mutants are typically produced by error-prone DNA amplification, sometimes in combination with DNA synthesis strategies, that aim to obtain between on average 1–3 mutations per gene (Galvao and de Lorenzo 2006). These DNAs are subsequently cloned into appropriate vectors and introduced into bacterial cells, or expressed on bacteriophages, or kept as in vitro transcription-translation systems. The next step consists of an appropriate library selection or screening strategy to find potential mutants fulfilling the intended new or altered ligand-binding properties. Screening can be based on the ligand-binding reaction itself, in which case one can, for example, present the ligand and capture mutant protein from a library display (Cirino and Qian 2013). In other cases, ligand binding can be detected indirectly through, e.g., ATPase activity of the TR or phosphorylation of the sensory protein (Ray et al. 2016). Selection can also be based on conditional or exclusive mutant growth in the presence of the new ligand, which has the advantage that in principle only positive genotypes survive (Galvao and de Lorenzo 2006; Choi et al. 2014). The problem in screening new sensing specificities from large libraries and in contrast to, for instance, screening of enzymatic variants using color reactions, is that sensor selection is indirect. The sensory element is coupled to a conditionally lethal output and only if the mutant displays the proper new properties, the cell can grow and multiply (Galvao and de Lorenzo 2006). Several strategies for conditional lethality and counterselection have been developed and have shown some degree of success (Galvao and de Lorenzo 2006). In practice, however, a significant proportion of false positives appears that have overcome the conditionality in a nonintended manner. As a third variant, sensor mutant libraries may be screened rather than selected for proper signaling in the presence of the new ligand. In this case, the sensory element is coupled to the synthesis of, for instance, a reporter protein (Beggah et al. 2008). Cells with outlier reporter protein production are then indicative for mutant protein behavior and can be detected and separated. This procedure has the disadvantage that high-throughput screening strategies are required for large mutant library sizes (Schallmey et al. 2014). On the other hand, there is less chance for selection of off-

11

Engineering of Sensory Proteins with New Ligand-Binding Capacities

229

target false-positive mutations. Particularly autofluorescent protein reporters can be easily measured in high-throughput on individual cells, on in vitro transcription/ translation systems in water–in–oil emulsion droplets, or on cells or microcolonies immobilized in microbeads using flow cytometry (Beggah et al. 2008; Duarte et al. 2017). Potential mutants with outlier fluorescence can be separated using fluorescence activated cell sorting (FACS) and recovered. As the signal increase in positive mutants need not be a priori extremely large, the separation of interesting mutants from the rest of the population remains challenging. Furthermore, single cell screening is prone to significant cell-to-cell variability, which adds to the complexity of identifying bona fide mutants with outlier fluorescence. To overcome these limitations, new strategies have been proposed recently. Van Rossum and coworkers used a conditional resistance to kanamycin in combination with a fluorescent bioreporter system (van Rossum et al. 2017), to reduce the complexity of a large mutant library to a smaller subset based on growth of surviving clones. Others showed how phenotypic variation among individual cells of different mutants can be reduced by screening libraries embedded and grown to microcolonies in alginate beads (Duarte et al. 2017).

Mutagenesis and Selection of Mutant Transcription Activators TRs are proteins that influence the transcription of a gene or a set of genes. TRs that carry ligand-sensing and DNA-binding domains can be used as part of a bioreporter system, by fusing a promoterless reporter gene downstream of the promotor that the TR controls (Fig. 1c). Both TR activators and repressors, single- or two-component systems, have been deployed within reporter gene circuitry (van der Meer 2010). The affinity of the TR towards its specific ligand(s) and to its operator site in the promoter region determine the sensitivity of the obtained bioreporter system (Nielsen et al. 2016; Berset et al. 2017). A number of well-characterized TRs have been used as test cases to attempt to obtain mutants with new ligand-binding properties. Early attempts focused on well-known members of activator and repressor TR families, such as TetR- (Scholz et al. 2003; de los Santos et al. 2016), LuxR-, LysR- (Jha et al. 2015), NtrC-, and AraC-families (Tang and Cirino 2011). Notably NahR and DntR, the LysR-type transcription activators for salicylate, were used for random as well as computation-inferred mutagenesis. Lonneborg and colleagues (Lonneborg et al. 2007) used crystal structure information to predict the binding pocket of DntR for salicylate and propose specific mutations that would result in binding of 2,4-dinitrotoluene. Interestingly, although a number of designed mutants were obtained, this procedure was at that point not extremely effective, apparently because the ligand-binding site in the mutant protein was slightly distorting the protein elsewhere, which influenced its activation process. The NtrC-type sigma54-dependent transcription activators XylR, DmpR, and HbpR, for xylenes, phenol, and 2-hydroxybiphenyl, respectively have been used in directed evolution and DNA shuffling experiments to screen for ligand-binding specificities expanding the original spectrum. Indeed, even with limited library

230

D. Tavares et al.

sizes of up to 105 clones mutant proteins could be retrieved that displayed activation by non-native ligands, such as dinitrotoluene (Garmendia et al. 2001; Galvao et al. 2007; Garmendia et al. 2008), chlorophenols (Wise and Kuske 2000), or 2chlorobiphenyl (Beggah et al. 2008). More recently, the crystal structure of the related NtrC-type protein MopR with its ligand phenol was solved (Ray et al. 2016). This was used to computationally infer amino acid changes in the binding pocket that resulted in binding and activation by methyl- and ethyl-substituted benzenes. Also the structure of LacI with its ligand allolactose was used to predict and build variant libraries for four new sugar ligands, which could subsequently be retrieved through selection and screening approaches (Taylor et al. 2016). These examples have shown that randomized mutagenesis strategies are more likely to show fruitful results when supported by computational predictions of the ligand-binding cavities in known or closely related crystal structures of the TRs. Although most examples have expanded or shifted the range of native ligands to closely related molecules, it shows that computationally inferred mutant production is a valuable method to obtain new sensory capacity for bioreporters. Protein sensory domains can be further shuffled with different other protein domains as blocks to create artificial TRs in biosensing pathways (Feng et al. 2015). It is to be expected that as predictions of structural protein domains improve further, computational-supported designs will gain further in advantage (Fig. 2).

Periplasmic Binding Proteins and Selection Periplasmic binding proteins have a conserved protein structure, named the bilobal structural fold (Chu Byron and Vogel 2011), which consists of two domains connected by a hinge region, with the binding pocket formed between the two domains. Based on the connecting hinges, PBPs have been classified in three major groups (I–III). Class I PBPs have three β-sheets connecting the two lobes, which in Class II are two β-sheets only. Class III PBPs have a single α-helix between the two lobes (Quiocho and Ledvina 1996; Berntsson et al. 2010). In the absence of ligand, PBPs adopt an open conformation, in which the binding site is exposed. Ligand-binding stabilizes the closed form of the protein, with the two lobes approaching each other and burying the ligand within the surrounding protein like a “Venus fly-trap” (Björkman and Mowbray 1998; Li et al. 2013). PBPs facilitate nutrient and trace mineral scavenging for the bacterial cell, by binding the ligand at high affinity and “presenting” the bound ligand to specific transport channels. Binding to the transport channel elicits ATP hydrolysis and compound transport. Some PBPs are additionally involved in chemotactic sensing and also interact in ligand-bound form with MCPs in the cytoplasmic membrane. Galactose-binding protein (GBP) and ribose-binding protein (RBP) of E. coli are two well-known examples involved in chemosensing as well as sugar scavenging. Ligand-bound GBP and RBP interact with the Trg chemoreceptor, which leads to a phosphorylation cascade that biases flagellar movement of the cell (Binnie et al. 1992; Eym et al. 1996).

11

Engineering of Sensory Proteins with New Ligand-Binding Capacities

231

Fig. 2 Strategies for obtaining new ligand-binding properties in sensory proteins. (a) Sensory protein classes: PBP periplasmic binding proteins, MCP methyl-accepting chemotaxis proteins, TR transcription regulators. (b) Mutagenesis strategies to design new sensing elements. (c–e) Screening and selection strategies to find and isolate potential candidate proteins with new or altered ligand-binding properties from mutant libraries. FACS fluorescence-activated cell sorting, PMT photomultiplier, FSC-A forward scatter area

PBPs constitute a widely distributed and evolutionary ancient protein family, members of which bind a large range of natural organic and inorganic ligands. The primary structure homology of PBPs that even bind structurally close ligands is surprisingly rather low. For example, GBP and RBP only share 29% amino acid conservation. Ligand-binding to PBPs has been exploited intensively for the development of biosensors. The configuration change of both protein lobes along the hinge upon ligand-binding can be followed by a variety of biochemical techniques. Most interesting for biosensing purposes is, for example, Föster resonance energy transfer (FRET) (Ye and Schultz 2003; Wu et al. 2012; Ko et al. 2017). For FRET to occur, two different fluorophores are attached to either side of the PBP lobes. The primary fluorophore (the donor) is excited by the illuminating light and its emission light excites the second fluorophore, which is measured. Depending on the type of PBP and the placement of the fluorophores, ligand binding can either diminish or increase the FRET signal (Medintz and Deschamps 2006). FRET on PBPs is commonly engineered by using native Cys-residues or modifying other PBP residues to Cys, which can be coupled in a thiol reaction to the fluorophore on the purified protein. The second fluorophore is frequently an autofluorescent protein (for example, yellow fluorescent protein), which is attached to the N-terminal region of the PBP (Ko et al. 2017). Recently, it was shown how fluorophore-bound amino acids, such as L-(7-hydroxycoumarin-4-yl)ethylglycine, can be incorporated

232

D. Tavares et al.

directly in the protein using the capacity of E. coli engineered amber mutants to incorporate nonnatural amino acids (Ko et al. 2017). PBP ligand binding can also be detected in competitive surface or liposome-based assays, leading to low sensitivities of detection in the nM-μM-range (Edwards et al. 2016). Different groups have demonstrated ligand detection by purified PBPs of, for example, thiamine (Hanes et al. 2011), glucose (Amiss et al. 2007; Helassa et al. 2014; Li et al. 2017), amino acids (Wada et al. 2003; Ko et al. 2017; Paul et al. 2017), phosphate (Solscheid et al. 2015), phosphonate (Rizk et al. 2006), or maltose (Fehr et al. 2002; Iijima and Hohsaka 2009; Park et al. 2009). Whereas most groups have exploited the natural binding properties of the investigated PBPs, very few have actually attempted to change their binding specificity. PBP binding specificity alteration through protein structure computational guidance was heralded in the early 2000s in a number of conceptual publications (Looger et al. 2003; Dwyer and Hellinga 2004). Computational modeling has been used to explain how mutations in the D-glucose/D-galactose-binding protein (GGBP) reduced affinity for glucose by 5000 times, bringing it to physiological human range (mM). The modified PBP may be used as a glucose blood biosensor (Amiss et al. 2007). Mutations improved the specificity of a leucine-binding protein (LBP) towards L-Leu by reducing crossreaction to structurally similar amino acids. Further modification increased the affinity for L-Leu by 14 times and led to moderate recognition of L-Met (Ko et al. 2017). PBPs have been integrated in whole cell living bioreporters thanks to a discovery in the group of Hazelbauer in 1994 (Baumgartner et al. 1994) that ligand-bound RBP and GBP can activate an E. coli hybrid chemoreceptor formed between Trg and EnvZ. This hybrid chemoreceptor (Trz1) consists of the 265 N-terminal amino acids of the Trg chemoreceptor linked to the 230 C-terminal amino acids of the EnvZ histidine kinase of the osmoregulation system. Ligand-bound GBP or RBP interaction to the periplasmic Trg’-domain of Trz1 triggers histidine kinase activity of the cytoplasmic EnvZ-domain, leading to autophosphorylation and subsequent phosphorylation of the cognate response regulator OmpR (Srividhya and Krishnaswamy 2004). Phosphorylated OmpR has increased affinity for the ompC promoter, and ligand-binding to RBP or GBP thus causes an increased transcription rate from this promoter. Coupling a reporter gene to the ompC promoter yields a bioreporter for galactose and ribose with good sensitivity in the low μM (galactose) to nM-range (ribose, 50 nM detection limit) (Reimer et al. 2014). In addition to proposing computational prediction of PBPs with new ligandbinding capacities on purified protein, several engineered PBP variants were included in the E. coli Trz1-OmpR platform. In this case, ligand interaction with the engineered mutant RBP would bind Trz1 and trigger reporter gene expression. One of the publications specifically reported the successful design of mutant proteins based on the native E. coli ribose binding protein (RbsB) yielding detection of nonnatural substrates trinitrotoluene (TNT), lactate or serotonin, down to the nMmM range. The publication claimed a design of a mutant protein (TNT.R3) that would effectively and sensitively recognize trinitrotoluene (TNT) with a constant of affinity (KD) of 2 nM (Looger et al. 2003). Unfortunately, this work was largely questioned and independent studies were unable to reproduce the initial findings

11

Engineering of Sensory Proteins with New Ligand-Binding Capacities

233

(Schreier et al. 2009; Reimer et al. 2014). Further subsequent studies have shown that RbsB becomes very sensitive to misfolding upon mutations that locate both in the ribose binding pocket as well as elsewhere in the protein (Antunes et al. 2011; Reimer et al. 2017). The initial idea of computational redesigned ligand-binding based on predictions of the (wild-type) binding pocket, which works successfully in TRs (see above), may have largely underestimated the necessary conformational characteristic changes in PBPs that determine its activity. Computational predictions of amino acid changes that would drastically alter the PBP’s ligand, and at the same time maintain correct protein stability and allow its intramolecular conformation changes are currently unreliable (Feldmeier and Höcker 2013; Yang and Lai 2017).

Bacterial Bioreporters Based on Methylaccepting Chemotaxis Proteins Methyl-accepting chemotaxis proteins (MCPs) are membrane-spanning proteins that provide sensory information to bacteria about chemical gradients in which they live. This permits the cells to swim towards possible attractants such as sugars or amino acids or retreat from toxic repellents. MCPs have four structural domains, which are called the ligand binding (LB), the transmembrane (TM), the histidine kinases, adenylate cyclases, methyl accepting proteins and phosphatases (HAMP), and the signal transmitter (ST) domains (Szurmant and Ordal 2004). The LB domain is responsible for the specificity of chemical recognition, which can consist of one or a group of closely related chemical compounds. LB domains have been clustered into two main groups (Lacal et al. 2010) with two subgroups each (Upadhyay et al. 2016; Matilla and Krell 2017), although this classification is not sufficient to predict their cognate ligands. The first cluster is comprised of the four helix bundle (4HB) and single CACHE (sCACHE) classes. E. coli MCPs all belong to the 4HB class (Matilla and Krell 2017). In contrast to the 4HB class with its four helical bundles, the sCACHE class also displays a β-sheet structure that interacts with the ligand. The second cluster is comprised of the helical bimodular (HBM) and double CACHE (dCACHE) classes. HBM in essence consists of two coupled 4HB domains, one of which is proximal and the other distal to the cytoplasmic membrane. Both 4HB domains in the HBM can have different ligand specificities. The dCACHE class is comprised of two sCACHE subdomains, but in contrast to HBM, only a single one of these binds the ligand (Matilla and Krell 2017). LB domains can bind ligands directly, like for the Tar-MCP of E. coli (Tajima et al. 2011), or interact with a PBP that is associated with the ligand, as is the case for Trg (Vyas et al. 1991). The membrane-spanning TM domain locates the MCPs to the cytoplasmic membrane. The HAMP domain is commonly seen as a linker to the ST domain (Mondejar et al. 2012). Some MCP-like proteins have several HAMPs, which are referred to as poly-HAMP arrays (Natarajan et al. 2014). The ST domain interacts with the soluble cytoplasmic response regulator (RR) and determines the further downstream reaction of the cell upon perception of the stimulus. Frequently, signal transduction is achieved by protein phosphorylation, either by the ST itself or with help of an associated kinase

234

D. Tavares et al.

(Bilwes et al. 1999). On the basis of extensive homology searches, seven different cytoplasmic ST-domains of MCPs have been recognized, relative to the number of helical bundles (named 24H–44H) (Alexander and Zhulin 2007). Chemotaxis mediated by MCPs is one of the best described two-component signaling pathways. E. coli has four MCPs involved in chemotaxis and one in aerotaxis, which are called Trg, Tar, Tsr and Tap, and Aer (Grebe and Stock 1998). Trg provides taxis towards ribose and galactose, Tar towards aspartate and maltose, Tsr towards serine and Tap towards dipeptides (Grebe and Stock 1998). The RR for MCPs in E. coli is CheY, which is phosphorylated by the ST-associated kinase CheA in presence of an MCP-activating ligand. This complex is stabilized by CheW (Weis et al. 2003). Phosphorylated CheY has increased affinity for FliM, a component of the flagellar motor. CheY~P–FliM binding will bias the rotation of the flagellum from anticlockwise into clockwise direction, leading to more frequent tumbling (Vladimirov and Sourjik 2009; Sampedro et al. 2015). When the cells perceive a gradient of attractant, the MCPs in the ligand-bound form reduce phosphorylation. This will lower the phosphorylation rate of CheY, which as a result of constant dephosphorylation by CheZ, decreases the intracellular CheY~P levels (Wadhams and Armitage 2004). The result is less frequent tumbling and more frequent long directional swims. This on average leads the motile cells to move in the direction of the higher attractant concentration. Repellents lead to the opposite reaction (Vladimirov and Sourjik 2009). MCP-chemotaxis responses are transient and to some extent relative (or insensitive to the actual chemical concentration itself). Longer exposure to higher attractant concentrations leads to a reset of the system through MCP methylation at the ST-domain by CheR (Antommattei et al. 2004). Demethylation is regulated by CheB which is activated by phosphorylation by CheA as CheY (Wadhams and Armitage 2004). The transient reaction of MCPs upon instant change of chemical concentration can be measured by FRET between CheY~P and CheZ and yields a concentration-dependent signal which can be interpreted as a biosensor measurement (Bi et al. 2016). Also the accumulation of motile cells as a result of chemotaxis in a constant chemical gradient can be quantified as a biosensor response (Roggo and van der Meer 2017). MCPs share the HAMP domain with a large range of other two-component signal transduction systems (Aravind and Ponting 1999; Mondejar et al. 2012). This actually allowed the construction of the first functional chimeric transmembrane receptor Trz1, mentioned above (Baumgartner et al. 1994), by fusing the LB domain of the chemoreceptor Trg to the ST-kinase domain of the osmosensor EnvZ. The structure of EnvZ is slightly different from MCPs as it has no LB domain but a cytoplasmic sensory domain sensitive to osmotic stress. Upon osmotic stress, the ST of EnvZ phosphorylates OmpR, which then activates the ompC promoter and expression of the porin OmpC to compensate the turgor pressure (Forst et al. 1988; Forst et al. 1989; Aiba and Mizuno 1990). In the Trz1 hybrid system, OmpR phosphorylation is triggered upon binding of the ribose-bound-RBP to the Trz1-LB domain. As mentioned, by fusing a reporter gene like gfp to the ompC promoter, a very sensitive and quantitative bioreporter for ribose can be obtained (Reimer et al. 2014). Understanding the mode of action of two-component systems was greatly improved by a resolved almost complete structure of the NarQ nitrate sensory

11

Engineering of Sensory Proteins with New Ligand-Binding Capacities

235

histidine kinase (Gushchin et al. 2017). The NarQ N-terminus is located inside the cytosol, after which the protein spans the cytoplasmic membrane in a first TM helix. In the periplasm, the NarQ-LB domain is then formed by four antiparallel helices, which connect to the membrane by a small loop between the second and third helix. A next TM helix connects the periplasmic LB domain to the cytosol via the HAMP domain. The HAMP domain consists of a helix–loop–helix motif with a parallel helix bundle and four rotations (Gushchin et al. 2017). NarQ forms a dimer whereas MCPs form trimers of dimers, which assemble themselves into MCP clusters located at the cell poles (Weis et al. 2003; Eismann and Endres 2015). Comparison of the NarQ-structure with and without its ligand nitrate indicated that binding of nitrate induces a displacement of the LB–helices of up to 1 Å, resulting into a “piston-like” shift in the TM helices. This shift translates into a “scissor-like” motion of the HAMP-domain through the interaction with the Nterminus of the first TM helix that pushes the second helix of the HAMP further into the cytoplasm. This motion displaces the end of the HAMP-domain 7 Å between the bound and unbound state and is assumed to lead to activation of the ST domain (Gushchin et al. 2017). The exploitation of MCPs has a good potential for future classes of biosensors or bioreporter cells. The natural variability of the LB-domains of MCPs deduced from genome information is large, although their cognate ligands are currently not well described and in most cases not predictable from protein homology (Sampedro et al. 2015). Some MCPs have been reported to react to environmentally relevant compounds, such as 2,4-dichlorophenoxyacetic acid (2,4-D; a previously widely applied herbicide) (Hawkins and Harwood 2002), naphthalene (Grimm and Harwood 1999), or toluene (Lacal et al. 2011). The ligand-properties of MCPs can be changed to some extent by site-directed or random mutagenesis. For example, Tar mutants have been described that permit a chemotactic response towards cysteic acid, phenylalanine, or glutamate (Derr et al. 2006). Bi and coworkers used structural prediction of the Tar LB-domain to produce binding to eight noncognate ligands, including a variety of aspartate derivates and phthalic acid (Bi et al. 2013). MCP LB-domains may be connected in a plug-and-play manner to other protein domains, in order to obtain quantitative standardized output. This principle was shown in a recent impressive work where a variety of MCP LB-domains were fused to the Tar cytoplasmic parts at sites close to the HAMP linker (Bi et al. 2016). Activation of the MCP-Tar hybrid by ligand-interaction was then quantified through FRET between CheY~P and CheZ. In this case, an yellow fluorescent protein (YFP) was fused to CheY and a cyan fluorescent protein (CFP) was fused to CheZ. CheY~P binding to CheZ triggers the FRET signal in seconds and is influenced by ligand interactions at the MCP-LB domains (Sourjik and Berg 2002; Bi et al. 2016). Since the response is ephemeral, the cells have to be exposed to instantly changing chemical concentrations in order to observe maximal FRET output, after which the signal within a few minutes gradually returns to baseline (Bi et al. 2016). MCP-LB domains can also be coupled to the cytoplasmic signaling domains of related two-component systems, in which case ligand-interaction with the receptor can be transmitted into de novo gene expression of a reporter gene. This was the concept behind the Trz1 fusion as mentioned earlier (Reimer et al. 2014). Apart

236

D. Tavares et al.

from the Trz1 Trg-EnvZ hybrid, also Tar-EnvZ hybrids (Zhu and Inouye 2003) DcuSEnvZ (Ganesh et al. 2013), and NarX-EnvZ (Lehning et al. 2017) have been constructed. Activation of the EnvZ autokinase activity would then in all cases lead to reporter gene production from the ompC or ompF promoter (Lehning et al. 2017). In contrast to the MCP-LB-domain–Tar fusions measured by FRET, very little response was observed from MCP-EnvZ hybrids (Lehning et al. 2017), except for Trz1. This suggests that obtaining correctly functioning hybrid proteins fused by HAMP linkers is still not trivial, and the rules for hybrid MCP functioning are ill understood. Finally, MCP sensory input for biosensors can be exploited from chemotaxis itself, by measuring the accumulation of motile chemotactic cells in chemical gradients. This exploitation has somewhat been hindered by the inherent difficulty to produce stable gradients, but microfluidic devices have been instrumental in this respect (Si et al. 2012; Wang et al. 2012, 2015). Recent work showed the quantitative and temporal response of E. coli motile and chemotactic cells to serine, aspartate, and methylaspartate in a microfluidic system (Roggo and van der Meer 2017).

Conclusions Sensory proteins are pivotal for the development of whole cell-based bacterial sensors. Past work has successfully exploited the wide variety of sensory protein classes, either directly, or in different plug-and-play manners. But the difficulty for future sensor development is to achieve a more straightforward manner to predict and mutate existing (or de novo) sensory proteins for noncognate ligands. For the case of PBPs, computational predictions of new ligand properties have not been successful (yet), and some of the past claims have proven irreproducible. For the case of TRs, computational-guided predictions have been more successful in obtaining new ligand-binding properties, perhaps because these proteins undergo less drastic conformational changes upon ligand interaction. It can be expected that the accuracy of the systems in free energy scoring function calculations, molecular dynamics involved on protein-ligand interactions, and prediction of conformational changes will improve in the near future by better algorithms.

References Aiba H, Mizuno T (1990) Phosphorylation of a bacterial activator protein, OmpR, by a protein kinase, EnvZ, stimulates the transcription of the ompF and ompC genes in Escherichia coli. FEBS Lett 261:19–22 Alexander RP, Zhulin IB (2007) Evolutionary genomics reveals conserved structural determinants of signaling and adaptation in microbial chemoreceptors. Proc Natl Acad Sci USA 104:2885–2890. https://doi.org/10.1073/pnas.0609359104 Amiss TJ, Sherman DB et al (2007) Engineering and rapid selection of a low-affinity glucose/ galactose-binding protein for a glucose biosensor. Protein Sci 16:2350–2359. https://doi.org/ 10.1110/ps.073119507 Antommattei FM, Munzner JB et al (2004) Ligand-specific activation of Escherichia coli chemoreceptor transmethylation. J Bacteriol 186:7556–7563. https://doi.org/10.1128/JB.186.22.75567563.2004

11

Engineering of Sensory Proteins with New Ligand-Binding Capacities

237

Antunes MS, Morey KJ et al (2011) Programmable ligand detection system in plants through a synthetic signal transduction pathway. PLoS One 6:e16292. https://doi.org/10.1371/journal. pone.0016292 Aravind L, Ponting CP (1999) The cytoplasmic helical linker domain of receptor histidine kinase and methyl-accepting proteins is common to many prokaryotic signalling proteins. FEMS Microbiol Lett 176:111–116 Baumann B, van der Meer JR (2007) Analysis of bioavailable arsenic in rice with whole cell living bioreporter bacteria. J Agric Food Chem 55:2115–2120 Baumgartner JW, Kim C et al (1994) Transmembrane signalling by a hybrid protein: communication from the domain of chemoreceptor Trg that recognizes sugar-binding proteins to the kinase/ phosphatase domain of osmosensor EnvZ. J Bacteriol 176:1157–1163 Bazin I, Tria SA et al (2017) New biorecognition molecules in biosensors for the detection of toxins. Biosens Bioelectron 87:285–298. https://doi.org/10.1016/j.bios.2016.06.083 Beggah S, Vogne C et al (2008) Mutant transcription activator isolation via green fluorescent protein based flow cytometry and cell sorting. Microb Biotechnol 1:68–78. https://doi.org/ 10.1111/j.1751-7915.2007.00008.x Belkin S, Yagur-Kroll S et al (2017) Remote detection of buried landmines using a bacterial sensor. Nat Biotechnol 35:308–310. https://doi.org/10.1038/nbt.3791 Berntsson RPA, Smits SHJ et al (2010) A structural classification of substrate-binding proteins. FEBS Lett 584:2606–2617. https://doi.org/10.1016/j.febslet.2010.04.043 Berset Y, Merulla D et al (2017) Mechanistic modeling of genetic circuits for ArsR arsenic regulation. ACS Synth Biol. https://doi.org/10.1021/acssynbio.6b00364 Bi S, Yu D et al (2013) Discovery of novel chemoeffectors and rational design of Escherichia coli chemoreceptor specificity. Proc Natl Acad Sci USA 110:16814–16819. https://doi.org/10.1073/ pnas.1306811110 Bi S, Pollard AM et al (2016) Engineering hybrid chemotaxis receptors in bacteria. ACS Synth Biol 5:989–1001. https://doi.org/10.1021/acssynbio.6b00053 Bilwes AM, Alex LA et al (1999) Structure of CheA, a signal-transducing histidine kinase. Cell 96:131–141 Binnie RA, Zhang H et al (1992) Functional mapping of the surface of Escherichia coli ribose-binding protein: mutations that affect chemotaxis and transport. Protein Sci 1:1642–1651 Björkman AJ, Mowbray SL (1998) Multiple open forms of ribose-binding protein trace the path of its conformational change. J Mol Biol 279:651–664. https://doi.org/10.1006/jmbi.1998.1785 Browning DF, Busby S (2004) The regulation of bacterial transcription initiation. Nat Rev Microbiol 2:1–9 Chang HJ, Mayonove P et al (2018) A modular receptor platform to expand the sensing repertoire of bacteria. ACS Synth Biol 7:166–175. https://doi.org/10.1021/acssynbio.7b00266 Choi S-L, Rha E et al (2014) Toward a generalized and high-throughput enzyme screening system based on artificial genetic circuits. ACS Synth Biol 3:163–171. https://doi.org/10.1021/ sb400112u Chu Byron CH, Vogel HJ (2011) A structural and functional analysis of type III periplasmic and substrate binding proteins: their role in bacterial siderophore and heme transport. J Biol Chem 392:39 Cirino PC and Qian SA (2013) Protein engineering as an enabling tool for synthetic biology. Synthetic biology: tools and applications, pp 23–42. https://doi.org/10.1016/B978-0-12394430-6.00002-9 Courbet A, Endy D et al (2015) Detection of pathological biomarkers in human clinical samples via amplifying genetic switches and logic gates. Sci Transl Med 7:289ra283. https://doi.org/ 10.1126/scitranslmed.aaa3601 Daunert S, Barrett G et al (2000) Genetically engineered whole-cell sensing systems: coupling biological recognition with reporter genes. Chem Rev 100:2705–2738 de los Santos EL, Meyerowitz JT et al (2016) Engineering transcriptional regulator effector specificity using computational design and in vitro rapid prototyping: developing a vanillin sensor. ACS Synth Biol 5:287–295. https://doi.org/10.1021/acssynbio.5b00090

238

D. Tavares et al.

Derr P, Boder E et al (2006) Changing the specificity of a bacterial chemoreceptor. J Mol Biol 355:923–932. https://doi.org/10.1016/j.jmb.2005.11.025 Duarte JM, Barbier I et al (2017) Bacterial microcolonies in gel beads for high-throughput screening of libraries in synthetic biology. ACS Synth Biol. https://doi.org/10.1021/acssynbio.7b00111 Dwyer MA, Hellinga HW (2004) Periplasmic binding proteins: a versatile superfamily for protein engineering. Curr Opin Struct Biol 14:495–504. https://doi.org/10.1016/j.sbi.2004.07.004 Edwards KA, Seog WJ et al (2016) High-throughput detection of thiamine using periplasmic binding protein-based biorecognition. Anal Chem 88:8248–8256. https://doi.org/10.1021/acs. analchem.6b02092 Eismann S, Endres RG (2015) Protein connectivity in chemotaxis receptor complexes. PLoS Comput Biol 11:e1004650. https://doi.org/10.1371/journal.pcbi.1004650 Eym Y, Park Y et al (1996) Genetically probing the regions of ribose-binding protein involved in permease interaction. Mol Microbiol 21:695–702. https://doi.org/10.1046/j.13652958.1996.261389.x Fehr M, Frommer WB et al (2002) Visualization of maltose uptake in living yeast cells by fluorescent nanosensors. Proc Natl Acad Sci USA 99:9846–9851. https://doi.org/10.1073/ pnas.142089199 Feldmeier K, Höcker B (2013) Computational protein design of ligand binding and catalysis. Curr Opin Chem Biol 17:929–933. https://doi.org/10.1016/j.cbpa.2013.10.002 Feng J, Jester BW et al (2015) A general strategy to construct small molecule biosensors in eukaryotes. elife 4. https://doi.org/10.7554/eLife.10606 Forst S, Delgado J et al (1988) Regulation of ompC and ompF expression in Escherichia coli in the absence of envZ. J Bacteriol 170:5080–5085 Forst S, Delgado J et al (1989) Phosphorylation of OmpR by the osmosensor EnvZ modulates expression of the ompF and ompC genes in Escherichia coli. Proc Natl Acad Sci USA 86:6052–6056 Galvao TC, de Lorenzo V (2006) Transcriptional regulators a la carte: engineering new effector specificities in bacterial regulatory proteins. Curr Opin Biotechnol 17:34–42 Galvao TC, Mencia M et al (2007) Emergence of novel functions in transcriptional regulators by regression to stem protein types. Mol Microbiol 65:907–919 Ganesh I, Ravikumar S et al (2013) Engineered fumarate sensing Escherichia coli based on novel chimeric two-component system. J Biotechnol 168:560–566. https://doi.org/10.1016/j. jbiotec.2013.09.003 Garmendia J, Devos D et al (2001) A la carte transcriptional regulators: unlocking responses of the prokaryotic enhancer-binding protein XylR to non-natural effectors. Mol Microbiol 42:47–59 Garmendia J, de las Heras A et al (2008) Tracing explosives in soil with transcriptional regulators of Pseudomonas putida evolved for responding to nitrotoluenes. Microb Biotechnol 1:236–246 Grebe TW, Stock J (1998) Bacterial chemotaxis: the five sensors of a bacterium. Curr Biol 8:154–157 Grimm AC, Harwood CS (1999) NahY, a catabolic plasmid-encoded receptor required for chemotaxis of Pseudomonas putida to the aromatic hydrocarbon naphthalene. J Bacteriol 181:3310–3316 Gushchin I, Melnikov I et al (2017) Mechanism of transmembrane signaling by sensor histidine kinases. Science 356. https://doi.org/10.1126/science.aah6345 Hanes JW, Chatterjee D et al (2011) Construction of a thiamin sensor from the periplasmic thiamin binding protein. Chem Commun (Camb) 47:2273–2275. https://doi.org/10.1039/c0cc04228e Harms H, Wells MC et al (2006) Whole-cell living biosensors – are they ready for environmental application? Appl Microbiol Biotechnol 70:273–280 Hawkins AC, Harwood CS (2002) Chemotaxis of Ralstonia eutropha JMP134(pJP4) to the Herbicide 2,4-Dichlorophenoxyacetate. Appl Environ Microbiol 68:968–972. https://doi.org/ 10.1128/aem.68.2.968-972.2002

11

Engineering of Sensory Proteins with New Ligand-Binding Capacities

239

Helassa N, Garnett JP et al (2014) A novel fluorescent sensor protein for detecting changes in airway surface liquid glucose concentration. Biochem J 464:213–220. https://doi.org/10.1042/ BJ20141041 Hynninen A, Virta M (2010) In: Belkin S, Gu MB (eds) Whole cell sensing system II: applications. Springer, Berlin/Heidelberg, pp 31–63 Iijima I, Hohsaka T (2009) Position-specific incorporation of fluorescent non-natural amino acids into maltose-binding protein for detection of ligand binding by FRET and fluorescence quenching. Chembiochem 10:999–1006. https://doi.org/10.1002/cbic.200800703 Ivask A, Rõlova T et al (2009) A suite of recombinant luminescent bacterial strains for the quantification of bioavailable heavy metals and toxicity testing. BMC Biotechnol 9:41. https://doi.org/10.1186/1472-6750-9-41 Jha RK, Chakraborti S et al (2015) Rosetta comparative modeling for library design: engineering alternative inducer specificity in a transcription factor. Proteins 83:1327–1340. https://doi.org/ 10.1002/prot.24828 Ko W, Kim S et al (2017) Engineering a periplasmic binding protein for amino acid sensors with improved binding properties. Org Biomol Chem 15:8761–8769. https://doi.org/10.1039/ C7OB02165H Korpela MT, Kurittu JS et al (1998) A recombinant Escherichia coli sensor strain for the detection of tetracyclines. Anal Chem 70:4457–4462 Lacal J, Garcia-Fontana C et al (2010) Sensing of environmental signals: classification of chemoreceptors according to the size of their ligand binding regions. Environ Microbiol 12:2873–2884. https://doi.org/10.1111/j.1462-2920.2010.02325.x Lacal J, Munoz-Martinez F et al (2011) Bacterial chemotaxis towards aromatic hydrocarbons in Pseudomonas. Environ Microbiol 13:1733–1744. https://doi.org/10.1111/j.14622920.2011.02493.x Lehning CE, Heidelberger JB et al (2017) A modular high-throughput in vivo screening platform based on chimeric bacterial receptors. ACS Synth Biol 6:1315–1326. https://doi.org/10.1021/ acssynbio.6b00288 Lewis C, Beggah S et al (2009) Novel use of a whole cell E. coli bioreporter as a urinary exposure biomarker. Environ Sci Technol 43:423–428 Li HY, Cao ZX et al (2013) Analysis of conformational motions and residue fluctuations for Escherichia coli ribose-binding protein revealed with elastic network models. Int J Mol Sci 14:10552–10569. https://doi.org/10.3390/ijms140510552 Li D, Su J et al (2017) Optical surface plasmon resonance sensor modified by mutant glucose/ galactose-binding protein for affinity detection of glucose molecules. Biomed Opt Express 8:5206–5217. https://doi.org/10.1364/BOE.8.005206 Lonneborg R, Smirnova I et al (2007) In vivo and in vitro investigation of transcriptional regulation by DntR. J Mol Biol 372:571–582 Looger LL, Dwyer MA et al (2003) Computational design of receptor and sensor proteins with novel functions. Nature 423:185–190. http://www.nature.com/nature/journal/v423/n6936/ suppinfo/nature01556_S1.html Matilla MA, Krell T (2017) Chemoreceptor-based signal sensing. Curr Opin Biotechnol 45:8–14. https://doi.org/10.1016/j.copbio.2016.11.021 Medintz IL, Deschamps JR (2006) Maltose-binding protein: a versatile platform for prototyping biosensing. Curr Opin Biotechnol 17:17–27. https://doi.org/10.1016/j.copbio.2006.01.002 Merulla D, Buffi N et al (2013) Bioreporters and biosensors for arsenic detection. Biotechnological solutions for a world-wide pollution problem. Curr Opin Biotechnol 24:534–541. https://doi. org/10.1016/j.copbio.2012.09.002 Mondejar LG, Lupas A et al (2012) HAMP domain-mediated signal transduction probed with a mycobacterial adenylyl cyclase as a reporter. J Biol Chem 287:1022–1031. https://doi.org/ 10.1074/jbc.M111.284067

240

D. Tavares et al.

Natarajan J, Schultz A et al (2014) Biochemical characterization of the tandem HAMP domain from Natronomonas pharaonis as an intraprotein signal transducer. FEBS J 281:3218–3227. https:// doi.org/10.1111/febs.12855 Nielsen AA, Der BS et al (2016) Genetic circuit design automation. Science 352:aac7341. https:// doi.org/10.1126/science.aac7341 Park K, Lee LH et al (2009) Detection of conformationally changed MBP using intramolecular FRET. Biochem Biophys Res Commun 388:560–564. https://doi.org/10.1016/j. bbrc.2009.08.049 Paton GI, Reid BJ et al (2009) Application of a luminescence-based biosensor for assessing naphthalene biodegradation in soils from a manufactured gas plant. Environ Pollut 157:1643–1648. https://doi.org/10.1016/j.envpol.2008.12.020 Paul S, Banerjee S et al (2017) Ligand binding specificity of the Escherichia coli periplasmic histidine binding protein, HisJ. Protein Sci 26:268–279. https://doi.org/10.1002/ pro.3079 Quiocho FA, Ledvina PS (1996) Atomic structure and specificity of bacterial periplasmic receptors for active transport and chemotaxis: variation of common themes. Mol Microbiol 20:17–25. https://doi.org/10.1111/j.1365-2958.1996.tb02484.x Ray S, Gunzburg MJ et al (2016) Structural basis of selective aromatic pollutant sensing by the effector binding domain of MopR, an NtrC family transcriptional regulator. ACS Chem Biol 11:2357–2365. https://doi.org/10.1021/acschembio.6b00020 Reimer A, Yagur-Kroll S et al (2014) Escherchia coli ribose binding protein based bioreporters revisited. Sci Rep 4:5626. https://doi.org/10.1038/srep05626 Reimer A, Maffenbeier V et al (2017) Complete alanine scanning of the Escherichia coli RbsB ribose binding protein reveals residues important for chemoreceptor signaling and periplasmic abundance. Sci Rep 7:8245. https://doi.org/10.1038/s41598-017-08035-5 Rizk SS, Cuneo MJ et al (2006) Identification of cognate ligands for the Escherichia coli phnD protein product and engineering of a reagentless fluorescent biosensor for phosphonates. Protein Sci 15:1745–1751. https://doi.org/10.1110/ps.062135206 Roggo C, van der Meer JR (2017) Miniaturized and integrated whole cell living bacterial sensors in field applicable autonomous devices. Curr Opin Biotechnol 45:24–33. https://doi.org/10.1016/j. copbio.2016.11.023 Roggo C, Picioreanu C et al (2018) Quantitative chemical biosensing by bacterial chemotaxis in microfluidic chips. Environ Microbiol 20:241–258. https://doi.org/10.1111/1462-2920.13982 Sampedro I, Parales RE et al (2015) Pseudomonas chemotaxis. FEMS Microbiol Rev 39:17–46. https://doi.org/10.1111/1574-6976.12081 Schallmey M, Frunzke J et al (2014) Looking for the pick of the bunch: high-throughput screening of producing microorganisms with biosensors. Curr Opin Biotechnol 26:148–154. https://doi. org/10.1016/j.copbio.2014.01.005 Scholz O, Kostner M et al (2003) Teaching TetR to recognize a new inducer. J Mol Biol 329:217–227 Schreier B, Stumpp C et al (2009) Computational design of ligand binding is not a solved problem. Proc Natl Acad Sci 106:18491–18496. https://doi.org/10.1073/pnas.0907950106 Selifonova O, Burlage R et al (1993) Bioluminescent sensors for detection of bioavailable Hg(II) in the environment. Appl Environ Microbiol 59:3083–3090 Shemer B, Yagur-Kroll S et al (2018) Aerobic transformation of 2,4-Dinitrotoluene by Escherichia coli and its implications for the detection of trace explosives. Appl Environ Microbiol 84. https://doi.org/10.1128/AEM.01729-17 Si G, Yang W et al (2012) A parallel diffusion-based microfluidic device for bacterial chemotaxis analysis. Lab Chip 12:1389–1394. https://doi.org/10.1039/c2lc21219f Siegfried K, Endes C et al (2012) Field testing of arsenic in groundwater samples of Bangladesh using a test kit based on lyophilized bioreporter bacteria. Environ Sci Technol 46:3281–3287. https://doi.org/10.1021/es203511k Solscheid C, Kunzelmann S et al (2015) Development of a Reagentless biosensor for inorganic phosphate, applicable over a wide concentration range. Biochemistry 54:5054–5062. https://doi. org/10.1021/acs.biochem.5b00449

11

Engineering of Sensory Proteins with New Ligand-Binding Capacities

241

Sourjik V, Berg HC (2002) Binding of the Escherichia coli response regulator CheY to its target measured in vivo by fluorescence resonance energy transfer. Proc Natl Acad Sci USA 99:12669–12674. https://doi.org/10.1073/pnas.192463199 Srividhya KV, Krishnaswamy S (2004) A simulation model of Escherichia coli osmoregulatory switch using E-CELL system. BMC Microbiol 4:44–44. https://doi.org/10.1186/1471-2180-4-44 Sticher P, Jaspers M et al (1997) Development and characterization of a whole cell bioluminescent sensor for bioavailable middle-chain alkanes in contaminated groundwater samples. Appl Environ Microbiol 63:4053–4060 Stock AM, Robinson VL et al (2000) Two-component signal transduction. Annu Rev Biochem 69:183–215 Szurmant H, Ordal GW (2004) Diversity in chemotaxis mechanisms among the bacteria and archaea. Microbiol Mol Biol Rev 68:301–319. https://doi.org/10.1128/MMBR.68.2.301319.2004 Tajima H, Imada K et al (2011) Ligand specificity determined by differentially arranged common ligand-binding residues in bacterial amino acid chemoreceptors Tsr and tar. J Biol Chem 286:42200–42210. https://doi.org/10.1074/jbc.M111.221887 Tang SY, Cirino PC (2011) Design and application of a mevalonate-responsive regulatory protein. Angew Chem Int Ed Engl 50:1084–1086. https://doi.org/10.1002/anie.201006083 Taylor ND, Garruss AS et al (2016) Engineering an allosteric transcription factor to respond to new ligands. Nat Methods 13:177–183. https://doi.org/10.1038/nmeth.3696 Tecon R, Beggah S et al (2010) Development of a multistrain bacterial bioreporter platform for the monitoring of hydrocarbon contaminants in marine environments. Environ Sci Technol 144:1049–1055 Tinberg CE, Khare SD et al (2013) Computational design of ligand-binding proteins with high affinity and selectivity. Nature 501:212–216 Trang PT, Berg M et al (2005) Bacterial bioassay for rapid and accurate analysis of arsenic in highly variable groundwater samples. Environ Sci Technol 39:7625–7630 Tropel D, van der Meer JR (2004) Bacterial transcriptional regulators for degradation pathways of aromatic compounds. Microbiol Mol Biol Rev 68:474–500 Turner K, Xu S et al (2007) Hydroxylated polychlorinated biphenyl detection based on a genetically engineered bioluminescent whole-cell sensing system. Anal Chem 79:5740–5745 Upadhyay AA, Fleetwood AD et al (2016) Cache domains that are homologous to, but different from PAS domains comprise the largest superfamily of extracellular sensors in prokaryotes. PLoS Comput Biol 12:e1004862. https://doi.org/10.1371/journal.pcbi.1004862 van der Meer JR (2006) Analytics with engineered bacterial bioreporter strains and systems. Curr Opin Biotechnol 17:1–3 van der Meer JR (2010) Bacterial sensors: synthetic design and application principles. Morgan & Claypool, San Rafael van der Meer JR, Belkin S (2010) Where microbiology meets microengineering: design and applications of reporter bacteria. Nat Rev Microbiol 8:511–522 van der Meer JR, Tropel D et al (2004) Illuminating the detection chain of bacterial bioreporters. Environ Microbiol 6:1005–1020 van Rossum T, Muras A et al (2017) A growth- and bioluminescence-based bioreporter for the in vivo detection of novel biocatalysts. Microb Biotechnol 10:625–641. https://doi.org/10.1111/ 1751-7915.12612 Vladimirov N, Sourjik V (2009) Chemotaxis: how bacteria use memory. Biol Chem 390:1097–1104. https://doi.org/10.1515/BC.2009.130 Vogne C, Beggah S et al (2010) In: Timmis KN (ed) Handbook of hydrocarbon and lipid microbiology, vol 5. Springer, Berlin/Heidelberg, pp 4429–4446 Vyas NK, Vyas MN et al (1991) Comparison of the periplasmic receptors for L-arabinose, Dglucose/D-galactose, and D-ribose. Structural and functional similarity. J Biol Chem 266:5226–5237 Wada A, Mie M et al (2003) Design and construction of glutamine binding proteins with a selfadhering capability to unmodified hydrophobic surfaces as reagentless fluorescence sensing devices. J Am Chem Soc 125:16228–16234. https://doi.org/10.1021/ja036459l

242

D. Tavares et al.

Wadhams GH, Armitage JP (2004) Making sense of it all: bacterial chemotaxis. Nat Rev Mol Cell Biol 5:1024–1037. https://doi.org/10.1038/nrm1524 Wang X, Long T et al (2012) Bacterial chemotaxis toward a NAPL source within a pore-scale microfluidic chamber. Biotechnol Bioeng 109:1622–1628. https://doi.org/10.1002/bit.24437 Wang X, Atencia J et al (2015) Quantitative analysis of chemotaxis towards toluene by Pseudomonas putida in a convection-free microfluidic device. Biotechnol Bioeng 112:896–904. https:// doi.org/10.1002/bit.25497 Way JC, Collins JJ et al (2014) Integrating biological redesign: where synthetic biology came from and where it needs to go. Cell 157:151–161. https://doi.org/10.1016/j.cell.2014.02.039 Weis RM, Hirai T et al (2003) Electron microscopic analysis of membrane assemblies formed by the bacterial chemotaxis receptor Tsr. J Bacteriol 185:3636–3643. https://doi.org/10.1128/ jb.185.12.3636-3643.2003 Willardson BM, Wilkins JF et al (1998) Development and testing of a bacterial biosensor for toluene-based environmental contaminants. Appl Environ Microbiol 64:1006–1012 Wise AA, Kuske CR (2000) Generation of novel bacterial regulatory proteins that detect priority pollutant phenols. Appl Environ Microbiol 66:163–169 Wu B, Wang Z et al (2012) Catalyst-free and site-specific one-pot dual labeling of a protein directed by two genetically incorporated noncanonical amino acids. Chembiochem 13:1405–1408. https://doi.org/10.1002/cbic.201200281 Xu T, Close DM et al (2013) Genetically modified whole-cell bioreporters for environmental assessment. Ecol Indic 28:125–141. https://doi.org/10.1016/j.ecolind.2012.01.020 Xu T, Close D et al (2014) Detection of organic compounds with whole-cell bioluminescent bioassays. Adv Biochem Eng Biotechnol 144:111–151. https://doi.org/10.1007/978-3-66243385-0_4 Yagur-Kroll S, Lalush C et al (2014) Escherichia coli bioreporters for the detection of 2,4dinitrotoluene and 2,4,6-trinitrotoluene. Appl Microbiol Biotechnol 98:885–895. https://doi. org/10.1007/s00253-013-4888-8 Yang W, Lai L (2017) Computational design of ligand-binding proteins. Curr Opin Struct Biol 45:67–73. https://doi.org/10.1016/j.sbi.2016.11.021 Ye K, Schultz JS (2003) Genetic engineering of an allosterically based glucose indicator protein for continuous glucose monitoring by fluorescence resonance energy transfer. Anal Chem 75:3451–3459. https://doi.org/10.1021/ac034022q Yoon Y, Kim S et al (2016) Use of tunable whole-cell bioreporters to assess bioavailable cadmium and remediation performance in soils. PLoS One 11:e0154506. https://doi.org/10.1371/journal. pone.0154506 Zhu Y, Inouye M (2003) Analysis of the role of the EnvZ linker region in signal transduction using a chimeric Tar/EnvZ receptor protein, Tez1. J Biol Chem 278:22812–22819. https://doi.org/ 10.1074/jbc.M300916200

Cell-Free Biosensors: Synthetic Biology Without Borders

12

Aidan Tinafar, Yu Zhou, Fan Hong, Kirstie L. Swingle, Anli A. Tang, Alexander A. Green, and Keith Pardee

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Accessibility of Biosensor Infrastructure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Storage and Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Companion Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Rapid Prototyping and Sensor Programmability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Definition of Cell-Free Biosensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Isothermal Amplification-Based Diagnostics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Nucleic Acid Sequence-Based Amplification (NASBA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Loop-Mediated Isothermal Amplification (LAMP) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Recombinase Polymerase Amplification (RPA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Drawbacks of Using Isothermal Amplification Strategies for Sensing . . . . . . . . . . . . . . . . . . . . . CFPE-Based Biosensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Recognition Components in Biosensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Riboregulators, Riboswitches, and Fluorogenic Aptamers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CRISPR-Enabled Nucleic Acid Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Functional DNA Molecules for Biosensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusion and Future Directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

244 245 246 247 248 248 249 251 251 252 253 255 255 258 258 262 265 273 273

Abstract

Cell-free biosensors can take many forms and can range in complexity from single enzymes to engineered systems of biological components that support A. Tinafar · K. Pardee (*) Leslie Dan Faculty of Pharmacy, University of Toronto, Toronto, ON, Canada e-mail: [email protected]; [email protected] Y. Zhou · F. Hong · K. L. Swingle · A. A. Tang · A. A. Green (*) Biodesign Center for Molecular Design and Biomimetics, The Biodesign Institute and the School of Molecular Sciences, Arizona State University, Tempe, AZ, USA e-mail: [email protected]; [email protected]; [email protected]; [email protected]; [email protected] © Springer Nature Switzerland AG 2022 G. Thouand (ed.), Handbook of Cell Biosensors, https://doi.org/10.1007/978-3-030-23217-7_130

243

244

A. Tinafar et al.

synthetic biology applications. This chapter will review the many recent innovations from this latter category and will explore how these more complex systems create synthetic networks to provide biosensors with signal amplification, programmability, high sensitivity, and even tolerance for analyte variation. In particular, cell-free biosensors that operate using isothermal amplification, coupled transcription and translation systems, and CRISPR-related mechanisms will be highlighted. Such DNA-/RNA-based technologies are an especially exciting category for cell-free biosensing, and here this rapidly evolving class of sensors, including toehold switch- and CRISPR-based systems, will be reviewed. Cellfree biosensors are also increasingly designed with companion hardware, and, in doing so, researchers are embedding the capacity for these otherwise laboratorybased reactions to be deployed in real-world applications. Among many innovations, this chapter will highlight how freeze-dried and paper-based systems, low-cost optical readers, and lateral flow devices are helping extend the reach of cell-free biosensors into new environments and applications. Taken together, the use of cell-free synthetic biology and engineered biochemical systems is an exciting category of biosensing and is on track to make significant contributions toward decentralizing the capacity for sensing. Keywords

Cell-free biosensors · Synthetic biology · Toehold switch · Isothermal amplification · CRISPR · Paper-based · Freeze-dried

Introduction Synthetic biology is an interdisciplinary field that brings together biology and engineering to create new biological functions. These efforts have resulted in cellbased biosensors (Kotula et al. 2014; Riglar et al. 2017; Kobayashi et al. 2004), bioproduction (Fossati et al. 2014; Torella et al. 2013; Zhang et al. 2012a), and even programmable logic (Green et al. 2017). Recent work has expanded these efforts to include in vitro applications that use the enzymes of transcription and translation in cell-free formats to deploy these genetically encoded tools beyond the laboratory. These advances are giving rise to a branch of synthetic biology devoted to in vitro devices that can deliver biotechnology in an abiotic and sterile format. As will be discussed, researchers from across the community are using this newfound venue to build synthetic biology-based cell-free biosensors (Pardee et al. 2014, 2016; Wen et al. 2017; Salehi et al. 2017, 2018; Didovyk et al. 2017; Duyen et al. 2016; Gootenberg et al. 2017). This new frontier in biosensing offers some exciting and unique features. Chief among these, when compared to cell-based biosensors, is that target analytes do not need to traverse a cell wall or membrane to reach the sensor apparatus. In a cell-free format, prepared samples of interest are placed directly into the cell-free reactions that host the molecular sensor. This has the advantage of ensuring the sensor is exposed to the full concentration of the analyte

12

Cell-Free Biosensors: Synthetic Biology Without Borders

245

and, importantly, introduces the possibility of detecting large biomolecular analytes like nucleic acids and proteins. In proof-of-concept work, this capacity for large biomolecules was used to create cell-free portable sequence-specific sensors for nucleic acids. Taking advantage of a pathogen’s genome as a barcode, sequences from the Sudan and Zaire strains of the Ebola virus were discriminated and identified. Additionally, sensors for other clinically relevant features, such as bacterial antibiotic resistance genes, were developed (Pardee et al. 2014). In the time since, efforts across the community have greatly extended these proof-of-concept capabilities to a range of analytes at clinically relevant concentrations (e.g., endocrine disruptors and quorum molecules) (Pardee et al. 2014; Wen et al. 2017; Salehi et al. 2017, 2018). This chapter will provide a brief overview of the most exciting aspects of the cell-free biosensing format and an in-depth review of key domains.

Background As with the cell-based biosensors described in the previous chapters, infrastructure is required to operate sensors in the cell-free format. For whole-cell biosensors, this infrastructure includes a cellular host, either prokaryotic or eukaryotic, a growth medium, and often some hardware to house the cells and quantify sensor output. The infrastructure for cell-free biosensors includes some of the same elements but with a key distinction: they can be made biosafe through filter sterilization. This feature imparts two key advantages. Firstly, the cell-free platform poses no risk for contamination in clinical settings or the food industry. Secondly, these systems do not experience the practical constraints associated with deploying cells out of the lab, such as maintaining viability and, importantly, containment of genetically modified organisms. The latter will likely make the regulatory path to approval, for deployment to clinical settings, easier for cell-free biosensors than their cell-based counterparts. Simple cell-free sensors can be created from a small number of purified enzymes, while others use complex networks of enzymes that recapitulate cellular transcription, translation, and ATP generation. These more complex systems, based on cellfree protein expression (CFPE) systems, are enabling new cell-free synthetic biology biosensing applications with increasing sophistication. CFPE systems have their origin in the 1940s and 1950s with simple cellular extract-based systems (Zamecnik and Frantz 1948; Gale and Folkes 1954) and were first brought to the fore in the 1960s when Escherichia coli cell-free extracts were used to discover the triplet nucleotide usage of codons in translation (Matthaei et al. 1962; Matthaei and Nirenberg 1961; Nirenberg and Matthaei 1961). In the decades since, these systems have undergone progressive improvements. Milestones include the coupling of transcription and translation, which allowed cell-free expression directly from DNA templates rather than from supplied mRNA (DeVries and Zubay 1967; Chen and Zubay 1983; Zubay 1973), and the collective work of many has resulted in the productive, high-yielding coupled transcription and translation CFPE reactions that

246

A. Tinafar et al.

are available today (Kigawa et al. 2004; Yang et al. 2012; Jewett et al. 2008; Jewett and Swartz 2004; Kim et al. 2006; Liu et al. 2005; Zawada and Swartz 2006). A variety of systems have been designed to carry out coupled transcription/ translation reactions outside of a cell. These systems generally fall under two categories: traditional extract-based CFPE systems (Schweet et al. 1958; Allen and Schwemt 1962; Rosenblum and Cooperman 2014) and the more recently developed reconstituted CFPE systems (Tuckey et al. 2014; Shimizu et al. 2001). Traditional extract-based systems often use cell lysates supplemented with additional tRNAs, amino acids, and compounds required for energy regeneration. In recent years, these systems have been optimized for reduced cost, enhanced productivity, rapid preparation, and uniform yields (Didovyk et al. 2017; Kwon and Jewett 2015; Sun et al. 2013). Lysates can be extracted from a wide range of cell types including bacterial (Didovyk et al. 2017; Kigawa et al. 2004; Jewett et al. 2008; Sun et al. 2013; Li et al. 2017a, 2018; Wang et al. 2018; Failmezger et al. 2018; Shin and Noireaux 2012; Caschera and Noireaux 2014; Wiegand et al. 2018; Kelwick et al. 2016; Moore et al. 2017, 2018), protozoan (Kovtun et al. 2010, 2011), plant (Harbers 2014; Buntru et al. 2014; Buntru et al. 2015), insect (Ezure et al. 2006), mammalian (Jackson and Hunt 1983; Mikami et al. 2006; Brödel et al. 2014; Martin et al. 2017; Tran et al. 2018; Burgenson et al. 2018), and fungal (Choudhury et al. 2014) sources. Reconstituted CFPE systems work by modular [re]construction of transcription and translation machinery from well-defined purified proteins – as opposed to cell lysates (Tuckey et al. 2014). A widely used example of this technology is the PURE system developed by Shimizu et al. (2001). While reconstituted systems are more costly than their extract-based counterparts, they have the advantage of containing a defined biochemical makeup and low background for sensor operation, which can provide a higher signal-to-noise ratio. Much of the cost associated with reconstituted systems comes from the labor-intensive nature of assembling such multi-protein systems, which involve culturing and purification steps for each constituent protein. To simplify this process, microbial consortia have been engineered to allow for the co-expression and co-purification of multiple proteins (Villarreal et al. 2018). More recently, Lavickova et al. have published a method where all non-ribosomal PURE proteins are prepared in a single co-culture and purification step. Here, they have reported productivity levels rivaling that of commercial PURE systems, leading to a 14-fold improvement in cost (Lavickova and Maerkl 2019). Developments such as these offer the potential to reduce the cost of reconstituted systems, thus making them more widely available for biosensing applications.

Accessibility of Biosensor Infrastructure When looking at biosensors across all categories, cell-based systems have the benefit of being able to self-replicate, a feature that could be important in applications where cost and distribution are key factors. Cell growth media are generally inexpensive and are comprised of easily sourced components. Similarly, the required basic culturing equipment (e.g., shaker, centrifuge) is also available in many labs. As a

12

Cell-Free Biosensors: Synthetic Biology Without Borders

247

result, once cell-based biosensors have been distributed, it should be – in theory – possible to easily maintain decentralized sensing capacity. This same argument can increasingly be applied to synthetic-biology-based cellfree biosensors. As with whole-cell biosensors, these CFPE-based sensors are also genetically encoded and essentially require the same biochemical components. Accordingly, it seems reasonable that the sustainable deployment of cell-free biosensors could also be made possible through decentralized production of CFPE systems. While this is not something that could be said about earlier CFPE protocols, which were labor-intensive and required costly infrastructure (Kigawa et al. 2004; Kim et al. 2006; Liu et al. 2005), these challenges have progressively been overcome, and the technology is now at a point where the preparation of CFPE systems essentially requires the same simple and widely available reagents and equipment (e.g., shaker, centrifuge) used to replicate cell-based biosensors. One such protocol, described by Kwon et al. (2015), uses sonication for the rapid production of CFPE reactions from cultures as small as 10 mL and as large as 10 L, using equipment common to most bioresearch settings (Kwon and Jewett 2015). This trend toward the scalable and accessible production of CFPE systems was extended further with the development of an autolysis protocol (Didovyk et al. 2017). Here the cultured E. coli used to make CFPE systems carry the phage lambda gene R, which is a lysin that degrades bacterial cell walls. The cells, however, remain intact until they are exposed to a freeze-thaw cycle, which breaches the inner bacterial membrane to trigger the equipment-free lysis of cells. Thus, this method simply requires the same equipment needed to propagate whole-cell biosensors, making sustained deployment of sensing possible through local production.

Storage and Distribution With greater accessibility to cell-free protein expression, members of the synthetic biology community began to experiment with designing and operating gene circuits outside of the cell (Sun et al. 2013; Shin and Noireaux 2012; Garamella et al. 2016; Takahashi et al. 2015). This work created a venue for dedicated cell-free synthetic biology applications but also a mode for rapid prototyping of biosensors and circuits for cell-based applications. While interesting in their own right, these applications have been restricted to the laboratory because of the need for storage of cell-free systems at
80  C. Protein stability, and perhaps the labile nature of some of the reagents, means that these biochemical systems require a constant cold chain, which makes deployment of cell-free synthetic-biology-based sensors impractical. With this challenge in mind, the cell-free community has worked to stabilize cellfree systems through lyophilization (Pardee et al. 2014; Smith et al. 2014). While it may not be intuitive that mixtures containing 35 to over 100 enzymes (Jewett and Swartz 2004; Liu et al. 2005; Shimizu et al. 2001) could be successfully reconstituted from a freeze-dried state, lyophilization of CFPE works remarkably well and enables storage and distribution of cell-free biosensors at room temperature. In one demonstration of this approach, it was found that freeze-dried cell-free

248

A. Tinafar et al.

(FD-CF) systems remained active for at least a year (Pardee et al. 2014), and, in the time since, stability of cell-free systems has been reported for over 2 years at room temperature (Chandler et al. 2015). CFPE has been further ruggedized with the use of cryoprotectants. Specifically, using the disaccharide trehalose, CFPE can be dried under ambient conditions and is tolerant of both elevated temperature (37  C) and ambient humidity for months (Karig et al. 2017).

Companion Devices As with any biosensor, the sensing unit in cell-free systems must be housed in way that allows for practical implementation by users. For both cell-based and cell-free biosensors, this frequently involves handling of liquids. Faced with this problem, the cell-free research community has developed a series of solutions that enable handling of biosensors in the field and assist with reading or quantifying their output. An early example of this was the development of paper-based synthetic gene networks. Here, inspired by tools like pH paper and chemistry-based paper tests, FD-CF systems were embedded into paper and other porous materials to create portable, field-ready tests (Pardee et al. 2014). In doing so, a basic in vitro device that can be handled outside of the lab was created. The cellular enzymes and ribosomes, which power gene circuit-based sensors, are embedded in the cellulose matrix and, upon activation, create paper-based fluorescent and colorimetric outputs. Similar solutions have been developed for biosensors across the spectrum of analytes, including nucleic acids, small molecules, and proteins (Cunningham et al. 2016; Hu et al. 2014). Such examples encompass paper-based and lateral-flow tests based on isothermal amplification and CRISPR-guided diagnostics (Pardee et al. 2016; Gootenberg et al. 2017; Seok et al. 2017; Rodriguez et al. 2015; Lafleur et al. 2016; Hongwarittorrn et al. 2017). This increasing portability of biosensors has been accompanied by the development of optical devices for detection and quantification (Pardee et al. 2014, 2016; Yamanaka et al. 2018), which have the potential to ultimately enable automated, portable biosensing.

Rapid Prototyping and Sensor Programmability Having discussed durability and practical means of deployment, this chapter will now turn to the development of specific molecular sensors. As mentioned above, the cell-free environment allows for rapid design-test cycles of sensors and their companion features. Unlike with cell-based systems, the genetically encoded instructions for sensing can be simply added to cell-free reactions; accordingly, cellular transformation or management of selection markers is not required. These DNA components can also be added directly as linear PCR products or as circular plasmids at precise concentrations, which allows for rapid screening of reaction parameters. Similarly, the biochemistry of cellular extracts can be easily augmented with specialized components, like repressor proteins, to ensure tight

12

Cell-Free Biosensors: Synthetic Biology Without Borders

249

transcriptional regulation upon rehydration, or substrates for reporter enzymes (e.g., CPRG). As will be discussed in detail, nucleic-acid-based biosensors, such as toehold switches, have the advantage that they can be created rationally with computer-aided design. This means that sensors can go from the identification of a need to design and prototyping in rapid succession. Previous work demonstrated a workflow of less than 6 weeks (Pardee et al. 2016), which has now been shortened to as little as 1 week. This programmable nature, common to many nucleic acid-based sensors, is an ideal complement to the capacity for rapid prototyping of the cell-free format, and these combined features mean that biosensors can be developed rapidly and at low cost. Clear opportunities for such biosensors include diagnostics for global health, neglected diseases, and emergency response (e.g., to pandemics), among many other applications outside of health, such as agriculture, environmental monitoring, and national security.

Definition of Cell-Free Biosensing Biosensors are molecular devices based on biological parts that recognize target analytes and generate a measurable output in response. At the mechanistic level, the output signal is produced by conformational changes induced by the recognition event. Early and widely recognized examples of biosensing can be traced back to the use of animals, such as canaries in coal mines for monitoring carbon monoxide levels or detection of illicit materials by canines (Slomovic et al. 2015). More recently single-cell organisms have translated this concept to the level of the cell, first as wild type and then as engineered sensor platforms. Earlier renditions of whole-cell biosensors took advantage of the natural response of wild-type cells to changes in specific analytes. In one example, the presence of herbicide was sensed by monitoring the electrical current produced by the photosynthetic activity of wild-type cyanobacteria cells mounted on an electrode. As herbicide concentrations increased, photosynthetic activity was inhibited, and the resulting electrochemical changes at the whole-cell level could be monitored (Rawson et al. 1989). Not surprisingly, such wild type-based systems had the disadvantage of being susceptible to interference from culturing conditions and often suffered from poor selectivity. Over time, with the advent of synthetic biology, cells could be engineered so as to contain poised analyte recognition components paired with tailored reporters. By allowing for rational design, synthetic biology ushered in new sensing strategies unconstrained by naturally evolved systems. These sensors could be specifically tuned for targeted molecular recognition rather than solely relying on the natural host’s general physiology. Some early synthetic biology-based designs took advantage of bioluminescence genes as reporters. In one instance, a promoter from the naphthalene degradation operon in Pseudomonas fluorescens was fused with the Vibrio fischeri luxCDABE bioluminescence cassette to sense the presence of naphthalene and salicylate (Burlage et al. 1990). Since that time, many whole-cell biosensors have been created to detect a wide range of

250

A. Tinafar et al.

analytes including hydrocarbons, specific ions, antibiotics, and nucleic acids (Slomovic et al. 2015; van der Meer and Belkin 2010; Trang et al. 2005; Bhadra and Ellington 2014; Isaacs et al. 2004; Green et al. 2014). Whole-cell biosensors often impose certain limitations, which stem from their dependence on cell viability, cell cycle, and membrane integrity. In contrast, cell-free biosensors can take advantage of biological parts without being limited by such shortcomings. Accordingly, many detection strategies have relied on using enzymes, nucleic acids, or antibodies outside of the cell (Slomovic et al. 2015). Some of the advantages and disadvantages of these cell-free systems are outlined in Table 1. Much of the modern conception of cell-free biosensing has been attributed to Clark et al., who proposed in the early 1960s that enzymes could be immobilized on the surface of electrochemical detectors to expand the analyte range for sensing (Ferrari 2006; Clark and Lyons 2006; Turner et al. 1987). Since then, a wealth of work has ensued to encompass a wide range of sensing components, analytes, and output systems (Slomovic et al. 2015; Turner et al. 1987; Mullis et al. 1986; Karig 2017). In the cell-free context, biosensors have been built from a diverse array of cellular components, including organelles, membranes, enzymes or enzyme components, receptors or binding domains, antibodies, nucleic acids, or organic molecules. The engineered outputs from cell-free biosensors are equally diverse. They may be optical, potentiometric, amperometric, conductimetric, impedimetric, calorimetric, acoustic, or mechanical (Turner et al. 1987; Banica 2012; Mehrotra 2016). Below, a subset of cell-free systems that have garnered considerable attention in recent years due to the advancements in synthetic biology will be discussed. The conversion of input information to output signals is called transduction. While much of the literature on biosensing has revolved around mechanisms of transduction, the focus here will be on the sensors’ biological and biomimetic components – particularly the network effects attained through their synergistic interactions. To illustrate these features, this chapter will focus on a few examples.

Table 1 Advantages and disadvantages of cell-free biosensing Advantages Cell-free sensors are not constrained by cell viability, cell cycle, or membrane integrity. Due to their open nature, there are virtually unlimited reaction formats available for sensing Physicochemical environment can be controlled allowing for sensing under a variety of reaction conditions They generally provide greater speed and flexibility compared to cell-based systems (Catherine et al. 2013) Cell-free sensors can exist in a biosafe format that is sterile and abiotic, allowing for unrestricted use in the clinic, food industry, or environmental monitoring

Disadvantages • Although capable of being augmented with membranes under certain conditions (Fenz et al. 2014), general lack of membranes makes cell-free systems susceptible to varying pH, salt concentrations, and contaminants • This general absence of membranes also makes it difficult to take advantage of surface receptors for sensing • The per-reaction costs tend to be higher when compared to cell-based systems

12

Cell-Free Biosensors: Synthetic Biology Without Borders

251

These will include isothermal nucleic acid amplification schemes, gene networks in cell-free transcription/translation CFPE systems, CRISPR-based diagnostic strategies, and protein-free nucleic acid-based tools. Cell-free sensing technology’s ability to make use of powerful biological components, without the constraints of cell-based systems, has made it an important domain for real-world applications. An incredibly successful example of this is the use of polymerase chain reaction (PCR), which provides robust and sensitive detection of nucleic acids (Mullis et al. 1986; Zhao et al. 2015; Mullis and Faloona 1987). However, despite its many advantages, PCR requires a thermal cycler with precise control over temperature for the repeated denaturation, annealing, and extension cycles of amplification, as well as technical expertise for protocol design and optimization. The combination of these factors has restricted its use primarily to laboratories. Thus, as molecular diagnostics become ever more important in medicine, there has been a significant effort to create portable biosensors that can deliver the advantages of PCR in a decentralized manner. Isothermal amplification techniques were the first to distribute the capacity for molecular detection. As the name suggests, these reactions can be carried out at a constant temperature without the need for the specialized equipment required for thermal cycling. Some isothermal amplification methods can be performed at room temperature (Alladin-Mustan et al. 2015). For many others, rudimentary heating strategies such as water baths (Pham et al. 2005), simple heating blocks (Chen and Ching 2017), chemical heaters (Curtis et al. 2012), or even body heat (Crannell et al. 2014) have successfully been used to achieve amplification. Not only can isothermal amplification reactions be used as stand-alone cell-free biosensors, they can also provide sequence-specific amplification for a second tier of sequence-specific biosensors that run on CFPE or CRISPR proteins. This combined approach can provide greater sensitivity and can reduce false-positive rates, which is a recognized challenge of isothermal amplification (Cordray and Richards-Kortum 2012).

Isothermal Amplification-Based Diagnostics Nucleic Acid Sequence-Based Amplification (NASBA) NASBA was created in 1990 as a way of amplifying single-stranded RNA (ssRNA) in a process that mimics retroviral RNA replication (Guatelli et al. 1990). NASBA’s sensitivity is comparable to that of RT-PCR (Burchill et al. 2002), and it can achieve billion-fold RNA amplification in under 2 h at 41  C. In addition to two primers, the system uses three enzymes: a reverse transcriptase, an RNase H, and a DNA-dependent RNA polymerase (Fig. 1). An initial heating step at 65  C is occasionally performed to deal with the secondary structure of the target RNA (Guatelli et al. 1990), but this may not always necessary (Pardee et al. 2016). The reaction begins with the hybridization of the reverse primer and the target RNA. The reverse transcriptase forms the complementary DNA using the target RNA as template. RNase H degrades the RNA template

252

A. Tinafar et al.

(5)

primer Reverse Transcription

(1)

(4) RNase H Degradation

RNA Transcription

(2)

primer (3)

dsDNA Synthesis

Fig. 1 Steps involved in the NASBA mechanism. (1) Target RNA is reverse transcribed using the first primer flanking one end of the target sequence. (2) RNase H degrades the template RNA strand. (3) The second primer flanking the other end of the target sequence with an overhang containing the sequence of a promoter is extended on the newly produced complementary DNA strand. (4) The double-stranded product can then be transcribed using a DNA-dependent RNA polymerase to make more RNA copies of the target sequence. (5) The resulting RNA product can be fed back into step 1 to achieve exponential amplification (Guatelli et al. 1990). (Adapted from Pardee et al. 2016)

from the resulting RNA/DNA duplex leaving the cDNA available for binding of the forward primer. The reverse transcriptase is then used to extend the forward primer, producing a dsDNA product. During the process of converting the target RNA into dsDNA, primers are also used to append a promoter sequence (e.g., T7 promoter sequence) to the amplified DNA. The dsDNA is then used as a template for the DNA-dependent RNA polymerase (e.g., T7 RNA polymerase) to create additional copies of the target RNA or its antisense complement. The newly transcribed RNA serves as template for further amplification.

Loop-Mediated Isothermal Amplification (LAMP) LAMP was first introduced by Notomi et al. in 2000 as a way of achieving rapid and highly sequence-specific accumulation of dsDNA (Notomi et al. 2000). It can produce approximately 109 (Pardee et al. 2016) copies of the target in less than 1 h. In its simplest form, the method uses four primers. Two inner primers termed FIP and BIP for forward inner primer and backward inner primer, respectively, are used for the bulk of the amplification. Two bumper primers F3 and B3 are used to displace

12

Cell-Free Biosensors: Synthetic Biology Without Borders

253

the extension products of FIP and BIP. Together, these four primers recognize six distinct sequences on the template, which is helpful for enforcing sequence specificity of the amplification. A strand-displacing DNA polymerase (e.g., Bst DNA polymerase large fragment) is used for extension of the primers. The inner primers are comprised of two sequences F1c/F2 and B1c/B2 for FIP and BIP, respectively, as shown in Fig. 2. Briefly, hybridization of FIP with the target DNA initiates the amplification process (Fig. 2a). The strand-displacing DNA polymerase then extends the FIP and unwinds the target dsDNA (Fig. 2b). F3 primer binds to the F3c region on the target leading to displacement of the newly synthesized DNA (Fig. 2c–d). This released ssDNA with a loop at the 50 end then serves as template for DNA synthesis primed by BIP and the B3 primer (Fig. 2e–g). This process forms a dumbbell-shaped DNA intermediate with seed loops that act as sites of annealing and extension without the need for template denaturation (Fig. 2h). The subsequent cycles of elongation and recycling are continued by the inner primer pair, resulting in cauliflower-like products with various numbers of stem-looped structures containing alternately inverted repeats of the target sequence (Zhao et al. 2015). Since positive LAMP reactions become cloudy due to the formation of magnesium pyrophosphate, successful DNA amplification by LAMP can be visually detected with the naked eye. Nonetheless, a number of colorimetric and fluorescence assays have been created for the detection of positive LAMP reactions (Goto et al. 2009), which further facilitate the use of this method in real-world applications. One drawback of LAMP is its strict requirement for higher temperatures (55–65  C) (Xing et al. 2017), which can be contrasted with those of RPA and NASBA (25–42  C) (Burchill et al. 2002; Xing et al. 2017; Piepenburg et al. 2006). LAMP has been used for highly specific detection of many pathogens including Shiga toxin producing E. coli, Salmonella, and Vibrio parahaemolyticus (Li et al. 2017b). Similar to LAMP can also be used to produce a real-time output. For instance, Cao et al. have developed a real-time fluorescence assay for the detection of the common smut of corn, caused by Ustilago maydis. Notably, this work demonstrated a LAMP detection sensitivity 200 times higher than that of conventional PCR (Cao et al. 2017).

Recombinase Polymerase Amplification (RPA) In 2006, Piepenburg et al. reported the RPA method, which amplifies target DNA at 37  C in 20 min or less with the detection limit as low as a single target copy (Piepenburg et al. 2006; Oriero et al. 2015; Lobato and O’Sullivan 2018). The method can be used to generate double- or single-stranded DNA from RNA or DNA templates. RPA takes advantage of recombinase proteins (e.g., UvsX) which bind to primers in the presence of ATP, forming a nucleoprotein complex. In its active form, the complex interrogates the template strand to find a homologous sequence. This leads to strand invasion by the primer at the cognate site. Primers can then be extended by a stranddisplacing polymerase (e.g., Bsu DNA polymerase), providing additional templates for the process. The recombinase-based nucleoprotein complex actively hydrolyses ATP,

254

A. Tinafar et al.

A F2

F3

5'

c F1

FIP 3'

F3c

F2

F1

B1c

B2c

B3c

F2c

F1c

B1

B2

B3

F2

F1

3' 5'

B c F1

F2c

C F3 F3c

D F1c

F1

F3

F1

E

F1c F3

F

B1c

B2c

B3c

BIP

B2 c B1

G B3c F1

B1

F1c

B3 c B1

3'

5'

F1 c

F2c

FIP

F2

H

Fig. 2 Initial steps involved in LAMP. (a) FIP (dotted green and solid blue) is designed such that its 50 end contains an overhang containing F1c (dotted green) and its 30 end contains the complement to F2c, F2 (solid blue). (b) Once annealed to F2c, FIP is extended so that the newly extended strand contains F1c and its complement, F1. (c) To release the new strand, the bumper primer F3

12

Cell-Free Biosensors: Synthetic Biology Without Borders

255

which results in its disassembly; thus, helper proteins (e.g., UvsY) and crowding agents are used to shift the equilibrium in favor of recombinase loading. To help with primer annealing, single-strand DNA-binding proteins are used to stabilize the displaced strand and to prevent primer ejection by branch migration (Piepenburg et al. 2006) (Fig. 3). Although both LAMP and RPA were first developed for DNA amplification, both methods can be used in concert with a reverse transcriptase for amplification of RNA. RPA tends to be quite fast; in one demonstration, real-time detection of norovirus genomic RNA from human stool samples was accomplished using fluorescent probes in as little as 6 min (Moore and Jaykus 2017).

Drawbacks of Using Isothermal Amplification Strategies for Sensing As noted above, reducing false-positive rates is a recognized challenge of using isothermal amplification alone (Cordray and Richards-Kortum 2012). This challenge can be addressed by using logic gates in their various forms. Although isothermal amplification systems have been used to construct logic gates (Xu et al. 2014), complex input processing in such systems tends to be uncommon. Some have offered a way to get around this limitation by introducing multiplexing at the amplification stage and relegating the task of information processing to the user (Nyan and Swinson 2016). In contrast, synthetic gene networks have emerged as a robust platform for utilizing logic gates (Green et al. 2017; Ando et al. 2011; Cameron et al. 2014). While these networks can operate in cells, they can also be hosted in CFPE-based systems (Pardee et al. 2014). Below, the use of such cell-free systems as biosensors will be explored, and integration with isothermal amplification to create more sophisticated sensors will be discussed.

CFPE-Based Biosensors For decades, cell-based biosensors have used engineered gene-circuit-based systems for high-resolution detection of target analytes (Slomovic et al. 2015; Burlage et al. 1990; van der Meer and Belkin 2010; Trang et al. 2005; Bhadra and Ellington 2014; ä Fig. 2 (continued) (solid red) complementary to F3c is extended by a strand-displacing polymerase. (d) The new strand is bumped off the template as the bumper is extended on the original template. (e) This results in the release of the newly extended strand that has a section, F1c, selfcomplementary to its inner sequence, F1. (f) A complementary strand is then produced from the inner primer, BIP, which anneals at the other end. (g) This complementary strand is bumped off using the second bumper, B3, and the strand-displacing polymerase. (h) The resulting new loop generated at the opposite end can now anneal to the first inner primer, FIP; as the process continues, more loops can act as annealing sites for the inner primers, FIP and BIP. Further extension and loop formation lead to the accumulation of a large number of concatemers under isothermal conditions (Notomi et al. 2000)

256

A. Tinafar et al.

A

Primer Recombinase

Single-Stranded Binding Protein

Polymerase

+

+

B UvsY Crowding Agents ATP

UvsY Crowding Agents

ADP+Pi

Fig. 3 The original mechanism developed for RPA. (a) UvsX recombinase forms a nucleoprotein complex with each primer. The complex scans template DNA for a complementary sequence. The displaced strand is bound by the single-strand DNA-binding protein Gp32. Primers are then extended using Bsu polymerase. The resulting two complementary amplicons are fed back into the mechanism to achieve exponential amplification. (b) UvsX binds to oligonucleotides in the presence of ATP to form a nucleoprotein complex. Upon hydrolysis of ATP, the complex disassembles and Gp32 replaces UvsX. The helper protein UvsY and the crowding agent Carbowax shift the equilibrium in favor of recombinase [re]loading (Piepenburg et al. 2006)

Isaacs et al. 2004; Green et al. 2014). These sensors can be generally characterized as systems where recognition events result in production or activation of transcriptionor translation-based reporters (Green et al. 2014; Cameron et al. 2014). In attempts to extend such capabilities beyond the laboratory, recent efforts have begun to translate this impressive body of work into the cell-free environment. As will be discussed below, new and exciting capabilities have evolved in taking gene circuit-based sensors out of the cell. In an early effort to extend the reach of gene circuit-based biosensors, portable and ruggedized cell-free systems were created by freeze-drying CFPE into porous

12

Cell-Free Biosensors: Synthetic Biology Without Borders

257

materials such as paper. Here CFPE reactions were used to host toehold switch-based sensors. These RNA-based sensors act as regulators of translation and can be rationally designed to detect specific single-stranded RNA or DNA targets (Green et al. 2014). This technology was first applied to the detection of antibiotic resistance genes and strain-specific Ebola viral sequences using freeze-dried paper-based systems that created colorimetric outputs. While it was exciting to be able to detect molecular sequences using paper-based cell-free biosensors, the detection limit was relatively high; it could detect ssRNA down to a concentration of 30 nM, which was insufficient for sensing many pathogens in clinical samples (Pardee et al. 2014). This challenge was solved by combining toehold sensing with a preceding NASBA isothermal amplification step, which boosted the detection limit to low femtomolar concentrations. Using this combined approach, clinically relevant concentrations of Zika virus could be detected. Importantly, as discussed, combining isothermal amplification with toehold switches provided the advantage of having two sequence-specific checkpoints for the generation of a positive output signal, which should provide greater specificity than either method alone. This system was also augmented with a novel CRISPR-/Cas9-based module making it capable of discriminating between closely related viral strains. Here, the differential PAM sites within the target regions of the Zika virus were used to distinguish between strains (e.g., American vs. African) with single base pair resolution (Pardee et al. 2016). The power of CFPE-based sensing can be harnessed beyond nucleic acid sequence detection. An impressive demonstration comes from the Freemont lab where cell-free expression systems have been used to tackle the challenge of monitoring Pseudomonas aeruginosa infections in cystic fibrosis patients by sensing a quorum molecule, 3-oxo-C12-HSL. Here they demonstrated that the abundance of this bacterial biomarker in human sputum can be measured at nanomolar quantities using gene circuit-based molecular recognition (Wen et al. 2017). A number of CFPE-based strategies have focused on the detection of contaminants in water supplies. An early example of this was aimed at sensing antibiotics by exploiting the inhibition of protein synthesis that is caused by many of these drugs (e.g., paromomycin, tetracycline, chloramphenicol, and erythromycin). This was done simply by monitoring the production of a colorimetric reporter protein in the presence of these compounds (Duyen et al. 2016). Another group has successfully shown that mercury contamination can be detected by utilizing the mercury(II)responsive transcriptional repressor MerR to control the expression of a reporter gene (Didovyk et al. 2017). Endocrine-disrupting chemicals (e.g., bisphenol A) mimic the function of natural estrogen hormones and have been associated with health concerns such as cancer and developmental problems. Conventional methods of detecting these compounds involve the use of yeast or human cell lines which can be relatively time-consuming or mass spectrometry which often necessitates high equipment costs. To offer an alternate detection strategy, CFPE was used to synthesize and host allosterically activated fusion proteins containing the ligand-binding domain of a hormone receptor and a β-lactamase reporter. It was shown that the switching of these proteins into

258

A. Tinafar et al.

active conformation, when exposed to endocrine-disrupting chemicals, can be used to detect these compounds in diluted samples of blood, urine, and wastewater at low nanomolar concentrations (Salehi et al. 2017, 2018). Considering the power and flexibility of cell-free sensing technologies, diagnostics, environmental monitoring, and microbial detection seem to be simply the tip of the iceberg. As the techniques develop and the costs are further reduced, translation of more of these technologies into real-world applications can certainly be expected. Having reviewed these fundamental cell-free operating environments, the focus of this chapter will now shift to the specific molecular mechanisms that carry out recognition functions in cell-free biosensors. Below selected mechanisms will be reviewed, with the goal of providing the reader with an overview of the diverse and creative designs that have been demonstrated to date.

Recognition Components in Biosensors While the focus here will be on cell-free biosensors, it is important to note that most research groups developing cell-free biosensors also have considerable experience in constructing cell-based systems. This highlights an important feature of cell-free biosensors, which is that many biosensors are transferrable between cell-based and cell-free formats. This is excellent news for cell-free biosensors in that it enables cell-free applications to draw from an expansive library of pre-existing cell-based sensors. In the long run, this interoperability between systems will also likely foster greater practical implementation of biosensors in both contexts. Below, a large body of biosensing work from cell-free and cell-based efforts to date will be reviewed, with the intention of promoting cross-pollination between these modalities. These sensors cover analytes that include small molecules, nucleic acids, and proteins, as well as other inputs such as light.

Riboregulators, Riboswitches, and Fluorogenic Aptamers RNA molecules play essential roles in living organisms, adopting functions far beyond their role as the intermediary between genes and proteins. Furthermore, RNAs can adopt a wide variety reactions of different structures for binding different ligands and catalyzing chemical reaction (Doherty and Doudna 2000). The versatility of RNAs in both function and structure makes them important regulatory components in nature for sensing and signal transduction via structural switching upon ligand binding. In recent years, with increasing understanding of these natural regulatory pathways, researchers have succeeded in manipulating biological processes with artificial devices such as riboregulators or riboswitches that act in response to different molecular stimuli. Such systems constitute a valuable toolkit for use in cell-free biosensing platforms.

12

Cell-Free Biosensors: Synthetic Biology Without Borders

259

Riboswitches for Nucleic Acid Sensing Post-transcriptional gene regulation in prokaryotes can be achieved by tuning the accessibility of the ribosomal binding site (RBS) to the ribosome. One of the first modular engineered riboregulator designs was reported in 2004 by Isaacs et al. (2004) (Fig. 4a). This riboregulator introduces a short RBS complementary sequence in cis to block recognition of the RBS by the ribosome. The cis-repressed RNA (crRNA) also features a pYrimidine-Uracil-Nucleotide-puRine (YUNR) motif in its loop region, which mediates a loop-linear intermolecular interaction with a transactivating RNA (taRNA) molecule. The taRNA-crRNA cognate interaction releases the sequestered RBS and activates downstream gene expression. However, this design was limited in orthogonality because of sequence constraints in crRNA design, which require the integration of the natural YUNR motif as well as invariable base pairing to RBS sequence. Later work (Rodrigo et al. 2012; Mutalik et al. 2012) expanded the size of orthogonal riboregulator libraries by mutating the YUNR motif in the loop domain and by expanding the loop size. These reports demonstrated that YUNR motif is dispensable for initiating RNA-RNA interactions in trans and further that these bimolecular RNA interactions are largely determined by RNA complex thermostability and the seed region that initiates binding.

A YUNR-mediated riboregulator: translational regulation YUNR

Ribosome

ON

taRNA

OFF Start codon

Gene

B Toehold switches: translational regulation Ribosome

OFF RBS

ON

Trigger RNA

Start codon Gene

C STARs: transcriptional regulation

UUUU

UUUU

OFF

Terminator

STAR

Gene

UUUU

Gene

ON Gene

RNAP

Fig. 4 Mechanisms of three engineered riboregulators. (a) Riboregulators reported by Isaacs et al. that employ a conserved YUNR motif (Isaacs et al. 2004). (b) Toehold switches that regulate gene expression at the translational level (Green et al. 2014). (c) STARs that regulate gene expression at the transcriptional level (Chappell et al. 2017)

260

A. Tinafar et al.

Building on these advances, Green et al. in 2014 reported a set of riboregulators generated purely through de novo computational design called toehold switches Green et al (2014). These riboregulators sequestered the start codon between RNA duplexes and concealed the RBS within the loop of an RNA hairpin (Fig. 4b). By achieving repression without base pairing to the RBS sequence or start codon, this design allowed toehold switches to be created that were responsive to trigger RNAs of nearly any arbitrary sequence. Rather than using a loop-linear or loop-loop interaction, binding between the trigger and the riboregulator was initiated by a single-stranded domain appended to the 50 end of the hairpin, which is referred to as a toehold. The resulting linear-linear interaction yielded kinetically and thermodynamically favorable interactions, enabling modulation of protein expression over two orders of magnitude. Taking advantage of the high dynamic range and orthogonality of the toehold switches, this work was extended to detect endogenous RNAs in E. coli (Green et al. 2014) and to construct multiple-input RNA-based computing circuits to enable complex translational regulation in living cells (Green et al. 2017). Importantly, as discussed above, toehold switches are well-suited to operate in cellfree environments where they have been used to detect antibiotic resistance mRNAs and pathogen nucleic acids in paper-based diagnostic systems (Pardee et al. 2014, 2016). Riboregulators that operate at the transcriptional level have also been widely studied. Using transcriptional regulation, the RNA input signals can generate RNA output signals. This use of the same type of biomolecule for input and output is a desirable property for constructing genetic circuits as it facilitates the construction of layered logic gates of greater complexity. Starting from natural transcriptional attenuators which promote the formation of a downstream intrinsic terminator upon binding with a complementary antisense RNA, Lucks et al. (2011) engineered orthogonal attenuator variants through rational mutagenesis of the attenuator hairpin loop and collar sequences. However, the dynamic range and orthogonal library size were still limited by designs preserving the endogenous loop-loop-mediated RNA interaction, which affects RNA folding and thermodynamic free energy. To achieve higher performance, the Lucks group (Chappell et al. 2017) optimized their transcriptional regulators using computational design and constructed high-performance and orthogonal small transcription activating RNAs (STARs) (Fig. 4c). STARs employ thermodynamically favorable linear-linear RNA interactions to reconstruct the sequestered terminator and use the 50 linear region as the seed for specific RNA recognition, achieving up to ~5,000–13,000-fold activation, which is the highest dynamic range by a riboregulator reported to date.

Riboswitches for Small Molecule and Protein Sensing RNA-based regulation can be achieved by RNA base pairing interactions like the RNA-sensing riboregulators discussed above or by interactions between RNA and small molecule or protein ligands that occur in riboswitches. In the latter case, the binding of a ligand to the RNA triggers allosteric structural switching that leads to changes in gene expression. Yen et al. (2004) incorporated a ribozyme into an mRNA sequence to promote self-cleavage of the RNA transcripts and suppression

12

Cell-Free Biosensors: Synthetic Biology Without Borders

261

A Cis regulation ON

OFF

inactive ribozyme

Ligand

active ribozyme aptamer

aptamer

mRNA

gene

AAAAA

gene

AAAAA

Ribozyme switch

B Trans regulation

aptamer

Ligand

OFF mRNA

anti_seq

gene

AAAAA

Antiswitch

Fig. 5 Mechanisms of riboswitches that give gene regulation in cis (Rooney et al. 2009) or in trans (Bayer and Smolke 2005). (a) Binding of a ligand triggers an allosteric structural change in the RNA to activate a self-cleaving ribozyme in cis. Cleavage of the RNA prevents translation of the downstream gene of interest. (b) In trans regulation, binding of the ligand causes exposure of an antisense RNA sequence. This antisense sequence then interacts in trans with an mRNA encoding the gene of interest

of translation. They then selected molecular ligands that bound to the ribozyme and inhibited RNA self-cleavage to encourage protein translation. This ribozyme-based regulation strategy requires small molecule screening, which cannot be rationally designed and is limited by the sequence of the ribozyme itself. In order to make ribozyme-based regulation more modular and extensible, Win and Smolke (Rooney et al. 2009) fused an RNA aptamer to a ribozyme to make an allosteric ribozyme responsive to small molecules (Fig. 5a). The binding of small molecules in the aptamer region or sensor region is able to either disrupt or reconstruct the ribozyme structure, achieving up- or downregulation of the protein of interest. Another work (Bayer and Smolke 2005) achieved trans regulation through modular, tunable allosteric riboswitches that regulate expression of transcripts in a ligand-binding manner (Fig. 5b). The conformational change upon binding with a target small molecule releases a sequestered antisense RNA sequence for binding with the mRNA for translational regulation. When combined with in vitro aptamer screening platforms, these allosteric aptamer-based regulation schemes can be extended to a vast array of potential target molecules.

Fluorogenic Aptamers and Sensor Systems Fluorogenic RNA aptamers are RNA aptamers that generate fluorescence upon binding of their otherwise non-fluorescent fluorogen ligands. As a result, these systems represent promising tools for signal transduction and readout in cell-free biosensors. Spinach, an RNA-fluorogen complex that mimics the green fluorescent

262

A. Tinafar et al.

protein, was identified by performing column-based systematic evolution of ligands by exponential enrichment (SELEX) in 2011 (Fig. 6a) (Paige et al. 2011). It has been used as a tag for messenger RNA (mRNA), and has also been adapted for use in protein-free biosensors for detection of proteins and small molecules (Paige et al. 2012; Song et al. 2013; Strack et al. 2014; You et al. 2015; Svensen and Jaffrey 2016). Current methods for sensing nucleic acids using fluorogenic aptamers can be grouped into three main categories that employ (1) destabilization of the fluorogenbinding site, (2) misfolding of the aptamer sequence, and (3) assembly of split aptamer halves. These general methods have all been implemented using the Spinach aptamer and can be used for in vitro detection and, in some cases, live cell imaging. In systems that destabilize the fluorogen-binding site (Fig. 6b), one of the stem regions of an aptamer is disrupted, and its sequence is extended at both ends with those complementary to the target molecule. When the target sequence is present, it hybridizes with the fluorogenic sensor stabilizing the aptamer fold and restoring aptamer fluorescence. This approach has been employed using a modified version of the Spinach aptamer for detection of mRNA and noncoding RNA (e.g., microRNA, small interfering RNA) (Ong et al. 2017; Aw et al. 2016). For the second class of sensors (Fig. 6c), the sequence of the aptamer is combined with additional sequences that create strong secondary structures to cause the aptamer to misfold and prevent fluorescence. Upon binding of the target DNA or RNA sequence, a strand displacement reaction occurs causing the aptamer to refold into its active conformation for binding to the fluorogen (Bhadra and Ellington 2014; Akter and Yokobayashi 2015; Ying et al. 2017; Huang et al. 2017). Sensors of this kind have been used with the Spinach aptamer for diagnostic applications (Bhadra and Ellington 2014; Huang et al. 2017) and in catalytic RNA amplifier circuits (Akter and Yokobayashi 2015). Lastly, there are binary probes that comprise two halves of aptamer sensors that are unable to hybridize except in the presence of the target DNA or RNA (Fig. 6d). Such a split aptamer probe for Spinach was reported by Kikuchi and Kolpashchikov in the sensing of a fragment of the inhA gene from Mycobacterium tuberculosis (Kikuchi and Kolpashchikov 2016). They found that the split aptamer design resulted in very low sensor leakage, with only a single base mismatched target producing detectable background fluorescence. Here, the Spinach aptamer has been used to illustrate the potential of aptamer-based cell-free biosensing, but these concepts can be extended to any other fluorogenic aptamer such as the malachite green and sulforhodamine B binding aptamers, Mango, Corn, Broccoli, and Red Broccoli (Ying et al. 2017; Song et al. 2017; Alam et al. 2017; Kolpashchikov 2005; Dolgosheina et al. 2014; Sato et al. 2015). These aptamers can also be combined to enable multiplexed detection.

CRISPR-Enabled Nucleic Acid Detection The CRISPR enzymes, which are programmable RNA-guided nucleases, have demonstrated great potential in diagnostic applications in recent years. The use of

12

Cell-Free Biosensors: Synthetic Biology Without Borders

a

263

b DFHBI

Spinach sensor

target molecule

c

d DFHBI

target molecule

split Spinach sensor target molecule

Spinach sensor

Fig. 6 Structures of the Spinach-DFHBI aptamer/fluorogen complex and Spinach-based sensor designs. (a) Sequence and secondary structure of Spinach-DFHBI (Huang et al. 2014). (b) A Spinach sensor with a destabilized fluorogen-binding site. The sensor remains non-fluorescent until hybridization with a target molecule. (c) A Spinach sensor that is triggered by strand displacement based on the design by Huang et al. (2017) (d) A split Spinach nucleic acid sensor. Target molecules are indicated in red, target-binding regions are indicated in blue, and the Spinach aptamer is indicated in green

264

A. Tinafar et al.

Fluorescence

CRISPR in biosensing applications began with the 2016 development of rapid and low-cost Zika virus biosensors used in paper-based cell-free reactions (Pardee et al. 2016). As described above, toehold switches were designed to respond to viral RNA sequences and to give a colorimetric readout. To achieve high detection sensitivity, the RNA in virus samples was amplified by a NASBA isothermal amplification reaction. This amplification process, when coupled with CRISPR/Cas9 cleavage, was able to achieve single-base discrimination for strain-specific detection. This strategy, termed NASBACC for NASBA CRISPR cleavage, exploits the selective cleavage of DNA by Cas9 in the presence of an NGG protospacer adjacent motif (PAM). If a mutation of interest abolishes or establishes a PAM motif, Cas9mediated DNA cleavage is affected, generating either full-length or truncated DNA products depending on if the wild-type or mutated sequence is present. Since only full-length transcripts can activate the toehold sensors, strain-specific detection of Zika virus was achieved. The reliance of this technique on mutations to NGG PAM motifs, however, limits its application to only the subset of mutations that affect PAM sites. Such mutations comprise only 14.6% of all potential singlebase substitutions. In 2017, Gootenberg et al. reported SHERLOCK (Specific High-Sensitivity Enzymatic Reporter UnLOCKing) for rapid and low-cost pathogen detection and genotyping (Gootenberg et al. 2017). They took advantage of the guide RNA-programmed collateral cleavage from the Cas13a enzyme to nonspecifically degrade FAM/quencher reporter RNA upon recognition of target RNA. When coupled with RPA isothermal amplification and T7 transcription to generate the target RNA, this detection method achieved attomolar sensitivity (Fig. 7). They later optimized this platform into SHERLOCKv2 with improvements in multiplexing and sensitivity (Gootenberg et al. 2018). By profiling nucleotide cleavage preferences of CRISPR enzymes, Gootenberg and colleagues were able to combine four orthogonal CRISPR enzymes for multichannel detection to distinguish different combinations of targets. Next, since the cleavage substrate of Cas13a can

RPA dsDNA transcription

RT-RPA RNA

target sequence

Cas13a-crRNA

cleavage reporter

Fig. 7 Schematic illustration of the SHERLOCK (Specific High-Sensitivity Enzymatic Reporter UnLOCKing) nucleic acid detection method (Gootenberg et al. 2017, 2018). Nucleic acids from a sample are amplified isothermally using RPA and transcribed. The amplified pathogen RNA is detected through a Cas13a guide RNA complex, which indiscriminately cleaves singlestranded RNA reporters after binding to the target RNA. SHERLOCK assay results can be measured using fluorescence or commercially available lateral flow test strips

12

Cell-Free Biosensors: Synthetic Biology Without Borders

265

serve as the activator for the CRISPR type-III effector nuclease Csm6, they coupled the two enzymes and achieved increased signal output and reaction kinetics for detection. This platform was also successfully adapted to a visual readout system with an instrument-free lateral flow detection assay that reaches a sensitivity of 2 aM in 2 h. Aiming at field assays for real-world samples, Myhrvold et al. (2018) developed HUDSON (heating unextracted diagnostic samples to obliterate nucleases) to extract RNA from viral particles in clinical samples. Pairing with SHERLOCKv2, this platform enabled instrument-free detection with attomolar sensitivity as well as high specificity to realize the detection of clinically relevant viral single-nucleotide polymorphisms. Similarly, the Doudna group (Chen et al. 2018) characterized the collateral cleavage of single-stranded DNA by Cas12a when bound to an ssDNA or dsDNA activator. Based on this property of the enzyme and RPA isothermal DNA amplification, they developed a one-pot detection method termed DNA endonucleasetargeted CRISPR trans reporter (DETECTR) with a fluorescence readout and attomolar sensitivity.

Functional DNA Molecules for Biosensing Although previous sections have described systems that primarily make use of the catalytic and fluorescence properties of proteins for detection of different analytes, the intrinsic properties of DNA and RNA enable them to be used without intervening proteins for biosensing applications. This section describes methods of using DNA molecules alone for detection and amplification, and it discusses how such DNA systems can be merged with the molecular recognition capabilities of proteins, in particular antibodies, for novel sensing modalities.

Functional DNA Molecules for Nucleic Acid and Protein Detection Functional DNA molecules have been widely used to construct biomolecular circuits and sensor systems. In general, functional nucleic acid molecules comprise aptamers, which bind to specific ligands, and DNAzymes, which can catalyze chemical reactions such as DNA cleavage. Aptamers enable nucleic acid circuits to detect diverse inputs, while DNAzymes can be used to display the results of a detection event. By exploiting the programmable nature of nucleic acid interactions, nucleic acid circuits that combine these functions with information-processing capabilities can be implemented. For example, as shown in Fig. 8a, Elbaz et al. constructed a DNA-based computational platform based on DNAzymes and their substrates. They demonstrated multilayered gate cascades, fan-out gates, and parallel logic gate operations (Elbaz et al. 2010). This system can also be used to regulate the controlled expression of antisense molecules, or aptamers, that act as inhibitors for enzymes. Orbach et al. exploited DNAzymes to construct reversible Toffoli and Fredkin logic gates (Orbach et al. 2012). Since DNAzymes are able to cut DNA strands in response to molecular targets such as metal ions or DNA strands with specific sequences, the cleavage can be implemented to cut a reporter to release signal for the detection of nucleic acids (Fig. 8b) (Wang et al.

266

A. Tinafar et al.

A

B

C +cDNA

eT

eT denaturation

target

D

1* 2*

1 2

3* 5

3

3

2*

3*

4

+

4* 4 H2

H1

5 12

3*

E

2*

+

+

2* 3 4 3*

2* 3* 4* 3 4 Assembled complex

Fig. 8 DNA-based circuits and biosensors. (a) Logic computation based on DNAzymes (Elbaz et al. 2010). Schematic showing the organization of DNAzyme subunits and the fluorophorelabeled substrate following the binding of sequences that activate the XOR gate. (b) DNA sensor for nucleic acid detection based on DNAzyme-mediated cleavage (Wang et al. 2011b). Upon adding the target DNA, the DNAzyme will be activated to cut the reporter strand to release the signal. (c) Electronic biosensor based on the conformational change of DNA upon target binding (Fan et al. 2003). The conformation of the DNA hairpin will change upon adding the target DNA and will lead to a corresponding change in the electrochemical signal. (d) Enzyme-free DNA circuit for signal amplification (Chen et al. 2013a). The target strand C1 is able to catalyze the interaction of two hairpin strands H1 and H2. (e) Toehold exchange probe for nucleic acid detection with single nucleotide specificity (Zhang et al. 2012b). The toehold exchange probe is programmed based on the equilibrium free energy in response to binding of the target DNA and produces a fluorescent readout signal

2011a, b) and metal ions (Hollenstein et al. 2008; Zhang et al. 2011; Zhou et al. 2017). For biosensors based on DNA aptamers, after target binding to the DNA aptamer, the conformation of the DNA aptamer will change, which can lead to variations in fluorescence signal as a result of changes in the distance between fluorophore-quencher pairs conjugated to different sites of the aptamer. Such target-induced fluorescence change can be used as a reporter signal to detect small molecules (Zuo et al. 2009) and proteins (Hamaguchi et al. 2001). For example, Yu et al. developed a split aptamer assay that is capable of detecting cocaine in saliva with high specificity and sensitivity (Yu et al. 2017). When DNA is modified with an electron transferable molecule, conformational changes induced by a target can affect electron transfer between the molecule and an electrode surface, leading to changes in electric current. Such signal modulations can

12

Cell-Free Biosensors: Synthetic Biology Without Borders

267

be implemented to construct electrochemical sensors for the detection of nucleic acids (Fan et al. 2003; Xia et al. 2010), ions (Xiao et al. 2007a; Porchetta et al. 2013), small molecules (Baker et al. 2006), and proteins (Xiao et al. 2005, 2007b). For example, Fan et al. developed an electrochemical sensor that involves an electroactive, ferrocene-tagged DNA stem-loop structure attached to a gold electrode surface. Hybridization of a target nucleic acid induces a substantial change in DNA conformation at the surface and a concomitant shift in the electron-transfer tunneling distance between the electrode and the redox active label. The detection limit of the sensor is as low as 10 pM. Baker et al. developed a rapid, reagent-less cocaine sensor by interfacing this system with a cocaine-binding aptamer (Baker et al. 2006).

Amplification and Polymorphism Detection Using Dynamic DNA Nanotechnology Novel molecular sensors have also been developed in dynamic DNA nanotechnology, an emerging field that aims to construct reconfigurable and autonomous devices that can respond, actuate, and compute (Zhang and Seelig 2011). Toehold-mediated strand displacement reactions are a common strategy used in dynamic DNA nanotechnology. In these interactions, a single-stranded DNA molecule is used to displace another single-stranded DNA from a pre-hybridized duplex through a short single-stranded toehold region. Toehold-mediated strand displacement can be used to engineer catalytic systems, such as catalytic hairpin reactions (CHR) (Yin et al. 2008), and polymerizing systems, such as hybridization chain reactions (HCR) (Dirks and Pierce 2004). CHR systems employ a single-stranded DNA trigger to catalyze the interaction between two hairpins (Fig. 8d), while in HCR systems the single-stranded DNA trigger initiates autonomous polymerization of two DNA hairpins to form extended DNA duplexes. These two purely DNA-based reaction systems can be used for signal amplification in nucleic acid biosensors without enzymes (Chen et al. 2013a; Allen et al. 2012; Niu et al. 2010). For example, Chen et al. successfully stacked a four-layer CHR cascade that yielded upward of 600,000-fold signal amplification using DNA alone. Aside from signal amplification without enzymes, dynamic DNA strand displacement can also be used to detect single-nucleotide polymorphisms (SNPs). As shown in Fig. 8e, Zhang et al. developed DNA sensors termed toehold exchange probes that exploit strand displacement reactions to detect SNPs with very high specificity (Zhang et al. 2012b). The design of toehold exchange probes relies on thermodynamic modeling with the aim of maximizing the signal difference between wildtype- and SNP-containing nucleic acid samples. These probes are able to function robustly under various temperatures and ion concentrations. Zhang and colleagues further applied the principles of dynamic DNA nanotechnology to detect very low concentrations of SNP alleles (down to 0.1% of a mixed population, which is relevant for liquid biopsies) (Wang and Zhang 2015) and detect mutations in double-stranded DNA (Chen et al. 2013b). Recently, a modular method for probe construction has also been developed to detect clinically relevant SNPs in complex sequences, such as long targets and those with multiple benign polymorphisms

268

A. Tinafar et al.

(Wang et al. 2017). The Seelig group also recently developed a molecular classifier that can be used to detect multiple RNA transcripts and to perform classification of the cell state (Lopez et al. 2018). Their strategy started from training a computational classifier on labeled gene expression data in silico. Then, a DNA probe set that was designed according to the trained model was used for classification through interaction with RNA inputs. The classifier can be applied in diagnostics for early detection of cancer or differentiating between viral and bacterial respiratory infections based on host gene expression.

DNA/Antibody Hybrid Systems for Protein and Small Molecule Detection The creation of systems that merge the molecular recognition capacity of antibodies with the programmability of DNA has opened up new in vitro detection modalities for cell-free biosensors. Along with allowing for signal transduction between proteins and nucleic acids, these systems have been utilized to control antibody activity (Janssen et al. 2013), detect specific biomolecules, (Heyduk et al. 2008; Porchetta et al. 2018; Zhang et al. 2014; Kjelstrup et al. 2018), and serve as platforms to promote DNA strand exchange (Janssen et al. 2015; Engelen et al. 2017). One of the first studies to use these hybrid structures employed peptide-DNA conjugates to control antibody activity by inhibiting antigen binding. This effect is achieved by using bivalent peptide-DNA conjugates to bind each of the antigenbinding sites on an antibody (Janssen et al. 2013). The conjugate structures are prepared using complementary DNA handles, which produce two peptides connected through a dsDNA linker (Fig. 9a). Use of the dsDNA linker enhances peptide binding, leading to a 500-fold stronger interaction with the antibody compared to the peptide alone, and leads to inhibition of antibody activity. In the presence of matrix metalloprotease 2, a protease overexpressed during tumor metastasis, the peptide-DNA strand linkers are cleaved at the linker sequence PLGLAG, which is recognized by the protease. Upon cleavage, binding activity is restored. Expanding on this concept, Janssen et al. were able to use DNA strand exchange to displace the DNA linker, implementing AND and OR logic to control antibody activity (Janssen et al. 2015) (Fig. 9b). This same concept was then used to control DNA circuits through antibody-initiated strand exchange (Engelen et al. 2017). Using similar methods, DNA/antibody hybrid systems have been used to implement several types of biosensors. In one report, complementary DNA strands were conjugated to antibodies, and each DNA strand was conjugated to an acceptor or donor of a Förster resonance energy transfer (FRET) pair (Heyduk et al. 2008). In the presence of the target antigen, the two antibodies containing the complementary DNA sequences bind, bringing the DNA and FRET pair in close proximity. This structural change allows the DNA strands to hybridize and cause a change in fluorescence. Another study was able to use DNA nanoswitches to detect antibodies in body fluids (Porchetta et al. 2018). This was done by designing three DNA strands (Fig. 9c). Strand 1 contains a self-complementary portion along with a stem loop

12

Cell-Free Biosensors: Synthetic Biology Without Borders

269

a Peptide-dsDNA conjugate

+

Block antibody binding sites

Add matrix metalloprotease; linker is cleaved

Antibody activity is restored

b

+

dsDNA with toehold

Add displacement strand

Bound strands released; antibody activity is restored

c Strand 1 Q

F

+

Strands hybridize; fluorescence is induced

Strand 2 Strand 3

d

Without Antibody a

A S

With Antibody a

A

S

d [AS] = [BS]

B

B S

B d

a

A

a

A

d B

[AS] > [BS]

S

d

Fig. 9 Methods of protein detection and protein function control using DNA/antibody hybrid systems. (a) A bivalent peptide-dsDNA binds to antibody binding sites to block antibody function. Function can be restored through cleavage of the linker sequence PLGLAG, which is recognized by matrix metalloprotease 2, to separate the dsDNA from the peptides. (b) Bivalent peptide-dsDNA with a toehold binds to the antibody arms to inhibit antibody function. Strand displacement is used to reverse binding by disrupting the dsDNA linker and converting them to monovalent peptidessDNAs. The peptide-ssDNAs are released and antibody function is restored. (c) A fluorescence assay for antibody detection. Strands 2 and 3 are conjugated to the target antigen, while strand 1 contains a fluorophore-quencher pair within a stem-loop region. Addition of the cognate antibody leads to co-localization of all three strands, which promotes interaction between strands 1 and 3 to disrupt the stem-loop and generate a fluorescence signal. (d) Protein and small molecule detection using strand displacement equilibrium dynamics. When combined in 1:1:1 molar ratios, the system forms equal concentrations of the bimolecular complexes AS and BS. Strands A and S are modified with an acceptor (“a”) and donor (“d”) FRET pair, respectively. After addition of an antibody that binds to the ligand conjugated to strand B, the shift in equilibrium will generate AS complexes and lead to an increased FRET signal

270

A. Tinafar et al.

containing a fluorophore/quencher pair and a DNA tail. Strand 2 is the complementary strand to the DNA tail and is conjugated to a target antigen. Strand 3 is the complementary sequence to the loop region of strand 1. Despite the complementary sequences of strands 1 and 3, the small loop region of strand 1 ensures that the two species will not interact strongly with one another when they are free in the solution. In the presence of the target antibody, however, the binding of strands 2 and 3 to the antibody will initiate interactions with strand 1 to disrupt its stem-loop structure. This strand displacement interaction frees the fluorophore from the quencher to induce fluorescence, signaling the presence of the target antibody. Another method uses strand displacement equilibrium dynamics to establish a competitive assay for the detection of small molecules and proteins (Zhang et al. 2014; Kjelstrup et al. 2018). This is done through the design of DNA strands A and B that bind competitively to overlapping regions of a third strand S (Fig. 9d). Strands A and S contain FRET pair fluorophores, and strand B is conjugated to a small molecule ligand. Both strands A and B are designed to bind to strand S with the same free energy and can displace one another through toehold-mediated interactions. Because of the balanced system thermodynamics, a 1:1:1 mixture of the three strands reaches an equilibrium where there is an equal concentration of AS and BS complexes. In the presence of a protein that binds to the ligand on strand B, strand B will interact with the protein and weaken its affinity for strand S. The resulting shift in equilibrium leads to the formation of additional AS complexes and a corresponding increase in FRET pair fluorescence. This assay can be used to detect small molecule ligands or any protein that binds to these small molecules, since the reaction equilibrium will shift depending on the concentration of free ligands or unbound proteins present in the reaction. Thus far, the assay has been applied to the detection of biotin, digoxigenin, vitamin D, folate, and digoxin in buffers and in plasma (Zhang et al. 2014; Kjelstrup et al. 2018).

Light-Sensitive Transcription Factors and Promoters Light-driven control systems can enable precise temporal control of cell-free sensors in portable, field-ready instruments. Several mechanisms have been developed for light-based control in cells and in vitro. These mechanisms have made use of phytochrome photoreceptors that utilize a two-component system (Levskaya et al. 2005; Tabor et al. 2011; Schmid et al. 2014; Fernandez-Rodriguez et al. 2017; Shimizu-Sato et al. 2002; Tabor et al. 2009), light-oxygen-voltage sensors (Ohlendorf et al. 2012; Moglich et al. 2009; Jayaraman et al. 2018), or photoswitchable azobenzenes (Kamiya et al. 2015) capable of altering DNA structure in response to light to promote or inhibit transcription. The Voigt lab began the development of light-sensitive promoters in prokaryotes by designing the first two-component system in E. coli that was responsive to red light (Levskaya et al. 2005). Natural two-component systems contain a membranebound light-responsive photoreceptor and an intracellular response molecule controlled by the photoreceptor. This intracellular response can then be used to control gene expression in a system upon exposure to light. To implement a light-sensitive two-component system in E. coli, the Voigt lab fused a phytochrome from

12

Cell-Free Biosensors: Synthetic Biology Without Borders

A

B

light

271

Inactive promoter trans isomers

cell wall 400 nm light

P

450 nm light

RNA P

ompC promoter

Active promoter

Reporter Gene cis isomers

RNA Polymerase

Fig. 10 Light-sensitive gene expression systems. (a) Photoreceptor-histidine kinase complex (Levskaya et al. 2005) is bound to the membrane surface of E. coli. In the absence of red light, histidine kinase phosphorylates the response regulator and gene expression occurs. Exposure to red light deactivates autophosphorylation, and gene expression is turned off. (b) Photoswitchable azobenzene can switch from cis to trans with UV light exposure or from trans to cis with blue light exposure (Kamiya et al. 2015). This change in conformation is used to inhibit transcription (trans) or to stabilize the DNA structure (cis) to enable the binding of RNA polymerase

cyanobacteria with histidine kinase domains in E. coli using the EnVZ-OmpR system. The EnVZ-OmpR system is used in nature to regulate the expression of porin during osmotic shock. Red light was used to deactivate gene expression by turning off autophosphorylation of the histidine kinase (Fig. 10a). They were also able to use this system for “bacterial photography,” using the reporter gene LacZ with a color indicator substrate. This method has since been made easier to incorporate into E. coli by condensing it from a three-plasmid to a two-plasmid system, and a large genetic circuit has been constructed to engineer cells to be responsive to red, green, and blue light simultaneously (Tabor et al. 2011; Schmid et al. 2014; Fernandez-Rodriguez et al. 2017). An edge detection program has also been implemented in the system by integrating NOT and AND logic to determine the edge interface between light and no light (Tabor et al. 2009). While such membrane-bound sensors may pose a challenge in cell-free systems, it is worth noting that many membrane-bound compartmentalized CFPE-based systems have been developed, some with proteins embedded within them (Shin and Noireaux 2012; Fenz et al. 2014; Nourian and Danelon 2013; Ishikawa et al. 2004). This includes the oligomerization and membrane insertion of proteins from in situ cell-free expression (Noireaux and Libchaber 2004). Such vesicle-based systems are where many of the early concepts of cell-free synthetic biology began, and, as these technologies develop, it is exciting to think about how light-actuated nano- and micro-biosensors could be used in synthetic biology-actuated therapeutics. A weakness of the original light-responsive system from the Voigt lab was its reliance on the chromophore phycocyanobilin for light absorption, which requires a separate plasmid for expressing a two-gene metabolic pathway for biosynthesis. The Möglich lab sought to remedy this issue by designing a one-plasmid system with the ability to turn gene expression on and off upon light exposure through the creation of

272

A. Tinafar et al.

pDawn and pDusk promoters (Ohlendorf et al. 2012). These promoters also act as a two-component system in E. coli, but they employ the YF1/FixJ system to respond to light and make use of the ubiquitous flavin mononucleotide as a chromophore. The YF1 protein in this system is a fusion of the light-oxygen-voltage (LOV) photosensor domain from the Bacillus subtilis YtvA protein with the phosphoacceptor and catalytic subdomains of the Bradyrhizobium japonicum FixL protein (Moglich et al. 2009). In the absence of illumination, YF1 acts as a histidine kinase that phosphorylates the transcription factor FixJ to enable it to bind to its promoter FixK2. Under blue light illumination, YF1 loses its kinase activity and can no longer activate FixJ to induce transcription. The pDusk promoter makes direct use of YF1/FixJ to turn off gene expression in response to blue light. In the pDawn promoter, gene expression is activated in response to blue light by using FixK2 to drive expression of the cI repressor of λ phage, which in turn halts transcription from the pR promoter. As a whole, the YF1/FixJ system provides a high degree of control over gene expression due to its ability to activate and deactivate gene expression within a few hours of light exposure and has demonstrated a sizeable 460-fold change in induction upon illumination. However, its direct use in cell-free systems has yet to be documented. Another promoter system using LOV with a similar design based on the EL222 protein has recently been used in cell-free systems (Jayaraman et al. 2018). It has been reported that this system shows higher than tenfold dynamic expression in the presence of blue light. Photoswitchable azobenzenes have also been utilized for light-induced gene expression in cell-free systems (Kamiya et al. 2015). This is achieved by the transition between cis and trans isomers of azobenzene derivatives when exposed to UV/blue light on a timescale of 5–10 min, which can either stabilize DNA structures or inhibit transcription (Fig. 10b). The time scales at which these light-driven reactions occur vary widely. For mechanisms using phytochrome photoreceptors that use a two-component system, reaction rates are dependent on gene expression rates in bacteria, placing them in a several-hour minimum time frame (Levskaya et al. 2005; Tabor et al. 2011; Schmid et al. 2014; Fernandez-Rodriguez et al. 2017; Shimizu-Sato et al. 2002; Tabor et al. 2009). For the initial study by Voigt et al. (Levskaya et al. 2005), exposure time of 4 h was reported. As system complexity increased, so too did exposure time, with blue, red, and green light sensors requiring 18 h of projection exposure (FernandezRodriguez et al. 2017) and edge detection needing 10 h of exposure (Tabor et al. 2009). In comparison, systems using LOV sensors have required shorter exposure times. Work from Möglich group has reported a minimum of 2 h required for pDawn activation but has stated that the results are still limited by gene expression in bacteria (Ohlendorf et al. 2012). Incorporating LOV sensors in cell-free systems has greatly reduced the exposure time required, with changes in red fluorescence protein (RFP) production observed in under 2 h (Jayaraman et al. 2018). Integrating photoswitchable azobenzenes in CFPE-based systems has condensed the required exposure time even further, with reported illumination times of 5–10 min and total reaction times of 80 min (Kamiya et al. 2015). The decrease in reaction times observed when moving from in vivo to CFPE systems is quite promising and

12

Cell-Free Biosensors: Synthetic Biology Without Borders

273

highlights the potential for using light-sensitive transcription factors and promoters in future biosensors and synthetic biology applications.

Conclusion and Future Directions These new synthetic, biology- and nanotechnology-based cell-free sensors showcase early, yet compelling examples of the field’s trajectory, which seems to be on track to bring low-cost and decentralized sensing to many domains. While simple cell-free biosensors have been around for decades, the synergistic effects provided by multiple-component systems – such as those highlighted here – offer new levels of sophistication, thus making these systems well-suited for field deployment and for the detection of a wide range of analytes. The community as a whole is striving to operationalize cell-free systems by improving sensitivity, speed, and practical implementation. The results to date are truly promising, with sensitivities reported down to the attomolar and zeptomolar range and reaction times that are unprecedented for portable sensing. Challenges remain and it will be exciting to see this area continue to be a focus of innovation. Thus far, much of the work has been on the development of assays, but as technologies move out of the laboratory setting and toward implementation under field conditions, more comprehensive and universally robust platforms will have to be developed and meet regulatory requirements. This would involve demonstration of sensors that are universally sensitive, accurate, multiplexed, rapid, and easy to operate. Efforts in this field have the potential to provide a driving force for the advancement of application-based synthetic biology and molecular engineering and ultimately contribute to ensuring that the molecular revolution reaches beyond the laboratory.

References Akter F, Yokobayashi Y (2015) RNA signal amplifier circuit with integrated fluorescence output. ACS Synth Biol 4:655–658 Alam KK, Tawiah KD, Lichte MF, Porciani D, Burke DHA (2017) Fluorescent split aptamer for visualizing RNA-RNA assembly in vivo. ACS Synth Biol 6:1710–1721 Alladin-Mustan BS, Mitran CJ, Gibbs-Davis JM (2015) Achieving room temperature DNA amplification by dialling in destabilization. Chem Commun 51:9101–9104 Allen EH, Schwemt RS (1962) Synthesis of hemoglobin in a cell-free system. J Biol Chem 237:760–767 Allen PB, Arshad SA, Li B, Chen X, Ellington AD (2012) DNA circuits as amplifiers for the detection of nucleic acids on a paperfluidic platform. Lab Chip 12:2951–2958 Ando H, Sinha S, Storni R, Aihara K (2011) Synthetic gene networks as potential flexible parallel logic gates. EPL (Europhys Lett) 93:50001 Aw SS, Tang MX, Teo YN, Cohen SM (2016) A conformation-induced fluorescence method for microRNA detection. Nucleic Acids Res 44:e92 Baker BR et al (2006) An electronic, aptamer-based small-molecule sensor for the rapid, label-free detection of cocaine in adulterated samples and biological fluids. J Am Chem Soc 128:3138–3139

274

A. Tinafar et al.

Banica F-G (2012) Chemical sensors and biosensors: fundamentals and applications. Wiley, Chichester Bayer TS, Smolke CD (2005) Programmable ligand-controlled riboregulators of eukaryotic gene expression. Nat Biotechnol 23:337–343 Bhadra S, Ellington AD (2014) A spinach molecular beacon triggered by strand displacement. RNA 20:1183–1194 Brödel AK, Sonnabend A, Kubick S (2014) Cell-free protein expression based on extracts from CHO cells. Biotechnol Bioeng 111:25–36 Buntru M, Vogel S, Spiegel H, Schillberg S (2014) Tobacco BY-2 cell-free lysate: an alternative and highly-productive plant-based in vitro translation system. BMC Biotechnol 14:37 Buntru M, Vogel S, Stoff K, Spiegel H, Schillberg S (2015) A versatile coupled cell-free transcription-translation system based on tobacco BY-2 cell lysates. Biotechnol Bioeng 112:867–878 Burchill SA, Perebolte L, Johnston C, Top B, Selby P (2002) Comparison of the RNA-amplification based methods RT-PCR and NASBA for the detection of circulating tumour cells. Br J Cancer 86:102–109 Burgenson D et al (2018) Rapid recombinant protein expression in cell-free extracts from human blood. Sci Rep 8:9569 Burlage RS, Sayler GS, Larimer F (1990) Monitoring of naphthalene catabolism by bioluminescence with nah-lux transcriptional fusions. J Bacteriol 172:4749–4757 Cameron DE, Bashor CJ, Collins JJ (2014) A brief history of synthetic biology. Nat Rev Microbiol 12:381–390 Cao Y et al (2017) Development of a real-time fluorescence loop-mediated isothermal amplification assay for rapid and quantitative detection of Ustilago maydis. Sci Rep 7:13394 Caschera F, Noireaux V (2014) Synthesis of 2.3 mg/ml of protein with an all Escherichia coli cellfree transcription–translation system. Biochimie 99:162–168 Catherine C, Lee K-H, Oh S-J, Kim D-M (2013) Cell-free platforms for flexible expression and screening of enzymes. Biotechnol Adv 31:797–803 Chandler M, Hayenga K, Heinsohn H (2015) Formulations for drying bacterial extracts Chappell J, Westbrook A, Verosloff M, Lucks JB (2017) Computational design of small transcription activating RNAs for versatile and dynamic gene regulation. Nat Commun 8:1–11 Chen H-W, Ching W-M (2017) Evaluation of the stability of lyophilized loop-mediated isothermal amplification reagents for the detection of Coxiella burnetii. Heliyon 3:e00415 Chen HZ, Zubay G (1983) Prokaryotic coupled transcription-translation. Methods Enzymol 101:674–690 Chen X, Briggs N, McLain JR, Ellington AD (2013a) Stacking nonenzymatic circuits for high signal gain. Proc Natl Acad Sci 110:5386–5391 Chen SX, Zhang DY, Seelig G (2013b) Conditionally fluorescent molecular probes for detecting single base changes in double-stranded DNA. Nat Chem 5. https://doi.org/10.1038/nchem.1713 Chen JS, Ma E, Harrington LB, Da Costa M, Tian X, Palefsky JM et al (2018;eaar6245) CRISPRCas12a target binding unleashes indiscriminate single-stranded DNase activity. Science 6245:1–8 Choudhury A, Hodgman CE, Anderson MJ, Jewett MC (2014) Evaluating fermentation effects on cell growth and crude extract metabolic activity for improved yeast cell-free protein synthesis. Biochem Eng J 91:140–148 Clark LC, Lyons C (2006) Electrode systems for continuous monitoring in cardiovascular surgery. Ann N Y Acad Sci 102:29–45 Cordray MS, Richards-Kortum RR (2012) Emerging nucleic acid-based tests for point-of-care detection of malaria. Am J Trop Med Hyg 87:223–230 Crannell ZA, Rohrman B, Richards-Kortum R (2014) Equipment-free incubation of recombinase polymerase amplification reactions using body heat. PLoS One 9:e112146 Cunningham JC, DeGregory PR, Crooks RM (2016) New functionalities for paper-based sensors lead to simplified user operation, lower limits of detection, and new applications. Annu Rev Anal Chem 9:183–202 Curtis KA et al (2012) Isothermal amplification using a chemical heating device for point-of-care detection of HIV-1. PLoS One 7:e31432

12

Cell-Free Biosensors: Synthetic Biology Without Borders

275

DeVries JK, Zubay G (1967) DNA-directed peptide synthesis. II. The synthesis of the alpha-fragment of the enzyme beta-galactosidase. Proc Natl Acad Sci U S A 57:1010–1012 Didovyk A, Tonooka T, Tsimring L, Hasty J (2017) Rapid and scalable preparation of bacterial lysates for cell-free gene expression. ACS Synth Biol 6:2198–2208 Dirks RM, Pierce NA (2004) Triggered amplification by hybridization chain reaction. Proc Natl Acad Sci U S A 101:15275–15278 Doherty EA, Doudna JA (2000) Ribozyme structures and mechanisms. Annu Rev Biochem 69:597–615 Dolgosheina EV et al (2014) RNA mango aptamer-fluorophore: a bright, high-affinity complex for RNA labeling and tracking. ACS Chem Biol 9:2412–2420 Duyen TTM et al (2016) Paper-based colorimetric biosensor for antibiotics inhibiting bacterial protein synthesis. J Biosci Bioeng 123:96–100 Elbaz J et al (2010) DNA computing circuits using libraries of DNAzyme subunits. Nat Nanotechnol 5:417 Engelen W, Meijer LHH, Somers B, de Greef TFA, Merkx M (2017) Antibody-controlled actuation of DNA-based molecular circuits. Nat Commun 8:14473 Ezure T et al (2006) Cell-free protein synthesis system prepared from insect cells by freeze-thawing. Biotechnol Prog 22:1570–1577 Failmezger J, Scholz S, Blombach B, Siemann-Herzberg M (2018) Cell-free protein synthesis from fast-growing vibrio natriegens. Front Microbiol 9:1146 Fan C, Plaxco KW, Heeger AJ (2003) Electrochemical interrogation of conformational changes as a reagentless method for the sequence-specific detection of DNA. Proc Natl Acad Sci 100:9134–9137 Fenz SF, Sachse R, Schmidt T, Kubick S (2014) Cell-free synthesis of membrane proteins: tailored cell models out of microsomes. Biochim Biophys Acta Biomembr 1838:1382–1388 Fernandez-Rodriguez J, Moser F, Song M, Voigt CA (2017) Engineering RGB color vision into Escherichia coli. Nat Chem Biol 13:706+ Ferrari M (2006) Biomolecular sensing, processing and analysis. BioMEMS Biomed. Nanotechnol. IV Fossati E et al (2014) Reconstitution of a 10-gene pathway for synthesis of the plant alkaloid dihydrosanguinarine in Saccharomyces cerevisiae. Nat Commun 5:3283 Gale EF, Folkes JP (1954) Effect of nucleic acids on protein synthesis and amino-acid incorporation in disrupted staphylococcal cells. Nature 173:1223–1227 Garamella J, Marshall R, Rustad M, Noireaux V (2016) The all E. coli TX-TL toolbox 2.0: a platform for cell-free synthetic biology. ACS Synth Biol 5:344–355 Gootenberg JS et al (2017) Nucleic acid detection with CRISPR-Cas13a/C2c2. Science 356:438–442 Gootenberg JS et al (2018) Multiplexed and portable nucleic acid detection platform with Cas13, Cas12a, and Csm6. Science 360:439–444 Goto M, Honda E, Ogura A, Nomoto A, Hanaki K-I (2009) Colorimetric detection of loop-mediated isothermal amplification reaction by using hydroxy naphthol blue. BioTechniques 46:167–172 Green AA, Silver PA, Collins JJ, Yin P (2014) Toehold switches: de-novo-designed regulators of gene expression. Cell 159:925–939 Green AA et al (2017) Complex cellular logic computation using ribocomputing devices. Nature 548:117–121 Guatelli JC et al (1990) Isothermal, in vitro amplification of nucleic acids by a multienzyme reaction modeled after retroviral replication. Proc Natl Acad Sci U S A 87:1874–1878 Hamaguchi N, Ellington A, Stanton M (2001) Aptamer beacons for the direct detection of proteins. Anal Biochem 294:126–131 Harbers M (2014) Wheat germ systems for cell-free protein expression. FEBS Lett 588:2762–2773 Heyduk E, Dummit B, Chang Y-H, Heyduk T (2008) Molecular pincers: antibody-based homogeneous protein sensors. Anal Chem 80:5152–5159 Hollenstein M, Hipolito C, Lam C, Dietrich D, Perrin DM (2008) A highly selective DNAzyme sensor for mercuric ions. Angew Chem Int Ed 47:4346–4350

276

A. Tinafar et al.

Hongwarittorrn I, Chaichanawongsaroj N, Laiwattanapaisal W (2017) Semi-quantitative visual detection of loop mediated isothermal amplification (LAMP)-generated DNA by distancebased measurement on a paper device. Talanta 175:135–142 Hu J et al (2014) Advances in paper-based point-of-care diagnostics. Biosens Bioelectron 54:585–597 Huang H et al (2014) A G-quadruplex–containing RNA activates fluorescence in a GFP-like fluorophore. Nat Chem Biol 10:686 Huang K et al (2017) FASTmiR: an RNA-based sensor for in vitro quantification and live-cell localization of small RNAs. Nucleic Acids Res 45:e130 Isaacs FJ et al (2004) Engineered riboregulators enable post-transcriptional control of gene expression. Nat Biotechnol 22:841–847 Ishikawa K, Sato K, Shima Y, Urabe I, Yomo T (2004) Expression of a cascading genetic network within liposomes. FEBS Lett 576:387–390 Jackson RJ, Hunt T (1983) Preparation and use of nuclease-treated rabbit reticulocyte lysates for the translation of eukaryotic messenger RNA. Methods Enzymol 96:50–74 Janssen BMG et al (2013) Reversible blocking of antibodies using bivalent peptide-DNA conjugates allows protease-activatable targeting. Chem Sci 4:1442–1450 Janssen BMG, van Rosmalen M, van Beek L, Merkx M (2015) Antibody activation using DNA-based logic gates. Angew Chem Int Ed 54:2530–2533 Jayaraman P et al (2018) Cell-free optogenetic gene expression system. ACS Synth Biol 7:986–994 Jewett MC, Swartz JR (2004) Mimicking the Escherichia coli cytoplasmic environment activates long-lived and efficient cell-free protein synthesis. Biotechnol Bioeng 86:19–26 Jewett MC, Calhoun KA, Voloshin A, Wuu JJ, Swartz JR (2008) An integrated cell-free metabolic platform for protein production and synthetic biology. Mol Syst Biol 4:220 Kamiya Y et al (2015) Synthetic gene involving Azobenzene-Tethered T7 promoter for the photocontrol of gene expression by visible light. ACS Synth Biol 4:365–370 Karig DK (2017) Cell-free synthetic biology for environmental sensing and remediation. Curr Opin Biotechnol 45:69–75 Karig DK, Bessling S, Thielen P, Zhang S, Wolfe J (2017) Preservation of protein expression systems at elevated temperatures for portable therapeutic production. J R Soc Interface. pii: 20161039. https://doi.org/10.1098/rsif.2016.1039. PMID: 28446704 Kelwick R, Webb AJ, MacDonald JT, Freemont PS (2016) Development of a Bacillus subtilis cellfree transcription-translation system for prototyping regulatory elements. Metab Eng 38:370–381 Kigawa T et al (2004) Preparation of Escherichia coli cell extract for highly productive cell-free protein expression. J Struct Funct Genom 5:63–68 Kikuchi N, Kolpashchikov DM (2016) Split spinach aptamer for highly selective recognition of DNA and RNA at ambient temperatures. Chembiochem 17:1589–1592 Kim T-W et al (2006) Simple procedures for the construction of a robust and cost-effective cell-free protein synthesis system. J Biotechnol 126:554–561 Kjelstrup MV, Nielsen LDF, Hansen-Bruhn M, Gothelf KV (2018) A DNA-based assay for digoxin detection. Biosens Bioelectron 8:19–30 Kobayashi H et al (2004) Programmable cells: interfacing natural and engineered gene networks. Proc Natl Acad Sci U S A 101:8414–8419 Kolpashchikov DM (2005) Binary malachite green aptamer for fluorescent detection of nucleic acids. J Am Chem Soc 127:12442–12443 Kotula JW et al (2014) Programmable bacteria detect and record an environmental signal in the mammalian gut. Proc Natl Acad Sci U S A 111:4838–4843 Kovtun O, Mureev S, Johnston W, Alexandrov K (2010) Towards the construction of expressed proteomes using a Leishmania tarentolae based cell-free expression system. PLoS One 5: e14388 Kovtun O et al (2011) Leishmania cell-free protein expression system. Methods 55:58–64

12

Cell-Free Biosensors: Synthetic Biology Without Borders

277

Kwon Y-C, Jewett MC (2015) High-throughput preparation methods of crude extract for robust cell-free protein synthesis. Sci Rep 5:8663 Lafleur LK et al (2016) A rapid, instrument-free, sample-to-result nucleic acid amplification test. Lab Chip 16:3777–3787 Lavickova B, Maerkl SJ (2019) A simple, robust, and low-cost method to produce the PURE cell-free system. ACS Synth Biol. 8(2):455–462. https://pubs.acs.org/doi/10.1021/acssynbio. 8b00427 Levskaya A et al (2005) Engineering Escherichia coli to see light – these smart bacteria `photograph’ a light pattern as a high-definition chemical image. Nature 438:441–442 Li J, Wang H, Kwon Y-C, Jewett MC (2017a) Establishing a high yielding streptomyces – based cell-free protein synthesis system. Biotechnol Bioeng 114:1343–1353 Li Y, Fan P, Zhou S, Zhang L (2017b) Loop-mediated isothermal amplification (LAMP): a novel rapid detection platform for pathogens. Microb Pathog 107:54–61 Li J, Wang H, Jewett MC (2018) Expanding the palette of Streptomyces -based cell-free protein synthesis systems with enhanced yields. Biochem Eng J 130:29–33 Liu DV, Zawada JF, Swartz JR (2005) Streamlining Escherichia coli S30 extract preparation for economical cell-free protein synthesis. Biotechnol Prog 21:460–465 Lobato IM, O’Sullivan CK (2018) Recombinase polymerase amplification: basics, applications and recent advances. TrAC Trends Anal Chem 98:19–35 Lopez R, Wang R, Seelig G (2018) A molecular multi-gene classifier for disease diagnostics. Nat Chem 10:746–754 Lucks JB, Qi L, Mutalik VK, Wang D, Arkin AP (2011) Versatile RNA-sensing transcriptional regulators for engineering genetic networks. Proc Natl Acad Sci 108:8617–8622 Martin RW et al (2017) Development of a CHO-based cell-free platform for synthesis of active monoclonal antibodies. ACS Synth Biol 6:1370–1379 Matthaei JH, Nirenberg MW (1961) Characteristics and stabilization of DNAase-sensitive protein synthesis in E. coli extracts. Proc Natl Acad Sci U S A 47:1580–1588 Matthaei JH, Jones OW, Martin RG, Nirenberg MW (1962) Characteristics and composition of RNA coding units. Proc Natl Acad Sci U S A 48:666–677 Mehrotra P (2016) Biosensors and their applications – a review. J Oral Biol Craniofacial Res 6:153–159 Mikami S, Masutani M, Sonenberg N, Yokoyama S, Imataka H (2006) An efficient mammalian cell-free translation system supplemented with translation factors. Protein Expr Purif 46:348–357 Moglich A, Ayers RA, Moffat K (2009) Design and signaling mechanism of light-regulated histidine kinases. J Mol Biol 385:1433–1444 Moore MD, Jaykus L-A (2017) Development of a recombinase polymerase amplification assay for detection of epidemic human noroviruses. Sci Rep 7:40244 Moore SJ, Lai H-E, Needham H, Polizzi KM, Freemont PS (2017) Streptomyces venezuelae TX-TL – a next generation cell-free synthetic biology tool. Biotechnol J 12:1600678 Moore SJ et al (2018) Rapid acquisition and model-based analysis of cell-free transcription–translation reactions from nonmodel bacteria. Proc Natl Acad Sci 115:E4340–E4349 Mullis KB, Faloona FA (1987) Specific synthesis of DNA in vitro via a polymerase-catalyzed chain reaction. Methods Enzymol 155:335–350 Mullis K et al (1986) Specific enzymatic amplification of DNA in vitro: the polymerase chain reaction. Cold Spring Harb Symp Quant Biol 51(Pt 1):263–273 Mutalik VK, Qi L, Guimaraes JC, Lucks JB, Arkin AP (2012) Rationally designed families of orthogonal RNA regulators of translation. Nat Chem Biol 8:447–454 Myhrvold C et al (2018) Field-deployable viral diagnostics using CRISPR-Cas13. Science 360:444–448 Nirenberg MW, Matthaei JH (1961) The dependence of cell-free protein synthesis in E. coli upon naturally occurring or synthetic polyribonucleotides. Proc Natl Acad Sci U S A 47:1588–1602

278

A. Tinafar et al.

Niu S, Jiang Y, Zhang S (2010) Fluorescence detection for DNA using hybridization chain reaction with enzyme-amplification. Chem Commun 46:3089–3091 Noireaux V, Libchaber A (2004) A vesicle bioreactor as a step toward an artificial cell assembly. Proc Natl Acad Sci U S A 101:17669–17674 Notomi T et al (2000) Loop-mediated isothermal amplification of DNA. Nucleic Acids Res 28:E63 Nourian Z, Danelon C (2013) Linking genotype and phenotype in protein synthesizing liposomes with external supply of resources. ACS Synth Biol 2:186–193 Nyan D-C, Swinson KL (2016) A novel multiplex isothermal amplification method for rapid detection and identification of viruses. Sci Rep 5:17925 Ohlendorf R, Vidavski RR, Eldar A, Moffat K, Moeglich A (2012) From dusk till dawn: one-plasmid systems for light-regulated gene expression. J Mol Biol 416:534–542 Ong WQ, Citron YR, Sekine S, Huang B (2017) Live cell imaging of endogenous mRNA using RNA-based fluorescence ‘Turn-On’ probe. ACS Chem Biol 12:200–205 Orbach R, Remacle F, Levine RD, Willner I (2012) Logic reversibility and thermodynamic irreversibility demonstrated by DNAzyme-based Toffoli and Fredkin logic gates. Proc Natl Acad Sci 109:21228–21233 Oriero EC, Jacobs J, Van Geertruyden JP, Nwakanma D, D’Alessandro U (2015) Molecular-based isothermal tests for field diagnosis of malaria and their potential contribution to malaria elimination. J Antimicrob Chemother 70:2–13 Paige JS, Wu KY, Jaffrey SR (2011) RNA mimics of green fluorescent protein. Science 333:642–646 Paige JS, Nguyen-Duc T, Song W, Jaffrey SR (2012) Fluorescence imaging of cellular metabolites with RNA. Science 335:1194 Pardee K et al (2014) Paper-based synthetic gene networks. Cell 159:940–954 Pardee K et al (2016) Rapid, low-cost detection of Zika virus using programmable biomolecular components in brief. Cell 165(5):1255–1266 Pham HM, Nakajima C, Ohashi K, Onuma M (2005) Loop-mediated isothermal amplification for rapid detection of Newcastle disease virus. J Clin Microbiol 43:1646–1650 Piepenburg O, Williams CH, Stemple DL, Armes NA (2006) DNA detection using recombination proteins. PLoS Biol 4:e204 Porchetta A, Vallée-Bélisle A, Plaxco KW, Ricci F (2013) Allosterically tunable, DNA-based switches triggered by heavy metals. J Am Chem Soc 135:13238–13241 Porchetta A et al (2018) Programmable nucleic acid nanoswitches for the rapid, single-step detection of antibodies in bodily fluids. J Am Chem Soc 140:947–953 Rawson DM, Willmer AJ, Turner AP (1989) Whole-cell biosensors for environmental monitoring. Biosensors 4:299–311 Riglar DT et al (2017) Engineered bacteria can function in the mammalian gut long-term as live diagnostics of inflammation. Nat Biotechnol 35:653–658 Rodrigo G, Landrain TE, Jaramillo A (2012) De novo automated design of small RNA circuits for engineering synthetic riboregulation in living cells. Proc Natl Acad Sci 109:15271–15276 Rodriguez NM et al (2015) Paper-based RNA extraction, in situ isothermal amplification, and lateral flow detection for low-cost, rapid diagnosis of Influenza A (H1N1) from clinical specimens. Anal Chem 87:7872–7879 Proc Natl Acad Sci U S A. (2009) 106(36):15514 PMID: 17709748 Rooney JE, Gurpur PB, Burkin DJ (2009) Correction for win and smolke, a modular and extensible RNA-based gene-regulatory platform for engineering cellular function. Proc Natl Acad Sci 106:15514 PMID: 17709748 Rosenblum G, Cooperman BS (2014) Engine out of the chassis: cell-free protein synthesis and its uses. FEBS Lett 588:261–268 Salehi ASM et al (2017) Cell-free protein synthesis approach to biosensing hTRβ-specific endocrine disruptors. Anal Chem 89:3395–3401 Salehi ASM et al (2018) Biosensing estrogenic endocrine disruptors in human blood and urine: a RAPID cell-free protein synthesis approach. Toxicol Appl Pharmacol 345:19–25

12

Cell-Free Biosensors: Synthetic Biology Without Borders

279

Sato S et al (2015) Live-cell imaging of endogenous mRNAs with a small molecule. Angew Chemie-International Ed 54:1855–1858 Schmid SR, Sheth RU, Wu A, Tabor JJ (2014) Refactoring and optimization of light-switchable Escherichia coli two-component systems. ACS Synth Biol 3:820–831 Schweet R, Lamfrom H, Allen E (1958) The synthesis of hemoglobin in a cell-free system. Proc Natl Acad Sci U S A 44:1029–1035 Seok Y et al (2017) A paper-based device for performing loop-mediated isothermal amplification with real-time simultaneous detection of multiple DNA targets. Theranostics 7:2220–2230 Shimizu Y et al (2001) Cell-free translation reconstituted with purified components. Nat Biotechnol 19:751–755 Shimizu-Sato S, Huq E, Tepperman JM, Quail PH (2002) A light-switchable gene promoter system. Nat Biotechnol 20:1041–1044 Shin J, Noireaux V (2012) An E. coli cell-free expression toolbox: application to synthetic gene circuits and artificial cells. ACS Synth Biol 1:29–41 Slomovic S, Pardee K, Collins JJ (2015) Synthetic biology devices for in vitro and in vivo diagnostics. PNAS 112:14429–14435 Smith MT, Berkheimer SD, Werner CJ, Bundy BC (2014) Lyophilized Escherichia coli-based cellfree systems for robust, high-density, long-term storage. BioTechniques 56:186–193 Song W, Strack RL, Jaffrey SR (2013) Imaging bacterial protein expression using genetically encoded RNA sensors. Nat Methods 10:873–875 Song W et al (2017) Imaging RNA polymerase III transcription using a photostable RNA-fluorophore complex. Nat Chem Biol 13:1187–1194 Strack RL, Song W, Jaffrey SR (2014) Using spinach-based sensors for fluorescence imaging of intracellular metabolites and proteins in living bacteria. Nat Protoc 9:146–155 Sun ZZ et al (2013) Protocols for implementing an Escherichia coli based TX-TL cell-free expression system for synthetic biology. J Vis Exp: e50762. https://doi.org/10.3791/50762 Svensen N, Jaffrey SR (2016) Fluorescent RNA aptamers as a tool to study RNA-modifying enzymes. Cell Chem Biol 23:415–425 Tabor JJ et al (2009) A synthetic genetic edge detection program. Cell 137:1272–1281 Tabor JJ, Levskaya A, Voigt CA (2011) Multichromatic control of gene expression in Escherichia coli. J Mol Biol 405:315–324 Takahashi MK et al (2015) Rapidly characterizing the fast dynamics of RNA genetic circuitry with cell-free transcription-translation (TX-TL) systems. ACS Synth Biol 4:503–515 Torella JP et al (2013) Tailored fatty acid synthesis via dynamic control of fatty acid elongation. Proc Natl Acad Sci U S A 110:11290–11295 Tran K et al (2018) Cell-free production of a therapeutic protein: expression, purification, and characterization of recombinant streptokinase using a CHO lysate. Biotechnol Bioeng 115:92–102 Trang PTK, Berg M, Viet PH, Van Mui N, Van Der Meer JR (2005) Bacterial bioassay for rapid and accurate analysis of arsenic in highly variable groundwater samples. Environ Sci Technol 39:7625–7630 Tuckey C, Asahara H, Zhou Y, Chong S (2014) Protein synthesis using a reconstituted cell-free system. Curr Protoc Mol Biol 108:16.31.1–16.3122 Turner APF, Karube I, Wilson GS (1987) Biosensors: fundamentals and applications. Oxford University Press, New York van der Meer JR, Belkin S (2010) Where microbiology meets microengineering: design and applications of reporter bacteria. Nat Rev Microbiol 8:511–522 Villarreal F et al (2018) Synthetic microbial consortia enable rapid assembly of pure translation machinery. Nat Chem Biol 14:29–35 Wang JS, Zhang DY (2015) Simulation-guided DNA probe design for consistently ultraspecific hybridization. Nat Chem 7:545 Wang F, Elbaz J, Teller C, Willner I (2011a) Amplified detection of DNA through an autocatalytic and catabolic DNAzyme-mediated process. Angew Chem Int Ed 50:295–299

280

A. Tinafar et al.

Wang F, Elbaz J, Orbach R, Magen N, Willner I (2011b) Amplified analysis of DNA by the autonomous assembly of polymers consisting of DNAzyme wires. J Am Chem Soc 133:17149–17151 Wang JS, Yan YH, Zhang DY (2017) Modular probes for enriching and detecting complex nucleic acid sequences. Nat Chem 9:1222 Wang H, Li J, Jewett MC (2018) Development of a Pseudomonas putida cell-free protein synthesis platform for rapid screening of gene regulatory elements. Synth Biol 3(1):ysy003. https:// academic.oup.com/synbio/article/3/1/ysy003/4994521 Wen KY et al (2017) A cell-free biosensor for detecting quorum sensing molecules in P. Aeruginosa-infected respiratory samples. ACS Synth Biol 6:2293–2301 Wiegand DJ, Lee HH, Ostrov N, Church GM (2018) Establishing a cell-free Vibrio natriegens expression system. bioRxiv 331645. https://doi.org/10.1101/331645 Xia F et al (2010) An electrochemical supersandwich assay for sensitive and selective DNA detection in complex matrices. J Am Chem Soc 132:14346–14348 Xiao Y, Lubin AA, Heeger AJ, Plaxco KW (2005) Label-free electronic detection of thrombin in blood serum by using an aptamer-based sensor. Angew Chem 117:5592–5595 Xiao Y, Rowe AA, Plaxco KW (2007a) Electrochemical detection of parts-per-billion lead via an electrode-bound DNAzyme assembly. J Am Chem Soc 129:262–263 Xiao Y, Lai RY, Plaxco KW (2007b) Preparation of electrode-immobilized, redox-modified oligonucleotides for electrochemical DNA and aptamer-based sensing. Nat Protoc 2:2875 Xing W et al (2017) Field evaluation of a recombinase polymerase amplification assay for the diagnosis of Schistosoma japonicum infection in Hunan province of China. BMC Infect Dis 17:164 Xu W, Deng R, Wang L, Li J (2014) Multiresponsive rolling circle amplification for DNA logic gates mediated by endonuclease. Anal Chem 86:7813–7818 Yamanaka ES, Tortajada-Genaro LA, Pastor N, Maquieira Á (2018) Polymorphism genotyping based on loop-mediated isothermal amplification and smartphone detection. Biosens Bioelectron 109:177–183 Yang WC, Patel KG, Wong HE, Swartz JR (2012) Simplifying and streamlining Escherichia colibased cell-free protein synthesis. Biotechnol Prog 28:413–420 Yen L et al (2004) Exogenous control of mammalian gene expression through modulation of RNA self-cleavage. Nature 431:471–476 Yin P, Choi HMT, Calvert CR, Pierce NA (2008) Programming biomolecular self-assembly pathways. Nature 451:318 Ying ZM, Wu Z, Tu B, Tan W, Jiang JH (2017) Genetically encoded fluorescent RNA sensor for ratiometric imaging of MicroRNA in living tumor cells. J Am Chem Soc 139:9779–9782 You M, Litke JL, Jaffrey SR (2015) Imaging metabolite dynamics in living cells using a spinachbased riboswitch. Proc Natl Acad Sci U S A 112:E2756–E2765 Yu H, Canoura J, Guntupalli B, Lou X, Xiao Y (2017) A cooperative-binding split aptamer assay for rapid, specific and ultra-sensitive fluorescence detection of cocaine in saliva. Chem Sci 8:131–141 Zamecnik PC, Frantz ID (1948) Incorporation in vitro of radioactive carbon from carboxyl-labeled dl-alanine and glycine into proteins of normal and malignant rat livers. J Biol Chem 175:299–314 Zawada J, Swartz J (2006) Effects of growth rate on cell extract performance in cell-free protein synthesis. Biotechnol Bioeng 94:618–624 Zhang DY, Seelig G (2011) Dynamic DNA nanotechnology using strand-displacement reactions. Nat Chem 3:103 Zhang X-B, Kong R-M, Lu Y (2011) Metal ion sensors based on DNAzymes and related DNA molecules. Annu Rev Anal Chem 4:105–128 Zhang F, Carothers JM, Keasling JD (2012a) Design of a dynamic sensor-regulator system for production of chemicals and fuels derived from fatty acids. Nat Biotechnol 30:354–359

12

Cell-Free Biosensors: Synthetic Biology Without Borders

281

Zhang DY, Chen SX, Yin P (2012b) Optimizing the specificity of nucleic acid hybridization. Nat Chem 4:208 Zhang Z, Hejesen C, Kjelstrup MB, Birkedal V, Gothelf KV (2014) A DNA-mediated homogeneous binding assay for proteins and small molecules. J Am Chem Soc 136:11115–11120 Zhao Y, Chen F, Li Q, Wang L, Fan C (2015) Isothermal amplification of nucleic acids. Chem Rev 115:12491–12545 Zhou W, Saran R, Liu J (2017) Metal sensing by DNA. Chem Rev 117:8272–8325 Zubay G (1973) In vitro synthesis of protein in microbial systems. Annu Rev Genet 7:267–287 Zuo X, Xiao Y, Plaxco KW (2009) High specificity, electrochemical sandwich assays based on single aptamer sequences and suitable for the direct detection of small-molecule targets in blood and other complex matrices. J Am Chem Soc 131:6944–6945

Engineering Prokaryote Synthetic Biology Biosensors

13

Xinyi Wan, Trevor Y. H. Ho, and Baojun Wang

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Synthetic Biology as an Enabling Platform for Rapid Construction and Optimization of Prokaryotic Biosensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A Streamlined Approach to Developing Novel Prokaryotic Biosensors . . . . . . . . . . . . . . . . . . . Efficient Sensor Optimization by Standardized and Modularized Genetic Parts . . . . . . . . . . Biosensor Improvement by Directed Evolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Development of New Sensing Modules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tools and Strategies from Synthetic Biology for Optimizing Biosensor Performance . . . . . . . . Properties of a Biosensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Strategies for Enhancing Selectivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Strategies for Lowering the Limit of Detection (LOD) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Strategies for Increasing Output Dynamic Range . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Strategies for Reducing Leakiness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Functional Expansion of Biosensors by Synthetic Biology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Memory Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Computation Modules to Integrate Signals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Modules to Reshape Response Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Reporter Modules for Interfacing with Different Detection Platforms . . . . . . . . . . . . . . . . . . . . . Biosafety Enhancing Modules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions and Future Directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

284 285 285 288 288 289 290 290 291 293 295 298 301 301 305 306 307 307 308 309

Abstract

Prokaryotic whole-cell biosensors are effective yet inexpensive, renewable, and environmentally friendly substitutes for many contemporary sensors and diagnostic devices. Unfortunately, many previously constructed prokaryotic biosensors are plagued by poor performance, which prohibits their use in real-life X. Wan · T. Y. H. Ho · B. Wang (*) School of Biological Sciences, University of Edinburgh, Edinburgh, UK Centre for Synthetic and Systems Biology, University of Edinburgh, Edinburgh, UK e-mail: [email protected] © Springer Nature Switzerland AG 2022 G. Thouand (ed.), Handbook of Cell Biosensors, https://doi.org/10.1007/978-3-030-23217-7_131

283

284

X. Wan et al.

applications. Biosafety concerns, promiscuous analyte detection, limited detection range, low output dynamic range, and high background expression are common issues. Engineering principles and strategies from the emerging field of synthetic biology offer unprecedented solutions. They accelerate new biosensor development and improve biosensor behavior. Addition of novel devices and modules from synthetic biology further augments functions beyond sensing and renders them safer. Thanks to this, prokaryotic whole-cell biosensors have enjoyed a renaissance in recent years, and they hold promise to address the increasing demand for marketable biosensors. Keywords

Whole-cell biosensor · Synthetic biology · Prokaryotes · Response curve · Limit of detection · Selectivity · Dynamic range · Leakiness

Introduction Whole-cell biosensors are domesticated or engineered cells that detect and report a target or condition of interest (Daunert et al. 2000; van der Meer and Belkin 2010; Wang and Buck 2012; Gui et al. 2017). They are viable alternatives to electronic or chemical sensors and have drawn increasing attention over the last three decades. Whole-cell biosensors are biodegradable and can be mass produced using inexpensive nutrients. So they are renewable, environmentally friendly, and cost-effective (van der Meer and Belkin 2010; Kim et al. 2018). Both prokaryotic and eukaryotic biosensors have been developed. This chapter limits its scope to only prokaryotic sensors, but the principles covered should also apply to their eukaryotic counterparts. Prokaryotic biosensors have been researched for various purposes, for instances, environment assessment (Stocker et al. 2003; Wang et al. 2013; Huang et al. 2015b; Hwang et al. 2016; Kim et al. 2016; Cayron et al. 2017), clinical diagnosis (Duan and March 2010; Saeidi et al. 2011; Archer et al. 2012; Gupta et al. 2013; Hwang et al. 2014; Kotula et al. 2014; Courbet et al. 2015; Danino et al. 2015; Cayron et al. 2017; Daeffler et al. 2017; Riglar et al. 2017; Ho et al. 2018; Watstein and Styczynski 2018), and controlled bioprocessing (Zhang and Keasling 2011; Zhang et al. 2012). Some less common applications include mineral surveying and landmine clearing (Cerminati et al. 2011; Belkin et al. 2017). Despite their advantages and demonstrated successes, many prokaryotic whole-cell biosensors fail to survive in or even reach the competitive biosensor market. A recurring concern is biosafety – whole-cell biosensors are often subjected to higher levels of legal and ethical scrutiny because there is a risk of releasing genetically engineered microorganisms into the wild (Dana et al. 2012). Yet with biosafety aside, many prokaryotic biosensors are still leaky with low output dynamic ranges and suffer from unsatisfactory detection limit and selectivity (Stocker et al. 2003; Amaro et al. 2011; De Mora et al. 2011; Siegfried et al. 2012; Wang et al. 2013; Huang et al. 2015a;

13

Engineering Prokaryote Synthetic Biology Biosensors

285

Kim et al. 2016; Merulla and van der Meer 2016). Nevertheless, recent advancements in synthetic biology have provided numerous solutions. Synthetic biology is the rational design of biological systems. It achieves so by applying established principles from engineering to biology and so is highly interdisciplinary (Endy 2005; Purnick and Weiss 2009; Cheng and Lu 2012; Wang and Buck 2012; Way et al. 2014; Bradley et al. 2016a, b; Bashor and Collins 2018; Xiang et al. 2018). These principles include: (1) Abstraction: complexity is managed through a hierarchy and less relevant information is hidden for clarity. (2) Standardization: genetic elements are standardized into reusable parts with measurable parameters that can be ranked on a common scale. (3) Modularization: genetic parts and devices are independent and interchangeable modules with defined functions, and (4) Rational and quantitative design: behavior of biological systems can be described and predicted by mathematical models. These principles and concepts enabled synthetic biologists to develop new strategies and devices, which substantially enhanced the performance of existing prokaryotic biosensors. In addition, they paved the way for developing novel biosensors with augmented capabilities. This chapter thus focuses on how synthetic biology facilitates biosensor development, creates tools for improving biosensors, and expands their capacities beyond sensing.

Synthetic Biology as an Enabling Platform for Rapid Construction and Optimization of Prokaryotic Biosensors A Streamlined Approach to Developing Novel Prokaryotic Biosensors Before the advent of genetic engineering, the development of a biosensor relied much on serendipity. Many biosensors were by-products from studying the stress responses or metabolic pathways of microbes. For instances, Microtox, one of the earliest biosensors, is a bioluminescent bacterium that glows weaker when its metabolism is disturbed by toxic chemicals (Bulich and Isenberg 1981), and a whole-cell naphthalene biosensor was developed by random transposition of a luciferase reporter into a naphthalene degradation pathway isolated from a soil bacterial species (King et al. 1990). In these examples, the biosensor development processes depended on the fortuitous discovery of a species or a strain with desirable responses towards the targets, and they were hardly generalizable or readily reproducible. In contrast, synthetic biology offered a formalized approach for biosensor development. A biosensor can be abstracted as a processor with a sensing module, a processing module, and an actuating module (Fig. 1a) (van der Meer and Belkin 2010; Wang and Buck 2012; Bradley and Wang 2015; Bernard and Wang 2017; Kim et al. 2018). Any naturally occurring response pathway in a prokaryote can also be dissected and classified in a similar fashion. Therefore, development of a biosensor is reduced to a task of identifying or creating an input module that can respond to the target and rewiring it to an observable output.

286

X. Wan et al.

Fig. 1 Architecture and engineering of synthetic biosensors. (a) Architecture of a modular synthetic biosensor. R, receptor. P, promoter. gfp, gene encoding a green fluorescent protein. rfp, gene encoding a red fluorescent protein. luxAB, genes encoding a bacterial luciferase for

13

Engineering Prokaryote Synthetic Biology Biosensors

287

This framework is illustrated by the classical example of the whole-cell biosensor for detecting arsenic in drinking water. Escherichia coli naturally protect themselves from high concentrations of arsenic through its endogenous arsenic resistant pathway (Fig. 1b) (Cervantes et al. 1994; Nealson et al. 2002; Silver and Phung 2005; Andres and Bertin 2016). When bounded by arsenite (III), the ArsR transcription factor would de-repress its cognate promoter ParsR which then triggers expression of molecular pumps that remove arsenite from the cell (Shi et al. 1996; Silver and Phung 2005; Chen and Rosen 2014; Saha et al. 2017). The ArsR-ParsR transcriptional regulation can thus be understood as a sensing module and the genes coding for the arsenic efflux pumps as an actuating module. The natural sensing module can then be isolated and wired to a downstream reporter gfp, and this reengineered E. coli would give a higher level of green fluorescence under an increased level of arsenic, serving the purpose of a biosensor (Fig. 1d). This framework is universal and can be adopted to quickly build new prokaryotic biosensors, because many other sensing modules have also been characterized. The majority of these sensing modules worked through either an allosterically controlled transcriptional regulator or a bacterial two-component system. The most wellstudied examples for the former category include modules for detecting metal ions like mercury (Fig. 1c and e) and nutrients such as arabinose and lactose (Misra et al. 1985; Barkay et al. 2003; van der Meer and Belkin 2010; Mahr and Frunzke 2016). Others include modules for reporting quorum sensing molecules that could indicate infections caused by pathogens or inflammation biomarkers (Saeidi et al. 2011; Archer et al. 2012; Gupta et al. 2013; Hwang et al. 2014). There are also examples for hypoxia responsiveness (Anderson et al. 2006; Yu et al. 2012), aromatic contamination, and DNT/TNT from landmines (Selifonova and Eaton 1996; Belkin et al. 2017). For two-component systems, notable modules include sensors for detecting green, red, and blue light, respectively (Olson et al. 2014; FernandezRodriguez et al. 2017), as well as sensors for zinc and lead (Fig. 3a) (Wang et al. 2013) and the aforementioned sensors for inflammation (Daeffler et al. 2017; Riglar et al. 2017). For both categories of sensing modules, the output from the module is almost always a transcriptional output. Therefore, they can be conveniently connected to an output module in a “plug-and-play” fashion to drive reporter gene expression.

ä Fig. 1 (continued) luminescent output. lacZ, gene coding β-galactosidase for colorimetric output. arg, acoustic reporter genes which express gas vehicles that are detectable by ultrasound. luxI and lasI, genes encoding synthases for quorum sensing molecules. (b) ars operon from E. coli’s chromosome and its role in arsenic regulation (Silver and Phung 2005; Chen and Rosen 2014). GlpF, an aquaglyceroporin. PST, phosphate-specific transport system. PIT, phosphate inorganic transport system. (c) mer operon from Shigella flexneri R100 plasmid and its role in mercury regulation (Misra et al. 1985; Barkay et al. 2003). CH3HgX, organic form of mercury. HgX, inorganic form of mercury. (d) An engineered E. coli biosensor with arsenic sensing and reporting function. (e) An engineered E. coli biosensor with mercury sensing and reporting function

288

X. Wan et al.

Efficient Sensor Optimization by Standardized and Modularized Genetic Parts In biotechnology, whole-cell mutagenesis remains the canonical technique to improve the characteristics of a strain. While it is still widely used in whole-cell biosensor optimization, they have further benefited from the modularization of genetic elements. The dissection of gene circuitries into stand-alone parts, e.g., promoters and ribosome-binding sites (RBS), encourages the scope of the otherwise global random mutagenesis, as well as their manifested effects, to be confined to a local region. This is exemplified by the work of Li et al. (2015). In pursuit of a better sensing module for arsenite, they simultaneously mutated the ArsR coding sequence and the ParsR promoter by error-prone polymerase chain reaction and identified mutations that reduced leakiness and boosted the output dynamic range (Li et al. 2015). Part standardization also led to the appearances of part libraries. Since the length of a sequence undergoing mutagenesis is shortened, the combinatorial sequence space to be explored is drastically shrunk. Therefore, sequence variations can be thoroughly exhausted, and this resulted in a large variety of elements with performance that spans the entire activity spectrum. For example, the Anderson promoter library was generated from saturation mutagenesis of a constitutive promoter, and there is a community RBS collection on iGEM that confers different strengths for translation initiation (http://parts.igem.org). They are particularly helpful if only a single part needs to be optimized – a part that was too strong and led to undesirable behavior in a biosensor could be swapped out by a weaker version effortlessly. This will be illustrated by an example of replacing an RBS to reduce biosensor leakiness in section “Managing Leakiness on a Translational Level.” In addition, these libraries enable full exploration of a design space for expression strengths. It has been suggested that a promoter library and an RBS library could be combinatorically combined to optimize the expression strength for any gene of interest (Kosuri et al. 2013).

Biosensor Improvement by Directed Evolution Synthetic biology has also provided new techniques for optimizing or altering biosensor behavior through directed evolution, which include phage-assisted continuous evolution (PACE) (Esvelt et al. 2011) and compartmentalized partnered replication (CPR) (Abil et al. 2017). In an example, a biosensor actuated through a split T7 RNAP polymerase had an improved signal-tonoise ratio after one split half was subjected to evolution by PACE (Pu et al. 2017). In another example, the transcriptional repressor for tryptophan has been evolved by CPR to respond specifically to halogenated tryptophan analogues (Ellefson et al. 2018).

13

Engineering Prokaryote Synthetic Biology Biosensors

289

Development of New Sensing Modules In biosensor development, there is a perennial and insatiable need for new sensing modules for new chemicals, biomarkers, or other targets. Due to strong chemical specificities in regulators and receptors, existing sensing modules intended for one target can rarely fully satisfy the sensing requirement of a different target, even though the two targets may be highly similar in structure. Thus, a search for a specific sensor module is almost always necessary. In nature, prokaryotes contribute the most in biodiversity. They thrive in all niches, including those inhabitable to other organisms, and respond to countless stimuli. If an environment has long been enriched with a compound, it is very likely to find in the vicinity a prokaryotic dweller evolved to detect and metabolize it (Nealson et al. 1991; Cervantes et al. 1994; Nealson et al. 2002). Consequently, a conventional practice to uncover a new biological sensing module is to sample and study bacteria in a target-enriched environment. The aforementioned naphthalene biosensor is one of such examples – the naphthalene metabolic pathway originated from a strain of Pseudomonas fluorescens living in the soil of a manufactured gas plant. This is still routinely practiced but it poses a limit in biosensor development. Nowadays, there are new strategies to accelerate sensing module discoveries.

Part Mining Part mining is the bioinformatics-guided search for biological parts from sequenced genomes and is a branch of genome mining. Genetic parts or proteins often have orthologues in closely related species which carry out similar tasks and have homology in sequences. Therefore, if a part of interest has a defined sequence feature, orthologs that potentially belong to the same family can be identified by performing sequence alignments across multiple genomes. Shortlisted candidates can then be synthesized and characterized, and parts that showed desirable responses can be grouped to form a new library. Many sequenced genomes, as well as metagenomes from unculturable prokaryotes, have been deposited into bioinformatics databases, which become a lucrative resource for part mining (Johns et al. 2018). For example, using sequences for a cadmium-responsive transcription factor from Staphylococcus aureus, researchers discovered a new sensing module for cadmium from the genome of Bacillus oceanisediminis, a bacterium that resides in sediment undersea (Kim et al. 2016). In addition, some databases contain annotated proteins or parts, so mining can also be done via a search of keywords in described or predicted biological functions. One demonstrated example is the creation of a library of sensing modules for aromatic compounds, obtained through part mining from the UniProt database (Xue et al. 2014). Antibody-Derived Domains as Universal Sensing Modules Many sensing modules have been derived from natural pathways, but their development into biosensor components often requires a thorough understanding of the

290

X. Wan et al.

underlying biochemistry, which is a time-consuming process. Moreover, some targets might not have a natural sensing module responsible for their detection. Some recent researches sought to circumvent these problems by turning to antibodyderived protein domains. Antibody-derived domains are protein fragments of antibodies created proteolytically or genetically. The leading examples are the single domain antibodies (sdAb) (Holliger and Hudson 2005). They retain the variable region responsible for strong and specific antigen recognition and binding, but their considerably smaller sizes allow stable and soluble expression from bacteria. Most importantly, a novel sdAb for any given epitope can be quickly and inexpensively created by screening a synthetic and combinatorial antibody domain library against the epitope. Identified sdAb can then be fused with other protein domains that could elicit sensing responses upon target binding. A proof-of-concept example was demonstrated by fusion of a caffeine-binding sdAb to a DNA-binding domain (Chang et al. 2018). sdAb co-localized on a caffeine molecule would then allow dimerization of the DNA-binding domain and restore its function as a transcriptional repressor.

Tools and Strategies from Synthetic Biology for Optimizing Biosensor Performance Many early biosensors have subpar performance compared to their electronic or chemical counterparts, rendering them uncompetitive in field applications. In recent years, tools and strategies developed from synthetic biology created new avenues to improve the characteristics of whole-cell prokaryotic biosensors. There are mathematical models that provide quantitative frameworks for sensor improvement (Ang et al. 2013; Mannan et al. 2017; Xiang et al. 2018), but this section focuses on tools and strategies that have been experimentally proven.

Properties of a Biosensor From the perspective of engineering, it is paramount to define measurable properties so that improvements can be gauged quantitatively. Different metrics are available for defining a whole-cell sensor’s performance (Daniel et al. 2013; Mannan et al. 2017). The one adopted in this chapter focuses on four important aspects most pertinent to applications: selectivity, limit of detection (LOD), output dynamic range, and leakiness (Fig. 2). Selectivity is a qualitative property that concerns how well the biosensor distinguishes the target of interest among others that are chemically similar (Fig. 2a). The other three properties define characteristics of the response curve of a biosensor, which mathematically describes how a sensor’s output varies with its input (Fig. 2b). Most biosensors have a sigmoidal response curve that monotonically increases with the target concentration. The LOD is the minimal concentration of the target that elicits an observable response. The output dynamic range refers to the

13

Engineering Prokaryote Synthetic Biology Biosensors

291

Fig. 2 Metric for defining the performance of a biosensor. (a) Schematic illustrating a biosensor’s selectivity. (b) A biosensor's response curve with the sensor’s leakiness, limit of detection (LOD), and output dynamic range annotated

ratio between the maximally achievable output and the basal output of the biosensor. Leakiness is the basal level of the biosensor when no target is present and has been known to jeopardize sensor applications: if colorimetric outputs are used, the response may be easily saturated even in the absence of targets and would restrict titrimetric analysis (Wackwitz et al. 2008).

Strategies for Enhancing Selectivity Some sensing modules are naturally promiscuous and can be triggered by targets with similar chemical properties. This is often the case for heavy metal sensing (Amaro et al. 2011; Wang et al. 2013). For example, CadC from S. aureus can sense Cd, Pb, and Zn; CmtR from Mycobacterium tuberculosis can sense Cd and Pb; and ArsR from E. coli can sense As, Sb, and Bi (Saha et al. 2017). These sensing modules are thus nonspecific and could give false positives when deployed in the field. Given that those transcription factors are allosterically regulated, an intuitive and routine approach is to randomly mutate their binding pockets, and then screen for mutants with increased selectivity towards the ion of interest. A number of successful cases have been reported: a mutated CueR remained sensitive to Cu2+ but no longer responded to Ag+ and became more sensitive to Au+ (Stoyanov and Browns 2003). In another example, MerR was mutated to detect Cd2+ rather than its original ligand Hg2+ (Hakkila et al. 2011). RcnR is normally regulated by both Ni and Co but could be mutated to only recognize the former (Cayron et al. 2017). However, as with general mutagenesis, this approach is timeconsuming and does not guarantee success in identifying a receptor with desirable traits. A general solution is to employ a genetic logic AND gate with two different receptors that can detect the same ligand (Bernard and Wang 2017). For example, two transcription factors can detect Zn2+, but both are nonspecific: ZraR detects Zn2+ and Pb2+, and ZntR detects Zn2+ and Cd2+ (Wang et al. 2013). Implementation of an

292

X. Wan et al.

AND logic using ZraR and ZntR as the inputs will yield a biosensor that responds to Zn2+ but not Pb2+ or Cd2+, thus increasing its specificity (Fig. 3a). This AND gate example was derived from the HrpRS activator complex and its cognate promoter PhrpL (Fig. 3b). PhrpL is a σ54-dependent promoter that is activated only when both HrpR and HrpS are present (Jovanovic et al. 2011; Wang et al. 2011a; Liu et al. 2018). Many other logic gates orthogonal to the HrpRS system are also available from the synthetic biology community: a T7 RNA polymerase-amber suppressor system that integrates two input signals (Anderson et al. 2007), activator-

Fig. 3 Biosensors specificity enabled by synthetic biology. (a) A zinc-specific biosensor using an AND gate (Wang et al. 2013). (a-i) A zinc/lead biosensor and its response curves for Zn2+ and Pb2+. (a-ii) A zinc/cadmium biosensor and its response curves for Zn2+ and Cd2+. (a-iii) A zinc-specific biosensor was generated by integrating both sensing modules from i and ii into an AND gate. gfp, gene encoding a green fluorescent protein. (b) The HrpR/HrpS hetero-regulation motif in the hrp (hypersensitive response and pathogenicity) system of Pseudomonas syringae pv. tomato DC3000 (Wang et al. 2011a; Wang and Buck 2014). The hrp system promotes pathogenicity of the bacteria in their plant host. The σ54-dependent hrpL promoter can be activated by the heterohexamers of the transcriptional activators HrpR and HrpS

13

Engineering Prokaryote Synthetic Biology Biosensors

293

chaperone systems (Moon et al. 2012), split T7 RNA polymerase-based systems (Shis and Bennett 2013; Schaerli et al. 2014), and recombinase-based Boolean gates (Bonnet et al. 2013; Courbet et al. 2015). These orthogonal logic gates provide the means to enhance biosensor specificity in more complicated cases.

Strategies for Lowering the Limit of Detection (LOD) Similar to solutions for biosensor selectivity, random mutagenesis on promoters or transcription factors used to be a popular way to generate mutated sensing modules capable of detecting lower ligand concentrations. However, the issue of LOD is intrinsically tied to the relative concentrations between the receptor and the ligand, so a more rational approach is to tune their respective densities in the cell.

LOD Improvement by Tuning Receptor Densities Depending on their modes of action, the LOD of a sensing module could be lowered by increasing or decreasing concentrations of receptors (Merulla et al. 2013; Wang et al. 2015; Wan et al. 2019). If the receptor is a transcriptional repressor that could be derepressed by an inducer (Fig. 4a), reducing its concentration could lead to both a lower LOD and a higher dynamic range. In the canonical allosteric transcriptional regulation paradigm, the binding between the inducer and the repressor, and between the repressor and the promoter are in equilibrium. A minimum concentration of inducer is always required to sufficiently inactivate the repressor and to allow the promoter to drive an observable expression of the reporter gene. Reducing the repressor concentration would therefore lower the specified demand on the inducer concentration, which effectively translates into a decreased LOD for the sensing module. This strategy has been demonstrated on the previously described arsenic biosensor, in which its LOD was improved by replacing the strong constitutive promoter that drives the repressor ArsR by a weaker variant (Fig. 4c). To improve the LOD in the opposite scenario, where the receptor is a transcriptional activator inducible by the inducer (Fig. 4b), the receptor density needs to be raised (Wang et al. 2015). In the case of the LuxR-Plux sensor that detects the ligand AHL, more LuxR present in the cells implies a higher probability to form the LuxRAHL complexes (Fig. 4d). Therefore, the Plux promoter can be activated with a lower concentration of AHL. It is noteworthy that receptor densities could also be dynamically controlled to give conditional LOD. A cadmium biosensor was integrated with a toggle switch, where the CadR repressor concentration was modulated by both the cadmium concentration and a LOD tuning ligand (Wu et al. 2009). Under a moderate level of the LOD tuning ligand, an increase in cadmium concentration would in turn reduce expression of the CadR repressor. This positive feedback mechanism lowered the LOD of the biosensor, and interestingly, the feedback could be quenched by increasing the concentration of the LOD tuning ligand.

294

X. Wan et al.

Fig. 4 Strategies for improving biosensors’ LOD. (a, b) A transcriptional repressor (TR) and a transcriptional activator (TA)-based biosensor. PC, constitutive promoter. PTR, TR’s cognate promoter. PTA, TA’s cognate promoter. Black dots, targets of interest. gfp, gene encoding a green

13

Engineering Prokaryote Synthetic Biology Biosensors

295

LOD Improvement by Tuning Intracellular Ligand Densities For ligands that interact with sensing modules intracellularly and do not diffuse freely across the membrane, increasing their cytoplasmic concentrations increases their availabilities and hence their probabilities to excite the sensing module. In fact, this strategy can be frequently found in nature. In E. coli for example, the transportations of arabinose and rhamnose across the cell membrane are governed by positive feedback loops (Altenbuchner and Mattes 2005; Fritz et al. 2014). Induction of the metabolism pathways drives the expression of the transmembrane transporter proteins, which subsequently facilitates the imports of the sugars. In biosensor design, if the transmembrane transporters for a ligand could be identified, the same outcome could be reproduced by overexpressing importers and knocking out exporters (Fig. 4e). For instance, disruption of the efflux transporters for Zn/Cd/Pb in P. putida strain KT2440 decreased the detection limits by up to 45fold (Hynninen et al. 2010). In another example, an engineered E. coli biosensor achieved a lower LOD for Ni through the introduction of several foreign Ni-uptake systems and the deletion of Ni efflux pumps (Cayron et al. 2017).

Strategies for Increasing Output Dynamic Range A number of native promoters responsive to heavy metals are relatively weak and so their derived biosensors suffer from limited output dynamic ranges (Stocker et al. 2003; De Mora et al. 2011; Siegfried et al. 2012; Wang et al. 2013; Huang et al. 2015a; Kim et al. 2016; Merulla and van der Meer 2016). Again, one option is to perform random mutagenesis on the promoter to maximize the transcriptional output. Recently, however, a more reliable approach has been developed for engineering hybrid σ70-based promoters. Chen et al. demonstrated that the dynamic ranges of these promoters could be quantitatively predicted by the choices of their
35 and
10 regions, which dictate the binding affinities between the promoters and the σ factors required for transcription initiation (Fig. 5a) (Cox et al. 2007; Brewster et al. 2012; Guzina and Djordjevic 2017; Chen et al. 2018). By inserting binding sites of inducible transcription factors around the
35 and
10 sequences, inducible promoters with improved dynamic ranges could be obtained. This example nonetheless is specific to σ70 promoters and remains inapplicable to promoters with uninsulated promoter core sequences.

ä Fig. 4 (continued) fluorescent protein. (c, d) Improving a biosensor’s LOD by tuning receptor densities (Wang et al. 2015). (c) shows a TR-based arsenic biosensor, and (d) shows a TA-based AHL biosensor. PStrong, strong constitutive promoter. PWeak, weak constitutive promoter. (e) Improving a biosensor’s LOD by increasing its targets’ intracellular density (Hynninen et al. 2010; Cayron et al. 2017). IS, importing system. ES, exporting system. R, receptor. PR, R’s cognate promoter

296

X. Wan et al.

Fig. 5 Strategies for improving biosensors’ output dynamic range. (a) Different σ70 binding sites of an inducible promoter yield different basal levels and output dynamics as a result of relative equilibrium constants of σ70 binding to the
10 and
35 regions, ln(Keq) =
(ΔG
10 + ΔG
35) (Chen et al. 2018). Non-consensus bases are underlined. ΔG
10 and ΔG
35 correspond to the relative changes in the binding energy due to changes in the
10 and
35 sites. R, receptor. PR, R’s cognate promoter which contains an operator for R (RO). yfp, gene encoding a yellow fluorescent protein. (b) Biosensors without transcriptional amplifiers (TAmp) (b-i), with two transcriptional repressor (TR)based amplifiers (b-ii), or with a transcriptional activator (TA)-based amplifier (b-iii), and their response curves (Hooshangi et al. 2005; Wang et al. 2014; Kim et al. 2016). iv, an amplifier with positive feedback (Nistala et al. 2010). PTR, TR’s cognate promoter. PTA, TA’s cognate promoter. gfp, gene encoding a green fluorescent protein. (c) A gain-tunable TAmp based on a HrpRSV system (Wang et al. 2014). This device scales the weak transcriptional input signal (I) linearly in response to a second “gain tuning” transcriptional input (βT)

13

Engineering Prokaryote Synthetic Biology Biosensors

297

A more universal solution from synthetic biology is to magnify the initially limited output through transcriptional amplifiers (TAmps). A TAmp is analogous to a buffer gate in electronics – it amplifies a transcriptional input signal before outputting it (Fig. 5b). In general, a TAmp takes the form of a transcriptional cascade that comprises a transcription factor (TF) and its cognate promoter PTF. TF is the output of the sensing module and the PTF drives expression of the next cascade or the final observable output. To qualify as a TAmp, PTF must have a higher maximal activity than the promoter upstream of TF, so that a minute transcriptional input could be converted into a much larger downstream output. Therefore, mathematically, a TAmp can be understood as a function that receives a transcriptional input from a relatively narrow domain and maps it to a much larger range. Early examples of TAmps were built using transcriptional repressors. Two examples of repressor-based cascades have been built, and one was shown to amplify weak promoter activities that would otherwise be unobservable (Fig. 5b-ii) (Karig and Weiss 2005; Hooshangi et al. 2005). However, the use of repressor-based amplifiers might not be suitable for sensing modules with positive relationships between inputs and outputs, because they would invert the response functions, unless they are cascaded in even numbers. New TAmps based on transcriptional activators have been developed. This is exemplified by a TAmp built from the HrpRS system described above (Fig. 5b-iii and c) (Wang et al. 2014; Wan et al. 2019). It readily accepts a wide range of transcriptional inputs and linearly amplifies them in an analog fashion. This property proved useful when the amplifier was used to significantly improve an arsenic sensor’s output dynamic range. The TAmp is also versatile – the amplification gain can be tuned by regulating the level of HrpS, which could be achieved either translationally, via changing the RBS, or post-translationally, by expressing the HrpS inhibitor HrpV. Another notable example of TAmp was based on the T7 RNA polymerase and its cognate promoter PT7 (Kim et al. 2016), which has been shown to improve both the LOD and the output dynamic range of a cadmium/lead biosensor. Provided that the sensing module is tightly regulated, amplification capacities for a TAmp could be further augmented through the incorporation of a positive feedback loop. In a proof-of-concept circuit, the output of a TAmp drives expression of an activator used in the TAmp (Nistala et al. 2010), so signal amplification not only applies on the input but also on the output (Fig. 5b-iv). Compared to the stand-alone TAmp architecture, this coupled TAmp significantly improved the observable output as well as the detection limit of a biosensor. Caution should be exercised when applying this coupled architecture to a potentially leaky sensing module, because a high basal output level could be significantly amplified and would lead to a reduced output dynamic range. Output amplification can also be realized using recombinase-based memory modules, provided that the maximal attainable transcription activity of the final output can surpass that of the sensing promoter. This will be described further in section “Recombinase-Based Memory Devices.”

298

X. Wan et al.

Strategies for Reducing Leakiness Biological processes are evolved to be inherently leaky and noisy, because it leaves room for variations and allows bet-hedging in face of sudden and drastic changes in the environment (Randall et al. 2011). One would therefore rarely find a natural sensing module that displays little to no leaky behavior. So it is important, and more appropriate, to view leakiness not as a curable property but as an unavoidable issue to be managed. Leakiness of a biosensor typically originates from the promoter within the sensing module (Wackwitz et al. 2008; Arpino et al. 2013; Adams et al. 2014; Merulla and van der Meer 2016), but it could be addressed by strategies that act on levels of transcription, translation, and post-translational modifications.

Managing Leakiness on a Transcriptional Level Receptor and Promoter Engineering Since the sensing promoter is almost always the source of leakiness, it is logical to start by engineering the promoter and its cognate transcription factor. Once again, random mutagenesis and random promoter truncation are two popular approaches – libraries of mutant sensing regulator-promoter pairs are screened to obtain less leaky and yet still functional variants (Li et al. 2015; Daeffler et al. 2017). Identified candidates usually have mutations that alter the promoters’ transcription initiation rates or the affinities between the transcriptional regulators and the promoters. The previously described cases of mutagenesis done on the ArsR-ParsR sensing module and the quantitative approach to engineer hybrid σ70 promoters are examples of such approaches (Fig. 5a) (Li et al. 2015; Chen et al. 2018). A similar but more predictable method is to vary the number and position of operators (Fig. 6a and b) (Merulla and van der Meer 2016; Zong et al. 2017; Chen et al. 2018; Wan et al. 2019). Operators are sequences within a promoter that serve as binding sites for transcriptional regulators, and there are some general rules regarding how they affect promoter leakiness. Operators for repressors may be placed in the distal region (upstream of
35), the core region (between
10 and
35), or the proximal region (downstream of
10) (Fig. 6a). If only a single operator for repressor exists in the promoter, maximal repression efficiency, and hence minimal leakiness, would be obtained when the operator is placed in the core region (Cox et al. 2007). Repression efficiency is further enhanced if the operator overlaps with part of the
10 or
35 regions (Chen et al. 2018). Adding an extra operator downstream of a promoter can also reduce leakiness. Should the first operator fail to recruit the repressor, the repressor bound to the extra operator can still inhibit readthrough of the RNA polymerase (Fig. 6b). This effect is known as “roadblocking,” and its efficiency can be tuned by varying the distance between the extra operator and the core region (Hao et al. 2014; Merulla and van der Meer 2016). Though it should be noted that the roadblocking efficiency depends on the maximum strength of the sensing promoter, the repressor concentration and the affinity between the repressor and the operator, and so the effects on different operator-repressor pairs would likely be variable.

13

Engineering Prokaryote Synthetic Biology Biosensors

299

Fig. 6 Strategies of tuning biosensors’ leakiness and output dynamics. (a) A transcriptional repressor (TR)-based inducible promoter, with an operator site (TRO) at the distal, core, or proximal region of the promoter. Repression efficiencies were shown to depend on the TRO’s location, which follows core  proximal  distal (Cox et al. 2007). PTR, TR’s cognate promoter. yfp, gene encoding

300

X. Wan et al.

Unlike their repressor counterparts, operators for activators can only be functional when placed in the distal region (Cox et al. 2007). Antisense Transcription Antisense transcription refers to the strategy of inserting a second promoter downstream of the sensing promoter but in an opposite direction (Pelechano and Steinmetz 2013). The second promoter will drive transcription of an RNA that is partially or fully complementary to the RNA produced from the sensing promoter. Antisense transcription interrupts RNA polymerase reading from the first promoter (Fig. 6c-iii) (Brophy and Voigt 2016), thus reducing its basal expression level and leakiness. If the antisense promoter is located at the 30 end of the target gene, the antisense RNA will be long enough to form a double-stranded RNA with the target mRNA and then triggers its degradation (Fig. 6c-ii) (Lasa et al. 2011). The efficiency of this strategy can be tuned by regulating the strength of the antisense promoter (Fig. 6c) (Brophy and Voigt 2016), but the maximum output expression from the first promoter may also significantly decrease if the antisense promoter is too strong. Antisense transcription may also exert translational interference: an anti-sense promoter that is positioned immediately downstream of the target promoter may lead to transcription of a short antisense RNA that inhibits translation initiation of the target gene (Fig. 6c-i) (Kawano et al. 2007).

Managing Leakiness on a Translational Level The expression rate of any gene of interest can be superficially regarded as a product of its transcription rate and translation rate. Reducing the translation efficiency is therefore an indirect way to counteract transcription rates that are either too high or too low. This is most helpful in scenarios where transcription elements are fixed and cannot be readily replaced. For example, in a transcriptional cascade, the upstream and downstream promoters are defined. Yet the effective intracellular concentrations of the transcriptional regulators can be modulated by changing their translation rates. This would alter their input-output characteristics and in turn impact the final observable leakiness of the biosensor (Wang et al. 2011a; Nielsen et al. 2016). In a specific example, an active recombinase under the control of a leaky inducible promoter might switch on the downstream signal even in absence of a target. By ä Fig. 6 (continued) a yellow fluorescent protein. (b) Transcriptional roadblocking effect. gfp, gene encoding a green fluorescent protein. (c) Antisense transcription as a tool to tune gene expression. Left panel: antisense promoter (Panti) can reduce PR’s leakiness by blocking entry of the ribosome to the reporter’s mRNA (c-i), triggering mRNA degradation(c-ii), or blocking the reading of the RNA polymerase (RNAP) along the gene (c-iii). Right panel: The sensor’s leakiness and output dynamics correlated with the strength of Panti (Brophy and Voigt 2016). R, receptor. PR, receptor’s cognate promoter. (d) Modifications of ribosome binding sites (RBS) for a transcriptional activator (TA), a TR, or the reporter change the leakiness and output dynamics of a biosensor (Wang et al. 2011a; Nielsen et al. 2016; Rubens et al. 2016). PTA, TA’s cognate promoter. (e) Schematics of a posttranslational regulation on an IPTG biosensor (Fernandez-Rodriguez and Voigt 2016). L represents a protein degradation tag LVA

13

Engineering Prokaryote Synthetic Biology Biosensors

301

attenuating its translation initiation rate and hence its concentration, this undesirable effect could be mitigated (Rubens et al. 2016). The most common way to tune translation is to modify the translation initiation rate, which is to a great extent governed by the sequence of the ribosome-binding site (RBS) that dictates its affinity towards the ribosome (Fig. 6d). RBS strengths can be described by a biophysical model and are generally predictable from the RNA sequences alone (Salis et al. 2009). A few RBS libraries with characterized translation initiation efficiencies are also available (http://parts.igem.org). However, translation rates of a protein have been shown to depend on myriad factors other than the RBS itself (Salis et al. 2009; Egbert and Klavins 2012; Kosuri et al. 2013; Mutalik et al. 2013; Wu et al. 2018). The recent work by Cambray et al. provided the most comprehensive understanding on how sequences other than the RBS affect translation efficiencies – RNA secondary structures in the vicinity of the start codon play a dominating role, while other factors, like codon usage, contribute to a much smaller extent (Cambray et al. 2018).

Managing Leakiness on a Post-translational Level In some published biosensors, protein degradation tags were attached to the regulator or reporter proteins to promote their clearance from the cell and reduce their effective concentrations (Andersen et al. 1998; Arpino et al. 2013; Cameron and Collins 2014; Bradley et al. 2016b). However, this method comes with a trade-off with a reduction in maximum output level (Fig. 6e, gray line). A solution that rescued the output level but maintained the lower level of leakiness has been recently developed (Fig. 6e) (Fernandez-Rodriguez and Voigt 2016; Wan et al. 2019). It achieved so by inserting a protease cleavage site between the reporter protein and the degradation tag, and the cognate protease is under the control of the sensing module. At lower target concentrations, a small amount of reporter was produced due to leaky expression, but they were quickly degraded due to the presence of the degradation tag. At high target concentrations, both the protease and the reporter would be strongly expressed, so the protease could cleave off the degradation tag from the latter, and the observable output could remain at a high level (Fig. 6e, orange dotted line). This protease-based regulation strategy is flexible and can be effortlessly applied to any leaky biosensors.

Functional Expansion of Biosensors by Synthetic Biology Memory Devices A biosensor with only sensing and actuating modules suffices to report the immediate bioavailability of a target. Yet when endowed with memory modules that allow record and retrieval of transient detection events, whole-cell biosensors become excellent platforms for continuous documentation of surroundings, because they are living organisms that can proliferate and colonize an environment with minimal maintenance from humans (Kotula et al. 2014). This could be useful in tracking a

302

X. Wan et al.

delayed response in a difficult-to-reach environment, like those in clinical diagnostic settings (Courbet et al. 2015). Memory devices can be broadly classified according to their reversibility and whether the memory is encoded in the DNA (i.e., genetic versus epigenetic). Synthetic biologists have created an overwhelming number of memory devices and they have been reviewed elsewhere (Inniss and Silver 2013; Roquet and Lu 2014; Bradley and Wang 2015). Given the scope of this chapter, only examples that have been successfully installed and demonstrated in biosensors will be covered.

Toggle Switches The synthetic toggle switch can be considered as the earliest epigenetic and reversible memory device from synthetic biology (Kotula et al. 2014; Zhang et al. 2014; Certain et al. 2017; Riglar et al. 2017). It achieved bistability through two repressors that mutually inhibit each other (Fig. 7a) (Gardner et al. 2000). A transient induction that upregulates the expression of either one of the repressors (TR1) downregulates the expression of the other (TR2) and allows the latter to dominate, which results in a flip from one state to another. The state of the system will then be stably maintained until the now dominant repressor (TR1) is repressed again. Since only two states are allowed, the toggle-switch converts any graded input response into a digital output, and the readout for a target will no longer be titrated, but this might improve robustness in a sensor and aid decision-making processes (Roquet and Lu 2014). The lambda phage cI/Cro genetic switch is a natural toggle switch which has high repression efficiency and modularity, but other strong repressors that display cooperative binding properties can also be used. Toggle switches have been widely used in whole-cell biosensors. For example, E. coli biosensors have been engineered to sense and record an antibiotic exposure or an inflammation in murine guts (Kotula et al. 2014; Riglar et al. 2017). Another toggle switch was part of a Pavlovian-like conditioning circuit in E. coli, where it could memorize a conditioned stimulus (Zhang et al. 2014). Recombinase-Based Memory Devices Site-specific recombinases are enzymes that perform genetic recombination on DNA flanked by specific recognition sites (aka recombination cassettes), resulting in either DNA excisions or inversions (Olorunniji et al. 2016). Biologically, recombinases belong to either the tyrosine or serine recombinase classes and can be further classified by their directionality and their permitted mode of actions (Wang et al. 2011b). For recombinases that can perform both excision and inversion, the mode of recombination depends on the relative orientations of the recombination sites: excision happens between aligned recombination sites and for inversion, antialigned sites (Fig. 7b). Both allow implementations of memory provided that the DNA within the recombination cassette is a functional biological part. A simple memory device using DNA excision can be built by inserting a recombination sites-flanked terminator between a constitutive promoter and a translational unit, which interrupts gene expression until its removal (Bonnet et al. 2013). In the case of inversion, the genetic part within the cassette would be purposefully

13

Engineering Prokaryote Synthetic Biology Biosensors

303

Fig. 7 Diagrams of genetic circuits for memorizing environmental signals. (a) A toggleswitch-based memory device. Initially, the toggle switch is at an OFF state, where the TR2 is expressed and TR1 expression is repressed. Upon target detection, the sensing circuit expresses TR1, which flips the device into an ON state. TR1 and LacZ will be continuously expressed even after the target is removed (Kotula et al. 2014; Riglar et al. 2017). R, receptor. PR, R’s cognate promoter. TR, transcriptional repressor. PTR, TR’s cognate promoter. lacZ, gene encoding βgalactosidase for colorimetric output. (b) A recombinase-based memory device. Upon sensing a particular target, the biosensor produces the Cre recombinase, which first flips the orientation of the

304

X. Wan et al.

inserted in a nonfunctional orientation, only to be restored when the recombinase is active. In both cases, the DNA is transformed from one sustained state into another, thereby conferring memory. Since the flanked DNA can only exist in either one of the two states, all recombinase-based memory devices are analog to digital converters. Their genetic nature also implies that the memory is heritable and would last even if the host cell is dead (Siuti et al. 2013). Many early synthetic memory devices were based on inversions using bidirectional recombinases like the Cre/lox and FLP/FRT systems. They tended to stochastically re-catalyze an inverted DNA back into the original orientation and create a mixed population of cells in either ON or OFF states, which posed an issue to robustness in memory (Schnütgen et al. 2003; Brophy and Voigt 2014). Workarounds to this issue include using mutated recombination sites that drive inversion equilibrium towards the product (Albert et al. 1995; Oberdoerffer 2003) or configurations that would lead to the disappearance of some recombination sites in the end state (Fig. 7b) (Schnütgen et al. 2003). The same mixed population issue also applies to unidirectional recombinases when their expression is at intermediate levels, but the problem could be alleviated through the use of feedback loops (Moon et al. 2011; Brophy and Voigt 2014; Folliard et al. 2017). Memory devices engineered from recombinase-recombination cassettes with unidirectionality provide irreversible memories – a temporal exposure to the target will flip the system into an irrevocable state (Siuti et al. 2013). To date, a large number of orthogonal unidirectional recombinases have been identified, and they allow different inputs to be remembered within the same cell (Yang et al. 2014). They were demonstrated to record conditions in gastrointestinal tracts, as well as pathogens in human serum and urine samples (Archer et al. 2012; Courbet et al. 2015; Mimee et al. 2015). The sequence of detection events can also be recorded in a state machine, which assigns a unique state to every possible sequence order by coordinating multiple unidirectional recombinases (Roquet et al. 2016). Biosensors built with irreversible memories would be single-use commodities. In contrast, reusable and rewritable memory devices can be constructed from recombinase-excisionase pairs with conditional bidirectionality (Fig. 7c) (Bonnet et al. 2012; Bonnet et al. 2013). The mechanism is illustrated using the integrase Bxb1 gp35 and its cognate excisionase Bxb1 gp47. The integrase alone will drive ä Fig. 7 (continued) gfp flanked by loxP sites, and then excises one of the two loxP sites through the lox511 sites (Schnütgen et al. 2003). PC, constitutive promoter. gfp, gene encoding a green fluorescent protein. (c) An integrase-based memory device switches the sensor’s output from gfp expression to rfp expression. The integrase and excisionase together restore the gfp expression (Bonnet et al. 2012). rfp, gene encoding a red fluorescent protein. (d) A CRISPR-based “biological tape recorder” system. The signals are recorded into the genomic CRISPR array (Sheth et al. 2017). When there is no signal, only the reference DNA will be recorded; when there are signals, the trigger DNA will be rapidly replicated and preferentially recorded into the CRISPR array. (e) The CAMERA recording system (Tang and Liu 2018) has two possible mechanisms: i, it uses Cas9 nucleases to record signals by shifting the ratio of two recording plasmids; ii, it uses Cas9-derived base editors to change DNA sequences upon sensing a signal

13

Engineering Prokaryote Synthetic Biology Biosensors

305

inversion towards one direction, but the direction is reversed in the presence of the excisionase.

CRISPR/Cas-Based Memory Devices CRISPR (Clustered Regularly Interspaced Short Palindromic Repeats) and CRISPRassociated (Cas) systems are bacterial immunity systems against phage infections. Their molecular mechanisms have been intensely studied and frequently reviewed elsewhere and so will not be covered in this section (Shalem et al. 2015; Jiang and Doudna 2017). CRISPR/Cas systems have innate memory functions to remember previously invaded viruses. They capture exogenous DNA from plasmids or phages and integrate them into the genomic CRISPR arrays as spacer sequences. This adaption process has been repurposed to yield a synthetic memory device, known as the “biological tape recorder” (Fig. 7d) (Sheth et al. 2017). A new spacer is incorporated into the array upon induction by a specific target, and multiple events over time can be recorded. The popularity of the CRISPR/Cas system, nonetheless, remains in its versatility in performing precise DNA cutting and editing. These properties have been also harnessed in a memory device named CAMERA (CRISPR-mediated analog multievent recording apparatus) (Fig. 7e) (Tang and Liu 2018). The system memorizes signals by one of the two ways: (1) An incoming signal modulates the activity of a DNA cleavage enzyme, which selectively cuts one of the two recording plasmids and thus alters the ratio between the two. (2) The signal instructs Cas9-derived base editors to modify designated DNA sequences. In both examples, the recorded information can be retrieved by sequencing the barcoded DNA sequences or by coupling the resulting change to an observable output. Other Notable Memory Devices The three types of memory devices described above were so far the most popular ones. Nonetheless, there are two other noteworthy examples that were also demonstrated in biosensors: (1) In a device named SCRIBE (Synthetic Cellular Recorders Integrating Biological Events), detection events lead to productions of hybrid RNA-ssDNA molecules that will undergo genetic recombination with the bacterial genome, modifying sequences on the latter (Farzadfard and Lu 2014). The strength of sensing correlates with the frequency of recombination and is therefore reflected on a population level, specifically, the proportion of cells that carries the modification. (2) An epigenetic and reversible memory device was constructed by DNA methylase and DNA-binding proteins (DBP) that are sensitive to DNA methylation. Target detection therefore triggers DNA methylation and precludes binding of the DBP, and memory reset is carried out via degradation of the methylase (Maier et al. 2017).

Computation Modules to Integrate Signals Most biosensors were built to detect a single input, but by receiving and processing multiple inputs, they can be used to sense a complex condition or a global

306

X. Wan et al.

Fig. 8 Detecting an environmental condition using multi-input AND gate and cell-cell communication. A three-input AND logic gate based on two HrpRS-based AND gates are separated into two different cell strains of a consortium (Wang et al. 2013). luxI, synthase of a quorum sensing molecule 3OC6HSL. rfp, gene encoding a red fluorescent protein

environment (Fig. 8) (Wang et al. 2013). Synthetic biology has offered numerous examples on biocomputation and interested readers are advised to consult references cited here (Wang et al. 2011a; Moon et al. 2012; Wang and Buck 2014; Ma et al. 2016; Nielsen et al. 2016; Roquet et al. 2016; Rubens et al. 2016; Xiang et al. 2018). Signal integration is of great interest in clinical diagnosis and biotherapy, where multiple signals define a specific disease state and determine if drugs should be administered. To date, no genetic circuits have been developed in prokaryotic biosensors to target multiple clinical biomarkers, but similar ideas have been proposed and tested. For example, a hypothetical E. coli that would invade tumor cells would only work when two conditions are satisfied: (1) it reaches a quorum due to colonization and (2) detects a hypoxic environment that is typically found in tumors (Anderson et al. 2006). The two conditions would thus require integration through an AND gate.

Modules to Reshape Response Function Other forms of signal processing can remodel the conventional sigmoidal response curve to facilitate biosensor readout. By connecting a sensing module to an incoherent feedforward loop, a biosensor can behave as a bandpass filter and only responds to a limited range of analyte concentrations (Peking iGEM 2013; Rubens et al. 2016). In another example, a coherent feedforward loop successfully transformed the response curve into a semilog sensing curve (Zhang et al. 2013). These changes in response functions could be helpful when the output sensing modules are wired to specific actuators that are not very responsive to sigmoidal inputs.

13

Engineering Prokaryote Synthetic Biology Biosensors

307

Reporter Modules for Interfacing with Different Detection Platforms Many biosensors were first developed using fluorescent reporters as the actuator, which facilitate their characterization in laboratory settings. Yet employing these biosensors in the field would be inconvenient since it must be accompanied with a fluorescent reader. The synthetic biology community has developed a diverse collection of reporters or actuators. They are standardized and so can easily substitute the fluorescent reporters. This allows the output signals to be observed through other means. For example, some reporter modules can convert detection events into electrical currents measurable by electrodes (Webster et al. 2014). Typically, colorimetric reporters are often used to allow direct observations of sensor output by naked eyes. They include chromoproteins, some fluorescent proteins, and enzymes that produce pigments or catalyze chromogenic substrates (Biran et al. 2003; Stocker et al. 2003; Fujimoto et al. 2006; Wackwitz et al. 2008; Joshi et al. 2009; De Mora et al. 2011; Joe et al. 2012; Shin 2012; Kotula et al. 2014; Pardee et al. 2014; Courbet et al. 2015; Danino et al. 2015; Huang et al. 2015a; Pardee et al. 2016a; Didovyk et al. 2017; Watstein and Styczynski 2018). This could drive down the operating costs of a biosensor because it obviates additional readout machines. Yet, there are also a few other interesting reporters that enable biosensors to operate in vivo: acoustic reporter genes were used to encode intracellular gas vesicles in E. coli (Bourdeau et al. 2018). These vesicles can scatter ultrasound waves that noninvasively penetrate living tissues and therefore lead to imaging contrast. Another example employs a luciferase reporter: E. coli entrapped in an ingestible micro-bio-electronic device (IMBED) can detect analytes that diffuse into their residing chemostat chamber. The cells then respond by emitting light which can be converted into digital signals through photodetectors. The signal can then be further relayed to computers via Wi-Fi (Mimee et al. 2018).

Biosafety Enhancing Modules As explained in the introduction, biosafety remains the bottleneck for field applications of prokaryotic whole-cell biosensors, and a large volume of work in synthetic biology was dedicated to this issue. Some representative work has been selected to illustrate various strategies in managing the biocontamination risks. “GeneGuard” was a stable and modular system for biosafety control in E. coli (Wright et al. 2015). Three safety modules were inserted into the genome: (1) A richmedia compatible auxotrophy selection marker ensures that the host cell would not survive if it leaves an industrial closed system. (2) Host-dependent origins of replication, as well as (3) toxin-antitoxin pairs prevent propagation of episomes that could have accidentally transferred to other organisms. The host cells can also be programed to commit suicide after a certain retention time in the environment. This could be achieved by putting genes encoding toxic

308

X. Wan et al.

products under the control of a synthetic timer (Ellis et al. 2009) or counting circuits (Friedland et al. 2009; Callura et al. 2010). Another method is to erect a “linguistic barrier” between the host cell and other prokaryotes in nature (Pinheiro et al. 2012; Wright et al. 2013). The genes that are expressed in the host cells are refactored, so they cannot be transcribed or translated in other prokaryotes in the wild.

Conclusions and Future Directions Prokaryotic whole-cell biosensors are suitable for a broad range of applications and are promising alternatives to other types of sensors. With the support of synthetic biology, the development and optimization of novel whole-cell biosensors has been vastly accelerated. Engineering principles facilitate biosensors to be constructed from a bottom-up approach and provide means to engineer new sensing modules on top of the myriad available options. Through the lens of rational design, researchers could pinpoint biological properties that limit selectivity, LOD, dynamic range, and leakiness and apply strategies that precisely tackle each issue and improve the overall sensing behavior of a biosensor. Finally, various parts and devices from the synthetic biology community augment functions of biosensors in memory and signal processing capacities, and make them easier and safer to be used in field applications. More tools and design principles that give rise to robust and reliable sensors will likely be discovered in the future, and they will further encourage whole-cell biosensors to be applied in real life scenarios. Synthetic biology alone, however, cannot solve all the problems present in wholecell biosensors. At the current stage, despite the fact that many prokaryotic biosensors had impeccable performance in the laboratory, only a few could be commercialized. There are three underlying reasons. First, the actual working environments for the biosensors are far more complicated than their laboratory counterparts, so requirements on LOD, selectivity, and robustness are more demanding. The major solution is to subject the biosensor to multiple rounds of rigorous tests using real environmental or clinical samples and progressively optimize their response behavior through genetic manipulation. Approaches described in this chapter should aid this process, but sample preprocessing steps like target purification or concentration might be helpful complementary methods to decrease biosensor LOD (Wen et al. 2017). There is also a dearth of durable, inexpensive, and user-friendly platforms for whole-cell biosensor storage and multiplexed sample testing. Some options are available, but they do not keep the cells alive over long periods of time. Hydrogels, which include alginate beads, agarose, and silica gels, can entrap prokaryotic cells and keep them hydrated and functional for around a month (Chang and Prakash 2001; Nassif et al. 2002; Papi et al. 2005; Sharma et al. 2010; Buffi et al. 2011; Power et al. 2011; Shin 2012; Courbet et al. 2015; Belkin et al. 2017; Wan et al. 2019). Better entrapping materials and storage conditions are needed for prolonging the shelf life of biosensors. An alternative is to find a different chassis that is

13

Engineering Prokaryote Synthetic Biology Biosensors

309

indifferent to harsh environments. In this sense, the Gram-positive bacteria B. subtilis are recently gaining ground. They can give physically and chemically resistant endospores that can be stored dry in a wide range of temperatures for years but can still return to a vegetative state when nourished with water and nutrients (Joshi et al. 2009; Volpetti et al. 2017). Cheap microfluidic devices have also been developed as platforms for biosensor applications, yet they remain inconvenient for multiplexed sample processing and on-site diagnosis (Buffi et al. 2011; Kim et al. 2016; Volpetti et al. 2017; Wan et al. 2019). Lastly, as long as they are still alive, whole-cell prokaryotic biosensors will carry the stigma of being potentially biohazardous when they exit laboratories. Biosafety concerns like ecosystem disruption and antibiotic resistant gene transfer will need to be addressed (Dana et al. 2012). Adequate intrinsic containment systems (biosafety circuits reviewed in section “Biosafety Enhancing Modules”) and physical containment systems like bioreactive nanofibers (Tong et al. 2014) and the aforementioned hydrogel/microfluidic device encapsulation will need to be in place for whole-cell biosensors to be safely deployed in the field. In recent years, cell-free systems have been proposed as a solution to circumvent the biosafety issues associated with whole-cell biosensing. A cell-free system consists of cell extracts or purified transcription/translation machineries mixed with energy supplements and amino acids (Perez et al. 2016). It is cheap and easy to produce (Kwon and Jewett 2015; Didovyk et al. 2017). A number of biosensors using cell-free extracts have already been shown to be capable of detecting heavy metals, pathogens, antibiotics, and viral RNA (Pellinen et al. 2004; Pardee et al. 2014, 2016a; Didovyk et al. 2017; Duyen et al. 2017; Wen et al. 2017; Ma et al. 2018). Furthermore, they could be freeze-dried on cellulose paper, which increases their stability and portability (Pardee et al. 2014 2016b). Theoretically, any wholecell biosensors can be converted into this paper-based cell-free system and become point-of-care diagnostic devices. Still, there will always be unbridgeable differences between an in vivo intracellular environment and a cell extract, which may cause unpredictable behavior during biosensor circuit migration. Therefore, until the moment that all circuits can be transferred systematically and flawlessly across the two platforms, whole-cell prokaryotic biosensors will remain a major player in the field of biosensors. Acknowledgments This work was supported by funds from UK BBSRC [BB/N007212/1], Leverhulme Trust [RPG-2015-445], Wellcome Trust [202078/Z/16/Z], EPSRC/BBSRC Global Challenges Research Fund Awards and Wellcome Trust Institutional Strategic Support Fund Award to BW, China Scholarship Council and the Scottish Universities Life Sciences Alliance to XW, and Darwin Trust PhD Scholarships to TYHH.

References Abil Z, Ellefson JW, Gollihar JD, Watkins E, Ellington AD (2017) Compartmentalized partnered replication for the directed evolution of genetic parts and circuits. Nat Protoc 12:2493–2512. https://doi.org/10.1038/nprot.2017.119

310

X. Wan et al.

Adams BL, Carter KK, Guo M, Wu HC, Tsao CY, Sintim HO, Valdes JJ, Bentley WE (2014) Evolved quorum sensing regulator, LsrR, for altered switching functions. ACS Synth Biol 3:210–219. https://doi.org/10.1021/sb400068z Albert H, Dale EC, Lee E, Ow DW (1995) Site-specific integration of DNA into wild-type and mutant lox sites placed in the plant genome. Plant J 7:649–659. https://doi.org/10.1046/j.1365313X.1995.7040649.x Altenbuchner J, Mattes R (2005) Escherichia coli. In: Production of recombinant proteins: novel microbial and eukaryotic expression systems. Wiley-VCH-Verl, Weinheim, pp 7–49 Amaro F, Turkewitz AP, Martín-González A, Gutiérrez JC (2011) Whole-cell biosensors for detection of heavy metal ions in environmental samples based on metallothionein promoters from Tetrahymena thermophila. Microb Biotechnol 4:513–522. https://doi.org/10.1111/j.17517915.2011.00252.x Andersen JB, Sternberg C, Poulsen LK, Bjørn SP, Givskov M, Molin S (1998) New unstable variants of green fluorescent protein for studies of transient gene expression in bacteria. Appl Environ Microbiol 64:2240–2246 Anderson JC, Clarke EJ, Arkin AP, Voigt CA (2006) Environmentally controlled invasion of cancer cells by engineered bacteria. J Mol Biol 355:619–627. https://doi.org/10.1016/j. jmb.2005.10.076 Anderson JC, Voigt CA, Arkin AP (2007) Environmental signal integration by a modular and gate. Mol Syst Biol 3:133. https://doi.org/10.1038/msb4100173 Andres J, Bertin PN (2016) The microbial genomics of arsenic. FEMS Microbiol Rev 40:299–322. https://doi.org/10.1093/femsre/fuv050 Ang J, Harris E, Hussey BJ, Kil R, McMillen DR (2013) Tuning response curves for synthetic biology. ACS Synth Biol 2:547–567. https://doi.org/10.1021/sb4000564 Archer EJ, Robinson AB, Süel GM (2012) Engineered E. coli that detect and respond to gut inflammation through nitric oxide sensing. ACS Synth Biol 1:451–457. https://doi.org/10.1021/ sb3000595 Arpino JAJ, Hancock EJ, Anderson J, Barahona M, Stan GBV, Papachristodoulou A, Polizzi K (2013) Tuning the dials of synthetic biology. Microbiol (United Kingdom) 159:1236–1253. https://doi.org/10.1099/mic.0.067975-0 Barkay T, Miller SM, Summers AO (2003) Bacterial mercury resistance from atoms to ecosystems. FEMS Microbiol Rev 27:355–384. https://doi.org/10.1016/S0168-6445(03)00046-9 Bashor CJ, Collins JJ (2018) Understanding biological regulation through synthetic biology. Annu Rev Biophys 47:399–423. https://doi.org/10.1146/annurev-biophys-070816-033903 Belkin S, Yagur-Kroll S, Kabessa Y, Korouma V, Septon T, Anati Y, Zohar-Perez C, Rabinovitz Z, Nussinovitch A, Agranat AJ (2017) Remote detection of buried landmines using a bacterial sensor. Nat Biotechnol 35:308–310. https://doi.org/10.1038/nbt.3791 Bernard E, Wang B (2017) Synthetic cell-based sensors with programmed selectivity and sensitivity. In: Methods in molecular biology, pp 349–363 Biran I, Rissin DM, Ron EZ, Walt DR (2003) Optical imaging fiber-based live bacterial cell array biosensor. Anal Biochem 315:106–113. https://doi.org/10.1016/S0003-2697(02)00700-5 Bonnet J, Subsoontorn P, Endy D (2012) Rewritable digital data storage in live cells via engineered control of recombination directionality. Proc Natl Acad Sci 109:8884–8889. https://doi.org/ 10.1073/pnas.1202344109 Bonnet J, Yin P, Ortiz ME, Subsoontorn P, Endy D (2013) Amplifying genetic logic gates. Science 340:599–603. https://doi.org/10.1126/science.1232758 Bourdeau RW, Lee-Gosselin A, Lakshmanan A, Farhadi A, Kumar SR, Nety SP, Shapiro MG (2018) Acoustic reporter genes for noninvasive imaging of microorganisms in mammalian hosts. Nature 553:86–90. https://doi.org/10.1038/nature25021 Bradley RW, Wang B (2015) Designer cell signal processing circuits for biotechnology. New Biotechnol 32:635–643. https://doi.org/10.1016/j.nbt.2014.12.009 Bradley RW, Buck M, Wang B (2016a) Recognizing and engineering digital-like logic gates and switches in gene regulatory networks. Curr Opin Microbiol 33:74–82. https://doi.org/10.1016/j. mib.2016.07.004

13

Engineering Prokaryote Synthetic Biology Biosensors

311

Bradley RW, Buck M, Wang B (2016b) Tools and principles for microbial gene circuit engineering. J Mol Biol 428:862–888. https://doi.org/10.1016/j.jmb.2015.10.004 Brewster RC, Jones DL, Phillips R (2012) Tuning promoter strength through RNA polymerase binding site design in Escherichia coli. PLoS Comput Biol 8:e1002811. https://doi.org/10.1371/ journal.pcbi.1002811 Brophy JAN, Voigt CA (2014) Principles of genetic circuit design. Nat Methods 11:508–520. https://doi.org/10.1038/nmeth.2926 Brophy JAN, Voigt CA (2016) Antisense transcription as a tool to tune gene expression. Mol Syst Biol 12:854–854. https://doi.org/10.15252/msb.20156540 Buffi N, Merulla D, Beutier J, Barbaud F, Beggah S, Van Lintel H, Renaud P, Roelof van der Meer J (2011) Development of a microfluidics biosensor for agarose-bead immobilized Escherichia coli bioreporter cells for arsenite detection in aqueous samples. Lab Chip 11:2369–2377. https:// doi.org/10.1039/c1lc20274j Bulich AA, Isenberg DL (1981) Use of the luminescent bacterial system for the rapid assessment of aquatic toxicity. ISA Trans 20:29–33 Callura JM, Dwyer DJ, Isaacs FJ, Cantor CR, Collins JJ (2010) Physiology using synthetic riboregulators. Proc Natl Acad Sci 107:15898–15903. https://doi.org/10.1073/pnas.1009747107 Cambray G, Guimaraes JC, Arkin AP (2018) Evaluation of 244,000 synthetic sequences reveals design principles to optimize translation in Escherichia coli. Nat Biotechnol 36:1005. https:// doi.org/10.1038/nbt.4238 Cameron DE, Collins JJ (2014) Tunable protein degradation in bacteria. Nat Biotechnol 32:1–8. https://doi.org/10.1038/nbt.3053 Cayron J, Prudent E, Escoffier C, Gueguen E, Mandrand-Berthelot MA, Pignol D, Garcia D, Rodrigue A (2017) Pushing the limits of nickel detection to nanomolar range using a set of engineered bioluminescent Escherichia coli. Environ Sci Pollut Res 24:4–14. https://doi.org/ 10.1007/s11356-015-5580-6 Cerminati S, Soncini FC, Checa SK (2011) Selective detection of gold using genetically engineered bacterial reporters. Biotechnol Bioeng 108:2553–2560. https://doi.org/10.1002/bit.23213 Certain LK, Way JC, Pezone MJ, Collins JJ (2017) Using engineered bacteria to characterize infection dynamics and antibiotic effects in vivo. Cell Host Microbe 22:263–268.e4. https://doi. org/10.1016/j.chom.2017.08.001 Cervantes C, Ji G, Ramirez J, Silver S (1994) Resistance to arsenic compounds in microorganisms. FEMS Microbiol Rev 15:355–367. https://doi.org/10.1111/j.1574-6976.1994.tb00145.x Chang TMS, Prakash S (2001) Procedures for microencapsulation of enzymes, cells and genetically engineered microorganisms. Mol Biotechnol 17:249–260. https://doi.org/10.1385/ MB:17:3:249 Chang HJ, Mayonove P, Zavala A, De Visch A, Minard P, Cohen-Gonsaud M, Bonnet J (2018) A modular receptor platform to expand the sensing repertoire of bacteria. ACS Synth Biol 7:166–175. https://doi.org/10.1021/acssynbio.7b00266 Chen J, Rosen BP (2014) Biosensors for inorganic and organic arsenicals. Biosensors 4:494–512. https://doi.org/10.3390/bios4040494 Chen Y, Ho JML, Shis DL, Gupta C, Long J, Wagner DS, Ott W, Josić K, Bennett MR (2018) Tuning the dynamic range of bacterial promoters regulated by ligand-inducible transcription factors. Nat Commun 9:64. https://doi.org/10.1038/s41467-017-02473-5 Cheng AA, Lu TK (2012) Synthetic biology: an emerging engineering discipline. Annu Rev Biomed Eng 14:155–178. https://doi.org/10.1146/annurev-bioeng-071811-150118 Courbet A, Endy D, Renard E, Molina F, Bonnet J (2015) Detection of pathological biomarkers in human clinical samples via amplifying genetic switches and logic gates. Sci Transl Med 7:289ra83. https://doi.org/10.1126/scitranslmed.aaa3601 Cox RS, Surette MG, Elowitz MB (2007) Programming gene expression with combinatorial promoters. Mol Syst Biol 3:145. https://doi.org/10.1038/msb4100187 Daeffler KN, Galley JD, Sheth RU, Ortiz-Velez LC, Bibb CO, Shroyer NF, Britton RA, Tabor JJ (2017) Engineering bacterial thiosulfate and tetrathionate sensors for detecting gut inflammation. Mol Syst Biol 13:923. https://doi.org/10.15252/msb.20167416

312

X. Wan et al.

Dana GV, Kuiken T, Rejeski D, Snow AA (2012) Synthetic biology: four steps to avoid a syntheticbiology disaster. Nature 483:29–29. https://doi.org/10.1038/483029a Daniel R, Rubens JR, Sarpeshkar R, Lu TK (2013) Synthetic analog computation in living cells. Nature 497:619–623. https://doi.org/10.1038/nature12148 Danino T, Prindle A, Kwong GA, Skalak M, Li H, Allen K, Hasty J, Bhatia SN (2015) Programmable probiotics for non-invasive urinary detection of cancer. Sci Transl Med 7:1–36. https:// doi.org/10.1126/scitranslmed.aaa3519 Daunert S, Barrett G, Feliciano JS, Shetty RS, Shrestha S, Smith-Spencer W (2000) Genetically engineered whole-cell sensing systems: coupling biological recognition with reporter genes. Chem Rev 100:2705–2738. https://doi.org/10.1021/cr990115p De Mora K, Joshi N, Balint BL, Ward FB, Elfick A, French CE (2011) A pH-based biosensor for detection of arsenic in drinking water. Anal Bioanal Chem 400:1031–1039. https://doi.org/ 10.1007/s00216-011-4815-8 Didovyk A, Tonooka T, Tsimring L, Hasty J (2017) Rapid and scalable preparation of bacterial lysates for cell-free gene expression. ACS Synth Biol 6:2198–2208. https://doi.org/10.1021/ acssynbio.7b00253 Duan F, March JC (2010) Engineered bacterial communication prevents Vibrio cholerae virulence in an infant mouse model. Proc Natl Acad Sci U S A 107:11260–11264. https://doi.org/10.1073/ pnas.1001294107 Duyen TTM, Matsuura H, Ujiie K, Muraoka M, Harada K, Hirata K (2017) Paper-based colorimetric biosensor for antibiotics inhibiting bacterial protein synthesis. J Biosci Bioeng 123:96–100. https://doi.org/10.1016/j.jbiosc.2016.07.015 Egbert RG, Klavins E (2012) Fine-tuning gene networks using simple sequence repeats. Proc Natl Acad Sci U S A 109:16817–16822. https://doi.org/10.1073/pnas.1205693109 Ellefson JW, Ledbetter MP, Ellington AD (2018) Directed evolution of a synthetic phylogeny of programmable Trp repressors article. Nat Chem Biol 14:361–367. https://doi.org/10.1038/ s41589-018-0006-7 Ellis T, Wang X, Collins JJ (2009) Diversity-based, model-guided construction of synthetic gene networks with predicted functions. Nat Biotechnol 27:465–471. https://doi.org/10.1038/ nbt.1536.DIVERSITY-BASED Endy D (2005) Foundations for engineering biology. Nature 438:449–453. https://doi.org/10.1038/ nature04342 Esvelt KM, Carlson JC, Liu DR (2011) A system for the continuous directed evolution of biomolecules. Nature 472:499–503. https://doi.org/10.1038/nature09929 Farzadfard F, Lu TK (2014) Genomically encoded analog memory with precise in vivo dna writing in living cell populations. Science 346:1256272–1256272. https://doi.org/10.1126/ science.1256272 Fernandez-Rodriguez J, Voigt CA (2016) Post-translational control of genetic circuits using Potyvirus proteases. Nucleic Acids Res 44:6493–6502. https://doi.org/10.1093/nar/gkw537 Fernandez-Rodriguez J, Moser F, Song M, Voigt CA (2017) Engineering RGB color vision into Escherichia coli. Nat Chem Biol 13:706–708. https://doi.org/10.1038/nchembio.2390 Folliard T, Steel H, Prescott TP, Wadhams G, Rothschild LJ, Papachristodoulou A (2017) A synthetic recombinase-based feedback loop results in robust expression. ACS Synth Biol 6:1663–1671. https://doi.org/10.1021/acssynbio.7b00131 Friedland AE, Lu TK, Wang X, Shi D, Church G, Collins JJ (2009) Synthetic gene networks that count. Science 324:1199–1202. https://doi.org/10.1126/science.1172005 Fritz G, Megerle JA, Westermayer SA, Brick D, Heermann R, Jung K, Rädler JO, Gerland U (2014) Single cell kinetics of phenotypic switching in the arabinose utilization system of E. coli. PLoS One 9:e89532. https://doi.org/10.1371/journal.pone.0089532 Fujimoto H, Wakabayashi M, Yamashiro H, Maeda I, Isoda K, Kondoh M, Kawase M, Miyasaka H, Yagi K (2006) Whole-cell arsenite biosensor using photosynthetic bacterium Rhodovulum sulfidophilum: Rhodovulum sulfidophilum as an arsenite biosensor. Appl Microbiol Biotechnol 73:332–338. https://doi.org/10.1007/s00253-006-0483-6

13

Engineering Prokaryote Synthetic Biology Biosensors

313

Gardner TS, Cantor CR, Collins JJ (2000) Construction of a genetic toggle switch in Escherichia coli. Nature 403:339–342. https://doi.org/10.1038/35002131 Gui Q, Lawson T, Shan S, Yan L, Liu Y (2017) The application of whole cell-based biosensors for use in environmental analysis and in medical diagnostics. Sensors 17:1623. https://doi.org/ 10.3390/s17071623 Gupta S, Bram EE, Weiss R (2013) Genetically programmable pathogen sense and destroy. ACS Synth Biol 2:715–723. https://doi.org/10.1021/sb4000417 Guzina J, Djordjevic M (2017) Mix-and-matching as a promoter recognition mechanism by ECF σ factors. BMC Evol Biol 17:1–14. https://doi.org/10.1186/s12862-016-0865-z Hakkila KM, Nikander PA, Junttila SM, Lamminmäki UJ, Virta MP (2011) Cd-specific mutants of mercury-sensing regulatory protein MerR, generated by directed evolution. Appl Environ Microbiol 77:6215–6224. https://doi.org/10.1128/AEM.00662-11 Hao N, Krishna S, Ahlgren-Berg A, Cutts EE, Shearwin KE, Dodd IB (2014) Road rules for traffic on DNA – systematic analysis of transcriptional roadblocking in vivo. Nucleic Acids Res 42:8861–8872. https://doi.org/10.1093/nar/gku627 Ho CL, Tan HQ, Chua KJ, Kang A, Lim KH, Ling KL, Yew WS, Lee YS, Thiery JP, Chang MW (2018) Engineered commensal microbes for diet-mediated colorectal-cancer chemoprevention. Nat Biomed Eng 2:27–37. https://doi.org/10.1038/s41551-017-0181-y Holliger P, Hudson PJ (2005) Engineered antibody fragments and the rise of single domains. Nat Biotechnol 23:1126–1136. https://doi.org/10.1038/nbt1142 Hooshangi S, Thiberge S, Weiss R (2005) Ultrasensitivity and noise propagation in a synthetic transcriptional cascade. Proc Natl Acad Sci U S A 102:3581–3586. https://doi.org/10.1073/ pnas.0408507102 Huang CW, Wei CC, Liao VHC (2015a) A low cost color-based bacterial biosensor for measuring arsenic in groundwater. Chemosphere 141:44–49. https://doi.org/10.1016/j. chemosphere.2015.06.011 Huang CW, Yang SH, Sun MW, Liao VHC (2015b) Development of a set of bacterial biosensors for simultaneously detecting arsenic and mercury in groundwater. Environ Sci Pollut Res 22:10206–10213. https://doi.org/10.1007/s11356-015-4216-1 Hwang IY, Tan MH, Koh E, Ho CL, Poh CL, Chang MW (2014) Reprogramming microbes to be pathogen-seeking killers. ACS Synth Biol 3:228–237. https://doi.org/10.1021/sb400077j Hwang IY, Koh E, Kim HR, Yew WS, Chang MW (2016) Reprogrammable microbial cell-based therapeutics against antibiotic-resistant bacteria. Drug Resist Updat 27:59–71. https://doi.org/ 10.1016/j.drup.2016.06.002 Hynninen A, Tõnismann K, Virta M (2010) Improving the sensitivity of bacterial bioreporters for heavy metals. Bioeng Bugs 1:132–138. https://doi.org/10.4161/bbug.1.2.10902 Inniss MC, Silver PA (2013) Building synthetic memory. Curr Biol 23:R812–R816. https://doi.org/ 10.1016/j.cub.2013.06.047 Jiang F, Doudna JA (2017) CRISPR–Cas9 structures and mechanisms. Annu Rev Biophys 46:505–529. https://doi.org/10.1146/annurev-biophys-062215-010822 Joe MH, Lee KH, Lim SY, Im SH, Song HP, Lee IS, Kim DH (2012) Pigment-based whole-cell biosensor system for cadmium detection using genetically engineered Deinococcus radiodurans. Bioprocess Biosyst Eng 35:265–272. https://doi.org/10.1007/s00449-011-0610-3 Johns NI, Gomes ALC, Yim SS, Yang A, Blazejewski T, Smillie CS, Smith MB, Alm EJ, Kosuri S, Wang HH (2018) Metagenomic mining of regulatory elements enables programmable species-selective gene expression. Nat Methods 15:323–329. https://doi.org/10.1038/ nmeth.4633 Joshi N, Wang X, Montgomery L, Elfick A, French CE (2009) Novel approaches to biosensors for detection of arsenic in drinking water. Desalination 248:517–523. https://doi.org/10.1016/j. desal.2008.05.096 Jovanovic M, James EH, Burrows PC, Rego FGM, Buck M, Schumacher J (2011) Regulation of the co-evolved HrpR and HrpS AAA+ proteins required for Pseudomonas syringae pathogenicity. Nat Commun 2:177. https://doi.org/10.1038/ncomms1177

314

X. Wan et al.

Karig D, Weiss R (2005) Signal-amplifying genetic circuit enables in vivo observation of weak promoter activation in the Rhl quorum sensing system. Biotechnol Bioeng 89:709–718. https:// doi.org/10.1002/bit.20371 Kawano M, Aravind L, Storz G (2007) An antisense RNA controls synthesis of an SOS-induced toxin evolved from an antitoxin. Mol Microbiol 64:738–754. https://doi.org/10.1111/j.13652958.2007.05688.x Kim HJ, Lim JW, Jeong H, Lee SJ, Lee DW, Kim T, Lee SJ (2016) Development of a highly specific and sensitive cadmium and lead microbial biosensor using synthetic CadC-T7 genetic circuitry. Biosens Bioelectron 79:701–708. https://doi.org/10.1016/j.bios.2015.12.101 Kim HJ, Jeong H, Lee SJ (2018) Synthetic biology for microbial heavy metal biosensors. Anal Bioanal Chem 410:1191–1203. https://doi.org/10.1007/s00216-017-0751-6 King JMH, DiGrazia PM, Applegate B, Burlage R, Sanseverino J, Dunbar P, Larimer F, Sayler GS (1990) Rapid, sensitive bioluminescent reporter technology for naphthalene exposure and biodegradation. Science 249:778–781. https://doi.org/10.1126/ science.249.4970.778 Kosuri S, Goodman DB, Cambray G, Mutalik VK, Gao Y, Arkin AP, Endy D, Church GM (2013) Composability of regulatory sequences controlling transcription and translation in Escherichia coli. Proc Natl Acad Sci U S A 110:14024–14029. https://doi.org/10.1073/ pnas.1301301110 Kotula JW, Kerns SJ, Shaket LA, Siraj L, Collins JJ, Way JC, Silver PA (2014) Programmable bacteria detect and record an environmental signal in the mammalian gut. Proc Natl Acad Sci 111:4838–4843. https://doi.org/10.1073/pnas.1321321111 Kwon YC, Jewett MC (2015) High-throughput preparation methods of crude extract for robust cellfree protein synthesis. Sci Rep 5:1–8. https://doi.org/10.1038/srep08663 Lasa I, Toledo-Arana A, Dobin A, Villanueva M, de los Mozos IR, Vergara-Irigaray M, Segura V, Fagegaltier D, Penades JR, Valle J, Solano C, Gingeras TR (2011) Genome-wide antisense transcription drives mRNA processing in bacteria. Proc Natl Acad Sci 108:20172–20177. https://doi.org/10.1073/pnas.1113521108 Li L, Liang J, Hong W, Zhao Y, Sun S, Yang X, Xu A, Hang H, Wu L, Chen S (2015) Evolved bacterial biosensor for arsenite detection in environmental water. Environ Sci Technol 49:6149–6155. https://doi.org/10.1021/acs.est.5b00832 Liu Q, Schumacher J, Wan X, Lou C, Wang B (2018) Orthogonality and burdens of heterologous and gate gene circuits in E. coli. ACS Synth Biol 7:553–564. https://doi.org/10.1021/ acssynbio.7b00328 Ma KC, Perli SD, Lu TK (2016) Foundations and emerging paradigms for computing in living cells. J Mol Biol 428:893–915. https://doi.org/10.1016/j.jmb.2016.02.018 Ma D, Shen L, Wu K, Diehnelt CW, Green AA (2018) Low-cost detection of norovirus using paperbased cell-free systems and synbody-based viral enrichment. Synth Biol 3:ysy018. https://doi. org/10.1093/synbio/ysy018/5102817 Mahr R, Frunzke J (2016) Transcription factor-based biosensors in biotechnology: current state and future prospects. Appl Microbiol Biotechnol 100:79–90. https://doi.org/10.1007/s00253-0157090-3 Maier JAH, Möhrle R, Jeltsch A (2017) Design of synthetic epigenetic circuits featuring memory effects and reversible switching based on DNA methylation. Nat Commun 8:15336. https://doi. org/10.1038/ncomms15336 Mannan AA, Liu D, Zhang F, Oyarzún DA (2017) Fundamental design principles for transcriptionfactor-based metabolite biosensors. ACS Synth Biol 6:1851–1859. https://doi.org/10.1021/ acssynbio.7b00172 Merulla D, van der Meer JR (2016) Regulatable and modulable background expression control in prokaryotic synthetic circuits by auxiliary repressor binding sites. ACS Synth Biol 5:36–45. https://doi.org/10.1021/acssynbio.5b00111 Merulla D, Hatzimanikatis V, van der Meer JR (2013) Tunable reporter signal production in feedback-uncoupled arsenic bioreporters. Microb Biotechnol 6:503–514. https://doi.org/ 10.1111/1751-7915.12031

13

Engineering Prokaryote Synthetic Biology Biosensors

315

Mimee M, Tucker AC, Voigt CA, Lu TK (2015) Programming a human commensal bacterium, bacteroides thetaiotaomicron, to sense and respond to stimuli in the Murine Gut Microbiota. Cell Syst 1:62–71. https://doi.org/10.1016/j.cels.2015.06.001 Mimee M, Nadeau P, Hayward A, Carim S, Flanagan S, Jerger L, Collins J, McDonnell S, Swartwout R, Citorik RJ, Bulović V, Langer R, Traverso G, Chandrakasan AP, Lu TK (2018) An ingestible bacterial-electronic system to monitor gastrointestinal health. Science 360:915–918. https://doi.org/10.1126/science.aas9315 Misra TK, Brown NL, Haberstroh L, Schmidt A, Goddette D, Silver S (1985) Mercuric reductase structural genes from plasmid R100 and transposon Tn501: functional domains of the enzyme. Gene 34:253–262. https://doi.org/10.1016/0378-1119(85)90134-9 Moon TS, Clarke EJ, Groban ES, Tamsir A, Clark RM, Eames M, Kortemme T, Voigt CA (2011) Construction of a genetic multiplexer to toggle between chemosensory pathways in Escherichia coli. J Mol Biol 406:215–227. https://doi.org/10.1016/j.jmb.2010.12.019 Moon TS, Lou C, Tamsir A, Stanton BC, Voigt CA (2012) Genetic programs constructed from layered logic gates in single cells. Nature 491:249–253. https://doi.org/10.1038/nature11516 Mutalik VK, Guimaraes JC, Cambray G, Lam C, Christoffersen MJ, Mai QA, Tran AB, Paull M, Keasling JD, Arkin AP, Endy D (2013) Precise and reliable gene expression via standard transcription and translation initiation elements. Nat Methods 10:354–360. https://doi.org/ 10.1038/nmeth.2404 Nassif N, Bouvet O, Noelle Rager M, Roux C, Coradin T, Livage J (2002) Living bacteria in silica gels. Nat Mater 1:42–44. https://doi.org/10.1038/nmat709 Nealson KH, Myers CR, Wimpee BB (1991) Isolation and identification of manganese-reducing bacteria and estimates of microbial Mn(IV)-reducing potential in the Black Sea. Deep Sea Res Part A 38:S907–S920. https://doi.org/10.1016/S0198-0149(10)80016-0 Nealson KH, Belz A, McKee B (2002) Breathing metals as a way of life: geobiology in action. Anton Leeuw Int J Gen Mol Microbiol 81:215–222. https://doi.org/10.1023/A:1020518818647 Nielsen AAK, Der BS, Shin J, Vaidyanathan P, Paralanov V, Strychalski EA, Ross D, Densmore D, Voigt CA (2016) Genetic circuit design automation. Science 352:aac7341. https://doi.org/ 10.1126/science.aac7341 Nistala GJ, Wu K, Rao CV, Bhalerao KD (2010) A modular positive feedback-based gene amplifier. J Biol Eng 4:4. https://doi.org/10.1186/1754-1611-4-4 Oberdoerffer P (2003) Unidirectional Cre-mediated genetic inversion in mice using the mutant loxP pair lox66/lox71. Nucleic Acids Res 31:140e. https://doi.org/10.1093/nar/gng140 Olorunniji FJ, Rosser SJ, Stark WM (2016) Site-specific recombinases: molecular machines for the Genetic Revolution. Biochem J 473:673–684. https://doi.org/10.1042/BJ20151112 Olson EJ, Hartsough LA, Landry BP, Shroff R, Tabor JJ (2014) Characterizing bacterial gene circuit dynamics with optically programmed gene expression signals. Nat Methods 11:449–455. https://doi.org/10.1038/nmeth.2884 Papi RM, Chaitidou SA, Trikka FA, Kyriakidis DA (2005) Encapsulated Escherichia coli in alginate beads capable of secreting a heterologous pectin lyase. Microb Cell Factories 4:35. https://doi.org/10.1186/1475-2859-4-35 Pardee K, A a G, Ferrante T, Cameron DE, Daleykeyser A, Yin P (2014) Paper-based synthetic gene networks. Cell 159:940–954. https://doi.org/10.1016/j.cell.2014.10.004 Pardee K, Green AA, Takahashi MK, Braff D, Lambert G, Lee JW, Ferrante T, Ma D, Donghia N, Fan M, Daringer NM, Bosch I, Dudley DM, O’Connor DH, Gehrke L, Collins JJ (2016a) Rapid, low-cost detection of Zika virus using programmable biomolecular components. Cell 165:1255–1266. https://doi.org/10.1016/j.cell.2016.04.059 Pardee K, Slomovic S, Nguyen PQ, Lee JW, Donghia N, Burrill D, Ferrante T, McSorley FR, Furuta Y, Vernet A, Lewandowski M, Boddy CN, Joshi NS, Collins JJ ((2016b)) Portable, on-demand biomolecular manufacturing. Cell 167:248–259.e12. https://doi.org/10.1016/j.cell.2016.09.013 Peking iGEM (2013) Aromatics Scouts. http://2013.igem.org/Team:Peking/Project/BioSensors. Accessed 3 Jun 2018 Pelechano V, Steinmetz LM (2013) Gene regulation by antisense transcription. Nat Rev Genet 14:880–893. https://doi.org/10.1038/nrg3594

316

X. Wan et al.

Pellinen T, Huovinen T, Karp M (2004) A cell-free biosensor for the detection of transcriptional inducers using firefly luciferase as a reporter. Anal Biochem 330:52–57. https://doi.org/10.1016/ j.ab.2004.03.064 Perez JG, Stark JC, Jewett MC (2016) Cell-free synthetic biology: engineering beyond the cell. Cold Spring Harb Perspect Biol 8:a023853. https://doi.org/10.1101/cshperspect.a023853 Pinheiro VB, Taylor AI, Cozens C, Abramov M, Renders M, Zhang S, Chaput JC, Wengel J, PeakChew SY, McLaughlin SH, Herdewijn P, Holliger P (2012) Synthetic genetic polymers capable of heredity and evolution. Science 336:341–344. https://doi.org/10.1126/science.1217622 Power B, Liu X, Germaine KJ, Ryan D, Brazil D, Dowling DN (2011) Alginate beads as a storage, delivery and containment system for genetically modified PCB degrader and PCB biosensor derivatives of Pseudomonas fluorescens F113. J Appl Microbiol 110:1351–1358. https://doi. org/10.1111/j.1365-2672.2011.04993.x Pu J, Zinkus-Boltz J, Dickinson BC (2017) Evolution of a split RNA polymerase as a versatile biosensor platform. Nat Chem Biol 13:432–438. https://doi.org/10.1038/nchembio.2299 Purnick PEM, Weiss R (2009) The second wave of synthetic biology: from modules to systems. Nat Rev Mol Cell Biol 10:410–422. https://doi.org/10.1038/nrm2698 Randall A, Guye P, Gupta S, Duportet X, Weiss R (2011) Design and connection of robust genetic circuits. Methods Enzymol 497:159–186. https://doi.org/10.1016/B978-0-12-385075-100007-X Riglar DT, Giessen TW, Baym M, Kerns SJ, Niederhuber MJ, Bronson RT, Kotula JW, Gerber GK, Way JC, Silver PA (2017) Engineered bacteria can function in the mammalian gut long-term as live diagnostics of inflammation. Nat Biotechnol 35:653–658. https://doi.org/10.1038/nbt.3879 Roquet N, Lu TK (2014) Digital and analog gene circuits for biotechnology. Biotechnol J 9:597–608. https://doi.org/10.1002/biot.201300258 Roquet N, Soleimany AP, Ferris AC, Aaronson S, Lu TK (2016) Synthetic recombinase-based State machines in living cells. Science 353:aad8559. https://doi.org/10.1126/science.aad8559 Rubens JR, Selvaggio G, Lu TK (2016) Synthetic mixed-signal computation in living cells. Nat Commun 7:11658. https://doi.org/10.1038/ncomms11658 Saeidi N, Wong CK, Lo T-M, Nguyen HX, Ling H, Leong SSJ, Poh CL, Chang MW (2011) Engineering microbes to sense and eradicate Pseudomonas aeruginosa, a human pathogen. Mol Syst Biol 7:521. https://doi.org/10.1038/msb.2011.55 Saha RP, Samanta S, Patra S, Sarkar D, Saha A, Singh MK (2017) Metal homeostasis in bacteria: the role of ArsR–SmtB family of transcriptional repressors in combating varying metal concentrations in the environment. BioMetals 30:459–503. https://doi.org/10.1007/s10534-017-0020-3 Salis HM, Mirsky EA, Voigt CA (2009) Automated design of synthetic ribosome binding sites to control protein expression. Nat Biotechnol 27:946–950. https://doi.org/10.1038/nbt.1568 Schaerli Y, Gili M, Isalan M (2014) A split intein T7 RNA polymerase for transcriptional ANDlogic. Nucleic Acids Res 42:12322–12328. https://doi.org/10.1093/nar/gku884 Schnütgen F, Doerflinger N, Calléja C, Wendling O, Chambon P, Ghyselinck NB (2003) A directional strategy for monitoring Cre-mediated recombination at the cellular level in the mouse. Nat Biotechnol 21:562–565. https://doi.org/10.1038/nbt811 Selifonova OV, Eaton RW (1996) Use of an ipb-lux fusion to study regulation of the isopropylbenzene catabolism operon of Pseudomonas putida RE204 and to detect hydrophobic pollutants in the environment. Appl Environ Microbiol 62:778–783 Shalem O, Sanjana NE, Zhang F (2015) High-throughput functional genomics using CRISPRCas9. Nat Rev Genet 16:299–311. https://doi.org/10.1038/nrg3899 Sharma J, Mahajan R, Gupta V (2010) Comparison and suitability of gel matrix for entrapping higher content of enzymes for commercial applications. Indian J Pharm Sci 72:223. https://doi. org/10.4103/0250-474X.65010 Sheth RU, Yim SS, Wu FL, Wang HH (2017) Multiplex recording of cellular events over time on CRISPR biological tape. Science 358:1457–1461. https://doi.org/10.1126/science.aao0958 Shi W, Dong J, Scott RA, Ksenzenko MY, Rosen BP (1996) The role of arsenic-thiol interactions in metalloregulation of the ars operon. J Biol Chem 271:9291–9297. https://doi.org/10.1074/ jbc.271.16.9291

13

Engineering Prokaryote Synthetic Biology Biosensors

317

Shin HJ (2012) Agarose-gel-immobilized recombinant bacterial biosensors for simple and disposable on-site detection of phenolic compounds. Appl Microbiol Biotechnol 93:1895–1904. https://doi.org/10.1007/s00253-011-3700-x Shis DL, Bennett MR (2013) Library of synthetic transcriptional AND gates built with split T7 RNA polymerase mutants. Proc Natl Acad Sci 110:5028–5033. https://doi.org/10.1073/ pnas.1220157110 Siegfried K, Endes C, Bhuiyan AFMK, Kuppardt A, Mattusch J, van der Meer JR, Chatzinotas A, Harms H (2012) Field testing of arsenic in groundwater samples of Bangladesh using a test kit based on lyophilized bioreporter bacteria. Environ Sci Technol 46:3281–3287. https://doi.org/ 10.1021/es203511k Silver S, Phung LT (2005) Genes and enzymes involved in bacterial oxidation and reduction of inorganic arsenic. Appl Environ Microbiol 71:599–608. https://doi.org/10.1128/AEM.71.2.599608.2005 Siuti P, Yazbek J, Lu TK (2013) Synthetic circuits integrating logic and memory in living cells. Nat Biotechnol 31:448–452. https://doi.org/10.1038/nbt.2510 Stocker J, Balluch D, Gsell M, Harms H, Feliciano J, Daunert S, Malik KA, van der Meer JR (2003) Development of a set of simple bacterial biosensors for quantitative and rapid measurements of arsenite and arsenate in potable water. Environ Sci Technol 37:4743–4750. https://doi.org/ 10.1021/es034258b Stoyanov JV, Browns NL (2003) The Escherichia coli copper-responsive copA promoter is activated by gold. J Biol Chem 278:1407–1410. https://doi.org/10.1074/jbc.C200580200 Tang W, Liu DR (2018) Rewritable multi-event analog recording in bacterial and mammalian cells. Science 360:eaap8992. https://doi.org/10.1126/science.aap8992 Tong H-W, Mutlu BR, Wackett LP, Aksan A (2014) Manufacturing of bioreactive nanofibers for bioremediation. Biotechnol Bioeng 111:1483–1493. https://doi.org/10.1002/bit.25208 van der Meer JR, Belkin S (2010) Where microbiology meets microengineering: design and applications of reporter bacteria. Nat Rev Microbiol 8:511–522. https://doi.org/10.1038/nrmicro2392 Volpetti F, Petrova E, Maerkl SJ (2017) A microfluidic biodisplay. ACS Synth Biol 6:1979–1987. https://doi.org/10.1021/acssynbio.7b00088 Wackwitz A, Harms H, Chatzinotas A, Breuer U, Vogne C, van der Meer JR (2008) Internal arsenite bioassay calibration using multiple bioreporter cell lines. Microb Biotechnol 1:149–157. https:// doi.org/10.1111/j.1751-7915.2007.00011.x Wan X, Volpetti F, Petrova E, French C, Maerkl SJ, Wang B (2019) Cascaded amplifying circuits enable ultrasensitive cellular sensors for toxic metals. Nat Chem Biol. https://doi.org/10.1038/ s41589-019-0244-3 Wang B, Buck M (2012) Customizing cell signaling using engineered genetic logic circuits. Trends Microbiol 20:376–384. https://doi.org/10.1016/j.tim.2012.05.001 Wang B, Buck M (2014) Rapid engineering of versatile molecular logic gates using heterologous genetic transcriptional modules. Chem Commun 50:11642–11644. https://doi.org/10.1039/ C4CC05264A Wang B, Kitney RI, Joly N, Buck M (2011a) Engineering modular and orthogonal genetic logic gates for robust digital-like synthetic biology. Nat Commun 2:508. https://doi.org/10.1038/ncomms1516 Wang Y, Yau YY, Perkins-Balding D, Thomson JG (2011b) Recombinase technology: applications and possibilities. Plant Cell Rep 30:267–285. https://doi.org/10.1007/s00299-010-0938-1 Wang B, Barahona M, Buck M (2013) A modular cell-based biosensor using engineered genetic logic circuits to detect and integrate multiple environmental signals. Biosens Bioelectron 40:368–376. https://doi.org/10.1016/j.bios.2012.08.011 Wang B, Barahona M, Buck M (2014) Engineering modular and tunable genetic amplifiers for scaling transcriptional signals in cascaded gene networks. Nucleic Acids Res 42:9484–9492. https://doi.org/10.1093/nar/gku593 Wang B, Barahona M, Buck M (2015) Amplification of small molecule-inducible gene expression via tuning of intracellular receptor densities. Nucleic Acids Res 43:1955–1964. https://doi.org/ 10.1093/nar/gku1388

318

X. Wan et al.

Watstein DM, Styczynski MP (2018) Development of a pigment-based whole-cell zinc biosensor for human serum. ACS Synth Biol 7:267–275. https://doi.org/10.1021/acssynbio.7b00292 Way JC, Collins JJ, Keasling JD, Silver PA (2014) Integrating biological redesign: where synthetic biology came from and where it needs to go. Cell 157:151–161. https://doi.org/10.1016/j. cell.2014.02.039 Webster DP, M a TA, Doud DFR, Chakravorty A, Holmes EC, Radens CM, Sureka S, J a G, Angenent LT (2014) An arsenic-specific biosensor with genetically engineered Shewanella oneidensis in a bioelectrochemical system. Biosens Bioelectron 62:320–324. https://doi.org/ 10.1016/j.bios.2014.07.003 Wen KY, Cameron L, Chappell J, Jensen K, Bell DJ, Kelwick R, Kopniczky M, Davies JC, Filloux A, Freemont PS (2017) A cell-free biosensor for detecting quorum sensing molecules in P. aeruginosa-infected respiratory samples. ACS Synth Biol 6:2293–2301. https://doi.org/ 10.1021/acssynbio.7b00219 Wright O, Stan G-B, Ellis T (2013) Building-in biosafety for synthetic biology. Microbiology 159:1221–1235. https://doi.org/10.1099/mic.0.066308-0 Wright O, Delmans M, Stan GB, Ellis T (2015) GeneGuard: a modular plasmid system designed for biosafety. ACS Synth Biol 4:307–316. https://doi.org/10.1021/sb500234s Wu CH, Le D, Mulchandani A, Chen W (2009) Optimization of a whole-cell cadmium sensor with a toggle gene circuit. Biotechnol Prog 25:898–903. https://doi.org/10.1002/btpr.203 Wu F, Zhang Q, Wang X (2018) Design of adjacent transcriptional regions to tune gene expression and facilitate circuit construction. Cell Syst 6:206–215.e6. https://doi.org/10.1016/j. cels.2018.01.010 Xiang Y, Dalchau N, Wang B (2018) Scaling up genetic circuit design for cellular computing: advances and prospects. Nat Comput 17:833–853. https://doi.org/10.1007/s11047-018-9715-9 Xue H, Shi H, Yu Z, He S, Liu S, Hou Y, Pan X, Wang H, Zheng P, Cui C, Viets H, Liang J, Zhang Y, Chen S, Zhang HM, Ouyang Q (2014) Design, construction, and characterization of a set of biosensors for aromatic compounds. ACS Synth Biol 3:1011–1014. https://doi.org/10.1021/ sb500023f Yang L, Nielsen AAK, Fernandez-Rodriguez J, McClune CJ, Laub MT, Lu TK, Voigt CA (2014) Permanent genetic memory with >1 byte capacity. Nat Methods 11:1261–1266. https://doi.org/ 10.1038/nmeth.3147 Yu B, Yang M, Shi L, Yao Y, Jiang Q, Li X, Tang LH, Zheng BJ, Yuen KY, Smith DK, Song E, Huang JD (2012) Explicit hypoxia targeting with tumor suppression by creating an “obligate” anaerobic Salmonella Typhimurium strain. Sci Rep 2:436. https://doi.org/10.1038/srep00436 Zhang F, Keasling J (2011) Biosensors and their applications in microbial metabolic engineering. Trends Microbiol 19:323–329. https://doi.org/10.1016/j.tim.2011.05.003 Zhang F, Carothers JM, Keasling JD (2012) Design of a dynamic sensor-regulator system for production of chemicals and fuels derived from fatty acids. Nat Biotechnol 30:354–359. https:// doi.org/10.1038/nbt.2149 Zhang H, Sheng Y, Wu Q, Liu A, Lu Y, Yin Z, Cao Y, Zeng W, Ouyang Q (2013) Rational design of a biosensor circuit with semi-log dose-response function in Escherichia coli. Quant Biol 1:209–220. https://doi.org/10.1007/s40484-013-0020-4 Zhang H, Lin M, Shi H, Ji W, Huang L, Zhang X, Shen S, Gao R, Wu S, Tian C, Yang Z, Zhang G, He S, Wang H, Saw T, Chen Y, Ouyang Q (2014) Programming a Pavlovian-like conditioning circuit in Escherichia coli. Nat Commun 5:3102. https://doi.org/10.1038/ncomms4102 Zong Y, Zhang HM, Lyu C, Ji X, Hou J, Guo X, Ouyang Q, Lou C (2017) Insulated transcriptional elements enable precise design of genetic circuits. Nat Commun 8:52. https://doi.org/10.1038/ s41467-017-00063-z

Cell-Free Synthetic Biology Biosensors

14

David K. Karig, Allison Reno, Lauren Elizabeth Franklin, and Andrea C. Timm

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cell-Free Protein Expression Platforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E. coli Cell-Free Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Variety of Available Platforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sensing Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Protein-Based Sensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Nucleotide-Based Sensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Regulation with Synthetic Gene Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Deployment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sample Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Encapsulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions and Future Directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

320 321 321 323 324 325 326 327 328 329 329 330 331 332 332

D. K. Karig (*) Department of Bioengineering, Clemson University, Clemson, SC, USA Research and Exploratory Development Department, Johns Hopkins University Applied Physics Laboratory, Laurel, MD, USA e-mail: [email protected] A. Reno · L. E. Franklin Department of Bioengineering, Clemson University, Clemson, SC, USA A. C. Timm Research and Exploratory Development Department, Johns Hopkins University Applied Physics Laboratory, Laurel, MD, USA © Springer Nature Switzerland AG 2022 G. Thouand (ed.), Handbook of Cell Biosensors, https://doi.org/10.1007/978-3-030-23217-7_134

319

320

D. K. Karig et al.

Abstract

The field of synthetic biology offers powerful tools for biosensor development that enable the integration of sensing, regulation, and response components. Cellfree protein expression systems offer an appealing platform for harnessing synthetic biology capabilities for a variety of sensing applications. Cell-free systems consist of components, typically derived from living cells, needed to reconstitute protein expression in vitro. In vitro operation can offer improved safety and flexibility profiles as compared to the use of living cells. Although cost, yield, and scale of cell-free protein expression were long barriers to practical applications of cell-free technology, these have all improved significantly, making biosensing applications appealing. Consequently, different sensing strategies have now been demonstrated for a variety of different ligands, using both protein and nucleic acid sensing components. In addition, a growing number and diversity of regulatory networks are being developed, and these can be incorporated to enable basic signal processing, integration, and output. Furthermore, several recent advances in robustness, shelf-life, and operation platforms have generated a variety of practical deployment options for cell-free biosensors. Therefore, this chapter discusses developments in cell-free biosensing strategies, regulatory networks, outputs, and deployment. Keywords

Cell-free · Synthetic biology · Biosensor · Protein expression · Gene networks

Introduction Synthetic biology offers transformative tools for biosensor development that expand the range of possible sensor targets and increase sophistication through the use of gene regulatory networks. While the majority of efforts have been conducted in living cells to date, cell-free protein expression systems present many advantages in terms of flexibility and simplicity of implementation and operation. Cell-free protein expression systems consist of machinery for transcription and translation, typically derived from cell extracts, in addition to reagents for fueling and stabilizing protein expression, including buffers, nucleotides, amino acids, and energy sources. Although cost, yield, and scale challenges have historically limited cell-free protein expression systems to basic research endeavors, advances in cell-free preparations have made them an increasingly appealing platform for applications (Silverman et al. 2019; Khambhati et al. 2019; Carlson et al. 2012; Zawada et al. 2011). Indeed, cellfree synthetic biology holds great promise for the development of low cost, deployable biosensors (Soltani et al. 2018). Cell-free biosensors offer a number of advantages over whole cell biosensors (Karig 2017). From a biosensor design and development perspective, cell-free platforms facilitate rapid screening and prototyping. No transformation steps are

14

Cell-Free Synthetic Biology Biosensors

321

required, and plasmids or PCR products harboring different genetic components may be readily combined in expression reactions to rapidly test different gene network implementations. Building on these advantages, a four-component genetic switch was developed in under 8 h (Sun et al. 2014), and new biosensors have been developed within a day (Pardee et al. 2014). From a safety perspective, cell-free systems can alleviate concerns associated with the release of genetically modified organisms (GMOs) into the environment (Perez et al. 2016; Smith et al. 2014b). From a performance perspective, cell-free systems offer several appealing features as well in terms of durability and efficiency. They can operate in the presence of many toxins that would inhibit or kill live cells. Thus, the target for sensing and/or other components of samples may be better tolerated to avoid false negatives (Pellinen et al. 2004). In addition, all energy resources can be devoted to the biosensor gene network, as opposed to supporting extraneous cell processes and self-replication. Finally, whereas living cells can evolve away from the intended application, this concern is alleviated in cell-free contexts. More generally, even compared to a wider array of biosensing strategies, cell-free protein systems offer a number of advantages in terms of cost, portability, and flexibility. Key proteins can be expressed at the time of application as opposed to purified ahead of time, potentially leading to cost savings. Cell-free synthetic biology offers a degree of modularity and scalability, as sensors can be linked to a variety of reporters and even gene regulatory networks, enabling sophisticated outputs and integration with other sensors. Due to the ease of measuring reporters such as fluorescent proteins, analysis can typically be performed on site. Therefore, complex, bulky, and/or costly instruments are not needed, such as qPCR machines or liquid chromatography–mass spectrometry (LC-MS) instruments. This chapter covers approaches for developing and deploying cell-free synthetic biology sensors. First, a brief overview will be given of cell-free protein expression systems and how they are prepared. Next, the sensing, regulation, and output aspects of cell-free biosensors will be covered. In particular, both protein-based and nucleic acid-based biosensing components will be discussed, and relevant examples of cellfree biosensors will be reviewed. Then, the use of gene regulatory networks in sensing applications and options for generating outputs will be discussed. Finally, the practical issues and strategies associated with deploying cell-free biosensors will be explored.

Cell-Free Protein Expression Platforms E. coli Cell-Free Systems Cell-free protein synthesis began with experiments by Nirenberg and Matthaei in which they developed an expression system using E. coli extracts (Nirenberg and Matthaei 1961). Since then, E. coli has been the subject of many efforts to improve the simplicity, cost effectiveness, and yield production of cell-free systems (Chong 2014; Carlson et al. 2012; Gregorio et al. 2019). E. coli systems are now powerful

322

D. K. Karig et al.

platforms for cell-free protein synthesis, abbreviated as CFPS, and are used either as the source of crude extracts in an extract system, or as the source of translational machinery toolbox components in the PURE system, which is later discussed (Shimizu et al. 2001). Extract systems often have a complex, time consuming, and technically challenging process for preparation of the cell extracts from cultured cells for use in CFPS systems. Cell extracts provide the transcription-translation machinery of the cell necessary to synthesize proteins in vitro (Fujiwara and Doi 2016). Therefore, preparation efforts focus on removing unnecessary and inhibitory parts of the cell such as the cell wall, cell membrane, and the functional genome of the cell (Gregorio et al. 2019). At a high level, extract preparation consists of four essential phases: culturing and harvesting the desired cells, obtaining a pellet and washing cells, lysing the cells, and clarifying the lysate (Dopp and Reuel 2018). Additional postprocessing steps such as lyophilization may also be included, as later discussed in the Deployment section (Dopp and Reuel 2018). Before an extract can be prepared, the desired strain must be selected, cultured, and then harvested. Once the cells have been harvested, the next crucial step to prepare a cell extract is to lyse the cells without damaging ribosomal activity in the lysate (Gregorio et al. 2019). There are many different techniques used to lyse cells, including homogenization, sonication, French press, freeze-thaw cycles, nitrogen cavitation, and bead beating. The most popular techniques for CFPS systems are French press and bead beating (Gregorio et al. 2019; Fujiwara and Doi 2016). Clarification of the lysate is the last step that prepares the cell extracts for use in a CFPS system. This is done through centrifugation, which separates the cell debris created during the previous step from the desired lysate components (Dopp and Reuel 2018). Centrifugation speeds, temperatures, time lengths, and other factors may vary depending on the cell type used and the desired purity and volume of the extract (Dopp and Reuel 2018). This step may also include run-off reactions and other postlysis processing such as dialysis or treatment with nucleases (Gregorio et al. 2019). Once all clarification and postlysis processing has been completed, the extract is ready to be utilized in a CFPS reaction. This involves the optimization and inclusion of additional reagents to fuel and stabilize protein expression, including amino acids, energy regeneration components, and nucleotide triphosphates. Both historically and at present, the E. coli extract system is the most commonly used extract system. This is due to the variety of advantages this system has over other crude extract based systems. One advantage is the low cost and relatively easy culture of E. coli compared to other cell types such as eukaryotic cells that require more expensive media and more complicated culturing methods (Carlson et al. 2012). Another key advantage is the high yield of E. coli systems in comparison to other crude extract systems (Carlson et al. 2012). The higher yield makes E. coli extract systems more cost and time efficient than other systems and also easier to extend to large scale or commercial applications (Zawada et al. 2011). Another important aspect of E. coli extract systems is the well-characterized nature of E. coli as a frequently used and studied model organism in microbiology and synthetic biology (Perez et al. 2016). The genetic information and transcription and translation

14

Cell-Free Synthetic Biology Biosensors

323

machinery of this organism are well studied, and a host of technologies, tools, and techniques are available. Another type of CFPS system that utilizes E. coli is the PURE system (Shimizu et al. 2001; Ohashi et al. 2010). Instead of crude extracts, the PURE system utilizes purified translational components derived from E. coli that are combined to create a functional translation system (Lu 2017). The PURE system is in some cases advantageous due to its minimal composition (Perez et al. 2016), but cost can limit the scale of applications (Carlson et al. 2012). However, recent methods have been developed to lower the cost (Lavickova and Maerkl 2019).

Variety of Available Platforms Beyond E. coli, CFPS systems based on crude extracts can be derived from a variety of different cell types, each associated with different challenges and application benefits. It is important to consider the different aspects of each type of extract system in order to select an extract to best suit application needs. While E. coli is the most popular choice for prokaryotic cell extract systems, there has been growing interest in developing expression platforms from other bacteria, including Pseudomonas putida (Wang et al. 2018), Bacillus subtilis (Kelwick et al. 2016), Bacillus megaterium (Moore et al. 2018), Vibrio natriegens (Failmezger et al. 2018; Des Soye et al. 2018; Wiegand et al. 2018), Streptomyces spp. (Li et al. 2017; Moore et al. 2017), Salmonella enterica, Pantoea agglomerans, and Klebsiella oxytoca (Yim et al. 2019). These bacteria each have different growth characteristics and inherent sensing and metabolic capabilities. While prokaryotic systems are advantageous for their relative simplicity, yield, and cost, many eukaryotic crude extract systems offer advantages for certain applications (Zemella et al. 2015). E. coli extract systems are naturally limited in their ability to produce certain post-translational modifications (Carlson et al. 2012). One recent direction is to augment E. coli with machinery from other organisms to enable the desired post-translational modifications (Jaroentomeechai et al. 2018). Another option is to utilize eukaryotic systems, which are naturally able to produce proteins with post-translational modifications. The three most commonly used eukaryotic extracts for CFPS systems are wheat-germ extracts, rabbit reticulocytes, and insect cell extracts (Carlson et al. 2012), although other systems based on Chinese Hamster Ovary (CHO) cells (Brödel et al. 2014) and even human cells (Machida et al. 2014) have been developed as well. However, eukaryotic extracts are typically more technically challenging, due to culture requirements as well as more complex procedures required for extract preparation once the desired cell line has been cultured. While simplified procedures have been developed for E. coli systems (Cai et al. 2015; Kim et al. 2006; Shrestha et al. 2012), additional processing steps are sometimes needed for eukaryotic systems. For example, with wheat germ systems, extensive washing steps are needed to remove inhibitors from the endosperm such as tritin, thionin, RNAse, DNAse, and proteases (Harbers 2014; Madin et al. 2000). In other systems, such as CHO, extract preparation procedures have included steps

324

D. K. Karig et al.

such as size exclusion chromatography (Brödel et al. 2014). Also, due to the longer half-life of mRNA in eukaryotes, treatment with nuclease and subsequent nuclease inactivation may be needed to deplete endogenous mRNA for efficient translation from the added or newly transcribed mRNA (Brödel et al. 2014; Pelham and Jackson 1976). Besides the differences in cell culture and extract preparation, eukaryotic systems typically have lower yields than prokaryotic extract systems. E. coli systems have exceeded protein yields of 2 mg/mL in batch mode, i.e., without continuous supply of fresh nutrients and removal of waste products (Caschera and Noireaux 2014). However, continuous exchange systems to supply nutrients and remove waste are usually needed to approach similar yields in eukaryotic systems (Sawasaki et al. 2002; Tran et al. 2018). Within the eukaryotic systems, wheat germ extract has the highest yield; however, it is limited in the post-translation modifications it can perform. Depending on the complexity of the protein necessary, a less efficient and lower yield extract such as rabbit reticulocytes or insect cell extracts could still be a better choice since they are capable of more complex protein synthesis (Carlson et al. 2012). Recently, yeast platforms have been developed using Saccharomyces cerevisiae (Gan and Jewett 2014) and Pichia pastoris (Aw and Polizzi 2019). These systems offer the advantages of wellcharacterized hosts that are easy to genetically modify, in addition to simple and inexpensive culture approaches. Collectively, ongoing efforts to simplify, optimize, and augment E. coli systems, to develop systems using novel microbes, and to improve eukaryotic systems are yielding an increasing array of platform choices for biosensing applications.

Sensing Methods A growing number of efforts have focused on developing sensors using cell-free protein expression systems (Soltani et al. 2018). These efforts can largely be divided into two classes – those which utilize protein-based sensing elements and those which use nucleotide-based sensing elements. Generally, protein-based sensors are robust and offer excellent sensitivity and specificity. If a protein for sensing a target ligand does not already exist, one must typically resort to protein engineering approaches (Lutz and Iamurri 2018) or directed evolution (Arnold 2018), and development can be challenging if no known protein detects analogous ligands. Nucleotide sensing components, on the other hand, can be developed for arbitrary chemicals using methods such as SELEX (Stoltenburg et al. 2007). In addition, as we later discuss, simple base pairing can be leveraged to detect essentially any viral sequence. Production of nucleotide-based sensors typically requires less energy than expression of protein-based components. However, for some nucleotide components, achieving stability and specificity can be challenging (Lakhin et al. 2013). Despite the relative challenges of each approach, they together enable the sensing of a wide array of chemicals, and even viruses.

14

Cell-Free Synthetic Biology Biosensors

325

Protein-Based Sensing Many efforts based on protein sensing elements have involved a single transcription factor, a promoter with a binding site for the transcription factor, and a reporter gene expressed from the promoter. When the ligand binds the transcription factor, binding of the transcription factor to the promoter is altered, and consequently reporter gene expression levels change. Perhaps the earliest CFPS biosensors were the mercury and tetracycline sensors presented by Pellinen et al. (2004). To sense mercury, the MerR transcription factor was expressed from a T7 promoter, and the firefly luciferase reporter was expressed from the mer operon promoter, which is activated by MerR in the presence of mercury. Likewise, to sense tetracycline, the TetR repressor was expressed from a T7 promoter, and firefly luciferase was expressed from the tet operon promoter, which is repressed by TetR in the absence of tetracycline. The results from these biosensors showed the strong benefits of a cell-free approach. The cell-free tetracycline sensor responded to tetracycline concentrations that were roughly tenfold lower than concentrations required to activate a similar whole-cell biosensor. Also, both cell-free sensors responded to levels of mercury and tetracycline that were toxic to whole cells (Pellinen et al. 2004). More recently, a number of different protein biosensors have been developed. Several efforts have focused on extending engineered quorum sensing to cell-free systems (Kawaguchi et al. 2008; Karig et al. 2011; Chappell et al. 2013; SchwarzSchilling et al. 2016; Halleran and Murray 2017). In whole cell synthetic biology, quorum sensing components from gram-negative bacteria have been widely used to coordinate population level behavior and engineer spatial regulation in living cells (Prindle et al. 2011; Danino et al. 2010; Basu et al. 2005; Brenner et al. 2007; Karig et al. 2018). Since many quorum sensing molecules indicate the presence of virulent bacteria, the ability to efficiently and specifically sense them has practical applications, in addition to future utility in constructing complex genetic regulatory circuits in cell-free systems. Gene circuits for detection of quorum sensing molecules express a receptor protein. The receptor can form a complex with cognate quorum sensing molecules, and this complex can activate a promoter expressing a reporter protein. Recently, Wen et al. developed cell-free biosensors for detecting P. aeruginosa in infected respiratory samples (Wen et al. 2017). Biosensor results were compared with LC-MS/MS (one false positive and one false negative out of 20 samples). Concentrations estimated based on the biosensor readout were generally within 3 nM of the LC-MS/MS determinations and at much lower cost (Wen et al. 2017). While many sensors consist of a transcription factor, a promoter, and a reporter gene, others consist of a single, multidomain protein that integrates sensing and reporting functions (Ribeiro et al. 2019). One class of multidomain sensor proteins commonly used in whole cells is based on periplasmic-binding proteins (PBPs). PBPs such as the maltose-binding protein consist of two domains joined by a ligandbinding hinge. Ligand binding causes the two domains to come into proximity with one another. One design strategy is to fuse one fluorescent protein to one domain and

326

D. K. Karig et al.

another fluorescent protein to the other to enable Fluorescence Resonance Energy Transfer (FRET) upon ligand binding (Lindenburg and Merkx 2014). In HeLa cell extracts, Pardee et al. demonstrated a FRET-based biosensor for glucose (Pardee et al. 2014; Takanaga and Frommer 2010). Salehi et al. developed a different variety of multidomain sensors for detecting ligands that bind nuclear hormone receptors (NHRs) in order to sense endocrine disrupting signals (Salehi et al. 2017). This platform utilizes an engineered fusion protein consisting of a maltose binding domain, a mini-intein domain to improve stability and solubility (Skretas and Wood 2005), a NHR ligand binding domain (Skretas et al. 2007), and a ß-lactamase reporter domain. When the ligand binds the NHR domain, the protein changes structure, and reporter enzyme accessibility and activity is increased. In this design, other NHRs, either natural or modified from human or animal sources, could be utilized in this sensor platform to detect different classes of ligands (Salehi et al. 2018). In general, multidomain protein sensors can generate fast responses, even though the design process and achievement of strong dynamic range is sometimes challenging (Erismann-Ebner et al. 2019). For instance, a FRET-based glucose sensor worked in cell extracts, but exhibited less than two-fold difference in output fluorescence intensity when ligand is present vs. absent (Pardee et al. 2014). Nonetheless, the signal-to-noise ratio was promising based on tight standard deviations across measurements. Moreover, response speed is often better than transcription factor approaches, since there is no initial delay incurred by expressing the transcription factor, followed by later expression of the reporter. For example, operation times for the endocrine disruptor sensors were faster than the quorum sensing detectors presented by Wen et al., which required initial accumulation of expressed receptor protein (Salehi et al. 2017; Wen et al. 2017). Regardless of the particular approach, one initial challenge can be finding or developing an appropriate ligand-binding protein. To greatly expand the detection space of protein-based sensors, Voydovic et al. developed metabolic transducers (Voyvodic et al. 2019). If no transcription factor exists to detect a ligand of interest, one may identify an enzyme to convert the ligand into one that can be detected by an existing sensor. They first optimized a benzoic acid sensor and then proceeded to develop sensors for hippuric acid and cocaine using the benzoic acid sensor. Specifically, they used the HipO enzyme from Campylobacter jejuni and CocE esterase from Rhodococcus sp. to, respectively, convert hippuric acid and cocaine into benzoic acid (Voyvodic et al. 2019).

Nucleotide-Based Sensing For sensing targets that have no known protein sensor, many cell-free sensing strategies have employed aptamers, which are short nucleic acid sequences that bind specific molecular targets. Selection procedures are available for identifying aptamers for tremendous variety of novel targets (Blind and Blank 2015; Darmostuk et al. 2015). Aptamer-based biosensors are generally designed such that the aptamer

14

Cell-Free Synthetic Biology Biosensors

327

changes conformation upon ligand binding, resulting in a corresponding alteration in transcriptional efficiency, translational efficiency, or enzymatic activity, depending upon the precise implementation. For example, Iyer and Doktycz demonstrated a DNA aptamer approach for engineering ligand-responsive promoters in cell-free systems (Iyer and Doktycz 2013). They inserted a DNA aptamer sequence near a T7 promoter such that transcription was altered upon ligand binding. However, most approaches have relied on RNA aptamers and regulation of translation. For instance, Ogawa designed riboswitches that function in eukaryotic cell-free systems and demonstrated sensors for theophylline, FMN, tetracycline, and sulforhodamine B. These riboswitches downregulate IRES-mediated translation in the presence of sufficient ligand concentrations (Ogawa 2011; Ogawa et al. 2017). Nucleic acid sensor components are particularly appropriate when the target itself is a specific nucleotide sequence, for instance, a sequence that indicates the presence of a pathogen (Green et al. 2014). Toehold switches offer a flexible approach for sensing specific RNA sequences (Green et al. 2014). In toehold switches, the ribosome binding site (RBS) for an output reporter gene is sequestered in the loop of a hairpin in the mRNA. The mRNA is also designed to contain a toehold region, which is complementary to the targeted RNA. When the targeted RNA sequence binds this toehold region, the hairpin opens, freeing the RBS to allow reporter translation. Pardee et al. utilized this method to detect Ebola (Pardee et al. 2014) and Zika (Pardee et al. 2016) RNA in E. coli extracts. Ma et al. similarly utilized toehold switches for the detection of norovirus (Ma et al. 2018). Toehold switches offer more sensitivity and relaxed genotype specificity (Ambert-Balay and Pothier 2013; Vyas et al. 2015) in comparison to lateral flow assays (Vinje 2015), which use antibodies that bind viral surface proteins. Toehold switches offer advantages over qPCR as well, namely, in the lack of a need for thermal cycling equipment. Cost profiles for cell-free toehold sensors are appealing as well (Ma et al. 2018; Pardee et al. 2014, 2016).

Regulation with Synthetic Gene Networks The above sensing methods offer a wide array of sensing and response capabilities, yet the deeper potential of synthetic biology lies in coupling sensing elements to gene circuits for signal processing and regulation. Gene regulatory networks may be used to improve responses to single targets and to integrate multiple targets. One opportunity for gene circuits is to amplify weak responses. Many promoters generate weak responses to signals, and amplifiers can be tuned to improve dynamic range, raise signals well above background levels, and even invert the output (Karig and Weiss 2005). More complex regulatory circuits can also be used to integrate multiple sensors and/or multiple signals. For example, specificity may be achieved through digital logic – a logical AND gate of multiple sensors with imperfect specificity may generate a response with high overall specificity. For instance, Wen et al. suggested the possibility of coupling sensors for multiple P. aeruginosa biomarkers through digital logic to improve detection of the pathogen (Wen et al. 2017). Synthetic gene

328

D. K. Karig et al.

networks may even integrate multiple sensing modalities (e.g., transcription factors, receptors, aptamers) and generate a single sensor read-out. Several cell-free synthetic biology developments lay the groundwork for integrating genetic circuits to develop sophisticated sensors. One cell-free “toolbox” contains sets of E. coli promoters and transcriptional activators that can be cascaded and combined (Shin and Noireaux 2012) to generate regulatory functions. A second toolbox includes components for regulating mRNA and protein stability (Garamella et al. 2016). In addition to these toolboxes, various multiinput promoters have been tested in cell-free systems (Isalan et al. 2005; Iyer et al. 2013; Shin and Noireaux 2012) and are particularly helpful for integrating multiple sensor responses and implementing digital logic. Using these and other similar components, several interesting regulatory behaviors have been demonstrated in cell-free systems: transcriptional cascades (Noireaux et al. 2003; Shin and Noireaux 2012), inducible feedback circuits (Karig et al. 2012; Shin and Noireaux 2012), digital logic (Iyer et al. 2013), pulse generators (Garamella et al. 2016), oscillators (Niederholtmeyer et al. 2015), and a pattern formation system (Isalan et al. 2005). As regulation components are incorporated into biosensors, one consideration is that all components may draw from the same pools of resources, such as nucleotides and amino acids. For example, Voyvodic et al. encountered this need to balance resource utilization between different components (Voyvodic et al. 2019). Fortunately, cell-free systems facilitate system optimization, as one can simply add different amounts of DNA encoding each system component to reactions. By contrast, with living cells, one would need to perform genetic modifications to alter plasmid copy numbers, promoter strength, and/or ribosome binding sites.

Output Generally, cell-free biosensor output consists of reporter gene expression, and the reporters used in cell-free systems to date are largely standard reporters derived from whole cell efforts, namely, fluorescent proteins, enzymatic reporters such as LacZ βgalactosidase, and colorimetric reporters such as chromoproteins. Nonetheless, appropriate choice of a reporter can have a significant impact on performance. As one example, Ma et al. switched from expressing the full length lacZ gene to αcomplementation, whereby the cell extract contains the omega portion of lacZ, and only the α portion is expressed in response to ligand. The use of α-complementation reduced detection time by 23 minutes (40% reduction) (Ma et al. 2018). Also, Voyvodic et al. demonstrated strong GFP responses for benzoic acid and hippuric acid sensors. However, detecting clinical concentrations of cocaine required switching to a luciferase reporter, since increases in autofluorescence in the GFP channel over time interfered with detection (Voyvodic et al. 2019). Indeed, few efforts to date have expanded output options by capitalizing on the flexible, open nature of cell-free systems, which facilitates the inclusion of custom molecules and enables gene circuit components and products to directly interact with surfaces. However, one significant step in this direction is the recent demonstration

14

Cell-Free Synthetic Biology Biosensors

329

of an electrochemical interface for cell-free biosensors (Sadat Mousavi et al. 2020). In this approach, a biosensor gene circuit expresses a restriction enzyme as an intermediate output. The expressed restriction enzyme then cleaves DNA reporter molecules in solution. This frees a cleaved portion of reporter DNA, labeled with the redox reporter methylene blue, to interact with complementary capture DNA. The capture DNA is bound to an electrode, and the proximity of methylene blue causes a measurable current change at the electrode. Multiple biosensors can be multiplexed through the use of multiple, orthogonally acting restriction enzymes, each paired with a corresponding reporter DNA, capture DNA, and electrode (Sadat Mousavi et al. 2020).

Deployment Sample Processing For a given application, one must assess the appropriate need for sample dilution, amplification, or purification, and how much to add to a cell-free reaction. Although cell-free systems can often remain responsive in the presence of conditions that are toxic to whole cells, as exemplified by the mercury and tetracycline sensors discussed earlier (Pellinen et al. 2004), some samples can still cause problematic inhibition of cell-free protein expression. Samples may contain nucleases (RNAse or DNAse), microbial contaminants, acids, bases, or salts that inhibit protein expression. In some cases, the ligand itself is inhibitory. This issue was encountered by Gräwe et al. in the development of a sensor for the date-rape drug gammahydroxybutyrate (GHB) (Grawe et al. 2019). A few studies generally offer encouragement for the prospects of cell-free biosensors for environmental, commercial, and clinical samples. Salehi et al. demonstrated cell-free protein production in a variety of environmental samples, including water from pond, snow, soil, storm, tap, and wastewater sources (Salehi et al. 2017). In addition, they tested expression in clinical samples, namely blood and urine (Salehi et al. 2017; Salehi et al. 2018). Also, Voyvodic et al. detected benzoic acid in beverage samples. Although detection was possible when beverage samples constituted 10% of the reaction volume, up to 75% inhibition was noted, prompting a 1:10 dilution of the sample (Voyvodic et al. 2019). When detecting hippuric acid in urine samples, they found minimal inhibition when a 1% final concentration of urine was used. However, in order to detect cocaine in urine samples at clinically relevant concentrations, sample dilution was not an option (Voyvodic et al. 2019). Again, for any sensing application, inhibitory effects of the target must be assessed, and optimal relative volumes of sample to include in cell-free reactions must be determined. Cell-free system compositions may ideally be optimized for targeted sample types. A few basic steps in this direction have been taken to date. Most formulations for handling environmental or clinical samples have included commercially available RNAse inhibitors (Pardee et al. 2014, 2016; Grawe et al. 2019; Voyvodic et al. 2019). For instance, without RNAse inhibitor, Salehi et al. found urine to be

330

D. K. Karig et al.

particularly inhibitory (Salehi et al. 2017), but inclusion of murine RNAse inhibitor significantly improved expression (Salehi et al. 2018). To lower costs, Gräwe et al. purified the endogenous E. coli RNase E inhibitor RraA and included it in extracts (Grawe et al. 2019). Also, for their GHB biosensor, Gräwe et al. found that inhibition by the targets was partly alleviated when the extracts were prepared from cells that pre-expressed the BlcR receptor (Grawe et al. 2019). In cases where the cell-free system cannot be easily reformulated to reduce inhibitory effects, one option is to implement ratiometric sensing, whereby a second reporter is constitutively expressed to quantify the expression capacity at the given ligand concentration. The sensor output is then defined as the ratio of the sensor reporter intensity to the constitutive reporter intensity. However, one must keep in mind that additional optics and processing are required. Besides mitigation of inhibitory effects, some applications require concentration, purification, or amplification of the target. For example, for sensing pathogen quorum sensing molecules in sputum, Wen et al. relied on solvent extraction (Wen et al. 2017). Also, to improve sensitivity of virus detection, Pardee et al. incorporated nucleic acid sequence-based amplification (NASBA), an isothermal amplification method, into their Zika sensing platform (Pardee et al. 2016). Interestingly, they went a step further and incorporated CRISPR/Cas9 cleavage into their NASBA amplification procedure in order to deplete non-targeted strains with single base pair resolution. This, in turn, enabled specific detection of a chosen strain (Pardee et al. 2016). Ma et al. developed different preprocessing procedures to enhance sensitivity by first concentrating the virus and then amplifying the signal. To concentrate the virus, they used streptavidin-coated magnetic beads in conjunction with biotinlabeled synthetic antibodies (synbodies) that bind norovirus (Ma et al. 2018). In general, the need for sample preprocessing depends on the complexity and content of the source material, as well as the degree to which sample concentrations match the detection range of the sensor. In the future, microfluidics offer a potentially appealing path to facilitating sample pre-processing. In particular, paper microfluidics offer an inexpensive, easily deployable approach for combining, filtering, and guiding flow of reagents and samples (Akyazi et al. 2018). As we shortly discuss, several studies have demonstrated cell-free sensing in porous substrates such as paper (Pardee et al. 2014, 2016; Grawe et al. 2019), constituting a step in this direction.

Stability For many years, the need for storing cell-free protein expression reagents at cold temperatures, e.g., 80  C, limited practical applications of cell-free protein expression. Recently, a few efforts have made strides in improving the shelf-life and delivery of cell-free protein expression systems through the drying and preserving of components (Hunt et al. 2017; Pardee 2018). Simple freeze-drying of reagents can offer stability at room temperature. Although measures must be taken to minimize

14

Cell-Free Synthetic Biology Biosensors

331

humidity, for instance, through the use of desiccants, this approach is potentially feasible for storage and distribution of small biosensors that could be distributed in sealed pouches and activated on demand. For example, Pardee et al. lyophilized cellfree reagents and used them to implement biosensor gene networks capable of detecting Ebola (Pardee et al. 2014) and Zika virus (Pardee et al. 2016). Salehi et al. also demonstrated function of the endocrine disruptor biosensors after lyophilization and reconstitution of cell-free expression reagents (Salehi et al. 2017). Beyond simple freeze-drying, other efforts have focused on the inclusion of stabilizing reagents to improve preservation (Kuroita et al. 2006; Peters and Bendzko 2002; Smith et al. 2014a). Recently, these efforts have been extended to realize longterm resilience, even above room temperature (Karig et al. 2017; Wilding et al. 2019) and without other specialized storage requirements such as desiccants or inert gas. These efforts offer promise to applications that demand additional degrees of robustness.

Encapsulation Although stabilization through drying and preservation addresses critical needs for many applications, certain applications require additional protection of cell-free reagents, for instance, through encapsulation. Encapsulation provides two vital functions for a cell-free system: protecting from contaminants and keeping the components of the cell-free system together. These functions are key requirements for large-scale or environmental sensing to ensure that transcription and translation are enabled. However, protection and consolidation must be achieved while still allowing the biosensor components to access the target ligands. Cell-free protein expression in basic emulsions and liposomes has been widely explored and has proved useful for characterization, assembly, directed evolution, and protein purification (Yue et al. 2019; Fallah-Araghi et al. 2012; Hori et al. 2017; Kuruma and Ueda 2015). Liposome-encapsulated cell-free systems were recently used in the synthesis of bacterial quorum signals and the response to these signals, which demonstrated that encapsulated cell-free reagents can be engineered to interact with the environment (Schwarz-Schilling et al. 2016). However, one of the main barriers of encapsulated systems is how to maintain stability within the system. Noireaux and Libchaber extended the stability of phospholipid vesicles by expressing α-hemolysin to optimize permeability and osmotic pressure (Noireaux and Libchaber 2004). However, many sensing applications may demand additional degrees of ruggedness. Another possible encapsulation strategy is to utilize polymer substrates. For example, alginate beads containing cell-free protein expression reagents were coated with silica to successfully enhance the environmental resilience of the system (Lim et al. 2009). DNA microgels have similarly been developed for cell-free expression (Kahn et al. 2016). Future efforts in this direction and in the field of minimal cells may further help to expand ruggedization options for cell-free biosensing (Schreiber et al. 2019; Vogele et al. 2018).

332

D. K. Karig et al.

Conclusions and Future Directions Cell-free protein expression systems can couple the relative robustness and flexibility of in vitro sensors to an array of synthetic biology capabilities, enabling the development of integrated sensing, response, and output functionalities. Indeed, recent years have seen significant developments in all essential areas needed to engineer such sophisticated sensors. This includes the development of more efficient and scalable CFPS systems (Carlson et al. 2012; Zawada et al. 2011), increasingly sophisticated cell-free gene circuits (Niederholtmeyer et al. 2015; To et al. 2018), an electrochemical interface for processing biosensor output (Sadat Mousavi et al. 2020), and methods for improving shelf-life (Karig et al. 2017; Wilding et al. 2019) and format (Pardee et al. 2014; Timm et al. 2015, 2016). Despite rapid recent advances in areas essential for cell-free synthetic biology sensors, a number of opportunities could potentially lead to transformative impacts. As the field of cell-free protein expression continues to grow and progress beyond the benchtop, more application specific developments, including procedures to efficiently optimize CFPS formulations for certain applications, are envisioned. The majority of preparation procedures have focused on improving yield, cost, or scale. While these are critical factors, in the realm of biosensing applications, robustness and repeatability may be more important than yield, for instance. Also, the growing interest in developing cell-free expression systems from different organisms will help the field of biosensing to capitalize on the diverse array of natural sensing and metabolic capabilities (Kelwick et al. 2016). In addition to new cell-free expression systems, another area ripe for development is biosensor output. To date, most efforts have relied on the same standard reporter types used in whole cells. The recent development of an electrochemical interface offers one path towards sophisticated transducers for enabling outputs that work seamlessly with other sensors and equipment (Efrat et al. 2018; Sadat Mousavi et al. 2020). Finally, future advances in minimal cell efforts may eventually lead to more rugged and flexible deployment formats. Collectively, these new directions, coupled with existing, recent developments point to a bright future for transformative progress in cell-free biosensing.

References Akyazi T, Basabe-Desmonts L, Benito-Lopez F (2018) Review on microfluidic paper-based analytical devices towards commercialisation. Anal Chim Acta 1001:1–17. https://doi.org/ 10.1016/j.aca.2017.11.010 Ambert-Balay K, Pothier P (2013) Evaluation of 4 immunochromatographic tests for rapid detection of norovirus in faecal samples. J Clin Virol 56(3):194–198. https://doi.org/10.1016/j. jcv.2012.11.001 Arnold FH (2018) Directed evolution: bringing new chemistry to life. Angew Chem Int Ed 57 (16):4143–4148 Aw R, Polizzi KM (2019) Biosensor-assisted engineering of a high-yield Pichia pastoris cell-free protein synthesis platform. Biotechnol Bioeng 116(3):656–666

14

Cell-Free Synthetic Biology Biosensors

333

Basu S, Gerchman Y, Collins CH, Arnold FH, Weiss R (2005) A synthetic multicellular system for programmed pattern formation. Nature 434(7037):1130–1134. https://doi.org/10.1038/ nature03461 Blind M, Blank M (2015) Aptamer selection technology and recent advances. Mol Ther –Nucleic Acids 4(1):e223 Brenner K, Karig DK, Weiss R, Arnold FH (2007) Engineered bidirectional communication mediates a consensus in a microbial biofilm consortium. Proc Natl Acad Sci U S A 104 (44):17300–17304. https://doi.org/10.1073/pnas.0704256104 Brödel AK, Sonnabend A, Kubick S (2014) Cell-free protein expression based on extracts from CHO cells. Biotechnol Bioeng 111(1):25–36 Cai Q, Hanson JA, Steiner AR, Tran C, Masikat MR, Chen R, Zawada JF, Sato AK, Hallam TJ, Yin G (2015) A simplified and robust protocol for immunoglobulin expression in Escherichia coli cell-free protein synthesis systems. Biotechnol Prog 31(3):823–831 Carlson ED, Gan R, Hodgman CE, Jewett MC (2012) Cell-free protein synthesis: applications come of age. Biotechnol Adv 30(5):1185–1194 Caschera F, Noireaux V (2014) Synthesis of 2.3 mg/ml of protein with an all Escherichia coli cellfree transcription–translation system. Biochimie 99:162–168 Chappell J, Jensen K, Freemont PS (2013) Validation of an entirely in vitro approach for rapid prototyping of DNA regulatory elements for synthetic biology. Nucleic Acids Res 41:3471 Chong S (2014) Overview of cell-free protein synthesis: historic landmarks, commercial systems, and expanding applications. Curr Protoc Mol Biol 108:16 30 11-11. https://doi.org/10.1002/ 0471142727.mb1630s108 Danino T, Mondragon-Palomino O, Tsimring L, Hasty J (2010) A synchronized quorum of genetic clocks. Nature 463(7279):326–330. https://doi.org/10.1038/nature08753 Darmostuk M, Rimpelova S, Gbelcova H, Ruml T (2015) Current approaches in SELEX: an update to aptamer selection technology. Biotechnol Adv 33(6):1141–1161 Des Soye BJ, Davidson SR, Weinstock MT, Gibson DG, Jewett MC (2018) Establishing a highyielding cell-free protein synthesis platform derived from Vibrio natriegens. ACS Synth Biol 7 (9):2245–2255 Dopp JL, Reuel NF (2018) Process optimization for scalable E. coli extract preparation for cell-free protein synthesis. Biochem Eng J 138:21–28 Efrat Y, Tayar AM, Daube SS, Levy M, Bar-Ziv RH (2018) Electric-field manipulation of a compartmentalized cell-free gene expression reaction. ACS Synth Biol 7(8):1829–1833 Erismann-Ebner K, Marowsky A, Arand M (2019) In-vitro characterization of mCerulean3_mRuby3 as a novel FRET pair with favorable bleed-through characteristics. Biosensors 9(1):33 Failmezger J, Scholz S, Blombach B, Siemann-Herzberg M (2018) Cell-free protein synthesis from fast-growing Vibrio natriegens. Front Microbiol 9:1146 Fallah-Araghi A, Baret J-C, Ryckelynck M, Griffiths AD (2012) A completely in vitro ultrahighthroughput droplet-based microfluidic screening system for protein engineering and directed evolution. Lab Chip 12(5):882–891 Fujiwara K, Doi N (2016) Biochemical preparation of cell extract for cell-free protein synthesis without physical disruption. PLoS One 11(4):e0154614. https://doi.org/10.1371/journal. pone.0154614 Gan R, Jewett MC (2014) A combined cell-free transcription-translation system from Saccharomyces cerevisiae for rapid and robust protein synthe. Biotechnol J 9(5):641–651 Garamella J, Marshall R, Rustad M, Noireaux V (2016) The all E. coli TX-TL toolbox 2.0: a platform for cell-free synthetic biology. ACS Synth Biol 5(4):344–355. https://doi.org/10.1021/ acssynbio.5b00296 Grawe A, Dreyer A, Vornholt T, Barteczko U, Buchholz L, Drews G, Ho UL, Jackowski ME, Kracht M, Luders J, Bleckwehl T, Rositzka L, Ruwe M, Wittchen M, Lutter P, Muller K, Kalinowski J (2019) A paper-based, cell-free biosensor system for the detection of heavy metals and date rape drugs. PLoS One 14(3):e0210940. https://doi.org/10.1371/journal.pone.0210940 Green AA, Silver PA, Collins JJ, Yin P (2014) Toehold switches: de-novo-designed regulators of gene expression. Cell 159(4):925–939

334

D. K. Karig et al.

Gregorio NE, Levine MZ, Oza JP (2019) A user’s guide to cell-free protein synthesis. Methods Protoc 2(1). https://doi.org/10.3390/mps2010024 Halleran AD, Murray RM (2017) Cell-free and in vivo characterization of Lux, Las, and Rpa quorum activation systems in E. coli. ACS Synth Biol 7(2):752–755 Harbers M (2014) Wheat germ systems for cell-free protein expression. FEBS Lett 588 (17):2762–2773 Hori Y, Kantak C, Murray RM, Abate AR (2017) Cell-free extract based optimization of biomolecular circuits with droplet microfluidics. Lab Chip 17(18):3037–3042 Hunt JP, Yang SO, Wilding KM, Bundy BC (2017) The growing impact of lyophilized cell-free protein expression systems. Bioengineered 8(4):325–330 Isalan M, Lemerle C, Serrano L (2005) Engineering gene networks to emulate Drosophila embryonic pattern formation. PLoS Biol 3(3):e64 Iyer S, Doktycz MJ (2013) Thrombin-mediated transcriptional regulation using DNA aptamers in DNA-based cell-free protein synthesis. ACS Synth Biol 3(6):340–346 Iyer S, Karig DK, Norred SE, Simpson ML, Doktycz MJ (2013) Multi-input regulation and logic with T7 promoters in cells and cell-free systems. PLoS One 8(10):e78442 Jaroentomeechai T, Stark JC, Natarajan A, Glasscock CJ, Yates LE, Hsu KJ, Mrksich M, Jewett MC, DeLisa MP (2018) Single-pot glycoprotein biosynthesis using a cell-free transcriptiontranslation system enriched with glycosylation machinery. Nat Commun 9(1):2686 Kahn JS, Ruiz RC, Sureka S, Peng S, Derrien TL, An D, Luo D (2016) DNA microgels as a platform for cell-free protein expression and display. Biomacromolecules 17:2019 Karig DK (2017) Cell-free synthetic biology for environmental sensing and remediation. Curr Opin Biotechnol 45:69–75. https://doi.org/10.1016/j.copbio.2017.01.010 Karig D, Weiss R (2005) Signal-amplifying genetic circuit enables in vivo observation of weak promoter activation in the Rhl quorum sensing system. Biotechnol Bioeng 89(6):709–718. https://doi.org/10.1002/bit.20371 Karig DK, Siuti P, Dar RD, Retterer ST, Doktycz MJ, Simpson ML (2011) Model for biological communication in a nanofabricated cell-mimic driven by stochastic resonance. Nano Commun Networks 2(1):39–49 Karig DK, Iyer S, Simpson ML, Doktycz MJ (2012) Expression optimization and synthetic gene networks in cell-free systems. Nucleic Acids Res 40(8):3763–3774 Karig DK, Bessling S, Thielen P, Zhang S, Wolfe J (2017) Preservation of protein expression systems at elevated temperatures for portable therapeutic production. J R Soc Interface 14(129). https://doi.org/10.1098/rsif.2016.1039 Karig D, Martini KM, Lu T, DeLateur NA, Goldenfeld N, Weiss R (2018) Stochastic turing patterns in a synthetic bacterial population. Proc Natl Acad Sci U S A 115(26):6572–6577. https://doi. org/10.1073/pnas.1720770115 Kawaguchi T, Chen YP, Norman RS, Decho AW (2008) Rapid screening of quorum-sensing signal N-acyl homoserine lactones by an in vitro cell-free assay. Appl Environ Microbiol 74 (12):3667–3671 Kelwick R, Webb AJ, MacDonald JT, Freemont PS (2016) Development of a Bacillus subtilis cellfree transcription-translation system for prototyping regulatory elements. Metab Eng 38:370–381 Khambhati K, Bhattacharjee G, Gohil N, Braddick D, Kulkarni V, Singh V (2019) Exploring the potential of cell-free protein synthesis for extending the abilities of biological systems. Front Bioeng Biotechnol 7:248 Kim T-W, Keum J-W, Oh I-S, Choi C-Y, Park C-G, Kim D-M (2006) Simple procedures for the construction of a robust and cost-effective cell-free protein synthesis system. J Biotechnol 126 (4):554–561 Kuroita T, Kawakami B, Kawamura Y, Nishikawa S, Endo Y (2006) Composition for cell-free protein synthesis. Google Patents Kuruma Y, Ueda T (2015) The PURE system for the cell-free synthesis of membrane proteins. Nat Protoc 10(9):1328–1344

14

Cell-Free Synthetic Biology Biosensors

335

Lakhin A, Tarantul V, Gening L (2013) Aptamers: problems, solutions and prospects. Acta Naturae (англоязычная версия) 5(4 (19)):34 Lavickova B, Maerkl SJ (2019) A simple, robust, and low-cost method to produce the PURE cellfree system. ACS Synth Biol 8(2):455–462 Li J, Wang H, Kwon YC, Jewett MC (2017) Establishing a high yielding streptomyces-based cellfree protein synthesis system. Biotechnol Bioeng 114(6):1343–1353 Lim SY, Kim K-O, Kim D-M, Park CB (2009) Silica-coated alginate beads for in vitro protein synthesis via transcription/translation machinery encapsulation. J Biotechnol 143(3):183–189 Lindenburg L, Merkx M (2014) Engineering genetically encoded FRET sensors. Sensors 14 (7):11691–11713 Lu Y (2017) Cell-free synthetic biology: engineering in an open world. Synth Syst Biotechnol 2 (1):23–27 Lutz S, Iamurri SM (2018) Protein engineering: past, present, and future. In: Protein engineering. Springer, New York, pp 1–12 Ma D, Shen L, Wu K, Diehnelt CW, Green AA (2018) Low-cost detection of norovirus using paperbased cell-free systems and synbody-based viral enrichment. Synth Biol 3(1):ysy018 Machida K, Mikami S, Masutani M, Mishima K, Kobayashi T, Imataka H (2014) A translation system reconstituted with human factors proves that processing of encephalomyocarditis virus proteins 2A and 2B occurs in the elongation phase of translation without eukaryotic release factors. J Biol Chem 289(46):31960–31971 Madin K, Sawasaki T, Ogasawara T, Endo Y (2000) A highly efficient and robust cell-free protein synthesis system prepared from wheat embryos: plants apparently contain a suicide system directed at ribosomes. Proc Natl Acad Sci 97(2):559–564 Moore SJ, Lai HE, Needham H, Polizzi KM, Freemont PS (2017) Streptomyces venezuelae TXTL–a next generation cell-free synthetic biology tool. Biotechnol J 12(4):1600678 Moore SJ, MacDonald JT, Wienecke S, Ishwarbhai A, Tsipa A, Aw R, Kylilis N, Bell DJ, McClymont DW, Jensen K, Polizzi KM, Biedendieck R, Freemont PS (2018) Rapid acquisition and model-based analysis of cell-free transcription–translation reactions from nonmodel bacteria. Proc Natl Acad Sci 115(19):E4340–E4349. https://doi.org/10.1073/ pnas.1715806115 Niederholtmeyer H, Sun ZZ, Hori Y, Yeung E, Verpoorte A, Murray RM, Maerkl SJ (2015) Rapid cell-free forward engineering of novel genetic ring oscillators. elife 4:e09771 Nirenberg MW, Matthaei JH (1961) The dependence of cell-free protein synthesis in E. coli upon naturally occurring or synthetic polyribonucleotides. Proc Natl Acad Sci U S A 47:1588–1602. https://doi.org/10.1073/pnas.47.10.1588 Noireaux V, Libchaber A (2004) A vesicle bioreactor as a step toward an artificial cell assembly. Proc Natl Acad Sci 101(51):17669–17674 Noireaux V, Bar-Ziv R, Libchaber A (2003) Principles of cell-free genetic circuit assembly. Proc Natl Acad Sci 100(22):12672–12677 Ogawa A (2011) Rational design of artificial riboswitches based on ligand-dependent modulation of internal ribosome entry in wheat germ extract and their applications as label-free biosensors. RNA 17(3):478–488 Ogawa A, Masuoka H, Ota T (2017) Artificial OFF-riboswitches that downregulate internal ribosome entry without hybridization switches in a eukaryotic cell-free translation system. ACS Synth Biol 6(9):1656–1662 Ohashi H, Kanamori T, Shimizu Y, Ueda T (2010) A highly controllable reconstituted cell-free system-a breakthrough in protein synthesis research. Curr Pharm Biotechnol 11(3):267–271 Pardee K (2018) Perspective: solidifying the impact of cell-free synthetic biology through lyophilization. Biochem Eng J 138:91–97 Pardee K, Green AA, Ferrante T, Cameron DE, DaleyKeyser A, Yin P, Collins JJ (2014) Paperbased synthetic gene networks. Cell 159(4):940–954 Pardee K, Green Alexander A, Takahashi Melissa K, Braff D, Lambert G, Lee Jeong W, Ferrante T, Ma D, Donghia N, Fan M, Daringer Nichole M, Bosch I, Dudley Dawn M, O’Connor David H,

336

D. K. Karig et al.

Gehrke L, Collins James J (2016) Rapid, low-cost detection of Zika virus using programmable biomolecular components. Cell. https://doi.org/10.1016/j.cell.2016.04.059 Pelham HR, Jackson RJ (1976) An efficient mRNA-dependent translation system from reticulocyte lysates. Eur J Biochem 67(1):247–256 Pellinen T, Huovinen T, Karp M (2004) A cell-free biosensor for the detection of transcriptional inducers using firefly luciferase as a reporter. Anal Biochem 330(1):52–57 Perez JG, Stark JC, Jewett MC (2016) Cell-free synthetic biology: engineering beyond the cell. Cold Spring Harb Perspect Biol 8:a023853 Peters LE, Bendzko P (2002) Method for producing complex multienzymatical, storage resistant reaction mixtures and use thereof. Google Patents Prindle A, Samayoa P, Razinkov I, Danino T, Tsimring LS, Hasty J (2011) A sensing array of radically coupled genetic ‘biopixels’. Nature 481(7379):39–44. https://doi.org/10.1038/ nature10722 Ribeiro LF, Amarelle V, Ribeiro LFC, Guazzaroni M, #xed, a-Eugenia (2019) Converting a periplasmic binding protein into a synthetic biosensing switch through domain insertion. Biomed Res Int 2019:15. https://doi.org/10.1155/2019/4798793 Sadat Mousavi P, Smith SJ, Chen JB, Karlikow M, Tinafar A, Robinson C, Liu W, Ma D, Green AA, Kelley SO, Pardee K (2020) A multiplexed, electrochemical interface for gene-circuitbased sensors. Nat Chem 12(1):48–55. https://doi.org/10.1038/s41557-019-0366-y Salehi AS, Shakalli Tang MJ, Smith MT, Hunt JM, Law RA, Wood DW, Bundy BC (2017) Cellfree protein synthesis approach to biosensing hTRbeta-specific endocrine disruptors. Anal Chem 89(6):3395–3401. https://doi.org/10.1021/acs.analchem.6b04034 Salehi ASM, Yang SO, Earl CC, Shakalli Tang MJ, Porter Hunt J, Smith MT, Wood DW, Bundy BC (2018) Biosensing estrogenic endocrine disruptors in human blood and urine: a RAPID cell-free protein synthesis approach. Toxicol Appl Pharmacol 345:19–25. https://doi.org/10.1016/j. taap.2018.02.016 Sawasaki T, Ogasawara T, Morishita R, Endo Y (2002) A cell-free protein synthesis system for high-throughput proteomics. Proc Natl Acad Sci 99(23):14652–14657. https://doi.org/10.1073/ pnas.232580399 Schreiber A, Stühn LG, Huber MC, Geissinger SE, Rao A, Schiller SM (2019) Self-assembly toolbox of tailored supramolecular architectures based on an amphiphilic protein library. Small 15:1900163 Schwarz-Schilling M, Aufinger L, Mückl A, Simmel F (2016) Chemical communication between bacteria and cell-free gene expression systems within linear chains of emulsion droplets. Integr Biol 8(4):564–570 Shimizu Y, Inoue A, Tomari Y, Suzuki T, Yokogawa T, Nishikawa K, Ueda T (2001) Cell-free translation reconstituted with purified components. Nat Biotechnol 19(8):751 Shin J, Noireaux V (2012) An E. coli cell-free expression toolbox: application to synthetic gene circuits and artificial cells. ACS Synth Biol 1(1):29–41 Shrestha P, Holland TM, Bundy BC (2012) Streamlined extract preparation for Escherichia colibased cell-free protein synthesis by sonication or bead vortex mixing. BioTechniques 53 (3):163–174 Silverman AD, Karim AS, Jewett MC (2019) Cell-free gene expression: an expanded repertoire of applications. Nat Rev Genet. https://doi.org/10.1038/s41576-019-0186-3 Skretas G, Wood DW (2005) Regulation of protein activity with small-molecule-controlled inteins. Protein Sci 14(2):523–532. https://doi.org/10.1110/ps.04996905 Skretas G, Meligova AK, Villalonga-Barber C, Mitsiou DJ, Alexis MN, Micha-Screttas M, Steele BR, Screttas CG, Wood DW (2007) Engineered chimeric enzymes as tools for drug discovery: generating reliable bacterial screens for the detection, discovery, and assessment of estrogen receptor modulators. J Am Chem Soc 129(27):8443–8457. https://doi.org/10.1021/ja067754j Smith MT, Berkheimer SD, Werner CJ, Bundy BC (2014a) Lyophilized Escherichia coli-based cellfree systems for robust, high-density, long-term storage. BioTechniques 56(4):186–193. https:// doi.org/10.2144/000114158

14

Cell-Free Synthetic Biology Biosensors

337

Smith MT, Wilding KM, Hunt JM, Bennett AM, Bundy BC (2014b) The emerging age of cell-free synthetic biology. FEBS Lett 588(17):2755–2761 Soltani M, Davis BR, Ford H, Nelson JAD, Bundy BC (2018) Reengineering cell-free protein synthesis as a biosensor: biosensing with transcription, translation, and protein-folding. Biochem Eng J 138:165–171. https://doi.org/10.1016/j.bej.2018.06.014 Stoltenburg R, Reinemann C, Strehlitz B (2007) SELEX – a (r)evolutionary method to generate high-affinity nucleic acid ligands. Biomol Eng 24(4):381–403. https://doi.org/10.1016/j. bioeng.2007.06.001 Sun ZZ, Yeung E, Hayes CA, Noireaux V, Murray RM (2014) Linear DNA for rapid prototyping of synthetic biological circuits in an Escherichia coli based TX-TL cell-free system. ACS Synth Biol 3(6):387–397. https://doi.org/10.1021/sb400131a Takanaga H, Frommer WB (2010) Facilitative plasma membrane transporters function during ER transit. FASEB J 24(8):2849–2858. https://doi.org/10.1096/fj.09-146472 Timm AC, Shankles PG, Foster CM, Doktycz MJ, Retterer ST (2015) Characterization of extended channel bioreactors for continuous-flow protein production. J Vac Sci Technol B Nanotechnol Microelectron Mater Process Meas Phenom 33(6):06FM02 Timm AC, Shankles PG, Foster CM, Doktycz MJ, Retterer ST (2016) Toward microfluidic reactors for cell-free protein synthesis at the point-of-care. Small 12(6):810–817 To AC-Y, Chu DH-T, Wang AR, Li FC-Y, Chiu AW-O, Gao DY, Choi CHJ, Kong S-K, Chan T-F, Chan K-M (2018) A comprehensive web tool for toehold switch design. Bioinformatics 34 (16):2862–2864 Tran K, Gurramkonda C, Cooper MA, Pilli M, Taris JE, Selock N, Han TC, Tolosa M, Zuber A, Peñalber-Johnstone C (2018) Cell-free production of a therapeutic protein: expression, purification, and characterization of recombinant streptokinase using a CHO lysate. Biotechnol Bioeng 115(1):92–102 Vinje J (2015) Advances in laboratory methods for detection and typing of norovirus. J Clin Microbiol 53(2):373–381. https://doi.org/10.1128/JCM.01535-14 Vogele K, Frank T, Gasser L, Goetzfried MA, Hackl MW, Sieber SA, Simmel FC, Pirzer T (2018) Towards synthetic cells using peptide-based reaction compartments. Nat Commun 9 (1):3862 Voyvodic PL, Pandi A, Koch M, Conejero I, Valjent E, Courtet P, Renard E, Faulon JL, Bonnet J (2019) Plug-and-play metabolic transducers expand the chemical detection space of cell-free biosensors. Nat Commun 10(1):1697. https://doi.org/10.1038/s41467-019-09722-9 Vyas K, Atkinson C, Clark DA, Irish D (2015) Comparison of five commercially available immunochromatographic tests for the detection of norovirus in faecal specimens. J Hosp Infect 91(2):176–178. https://doi.org/10.1016/j.jhin.2015.06.013 Wang H, Li J, Jewett MC (2018) Development of a Pseudomonas putida cell-free protein synthesis platform for rapid screening of gene regulatory elements. Synth Biol 3(1). https://doi.org/ 10.1093/synbio/ysy003 Wen KY, Cameron L, Chappell J, Jensen K, Bell DJ, Kelwick R, Kopniczky M, Davies JC, Filloux A, Freemont PS (2017) A cell-free biosensor for detecting quorum sensing molecules in P. aeruginosa-infected respiratory samples. ACS Synth Biol 6(12):2293–2301. https://doi.org/ 10.1021/acssynbio.7b00219 Wiegand DJ, Lee HH, Ostrov N, Church GM (2018) Establishing a cell-free Vibrio natriegens expression system. ACS Synth Biol 7(10):2475–2479 Wilding KM, Long Zhao E, Earl CC, Bundy BC (2019) Thermostable lyoprotectant-enhanced cellfree protein synthesis for on-demand endotoxin-free therapeutic production. New Biotechnol. https://doi.org/10.1016/j.nbt.2019.07.004 Yim SS, Johns NI, Park J, Gomes AL, McBee RM, Richardson M, Ronda C, Chen SP, Garenne D, Noireaux V (2019) Multiplex transcriptional characterizations across diverse bacterial species using cell-free systems. Mol Syst Biol 15(8):e8875 Yue K, Zhu Y, Kai L (2019) Cell-free protein synthesis: chassis toward the minimal cell. Cell 8 (4):315

338

D. K. Karig et al.

Zawada JF, Yin G, Steiner AR, Yang J, Naresh A, Roy SM, Gold DS, Heinsohn HG, Murray CJ (2011) Microscale to manufacturing scale-up of cell-free cytokine production – a new approach for shortening protein production development timelines. Biotechnol Bioeng 108(7):1570–1578 Zemella A, Thoring L, Hoffmeister C, Kubick S (2015) Cell-free protein synthesis: pros and cons of prokaryotic and eukaryotic systems. Chembiochem 16(17):2420–2431

Genetic Circuit Design Principles

15

Jing Wui Yeoh, Salvador Gomez-Carretero, Wai Kit David Chee, Ai Ying Teh, and Chueh Loo Poh

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Biosensors and Their Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Biosensing Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Molecule-Based and Cell-Based Biosensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sensing Architectures in Cell-Based Biosensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Advantages and Challenges of Cell-Based Biosensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Synthetic Biology and the Rational Design of Cell-Based Biosensors . . . . . . . . . . . . . . . . . . . . . Principles in Design and Construction of Genetic Circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Choice of Host Chassis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Insights into the Basic Components of Genetic Circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Plasmid Design and Construction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Primer Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DNA Assembly . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Experimental Design Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Model-Driven Approach Toward Rational Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Modelling Framework . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tuning Steady-State Response Curves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Deterministic Kinetic Modelling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Design and Modelling Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A Rational Design Optimization Workflow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

340 341 341 343 343 346 347 348 348 349 352 352 353 356 360 361 362 363 364 368

J. W. Yeoh · S. Gomez-Carretero Department of Biochemistry, Yong Loo Lin School of Medicine, National University of Singapore, Singapore, Singapore NUS Synthetic Biology for Clinical and Technological Innovation (SynCTI), Centre for Life Sciences, National University of Singapore, Singapore, Singapore W. K. D. Chee · A. Y. Teh · C. L. Poh (*) Department of Biomedical Engineering, Faculty of Engineering, National University of Singapore, Singapore, Singapore NUS Synthetic Biology for Clinical and Technological Innovation (SynCTI), Centre for Life Sciences, National University of Singapore, Singapore, Singapore e-mail: [email protected]; [email protected] © Springer Nature Switzerland AG 2022 G. Thouand (ed.), Handbook of Cell Biosensors, https://doi.org/10.1007/978-3-030-23217-7_171

339

340

J. W. Yeoh et al.

Tuning Circuit Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Failure Modes and Engineering Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions and Future Perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

370 371 374 375

Abstract

Synthetic biology is a rapidly developing field which aims to repurpose natural biological systems through the rational design of existing/new biological parts and systems, with a strong emphasis on applying engineering principles. Over the years, synthetic gene circuits have been designed to control and manipulate cellular gene expressions in practical applications including biosensing, biomanufacturing, bioremediation, and biotherapy. However, the design of genetic circuits to achieve the desired performances remains a daunting challenge compounded by typical issues of modularity, orthogonality, context dependency, stability, and predictability. To address this challenge, there has been a great advancement in our ability to design gene circuit, with new tools being developed and design principles being elucidated. As such, this chapter aims to review the underpinning process involved in gene circuit design, with the emphasis on applying it to cell-based biosensors. Accordingly, appropriate computer-aided design tools to be used during the design and construction phases as well as modelling tools that facilitate the rational design-build-test-learn cycle will be explored. Lastly, a compilation of the common failure modes as faced by typical users and recommended potential engineering solutions is presented. Moving forward, adopting better characterized genetic parts and their interfaces, embedded with appropriate design rules and principles, accompanied by advanced computer-aided tools, the full capabilities of synthetic biology can be realized as previously anticipated. Keywords

Cell-based biosensor · Gene circuits · Design principles · Rational designs · Tuning capabilities · Model-driven approach

Introduction The design of cell-based biosensors is an interdisciplinary field involving several concepts from engineering, biochemistry, and biology. It is therefore useful to establish a common ground of knowledge before diving into further concepts and design techniques. In this introductory section, the authors will therefore present a brief review of the field, with topics spanning from the history of biosensors to the types of biosensors followed by the methodology of cell-based biosensor design, among others.

15

Genetic Circuit Design Principles

341

Biosensors and Their Applications A biosensor is currently defined by the IUPAC as “a device that uses specific biochemical reactions mediated by isolated enzymes, immunosystems, tissues, organelles or whole cells to detect chemical compounds usually by electrical, thermal or optical signals” (Gold Book 2014; Mehrotra 2016). In this sense, the first reported biosensor is the glucose sensor by Leland Clark in 1962, which sparked the race to detect and measure every possible relevant biomolecule (Mehrotra 2016). It is, however, in the last two to three decades where the development of biosensors has become a field of major economic importance. This is due to the increasing demands of high-quality healthcare, the growing concerns over environmental issues, and the emergence of the bioeconomy, defined as “the production of renewable biological resources and the conversion of these resources and waste streams into value added products, such as food, feed, bio-based products and bioenergy” (Mccormick and Kautto 2013; Ronzon et al. 2017). Healthcare is, as stated above, a main area of application for biosensors. Prominent examples include point-of-care systems for the diagnosis of infectious diseases, where pathogen identification is crucial, as well as in cancer detection (Wu and Qu 2015; Holowko et al. 2016; Alahi and Mukhopadhyay 2017; Jayaraman et al. 2017). A novel area of research is theranostics, aimed at creating technologies that can provide both detection and treatment (Kelkar and Reineke 2011). The food industry is also a key player in the potential widespread use of biosensors, particularly in the area of contamination prevention during food manufacturing and food spoiling detections in supermarkets (Mustafa and Andreescu 2018). Environmental protection is another area where the use of biosensors is imperative, with the detection of heavy-metal pollutants and other high-risk contaminants as well as carbon emission monitoring (Justino et al. 2017). Finally, another area of growing concern is personal security, where there is a demand for biosensors capable of detecting security threats such as explosives and chemical and biological toxic agents (Walper et al. 2018).

Biosensing Parameters To design biosensors with an adequate performance, several design parameters need to be considered. The main parameters are presented below: Selectivity: Sometimes referred to as specificity, it is the capacity of the biosensor to detect only the analyte of interest, therefore not being responsive to other biomolecules in the system. Sensitivity: The minimum concentration of the analyte the biosensor can detect. It is often modelled as the slope of the biosensor response (Fig. 1a). Functional range: The interval of analyte concentrations that the biosensor can measure (Fig. 1a).

342

J. W. Yeoh et al.

Fig. 1 Main parameters used in sensor design. (a) Sensitivity (Δy/Δx), functional range (xmax
xmin), and dynamic range (ymax
ymin). (b) Curve 1: theoretical linear response curve. Curves 2, 3, and 4: linearization of the output curve in different ranges of the input. (c) Curve 1: output with offset zero. Curves 2 and 3: output with offset different from zero. (d) Signal-to-noise ratio: SNR1 > SNR2. (e) Response time (tr): tr1 < tr2 < tr3. (f) Hysteresis curve

Dynamic range: The interval of outputs that the biosensor can exhibit. The wider this is, the easier will be for the subsequent subsystems to read the biosensor output (Fig. 1a). Accuracy: It refers to the closeness of the measured value of a quantity with respect to the real value of the quantity, which is measured with a more accurate reference system. Deviations in accuracy can often be fixed using calibration. Precision: It refers to the closeness of repeated measurements of the same value of a certain physical quantity. Linearity: It is a measure of the likeness of the biosensor response to a straight line. Linearity is a desirable feature of a biosensor as it simplifies the interpretation of the measurements and processing by subsequent systems. If the output curve is not linear, it is also a common practice to work with a linear approximation of the curve in different ranges of the input (Fig. 1b). Offset: Closely related to linearity, it refers to the value of the output in the absence of an input. An offset value of zero is generally preferred, which can often be achieved using calibration (Fig. 1c). Signal-to-noise ratio (SNR): The level of the signal of interest with respect to the background noise (Fig. 1d).

15

Genetic Circuit Design Principles

343

Response time: The time needed for the output to reach the 95% of the final output value (Fig. 1e). Hysteresis: It measures the difference in biosensor performance while detecting variations in analyte concentration in the ascending (increasing concentrations) or the descending (decreasing concentrations) direction. When a biosensor shows hysteresis, the output of a certain input value is influenced by the previous values of the input. The presence of hysteresis in a biosensor is generally non-desirable as it takes the biosensor away from its ideal behavior (Fig. 1f).

Molecule-Based and Cell-Based Biosensors The current frantic race in the development of biosensors with better performance has led to a multitude of detection technologies. Traditionally, biosensors consist of “molecule-based” detection devices, where the molecules needed to produce an output response are the only biological elements involved. Many molecule-based biosensing technologies exist, employing enzymes or other biomolecules either dissolved or suspended in liquid phase, functionalizing a surface or in a combination of them. Common detection responses include colorimetric changes, like in home pregnancy tests, and signals of electrochemical origin interpretable by a subsequent electronic device, like in a glucose monitor (Lee 2008; Mehrotra 2016; Carpenter et al. 2018). Recently, the development in synthetic biology has greatly advanced the development of cell-based biosensors. In these biosensors, the sophisticated physiology of whole cells, typically yeast or bacteria, is used in the biodetection process. By employing whole cells as biosensors, one could take advantage of the optimization process undergone by cells during millions of years of evolution and harness it for the detection procedure. Furthermore, the applications of cell-based biosensor are not limited to in vitro applications. There have been recent reports demonstrating their potential as living biosensors to monitor the gut environment in vivo (Saltepe et al. 2017; ▶ Chap. 13, “Engineering Prokaryote Synthetic Biology Biosensors”). Despite still in its infancy, research on cell-based biosensors had already produced interesting solutions in areas such as diagnostics, theranostics, environment monitoring, and personal security (Saltepe et al. 2017; ▶ Chap. 13, “Engineering Prokaryote Synthetic Biology Biosensors”).

Sensing Architectures in Cell-Based Biosensors Similar to molecule-based biosensors, cell-based detection is typically based on three stages (Fig. 2a): single or multiple sensing of the input, signal processing, and actuation to generate an observable output response. This general architecture can, however, be implemented in a large variety of ways due to the large number of genetic structures available in different types of cells. In terms of input detection, sensing of chemical ions, small metabolites, and large proteins have been abundantly reported (Saltepe et al. 2017; ▶ Chap. 13,

344

J. W. Yeoh et al.

Fig. 2 Architecture of a cell-based biosensor. (a) The input quantity is detected by the sensing element. This will generally cause the induction or repression of a certain genetic element. These events then trigger a series of biochemical interactions that will transform the input signal, which can be regarded as the signal processing stage of the biosensor and is schematically represented as a series of logic gates. The result of the signal processing stage is generally again the induction or repression of a genetic element. This output genetic element is considered the actuating part of the biosensor, and its purpose is the generation of a reporter which can produce a measurable output quantity. (b) In the signal processing stage, a series of logical operations will be applied to the input. These operations can be achieved at transcriptional level (left panel, showing transcriptional repression on top and, at the bottom, transcriptional activation), at posttranscriptional level (center panel, showing a riboregulator), and at posttranslational level (right panel, showing the unaffected protein of interest on top and, at the bottom, how tagging with a small protein induced degradation of the protein of interest by the proteolytic machinery of the cell). RNAP RNA polymerase, RBS ribosome binding site, CDS coding sequence, crRNA cis-repressing RNA, CR cis-repressing sequence, AUG a start codon (adenine, uracil, and guanine, coding for the amino acid methionine), taRNA trans-activating RNA, TA trans-activating sequence

“Engineering Prokaryote Synthetic Biology Biosensors”). A recent study developed a whole-cell sensing platform utilizing Escherichia coli (E. coli) surface-displaying nanobodies (single-domain antibodies) which can bind selectively to a target protein

15

Genetic Circuit Design Principles

345

analyte that could be hard to diffuse across the cell membrane (Kylilis et al. 2019). In addition, detection of physical quantities as diverse as light of a wide array of wavelengths (Hughes 2018; Liu et al. 2018), temperature (Sengupta and Garrity 2013), pH (Krulwich et al. 2011), pressure (Booth 2014; Persat et al. 2015), osmolarity (Wood 1999; Yuan et al. 2017), viscosity (Anderson et al. 2009), and redox stimuli has also been reported (Green et al. 2004; Tschirhart et al. 2017). Then, the detection of the quantity of interest is commonly transduced into an internal biochemical signal by means of transcription factors, repressing or inducing gene transcription initiation by influencing the promoter region (Ang et al. 2013; Brophy and Voigt 2014; ▶ Chap. 13, “Engineering Prokaryote Synthetic Biology Biosensors”). Once the detected quantity has been transduced into a biochemical signal within the cell, a subsequent cascade of cell-signalling events can be triggered to elicit a response. This corresponds to the signal processing stage of the biosensing procedure. To better understand this signal processing stage, the biochemical interactions produced in the cell can be compared to logic gates present in digital electronic circuits. For example, if two biochemical events are needed for a third biochemical event to occur, the interaction is equivalent to an AND gate. While if only one of the biochemical events is sufficient, the interaction can be compared to an OR gate (Fig. 2a). Similarly, if the presence of one biochemical event causes another biochemical event to stop, the interaction is deemed as a NOT gate (Fig. 2a). Several distinct mechanisms have been used to incorporate genetic circuits that exhibit logic-gate functions into cell-based biosensors (Fig. 2b). One typical example is the use of transcription factors, implementing a network of repressing and inducing elements leading to complex logical circuits that control gene expression at transcription level (Ang et al. 2013; Brophy and Voigt 2014; Saltepe et al. 2017; ▶ Chap. 13, “Engineering Prokaryote Synthetic Biology Biosensors”). Gene regulation can also be achieved at posttranscriptional level. One common strategy is the use of nucleic acids to control the translation of mRNA. This includes strategies such as riboregulators, riboswitches, RNA interference, and the use of aptamers (Ang et al. 2013; Saltepe et al. 2017; ▶ Chap. 13, “Engineering Prokaryote Synthetic Biology Biosensors”). Finally, control at posttranslational level has also been achieved, mainly by the use of enzymes controlling the amount or the function of the protein of interest (Ang et al. 2013; ▶ Chap. 13, “Engineering Prokaryote Synthetic Biology Biosensors”). The last step in the sensing procedure is the output transduction from the internal biochemical language of the cell to an external detectable phenomenon. A large variety of options is also available here, with the main examples being colorimetric, fluorescent, or bioluminescent responses (Saltepe et al. 2017; ▶ Chap. 13, “Engineering Prokaryote Synthetic Biology Biosensors”), the production of particular chemicals detectable by subsequent analytical procedures (Zhao et al. 2018), chemotactic responses (Saltepe et al. 2017), and the generation of an electrochemical signal detectable by means of a subsequent electronic device (Tschirhart et al. 2017).

346

J. W. Yeoh et al.

Advantages and Challenges of Cell-Based Biosensors The use of the untapped potential of cell physiology is posed to revolutionize the field of biosensors. Cells are certainly the most sophisticated molecular machines, far beyond any current fully synthetic attempt. As a result, the use of cell-based biosensors could allow us to achieve the most advanced device possible in a size of just a few micrometers, incorporating, potentially in a completely autonomous manner, sensing of one or several phenomena, a complex signal processing logic and a whole palette of output responses. This would make cell-based biosensors the most adequate technology for environments where no human intervention is possible, like in in vivo applications (Saltepe et al. 2017; ▶ Chap. 13, “Engineering Prokaryote Synthetic Biology Biosensors”). This high sophistication would also allow the incorporation of both sensing and actuation capabilities into one single cell. Relevant applications include theranostics, where cells could detect, for example, the type of infecting pathogen and release the adequate toxin, bioremediation, with cells detecting the type of pollutant and responding accordingly, and biofabrication, where factory cells could detect the phase of the production cycle and activate the corresponding biosynthesis pathway. In addition, the advanced cellular machinery makes the use of biosensors very interesting from an economic point of view. In particular, cell division makes fabrication of the biosensor virtually inexpensive, therefore lowering the barriers for scaling biosensing platforms up to industrial application level. However, the advantages of cell-based biosensors also create challenges to their use. Since cell-based biosensors can act autonomously, they may behave in unexpected ways if accidentally released. This requires the use of biocontainment technologies to prevent health and environmental issues. A common solution is the use of kill switches (Gómez-tatay and Hernández-andreu 2019; Wang and Zhang 2019). The major challenge, however, is the complexity of cell-based biosensors, which affects the use of the cell-based biosensor technology in a variety of ways. One is the high variability and lack of reproducibility found when dealing with bacterial behavior, which originates partially in the stochastic nature of cell behavior and the variability in human-performed tasks, but fundamentally in the lack of understanding of cell physiology. This hinders the development and commercialization of novel cell-based biosensor technologies. The challenges originated by the elevated cell complexity can therefore be traced back to the inability to accurately predict cell behavior. To solve this key issue, several research approaches have been proposed. One important research area is the development of minimal cells. These are cells that, while still maintaining the required properties to perform their intended function, have a minimum level of internal complexity, which results in larger predictability (Hutchison et al. 2016; Glass et al. 2017). Another key contribution comes from the progressive transformation of the design and construction of biological systems into an engineering discipline, termed “synthetic biology” or “engineering biology” (Cameron et al. 2014; Raimbault et al. 2016; Shapira et al. 2017).

15

Genetic Circuit Design Principles

347

Synthetic Biology and the Rational Design of Cell-Based Biosensors The discipline of synthetic biology aims at applying engineering principles to achieve a rational design of biological systems with predictable behavior. This design would be based on the use of an interconnected network of well-characterized elements termed “parts.” Similar to traditional engineering disciplines, synthetic biology employs the principles of modularity, characterization and standardization, following a systematic design-build-test-learn (DBTL) cycle (Fig. 3) supported by the use of computer-aided design (CAD) tools. Although still in its infancy, the development and use of synthetic biology is already a key factor for the construction of increasingly robust biological systems with improved performance (Cameron et al. 2014; Raimbault et al. 2016; Shapira et al. 2017). When applying a synthetic biology approach to develop cell-based biosensors, all the necessary tasks are arranged around the DBTL cycle. Following this methodology, the first phase would be the design of the system general structure. The main aspect to consider here is the application of the biosensor, which would affect the rest of the design. For example, a biosensor employed for in vivo detection would avoid the use of any potentially pathogenic host, while a biosensor involved in bioproduction might need a certain growth rate as well as to withstand the culture conditions (e.g., pH and temperature) of the production process. Another important aspect is to establish the specifications needed for the biosensing procedure, briefly mentioned in section “Biosensing Parameters.” It is worth noting that, although a designer might always want the best possible value in every parameter of the biosensor, this is often not possible and a compromise is usually needed. In addition, the biosensing specifications might also affect the selection of the host chassis, since the genetic makeup of the host might affect parameters of the biosensor performance such as the selectivity. These considerations will not be covered in detail in this chapter due to its application-specific nature as well as the lack of reliable design guidelines. At the end of the design phase, the designer will have a candidate gene circuit design based on the interconnection of parts in schematic format. The next step will therefore be the initial modelling of the gene circuit followed by the actual construction of the gene circuit and its incorporation into the host. The system is then tested

Fig. 3 A systematic design-build-test-learn (DBTL) cycle and the tasks underlie the individual phases

348

J. W. Yeoh et al.

and characterized. Then, the information generated in the testing phase will be analyzed and used in the next iteration of the circuit design, repeating the cycle until the designer is satisfied with the results. The following sections aim at introducing the reader to the different stages of the DBTL cycle in more detail. Section “Principles in Design and Construction of Genetic Circuits” covers the important considerations during the design and construction phases of genetic circuits, with a brief mention of reporter genes commonly used in the testing phase. Section “Model-driven Approach towards Rational Design” focuses on the learning phase and, particularly, the use of a model-driven approach in design optimization . Finally, section “Tuning Circuit Performance” provides useful strategies to improve the circuit performance of biosensors and a list of common problems/failure modes with possible solutions.

Principles in Design and Construction of Genetic Circuits This section presents the important design considerations and guidelines for gene circuit construction. A list of important online platforms and experimental tools is also provided at the end of the section for reference.

Choice of Host Chassis To date, advances in synthetic biology have continually streamlined the development of cell-based biosensors with unprecedented capabilities in the field of environment monitoring and healthcare. Soil-based bacteria such as Pseudomonas fluorescens, Pseudomonas putida, or Staphylococcus aureus have been engineered to detect traces of xenobiotics and heavy metals in the soil (Chang et al. 2017). Similarly, prokaryotic-based biosensors and yeast-based biosensors are deployed in food monitoring to trace the presence of residual heavy metals and the presence of antibiotics resistance bacteria. In the healthcare application, engineered Escherichia coli (E. coli) have been developed as an in vivo diagnostics solution to detect liver metastasis in the urine, and yeast-based antibody panel has been created to serve as a point-of-care solution for biomarker detection. Among the large variety of choice of host chassis for biosensing, E. coli bacteria are commonly used as the workhorse for gene cloning/expression due to the high manipulability for gene cloning, minimal biosafety risks of several nonpathogenic strains, the ease of growing rapidly, and its commercial scalability. In this chapter, the authors thus elucidate the genetic circuit design principles surrounding the use of E. coli as the host chassis. In the typical design workflow, a designer would need to select an appropriate chassis for molecular cloning and, subsequently, decide on the expressional (terminal) chassis to allow for proper functioning of the biosensor in the site-specific application. In molecular cloning, commercial cloning strains such as E. coli DH10B and DH5α are preferably used, owing to its high transformation efficiency and the ability to assimilate large plasmid sizes (~10 kbp). This is because these cloning strains housed beneficial mutations such as recA1 (reduce homologous recombination from occurring) and

15

Genetic Circuit Design Principles

349

endA1 (reduce nonspecific cleavage of double-stranded DNA) to improve the stability of the plasmid. Depending on the complexity and size of the designed plasmids, the designer can consider either using heat shock method on chemically competent cells for smaller and less complex plasmid ( > > > > < > > > > > > :

½xn , ½x þ K x n n


1
K maxrep

½xn ½x þ K x n

Inducible system

1, Constitutive system  Repressible system n ,

ð5Þ

364

J. W. Yeoh et al.

d ½ Pn  ¼ Sp ½mRNA
K mat ½Pn  dt

ð6Þ

d ½ Pm  ¼ K mat ½Pn 
dP ½Pm  dt

ð7Þ

Here, SmRNA and dmRNA refer to the promoter-dependent mRNA synthesis rate and degradation rate respectively, and SP and dP correspond to the RBS-dependent protein synthesis rate and protein degradation rate. The protein degradation rate considers both cell growth dependent dilution rate and active degradation rate of the protein. R(x) is a function used to account for the promoter-dependent effect in response to the availability of ligands (activators or repressors), with n and Kx being related to ligand binding cooperativity and the ligand concentration producing half occupancy and Kmaxrep denoting the maximum repression capacity. Fluorescent proteins (FPs) are extensively used as reporters for characterizing the responses of cellular biosensor designs. However, nascent FPs have to undergo maturation steps before they can fluoresce. High variations of maturation time have been reported and found to be closely related to the cell strains and culture conditions (Balleza et al. 2018). The protein maturation kinetics can be described using Eqs. 6 and 7 to better capture the output response dynamics. All these parameters could be tuned using the relevant experimental approaches as listed in Table 3 to achieve the desired ultimate response behavior as demonstrated in Fig. 7. These equations form the basis for developing models to describe more complex gene circuits such as logic gates, which often constitute the sensing and signal processing modules of a biosensor. The model frameworks for routinely used logic gates like NOT, AND, and OR gates can be acquired from the model bank of the BioModel Selection System (BMSS) accessible at https://github.com/EngBioNUS/ BMSSlib (Yeoh et al. 2019).

Design and Modelling Tools In view of the increasing importance in applying the model-driven approach to facilitate the experimental implementation, a growing number of computer-aided platforms have been developed to automate or ease the efforts of model development and simulation processes. Most of the tools available employ forward engineering approach to simulate and predict circuit behavior based on previous characterized databases, whereas some tools focus on reverse engineering to fit the experimental measurements and provide mechanistic insights. These design and modelling tools are reviewed in terms of their approaches and functionalities in Table 4. Exhaustive lists of other tools have been featured at http://sbolstandard.org/applications/ and in this reference (Appleton et al. 2017). Integrating both approaches is prerequisite to effectively derive insightful information from experiments, which in turn empowers the simulation and prediction processes for achieving high-performance modeldriven design and optimization approaches.

15

Genetic Circuit Design Principles

365

Table 3 Representative response function tuning with reference to Fig. 7 using experimental approaches Response Increase Kmax to achieve vertical extension Reduce leakiness, where the sum of Kb and Kmax remains constant

To achieve vertical scaling where Kb and Kmax are scaled equally

Lower Km to improve sensitivity (horizontal scaling)

Experimental tuning approaches - Tune the activation potency of the binding ligand - Modify promoter with multiple operator sites - Use orthogonal RNAPs - Perform mutagenesis or random truncation on promoter - Add degradation tag to the reporter protein (reduce maximum output level as well) - Insert a protease cleavage site between reporter protein and degradation tag, with the cognate protease regulated by the sensing module - Increase the strength of repression for transcriptional derepression system - Highly dependent on the location of repressor binding site: position between the core region
35 and
10 is stronger than the proximal location at downstream of
10 followed by the distal location upstream of
35 (▶ Chap. 13, “Engineering Prokaryote Synthetic Biology Biosensors”) - Add an extra repressor binding site downstream, and tune the distance between the binding sites - Insert multiple hairpins into 30 UTR of mRNA amenable to cleavage by ribonuclease - Add multiple transcriptional attenuator sequence into the 50 UTR - Adjust translation efficacy through RBS sequence or RNA secondary structure near to the start codon - Modify plasmid copy number (origin of replication) - Change gene location: higher expression with reduced proximity to the origin - Modify promoter strength, mutated upstream sequence - Modify 20 nucleotide downstream sequence of the transcription start site, which could affect the promoter escape - Adjust the spacer length - Modify RBS strength, vary the sequence of the 16S rRNA - Use orthogonal RNAP or ribosome - Modify start codon, codon optimization (translation elongation) - Tuning mRNA degradation rate - Modify the binding affinity of the ligand to the promoter - Change the sequence of the binding sites for the RNAPs,
10 and
35 hexameric upstream sequence - For transcriptional derepression system, lower the strength of the promoter expressing the underlying repressor - Add an extra repressor binding site downstream, and (continued)

366

J. W. Yeoh et al.

Table 3 (continued) Response

Experimental tuning approaches

Adjust Kb to achieve vertical shifting Adjust/increase n to improve response steepness

tune the distance between the binding sites - Increase the number of amplifier layers (Wan et al. 2019) - Tune the ligand-independent constitutive source - Adjust the cooperative binding of multiple ligands to the promoter - Vary the position of operator sites - Add transcriptional attenuator sequence into the 50 UTR - Increase cascade length (Wan et al. 2019) - Use positive feedback loop

Table 4 Model-based CAD tools Tools Matlab SimBiology Toolbox (MathWorks)

COPASI (Hoops et al. 2006)

SynBioSS (Weeding et al. 2010)

TinkerCell (Chandran et al. 2010)

RBS Calculator (Salis 2011)

Descriptions/functions - Provides a block diagram editor and programmatic tools to build model and simulate and analyze dynamic systems - Enables import and export models in SBML format - Supports deterministic and stochastic simulations - Supports sensitivity analysis and parameter inferences - Input: user-defined model based on reaction network - Enables import and export models in SBML - Supports deterministic kinetic or stochastic simulations - Supports time-series and steady-state simulations - Supports model analysis (sensitivity analysis) and parameter estimations - Input: user adding biological components - A software suite consists of three components: Designer, WIKI, and Simulator - Linked to BioBrick’s biological components - Enables generation of protein reactions (multimerization) and additional reactions such as leakiness, degradation, transport, etc. - Automatically generates a kinetic model from the desired construct - Retrieves required kinetic information from database in WIKI - Simulates dynamic behavior of biological systems - Outputs a SBML or NetCDF file - Supports detailed visual diagram for simulations - Default dynamic rate equation-based models automatically derived from structure - Allows deterministic and stochastic simulations and few other analyses supported by COPASI - Supports SBOL standard files - Input: mRNA sequence and/or first 35 nucleotides of a protein CDS (continued)

15

Genetic Circuit Design Principles

367

Table 4 (continued) Tools

Cello (Nielsen et al. 2016)

iBioSim (Watanabe et al. 2018)

Tellurium (Choi et al. 2018)

BMSS (Yeoh et al. 2019)

Descriptions/functions - Predicts the translation initiation rate for every start codon of mRNA transcript - Designs and optimizes synthetic RBS sequence to achieve the desired translation rate for protein expression - Based on thermodynamic model (Gibbs free energy of ribosome binding) combined with stochastic optimization (simulated annealing) - Input: user-defined desired circuit function using Verilog language, and specify the sensors, actuators, and user constraints file (UCF) - Gate assignment composed of NOT/NOR logic based on repressors - Maps steady-state response function - Assignment algorithms include breadth-first search, hill climbing, and simulated annealing - Predicts gene circuit performance factoring in growth, load, and population variability - Provides the complete DNA sequence with the selected circuit configuration which are compliant to the specified Eugene rules - Supports both systems and synthetic biology applications - Capable of importing and exporting SBML and SBOL standard files - For design, modelling, and analysis of genetic circuits - Supports both deterministic and stochastic simulations - The default model parameters are inferred and can be refined using model editor - A Python-based environment for model building, simulation, and analysis - Supports various standards including SBML, SBOL, SEDML (Simulation Experiment Description Markup Language), and COMBINE archive - Enables deterministic and stochastic simulations and steadystate analyses - Supports standard libraries and other advanced modules for wide applications (visualization, multidimensional parameter scanning, and bifurcation analysis) - Input: time-series characterization data - Automates the model selection process and outputs the best model candidate with underlying mechanistic insights on the expression profiles - Selects model based on trade-off between goodness of fit and model complexity - Based on deterministic kinetic model - Contains model bank storing models for inducible and constitutive promoter systems, logic-gate system (NOT, AND, and OR gates) - Exports SBML file to enable further cross-platform analysis

368

J. W. Yeoh et al.

Interoperability between tools via a common standard is a long-standing effort in synthetic biology. This aims to overcome the reproducibility challenge and could serve as a common integrated knowledge base for sharing and supporting reusability. A list of standards can be found at http://co.mbine.org/standards. Synthetic Biology Open Language (SBOL) data and Systems Biology Markup Language (SBML) are the omnipresent community standards used in synthetic biology for communicating gene circuit designs (Hucka et al. 2003; Galdzicki et al. 2014). SBOL captures structural information of the synthetic biological parts and at the DNA levels, devices, and systems (Galdzicki et al. 2014), whereas SBML supports quantitative behavioral model representation of the biological devices or systems (Hucka et al. 2003). There have been community-driven efforts such as SynBioHub repository (https://synbiohub.org/) in providing computational access for software and data integration based upon the SBOL standard (McLaughlin et al. 2018) and COmputational Modeling in BIology NEtwork (COMBINE) archive developed by COMBINE community to embed all the relevant standard files in a single ZIP container to facilitate the reproduction of modelling and simulation experiments (https:// combinearchive.org/index/) (Bergmann et al. 2014). Conversion between the two main standards, SBOL and SBML, could further enhance the consistency and transferability, as well as facilitate the design automation process (Nguyen et al. 2016; Misirli et al. 2018).

A Rational Design Optimization Workflow The ability to implement a robust rational design of synthetic gene circuits, similar to what has been accomplished by other engineering disciplines, is a constant goal in synthetic biology. This, however, is inevitably deterred by the five hard truths of synthetic biology: undefined parts, unpredictable circuitry, unwieldy complexity, incompatible parts, and variability (Kwok 2010). One of the basis to access the rational design capability is the use of well-characterized parts and well-characterized interfaces (Freemont and Kitney 2012; Appleton et al. 2017; Karamasioti et al. 2017). Built on these two key criteria, accompanied by advances in computer-based aided design and modelling tools, one could simulate before building and fine-tune the circuit performances in an unprecedented way and scale. Harnessing the power of existing available model-based CAD tools as elaborated in Table 4, a potential rational design workflow that combines both reverse and forward engineering approaches, as shown in Fig. 8a, can be used to engineer a functional gene circuit and fine-tune the circuit performances. To achieve a robust predictive power, it is essential to account for the variability of biological and physical contexts of the characterized parts or gene circuits instead of using databases with distinctive preestablished contextual settings. One alternative could be the use of an automated model selection tool which serves as a knowledge base and hypothesis-testing tool to store the models carrying different hypotheses and to identify the most appropriate model to describe data characterized under a particular condition. The role of this tool coupled with other aforementioned simulation and analysis tools in facilitating the rational design cycle is illustrated in Fig. 8b.

15

Genetic Circuit Design Principles

369

Fig. 8 A rational design workflow for model-driven functional gene circuit design and optimization supported by modelling tools. (a) Cello enables high-level input-output specifications written in Verilog, along with genetic parts, and structural design rules specified in Eugene language. Output is the recommended circuit construct to be implemented with predicted steadystate profiles. Dynamic models can be generated and simulated by iBioSim using structural SBOL data. This is then followed by the actual gene circuit construction and its characterization. The characterization data can then be fed into BMSS for automated model selection via hypothesis testing and providing insights into the gene circuit behavior. Further model analyses such as sensitivity analysis and parameter tuning can again be performed using COPASI or iBioSim to optimize the circuit performances. (b) To account for the context-dependent nature of the biological circuit for improved model predictive power, BMSS could serve as an automated hypothesis-testing tool to generate a customized model for data measured under a specific condition. Forward analysis and model prediction can subsequently be executed, which in turn complement the rational design optimization process

370

J. W. Yeoh et al.

Tuning Circuit Performance A biosensor performance is typically evaluated through the response curve, functional dynamic range, sensitivity and specificity, activation time, stability, ease of use, and so forth. However, there are several challenges to be addressed to achieve the desired requirements for useful applications. Here, the authors will explore some of the key issues and the potential solutions to counteract the problems. Context Dependence: Phenomena where unintended construct-construct interactions or host-construct interactions occur, often attributed to the use of shared resources (building blocks, RNAPs, ribosomes, tRNA, protease) or toxicity, which could critically impact cell growth as well as gene circuit stability and functionality. This is particularly relevant as the complexity of synthetic constructs increases, often leading to unpredictable behavior. Host-aware synthetic constructs with burdendriven negative feedback regulators could potentially balance the resources allocation between biomass and expression (Ceroni et al. 2015, 2018; Huang et al. 2018; Boo et al. 2019). By fine-tuning the feedback regulators, the coupling effects could be mitigated. A more global alternative to resource relocation for reduced burden could be the use of native MazF ribonuclease to cleave unprotected mRNA (Venturelli et al. 2017). The influence of context dependence can be further extended into environmental factors such as culture media compositions, temperature, etc. which could also contribute to host burden and poor cell fitness (Yeoh et al. 2019). It is therefore important to characterize the individual parts and interfaces before composing into more complex genetic circuits. Modularity/Orthogonality: Modularity refers to the conserved function or consistent response observed across changing physical or genetic contexts, which is the core criteria for predictive design of gene circuits. Orthogonality, a complementary feature, implies lack of undesirable interactions or crosstalk among elements and with the host genetic context. These two properties are the principal criteria for constructing large genetic circuit from composable modular parts, promoting part compatibility and scalability. A proper insulation between parts or modules to achieve modularity can be achieved via adding standardized spacers between parts, cleaving 50 UTR by ribozymes, or using a bicistronic design which incorporates an upstream RBS to disrupt mRNA structure across the junction between cistrons (Davis et al. 2011; Mutalik et al. 2013; Nielsen et al. 2016; Carr et al. 2017). Meanwhile, orthogonality requires combinatorial screening experiments to determine part compatibility or the use of orthogonal machineries exogenous to the native host cells to avoid crosstalk or gene coupling (An and Chin 2009; Rhodius et al. 2013). Specificity/Selectivity: Most of the sensing modules are promiscuous which can be triggered by ligands with similar properties, especially in the case of heavy metals. Mutagenesis is a routine approach employed to modify the binding pockets to increase specificity (Cayron et al. 2017). Another, less time-consuming alternative, could be the use of an AND logic gate which is composed of two or more nonspecific sensing modules that can detect the same ligand, as demonstrated by

15

Genetic Circuit Design Principles

371

zinc biosensor derived from HrpRS transcription activators constructed by Bernard and Wang (2017) (Wang et al. 2013; Bernard and Wang 2017). AND gates can also be used to detect the simultaneous presence of multiple specific inputs (Wang et al. 2013). Sensitivity/Limit of Detection: To improve the sensitivity of biosensors, random mutagenesis can be performed on transcription factors or promoters to detect lower ligand concentrations. Alternatively, altering ligand membrane transport systems, such as increasing importers or/and reducing exporters, could facilitate the accumulation of ligands within the cells, thus improving the detection limit (Hynninen et al. 2010). For transcriptional derepression systems, lowering the strength of the promoters expressing repressors could lower the ligand concentration demand needed to derepress the transcriptional system (Wan et al. 2019). For the transcriptional activator counterparts, higher sensitivity can be achieved by increasing the promoter strength for activator expression to incur higher binding probability (▶ Chap. 13, “Engineering Prokaryote Synthetic Biology Biosensors”). In addition, cascaded amplifiers were also found to boost sensitivity beyond amplifying output dynamic range (Wan et al. 2019). Dynamic Range: Aside from performing random mutagenesis to improve the transcriptional output range, transcriptional amplifiers consisting of a transcriptional factor and its cognate promoter could also be adopted to extend the output range (Kim et al. 2016). The dynamic range can be further augmented by incorporating positive feedback loop (Nistala et al. 2010). Response Curves Mapping: Matching the output dynamic range with the input functional range between consecutive modules for complex circuit construction is crucial to maximize the responsive dynamic range of biosensors (Nielsen et al. 2016). This could be accomplished by tuning the RBS strengths to achieve the desired expression level, extending the circuit with modules exhibiting relevant response function, or modifying the circuit topology. Response Time: For cell-based biosensors using fluorescent proteins as reporters, lag response time could be challenging due to the protein maturation time needed for stable fluorescence (Gui et al. 2017). For example, 50 fluorescent proteins spanning the visible spectrum were found to display significantly different maturation kinetics which could be a key consideration in the reporter selection process (Balleza et al. 2018). Despite this, fluorescent proteins are highly stable and have no additional substrate requirement. In contrast, bioluminescence, which exhibits rapid response, requires O2 or/and ATP to emit (Gui et al. 2017).

Failure Modes and Engineering Solutions Assembling multiple fundamental parts or logic gates to build complex circuitries in biosensors can be problematic due to the nonlinear behavior of biological system, caused by interactions at different levels and their interdependencies. This inherently renders the context-dependent behavior of the actual biosensors, which often results in low predictability of model-driven approach and undesirable performances in

372

J. W. Yeoh et al.

experimental designs. In response to these, herein, the authors compile some of the common phenomena as observed in typical genetic circuitries for biosensors, the corresponding failure modes, and the potential engineering solutions to address the problems as a reference in Table 5. Table 5 Failure modes encountered and potential engineering solutions when designing simple and layered gene circuitries for biosensors Phenomenon Circuit when paired with different promoters

Failure mode Transcriptional interference: promoters generated transcripts with different 50 untranslated regions (50 UTRs), thus different response function (Nielsen et al. 2016)

Circuit with tandem promoters

“Road-blocking” phenomenon: downstream promoter reduces upstream transcription (Nielsen et al. 2016)

Input switches

Genetic crosstalk, homologous recombination

Engineering solution - Downstream insulation: use a hammerhead ribozyme and downstream hairpin (RiboJ) to cleave away the 50 UTRs to eliminate transfer function discrepancies (Nielsen et al. 2016) - Upstream insulation: add promoter spacers (e.g., ~15 nucleotides of random sequence) (Nielsen et al. 2016) - Use insulated promoter-gene cassettes flanked by controlled sequences at both upstream and downstream of the transcriptional initiation site - Use CRISPR editing to remove undesired nucleotide at 50 UTRs upstream of RBS and gene coding (Brophy and Voigt 2014) - Transcriptional insulation: add unique and strong terminators with non-repetitive sequences to avoid homologous recombination (Nielsen et al. 2016) - Separate the promoters into different expression cassettes/ plasmids (Wong et al. 2015) - Measure propensity of promoters to roadblock and used to create Eugene rules/constraints (Nielsen et al. 2016) - Separate the promoters into distinct expression cassettes (Wong et al. 2015) - Check pairwise compatibility (Wong et al. 2015) - Perform mutagenesis to identify orthogonal pairs (Wong et al. 2015) - Avoid repetitive usage of the same parts (continued)

15

Genetic Circuit Design Principles

373

Table 5 (continued) Phenomenon Layered circuit

Failure mode Resources competition: parallel usage of the shared resources

Transcriptional read-through

Logic gates (AND, OR, etc.)

Stoichiometric mismatch: skewed responses or leakiness

Logic gates (NOT gate)

Insufficient repression capability

Toxicity

Growth inhibition; sequence mutation in plasmid or/and genome

Homologous recombination

Mutations that result in deletion of chain of nucleotides

Engineering solution - Use orthogonal RNAPs (such as T7 RNAPs) or ribosomes - Use low plasmid copy for the non-reporting modules and high plasmid copy for the reporter module to increase detection limit (Wan et al. 2019) - Use strong, tandem terminators (Brophy and Voigt 2014) - Employ antisense transcription (insert the second promoter downstream in an opposite direction) (▶ Chap. 13, “Engineering Prokaryote Synthetic Biology Biosensors”) - Characterize the different parts individually to understand their expression profiles - Use model to predict the stoichiometric balance configurations (Wong et al. 2015) - Perform mutagenesis on operator binding region to identify the optimal sequence - Place the repressor expression near to origin (higher expression especially for fast-growing cells) (Shis et al. 2018) - Use multiple repressor binding sites, adjust the position, and distance between binding sites (Ang et al. 2013) - Reduce the expression level (use low copy number plasmid, weaker promoters, or RBSs) - Introduce site mutagenesis in proximity of active site of target protein to identify the proteins with less toxicity (Temme et al. 2012) - Split the protein into several fragments (Segall-shapiro et al. 2014) - Check for sequence similarities for plasmid and with genome (NCBI BLAST) - Avoid long identical nucleotide sequences (>25 bp)

374

J. W. Yeoh et al.

Conclusions and Future Perspectives Rapid advancement in synthetic biology over the years has fueled the exciting development of biosensor for many different applications. This has resulted in a better understanding of how genetic circuits should be designed for biosensor application, as well as tools that can assist in their development. Despite these advancements, challenges including the non-ideal gene circuit, inadequate tuning capabilities, and complexity of biological systems still limit the development of cellbased biosensors toward practical real-world applications. In this vein, the synthetic biology field, empowered by the valuable insights from computer-aided tools, could provide a streamlined rational design process into developing more reliable biosensors with desired capabilities. Clearly, there is still a gap in the understanding of the intricacies of nonlinear interplays among biological components. As the circuit complexities increase, greater emphasis should be placed on maintaining modularity and orthogonality of individual genetic elements in order to make the biological system more predictable. Model-driven approach coupling both forward and reverse engineering approaches should be leveraged upon to better shed light on the system behavior and the underlying interactions to give better biological prediction. In addition, this approach could be particularly useful for fine-tuning the gene circuit design configurations to achieve the desired performance in large combinatorial studies. The outlook of future successes in cell-based biosensor development is pinned on the importance of addressing biosafety issues related to their uses. From the lens of biosafety, the cell-based biosensors are regulated as genetically modified organisms (GMOs) that might pose environmental/health threats of different levels, if they do “escape” from their designated environments. This makes the inclusion of biocontainment systems into biosensor applications more essential than ever. Two main biocontainment strategies are commonly employed. The first focuses on limiting the spread of genetic materials to other environmental organisms via horizontal gene transfer (HGT) (Saltepe et al. 2017). Typically, plasmids containing the biosensor cassette are preferably utilized as opposed to directly introducing the cassette into the genome. To further reduce the chances of HGT to occur, additional regulatory systems have been incorporated to control the gene expression (Gallagher et al. 2015), and kill switches, which are sensitive to various chemical and physical stimuli, were deployed to express toxin molecules in the non-designated environment (Pedrolli et al. 2018). Secondly, the other method is the use of orthogonal biology or xenobiology, which incorporates nonnatural amino acids and nucleotides as building blocks, that does not interfere with the host’s activities and is anticipated to halt the transfer of genetic information into the environment (Diwo and Budisa 2019). In short, only until the biosafety issue is resolved that the full capabilities of the cell-based biosensor would be embraced and be applied into real-world applications.

15

Genetic Circuit Design Principles

375

References Alahi M, Mukhopadhyay S (2017) Detection methodologies for pathogen and toxins: a review. Sensors 17:1885. https://doi.org/10.3390/s17081885 Almquist J, Cvijovic M, Hatzimanikatis V, Nielsen J (2014) Kinetic models in industrial biotechnology – improving cell factory performance. Metab Eng 24:38–60. https://doi.org/10.1016/j. ymben.2014.03.007 An W, Chin JW (2009) Synthesis of orthogonal transcription- translation networks. PNAS 106:8477–8482 Anderson JK, Smith TG, Hoover TR (2009) Sense and sensibility: flagellum- mediated gene regulation. Trends Microbiol 18:30–37. https://doi.org/10.1016/j.tim.2009.11.001 Ang J, Harris E, Hussey BJ, Kil R, Mcmillen DR (2013) Tuning response curves for synthetic biology. ACS Synth Biol 2:547–567 Aoki SK, Lillacci G, Gupta A, Baumschlager A, Schweingruber D, Khammash M (2019) A universal biomolecular integral feedback controller for robust perfect adaptation. Nature 570:533. https://doi.org/10.1038/s41586-019-1321-1 Appleton E, Madsen C, Roehner N, Densmore D (2017) Design automation in synthetic biology. Cold Spring Harb Perspect Biol 9:a023978 Ay A, Arnosti DN (2011) Mathematical modeling of gene expression: a guide for the perplexed biologist. Crit Rev Biochem Mol Biol 46:137–151. https://doi.org/10.3109/ 10409238.2011.556597 Bae S, Park J, Kim J (2014) Cas-OFFinder: a fast and versatile algorithm that searches for potential off-target sites of Cas9 RNA-guided endonucleases. Bioinformatics 30:1473–1475. https://doi. org/10.1093/bioinformatics/btu048 Balleza E, Kim JM, Cluzel P (2018) Systematic characterization of maturation time of fluorescent proteins in living cells. Nat Methods 15:47. https://doi.org/10.1038/nmeth.4509 Baronas R, Ivanauskas F, Kulys J (2009) Mathematical modeling of biosensors: an introduction for chemists and mathematicians. Springer Science & Business Media, Dordrecht Bergmann FT, Adams R, Moodie S, Cooper J, Glont M, Golebiewski M, Hucka M, Laibe C, Miller AK, Nickerson DP, Olivier BG, Rodriguez N, Sauro HM, Scharm M, Soiland-reyes S, Waltemath D, Yvon F (2014) COMBINE archive and OMEX format: one file to share all information to reproduce a modeling project. BMC Bioinf 15:369 Bernard E, Wang B (2017) Synthetic cell-based sensors with programmed selectivity and sensitivity. In: Biosensors and biodetection. Humana Press, New York, pp 349–363 Bhosekar A, Ierapetritou M (2018) Advances in surrogate based modeling, feasibility analysis, and optimization: a review. Comput Chem Eng 108:250–267. https://doi.org/10.1016/j. compchemeng.2017.09.017 Bird LE, Rada H, Flanagan J, Diprose JM, Gilbert RJ, Owens RJ (2014) Application of in-fusion™ cloning for the parallel construction of E. coli expression vectors. In: DNA cloning and assembly methods. Humana Press, Totowa, New Jersey, pp 209–234 Boo A, Ellis T, Stan G (2019) Host-aware synthetic biology. Curr Opin Syst Biol 14:66–72. https:// doi.org/10.1016/j.coisb.2019.03.001 Booth IR (2014) Bacterial mechanosensitive channels: progress towards an understanding of their roles in cell physiology. Curr Opin Microbiol 18:16–22. https://doi.org/10.1016/j. mib.2014.01.005 Briat C, Gupta A, Khammash M (2016) Antithetic integral feedback ensures robust perfect adaptation in Noisy biomolecular networks. Cell Syst 2:15–26. https://doi.org/10.1016/j. cels.2016.01.004 Brophy JAN, Voigt CA (2014) Principles of genetic circuit design. Nat Methods 11:508–520. https://doi.org/10.1038/nmeth.2926. Principles

376

J. W. Yeoh et al.

Cameron DE, Bashor CJ, Collins JJ (2014) A brief history of synthetic biology. Nat Rev Microbiol 12:381. https://doi.org/10.1038/nrmicro3239 Carpenter AC, Paulsen IT, Williams TC (2018) Blueprints for biosensors: design, limitations, and applications. Genes (Basel) 9:375. https://doi.org/10.3390/genes9080375 Carr SB, Beal J, Densmore DM (2017) Reducing DNA context dependence in bacterial promoters. PLoS One 12:e0176013 Casini A, Macdonald JT, De JJ, Christodoulou G, Freemont PS, Baldwin GS, Ellis T (2014) Onepot DNA construction for synthetic biology: the modular overlap-directed assembly with linkers (MODAL) strategy. Nucleic Acids Res 42:e7. https://doi.org/10.1093/nar/gkt915 Casini A, Storch M, Baldwin GS, Ellis T (2015) Bricks and blueprints: methods and standards for DNA assembly. Nat Rev Mol Cell Biol 16:568. https://doi.org/10.1038/nrm4014 Cayron J, Prudent E, Escoffier C, Gueguen E (2017) Pushing the limits of nickel detection to nanomolar range using a set of engineered bioluminescent Escherichia coli. Environ Sci Pollut Res 24:4–14. https://doi.org/10.1007/s11356-015-5580-6 Ceroni F, Algar R, Stan G, Ellis T (2015) Quantifying cellular capacity identifies gene expression designs with reduced burden. Nat Methods 12:415. https://doi.org/10.1038/nmeth.3339 Ceroni F, Boo A, Furini S, Gorochowski TE, Borkowski O, Ladak YN, Awan AR, Gilbert C, Stan G, Ellis T (2018) Burden-driven feedback control of gene expression. Nat Methods 15:387. https://doi.org/10.1038/nmeth.4635 Chandran D, Bergmann FT, Sauro HM (2010) Computer-aided design of biological circuits using TinkerCell. Bioeng Bugs 1:274–281 Chang H, Voyvodic PL, Zúñiga A, Bonnet J (2017) Microbially derived biosensors for diagnosis, monitoring and epidemiology. Microb Biotechnol 10:1031–1035. https://doi.org/10.1111/17517915.12791 Chao R, Yuan Y, Zhao H (2015) Recent advances in DNA assembly technologies. FEMS Yeast Res 15:1–9. https://doi.org/10.1111/1567-1364.12171 Chen Y, Liu P, Nielsen AAK, Brophy JAN, Clancy K, Peterson T, Voigt CA (2013) Characterization of 582 natural and synthetic terminators and quantification of their design constraints. Nat Methods 10:659. https://doi.org/10.1038/nmeth.2515 Choi K, Medley JK, König M, Stocking K, Smith L, Gu S, Sauro HM (2018) Tellurium: An extensible python-based modeling environment for systems and synthetic biology. Biosystems 171:74–79. https://doi.org/10.1016/j.biosystems.2018.07.006 Craney A, Hohenauer T, Xu Y, Navani NK, Li Y (2007) A synthetic luxCDABE gene cluster optimized for expression in high-GC bacteria. Nucleic Acids Res 35:e46. https://doi.org/ 10.1093/nar/gkm086 Daniel R, Almog R, Ron A, Belkin S, Shacahm Y (2008) Modeling and measurement of a wholecell bioluminescent biosensor based on a single photon avalanche diode. Biosens Bioelectron 24:882–887. https://doi.org/10.1016/j.bios.2008.07.026 Davis JH, Rubin AJ, Sauer RT (2011) Design, construction and characterization of a set of insulated bacterial promoters. Nucleic Acids Res 39:1131–1141. https://doi.org/10.1093/nar/gkq810 de Kok S, Stanton LH, Slaby T, Durot M, Holmes VF, Patel KG, Platt D, Shapland EB, Serber Z, Dean J, Newman JD, Chandran SS (2014) Rapid and reliable DNA assembly via ligase cycling reaction. ACS Synth Biol 3:97–106. https://doi.org/10.1021/sb4001992 Diwo C, Budisa N (2019) Alternative biochemistries for alien life: basic concepts and requirements for the design of a robust biocontainment system in genetic isolation. Genes (Basel) 10:17. https://doi.org/10.3390/genes10010017 Doyle EL, Booher NJ, Standage DS, Voytas DF, Brendel VP, Vandyk JK, Bogdanove AJ (2012) TAL effector-nucleotide Targeter (TALE-NT) 2.0: tools for TAL effector design and target prediction. Nucleic Acids Res 40:W117–W122. https://doi.org/10.1093/nar/gks608 Ellis T, Adie T, Baldwin GS (2011) DNA assembly for synthetic biology: from parts to pathways and beyond. Integr Biol 3:109–118. https://doi.org/10.1039/c0ib00070a Enghiad B, Zhao H (2017) Programmable DNA-guided artificial restriction enzymes. ACS Synth Biol 6:752–757. https://doi.org/10.1021/acssynbio.6b00324

15

Genetic Circuit Design Principles

377

Engler C, Kandzia R, Marillonnet S (2008) A one pot, one step, precision cloning method with high throughput capability. PLoS One 3:e3647. https://doi.org/10.1371/journal.pone.0003647 Fan M, Ferenc M, Juccheim A, LaManna C, Monroe M, Morgan K, Niehaus J, Patrick M, Swanson L, Julian T-P (2014) Plasmids 101: a desktop resource. Available at https://www.addgene.org Freemont PS, Kitney RI (2012) Synthetic biology – a primer, rev edn. World Scientific Publishing Company Freemont PS, Kitney RI (2015) Synthetic biology – a primer, rev edn. World Scientific Publishing Company Galdzicki M, Clancy KP, Oberortner E, Pocock M, Quinn JY, Rodriguez CA, Roehner N, Wilson ML, Adam L, Anderson JC, Bartley BA, Beal J, Chandran D, Chen J, Densmore D, Endy D, Grünberg R, Hallinan J, Hillson NJ, Johnson JD, Kuchinsky A, Lux M, Misirli G, Peccoud J, Plahar HA, Sirin E, Stan G, Villalobos A, Wipat A, Gennari JH, Myers CJ, Sauro HM (2014) The synthetic biology open language (SBOL) provides a community standard for communicating designs in synthetic biology. Nat Biotechnol 32:545–550. https://doi.org/10.1038/nbt.2891 Gallagher RR, Patel JR, Interiano AL, Rovner AJ, Isaacs FJ (2015) Multilayered genetic safeguards limit growth of microorganisms to defined environments. Nucleic Acids Res 43:1945–1954. https://doi.org/10.1093/nar/gku1378 Gibson DG, Young L, Chuang R, Venter JC, Iii CAH, Smith HO, America N (2009) Enzymatic assembly of DNA molecules up to several hundred kilobases. Nat Methods 6:343. https://doi. org/10.1038/NMETH.1318 Gibson DG, Smith HO, Iii CAH, Venter JC, Merryman C (2010) Chemical synthesis of the mouse mitochondrial genome. Nat Methods 7:901. https://doi.org/10.1038/nmeth.1515 Glass JI, Merryman C, Wise KS, Iii CAH, Smith HO (2017) Minimal cells – real and imagined. Cold Spring Harb Perspect Biol 9:a023861 Gold Book (2014) International Union of Pure and Applied Chemistry Compendium of chemical terminology. International Union of Pure and Applied Chemistry, Research Triangle Park Gómez-tatay L, Hernández-andreu JM (2019) Biosafety and biosecurity in synthetic biology: a review. Crit Rev Environ Sci Technol 49:1587–1621. https://doi.org/10.1080/10643389.2019.1579628 Gounaris CA, Floudas CE (2009) A review of recent advances in global optimization. J Glob Optim 45:3–38. https://doi.org/10.1007/s10898-008-9332-8 Green J, Paget MS, Bank W (2004) Bacterial redox sensors. Nat Rev Microbiol 2:954. https://doi. org/10.1038/nrmicro1022 Grote A, Hiller K, Scheer M, Munch R, Nortemann B, Hempel DC, Jahn D (2005) JCat: a novel tool to adapt codon usage of a target gene to its potential expression host. Nucleic Acids Res 33: W526–W531. https://doi.org/10.1093/nar/gki376 Gui Q, Lawson T, Shan S, Yan L, Liu Y (2017) The application of whole cell-based biosensors for use in environmental analysis and in medical diagnostics. Sensors 17:1623. https://doi.org/ 10.3390/s17071623 Haeussler M, Schönig K, Eckert H, Eschstruth A, Mianné J, Renaud J, Schneider-maunoury S, Shkumatava A, Teboul L, Kent J, Joly J (2016) Evaluation of off-target and on-target scoring algorithms and integration into the guide RNA selection tool CRISPOR. Genome Biol 17:148. https://doi.org/10.1186/s13059-016-1012-2 Hancock EJ, Stan G, Arpino JAJ, Stan G (2015) Simplified mechanistic models of gene regulation for analysis and design. J R Soc Interface 12:20150312 Harrison ME, Dunlop MJ (2012) Synthetic feedback loop model for increasing microbial biofuel production using a biosensor. Front Microbiol 3:360. https://doi.org/10.3389/ fmicb.2012.00360 Heigwer F, Kerr G, Walther N, Glaeser K, Pelz O, Breinig M, Boutros M (2013) E-TALEN: a web tool to design TALENs for genome engineering. Nucleic Acids Res 41:e190. https://doi.org/ 10.1093/nar/gkt789 Hillson NJ, Rosengarten RD, Keasling JD (2012) j5 DNA assembly design automation software. ACS Synth Biol 1:14–21. https://doi.org/10.1021/sb2000116

378

J. W. Yeoh et al.

Holowko MB, Wang H, Jayaraman P, Poh CL (2016) Biosensing Vibrio cholerae with genetically engineered Escherichia coli. ACS Synth Biol 5:1275–1283. https://doi.org/10.1021/ acssynbio.6b00079 Hoops S, Sahle S, Gauges R, Lee C, Simus N, Singhal M, Xu L, Mendes P, Kummer U (2006) COPASI – a COmplex PAthway SImulator. Bioinformatics 22:3067–3074. https://doi.org/ 10.1093/bioinformatics/btl485 Huang H, Qian Y, Del Vecchio D (2018) A quasi-integral controller for adaptation of genetic modules to variable ribosome demand. Nat Commun 9:5415. https://doi.org/10.1038/s41467018-07899-z Hucka M, Sauro HM, Doyle J, Kitano H (2003) The systems biology markup language (SBML): a medium for representation and exchange of. Bioinformatics 19:524–531. https://doi.org/ 10.1093/bioinformatics/btg015 Hughes RM (2018) A compendium of chemical and genetic approaches to light-regulated gene transcription. Crit Rev Biochem Mol Biol 53:453–474. https://doi.org/10.1080/ 10409238.2018.1487382 Hutchison CA, Chuang RY, Noskov VN, Assad-Garcia N, Deerinck TJ, Ellisman MH, Gill J, Kannan K, Karas BJ, Ma L, Pelletier JF (2016) Design and synthesis of a minimal bacterial genome. Science 351(6280):aad6253. https://doi.org/10.1126/science.aad6253 Hwang IY, Tan MH, Koh E, Ho CL, Poh CL, Chang MW (2014) Reprogramming microbes to be pathogen-seeking killers. ACS Synth Biol 3:228–237. https://doi.org/10.1021/sb400077j Hynninen A, Tõnismann K, Virta M (2010) Improving the sensitivity of bacterial bioreporters for heavy metals. Bioeng Bugs 1:132–138 Jayaraman P, Devarajan K, Chua TK, Zhang H, Gunawan E, Poh CL (2016) Blue light-mediated transcriptional activation and repression of gene expression in bacteria. Nucleic Acids Res 44:6994–7005. https://doi.org/10.1093/nar/gkw548 Jayaraman P, Holowko MB, Yeoh JW, Lim S, Poh CL (2017) Repurposing a two-component system-based biosensor for the killing of Vibrio cholerae. ACS Synth Biol 6:1403–1415. https:// doi.org/10.1021/acssynbio.7b00058 Jayaraman P, Yeoh JW, Zhang J, Poh CL (2018) Programming dynamic control of bacterial gene expression with the chimeric ligand and light-based promoter system. ACS Synth Biol 7:2627. https://doi.org/10.1021/acssynbio.8b00280 Justino C, Duarte A, Rocha-Santos T (2017) Recent progress in biosensors for environmental monitoring: a review. Sensors 17:2918. https://doi.org/10.3390/s17122918 Karamasioti E, Lormeau C, Stelling J (2017) Computational design of biological circuits: putting parts into context. Mol Syst Des Eng 2:410–421. https://doi.org/10.1039/c7me00032d Kelkar SS, Reineke TM (2011) Theranostics: combining imaging and therapy. Bioconjug Chem 22:1879–1903. https://doi.org/10.1021/bc200151q Kery MB, Feldman M, Livny J, Tjaden B (2014) TargetRNA2: identifying targets of small regulatory RNAs in bacteria. Nucleic Acids Res 42:W124–W129. https://doi.org/10.1093/nar/ gku317 Kim HJ, Lim JW, Jeong H, Lee S, Lee D, Kim T, Lee SJ (2016) Development of a highly specific and sensitive cadmium and lead microbial biosensor using synthetic CadC-T7 genetic circuitry. Biosens Bioelectron 79:701–708. https://doi.org/10.1016/j.bios.2015.12.101 Knight T (2003) Idempotent vector design for standard assembly of biobricks. https://www. addgene.org Krulwich TA, Sachs G, Padan E (2011) Molecular aspects of bacterial pH sensing and homeostasis. Nat Rev Microbiol 9:330. https://doi.org/10.1038/nrmicro2549 Kulkarni VV, Stan G-B, Raman K (eds) (2014a) A systems theoretic approach to systems and synthetic biology I: models and system characterizations. Springer, Dordrecht Kulkarni VV, Stan G-B, Raman K (eds) (2014b) A systems theoretic approach to systems and synthetic biology II: analysis and design of cellular systems. Springer, Dordrecht Kwok R (2010) Five hard truths for synthetic biology. Nat News 463:288–290 Kylilis N, Riangrungroj P, Lai H-E, Salema V, Fernández LA, Stan G-BV, Freemont PS, KM P (2019) Whole-cell biosensor with tunable limit of detection enables low-cost agglutination

15

Genetic Circuit Design Principles

379

assays for medical diagnostic applications. ACS Sensors 4:370–378. https://doi.org/10.1021/ acssensors.8b01163 Labun K, Montague TG, Krause M, Cleuren YNT, Valen E (2019) CHOPCHOP v3: expanding the CRISPR web toolbox beyond genome editing. Nucleic Acids Res 47:W171–W174. https://doi. org/10.1093/nar/gkz365 Lee TM (2008) Over-the-counter biosensors: past, present, and future. Sensors 8:5535–5559. https://doi.org/10.3390/s8095535 Lin D, Callaghan CAO (2018) MetClo: methylase-assisted hierarchical DNA assembly using a single type IIS restriction enzyme. Nucleic Acids Res 46:e113. https://doi.org/10.1093/nar/ gky596 Liu Z, Zhang J, Jin J, Geng Z, Qi Q, Liang Q (2018) Programming bacteria with light – sensors and applications in synthetic biology. Front Microbiol 9:2692. https://doi.org/10.3389/ fmicb.2018.02692 Mann M, Wright PR, Backofen R (2017) IntaRNA 2.0: enhanced and customizable prediction of RNA-RNA interactions. Nucleic Acids Res 45:W435–W439. https://doi.org/10.1093/nar/ gkx279 Mannan AA, Liu D, Zhang F (2017) Fundamental design principles for transcription-factor-based metabolite biosensors. ACS Synth Biol 6:1851–1859. https://doi.org/10.1021/acssynbio.7b00172 Mccormick K, Kautto N (2013) The bioeconomy in Europe: An overview. Sustainability 5:2589–2608. https://doi.org/10.3390/su5062589 McLaughlin JA, Myers CJ, Zundel Z, Mlslrll G, Zhang M, Ofiteru ID, Goñi-Moreno A, Wipat A (2018) SynBioHub: a standards-enabled design repository for synthetic biology. ACS Synth Biol 7:682–688. https://doi.org/10.1021/acssynbio.7b00403 Mehrotra P (2016) Biosensors and their applications – a review. J Oral Biol Craniofacial Res 6:153–159. https://doi.org/10.1016/j.jobcr.2015.12.002 Misirli G, Nguyen T, Mclaughlin JA, Vaidyanathan P, Jones TS, Densmore D, Myers C, Wipat A (2018) A computational workflow for the automated generation of models of genetic designs. ACS Synth Biol 8:1548. https://doi.org/10.1021/acssynbio.7b00459 Morio J (2011) Global and local sensitivity analysis methods for a physical system. Eur J Phys 32:1577. https://doi.org/10.1088/0143-0807/32/6/011 Mustafa F, Andreescu S (2018) Chemical and biological sensors for food-quality monitoring and smart packaging. Foods 7:168. https://doi.org/10.3390/foods7100168 Mutalik VK, Guimaraes JC, Cambray G, Lam C, Christoffersen MJ, Mai Q, Tran AB, Paull M, Keasling JD, Arkin AP, Endy D (2013) Precise and reliable gene expression via standard transcription and translation initiation elements. Nat Methods 10:354. https://doi.org/10.1038/ nmeth.2404 Nguyen T, Roehner N, Zundel Z, Myers CJ (2016) A converter from the systems biology markup language to the synthetic biology open language. ACS Synth Biol 5:479–486. https://doi.org/ 10.1021/acssynbio.5b00212 Nielsen AAK, Der BS, Shin J, Vaidyanathan P, Paralanov V, Strychalski EA, Ross D, Densmore D, Voigt CA (2016) Genetic circuit design automation. Science 352(6281):aac7341. https://doi. org/10.1126/science.aac7341 Nistala GJ, Wu K, Rao CV, Bhalerao KD (2010) A modular positive feedback-based gene amplifier. J Biol Eng 4:1–8 Nonaka G, Blankschien M, Herman C, Gross CA, Rhodius VA (2006) Regulon and promoter analysis of the E. coli heat-shock factor, α32, reveals a multifaceted cellular response to heat stress. Genes Dev 20:1776–1789. https://doi.org/10.1101/gad.1428206.but Olson EJ, Hartsough LA, Landry BP, Shroff R, Tabor JJ (2014) Characterizing bacterial gene circuit dynamics with optically programmed gene expression signals. Nat Methods 11:449. https://doi. org/10.1038/nmeth.2884 Oyarzún DA, Stan G-BV (2012) Synthetic gene circuits for metabolic control: design trade-offs and constraints. J R Soc Interface 10:20120671 Oyarzún DA, Lugagne J-B, Stan G-BV (2015) Noise propagation in synthetic gene circuits for metabolic control. ACS Synth Biol 4:116–125. https://doi.org/10.1021/sb400126a

380

J. W. Yeoh et al.

Pédelacq J-D, Cabantous S, Tran T, Thomas C, Terwilliger TC, Waldo GS (2006) Engineering and characterization of a superfolder green fluorescent protein. Nat Biotechnol 24:79–89. https://doi. org/10.1038/nbt1172 Pedrolli DB, Ribeiro NV, Squizato PN, de Jesus VN, Aqa T, Gracindo A, Cesar MB, Paulo J, Freire C, Costa AFM, Lins MRCR, Correa GG, Cerri MO (2018) Engineering microbial living therapeutics: the synthetic biology toolbox. Trends Biotechnol 37:100. https://doi.org/ 10.1016/j.tibtech.2018.09.005 Persat A, Nadell CD, Kim MK, Ingremeau F, Siryaporn A, Drescher K, Wingreen NS, Bassler BL, Gitai Z, Stone HA (2015) The mechanical world of Bacteria. Cell 161:988–997. https://doi.org/ 10.1016/j.cell.2015.05.005 Qing G, Ma L, Khorchid A, Swapna GVT, Mal TK, Takayama MM, Xia B, Phadtare S, Ke H, Acton T, Montelione GT, Ikura M, Inouye M (2004) Cold-shock induced high-yield protein production in Escherichia coli. Nat Biotechnol 22:877. https://doi.org/10.1038/nbt984 Quan J, Tian J (2014) Circular polymerase extension cloning. In: DNA cloning and assembly methods. Humana Press, pp 103–117 Raimbault B, Cointet J, Joly P (2016) Mapping the emergence of synthetic biology. PLoS One 11: e0161522. https://doi.org/10.1371/journal.pone.0161522 Rhodius VA, Segall-shapiro TH, Sharon BD, Ghodasara A, Orlova E, Tabakh H, Burkhardt DH, Clancy K, Peterson TC, Gross CA, Voigt CA (2013) Design of orthogonal genetic switches based on a crosstalk map of σs, anti-σs, and promoters. Mol Syst Biol 9:1–13. https://doi.org/ 10.1038/msb.2013.58 Ronzon T, Lusser M, Klinkenberg M, Landa L, Sanchez Lopez J, M’Barek R, Hadjamu G, Belward A, Camia A, Giuntoli J, Cristobal J, Parisi C, Ferrari E, Marelli L, Torres de Matos C, Gomez Barbero M, Rodriguez Cerezo E (2017) JRC science for policy report. Bioeconomy report 2016. https://www.addgene.org Salis HM (2011) The ribosome binding site calculator. Methods Enzymol 498:19–42 Saltepe B, Kehribar ES, Su Yirmibeşoğlu SS, Şafak Şeker UO (2017) Cellular biosensors with engineered genetic circuits. ACS Sensors 3:13–26. https://doi.org/10.1021/ acssensors.7b00728 Segall-shapiro TH, Meyer AJ, Ellington AD, Sontag ED, Voigt CA (2014) A “resource allocator” for transcription based on a highly fragmented T 7 RNA polymerase. Mol Syst Biol 10:742 Sengupta P, Garrity P (2013) Sensing temperature. Curr Biol 23:R304–R307. https://doi.org/ 10.1016/j.cub.2013.03.009 Shao Z, Zhao H, Zhao H (2009) DNA assembler, an in vivo genetic method for rapid construction of biochemical pathways. Nucleic Acids Res 37:e16. https://doi.org/10.1093/nar/gkn991 Shapira P, Kwon S, Youtie J (2017) Tracking the emergence of synthetic biology. Scientometrics 112:1439–1469. https://doi.org/10.1007/s11192-017-2452-5 Shaw WM, Yamauchi H, Mead J, Wigglesworth M, Ladds G, Ellis T (2019) Engineering a model cell for rational tuning of GPCR Signaling. Cell 177:782–796. https://doi.org/10.1016/j. cell.2019.02.023 Shis DL, Bennett MR, Igoshin OA (2018) Dynamics of bacterial gene regulatory networks. Annu Rev Biophys 47:447–467 Storch M, Casini A, Mackrow B, Fleming T, Trewhitt H, Ellis T, Baldwin GS (2015) BASIC: a new biopart assembly standard for idempotent cloning provides accurate, single-tier DNA assembly for synthetic biology. ACS Synth Biol 4:781–787. https://doi.org/10.1021/sb500356d Taylor GM, Mordaka PM, Heap JT (2019) Start-stop assembly: a functionally scarless DNA assembly system optimized for metabolic engineering. Nucleic Acids Res 47:e17. https://doi. org/10.1093/nar/gky1182 Temme K, Hill R, Segall-shapiro TH, Moser F, Voigt CA (2012) Modular control of multiple pathways using engineered orthogonal T7 polymerases. Nucleic Acids Res 40:8773–8781. https://doi.org/10.1093/nar/gks597

15

Genetic Circuit Design Principles

381

Torella JP, Boehm CR, Lienert F, Chen J, Way JC, Silver PA (2014) Rapid construction of insulated genetic circuits via synthetic sequence-guided isothermal assembly. Nucleic Acids Res 42:681–689. https://doi.org/10.1093/nar/gkt860 Trubitsyna M, Michlewski G, Cai Y, Elfick A, French E (2014) PaperClip: rapid multi-part DNA assembly from existing libraries. Nucleic Acids Res 42:e154. https://doi.org/10.1093/nar/gku829 Tschirhart T, Kim E, Mckay R, Ueda H, Wu H, Pottash AE, Zargar A, Negrete A, Shiloach J, Payne GF, Bentley WE (2017) Electronic control of gene expression and cell behaviour in Escherichia coli through redox signalling. Nat Commun 8:14030. https://doi.org/10.1038/ncomms14030 Venturelli OS, Tei M, Bauer S, Chan LJG, Petzold CJ, Arkin AP (2017) Programming mRNA decay to modulate synthetic circuit resource allocation. Nat Commun 8:15128. https://doi.org/ 10.1038/ncomms15128 Walper SA, Lasarte Aragonés G, Sapsford KE, III BCW, Rowland CE, Breger JC, Medintz IL (2018) Detecting biothreat agents: from current diagnostics to developing sensor technologies. ACS Sensors 3:1894–2024. https://doi.org/10.1021/acssensors.8b00420 Wan X, Volpetti F, Petrova E, French C, Maerkl SJ, Wang B (2019) Cascaded amplifying circuits enable ultrasensitive cellular sensors for toxic metals. Nat Chem Biol 15:540–548. https://doi. org/10.1038/s41589-019-0244-3 Wang F, Zhang W (2019) Synthetic biology: recent progress , biosafety and biosecurity concerns, and possible solutions. J Biosaf Biosecurity 1:22–30. https://doi.org/10.1016/j. jobb.2018.12.003 Wang B, Barahona M, Buck M (2013) A modular cell-based biosensor using engineered genetic logic circuits to detect and integrate multiple environmental signals. Biosens Bioelectron 40:368–376. https://doi.org/10.1016/j.bios.2012.08.011 Watanabe L, Nguyen T, Zhang M, Zundel Z, Zhang Z, Madsen C, Roehner N, Myers CJ (2018) iBioSim 3: a tool for model-based genetic circuit design. ACS Synth Biol 8:1560. https://doi. org/10.1021/acssynbio.8b00078 Weber E, Engler C, Gruetzner R, Werner S, Marillonnet S (2011) A modular cloning system for standardized assembly of multigene constructs. PLoS One 6:e16765. https://doi.org/10.1371/ journal.pone.0016765 Weeding E, Houle J, Kaznessis YN (2010) SynBioSS designer: a web-based tool for the automated generation of kinetic models for synthetic biological constructs. Brief Bioinform 11:394–402. https://doi.org/10.1093/bib/bbq002 White BA (1993) PCR protocols: current methods and applications. Springer Science & Business Media, Totowa Wong A, Wang H, Poh CL, Kitney RI (2015) Layering genetic circuits to build a single cell bacterial half adder. BMC Biol 13:1–16. https://doi.org/10.1186/s12915-015-0146-0 Wood JM (1999) Osmosensing by Bacteria: signals and membrane-based sensors. Microbiol Mol Biol Rev 63:230–262 Wu L, Qu X (2015) Cancer biomarker detection: recent achievements and challenges. Chem Soc Rev 44:2963–2997. https://doi.org/10.1039/c4cs00370e Yeoh JW, Ivan Ng KB, Teh AY, Zhang J, David Chee WK, Poh CL (2019) An automated biomodel selection system (BMSS) for gene circuit designs. ACS Synth Biol 8:1484. https://doi.org/ 10.1021/acssynbio.8b00523 Yuan J, Jin F, Glatter T, Sourjik V (2017) Osmosensing by the bacterial PhoQ/PhoP two-component system. Proc Natl Acad Sci 114:E10792–E10798. https://doi.org/10.1073/pnas.1717272114 Zhao EM, Zhang Y, Mehl J, Park H, Lalwani MA, Toettcher JE, Avalos JL (2018) Optogenetic regulation of engineered cellular metabolism for microbial chemical production. Nature 555:683. https://doi.org/10.1038/nature26141 Zou R, Zhou K, Stephanopoulos G, Too HP (2013) Combinatorial engineering of 1-deoxy-DXylulose 5-phosphate pathway using cross-lapping in vitro assembly ( CLIVA ) method. PLoS One 8:e79557. https://doi.org/10.1371/journal.pone.0079557

Fundamental Building Blocks of Whole-Cell Biosensor Design

16

Ke Yan Wen, Jack W. Rutter, Chris P. Barnes, and Linda Dekker

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sensing Mechanisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Transcriptional . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Translational . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Posttranslational . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Design Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Inputs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chassis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Optimizing Biosensor Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tuning Design Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Enhancing Sensing Mechanisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Environmental Compatibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions and Future Directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

384 385 385 386 387 388 388 390 392 394 394 397 398 398 399

Abstract

Whole-cell biosensors are one of the most popular applications of synthetic biology. Biological sensing mechanisms originating from natural systems are coupled to response modules in living cells to create engineered organisms that can sense target compounds and produce a measurable output. This chapter reviews the fundamental building blocks of whole-cell biosensor design and the methods that are available for the development of whole-cell biosensors. A wide variety of target molecules have been shown to be detectable by whole-cell biosensors, ranging from heavy metals to metabolic products. The mechanism of detection can occur at the transcriptional, translational, or posttranslational K. Y. Wen · J. W. Rutter · C. P. Barnes (*) · L. Dekker Division of Biosciences, University College London, London, UK e-mail: [email protected]; [email protected]; [email protected]; linda. [email protected] © Springer Nature Switzerland AG 2022 G. Thouand (ed.), Handbook of Cell Biosensors, https://doi.org/10.1007/978-3-030-23217-7_181

383

384

K. Y. Wen et al.

level within the cell. Further design considerations such as inputs, output, and chassis need to be selected depending on the purpose of the biosensor. Once a design has been finalized, it is common that optimization of the device is necessary in order to meet performance requirements. This can be done through methods such as modification of the genetic design, mutation of the sensing element, or addition of control elements such as feedback loops or amplifiers. Creating a mathematical model of the biosensing system is a vital step in this process as it enables a priori prediction of which parts to modify in order to achieve the desired behavior. Taking into account the fundamental building blocks and design choices enables the development of whole-cell biosensors that can be used for real-world applications in the future.

Introduction Sensors capable of detecting specific compounds are required in a wide range of areas, such as environmental monitoring, healthcare, and industrial processes. Whole-cell biosensors have emerged in recent years to address these needs and are one of the most popular applications of synthetic biology. By taking advantage of the native ability of cells to sense and respond to their environment, biosensor devices can be created that are specific, highly sensitive toward their target compound, and produce an easily monitored response. Commercial biosensing systems using bacterial cells have existed for over 30 years, such as the Microtox (Curtis et al. 1982; De Zwart and Slooff 1983) and BioTox kits (reviewed in Kokkali and Van Delft 2014). These systems contain naturally bioluminescent bacteria to monitor environmental samples, based on the principle that compounds which are toxic to metabolic processes will lead to a reduction in overall bioluminescence. Unlike such non-specific methods however, genetically engineered biosensors capable of sensing single compounds can be constructed for more targeted monitoring. They generally incorporate multiple elements to enable the three key functions of sensors – detection, transduction, and output – to exist within a single cell. In this way, sensing mechanisms can be linked to measurable outputs such as fluorescence for quantification, or signal processing elements to produce desired cellular behavior (such as the production of a toxin). In this chapter, types of whole-cell biosensors developed in synthetic biology will be reviewed and categorized by the different types of sensing mechanisms, with a focus on microbial-based systems. The various design options for input, output, and chassis that are available will then be discussed. For biosensors to suit the envisaged applications, they must fulfil key requirements such as high sensitivity toward relevant concentration ranges and low background signal response. How biosensors can be optimized to obtain the desired performance using tools, such as mathematical modelling or directed evolution of genetically encoded parts, will also be summarized. While this is a rapidly growing area of research, this chapter aims to provide a snapshot of the current methods and parts that are available for the design of novel whole-cell biosensors.

16

Fundamental Building Blocks of Whole-Cell Biosensor Design

385

Sensing Mechanisms Nature has provided a plethora of biological sensing systems and signalling pathways that can be exploited within biosensing systems using synthetic biology. The sensing mechanism is a fundamental building block of whole-cell biosensor design and can be transcriptional, translational, or posttranslational, depending on the application.

Transcriptional Transcription factor (TF) systems controlling microbial physiology at the transcriptional level allow gene expression to be controlled in the presence of a compound of interest (Fig. 1). Transcriptional sensing mechanisms consist of an activator or repressor protein regulating the transcriptional activity of a specific promoter. Upon interaction with a small molecule or environmental stress signal, the TF will undergo a conformational change to decrease or increase its DNA binding affinity. Most transcriptional biosensors are built by linking environment-responsive promoters to engineered gene circuits for programmed transcriptional changes (Khalil and Collins 2010). Promoters from the Escherichia coli lac, tet, and ara operons have been extensively characterized and are frequently incorporated in synthetic designs. Many databases are available to gain information on prokaryotic

Fig. 1 Major types of transcription-regulating sensing mechanisms used in engineered biosensors

386

K. Y. Wen et al.

transcription factors and regulons (Munch et al. 2003; Wilson et al. 2008; Novichkov et al. 2013). A promoter’s sensitivity to a molecule can be tuned by adding, deleting, or modifying activator and repressor sites, and additional control over modularity and specificity can be achieved by embedding environment-responsive promoters in engineered gene networks (discussed in more detail below). An emerging trend within transcriptional-based biosensors is to engineer synthetic promoters rather than utilizing natural ones (Dekker and Polizzi 2017). Bacterial transmembrane signalling systems are another important source of sensing mechanisms for biosensors. Some compounds are unable to cross the cytoplasmic membrane, and therefore the sensing mechanism must be located on the exterior of the cell. The three main systems are one-component systems, two component systems (TCS), and extracytoplasmic function (ECF) σ factors (Jung et al. 2018). One-component signalling systems consist of a membrane-integrated sensor domain and an intracellular DNA-binding domain; they are considered the simplest class of transmembrane signalling since they do not contain a phosphoryl acceptor domain (Jung et al. 2018). CadC is an example of a one-component system which is activated by low pH and the presence of external lysine (Fritz et al. 2009). Two-component systems (TCS) are widely found in prokaryotes and typically consist of a histidine kinase (HK) and response regulator (RR). The HK acts as a sensor and phosphorylates the RR which then regulates expression of the effector gene. The modular design of TCS can be exploited to create sensor kinases with novel target specificities. Levskaya et al. (2005) developed an E. coli-based optical sensor where a synthetic sensor kinase was constructed by fusing a membranebound cyanobacterial photoreceptor to an E. coli intracellular histidine kinase, which allowed cells to identify and report the presence of red light. In another example, Riglar et al. (2017) engineered a commensal E. coli strain with the TtrR/TtrS twocomponent system. This was coupled to a memory device output to colonize the murine gut and sense the presence of tetrathionate, showing that inflammation could be monitored up to 6 months after initial application. ECF σ factors are alternative σ factors that recruit RNA polymerase to target promoter regions and are the third class of bacterial signal transduction proteins. Cognate anti-σ factors bind the ECF σ factors and only release them in the presence of activating signal (Helmann 2002). Most of the research so far in refactoring ECFs has looked at their utility for creating highly orthogonal genetic switches, such as in Rhodius et al. (2013), but they also have potential to be used for biosensing purposes since they have been found to respond to a wide diversity of signals (Mascher 2013).

Translational Many regulatory RNA molecules are natural environmental sensors and are therefore suitable for biosensing applications. Translational biosensors are typically built by linking RNA aptamer domains to RNA regulatory domains (Khalil and Collins 2010). Riboswitches bind specific small-molecule ligands through aptamer domains and induce conformational changes in the 5’ UTR of their own mRNA, thereby

16

Fundamental Building Blocks of Whole-Cell Biosensor Design

387

regulating gene expression. They can respond rapidly to signals since the RNA has already been transcribed and can therefore be immediately bound and perform its regulatory function (Liu et al. 2017). A riboswitch calculator was recently developed which has opened up new opportunities for using riboswitches as biological sensors (Espah Borujeni et al. 2016). Another type of translational sensing system is the RNA toehold switch, which can be designed to bind and detect virtually any RNA sequence (Green et al. 2014). The toehold switch sequesters the RBS and start codon in a hairpin structure, preventing translation unless in the presence of a complementary trigger RNA. Green et al. showed that toehold switches could be integrated into the E. coli genome to regulate gene expression in response to endogenous RNAs. Although originally designed for use in bacteria, toehold switches have also been implemented in cellfree paper-based biosensors for a range of diagnostic applications (Pardee et al. 2014, 2016; Takahashi et al. 2018).

Posttranslational Posttranslational sensing mechanisms are based on proteins that are constitutively produced but become activated in the presence of the target. An example of a posttranslational sensing mechanism is an enzyme that catalyzes the reaction of the target with one or more secondary substrates to a produce a signal (Goers et al. 2013). Förster resonance energy transfer (FRET) is a type of energy transfer from a donor molecule to an acceptor molecule and can be used as an intracellular sensing mechanism. In fluorescence-based methods, a ligand binding domain is positioned between two fluorophores and induces a conformational change when bound by the target ligand, resulting in a shift in emission (reviewed in Lindenburg and Merkx 2014). Bioluminescence resonance energy transfer (BRET) sensors use a bioluminescent enzyme as the donor molecule instead (Komatsu et al. 2018). Although genetically encoded FRET sensors have largely been developed for studying intracellular interactions in mammalian cells, whole-cell biosensors using FRET principles have also been designed enabling detection of small molecules in E. coli and yeast cells (Sourjik and Berg 2002; Fehr et al. 2002; Ameen et al. 2016). Sensing extracellular proteins poses a challenge for whole-cell biosensors since there is a physical barrier of cell membrane and/or cell wall between the molecule of interest and the detection element inside the cell. To circumvent this, different approaches have been proposed based on displaying the detection element on the cell surface. These types of mechanisms rely on a protein or fusion protein that is exposed to the external environment outside of the cell. When these proteins interact with the component, they are designed to detect modifications such as activation or cleavage that can be enabled. In one study, the enzymatic activity of the target protein elastase from the Schistosomiasis parasite was detected by displaying a polyhistidine-tag anchored to the cell surface of either E. coli or Bacillus subtilis with an elastase-specific peptide sequence (Webb et al. 2016). Presence of active

388

K. Y. Wen et al.

elastase cleaved off the tag and resulted in loss of a fluorescent label binding to the tag. Another mode of extracellular protein detection utilized surface display of camelid nanobodies in E. coli cells (Kylilis et al. 2019). When the target human fibrinogen protein, a biomarker for cardiac disease, was present, selective binding of the nanobodies to the target molecules resulted in cell agglutination which could be visualized by eye.

Design Considerations When constructing a whole-cell biosensor, there are many design choices at different levels, including the type of target, the output module, and the choice of chassis, all of which must be considered specifically for the application at hand.

Inputs The input is the target for detection and is generally a compound of interest specific to the area of application. The input choice will often, but not always, determine the sensing mechanism of the biosensor. Inputs can be chemical elements (heavy metals and metalloids), small molecules, proteins, or physical phenomena, all of which will be summarized briefly below. A variety of natural mechanisms can be found in microorganisms to sense and detoxify elements such as heavy metals, enabling cells to live in otherwise harsh environments. These have been exploited for the production of biosensors, such as metal-responsive TFs from the MerR and ArsR/SmtB families. The MerR family of TFs works as activators. This includes the GolS protein from Salmonella enterica, which was inserted into E. coli to create a bacterial biosensor for gold ion complexes, with a lower limit of detection of 2 ppb (Zammit et al. 2013). By introducing a single mutation in GolS and using an E. coli strain with a knockout Zn(II) transporter zntA gene, Cerminati et al. (2015) created a broad-spectrum biosensor that was also responsive to Hg, Pb, or Cd salts in solution. ArsR is a transcriptional repressor from the E. coli chromosomal ars operon, which binds to its cognate promoter and is released when bound to arsenate or arsenite ions. This was the focus of an early iGEM project by Edinburgh in 2006 (Aleksic et al. 2007; Joshi et al. 2009; French et al. 2011a) and has continued to be developed with the aim of preventing arsenic poisoning in areas with contaminated groundwater. Eukaryotic metal-sensing systems are also available, such as a biosensor in Saccharomyces cerevisiae which used the cup1 promoter and native Ace1 protein to sense copper ions in solution (Shetty et al. 2004). Gutiérrez et al. (2015) argue that eukaryotic systems can be advantageous for monitoring toxicity due to their closer similarity to the human genome than bacteria. Certain compounds, such as carotenoids, are naturally pigmented and can thus be directly visualized or measured with absorbance. However, this is rare; for most molecules of interest, it is necessary to develop specific biosensors. As mentioned

16

Fundamental Building Blocks of Whole-Cell Biosensor Design

389

above, sugars are well-known inducers of gene expression in bacteria, such as arabinose and lactose induction of the AraC and LacI transcriptional regulators, respectively. In order to avoid inducers that also function as alternative carbon sources for the cells, they can be replaced with chemical analogues such as isopropyl-β-D-1-thiogalactoside (IPTG) for LacI (Lewis et al. 1996), or the transcriptional regulator can be mutated to accept rarer isomers, e.g., D-arabinose for AraC (Tang et al. 2008). Another mutant of AraC was identified to create a biosensor specific to mevalonate, a precursor in isoprenoid synthesis pathways (Tang and Cirino 2011). Glucose sensing has also been engineered for a bacterial biosensor using the native E. coli pCpxP promoter (Courbet et al. 2015). Amino acids are another useful target for detection since they are precursors or intermediates in many metabolic pathways of interest. In yeast, a biosensor for tryptophan was developed by modifying the Aro80p TF and associated ARO9 promoter, which the authors then used to screen for strains with improved production of the polymer-precursor muconic acid (Leavitt et al. 2016, 2017). The bacterial species Corynebacterium glutamicum is used industrially to produce amino acids, and Mustafi et al. (2012) created a biosensor from the native Lrp TF which allowed them to identify strains with increased intracellular L-valine, L-leucine, and L-isoleucine levels. In the same species, a biosensor sensitive to L-lysine, L-arginine, and L-histidine – composed of the LysG TF controlling EYFP expression – was used to select mutants with improved lysine secretion (Binder et al. 2012) or increased L-arginine and L-histidine accumulation in cells (Schendzielorz et al. 2014). A very different approach to amino acid sensing was taken by Utsumi et al. (1989), who modified the two-component system EnvZ-OmpR in E. coli by replacing the N-terminal region of EnvZ with the periplasmic sensing domain of the chemoreceptor Tar to create a biosensor that could detect aspartate in the growth media. Metabolites with relevance for medical research or clinical applications are also common biosensor targets. Win and Smolke (2007) created riboswitches to control gene expression in S. cerevisiae in response to two drug compounds theophylline and tetracycline. Sensors for the immediate precursors of serotonin and dopamine neurotransmitters were designed using a similar approach, in this case by coupling aptamers to the fluorophore-binding “Broccoli” RNA sequence in E. coli (Porter et al. 2017). Nitrogen oxides are a biomarker implicated in inflammation and can be detected using a biosensor based on the pYeaR promoter from E. coli (Courbet et al. 2015). Tetrathionate is another inflammation marker in the gut for which a bacterial biosensor was developed, as mentioned above (Riglar et al. 2017). Quorum-sensing molecules are a family of compounds with potentially significant applications for the detection and diagnosis of pathogens. For example, a biosensor for the detection of the acyl homoserine lactone (AHL) 3-oxo-C12-HSL, produced by the opportunistic pathogen Pseudomonas aeruginosa, was built using the LasR transcription factor and coupled to a bacterial killing mechanism (Saeidi et al. 2014; Hwang et al. 2017). As discussed in posttranslational mechanisms for biosensing, proteins with relevance for human health such as human fibrinogen have also been targets for biosensors (Kylilis et al. 2019).

390

K. Y. Wen et al.

Besides chemical compounds or elements, biosensors can be made for the sensing of physical phenomenon. Light sensing has been engineered through various mechanisms: via a fusion of the photoreceptor from Synechocystis phytochrome Cph1 with the EnvZ-OmpR two-component system in E. coli (Levskaya et al. 2005); by importing the bacterial EL222 transcription factor into mammalian cells (MottaMena et al. 2014); and by fusing the photoreceptor VVD domain from filamentous fungus Neurospora crassa to split T7 RNA polymerase fragments in E. coli (Han et al. 2017). Temperature has also been used as an input, such as with the temperature-sensitive λ repressor controlling gene expression in a toggle switch in E. coli (Gardner et al. 2000) and with an engineered eukaryotic transcription factor named cold-inducible transactivator (Weber et al. 2003). The pH of a solution can be determined using biosensors as well, such as with a riboswitch-based system (Nechooshtan et al. 2009). This overview shows that biosensors have been constructed for an enormous variety of targets. In metabolic engineering, real-time monitoring of biosynthetic pathways is necessary to identify best-performing strains and to identify bottlenecks in metabolite turnover. Targets in healthcare applications include biomarkers of disease in the human gut and blood. Environmental conditions, such as the pollution of water by heavy metals, can also be monitored using biosensors. As biosensors are indispensable tools for many synthetic biology applications, it is foreseeable that the diversity of targets for which sensors have been developed will only continue to expand in the coming years.

Output Within biosensors, reporters are used to convert an environmental stimulus into a detectable output. However, the selection of a suitable reporter gene or output can be daunting due to the vast range of options available. In many cases a reporter will only be suitable for specific applications; therefore, it is important to consider viable options within the context of the final biosensor application. The most commonly reported biosensor outputs are fluorescent proteins (examples provided within Table 1). These require the use of specialist equipment that is able to stimulate and record the emission of these proteins. In some cases, such as green fluorescent protein (GFP), they require the presence of oxygen in order to fluoresce (Tsien 1998). This means they are not ideal for biosensors which are designed to work in anaerobic environments; for example, as diagnostics in the mammalian gut (a concept which is discussed further by Ozdemir et al. (2018)). However, fluorescent proteins are an attractive option due to the availability of a huge collection of variants; a review of which is given by Rodriguez et al. (2017). This allows engineers to design biosensors which use ratiometric signals (Shynkar et al. 2007; Ding et al. 2015) or multiplexing of multiple variants to report on diverse input conditions (Lin et al. 2019). Additional outputs that can be used as biosensor reporters are summarized in Table 1 alongside their advantages and disadvantages. These include electrical,

16

Fundamental Building Blocks of Whole-Cell Biosensor Design

391

Table 1 Summary of outputs reported in biosensor designs Output GFP (e.g., sfGFP, EGFP) BFP (e.g., EBFP, mTagBFP) RFP (e.g., mCherry, TurboRFP) lux (bacterial luciferase) luc (firefly luciferase)

Type Fluorescent

Advantages High stability, no substrate needed, many variants available, does not require ATP

Disadvantages May have long maturation times, require oxygen, autofluorescent backgrounds, require excitation

Examples (Cerminati et al. 2015; Jiang et al. 2016; Goers et al. 2017; Lin et al. 2019)

Luminescent

Easy measurement, quick response, high stability

(Burlage et al. 1994; Weitz et al. 2001; Kobras et al. 2017)

crtA xylE lacZ

Colorimetric

Conductometry Potentiometry Amperometry Voltammetry Motility (swimming) Cell agglutination

Electrical

Detectable by eye, high sensitivity, rapid response, low limit of detection Rapid response, high sensitivity

Require substrates (not in the case of luxCDABE) (Kobras et al. 2017), may require oxygen May require substrate, may require cell lysis

Behavioral changes

Detectable by eye, may not require active translation of a reporter protein

Require immobilization

Not easily quantifiable, low sensitivity

(Hansen and Sørensen 2000; Maeda et al. 2006; French et al. 2011b; Zammit et al. 2013) (Jensen et al. 2010; Golitsch et al. 2013) (Topp and Gallivan 2008; Kylilis et al. 2019)

colorimetric, pH-modifying, or physiological reporters. While the majority of these output modules rely on a further external measurement tool to quantify levels of output, physiological reporters are particularly interesting because they do not require the use of specialist equipment and may be performed by any lab with the facilities to culture bacteria. For example, Derr et al. (2006) linked detection of a specific target to an increase in cell motility to develop chemoreceptors with an improved sensitivity to glutamate. As this can easily be visualized by the eye, such outputs may be developed for relatively cheap use in remote environments or “infield” applications. Alongside these basic reporters, it is also possible to integrate additional genetic elements into biosensor design. One example is the phage lambda cl/cro memory element reported by Kotula et al. (2014); this is a bistable switch, which flips from a cl to a Cro state upon exposure to anhydrotetracycline (aTc) and remains in an ON state even after removal of the aTc inducer. Further work showed this memory element could be applied to record tetrathionate exposure events within the

392

K. Y. Wen et al.

mammalian gut of a mouse model (Riglar et al. 2017). Another example is the cascading amplifiers reported by Wan et al. (2019); the authors designed a modular signal amplifier which could be used to drastically change the sensitivity and response curves of arsenic and mercury biosensors. Genetic elements such as these allow for further signal processing within outputs and the tailoring of biosensors for specific applications. In the future, it is likely that the use of biosensors will start to evolve from purely diagnostic or reporting tools toward integration into bioremediation or therapeutic applications. As such, there will be the development of more complex outputs which are able to perform a specific function within a host or environment. The P. aeruginosa-sensing E. coli biosensor discussed above (Saeidi et al. 2014) is an example of this, where the AHL detection element was coupled to the production of lysin E7 (a lysis protein), pyocin S5 (an anti-P. aeruginosa toxin), and dispersin B (an anti-biofilm enzyme) (Hwang et al. 2017). The authors subsequently went on to show that this strain was able to prevent P. aeruginosa infection in both mouse and Caenorhabditis elegans models of infection. This gives an effective example of how whole-cell biosensors may be further engineered to perform therapeutic applications through the integration of multiple complex outputs.

Chassis As with many design aspects, the most suitable choice of chassis for a biosensor depends heavily upon the environment in which it is intended to operate and the function that the biosensor is intended to perform. There is a huge array of possible chassis which can be chosen during biosensor design, including prokaryote (Kotula et al. 2014; Webb et al. 2016; Daeffler et al. 2017; Riglar et al. 2017; Chen et al. 2019), eukaryote (Gutiérrez et al. 2015; Shaw et al. 2019), and minicells (Rampley et al. 2017) (summarized within Fig. 2). Even within a single species, different strains may be more suited to a specific application over others. Therefore, the choice of an appropriate chassis is a vital part of biosensor design, and all options should be evaluated carefully before deciding on a final candidate. The most commonly used chassis for biosensors is the Gram-negative bacteria E. coli. This is due to familiarity and the range of tools developed for manipulating this species (French et al. 2011b). To date there have been numerous E. coli strains involved in the development of biosensors including E. coli Nissle (Daeffler et al. 2017; Chen et al. 2019), MG1655 (Kotula et al. 2014; Holowko et al. 2016), NGF-1 (Ziesack et al. 2018), and members of the Keio knockout collection (derivatives of K12) (Baba et al. 2006), among others. Of special interest is the E. coli Nissle strain which is a commensal, non-pathogenic strain that has been shown to confer numerous health benefits to a human host (Sonnenborn 2016; Secher et al. 2017, 2018). As such, this strain is ideally suited for biosensors which are designed to operate in vivo. To date, this strain has been incorporated into multiple biosensors for detecting factors relevant to human health (Daeffler et al. 2017; Chen et al. 2019). Other E. coli biosensors include sensors for detecting lactate (Goers et al. 2017), gold (Zammit

16

Fundamental Building Blocks of Whole-Cell Biosensor Design

393

Fig. 2 Summary of advantages and disadvantages of common microbial chassis options for biosensor design

et al. 2013), arsenic (Wan et al. 2019; Jia et al. 2019), and mercury (Cai et al. 2018). However, despite the popularity of E. coli in general, there are a number of limitations which affect its suitability toward certain applications. For example, E. coli does not have a dormant state and requires additional processes such as freeze-drying to be suitable for long-term storage and distribution (Pedahzur et al. 2004). This adds additional layers of complexity to their field deployment to remote areas. Also, some strains may secrete endotoxins which can severely limit their in vivo use (French et al. 2011b). Alongside E. coli, B. subtilis is another popular chassis option for biosensors. This species is the model Gram-positive bacteria and is generally regarded as safe (Webb et al. 2016). B. subtilis is able to create endospores (Errington 2003; Date et al. 2007), a process that occurs when the cells encounter unfavorable conditions and can potentially facilitate transport outside of the laboratory environment. In addition, they are able to readily take up linear portions of DNA and incorporate them into their genome through homologous recombination (French et al. 2011b). This means it is an easier process to integrate genetically encoded biosensors into the chromosome than in E. coli. To date, examples of B. subtilis biosensors include a sensor used in the detection of antimicrobial compounds that disrupt the cell wall (Kobras et al. 2017) and a strain for detecting parasitic schistosome cercariae infection (Webb et al. 2016). Other reported bacterial chassis include Lactobacillus reuteri (Lubkowicz et al. 2018) and Pseudomonas putida (Fernández et al. 2016), among others. Eukaryotic cells may offer advantages over bacterial chassis such as the ability to host eukaryotic sensing modalities (such as G-protein coupled receptors) and in certain cases the ability to tolerate harsh conditions (Adeniran et al. 2014). Within eukaryote chassis, yeast is a prominent example – particularly the model organism S. cerevisiae (Akyilmaz et al. 2006; Cevenini et al. 2018; Baumann et al. 2018). Other eukaryote chassis include microalgae (Wong et al. 2018) and protozoa (Amaro et al. 2014). It has recently been proposed that “minicells,” small chromosome-free cells formed during aberrant bacterial cell division (Farley et al. 2016), are a viable

394

K. Y. Wen et al.

alternative to the previous chassis discussed. As they lack native host genes and the ability to reproduce, they may offer a platform for synthetic biology applications that bypasses concerns associated with genetically modified organisms (GMOs). One example of minicells is the SimCell chassis reported by Rampley et al. (2017). The authors derived chromosome-free minicells from the abnormal cell division of the E. coli MC1000 ΔminD strain and demonstrated that plasmid-encoded biosensors for glucarate, acrylate, and arabinose could function in these minicells. Chen et al. went on to develop an aspirin-inducible biosensor in SimCells, which showed low levels of background expression and induction below clinically safe concentrations of aspirin. Thus the authors propose that SimCells could be a useful system for therapeutics delivery due to their small size and inability to reproduce (Chen et al. 2019). These examples are by no means an exhaustive list of all the possible chassis which can be chosen as part of biosensor design. However, they serve to highlight that there is not one key type of chassis which is superior to all others and that the advantages and disadvantages of each type must be considered within the specific context of the proposed biosensor.

Optimizing Biosensor Performance For a whole-cell biosensor to be used in a specific application, its performance must first be evaluated and compared to the requirements of the intended environment of interest. There are several key features that should be determined: (1) linear range; (2) sensitivity; (3) specificity; (4) dynamic range; (5) threshold; and (6) lower limit of detection (Fig. 3). Although there are various definitions for these terms, a summary of terms is provided in Table 2 to clarify terms used in this work. It is likely that a biosensor assembled from native biological parts will not immediately behave in the desired way for a specific application. For example, a whole-cell biosensor may be activated at a different concentration of inducer than the anticipated environmental concentration in test conditions. As such, there are numerous approaches that can be taken to modify the design of biosensor parts and tune their behavior. This section will discuss how the construction of circuit parts can be optimized to improve response, the use of modelling to identify the optimal circuit design, how sensing mechanisms can be modified by methods such as directed evolution, and finally how the overall system can be designed in order to suit the intended environment.

Tuning Design Parameters A whole-cell biosensor is typically made up of multiple components, including genetic control elements that determine the level of expression for signal transduction and reporter genes. These can be used as dials to tune the overall behavior of the biosensor. A good review of modifying genetic design to tune the behavior of

16

Fundamental Building Blocks of Whole-Cell Biosensor Design

395

Fig. 3 A typical biosensor response curve, annotated with common parameters used to evaluate biosensor performance Table 2 Definitions of key features of biosensor performance Feature Linear range Sensitivity Specificitya Dynamic range Threshold (K1/2) Lower limit of detection

Definition Concentration range of inducer over which the biosensor can produce a concentration-dependent output. Also sometimes referred to as dose response Gradient of the response curve over the linear range Response of the biosensor to its target analyte versus its response to structurally similar target compounds Maximum increase in output expression divided by the basal (uninduced) output level Concentration of inducer that activates 50% of maximum output Concentration at which 95% of samples would have a measured signal above background level (Armbruster and Pry 2008)

a

Differs to the use of the terms for medical diagnostics, which are based on an evaluation of a test’s ability to distinguish true positives and true negatives

cellular systems is given in Ang et al. (2013). Here, a few examples of how wholecell biosensors have been improved using these methods will be provided. It has been shown that the promoter strength of the transcriptional regulator can impact biosensor output. Wang et al. (2015) decreased the expression of either the TetR repressor or LuxR activator in E. coli genetic circuits by placing them under the control of constitutive promoters of varying strength or under an inducible promoter. They found that reducing the strength of the promoter for TetR increased the sensitivity and dynamic range of the response to the inducer aTc, and the reverse was true in the case of the activator LuxR and its cognate inducer 3OC6-HSL. When this was applied to an arsenic biosensor based on the ArsR repressor, lower

396

K. Y. Wen et al.

expression of the transcriptional regulator did indeed lead to increased sensitivity and dynamic range (Wan et al. 2019). Interestingly, when this was applied to MerR, which is both a repressor in the absence of inorganic mercury and an activator when bound to mercury, it was found that lower promoter strength also improved detection limit and dynamic range. Besides modifying the promoter controlling the signal transducer, it is also possible to modify the promoter regulating expression of the output genes. Trantidou et al. created a whole-cell E. coli lactate biosensor based on the transcriptional elements of the lldPRD operon, where a modification in the lldPRD promoter resulted in increased output of the GFP reporter in response to lactate (Trantidou et al. 2018). Merulla and Van der Meer (2016) introduced additional operator sites in the Pars promoter regulated by ArsR, which created a transcriptional roadblock that could be relieved by addition of arsenite inducer. This lead to lower background expression and improved dynamic range, and could be tuned by changing the distance between the operator sites (Merulla and van der Meer 2016; Wan et al. 2019). Another method for engineering bacterial regulated promoters was demonstrated by modifying the 10 and 35 sites in the promoter sequence based on predicted binding energies to create AraC- and LasR-regulated promoters with varied dynamic ranges (Chen et al. 2018). The authors used this to further develop hybrid promoters regulated by two or three ligands, but this approach would also be useful for improving transcriptionally regulated biosensors. The integration of feedback loops into biosensor circuits can optimize biosensor performance. Positive feedback loops have shown enhanced sensitivity and higher output signals (Cai et al. 2018; Jia et al. 2018). A whole-cell biosensor for lead detection was improved by reconfiguring the regulatory elements and incorporating positive feedback loops to the circuits, where the output signal was 1.5–2 times stronger when the positive feedback loop was present (Jia et al. 2018). The addition of a positive feedback loop in a mercury biosensor led to improved sensitivity and fluorescence intensity (Cai et al. 2018). In another approach, a feedback control architecture was proposed to improve orthogonality in two-component systems (Steel et al. 2019). This architecture, which involves the addition of a substrate that acts as phosphate sink, could open the way to more nuanced approaches to incorporating feedback into biosensor systems. It can be difficult to determine experimentally which element in a genetically encoded biosensor should be changed to obtain the desired behavior due to the large number of parts and the possibility that changing one part will affect another part of the system in non-intuitive ways. This has been studied mathematically for the case of transcription factor-based biosensors, where relationships – or constraints – between threshold and dynamic range were found to hold (Mannan et al. 2017). By building a mathematical model of the system, the biosensor designer can identify a priori which parameter values will give the desired sensor behavior and then implement this experimentally. For example, a biosensor for the Vibrio cholerae signalling molecule CAI-I was constructed using a three-part sensing module: a transmembrane receptor CqsS, a phosphorelay protein LuxU, and a transcription factor LuxO (Holowko et al. 2016). In order to identify the optimum design, a model

16

Fundamental Building Blocks of Whole-Cell Biosensor Design

397

was constructed of the system and sensitivity analysis performed which identified that the promoter regulating LuxO expression was a critical factor.

Enhancing Sensing Mechanisms A range of methods have been developed to optimize existing sensing mechanisms (Salis et al. 2009; Ellefson et al. 2018; Meyer et al. 2019). Optimizing performance generally requires screens to be performed in the presence of an inducer to select for maximal activation and in the absence of an inducer to select against leakiness (Yokobayashi et al. 2002). Selection methods are often carried out by sorting cells based on low and high fluorescence, using separate positive and negative selections (Taylor et al. 2016; Meyer et al. 2019) or by using selection markers that can be either toxic or selective depending on the environment (such as the tetracycline/H+ antiporter TetA (Muranaka et al. 2009) or herpes simplex virus thymidine kinase (Tashiro et al. 2011)). Directed evolution is a powerful approach for artificially improving biosensor properties in a short period of time. First, random mutagenesis using techniques such as error prone PCR can be performed to create libraries of sensor variants. Several commercially available kits are available to undertake this approach, as well as a more simplistic approach of using mutagenic agents such as UV. Next, the variants with the desired function should be selected followed by amplification to generate the template for the following round. Meyer et al. (2019) developed a directed evolution strategy to simultaneously select for lower background, high dynamic range, improved sensitivity, and low crosstalk. This enabled them to generate a collection of biosensors that exhibited >100-fold induction with low background and high specificity. Rational design is another approach to optimize existing sensor mechanisms, which can be used when information about the structure of sensor proteins is available. A recent study developed a method for rewiring TCSs by swapping DNA-binding domains and demonstrated that the two largest families of RRs are highly modular and can be rewired to construct “new” functional TCSs that can be used as biosensors (Schmidl et al. 2019). Another example is the model-guided design of a B. subtilis nitrate sensor composed of the two-component system NarX/ NarL-YdfL (Landry et al. 2018). By building a mathematical model of the twocomponent system, the authors identified that changing the phosphatase activity of the sensor kinase could decrease the detection threshold. This was validated experimentally and applied to the detection of fertilizer levels in soil. It was also demonstrated that this method could be used to improve previously developed TCS sensors for aspartate, thiosulfate, and tetrathionate. This study was dependent on known mutations in the sensor proteins to modify their function however, and such information about the sequence-function relationship may not always be available to guide rational design. A combination of methods including computational protein design, single-amino acid saturation mutagenesis, and error-prone PCR was used to allow the redesign of

398

K. Y. Wen et al.

bacterial allosteric TFs to respond to new molecules (Taylor et al. 2016). This study resulted in the E. coli lac repressor, LacI, responding to one of four new inducers – fructose, gentiobiose, lactitol, and sucralose – demonstrating that combining multiple approaches can be synergistic in optimizing existing sensing mechanisms.

Environmental Compatibility Biosensing applications have been hindered partly because of regulatory concerns with the use of GMOs in the field. A variety of features have been proposed to ensure suitability of whole-cell biosensors for their intended operational environment. Alternative selection markers and genetic circuits are the most common strategies for introducing biocontainment measures to biosensors. In a research environment, biosensors constructed as synthetic circuits are generally maintained in a host bacterium on plasmids with antibiotic resistance as a selection tool. If a biosensor was to be used in a clinical application, however, this would not be suitable as they could spread antibiotic resistance by horizontal gene transfer of resistance genes, and the use of antibiotics would lead to the disruption of normal microflora. Alternative selection systems have been successfully implemented in other synthetic circuits such as auxotrophy, toxin-antitoxin systems, bacteriocins, and active partitioning mechanisms (Velur Selvamani et al. 2014; Danino et al. 2015; Fedorec et al. 2019). GeneGuard, a modular plasmid system designed for biosafety, which utilized auxotroph and toxin-antitoxin systems and did not contain antibiotic resistance cassettes, was demonstrated to be an efficient vector for heavy metal biosensors (Wright et al. 2015). Cellular circuits that can lead to cell death when activated, known as kill switches, can also be implemented into biosensor design. Two-layered kill switches such as the Deadman switch and Passcode switch are useful for many biosensor applications (Lee et al. 2018). Next-generation biocontainment systems for engineered organisms such as xenobiology are reviewed further in Lee et al. (2018). The application of a whole-cell biosensor is an important consideration with regard to sensor design and its environmental compatibility. For example, Trantidou et al. (2018) embedded a whole-cell E. coli lactate biosensor in a vesicle-based cell mimic to monitor lactate levels in the external environment. Shielding a whole-cell biosensor from its external environment could be critical in applications such as cellbased therapies, where encapsulation could protect the cell from the host’s immune response. Even in settings where the biosensors are used for in vitro monitoring, bacterial whole cells can be encapsulated in hydrogel beads to create easy-to-use diagnostic tools (Courbet et al. 2015) or freeze-dried to improve storage and transport requirements (Pedahzur et al. 2004; Webb et al. 2016).

Conclusions and Future Directions This chapter has described the fundamental building blocks and design choices that underlie the current construction process of whole-cell biosensor systems. As sensing applications become increasingly complex, single channel readout may be

16

Fundamental Building Blocks of Whole-Cell Biosensor Design

399

replaced with sensors that have multiple inputs and more complex information processing capabilities. While this can be achieved with genetic logic gates (Nielsen et al. 2016) or by leveraging other aspects of biological computing (Prindle et al. 2012; Dalchau et al. 2018), engineered consortia offer a powerful alternative, where information is integrated and processed in a distributed fashion (Regot et al. 2011; Tamsir et al. 2011). Despite the efforts in synthetic biology toward the development of whole-cell biosensors in the past decade and a half, there are limited cases where they have made it into real-world applications. As methods for containment improve and the regulatory landscape adapts, it is envisaged that the engineering approaches described here will be further developed and these useful sensing systems will become widely adopted.

References Adeniran A, Sherer M, Tyo KEJ (2014) Yeast-based biosensors: design and applications. FEMS Yeast Res. https://doi.org/10.1111/1567-1364.12203 Akyilmaz E, Yaşa İ, Dinçkaya E (2006) Whole cell immobilized amperometric biosensor based on Saccharomyces cerevisiae for selective determination of vitamin B1 (thiamine). Anal Biochem 354:78–84. https://doi.org/10.1016/j.ab.2006.04.019 Aleksic J, Bizzari F, Cai Y et al (2007) Development of a novel biosensor for the detection of arsenic in drinking water. Synth Biol IET 1:87–90. https://doi.org/10.1049/iet-stb:20060002 Amaro F, Turkewitz AP, Martín-González A, Gutiérrez JC (2014) Functional GFP-metallothionein fusion protein from Tetrahymena thermophila: a potential whole-cell biosensor for monitoring heavy metal pollution and a cell model to study metallothionein overproduction effects. Biometals 27:195–205. https://doi.org/10.1007/s10534-014-9704-0 Ameen S, Ahmad M, Mohsin M et al (2016) Designing, construction and characterization of genetically encoded FRET-based nanosensor for real time monitoring of lysine flux in living cells. J Nanobiotechnology 14:49. https://doi.org/10.1186/s12951-016-0204-y Ang J, Harris E, Hussey BJ et al (2013) Tuning response curves for synthetic biology. ACS Synth Biol 2:547–567. https://doi.org/10.1021/sb4000564 Armbruster DA, Pry T (2008) Limit of blank, limit of detection and limit of quantitation. Clin Biochem Rev 29(Suppl 1):S49–S52 Baba T, Ara T, Hasegawa M, et al (2006) Construction of Escherichia coli K-12 in-frame, single-gene knockout mutants: the Keio collection. Mol Syst Biol 2. https://doi.org/10.1038/ msb4100050 Baumann L, Rajkumar AS, Morrissey JP et al (2018) A yeast-based biosensor for screening of short- and medium-chain fatty acid production. ACS Synth Biol 7:2640–2646. https://doi.org/ 10.1021/acssynbio.8b00309 Binder S, Schendzielorz G, Stäbler N et al (2012) A high-throughput approach to identify genomic variants of bacterial metabolite producers at the single-cell level. Genome Biol 13:R40. https://doi.org/10.1186/gb-2012-13-5-r40 Burlage RS, Palumbo AV, Heitzer A, Sayler G (1994) Bioluminescent reporter bacteria detect contaminants in soil samples. Appl Biochem Biotechnol 45–46:731–740. https://doi.org/ 10.1007/BF02941845 Cai S, Shen Y, Zou Y et al (2018) Engineering highly sensitive whole-cell mercury biosensors based on positive feedback loops from quorum-sensing systems. Analyst 143:630–634. https://doi. org/10.1039/C7AN00587C Cerminati S, Soncini FC, Checa SK (2015) A sensitive whole-cell biosensor for the simultaneous detection of a broad-spectrum of toxic heavy metal ions. Chem Commun 51:5917–5920. https://doi.org/10.1039/C5CC00981B

400

K. Y. Wen et al.

Cevenini L, Lopreside A, Calabretta MM et al (2018) A novel bioluminescent NanoLuc yeastestrogen screen biosensor (nanoYES) with a compact wireless camera for effect-based detection of endocrine-disrupting chemicals. Anal Bioanal Chem 410:1237–1246. https://doi.org/ 10.1007/s00216-017-0661-7 Chen Y, Ho JML, Shis DL et al (2018) Tuning the dynamic range of bacterial promoters regulated by ligand-inducible transcription factors. Nat Commun 9:64. https://doi.org/10.1038/s41467017-02473-5 Chen JX, Steel H, Wu Y-H, et al (2019) Development of aspirin-inducible biosensors in Escherichia coli and SimCells. Appl Environ Microbiol 85. https://doi.org/10.1128/AEM.02959-18 Courbet A, Endy D, Renard E et al (2015) Detection of pathological biomarkers in human clinical samples via amplifying genetic switches and logic gates. Sci Transl Med 7:289ra83–289ra83. https://doi.org/10.1126/scitranslmed.aaa3601 Curtis C, Lima A, Lozano S, Veith G (1982) Evaluation of a bacterial bioluminescence bioassay as a method for predicting acute toxicity of organic chemicals to fish. In: Aquatic toxicology and hazard assessment. ASTM International, West Conshohocken, pp 170–178 Daeffler KN, Galley JD, Sheth RU et al (2017) Engineering bacterial thiosulfate and tetrathionate sensors for detecting gut inflammation. Mol Syst Biol 13:923. https://doi.org/10.15252/ msb.20167416 Dalchau N, Szép G, Hernansaiz-Ballesteros R et al (2018) Computing with biological switches and clocks. Nat Comput 17:761–779. https://doi.org/10.1007/s11047-018-9686-x Danino T, Prindle A, Kwong GA et al (2015) Programmable probiotics for detection of cancer in urine. Sci Transl Med 7:289ra84–289ra84. https://doi.org/10.1126/scitranslmed.aaa3519 Date A, Pasini P, Daunert S (2007) Construction of spores for portable bacterial whole-cell biosensing systems. Anal Chem 79:9391–9397. https://doi.org/10.1021/ac701606g De Zwart D, Slooff W (1983) The Microtox as an alternative assay in the acute toxicity assessment of water pollutants. Aquat Toxicol 4:129–138. https://doi.org/10.1016/0166445X(83)90050-4 Dekker L, Polizzi KM (2017) Sense and sensitivity in bioprocessing – detecting cellular metabolites with biosensors. Curr Opin Chem Biol 40:31–36. https://doi.org/10.1016/j.cbpa.2017.05.014 Derr P, Boder E, Goulian M (2006) Changing the specificity of a bacterial chemoreceptor. J Mol Biol 355:923–932. https://doi.org/10.1016/j.jmb.2005.11.025 Ding Y, Li J, Enterina JR et al (2015) Ratiometric biosensors based on dimerization-dependent fluorescent protein exchange. Nat Methods 12:195–198. https://doi.org/10.1038/nmeth.3261 Ellefson JW, Ledbetter MP, Ellington AD (2018) Directed evolution of a synthetic phylogeny of programmable Trp repressors. Nat Chem Biol 14:361–367. https://doi.org/10.1038/s41589-0180006-7 Errington J (2003) Regulation of endospore formation in Bacillus subtilis. Nat Rev Microbiol 1:117–126. https://doi.org/10.1038/nrmicro750 Espah Borujeni A, Mishler DM, Wang J et al (2016) Automated physics-based design of synthetic riboswitches from diverse RNA aptamers. Nucleic Acids Res 44:1–13. https://doi.org/10.1093/ nar/gkv1289 Farley MM, Hu B, Margolin W, Liu J (2016) Minicells, back in fashion. J Bacteriol 198:1186–1195. https://doi.org/10.1128/JB.00901-15 Fedorec AJH, Ozdemir T, Doshi A et al (2019) Two new plasmid post-segregational killing mechanisms for the implementation of synthetic gene networks in Escherichia coli. iScience 14:323–334. https://doi.org/10.1016/j.isci.2019.03.019 Fehr M, Frommer WB, Lalonde S (2002) Visualization of maltose uptake in living yeast cells by fluorescent nanosensors. Proc Natl Acad Sci 99:9846–9851. https://doi.org/10.1073/ pnas.142089199 Fernández M, Morel B, Ramos JL, Krell T (2016) Paralogous regulators ArsR1 and ArsR2 of Pseudomonas putida KT2440 as a basis for arsenic biosensor development. Appl Environ Microbiol 82:4133–4144. https://doi.org/10.1128/AEM.00606-16 French CE, de Mora K, Joshi N et al (2011a) Synthetic biology and the art of biosensor design. In: The science and applications of synthetic and systems biology: workshop summary. National Academies Press, Washington, DC

16

Fundamental Building Blocks of Whole-Cell Biosensor Design

401

French CE, de Mora K, Joshi N et al (2011b) Synthetic biology and the art of biosensor design. In: Institute of Medicine (US) forum on microbial threats. National Academies Press (US), Washington, DC Fritz G, Koller C, Burdack K et al (2009) Induction kinetics of a conditional pH stress response system in Escherichia coli. J Mol Biol 393:272–286. https://doi.org/10.1016/J. JMB.2009.08.037 Gardner TS, Cantor CR, Collins JJ (2000) Construction of a genetic toggle switch in Escherichia coli. Nature 403:339 Goers L, Kylilis N, Tomazou M et al (2013) Engineering microbial biosensors. Methods Microbiol 40:119–156. https://doi.org/10.1016/B978-0-12-417029-2.00005-4 Goers L, Ainsworth C, Goey CH et al (2017) Whole-cell Escherichia coli lactate biosensor for monitoring mammalian cell cultures during biopharmaceutical production. Biotechnol Bioeng 114:1290–1300. https://doi.org/10.1002/bit.26254 Golitsch F, Bücking C, Gescher J (2013) Proof of principle for an engineered microbial biosensor based on Shewanella oneidensis outer membrane protein complexes. Biosens Bioelectron 47:285–291. https://doi.org/10.1016/j.bios.2013.03.010 Green AA, Silver PA, Collins JJ, Yin P (2014) Toehold switches: De-novo-designed regulators of gene expression. Cell 159:925–939. https://doi.org/10.1016/j.cell.2014.10.002 Gutiérrez JC, Amaro F, Martín-González A (2015) Heavy metal whole-cell biosensors using eukaryotic microorganisms: an updated critical review. Front Microbiol 6. https://doi.org/ 10.3389/fmicb.2015.00048 Han T, Chen Q, Liu H (2017) Engineered photoactivatable genetic switches based on the bacterium phage T7 RNA polymerase. ACS Synth Biol 6:357–366. https://doi.org/10.1021/ acssynbio.6b00248 Hansen LH, Sørensen SJ (2000) Detection and quantification of tetracyclines by whole cell biosensors. FEMS Microbiol Lett 190:273–278. https://doi.org/10.1111/j.1574-6968.2000.tb09298.x Helmann JD (2002) The extracytoplasmic function (ECF) sigma factors. Adv Microb Physiol 46:47–110 Holowko MB, Wang H, Jayaraman P, Poh CL (2016) Biosensing vibrio cholerae with genetically engineered Escherichia coli. ACS Synth Biol 5:1275–1283. https://doi.org/10.1021/ acssynbio.6b00079 Hwang IY, Koh E, Wong A et al (2017) Engineered probiotic Escherichia coli can eliminate and prevent Pseudomonas aeruginosa gut infection in animal models. Nat Commun 8:15028. https://doi.org/10.1038/ncomms15028 Jensen HM, Albers AE, Malley KR et al (2010) Engineering of a synthetic electron conduit in living cells. Proc Natl Acad Sci 107:19213–19218. https://doi.org/10.1073/pnas.1009645107 Jia X, Zhao T, Liu Y, et al (2018) Gene circuit engineering to improve the performance of a whole-cell lead biosensor. FEMS Microbiol Lett 365. https://doi.org/10.1093/femsle/fny157 Jia X, Bu R, Zhao T, Wu K (2019) Development of a sensitive and specific whole-cell biosensor for arsenic detection. Appl Environ Microbiol. https://doi.org/10.1128/AEM.00694-19 Jiang H, Jiang D, Shao J et al (2016) High-throughput living cell-based optical biosensor for detection of bacterial lipopolysaccharide (LPS) using a red fluorescent protein reporter system. Sci Rep 6:36987. https://doi.org/10.1038/srep36987 Joshi N, Wang X, Montgomery L et al (2009) Novel approaches to biosensors for detection of arsenic in drinking water. Desalination 248:517–523. https://doi.org/10.1016/J. DESAL.2008.05.096 Jung K, Fabiani F, Hoyer E, Lassak J (2018) Bacterial transmembrane signalling systems and their engineering for biosensing. Open Biol 8:180023. https://doi.org/10.1098/rsob.180023 Khalil AS, Collins JJ (2010) Synthetic biology: applications come of age. Nat Rev Genet 11:367–379. https://doi.org/10.1038/nrg2775 Kobras CM, Mascher T, Gebhard S (2017) Application of a bacillus subtilis whole-cell biosensor (PliaI-lux) for the identification of cell wall active antibacterial compounds. Methods Mol Biol 1520:121–131 Kokkali V, van Delft W (2014) Overview of commercially available bioassays for assessing chemical toxicity in aqueous samples. TrAC Trends Anal Chem 61:133–155. https://doi.org/ 10.1016/j.trac.2014.08.001

402

K. Y. Wen et al.

Komatsu N, Terai K, Imanishi A et al (2018) A platform of BRET-FRET hybrid biosensors for optogenetics, chemical screening, and in vivo imaging. Sci Rep 8:8984. https://doi.org/10.1038/ s41598-018-27174-x Kotula JW, Kerns SJ, Shaket LA et al (2014) Programmable bacteria detect and record an environmental signal in the mammalian gut. Proc Natl Acad Sci 111:4838–4843. https://doi. org/10.1073/pnas.1321321111 Kylilis N, Riangrungroj P, Lai H-E et al (2019) Whole-cell biosensor with tunable limit of detection enables low-cost agglutination assays for medical diagnostic applications. ACS Sens 4:370–378. https://doi.org/10.1021/acssensors.8b01163 Landry BP, Palanki R, Dyulgyarov N et al (2018) Phosphatase activity tunes two-component system sensor detection threshold. Nat Commun 9:1433. https://doi.org/10.1038/s41467-01803929-y Leavitt JM, Tong A, Tong J et al (2016) Coordinated transcription factor and promoter engineering to establish strong expression elements in Saccharomyces cerevisiae. Biotechnol J 11:866–876. https://doi.org/10.1002/biot.201600029 Leavitt JM, Wagner JM, Tu CC et al (2017) Biosensor-enabled directed evolution to improve Muconic acid production in Saccharomyces cerevisiae. Biotechnol J 12:1600687. https://doi. org/10.1002/biot.201600687 Lee JW, Chan CTY, Slomovic S, Collins JJ (2018) Next-generation biocontainment systems for engineered organisms. Nat Chem Biol 14:530–537. https://doi.org/10.1038/s41589-018-0056-x Levskaya A, Chevalier AA, Tabor JJ et al (2005) Engineering Escherichia coli to see light. Nature 438:441–442. https://doi.org/10.1038/nature04405 Lewis M, Chang G, Horton NC et al (1996) Crystal structure of the lactose operon repressor and its complexes with DNA and inducer. Science 271:1247–1254. https://doi.org/10.1126/ science.271.5253.1247 Lin C, Zhang Q-X, Yeh Y-C (2019) Development of a whole-cell biosensor for the determination of tyrosine in urine for point-of-care diagnostics. Anal Methods 11:1400–1404. https://doi.org/ 10.1039/C9AY00070D Lindenburg L, Merkx M (2014) Engineering genetically encoded FRET sensors. Sensors 14:11691–11713. https://doi.org/10.3390/s140711691 Liu Y, Liu Y, Wang M (2017) Design, optimization and application of small molecule biosensor in metabolic engineering. Front Microbiol 8:2012. https://doi.org/10.3389/fmicb.2017.02012 Lubkowicz D, Ho CL, Hwang IY et al (2018) Reprogramming probiotic Lactobacillus reuteri as a biosensor for Staphylococcus aureus derived AIP-I detection. ACS Synth Biol 7:1229–1237. https://doi.org/10.1021/acssynbio.8b00063 Maeda I, Yamashiro H, Yoshioka D et al (2006) Colorimetric dimethyl sulfide sensor using Rhodovulum sulfidophilum cells based on intrinsic pigment conversion by CrtA. Appl Microbiol Biotechnol 70:397–402. https://doi.org/10.1007/s00253-005-0117-4 Mannan AA, Liu D, Zhang F, Oyarzún DA (2017) Fundamental design principles for transcriptionfactor-based metabolite biosensors. ACS Synth Biol 6:1851–1859. https://doi.org/10.1021/ acssynbio.7b00172 Mascher T (2013) Signaling diversity and evolution of extracytoplasmic function (ECF) σ factors. Curr Opin Microbiol 16:148–155 Merulla D, van der Meer JR (2016) Regulatable and modulable background expression control in prokaryotic synthetic circuits by auxiliary repressor binding sites. ACS Synth Biol 5:36–45. https://doi.org/10.1021/acssynbio.5b00111 Meyer AJ, Segall-Shapiro TH, Glassey E et al (2019) Escherichia coli “Marionette” strains with 12 highly optimized small-molecule sensors. Nat Chem Biol 15:196–204. https://doi.org/10.1038/ s41589-018-0168-3 Motta-Mena LB, Reade A, Mallory MJ et al (2014) An optogenetic gene expression system with rapid activation and deactivation kinetics. Nat Chem Biol 10:196–202. https://doi.org/10.1038/ nchembio.1430

16

Fundamental Building Blocks of Whole-Cell Biosensor Design

403

Munch R, Hiller K, Barg H et al (2003) PRODORIC: prokaryotic database of gene regulation. Nucleic Acids Res 31:266–269. https://doi.org/10.1093/nar/gkg037 Muranaka N, Sharma V, Nomura Y, Yokobayashi Y (2009) An efficient platform for genetic selection and screening of gene switches in Escherichia coli. Nucleic Acids Res 37:e39–e39. https://doi.org/10.1093/nar/gkp039 Mustafi N, Grünberger A, Kohlheyer D et al (2012) The development and application of a single-cell biosensor for the detection of l-methionine and branched-chain amino acids. Metab Eng 14:449–457. https://doi.org/10.1016/j.ymben.2012.02.002 Nechooshtan G, Elgrably-Weiss M, Sheaffer A et al (2009) A pH-responsive riboregulator. Genes Dev 23:2650–2662. https://doi.org/10.1101/gad.552209 Nielsen AAK, Der BS, Shin J et al (2016) Genetic circuit design automation. Science 352:aac7341 Novichkov PS, Kazakov AE, Ravcheev DA et al (2013) RegPrecise 3.0–a resource for genomescale exploration of transcriptional regulation in bacteria. BMC Genomics 14:745. https://doi. org/10.1186/1471-2164-14-745 Ozdemir T, Fedorec AJH, Danino T, Barnes CP (2018) Synthetic biology and engineered live biotherapeutics: toward increasing system complexity. Cell Syst 7:5–16. https://doi.org/ 10.1016/j.cels.2018.06.008 Pardee K, Green AA, Ferrante T et al (2014) Paper-based synthetic gene networks. Cell 159:940–954. https://doi.org/10.1016/j.cell.2014.10.004 Pardee K, Green AA, Takahashi MK et al (2016) Rapid, low-cost detection of zika virus using programmable biomolecular components. Cell 165:1255–1266. https://doi.org/10.1016/J. CELL.2016.04.059 Pedahzur R, Rosen R, Belkin S (2004) Stabilization of recombinant bioluminescent bacteria for biosensor applications. Cell Preserv Technol 2:260–269. https://doi.org/10.1089/cpt.2004.2.260 Porter EB, Polaski JT, Morck MM, Batey RT (2017) Recurrent RNA motifs as scaffolds for genetically encodable small-molecule biosensors. Nat Chem Biol 13:295–301. https://doi.org/ 10.1038/nchembio.2278 Prindle A, Samayoa P, Razinkov I et al (2012) A sensing array of radically coupled genetic “biopixels”. Nature 481:39–44. https://doi.org/10.1038/nature10722 Rampley CPN, Davison PA, Qian P et al (2017) Development of SimCells as a novel chassis for functional biosensors. Sci Rep 7:7261. https://doi.org/10.1038/s41598-017-07391-6 Regot S, MacIa J, Conde N et al (2011) Distributed biological computation with multicellular engineered networks. Nature 469:207–211. https://doi.org/10.1038/nature09679 Rhodius VA, Segall-Shapiro TH, Sharon BD et al (2013) Design of orthogonal genetic switches based on a crosstalk map of σs, anti-σs, and promoters. Mol Syst Biol 9:702–702. https://doi. org/10.1038/msb.2013.58 Riglar DT, Giessen TW, Baym M et al (2017) Engineered bacteria can function in the mammalian gut long-term as live diagnostics of inflammation. Nat Biotechnol 35:653–658. https://doi.org/ 10.1038/nbt.3879 Rodriguez EA, Campbell RE, Lin JY et al (2017) The growing and glowing toolbox of fluorescent and photoactive proteins. Trends Biochem Sci 42:111–129. https://doi.org/10.1016/j.tibs.2016.09.010 Saeidi N, Wong CK, Lo T-M et al (2014) Engineering microbes to sense and eradicate Pseudomonas aeruginosa, a human pathogen. Mol Syst Biol 7:521–521. https://doi.org/10.1038/ msb.2011.55 Salis H, Tamsir A, Voigt C (2009) Engineering bacterial signals and sensors. In: Bacterial sensing and signaling. KARGER, Basel, pp 194–225 Schendzielorz G, Dippong M, Grünberger A et al (2014) Taking control over control: use of product sensing in single cells to remove flux control at key enzymes in biosynthesis pathways. ACS Synth Biol 3:21–29. https://doi.org/10.1021/sb400059y Schmidl SR, Ekness F, Sofjan K, et al (2019) Rewiring bacterial two-component systems by modular DNA-binding domain swapping. Nat Chem Biol 1. https://doi.org/10.1038/s41589019-0286-6

404

K. Y. Wen et al.

Secher T, Kassem S, Benamar M et al (2017) Oral administration of the probiotic strain Escherichia coli Nissle 1917 reduces susceptibility to neuroinflammation and repairs experimental autoimmune encephalomyelitis-induced intestinal barrier dysfunction. Front Immunol 8:1096. https://doi.org/10.3389/fimmu.2017.01096 Secher T, Maillet I, Mackowiak C et al (2018) The probiotic strain Escherichia coli Nissle 1917 prevents papain-induced respiratory barrier injury and severe allergic inflammation in mice. Sci Rep 8:11245. https://doi.org/10.1038/s41598-018-29689-9 Shaw WM, Yamauchi H, Mead J et al (2019) Engineering a model cell for rational tuning of GPCR signaling. Cell 177(3):782–796.e27. https://doi.org/10.1016/j.cell.2019.02.023 Shetty RS, Deo SK, Liu Y, Daunert S (2004) Fluorescence-based sensing system for copper using genetically engineered living yeast cells. Biotechnol Bioeng 88:664–670. https://doi.org/ 10.1002/bit.20331 Shynkar VV, Klymchenko AS, Kunzelmann C et al (2007) Fluorescent biomembrane probe for ratiometric detection of apoptosis. J Am Chem Soc 129:2187–2193. https://doi.org/10.1021/ ja068008h Sonnenborn U (2016) Escherichia coli strain Nissle 1917—from bench to bedside and back: history of a special Escherichia coli strain with probiotic properties. FEMS Microbiol Lett 363:fnw212. https://doi.org/10.1093/femsle/fnw212 Sourjik V, Berg HC (2002) Receptor sensitivity in bacterial chemotaxis. Proc Natl Acad Sci 99:123–127. https://doi.org/10.1073/pnas.011589998 Steel H, Sootla A, Smart B et al (2019) Improving orthogonality in two-component biological signalling systems using feedback control. IEEE Control Syst Lett 3:326–331. https://doi.org/ 10.1109/LCSYS.2018.2871663 Takahashi MK, Tan X, Dy AJ et al (2018) A low-cost paper-based synthetic biology platform for analyzing gut microbiota and host biomarkers. Nat Commun 9:3347. https://doi.org/10.1038/ s41467-018-05864-4 Tamsir A, Tabor JJ, Voigt CA (2011) Robust multicellular computing using genetically encoded NOR gates and chemical ‘wires’. Nature 469:212–215. https://doi.org/10.1038/nature09565 Tang S-Y, Cirino PC (2011) Design and application of a mevalonate-responsive regulatory protein. Angew Chemie Int Ed 50:1084–1086. https://doi.org/10.1002/anie.201006083 Tang SY, Fazelinia H, Cirino PC (2008) AraC regulatory protein mutants with altered effector specificity. J Am Chem Soc 130:5267–5271. https://doi.org/10.1021/ja7109053 Tashiro Y, Fukutomi H, Terakubo K et al (2011) A nucleoside kinase as a dual selector for genetic switches and circuits. Nucleic Acids Res 39:e12–e12. https://doi.org/10.1093/nar/gkq1070 Taylor ND, Garruss AS, Moretti R et al (2016) Engineering an allosteric transcription factor to respond to new ligands. Nat Methods 13:177–183. https://doi.org/10.1038/nmeth.3696 Topp S, Gallivan JP (2008) Random walks to synthetic riboswitches – a high-throughput selection based on cell motility. Chembiochem 9:210–213. https://doi.org/10.1002/cbic.200700546 Trantidou T, Dekker L, Polizzi K et al (2018) Functionalizing cell-mimetic giant vesicles with encapsulated bacterial biosensors. Interface Focus 8:20180024. https://doi.org/10.1098/ rsfs.2018.0024 Tsien RY (1998) The green fluorescent protein. Annu Rev Biochem 67:509–544. https://doi.org/ 10.1146/annurev.biochem.67.1.509 Utsumi R, Brissette R, Rampersaud A et al (1989) Activation of bacterial porin gene expression by a chimeric signal transducer in response to aspartate. Science 245:1246–1249. https://doi.org/ 10.1126/science.2476847 Velur Selvamani R, Telaar M, Friehs K, Flaschel E (2014) Antibiotic-free segregational plasmid stabilization in Escherichia coli owing to the knockout of triosephosphate isomerase (tpiA). Microb Cell Factories 13:58. https://doi.org/10.1186/1475-2859-13-58 Wan X, Volpetti F, Petrova E et al (2019) Cascaded amplifying circuits enable ultrasensitive cellular sensors for toxic metals. Nat Chem Biol 15(5):540–548. https://doi.org/10.1038/s41589-0190244-3

16

Fundamental Building Blocks of Whole-Cell Biosensor Design

405

Wang B, Barahona M, Buck M (2015) Amplification of small molecule-inducible gene expression via tuning of intracellular receptor densities. Nucleic Acids Res 43:1955–1964. https://doi.org/ 10.1093/nar/gku1388 Webb AJ, Kelwick R, Doenhoff MJ et al (2016) A protease-based biosensor for the detection of schistosome cercariae. Sci Rep 6:24725. https://doi.org/10.1038/srep24725 Weber W, Marty RR, Link N et al (2003) Conditional human VEGF-mediated vascularization in chicken embryos using a novel temperature-inducible gene regulation (TIGR) system. Nucleic Acids Res 31:e69 Weitz HJ, Ritchie JM, Bailey DA et al (2001) Construction of a modified mini-Tn 5 lux CDABE transposon for the development of bacterial biosensors for ecotoxicity testing. FEMS Microbiol Lett 197:159–165. https://doi.org/10.1111/j.1574-6968.2001.tb10598.x Wilson D, Charoensawan V, Kummerfeld SK, Teichmann SA (2008) DBD – taxonomically broad transcription factor predictions: new content and functionality. Nucleic Acids Res 36:D88–D92. https://doi.org/10.1093/nar/gkm964 Win MN, Smolke CD (2007) A modular and extensible RNA-based gene-regulatory platform for engineering cellular function. Proc Natl Acad Sci 104:14283–14288. https://doi.org/10.1073/ pnas.0703961104 Wong LS, Judge SK, Voon BWN et al (2018) Bioluminescent microalgae-based biosensor for metal detection in water. IEEE Sensors J 18:2091–2096. https://doi.org/10.1109/JSEN.2017.2787786 Wright O, Delmans M, Stan G-B, Ellis T (2015) GeneGuard: a modular plasmid system designed for biosafety. ACS Synth Biol 4:307–316. https://doi.org/10.1021/sb500234s Yokobayashi Y, Weiss R, Arnold FH (2002) Directed evolution of a genetic circuit. Proc Natl Acad Sci U S A 99:16587–16591. https://doi.org/10.1073/pnas.252535999 Zammit CM, Quaranta D, Gibson S et al (2013) A whole-cell biosensor for the detection of gold. PLoS One 8:e69292. https://doi.org/10.1371/journal.pone.0069292 Ziesack M, Karrenbelt MAP, Bues J et al (2018) Escherichia coli NGF-1, a genetically tractable, efficiently colonizing murine gut isolate. Microbiol Resour Announc 7:e01416–e01418. https://doi.org/10.1128/MRA.01416-18

Mammalian Cell-Based Biosensors

17

Karl-Heinz Feller

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Reporter Gene Constructs and Optical Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Electric Cell-substrate Impedance Sensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Quartz Microbalance Sensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Three-Dimensional Cell-Based Biosensors Based on Synthetic Scaffolds . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

407 410 420 427 428 430

Abstract

The chapter deals with the start-of the art of mammalian cell-based biosensors given the main emphasize to optical, electrochemical, impedimetric (ECIS) and quartz microbalance (QMB) sensing. Because fluorescence sensing is the most common detection mode, the corresponding development of reporter gene constructs is dealt in detail. In three-dimensional cell-constructs the scaffold is very important for the structure of the cells, the survival time, the interaction with the neighbor cell or with other kinds of cells therefore a small chapter will deal with the material basis and special properties various scaffolds have (in dependence upon the cell line used).

Introduction At the beginning let us start with the explanation what is meant if we speak about cell-based biosensors. This is simply necessary because we have a rather strong confusion in the literature about that point. “Cell-based biosensor” means in general that the cell is used as a sensor to measure the response of the cell after a certain interaction with what we call in general an “analyte,” a substance that has an K.-H. Feller (*) Ernst-Abbe-University Jena, Jena, Germany e-mail: [email protected] © Springer Nature Switzerland AG 2022 G. Thouand (ed.), Handbook of Cell Biosensors, https://doi.org/10.1007/978-3-030-23217-7_193

407

408

K.-H. Feller

influence on the cell (growth, shape, proliferation, apoptosis, and necrosis). Wholecell biosensors incorporate living cells, such as mammalian cells “inside” a sensing device for a conversion of initial inputs into cellular response as primary transducer. In this sense whole-cell biosensors are a type of biosensor that incorporates living cells as biorecognition element. The sensitive biological element, in this case a cell is a biologically derived material (e.g., genetically modified) that interacts with, binds with, or recognizes the analyte under study. Cells consist of naturally evolved receptor, ion channels, and enzymes that can be used as targets for biological or biologically active analytes. Thus, whole cell–based biosensors are able to measure functional information and the effects of the analyte on the physiological function of living cells. It should be mentioned that the definition is not as simple as it seems. There are borderline cases where the cell is the biorecognition element as well as the “analyte” rolled into one. This is for instance the case for the wound healing application of mammalian biosensors discussed in this chapter. On the other hand, a sensor, which measures the response of an external cell to an external induction by means of chemical or biological components (receptors), is simple a chemical or biosensor in a special application but no “cell-based biosensor.” Various whole cell–based biosensors have been reported in the literature for the last 20 years, and these reports have shown great potential for their use in the areas of pollution detection in environmental and in biomedical diagnostics. Unlike other reviews of this growing field, Qingyuan Gui et al. (2017) argue that: 1. The selection of reporter genes and their regulatory proteins are directly linked to the performance of cellular biosensors. 2. Broad enhancements in microelectronics and information technologies have also led to improvements in the performance of these sensors. 3. Their future potential is most apparent in their use in the areas of medical diagnostics and in environmental monitoring. 4. Currently the most promising work is focused on the better integration of cellular sensors with nano- and micro-scaled integrated chips. With better integration, it may become practical to see these cells used as real-time portable devices for diagnostics at the bedside and for remote environmental toxin detection, and this in situ application will make the technology commonplace and thus as unremarkable as other ubiquitous technologies. The conventional cell-based biosensors usually use group of cells recording to study behavior of cell populations, what can give broad spectrum of cell sensitivity. Compared with molecular biosensors, cell-based biosensors are expected to respond optimally to bioactive analytes. Additionally, cell-based biosensors are characterized by many advantages, for example, fast response time, long-term recording, and label-free experimentation. However, cell-based biosensors still suffer from some intrinsic shortcomings. The common problems faced by the optimization of cellbased biosensors include how to achieve satisfactory stability, how to improve the selectivity of a sensor cells, and how to prolong the cells’ lifetime.

17

Mammalian Cell-Based Biosensors

409

This chapter of the book will deal with the progress in the development of mammalian cell–based biosensors with some hints to the situation for eukaryotic cell–based biosensors in general. For a critical review of the advantages and disadvantages of using eukaryotic microorganisms (yeasts, microalgae, and ciliated protozoa, but no mammalian cells) to design whole-cell biosensors for monitoring environmental heavy metal pollution, see the review of J. C. Gutiérrez et al. (2015). Another review (Lu et al. 2020) focuses on the properties of mammalian cell–based biosensors and applications in the detection of foodborne pathogens (bacteria and viruses) and toxins (bacterial toxins, mycotoxins and marine toxins). The authors discuss mammalian cell adhesion and how it is involved in the establishment of 3D cell culture models for mammalian cell-based biosensors (Lu et al. 2020). For a more detailed discussion see the part concerning 3D cell culture and the role the scaffold plays in this respect at the end of this chapter. Cell-based biosensors have emerged as powerful functional tools for the rapid detection of hazards and threats associated with food, agriculture, environment, and biosecurity (Sun et al. 2015; El-Ali et al. 2006; Pancrazio et al. 1999; Gupta et al. 2019; Liu et al. 2014a). Cell-based biosensors detect the functional aspects of a host– hazard interaction and render an accurate estimation of the risks. Assessing hazardinduced physiological responses, such as receptor–ligand interactions, signal transduction, gene expression, membrane damage, apoptosis, and oncosis (Ischemic cell death) of living sensing organisms can provide insight into the basis of toxicity for a particular hazard. This review highlights the progress made in developing mammalian cell–based biosensors for pathogens and toxins, with special emphasis on multidisciplinary approaches that combine molecular biology and microbiology with methods used in physics and engineering, which led to the development of a three-dimensional cell-culture system and high-throughput screening employing optical and electrical systems. The nowadays wide-ranging application of cell-based eukaryotic biosensors is based on different advantages against classical animal tests, detailed discussed in the literature (Palmer et al. 2011). Therefore, cells are able due to external stimuli to react in an appropriate physiological manner. Cell typical functions can be investigated separately without overlap of other cell complex organ or body functions. Cultivation of specific cell types in 2D or 3D configuration enables a simple detection by usage of various equipment (Stenger et al. 2001; Li et al. 2003; O’Shaughnessy and Pancrazio 2007; Pampaloni et al. 2007). Whereas bacterial or prokaryotic systems are often applied in environmental control, for example, for the investigation of water or soil samples (Lagarde and Jaffrezic-Renault 2011; D’Souza 2001; Sorensen et al. 2006; Liu et al. 2014b; Wang et al. 2010; Xing et al. 2006; Wegener 2015; Garzón et al. 2019), it is obvious that for the investigation of samples of cosmetically, pharmaceutical or medical origin higher cells are used. Reporter genes respectively proteins and their derived assays are very valuable, nearly indispensable tools in cell and molecular biological research. They have been used since 40 years very successfully for the visualization and tracing of spatial and time-dependent gene and protein expression pattern. The central concept of all established reporter genes, like β-glucuronidase (GUS),

410

K.-H. Feller

β-galactosidase (lacZ), luciferase (luc), chloramphenicol-acetyl-transferase (CAT), or the green fluorescing protein (GFP), is based on the linkage of the reporter gene with a regulatory sequence, and produces a simple to detecting signal which depends after insertion in a biological system upon the modulated expression (Ghim et al. 2010; Jiang et al. 2016). A common characteristic of all enzymatic reporter systems is the application of appropriate low molecular substrates for the qualitative (histochemical) or quantitative (photometrical or fluorometrical) detection of the enzyme activity. The necessary in vitro detection methods are normally destructive (e.g., cell disruption, production of protein extracts, application of appropriate substrates) and prevent therefore a real-time analysis. One further requirement for the application of these systems is the absence of related endogenous activities in the biological systems to investigate to prevent the detection of false positive signals. Furthermore it is necessary that the cells or the tissue can penetrate the appropriate substrates complete, which becomes more problematic as more complex the investigation objects are.

Reporter Gene Constructs and Optical Detection Of great importance are nowadays light-emitting reporter proteins. Luciferases, for example, produce via oxidative decarboxylation light pulses and exhibit a high sensitivity as well as a low background signal. Disadvantageous is the eukaryotic luciferase system; however, due to the necessary cell disruption and the addition of cost-intensive substrate, which prevents a living cell analysis, respectively, a realtime measurement in eukaryotes. In contrary to this the substrate of the prokaryotic luciferase (LuxAB) is produced via the proteins LuxC, D, and E, which all code the Lux-Operon (LuxCDABE) (Meighen 1991). The limitation of the system is however that a usage within higher cells is up to now not possible (Gupta et al. 2003). The 27 kDa large GFP, however, can be excited in vitro via irradiation with ultraviolet or blue light. Remarkable is that GFP is a natural fluorescing protein, which contains a light-emitting chromophore (fluorophore) within his primary amino sequence (Kain 1999). It is produced by means of the three sequenced amino acids Ser-Tyr-Gly of the positions 65–67 via sequential cyclization and oxidation steps. The increasing use of GFPs in the following years after its cloning in 1992 (Prasher et al. 1992) lead, however, to the discovery of a number of limitations. In this respect especially the thermal sensitivity (TOpt: 20–23
C) of the protein folding and the spectral properties must be announced. Examining the fluorescence spectrum of wtGFP (Chalfie et al. 1994) it becomes clear that light of 395 nm and 475 nm is absorbed very strong and is emitted at 508 nm. Specific mutations lead on the one hand to improved absorption/emission (higher emission intensity) and on the other hand to a shift of the otherwise cell damaging absorption peak in the UV. Besides, the elimination of the mentioned limitations and the improvement of the properties of the GFPs lead to the various mutations accompanied by the invention of further fluorescing proteins to the generation of whole color range of fluorescing proteins.

17

Mammalian Cell-Based Biosensors

411

Conventional gene expression analysis usually engages measuring single time point with techniques such as northern blots, reverse transcription polymerase chain reaction (RT-PCR), or DNA microarrays. Taking advantage of these techniques, dynamics can only be approximated by assembling average results from separate cell populations, one for each time point. As seen in Fig. 1 rather different whole cell–based biosensors are meanwhile available. (a) Electric cell–substrate impedance sensing (ECIS) measures the changes in impedance of a cell-culture system (Giaever and Keese 1984, 1993). In this approach, microelectrodes are constructed beneath the cell attachment platform.

Fig. 1 Different whole cell–based biosensors (with permission from (Giaever and Keese 1993)). (a) Phase-sensitive impedance measurement array where the cell monolayer is deposited on a gold film electrode. (b) Engineered B lymphocytes express a bioluminescent protein (the so-called CANARY system). (c) Bioelectronics recognition assay (BERA) uses mammalian cell membranes, the binding of analytes cause changes in the membrane potential

412

K.-H. Feller

When no cells are growing, electric current in the form of ions can flow freely from the surface to the electrodes. Cells growing on the electrode “impede” the flow of current and thus will increase the resistance of the system (measured in ohm) as cell membranes act as insulators. A complete cell monolayer would produce the highest resistance, whereas electrodes without any cells will produce minimal “base-line” resistance. The addition of analytes that either rupture the cell monolayer or cause disturbances, for example by producing “openings” that cause a drop in resistance, indicates the presence of potentially toxic substances. To a more detailed discussion see the part below in this chapter. (b) In the cellular analysis and notification of antigen risks and yields (CANARY) system, engineered B lymphocytes express a bioluminescent calcium-responsive protein in their cytosol and pathogen-specific immunoglobulins (Igs) on their surface. The binding of the antigen to the Igs initiates a signal transduction cascade that results in the increase of the intracellular calcium flux and thus the emission of light, which is detected with a luminometer (Rider et al. 2003). (c) The bioelectric recognition assay (BERA) utilizes mammalian cell membranes that have been engineered to carry antibodies or other molecules, which are inserted by an electroporation method (so-called electroinsertion) (Kintzios 2006; Moschopoulou et al. 2008). The binding of analytes such as virus particles to the antibodies triggers changes in the membrane potential, which are measured by microelectrodes. Biosensors incorporating mammalian cells have a distinct advantage of responding in a manner that can offer insight into the physiological effect of an analyte. Several approaches for transduction of cellular signals are discussed; these approaches include measures of cell metabolism, impedance, intracellular potentials, and extracellular potentials. Among these approaches, networks of excitable cells cultured on microelectrode arrays are uniquely poised to provide rapid, functional classification of an analyte and ultimately constitute a potentially effective cell-based biosensor technology. The usage of genetically modified cells as biosensors having a reporter gene construct connected with a stress-sensitive promoter enables the generation of a functional information with respect to the influence the cell-damaging substance has on the cell physiology (Palmer et al. 2011). Cell-based biosensors of this kind express a spectroscopic active reporter molecule as answer to the chemical “effector.” Advantages and disadvantages have been discussed in detail by several authors (Palmer et al. 2011; Ghim et al. 2010; Banerjee and Bhunia 2009). Besides the enormous progress on that field the establishment of versatile stress-sensitive reporter gene assays within the last three decades, the fusion of a promoter with the DNA sequence of a fluorescing protein seems to be still the method of choice to establish a sensitive, time-resolved, and reliable measurement. Let us go more in detail by discussing two different approaches for the development of a whole cell biosensors assay with optical detection. One very successful approach is the use of fluorescence by means, for example, of green fluorescing proteins (GFPs) to create a detection signal: In this case, as seen in Fig. 2, the analyte interacts with the cell in the sensor array via the cell membrane with the regulatory protein starting a whole signal cascade

17

Mammalian Cell-Based Biosensors

413

Fig. 2 Signal cascade in eukaryotic biosensors (with permission of (Daunert et al. 2000))

inside the cell. Via the reporter plasmid and the transcription into the mRNA, the translation of the signal to the reporter protein is forwarded. The reporter protein expresses after initiation, for example, GFPs which on its part after irradiation emit a fluorescence signal easily to detect and very sensitive. This method has some advantages over other competitive methods, for example, it (a) Responds to external stimuli in a physiologically relevant manner (b) Provides more comprehensive and complex functional information (signaling events, protein synthesis, apoptotic or necrotic cell death) (c) Is modifiable/customizable and applicable to HTS (d) Gives a report of bioavailability (e) Is adaptable to a wide range of transducer methods On the other hand the main disadvantages are: • Temperature and pH sensitive • Short lifetime of the cells on surfaces In Table 1 a selection of typical biosensors based on eukaryotic cells, which are used in cytotoxicity tests, are summarized without guaranty of completeness. It is obvious that rather different detection methods are applied. To be mentioned that bioluminescence as detection principle has been never used with mammalian cells as sensor elements. Disregarding disadvantages (which could be overcome) the application of whole cell biosensors of nanoscale materials in medicine and life science is rapidly increasing.

414

K.-H. Feller

Table 1 Selection of typical eukaryotic cell-based biosensors used in cytotoxicity tests Whole cell–based biosensor HaCaT cells (keratinocytes, human) Ped-2E9 cells (lymphocytes, murine) Vero cells (kidney epithelial cells, monkey) Alcaligenes eutrophus (a gram negative Bacillus) HUVECs cells (umbilical vein endothelial cells, human) Primary cells (cortical cells, rat) HepG2 (hepatoblastoma, human) Chlorella vulgaris (algae) Pseudokirchneriella subcapitata (algae) NIH/3T3 (embryonic fibroblast; mouse)

Analyte plant extracts, heavy metals

Detection method Fluorescence

References Hofmann et al. (2013, 2014)

Listeria monocytogenes strains, Bacillus cereus strains and toxins from Listeria/Bacillus Mycotoxins (t-2, zearalenone)

Laser scanning cytometry

Banerjee et al. (2007)

Spectrophotometry

Bouaziz et al. (2013)

Heavy metals

Bioluminescence/ cyclic voltammetry Ion-selective electrode (ISE)

Corbisier et al. (1999)

Acetylcholine

Potentiometry

Xu et al. (2005)

Copper ions, titanium dioxide, nanoparticles

Bioluminescence

Herbicides

Fluorescence

CuO, ZnO, and TiO2 – nanoparticles Titanium dioxide nanoparticles

Fluorescence

McElwee et al. (2009), Chen and Taniguchi (2012) Védrine et al. (2003) Aruoja et al. (2009) Chen et al. (2011a)

Histamine

Bioluminescence

May et al. (2004a)

This is mainly due to the need of fast, efficient, reliable, and low-cost in vitro test systems as an alternative to animal testing to characterize the effects of chemicals or nanoparticles (NP) to the human body. As seen from Table 2 besides electrical and optical methods with a broad area of application, the determination of the cytotoxicity induced by oxidative stress and cellular uptake of chemicals, NP, and toxins into the cell is of great importance. Concerning the test of toxicity of nanoparticles, see the detailed discussion at the end of this chapter. Before discussing the impedance-based methods let us briefly overview the optical methods in whole cell–based biosensor development. The general question behind the whole cell–based biosensors is as mentioned the development of a cell-based sensor chip for the complex description of physiological properties of chemicals and plant extracts toward a human cell line. There are many methods like MTT, XTT Almar Blue, or LDH tests, which are in normal case endpoint measurements and allow no real-time monitoring. If we speak about the

Table 2 Overview of whole cell biosensors for different application areas (with permission from (Banerjee and Bhunia 2009))

17 Mammalian Cell-Based Biosensors 415

416

K.-H. Feller

necessity to discriminate toxic effects, then we have on the one hand weak reactions (cellular stress, irritation) in mind and, on the other hand, we expect strong reactions (cell death as apoptosis or necrosis). The answer is to look for early biomarkers (intracellular signaling events) which trigger interaction between external stress upon the cell and the reaction of the cell to this external stimulus. Looking back at Fig. 2 it is obvious that the signal cascade in the cell gives many possibilities to develop a signal chain to monitor the external stress (e.g., the production of reactive oxygen species (ROS) or others). In Fig. 3 a scheme of the possible biomarkers in mammalian cells is depicted. Possible realizations are the usage of the respiratory chain or the MAP kinase chain (with the heat shock proteins as excellent biomarkers) as possible starting point for the development of a promoter-reporter gene construct to measure the stress induction (see Fig. 4). The next step is the identification of a powerful biomarker for the detection of, for example, stress induction on the mammalian cells (see Fig. 3). Precondition for the establishment of a sensor cell line is the successful cloning of the construct, which enables the stress-induced expression of products able to be used as a detectable signal in accordance with the control of the functional promoter region of the corresponding gene. Later a stable transfection leads to the establishment of a stable reporter assay. Together with the promoter construct, it leads to the necessary intracellular signaling event (e.g., the abovementioned expression of green

Fig. 3 Scheme of the possible biomarkers to detect reactive oxygen species (ROS) in mammalian cells

17

Mammalian Cell-Based Biosensors

417

Fig. 4 Scheme of the promoter-reporter gene construct to measure stress induction on mammalian cells

fluorescing proteins (GFPs)). The fluorescence of the GFPs is then a direct measure of the irritation of the cells (see right picture of Fig. 4, which shows the green fluorescence of GFP in keratinocytes after irritation). The external stimulus can activate various intracellular signaling pathways like HSPs, NF-κB, MAPK, Rho GTPase, p38-Nrf-2, and SAP/ JNK signaling as shown already in Fig. 3. Different kinds of response elements/biomarkers/receptors (e.g., AP-1(activator protein 1), STAT3 (Signal transducer and activator of transcription 3), HSP (Heat Shock Protein), ISRE/IRF (IFN-stimulated response elements/Interferon regulatory factor) were until now developed in combination with various reporter systems (Hofmann et al. 2014; King et al. 2007; Chen et al. 2011b) to monitor gene expression in mammalian cells after exposure to potential hazardous agent. Undeniably, selecting the right biomarker is necessary for the later proper assessment of the toxic effects of analyzed compounds. The nuclear factor-κB, for example, is a pleiotropic mediator of inducible and specific gene regulation involving various biological activities including immune response, inflammation, cell proliferation, and death. It has been shown (Lindquist and Craig 1988) that heat shock proteins are good biomarkers due to their similar behavior with respect to the application of heat and stress. In living systems, heat shock is a general term for any condition that leads to significant protein damage or unfolding; the name originates from the fact that higher temperatures lead to protein denaturation. Under such conditions, a variety of genes is induced. In this respect heat shock proteins are favored for their use as early biomarkers of stress induction upon the cell. By cloning the biomarker into the signal cascade (e.g., see the vector card in Fig. 5) the promoter-reporter gene construct can be inserted into the sensing cell. The promoter-less vector pAcGFP1-1 enables the observation of the transcription of different, in the multiple cloning site (MCS) inserted promoters. Without insertion of a functional promoter, there is no expression of GFP. Heat shock proteins differ very much in their sensitivity to external stress as seen in Fig. 6. In (Hofmann et al. 2014) it was shown that the two chosen heat shock proteins (HSP27 and HSP72) have very different relative mRNA levels under stress. Whereas HSP27 shows only a slight increase in signal, the relative mRNA level of

418

K.-H. Feller

Fig. 5 Vector card pAcGFP1-1. The promoter-less vector pAcGFP1-1 enables the observation of the transcription of different, in the multiple cloning site (MCS) inserted promoters

HSP72 increases by orders of magnitude making this protein a good candidate for toxicity testing (Fig. 7). The results of the analysis of the dose-responsm reationship of CdCl2 on transfected HaCaT (Human adult low Calcium high Temperature) cells by means of MTT test (yellow tetrazolium reduced to purple formazan) in deficit (1% FBS) and complete medium (10 % FBS) (t ¼ 24 h) is shown in Figure 7. At 80 % viability (1% FBS) the corresponding CdCl2 concentration is about 25 μM. By using these conditions to follow the stress induction in transfected HaCaT cells it gives clear evidence of a linear correlation of the GFP fluorescence signal in dependence on the applied toxin concentration. As seen from Fig. 8 the relative fluorescence signal increases nearly linearly with the concentration of up to 35 μM. At higher concentrations, the signal decreases due to the strongly affected and less proliferating cells (seen in the microscope by morphological changes of the cells in the sample). Remarkable is the fact that the data give clear evidence about the high sensitivity of the method. Comparison of the sensitivity of the method with commercially available sensitivity tests shows orders of magnitude with higher sensitivity of the fluorescence detection (the LOD in this case is around 7 μM CdCl2 concentration). Such a system is favored for toxicity screening at the sublethal level and short incubation times. Figure 9 shows an example of images (microscopic images (amplification 200)) and histogram of flow cytometry before (control) and after stress induction (6 h 25

17

Mammalian Cell-Based Biosensors

419

Fig. 6 Relative mRNA level of two heat shock protein stress markers (HSP72 and HSP27) (with permission from (Hofmann et al. 2013)

μM CdCl2). The flow cytometry gives clear evidence of strong stress induction when CdCl2 is added to the sample (represented by the GFP+ signal). As an alternative to the fluorescence-based detection system, similar biosensors with an absorbance readout have been developed (see for instance (DubiakSzepietowska et al. 2016a, b)), which, for example, allows monitoring the metabolic activity in hepatocyte cells under a broad range of conditions in a real time with superior chemical range and sensitivity (see Fig. 10). In this case the SEAP (human secreted alkaline phosphatase) activity revealed in cell culture medium is directly proportional to changes in intracellular amount of SEAP mRNA and protein. SEAP catalyzes the hydrolysis of p-nitrophenyl phosphate producing a yellow end product that can be measured spectrophotometrically at 405 nm. Therefore, SEAP reporter system is well suited to high-throughput applications. Another important advantage is that background noise of the endogenous alkaline phosphatases is nearly absent. These approaches allow testing of more chemicals at lower cost, can also help characterize the underlying mechanisms by which chemicals interact with human cells and provide physiologically relevant data in response to the analyte and to measure the bioavailability of the analyte. Besides, developed reporter gene assay enable detect toxicity in very early stage, before cell death, in short time even after 2 h time exposure as opposed to widely used commercially available cell-based assays like MTT, XTT Almar Blue, or LDH. Finally, selected screening pathway (NF-κB signaling) can be activated with wide spectrum of stimuli.

420

K.-H. Feller

Fig. 7 Cell viability of gene-modified keratinocytes at different culture conditions (1% and 10% fetal bovine serum (FBS)) measured with MTT test (with permission from (Hofmann et al. 2013)

The validation and optimization of the stably transfected sensor cells (NF-κB_hepG2 cells) was carried out with activators (TNF-α, LPS) of NF-κB pathway. Genetic modifications display any influence on cells sensitivity and growth rate. The disadvantage of the absorption-based sensor assay is the lower sensitivity and detection limit of the absorption. The main advantage is that stray light often disturbing the sensitive detection of fluorescence in a chip with lots of straying interfaces is not limiting the absorption detection (Fig. 11). Such a system was used, for example, for the toxicity testing of nanoparticles (shown for Ag nanoparticles in matrigel environment) (see (Dubiak-Szepietowska et al. 2016a, b) for a more detailed discussion). Comparison of the dose and time dependent detection of the nanomaterials toxicity between sensor cells in 2D and 3D environment showed higher precision of NF-κB_HepG2 cells in Matrigel. This result is clearly indicated from the lower signal of the relative absorbance of the 3D structures in comparison to the 2D monolayer (the dashed blue line).

Electric Cell-substrate Impedance Sensing Electric cell-substrate impedance sensing or ECIS refers to a noninvasive biophysical approach to monitor living animal cells in vitro, that is, within a welldefined laboratory environment. In ECIS the cells are grown on the surface of small

17

Mammalian Cell-Based Biosensors

421

Fig. 8 Fluorescence intensity of transfected HaCaT cell under the action of CdCl2 at different incubation times (with permission from (Hofmann et al. 2013)

and planar gold-film electrodes, which are deposited on the bottom of a cell culture dish. The AC impedance of the cell-covered electrode is then measured at one (mostly at 40 kHz) or several frequencies as a function of time. Due to the insulating properties of their membranes, the cells behave like dielectric particles so that the impedance increases with increasing coverage of the electrode until a confluent (i.e., continuous) layer of cells is established. In confluent cell layers, the threedimensional shape of the cells mainly determines the measured impedance. If cell shape changes occur, the current pathways through and around the cell bodies change as well, leading to a corresponding increase or decrease of impedance. Thus, by recording time-resolved impedance measurements, cell shape changes can be followed in real time with sub-microscopic resolution and can be used for bioanalytical purpose. As the shape of cells responds very sensitively to alterations in metabolism as well as chemical, biological, or physical stimuli, the ECIS technique is applied in various experimental settings in cell biological research laboratories. It can be used as a sensor in cytotoxicity studies, drug development, or as a noninvasive means to follow cell adhesion to in vitro surfaces. Typical applications include measurement of cell attachment and spreading, proliferation, barrier function of endothelia and epithelia, cellular micromotility, lateral migration and wound healing, receptor activation and signal transduction, cell differentiation, and cytotoxicity studies. Equipment based on the ECIS technique are also dedicated to monitor the

422

K.-H. Feller

Fig. 9 Phase contrast, fluorescence images, and flow cytometry results of a reporter promoter construct under stress (HSP72 from above) (above control measurement without stress induction, below at high cadmium chloride level)

Fig. 10 Scheme of an absorption-based biosensor system which uses SEAP (human secreted alkaline phosphatase) as reporter gene and the color change during hydrolysis of the substrate (p-nitrophenyl phosphate) as detecting signal (with permission from (Dubiak-Szepietowska et al. 2016a))

17

Mammalian Cell-Based Biosensors

423

Fig. 11 Relative response of sensor cells (HepG2) in Matrigel after stimulation with AgNP at various incubation times in two-dimensional layer (dashed blue lines) and in three-dimensional organization in matrigel (with permission from (Dubiak-Szepietowska et al. 2016a))

chemokinetic activity of adherent cells spread on the electrode surface (micromotion) as well as their chemotactic activities in ECIS-based wound healing assays (Wegener et al. 2000a) (Fig. 12). Migration of cells plays an important role in processes such as development, wound healing, and cancer progression. For this reason, a diverse set of assays has been developed to investigate the migratory activity of cells. In vitro NRK cells (normal rat kidney cells) grown on circular gold film electrodes stained by a dualcolor vital stain: (1) prior to wounding, (2) immediately after wounding, (3) at ~50% recovery, and (4) after full recovery. In the so-called electric wound-healing assay, the cell-covered thin-film electrodes are used both for creating a defined wound in a confluent cell layer and for real-time documentation of wound closure. Lethal voltages (or currents) are applied to the electrodes for several seconds, which leads to immediate and irreversible permeabilization of the cells that are residing on the electrode. As the insulating passivation layer protects the cells around the electrode, a well-defined and very reproducible wound is created that is restricted to the surface area of the electrode (Fig. 13 B2). Intact cells at the wound edge sense the lesion and start to migrate into

424

K.-H. Feller

Fig. 12 Schematic of the ECIS principle (with permission from (Wegener et al. 2000a))

the open spaces that have been created upon the electrode (Fig. 13 B3). For a more detailed discussion see the excellent reviews in (Stolwijk and Wegener 2019; Keese et al. 2001)). Figure 14 shows that impedimetric measurements are a very useful tool for the recording of stress induction at higher concentrations (where the optical methods fail). At low concentrations of CdCl2 there is no remarkable effect upon the impedance. Starting with 35% of CdCl2 the conformational changes of the cells on the electrode surface become visible. At 60 μM concentration of CdCl2 the signal reaches after 20 h already the cell-free signal level. At 100 μM the cells die after incubation. There are several reports about similar biosensors based on changes in impedance or optical properties in the literature (Prendecka et al. 2018; Michaelis et al. 2012). For example, K.M. L. May et al. (2004b) reported the development of a novel wholecell potentiometric biosensor for screening of toxins. The constructed biosensor consists of a confluent monolayer of human umbilical vein endothelial cells (HUVECs) attached to an ion-selective cellulose triacetate (CTA) membrane modified with a covalently attached RGD (arginine-glycine-aspartic acid) peptide sequence. When the HUVECs form a confluent monolayer, ion transport is almost completely inhibited, thereby reducing the response of the ion-selective electrode (ISE). When the monolayer is exposed to agents that increase its permeability (e.g., toxins), ions can diffuse through the membrane, and a potential response from the ISE is achieved (May et al. 2004b). Hui Jiang et al. (2016) reported the development of sensitive, user-friendly, highthroughput LPS detection in a 96-well microplate using a transcriptional biosensor system, based on 293/hTLR4A-MD2-CD14 (based on HEK-293 cells) cells that are transformed by a red fluorescent protein (mCherry) gene under the transcriptional control of an NF-κB response element. The recognition of LPS activates the

17

Mammalian Cell-Based Biosensors

425

Fig. 13 Fluorescence micrographs of cells at different points of the wound healing/migration process (green – vital cells, red – dead cells). B1 before, B2 immediately after the wounding electrical pulse, B3 after 50% wound healing, and B4 after complete wound healing (with permission from (Stolwijk and Wegener 2019))

biosensor cell, TLR4, and the co-receptor-induced NF-κB signaling pathway, which results in the expression of mCherry fluorescent protein. Summarizing this part of the chapter it can be stated that the external stimulus upon the cell can have broad variety of influences, which result in the metabolism of the cell cytoskeleton and leads to complex changes in the cell morphology. Monitoring the cytotoxicity of drugs, toxin, and pollutants in contact to adherent cells is one of the most important and widely applied impedance-based cellular assays (Arndt et al. 2004; Hong et al. 2011). The spectrum of stressors comprises chemical, biological, and physical challenges to the cells. The biggest asset of noninvasive and label-free devices in this respect is the opportunity for continuous and automated observation over extended time periods so that there is no need to predefine the exposure time a priori. When the cells are followed continuously for a prolonged time, dose-response relationships can be established for different exposure times from a single experiment providing significant extra information. Upon exposure to eventually lethal challenges, the cells will undergo either apoptosis or necrosis (Xie et al. 2012).

426

K.-H. Feller

Fig. 14 Label-free impedimetric recording of stress induction upon HaCaT cells (6 h of incubation) (with permission from (Hofmann et al. 2013))

Fig. 15 Scheme of possible influences upon the cells by external stimulus (adapted with permission from (Xie et al. 2012))

17

Mammalian Cell-Based Biosensors

427

It was the aim of the chapter to show how such changes can be used to characterize the effect an external stimulus/a stress induction can have upon the sensor cell. For a short overview see Fig. 15 where various influences upon the cells connected with changes in the cell morphology are summarized. In this manner, all of them can be more or less sensitive be detected via impedance changes.

Quartz Microbalance Sensing One widely used technique to monitor the behavior of mammalian cells under stress is the quartz microbalance sensor (e.g., (Wegener et al. 1998; Lord et al. 2006a; Janshoff et al. 2000)). It uses piezoelectric resonators and the associated acoustic waves. The quartz crystal microbalance (QCM) is by far the most widely known and most often applied device from this category (Janshoff et al. 2000). The QCM has a long track record as a mass-sensitive tool to study adsorption reactions at the solid– liquid interface. It operates noninvasively and with a superb time resolution that is much better than necessary for most cell-related studies. The core component of this technique is a thin, disk-shaped piezoelectric (AT-cut) quartz crystal sandwiched between two gold-film electrodes (see Fig. 16 for a scheme of the measuring principle). When an oscillating potential difference is applied between the surface electrodes, the piezoelectric resonator performs mechanical shear oscillations parallel to the crystal faces at the resonator’s fundamental resonance frequency. This mechanical oscillation is highly sensitive for any changes at the resonator surface, so that adsorption or desorption processes are measurable as decrease or increase of the resonance frequency (Janshoff et al. 2000).

Fig. 16 Principal scheme of a quartz microbalance (QMB) sensor where cells are attached to the surface of the gold electrodes and interact with toxins connected with Δf and ΔD changes

428

K.-H. Feller

Later it has been recognized that QCM readings of confluent cell layers report on cellular micromechanics, which is dominated by the intracellular cytoskeleton (Wegener et al. 2000b). Recognizing this connection has led to many applications of QCM devices in cell-based bioanalytics (Heitmann and Wegener 2007). Such experiments are normally performed by simultaneously monitoring the shifts in frequency (Δf) and energy dissipation (ΔD) of a silica-coated QCM-D crystal. For instance, the QCM-D has been used as a transducer in cell-based cytotoxicity assays (Marx et al. 2001, 2007) as well as to analyze the cellular response to nanomaterials (Pietuch et al. 2015; Tarantola et al. 2009). However, the results are sometimes difficult to interpret and leave considerable ambiguities – in particular when the resonance frequency is the only parameter being measured. Haibo et al. (2017) investigated the cytoskeleton affecting drug-induced viscoelastic changes of human umbilical vein endothelial cells (HUVECs). This is one of the very rare cases where a QCM sensor was used to investigate drug-cell interactions on mammalian cells by the QCM technique. The results of various studies (e.g., Lord et al. (2006b)) raise interesting questions regarding the early stages of cell response, the influence of drug/chemical concentration, and the characteristics of the chemical perturbation on the cell morphology, thus motivating the need for further research into detection limits and response kinetics of QCM-based whole-cell biosensors.

Three-Dimensional Cell-Based Biosensors Based on Synthetic Scaffolds Let us discuss at the end of the chapter besides the genetic aspects of mammalian cell–based biosensors briefly the question of the matrix in which the mammalian cells are embedded. This is a crucial point, because it is well known that cells that grow in traditional two-dimensional (2D) cell culture monolayers lack physiologically relevant environmental conditions. They mostly exhibit drastic differences of their physical and biochemical properties in comparison to intact biological systems. Therefore, a lot of different methods have been developed to enforce the cells to grow as three-dimensional (3D) structures. Different types of 3D culture exist today, including hydrogel scaffold-based models, which possess a complex structure mimicking the extracellular matrix. These hydrogels can be made of polymers (natural or synthetic) or low-molecular-weight gelators that, via the supramolecular assembly of molecules, allow the production of a reproducible hydrogel with tunable mechanical properties. For a more detailed discussion, see the excellent review of T. Sayde et al. (2021). In two-dimensional systems the problem of building up the required structure is mostly easily solved. The cells grow adherently attached to a surface (or an electrode) in more or less ideal monolayer structure. This is for three-dimensional organization more complicated. A major requirement is biocompatibility of material, and therefore currently material developers combine semisynthetic or biohybrid hydrogels. These kinds of structures usually consist of a synthetic hydrophilic polymer, which is joined to a polysaccharide or protein moiety. Semisynthetic hydrogels are produced in copolymerization reactions between the polymer

17

Mammalian Cell-Based Biosensors

429

Fig. 17 SEM images of hydrogels: (a) collagen (I) gel, (b) Matrigel, (c) gelatin gel 10%(v/v) crosslinked with transglutaminase (with permission from (Dubiak-Szepietowska et al. 2016b))

precursor and the biological conjugate. Such reactions typically adapt the already mentioned cross-linking methods. In this way, it is possible to create a promoting hydrogel with highly defined properties (Vats and Benoit 2013). Hydrogel materials generally exhibit good biocompatibility and high permeability for oxygen, nutrients, and other water-soluble metabolites, making them an attractive scaffold for three-dimensional cell cultivation. A variety of hydrogels has been developed as microenvironment for cells (see the review of Vats and Benoit (Stolwijk and Wegener 2019)). In (Dubiak-Szepietowska et al. 2016b) three-dimensional environments for human cells were developed and tested using three different types of hydrogels including transglutaminase-cross-linked gelatin, collagen type I and growth factor–depleted Matrigel (see Fig. 17). Cells grown in Matrigel exhibited the greatest cell proliferation and spheroid diameter. The time- and dose-dependent toxicity of nanoparticles demonstrates higher cytotoxic effect when HepG2 cells grown as monolayer than embedded in hydrogels. The experimental results provided evidence that cell environment has significant influence for cell sensitivity and that liver spheroid is a useful and novel tool to examine nanoparticles dosing effect even at the level of in vitro studies. As

430

K.-H. Feller

expected, relative response of sensor cells in 3D environment was lower compared to monolayer cell culture (see Fig. 11). Higher resistance of cells to nanomaterials in 3D environment may explain this. In the monolayer culture, cells for toxicity testing of nanoparticle testing are in contact with about 50% of their surface with a solution of nanoparticles, what results in high mortality of cells. Additionally, cells grown on flat surfaces possess an abnormal architecture and are typically under mechanical stress. This should result in lower tolerance to negative external environmental stimuli. Cells in 3D (representing their native shape) are on the other hand less susceptible to environmental pressures. Moreover, the most straightforward explanation for increased resistance in 3D cultures to nanoparticle-mediated toxicity (see, e.g., Fig. 11) is that the cells inside the spheroids may be protected from nanoparticle penetration by the cells in the outer layer of the spheroid. In addition, the lower toxicity of nanoparticles in the 3D cultures may be due to the hindered diffusion through the hydrogels and, therefore, decreased nanoparticles uptake and penetration into the inner layers of the spheroids (Stolwijk and Wegener 2019). Since, cells in native tissue create multilayer structures with hindered diffusion of chemicals and uneven exposure of cells to external stimuli, the developed 3D systems seem to mimic native tissue more closely. Summarizing this chapter it is shown that mammalian cell–based biosensors are widely developed with a broad variety of genetic principles and modifications resulting in a very powerful tool for the investigation of time- and toxicity-dependent phenomena of the cells in the vicinity of toxins or other stress-inducing agents. This will give wide possibilities such constructs to use in environmental, biochemical, pharmaceutical, and medical application. Furthermore, they can mimic the very complex natural behavior of human tissue and organs and so are very helpful in drug and toxicity screening on one hand. On the other side, they reduce the disturbing background, which hinders the reliable and unambiguous assignment of detected effect on the cell level to cellular reasons. They are cheap, stable, and versatile and can be easily adapted to special purposes even for personalized medical applications. We will see in near future much more complex constructs to mimic much better than the present developments the natural conditions of human tissue and organs. Organs-on-a-chip will give a much better possibility to test new drugs for the mechanism of action and side effects without the need of problematic and timeand money-consuming patient tests. Finally, in the near future we will have real artificial organs (like liver, kidney, and others).

References Arndt S, Seebach J, Psathaki K, Galla HJ, Wegener J (2004) Bioelectrical impedance assay to monitor changes in cell shape during apoptosis. Biosens Bioelectron 19:583 Aruoja V, Dubourguier H-C, Kasemets K, Kahru A (2009) Toxicity of nanoparticles of CuO, ZnO and TiO2 to microalgae Pseudokirchneriella subcapitata. Sci Total Environ 407:1461 Banerjee P, Bhunia AK (2009) Mammalian cell-based biosensors for pathogens and toxins. Trends Biotechnol 27:179

17

Mammalian Cell-Based Biosensors

431

Banerjee P, Lenz D, Robinson JP, Rickus JL, Bhunia AK (2007) A novel and simple cell-based detection system with a collagen-encapsulated B-lymphocyte cell line as a biosensor for rapid detection of pathogens and toxins. Lab Investig 88:196 Bouaziz C, Bouslimi A, Kadri R, Zaied C, Bacha H, Abid-Essefi S (2013) The in vitro effects of zearalenone and T-2 toxins on Vero cells. Exp Toxicol Pathol 65:497 Chalfie M, Tu Y, Euskirchen G, Ward WW, Prasher DC (1994) Green fluorescent protein as a marker for gene expression. Science 263:802 Chen P, Taniguchi A (2012) Detection of DNA damage response caused by different forms of titanium dioxide nanoparticles using sensor cells. J Biosens Bioelectron 3. https://doi.org/10. 4172/2155-6210.1000129 Chen P, Migita S, Kanehira K, Sonezaki S, Taniguchi A (2011a) Development of sensor cells using NF- κB pathway activation for detection of nanoparticle-induced inflammation. Sensors 11: 7219 Chen P, Migita S, Kanehira K, Sonezaki S, Taniguchi A (2011b) Development of sensor cells using NF-κB pathway activation for detection of nanoparticle-induced inflammation. Sensors 11:7219 Corbisier P, van der Lelie D, Borremans B, Provoost A, Lorenzo V, Brown NL, Lloyd JR, Hobman JL, Csöregi E, Johansson G, Mattiasson B (1999) Whole cell- and protein-based biosensors for the detection of bioavailable heavy metals in environmental samples. Anal Chim Acta 387:235 D’Souza SF (2001) Microbial biosensors. Biosens Bioelectron 16:337 Daunert S, Barrett G, Feliciano JS, Shetty RS, Shrestha S, Smith-Spencer W (2000) Genetically engineered whole-cell sensing systems: coupling biological recognition with reporter genes. Chem Rev 100:2705 Dubiak-Szepietowska M, Karczmarzyk A, Winkler T, Feller K-H (2016a) A cell-based biosensor for nanomaterials cytotoxicity assessment in three dimensional cell culture. Toxicology 370:60 Dubiak-Szepietowska M, Karczmarzyk A, Jӧnsson-Niedziółka M, Winkler T, Feller K-H (2016b) Development of complex-shaped liver multicellular spheroids as a human-based model for nanoparticle toxicity assessment in vitro. Toxicol Appl Pharmacol 294:78 El-Ali J, Sorger PK, Jensen KF (2006) Cells on chips. Nature 442:403 Garzón V, Pinacho DG, Bustos R-H, Garzón G, Bustamante S (2019) Optical biosensors for therapeutic drug monitoring. Biosensor 9:132. https://doi.org/10.3390/bios9040132 Ghim CM, Lee SK, Takayama S, Mitchell RJ (2010) The art of reporter proteins, in science: past, present and future applications. BMB Rep 43:451 Giaever I, Keese CR (1984) Monitoring fibroblast behavior in tissue culture with an applied electric field. Proc Natl Acad Sci U S A 81:3761 Giaever I, Keese CR (1993) A morphological biosensor for mammalian cells. Nature 366:591 Gui Q, Lawson T, Shan S, Yan L, Liu Y (2017) The application of whole cell-based biosensors for use in environmental analysis and in medical diagnostics. Sensors 17:1623 Gupta RK, Patterson SS, Ripp S, Simpson ML, Sayler GS (2003) Expression of the photorhabdus luminescense lux genes (luxA, B, C, D, and E) in Saccharomyces cerevisiae. FEMS Yeast Res 4: 305 Gupta N, Renugopalakrishnan V, Liepmann D, Paulmurugan R, Malhotra BD (2019) Cell-based biosensors: recent trends, challenges and future perspectives. Biosens Bioelectron 141:111435 Gutiérrez JC, Amaro F, Martin-Gonzáles A (2015) Heavy metal whole-cell biosensors using eukaryotic microorganisms: an updated critical review. Front Microbiol 6. https://doi.org/10. 3389/fmicb.2015.00048 Haibo S, Tiean Z, Jiajin Hu H (2017) A high-throughput QCM chip configuration for the study of living cells and cell-drug interactions. Anal Bioanal Chem 409:6463–6473 Heitmann V, Wegener J (2007) Monitoring cell adhesion by piezoresonators: impact of increasing oscillation amplitudes. Anal Chem 79:3392–3400 Hofmann U, Michaelis S, Winckler T, Wegener J, Feller K-H (2013) A whole-cell biosensor as alternative to skin irritation tests. Biosens Bioelectron 39:156 Hofmann U, Priem M, Bartzsch C, Winckler T, Feller K-H (2014) A sensitive sensor cell line for the detection of oxidative stress responses in cultured human keratinocytes. Sensors 14:11293 Hong J, Kandasamy K, Marimuthu M, Choi CS, Kim S (2011) Electrical cell-substrate impedance sensing as a non-invasive tool for cancer cell study. Analyst 136:237–245

432

K.-H. Feller

Hui J, Jiang D, Shao J, Sun X, J. (2016) Wang, high-throughput living cell-based optical biosensor for detection of bacterial lipopolysaccharide (LPS) using a red fluorescent protein reporter system. Sci Rep 6:36987. https://doi.org/10.1038/srep36987 Janshoff A, Galla HJ, Steinem C (2000) Piezoelectric mass-sensing devices as biosensors-an alternative to optical biosensors? Angew Chem Int Ed Engl 39:4004–4032 Jiang H, Jiang D, Shao J, Sun X, Jiasheng W (2016) High-throughput living cell-based optical biosensor for detection of bacterial lipopolysaccharide (LPS) using a red fluorescent protein reporter system. Sci Rep 6:36987. https://doi.org/10.1038/srep36987 Kain SR (1999) Green fluorescent protein (GFP): applications in cell-based assays for drug discovery. Drug Discov Today 4:304 Keese CR, Wegener J, Walker SH, Giaever I (2001) Electrical wound-healing assay for cells in vitro. Proc Natl Acad Sci U S A 101:1554 King KR, Wang S, Irimia D, Jayaraman A, Toner M, Yarmush ML (2007) A high-throughput microfluidic real-time gene expression living cell array. Lab Chip 7:77 Kintzios SE (2006) Cell-based biosensors in proteomic analysis. Front Drug Des Discov 2:225 Lagarde F, Jaffrezic-Renault N (2011) Cell-based electrochemical biosensors for water quality assessment. Anal Bioanal Chem 400:947 Li N, Tourovskaia A, Folch A (2003) Biology on a chip: microfabrication for studying the behavior of cultured cells. Crit Rev Biomed Eng 31:423 Lindquist S, Craig EA (1988) The heat-shock proteins. Annu Rev Genet 22:631–677 Liu Q, Wu C, Cai H, Hu N, Zhou J, Wang P (2014a) Cell-based biosensors and their application in biomedicine. Chem Rev 114:6423 Liu F, Nordin AN, Li F, Voiculescu I (2014b) A lab-on-chip cell-based biosensor for label-free sensing of water toxicants. Lab Chip 14:1270 Lord MS, Mod C, Foss M, Duch M, Simmons A, Pedersen FS, Milthorpe BK, Besenbacher F (2006a) Monitoring cell adhesion on tantalum and oxidised polystyreneusing a quartz crystal microbalance with dissipation. Biomaterials 27:4529–4537 Lord MS, Modin C, Foss M, Duch M, Simmons A, Pedersen FS, Milthorpe BK, Besenbacher F (2006b) Monitoring cell adhesion on tantalum and oxidized polystyrene using a quartz crystal microbalance with dissipation. Biomaterials 27:4529–4537 Lu X, Ye Y, Zhang Y, Sun X (2020) Current research progress of mammalian cell-based biosensors on the detection of foodborne pathogens and toxins. Crit Rev Food Sci Nutr. https://doi.org/10. 1080/10408398.2020.1809341 Marx KA, Zhou T, Montrone A, Schulze H, Braunhut SJ (2001) A quartz crystal microbalance cell biosensor: detection of microtubule alterations in living cells at nM nocodazole concentrations. Biosens Bioelectron 16(9–12):773–782 Marx KA, Zhou T, Montrone A, McIntosh D, Braunhut SJ (2007) A comparative study of the cytoskeleton binding drugs nocodazole and taxol with a mammalian cell quartz crystal microbalance biosensor: different dynamic responses and energy dissipation effects. Anal Biochem 361:77–92 May KML, Wang Y, Bachas LG, Anderson KW (2004a) Development of a whole-cell-based biosensor for detecting histamine as a model toxin. Anal Chem 76:4156 May KML, Wang Y, Bachas LG, Anderson KW (2004b) Development of a whole-cell based biosensor for detecting histamine as a model toxin. Anal Chem 76:4156 McElwee MK, Song MO, Freedman JH (2009) Copper activation of NF- κB signaling in HepG2 cells. J Mol Biol 393(5):1013 Meighen EA (1991) Molecular biology of bacterial bioluminescence. Microbiol Rev 55:123 Michaelis S, Robelek R, Wegener J (2012) Adv Biochem Engin/Biotechnol 126:33 Moschopoulou G, Vitsa K, Bem F, Vassilakos N, Perdikaris A, Blouhos B, Yialouris C, Frosyniotis D, Anthopoulos I, Mangana O, Nomikou K, Rodeva V, Kostova D, Grozeva S, Michaelides A, Simonian A, Kintzios S (2008) Engineering of the membrane of fibroblast cells with virus-specific antibodies: a novel biosensor tool for virus detection. Biosens Bioelectron 24:1033 O’Shaughnessy J, Pancrazio JJ (2007) Broadband detection of environmental neurotoxicants. Anal Chem 79:8838

17

Mammalian Cell-Based Biosensors

433

Palmer AE, Qin Y, Park JG, McCombs JE (2011) Design and application of genetically encoded biosensors. Trends Biotechnol 29:144 Pampaloni F, Reynaud EG, Stelzer EH (2007) The third dimension bridges the gap between cell culture and live tissue. Nat Rev Mol Cell Biol 8:839 Pancrazio JJ, Whelan JP, Borkholder DA, Ma W, Stenger DA (1999) Development and application of cell-based biosensors. Ann Biomed Eng 27:697 Pietuch A, Bruckner BR, Schneider D, Tarantola M, Rosman C, Sonnichsen C, Janshoff A (2015) Mechanical properties of MDCK II cells exposed to gold nanorods. Beilstein J Nanotechnol 6: 223–231 Prasher DC, Eckenrode VK, Ward WW, Prendergast FG, Cormier MJ (1992) Primary structure of the Aequorea victoria green-fluorescent protein. Gene 111:229 Prendecka M, Frankowski J, Sobieszek G, Kapka-Skrzypczak L, Skwarek-Dziekanowska A, Małecka-Massalska T (2018) Electric cell substrate impedance sensing (ECIS) as a unique technique in cancer metastasis research. J Pre Clin Clin Res 12:142 Rider TH, Petrovick MS, Nargi FE, Harper JD, Schwoebel ED, Mathews RH, Blanchard DJ, Bortolin LT, Young AM, Chen J, Hollis MA (2003) A B cell-based sensor for rapid identification of pathogens. Science 301:213–215 Saydé T, El Hamoui O, Alies B, Gaudin K, Lespes G, Battu S (2021) Biomaterials for threedimensional cell culture: from applications in oncology to nanotechnology. Nano 11:481–509. https://doi.org/10.3390/nano11020481 Sorensen SJ, Burmolle M, Hansen LH (2006) Making bio-sense of toxicity: new developments in whole-cell biosensors. Curr Opin Biotechnol 17:11 Stenger DA, Gross GW, Keefer EW, Shaffer KM, Andreadis JD, Ma W, Pancrazio JJ (2001) Detection of physiologically active compounds using cell-based biosensors. Trends Biotechnol 19:304 Stolwijk JA, Wegener J (2019) Impedance-based assays along the life span of adherent mammalian cells in vitro: from initial adhesion to cell death. Biorev. https://doi.org/10.1007/11663_2019_7 Sun J-Z, Kingori GP, Si R-W, Zhai D-D, Liao Z-H, Sun D-Z, Zheng T, Yong Y-C (2015) Microbial fuel cell-based biosensors for environmental monitoring: a review. Water Sci Technol 71(6):801 Tarantola M, Schneider D, Sunnick E, Adam H, Pierrat S, Rosman C, Breus V, Sonnichsen C, Basche T, Wegener J, Janshoff A (2009) Cytotoxicity of metal and semiconductor nanoparticles indicated by cellular micromotility. ACS Nano 3:213–222 Vats K, Benoit DS (2013) Dynamic manipulation of hydrogels to control cell behavior: a review. Tissue Eng Part B Rev 19:455–469 Védrine C, Leclerc J-C, Durrieu C, Tran-Minh C (2003) Optical whole-cell biosensor using Chlorella vulgaris designed for monitoring herbicides. Biosens Bioelectron 18:457 Wang T-H, Hui G-H, Deng S-P (2010) A novel sweet taste cell-based sensor. Biosens Bioelectron 26:929 Wegener J (2015) Cell-based microarrays for in vitro toxicology. Annu Rev Anal Chem 8:335 Wegener J, Janshoff A, Galla HJ (1998) Cell adhesion monitoring using a quartz crystal microbalance: comparative analysis of different mammalian cell lines. Eur Biophys J 28:26–37 Wegener J, Keese CR, Giaever I (2000a) Electric cell-substrate impedance sensing as a non-invasive means to follow the kinetics of cell spreading on artificial surfaces. Exp Cell Res 259:158 Wegener J, Seebach J, Janshoff A, Galla HJ (2000b) Analysis of the composite response of shear wave resonators to the attachment of mammalian cells. Biophys J 78:2821–2833 Xie F, Xu Y, Wang L, Mitchelson K, Xing W, Cheng J (2012) Use of cellular electrical impedance sensing to assess in vitro cytotoxicity of anticancer drugs in a human kidney cell nephrotoxicity model. Analyst 137:1343–1350 Xing JZ, Zhu L, Gabos S, Xie L (2006) Microelectronic cell sensor assay for detection of cytotoxicity and prediction of acute toxicity. Toxicol In Vitro 20:995 Xu G, Ye X, Qin L, Xu Y, Li Y, Li R, Wang P (2005) Cell-based biosensors based on lightaddressable potentiometric sensors for single cell monitoring. Biosens Bioelectron 20:1757

Mammalian Synbio Sensors

18

Fabiana Tedeschi and Velia Siciliano

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Transcriptional Biosensors’ Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Input Sensing Coupled to Synthetic Transcriptional Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Biosensors with Endogenous Transcription Factors Coupled to Synthetic Promoters . . . . . . . . Other Sensors That Connect to a Transcriptional Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Biosensors That Respond to Light . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Post-transcriptional Sensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Riboswitches and Aptazymes-Based Biosensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RNAi-Based Biosensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Post-translational Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusion and Future Perspective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

436 437 437 439 441 443 445 445 446 449 450 451

Abstract

Mammalian cells behave as computers: they operate as information-processing systems that dynamically integrate and respond to diverse environmental signals. This is possible thank to a myriad of regulatory circuits consisting of sensitive elements that recognize and bind analytes, and of transducer modules that filter the signals and initiate a cellular response modulating theirs or the behavior of heterologous pathways. Similarly, synthetic biology has prompted the design of sensors that are connected to precise actuation to either shed light on the behavior of regulatory networks or to implement novel cellular functionalities with diagnostic or therapeutic purposes. Synthetic biosensors are based on the coupling of sensing and actuation modules that are finely balanced to obtain the desired modularity and specificity. Classically, sensors are defined based on the source of the signals detected: F. Tedeschi · V. Siciliano (*) Synthetic and Systems Biology Lab for Biomedicine, Istituto Italiano di Tecnologia, Naples, Italy e-mail: [email protected]; [email protected] © Springer Nature Switzerland AG 2022 G. Thouand (ed.), Handbook of Cell Biosensors, https://doi.org/10.1007/978-3-030-23217-7_190

435

436

F. Tedeschi and V. Siciliano

biosensors for environmental signals, biosensors for extracellular chemicals, and biosensors for intracellular chemicals and metabolites. Mammalian synthetic biosensors require further classification that relies on the type of activity involved in the signal-to-gene-activation cascade: transcriptional, posttranscriptional, and translational sensors. This chapter provides an overview of mammalian synthetic biology-based sensors and their architecture and connection to downstream signal processing. Keywords

Mammalian synthetic biology · Biosensors · Protein sensors · MicroRNA sensors · Synthetic transcriptional systems · RNA regulation

Introduction Mammalian synthetic biology has two fundamental aims: (i) to better uncover regulatory network principles by relating circuits topology to gene expression dynamics and (ii) to provide novel gene-based and cell-based therapeutic strategies (Caliendo et al. 2019). Rewiring cellular activities, metabolic pathways, and functionality require the design of complex genetic circuits (Black et al. 2017) with sophisticated control of timing, localization, and strength of transgene expression along with mathematical models that help to predict circuits’ performances (Di Bernardo et al. 2012). Gene expression can be regulated by the development of modular smart interfaces composed by sensing and actuating modules that respond to intra- or extracellular stimuli and activate desired outputs in a highly specific and sensitive manner (Ausländer and Fussenegger 2013; Cella and Siciliano 2019). Thus, different molecular species that control signaling cascades are embedded into complex circuits that organize cellular operations at the DNA, RNA, and protein levels. Gene switches in particular represent important computational units, with continuous or discontinuous input-output characteristics, useful to fine-tune gene expression. Mammalian synthetic biologists have implemented a variety of biosensor’ architectures. They vary by the type of recognition element (e.g., proteins or nucleic acids) and whether transduction elements are biological or not biological. In some biosensors, such as whole-cell biosensors, the recognition and transduction elements are entirely biological, whereas other types of designs couple the biological recognition to a physical or chemical transducer. Genetically encoded biosensors often exploit natural macromolecules that cells use to sense their surrounding environment and adjust gene expression to maintain homeostasis under changing conditions. This includes components from natural signaling pathways, protein transcriptional repressors and activators, or riboswitches. Engineered proteins and nucleic acids can also be part of sensing circuits that are orthogonal to the native cellular machinery in an effort to reduce crosstalk and sensor interference. Overall, the interaction of the analyte with the recognition

18

Mammalian Synbio Sensors

437

element may initiate transcription or translation of the reporter gene, or the output of a constitutively expressed protein may be influenced via a posttranslational mechanism (Collins and Khalil 2010; Saxena et al. 2017). Here we describe different synthetic biosensors that rely on/are coupled to transcriptional, posttranscriptional, translational, and posttranslational regulation.

Transcriptional Biosensors’ Architecture Transcription is an orchestrated process that involves tight coordination between transcription factors (TFs) and DNA regulatory sequences in the distal promoter regions, including enhancers and silencers. Transcription-based biosensors require accurate selection of transcription factors (TFs) and promoters to exert robust control of gene expression and can be of two different types. One consists of newly designed transcription factors and associated synthetic promoters whereas the other relies on natural signaling cascades that result in transcription activation. Natural promoters in mammalian cells are composed of a core promoter and proximal promoter elements, and typically spans less than 1 kb base pairs. Synthetic promoters are usually much smaller and are composed of repeated DNA-binding sites for the selected transcription factor and a core promoter that serves as a docking site for the basic transcriptional machinery assembly, defining the position of the transcription start site (TSS) as well as the direction of transcription. Synthetic promoters are generated using a building block approach by inserting cis-regulatory elements, obtained from endogenous promoters, upstream of a minimal promoter to enhance gene expression (Hammer et al. 2006; Roberts et al. 2017). Many parameters of a synthetic promoter can be adjusted, such as the choice of core promoter, the number of operator sites, the spacing between operator sites, and the binding affinity of the operator site for its respective transcription factor. All these parameters can alter the expression dynamics, such as strength, leakiness, and sensitivity of the synthetic promoter (Saxena et al. 2017).

Input Sensing Coupled to Synthetic Transcriptional Systems Transcriptional repressors and activators typically contain a C-terminal ligand-binding domain (Wan et al. 2019). This sensor domain responds to small molecules or environmental stress signal (salt, osmosis, pH, oxygen, redox, light, or radiation, etc.), while a N-terminal DNA-binding domain recognizes a cognate DNA sequence (generally called operator) and regulates gene expression. The binding with a cellular signal induces a conformational change in the TF structure that enables gene expression activation or repression by affecting the binding affinity between the RNA polymerase (RNAP) and the regulated promoter. Partially differently from these, synthetic transcription factors (sTFs) coupled to cognate synthetic promoters consist of the DNA-binding domain and an activation

438

F. Tedeschi and V. Siciliano

(i.e., herpes simplex virus transcriptional activator VP16) or repressor (i.e., KRAB) domain to recruit the RNAP. Associated synthetic promoters include responsive elements for the sTF upstream a minimal core promoter (Ausländer and Fussenegger 2013; Stanton et al. 2014). Synthetic TFs need two key features: they should be orthogonal (influencing only their own cognate promoters, with minimal off-target effects) and programmable (to accommodate a wide range of user-specified transcriptional start sites). With this strategy is easier to couple intracellular or extracellular molecules to a new genetic program avoiding any cross-talk with endogenous processes. Dealing with the need of generating promoter libraries, bioengineers have generated new synthetic systems converting transcription factors expressed in bacteria into optimized gene expression activators and repressors for mammalian cells. For example, Stanton and co-authors have shown a platform to transfer bacteria regulators belonging to the TetR family in mammalian cells and successfully demonstrated them in human embryonic kidney (HEK293) and Chinese hamster ovary (CHO) cells (Stanton et al. 2014). Here, starting from the same prokaryotic TF, they generated repressors and activators. They both contain the same DNA-binding domain. Repressors include a nuclear localization signal (NLS) to ensure their nuclear translocation, blocking the RNAPII processing by steric hindrance (Fig. 1, right), whereas activators additionally include a VP16 activation domain to recruit the transcriptional machinery. Promoters consist of multiple operator sequences for the regulators (Fig. 1, left). Notably among the newly generated activators and repressors, the authors characterized a new inducible system by the small molecule DAPG showing 50-fold induction (Stanton et al. 2014). A similar class of synthetic gene switches makes use of the Saccharomyces cerevisiae regulatory protein GAL4 with its operator known as upstream activator sequence (UAS). By fusing effector domains (i.e., VP16) to Gal4, synthetic transcription factors were created allowing to develop highly selective induction systems that modulate transcription from chimeric UAS-responsive promoters in several Activator

Activator

EYFP

EYFP

Repressor

Fig. 1 Design of synthetic transcriptional systems from Stanton et al. (2014). On the left, a synthetic activator composed by a DNA-binding domain and a transactivation domain binds to a synthetic promoter that includes multiple responsive sequences for the activator. On the right, the design of a repressor system. The repressor blocks transcription by steric hindrance and compete with the activator for regulation of gene expression

18

Mammalian Synbio Sensors

439

species, including mammalian cells (Braselmann et al. 1993). A synthetic transcription factor was created by fusing the DNA-binding domain of Gal4 to the hormonebinding domain of the human estrogen receptor (ER) to develop an estrogendependent induction system that regulate the transcription of Gal4-responsive genes in a hormone-dependent manner. To make this system highly selective, an optimal synthetic Gal4-responsive promoter with low basal activity and high inducibility by Gal-ER was constructed by introducing four Gal4-binding sites, an inverted CCAAT element, a TATA box, and the adenovirus major late initiation region. Moreover, the transactivation domain of the herpes virus protein VP16 onto its C terminus was fused to the transcription factor Gal-ER to render it more potent and less cell-type dependent. Although these transcription factors continue to be used in specific applications, they are increasingly being replaced by programmable transcription factors. The most common DNA-binding platforms used to construct synthetic TFs are transcription activator-like effectors (TALEs) (Kim and Kini 2017) and the clustered regularly interspaced short palindromic repeat (CRISPR)-associated (Cas) system that allows to target virtually any DNA sequences, in contrast to TetR-like proteins and GAL4 in which operator sequences are defined (Moscou and Bogdanove 2009; Boch and Bonas 2010; Jinek et al. 2012; Maeder et al. 2013; Qi et al. 2013). The catalytically dead Cas9 (dCas9) protein has no endonuclease activity and can be genetically fused to effector domains to activate or repress the transcription of genes of interest. It can be directed by a ~100 nt single guide RNA (sgRNA) to a target genomic DNA sequence that is complementary to the first 20 nt of the sgRNA. As the DNA-binding specificity is exclusively governed by the short sgRNA sequence, simple sgRNA cloning procedures can be applied to test large numbers of potential DNA target sites. Heterologous effector domains can be fused to dCas9 to inhibit or partially stimulate synthetic or endogenous mammalian genes, without disrupting its ability to complex and function with sgRNAs, and the expression of multiple guide RNAs in a single cell might enable synergistic activation of endogenous target genes (Maeder et al. 2013). The potentiality of connecting specific sensing to multilayered gene regulation with several transcriptional units and multiplexed regulation of gene expression (i.e., by dCas9) led to the need of designing sequences orthogonal and insulated from endogenous and synthetic genomes. To this end, publicly accessible computational tools, such as R2oDNA algorithm, aid the design of orthogonal sequences that satisfy user-specified constraints such as forbidden sequence motifs and minimum folding free energies (Casini et al. 2014; MacDonald and Siciliano 2017).

Biosensors with Endogenous Transcription Factors Coupled to Synthetic Promoters Biosensors that drive endogenous transcription initiation usually rely on protein recognition elements of extracellular analytes. These can either bind transmembrane receptors or cross the cell membrane and bind repressors (Rosenbaum et al. 2009) or activators (Katzenellenbogen et al. 2000). For transmembrane receptors, interaction

440

F. Tedeschi and V. Siciliano

of the analyte and the extracellular protein recognition element leads to conformational changes that result in activation of an intracellular signaling cascade, often mediated by the phosphorylation of a transcription factor. For repressors and activators, the analyte interacts directly with the transcription factor, determining a conformational change that results in a different affinity for the DNA-binding sequence. Thus, the transcription factor will either release from a promoter thereby relieving steric hindrance or bind and recruit RNA polymerase. In all cases, the presence of the analyte results in the expression of genes downstream of a regulated promoter. Capitalizing on natural signaling cascades, one way to rewire cellular behavior is to engineer downstream synthetic promoters that respond to the naturally activated transcription factor but will initiate a different genetic program. Following this notion, sophisticated therapeutic sensor–effector devices were designed to autonomously detect and treat pathological conditions, and some of them showed to work as expected in animal models of human diseases. For example, a synthetic reward-based dopamine-triggered signaling network was designed to control blood pressure in hypertensive mice (Rössger et al. 2013). This system is based on a synthetic promoter that contain cAMP response elements (PCRE) which induces the expression of the atrial natriuretic peptide (ANP), a powerful vasodilator attenuating high blood pressure when dopamine is sensed (Fig. 2a). Specifically, dopamine binding to the human dopamine receptor DRD1 activates adenylyl cyclase through the G protein Gsα, which catalyzes the conversion of ATP into cAMP. The intracellular cAMP surge and consequent activation protein kinase A (PKA) results in CREB1 activation by phosphorylation. CREB1 binds to its synthetic promoter (PCRE) containing consensus cAMP-responsive elements, driving ANP expression. The synthetic dopamine sensor–effector device was effective to successfully manage antihypertensive treatment in animal models that were stimulated by food, sexual arousal, or addictive drugs to trigger a dopamine release in the brain. Synthetic biosensors have proven remarkable applicability. In a pioneering work, Kemmer and co-authors engineered a biosensor that responds to uric acid by using a modified version of the Deinococcus radiodurans’ uricase regulator (HucR) (Fig. 2b). HucR binds the operator site (hucO) in the absence of uric acid, while in the presence of the metabolite, it dissociates from DNA, thereby allowing expression of a downstream gene. In the uric acid–responsive expression network (UREX), HucR and hucO were reengineered to optimize their expression in mammalian cells. HucR was fused to the C terminus of the Kruppel-associated box (KRAB) protein domain, resulting in a chimeric mammalian urate-dependent trans-silencer (mUTS). A reporter gene encoding human placental secreted alkaline phosphatase (SEAP) was driven by a synthetic promoter (PUREX8), composed by eight tandem hucO modules (hucO8) downstream of a simian virus 40 promoter. Therefore, in the absence of uric acid, mUTS binds hucO8and silences SEAP expression; in the presence of uric acid, mUTS is released from hucO 8 and thereby inducing the transgene expression (Kemmer et al. 2010).

18

Mammalian Synbio Sensors

441

Fig. 2 (a) A synthetic dopamine biosensor is used for hypertension control (Rössger et al. 2013). Dopamine binding to DRD1 induces the ATP catalysis into cAMP through the G protein GSa activation. cAMP activates PKA, which in turn phosphorylates CREB1. The active form of CREB1 binds to its synthetic promoter (PCRE) triggering the transcription of the ANP, a powerful vasodilator attenuating high blood pressure in response to dopamine. (b) A biosensor of uric acid (Kemmer et al. 2010). A chimeric mammalian urate-dependent trans-silencer (mUTS) is developed by fusing a bacterial uricase regulator (HucR) to the transcriptional-silencing domain KRAB. In the absence of uric acid, mUTS binds to an engineered synthetic promoter (PUREX8) silencing the expression of SEAP

Other Sensors That Connect to a Transcriptional Response Often, biosensors can be coupled to multilayered regulation where physiological cues are closely intertwined to complex synthetic systems. For example, a sophisticated genetically encoded anti-inflammatory mammalian cell device was developed to detect and suppress inflammatory flare-ups, before they become harmful (Smole et al. 2017). In this work, the activation of an endogenous TF coupled to an engineered synthetic promoter leads to the expression of a synthetic transcription factor to activate a downstream signaling cascade. In detail, a synthetic NF-kB-responsive promoter monitors the physiologically relevant markers of inflammation and triggers the expression of the synthetic transcriptional activator Gal4-VP16, which drives anti-inflammatory proteins production, while increasing its own expression with a positive feedback amplifier (Fig. 3). As a safety and regulatory feature, a thresholder was included to decrease

442

F. Tedeschi and V. Siciliano

Gal4-VP16

III. EFFECTOR

I. SENSOR TRE

NF-kB responsive element

TRE

Gal4-VP16

II. POSITIVE FEEDBACK AMPLIFIER

TRE

UAS

UAS

Anti-inflammatory proteins

Gal4-VP16

CMV

rtTR-KRAB Gal4 IV. TRESHOLDER

rtTR-KRAB Gal4

Doxy

V. INDUCIBLE OFF SWITCH

Fig. 3 A synthetic anti-inflammatory mammalian cell device is to prevent inflammation-related damages (Smole et al. 2017): (i) a NF-kB-responsive element is the biosensor that monitors the physiologically relevant markers of inflammation; (ii) a positive-feedback amplifier module prevents premature inactivation of the input signals; (iii) an effector induces sustained production of anti-inflammatory proteins; (iv) a thresholder restricts the activation of the positive-feedback loop until the signal is sufficiently strong and decreases the transcriptional leakage of the system; and (v) a tetracycline-inducible OFF switch (implemented as the rtTR-KRAB repressor) resets the device by external chemical signal and shuts down production of the anti-inflammatory effectors when they were no longer required

the transcriptional leakage of the system and to restrict activation of the positivefeedback loop until the signal is sufficiently strong; and a tetracycline-inducible OFF switch (implemented as the rtTR-KRAB repressor – detailed description of such system in the following section) to reset the device by external chemical signal and shut down production of the anti-inflammatory effectors when they were no longer required. Biosensors connected to transcriptional regulation were designed also to detect extracellular or intracellular proteins. In the Tango system (Barnea et al. 2008), G-protein coupled receptors (GPCR) or insulin-like growth factor 1 receptors (IGF1R), which detect vasopressin or IGF1, respectively, were designed to recruit the tobacco TEV protease (TEVp) in proximity of the responsive cleavage site (CS) fused to a membrane-tethered TF. TEVp-mediate cleavage of the CS releases the TF which translocates into the nucleus to activate reporter genes (Fig. 4a). Platforms that expand the diversity of sensors to different signals were demonstrated both for extracellular and intracellular information processing. Generalized extracellular molecule sensor (GEMS) couple antibody fragments that recognize different inducers (i.e., nicotine and prostate-specific antigen-PSA) to JAK/STAT, MAPK, PLGC, or PI3K/Akt signaling. These were then activating synthetic promoters to drive desired output expression (Scheller et al. 2018). A different platform was built to sense intracellular proteins (Siciliano et al. 2018). In this architecture, proteins are sensed by two different intrabodies. The first is fused to TEVp, and the second is anchored to the membrane and fused to a CS

18

Mammalian Synbio Sensors

443

(a)

(b) Ligand Transmembrane receptor scFv1

scFv2 POI

Arrestin

CS TF

CS TEVp

TEVp TF

TF

TF Reporter gene

Reporter gene

Fig. 4 (a) Tango system-based circuit (Barnea et al. 2008). Transmembrane receptors, such as GPCR or IGF1R receptors, are fused to a TEVp cleavage site (CS) and a TF. AVPR2 (ligand) binding to the receptor recruits the arrestin fused to TEVp close the cleavage site, enabling the release of TF and its translocation into the nucleus to activate target gene expression. (b) Modular intrabody-based protein sensing device for transcriptional control (Siciliano et al. 2018). Intracellular target protein is detected by two intrabodies, one fused to a membrane-anchored TF including a CS, and the second to the TEVp. When intrabodies bind to target protein, TEVp is recruited close the CS, thus releasing the TF that can translocate into the nucleus and activate gene expression

and a TF. In presence of the protein, the binding of the intrabodies brings TEVp to the CS and similar to Barnea’s system, the TF translocates to the nucleus, activating target gene expression through a synthetic promoter (Fig. 4b). This platform was demonstrated for four different protein biomarkers of Huntington’s disease, HCV and HIV infection, and reported target immunomodulation in HIV-infected T cells.

Biosensors That Respond to Light Beyond chemically regulated systems, sensors that respond to electricity, temperature, and blue light can serve as input to mammalian transcriptional switches. Notably, optogenetics, in which genetically encoded light-sensitive proteins are used to regulate cellular gene expression, has emerged during the past decade with a broad range of applications (Kim and Kini 2017; Repina et al. 2017; Rost et al. 2017). This is mostly due to some advantages over conventional chemical-based switches, such as potential toxic or off-target effects of the regulatory small molecule, and reversibility of the action (Weber and Fussenegger 2012). These systems

444

F. Tedeschi and V. Siciliano

are specifically responsive to different wavelengths and allow noninvasive and reversible gene expression regulation with high spatiotemporal resolution. The light-regulated mammalian gene expression systems can be categorized into three classes depending on their mode of function. The majority of the systems are two-hybrid systems, which are based on the interaction, light-mediated, between two proteins. The second class is based on light-induced protein homodimerization of otherwise monomeric, inactive regulators. The third-class relies on the light-inducible nuclear import of a TF. The transport into the nucleus can be regulated directly, by fusing a light-inducible nuclear localization signal (NLS) to the TF, or indirectly by the light-inducible dimerization of the TF with an NLS-harboring protein. A synthetic light-controlled transcription device of this kind was developed to restore glucose homeostasis in mice, by using the constitutively expressed human Gprotein coupled receptor (GPCR) melanopsin as sensor module, which activates the endogenous NFAT signaling upon blue-light illumination. The activation of this pathway induces the Glucagon-like peptide 1 (GLP1) secretion into the bloodstream, restoring glucose homeostasis (Ye et al. 2011). Recently, a two-component blue light-responsive optogenetic OFF switch (“Blue-OFF”) was built for rapid and quantitative downregulation of target proteins by combining transcriptional repression with regulation of protein stability, upon illumination with a single wavelength (Baaske et al. 2018). This system consists of two blue light-responsive protein modules: a novel, light-responsive repressor, KRAB-EL222, and the protein degradation module B-LID (blue light-inducible degradation domain). Both components utilize light-oxygen-voltage (LOV) domains. EL222 is a photosensitive transcription factor composed of a light-sensitive LOV domain and a helix-turn-helix (HTH) DNA-binding domain, which mediates light-induced transcription activation. In the dark, the LOV domain binds the HTH domain, precluding dimerization of the transcription factor. Blue light illumination disrupts the inhibitory LOV-HTH interactions and allows EL222 to homodimerize and activate the transcription. In Blue-OFF system, EL222 is used like a light-inducible transcriptional repressor by fusing it to the KRAB transrepressor domain. An example of light-inducible nuclear localization signal was developed introducing a NLS into the sequence of the C-terminal Jα helix of the LOV2 domain of Avena sativa phototropin 1, termed LINus (Niopek et al. 2014). LINuS-mediated nuclear import is fast and reversible, and can be tuned at different levels, such as by introducing mutations that alter AsLOV2 domain photo-caging properties. Optogenetic tools were used also to improve chemically regulated systems normally used in mammalian cells, such as the Tet-OFF/ON system with faster dynamical properties and spatiotemporal control. Specifically, the photoactivatable (PA)-Tet-OFF/ON system utilizes Cry2 and CIB1 cryptochrome domains by Arabidopsis thaliana that respond to blue light, to achieve a large dynamic range of downstream gene expression, rapid activation, and deactivation kinetics, preserving the tight regulation by Doxycycline treatment like the original Tet system (Yamada et al. 2018).

18

Mammalian Synbio Sensors

445

Optogenetic tools may have a groundbreaking impact in the future, although limitations of tissue penetration are still detrimental to the development of new therapeutics for the clinical practice.

Post-transcriptional Sensors Although the largest number of mammalian synthetic circuits is coupled to transcriptional regulation via endogenous transcription factors acting as sensors of cell states, post-transcriptional circuits/sensors are gaining increasing importance in gene regulation by affecting translation or mRNA stability.

Riboswitches and Aptazymes-Based Biosensors Riboswitches are a class of regulatory RNAs that fine-tune target genes expression through small molecule-induced conformational switching (Nudler and Mironov 2004; Mandal and Breaker 2004; Tucker and Breaker 2005). Riboswitches-based sensors contain a ligand-binding aptamer domain that acts as a sensor and an RNA regulatory domain, called transducer module, for downstream gene regulation. Numerous natural riboswitches are present in bacteria where they represent one of the main posttranscriptional control mechanisms, enabling gene expression regulation in response to intracellular metabolite concentrations (Winkler and Breaker 2005). Engineered riboswitches have been developed to construct RNA-based regulatory systems also for eukaryotic organisms. Several studies showed that insertion of aptamers into the 50 -UTR in eukaryotic cells can inhibit translation initiation by adding a small molecule (Werstuck and Green 1998; Suess et al. 2003), while aptamers inserted into the 30 -UTR act via pre-mRNA splicing inhibition (Kim et al. 2005). Another means to regulate degradation rate of mRNAs are aptazymes, namely allosteric self-cleaving ribozymes. These catalytic riboswitches result from the fusion between aptamers and ribozymes and are characterized by the self-cleavage activity of the ribozyme, allosterically regulated by the aptamer (Zhang et al. 2010). The integration of a self-cleaving aptazyme in the 50 - or 30 -UTR region of a target mRNA allows to remove mRNA-stabilizing elements, such as the 50 -cap and the poly(A) tail, triggering a ligand-dependent RNA degradation and preventing the translation in a species-independent manner. Several aptazymes were developed to regulate gene expression in both bacteria and eukaryotes (Yen et al. 2004; Win and Smolke 2007; Wieland and Hartig 2008). For example, Chen and co-authors assembled a synthetic aptazyme to develop a synthetic small-molecule-responsive RNA-based regulatory system that controls mammalian T-cell proliferation (Chen et al. 2010). In this system, the ribozyme-based device is embedded in the 30 -UTR of a gene encoding a proliferative cytokine and was designed in order to induce a conformational change in response to the drug binding. In the absence of the drug, the ribozyme is active and

446

F. Tedeschi and V. Siciliano

Fig. 5 (a) Synthetic aptazyme to control mammalian T-cell proliferation (Chen et al. 2010). In this small-molecule-responsive regulatory system, the ribozyme is placed in the 30 -UTR of IL-2 gene and its activity can be regulated by a drug. In the absence of the drug, the ribozyme is active and its self-cleavage activity leads to IL-2 mRNA degradation, resulting in cell growth inhibition. The drug binding induces a conformational change and the aptazyme becomes inactive, enabling cytokine expression and promoting cell proliferation. (b) miRNA-responsive CRISPR-Cas9 system (miRCas9 switch) in which Cas9-expression is regulated in response to specific miRNAs by using the RNA-binding protein (RBP) L7Ae. L7Ae inhibits Cas9 translation and is in turn regulated by miRNA. miRNAs knock down RBP translation, resulting in Cas9 upregulation (Hirosawa et al. 2017). (c) miRNA-responsive CRISPR-Cas9 system to inhibit Cas9 activity by using a Cas9 antagonist protein, AcrllA4. AcrllA4 activity is in turn regulated by miRNA, so that Cas9 system is activated specifically in target cells in which the antagonist expression is repressed by endogenous miRNA (Hirosawa et al. 2019)

its self-cleavage activity results in target mRNA degradation and reduced cytokine production with following cell growth inhibition. The presence of the drug input stabilizes the ribozyme-inactive conformation, preserving the target mRNA stability and resulting in cytokine production and autocrine cell growth (Fig. 5a).

RNAi-Based Biosensors Posttranscriptional sensors can also take advantage of RNA interference (RNAi), a silencing mechanism that leads to sequence-specific mRNA degradation or translational inhibition (Mittal 2004; Meister and Tuschl 2004), in a spatial and temporal fashion. Inducible RNAi systems can be obtained by employing aptamers that when integrated into siRNAs or miRNAs can modulate their function in a ligand-responsive manner. By incorporating an RNA aptamer for theophylline in the loop region of a short hairpin RNA (shRNA), An and colleagues designed a system that enables a dose-dependent inhibition of RNAi by theophylline. They demonstrated that the small molecule inhibits cleavage of the aptamer-fused shRNA by Dicer in vitro and leads to inhibition of siRNA production in vivo (An et al. 2006).

18

Mammalian Synbio Sensors

447

Another biosensor based on conditional RNAi system was developed by combining a drug-inducible allosteric ribozyme with a miRNA precursor analogue (Kumar et al. 2009). The RNA transcript underlying this system includes a structural analogue of primary miRNA (pri-miRNA) responsible for RNAi, an aptazyme with an embedded theophylline aptamer, and an inhibitory strand attached to the 50 end of the ribozyme that forms a stable stem with the 50 single-stranded portion of the primiRNA module, thus preventing the silencing mechanism. RNAi is activated by theophylline that induces the ribozyme self-cleavage and allows the exposure of the pri-miRNA portion originally masked by the inhibitory strand. Therefore, Drosha can process the pri-miRNA and pre-miRNA produced is then transported into the cytoplasm to be digested by Dicer and induce RNAi. miRNAs can be very sensitive to specific cell state or tissue specific and have been thus used as biomarker of lineage. For this reason, they have been widely used in synthetic biology as biosensors to regulate transgene expression in specific cell types. For example, miRNA-responsive synthetic mRNA systems (called miRNA switches) have been developed to purify target cells based on endogenous miRNA activity. The principle of miRNA switches is to discriminate cell types based on miRNA expression. This leads to the regulation of a target mRNA (of choice) that include target sites for the selected miRNA. This strategy was used to detect human-induced pluripotent stem cells (hiPSCs) in a heterogeneous population by using a miRNA switch that respond to microRNA302a-5p (miR-302a). MiR-302a is highly and specifically expressed in human pluripotent stem cells while decreases to basal levels during the differentiation (Parr et al. 2016). With a similar approach, Miki and co-authors designed miRNA switches to purify several target cell populations, such as cardiomyocytes, endothelial cells, hepatocytes, and insulin-producing cells differentiated from human pluripotent stem cells (PSCs) without affecting cellular properties (Miki et al. 2015a). To build these switches, the authors first identified specific miRNAs for the target populations and then synthesized in vitro mRNAs containing miRNA target sites in the 50 UTR or 30 UTR of the target genes. They also created a miRNA-responsive system to control apoptosis that specifically kills nontarget cells in order to purify target population without cell sorting. Further, same authors have developed a synthetic RNA-based miRNA-responsive CRISPR-Cas9 system (miR-Cas9 switch) by combining the miRNA switch and CRISPR-Cas9 system so that Cas9 activity can be modulated by endogenous miRNAs (Hirosawa et al. 2017). Specifically, a miRNA-complementary sequence was placed in the 50 -UTR region of mRNA encoding Streptococcus pyogenes Cas9. For example, a miR-Cas9 ON switch was developed to upregulate the Cas9 expression in response to a target miRNA by using the RNA-binding protein L7Ae. In this system, L7Ae suppresses Cas9 translation by interacting with its binding K-turn motifs placed in the 50 -UTR region of Cas9-coding mRNA, and L7Ae translation is in turn suppressed by the presence of miRNA target sites in the 50 -UTR region of L7Ae-coding mRNA (Fig. 5b). In this way, in the presence of target miRNA activity, L7Ae is repressed resulting in Cas9 upregulation. However,

448

F. Tedeschi and V. Siciliano

the miR-Cas9 ON switch showed leaky Cas9 expression, probably due to a not complete Cas9 repression by L7Ae. This system was improved by using an antagonist of Cas9, the anti-CRISPR (Acr) protein derived from Listeria monocytogenes prophage AcrllA4, which was in turn regulated by selected miRNAs (Hirosawa et al. 2019). By combining the miRNA-responsive AcrllA4 switch with the CRISPR-Cas9 system, it was possible to generate a miRNA-responsive CRISPR-Cas9 ON system that allows the Cas9 system activation specifically in target cells in which AcrllA4 is repressed by endogenous miRNA (Fig. 5c). miRNA biosensors that couple specific signature of miRNA up- and downregulated, enable the identification of cancer cells. Xie et al. have reported a miRNA-based HeLa cell-specific classifier that monitors the levels of two specific endogenous microRNA sets and triggers a pro-apoptotic response if the miRNA levels match a specific profile (Xie et al. 2011). The HeLa-high miRNA subset (miR21, miR-17, and miR-30a) prevents the expression of the reverse tetracyclinedependent transactivator (rtTA) and lactose operon repressor (LacI) in a two-level cascade designed to repress the pro-apoptotic gene hBax. Instead, the HeLa-low miRNA subset (miR-141, miR-142(3p), and miR-146a) targets the translation of hBax. Therefore, hBax will be exclusively expressed in the presence of high levels of HeLa-high miRNAs and low levels of HeLa-low miRNAs, resulting in specific destruction of matching cancer cells (Fig. 6a). Another means to control RNA regulation with important implications in the construction of biosensors are RNA-binding proteins (RBPs) that can tune translation or degradation of target mRNAs. This strategy show also advantages in the construction of RNA-encoded networks which are gaining attention as they do not integrate into host genome, have rapid dynamics, and no epigenetic silencing. L7Ae and Ms2-cNOT7 are two RBPs recently adapted to mammalian mRNA regulation (Wroblewska et al. 2015a). The first binds K-turn motifs in the 50 UTR of target mRNA, whereas the second binds motifs in the 30 UTR (Ms2) and induce mRNA degradation by polyA chopping (cNOT7). In addition, multilayered gene regulation using the abovementioned RBPs has been achieved by integrating multiple biosensors in mammalian cells. Specifically, these were reengineered to respond to viral proteases, creating interconnecting modular parts through protein-RNA and protein-protein interactions (Cella et al. 2018). L7Ae responsive to TEVp was engineered by inserting the TEV cleavage site (TCS) in the RBP protein structure. The new TEV responsive L7Ae was then used to create different regulatory circuits including protein sensors, cascades, and switches. In the protein sensing device, the target protein is detected by two intrabodies fused to a protease-responsive L7Ae and a tobacco etch virus protease (TEVp), respectively, so that the intracellular expression of a target protein results in proteasemediated cleavage of L7Ae, thus enabling target mRNA translation (Fig. 6b). Another circuit is a three-stage cascade based on engineered proteases, each of which includes a cleavage site for orthogonal proteases, whose function was connected to a protease-responsive MS2-cNOT7. Likewise, this chimeric RBP

18

Mammalian Synbio Sensors

449

Fig. 6 (a) A miRNA-biosensor to classify cancer cells is based on the expression of two endogenous miRNA sets, HeLa-high miRNAs and HeLa-low miRNAs (Xie et al. 2011). HeLahigh miRNAs suppress the expression of the reverse tetracycline-dependent transactivator (rtTA) and the lactose operon repressor (LacI), which represses the expression of the pro-apoptotic gene hBax. HeLa-low miRNAs inhibit directly hBax translation. hBax will be only expressed in the cells where the miRNA levels match a specific profile, characterized by high levels of HeLa-high miRNAs and low levels of HeLa-low miRNAs. (b) Protein-based devices for posttranscriptional control by RNA-binding proteins (RBPs) (Cella et al. 2018). Intrabodies recognizing the intracellular target protein are fused to a protease-responsive L7Ae and a tobacco etch virus protease (TEVp). The intracellular target protein expression induces a TEV-mediated cleavage of L7Ae, thus disrupting its repressive function and enabling the target gene translation. (c) A three-stage cascade is designed by connecting a protein-protein regulation to protein-RNA interaction (Cella et al. 2018). Translation of target transcripts is regulated by the fusion protease-responsive MS2-CScNOT7 that binds to the 30 -UTR and cuts the polyA of the target mRNA inducing degradation. RBP activity is inhibited by a first engineered protease (Protease 1) that cleaves the protease functionally disrupting it, thus restoring the transcript stability. Protease 1 includes a CS for a second protease (Protease 2) so that when this is present, MS2-cNOT7 activity is restored and the translation is repressed

was engineered by inserting a cleavage site between the two domains that compose it, so that its activity may be inhibited by interaction with a first protease (protease 1), thus restoring the mRNA target translation. Protease 1 was in turn engineered to respond to a second protease (protease 2) similarly to L7Ae, such that it can in turn be cleaved, restoring posttranscriptional repression (Fig. 6c).

Post-translational Control In addition to post-transcriptional control, sensors have also been developed to regulate gene expression at the post-translational level, modulating protein stability and inducing protein degradation. Traditional methods for protein function depletion include gene editing or RNAi (Sauer and Henderson 1988; Elbashir et al. 2001), but in both the cases, the depletion is slow-acting and not easily reversible. Novel systems have been designed to overcome these limitations and achieve a rapid and specific protein function modulation by combining pharmacologic

450

F. Tedeschi and V. Siciliano

manipulation with genetic engineering. Several strategies are based on the tagging of a protein of interest with a degradation signal that leads to the protein degradation by ubiquitin-proteasome system. For example, the auxin-inducible degron (AID) system, described for the first time in plants (Abel and Theologis 1996), enables rapid and auxin-inducible degradation of target proteins also in mammalian cells, thus representing a powerful tool to control protein expression and study protein function (Nishimura et al. 2009). Destabilizing domains (DDs) or ligand-induced degradation (LID) domains were also used to regulate protein stability in a specific and ligand-dependent manner. The genetic fusion of DDs to a protein of interest results in the instability of the entire fusion protein with following degradation by the proteasome. By adding a specific small-molecule that targets the DD, the fusion protein can be stabilized thus conferring speed, reversibility, and dose-dependence to this method. For example, several DDs were constructed by engineering the human FK506-binding protein (FKBP12), the E. coli dihydrofolate reductase (ecDHFR) protein, and the ligandbinding domain of the human estrogen receptor, so that they are unstable in the absence of their high-affinity ligands (Banaszynski et al. 2006; Iwamoto et al. 2010; Miyazaki et al. 2012). In contrast to the DD system in which the protein is stabilized by ligand administration, the fusion of LID domains to the target protein makes the fusion protein stable in the absence of ligand and leads to its rapid degradation in the presence of a high-affinity ligand (Bonger et al. 2011). A versatile protein-tagging system, called HaloTag, was designed by using a modified bacterial haloalkane dehalogenase as a sensor part that is fused to a protein of interest. When hydrophobic small molecules covalently bind to the HaloTag, the fusion protein is marked as unfolded and is degraded by the proteasome (Los et al. 2008; Neklesa et al. 2011).

Conclusion and Future Perspective Mammalian synthetic biosensors respond to the urgency of better linking genetically encoded circuits with intracellular and extracellular environment. While initial synthetic networks were designed to gain insights about the biological phenomena and, as such, needed to be insulated from the cellular context to avoid cellular crosstalks, biomedical and industrial applications require fast response to changes in the cellular states. In this chapter, we have focused on the structural architecture of biosensors developed in mammalian cells that can foster much rapid development of therapies and cell response. These include regulatory devices encompassing transcriptional, posttranscriptional, and translational regulation, defining a set of toolboxes available for synthetic biologist for any desired applications (Cella and Siciliano 2019). The different types of regulation coupled to biosensing have advantages and disadvantages. Transcriptional activation or repression is the most frequent network topology associated to molecule sensing (Stanton et al. 2014). This offers the possibility to create programmable circuits that can respond to more than

18

Mammalian Synbio Sensors

451

one input at time, thanks to a large variety of synthetic transcriptional factors available, and even more with new systems based on TALEN or dCas9 (Chavez et al. 2015; Chen et al. 2018) which allow to theoretically target any endogenous or synthetic promoter of interest. These new libraries though require also attentive design of target sequences to avoid cross-talks with endogenous sequences. To address this potential issue, many tools have been developed to enable the design of user specific sequences of interest. One concern still arising is that these systems are DNA-based, requiring integration in the host genome with the risk of genomic mutations, or epigenetic silencing. However, they are still considered a state-of-the art method to convert molecule sensing to cellular response (Smole et al. 2017), either for well-defined systems (system designed for specific sensing) or for modular framework that can recognize multiple targets by just swapping the sensing parts (Siciliano et al. 2018; Scheller et al. 2018). RNA-encoded network on the other side addresses the issues related to DNA-based ones, but connecting sensing to RNA regulation was a challenge until few years ago when synthetic biologist started to create novel RNA regulatory systems. Indeed, miRNA sensors are the most widely used in synthetic biology, as they are valid biomarkers specific of cell lineage and state (Miki et al. 2015). Thus, in the case of miRNA regulation, sensor and output may coincide, without needing intermediate responses. The recent use of RNAbinding proteins (RBPs) increased the toolbox of regulators available (Wroblewska et al. 2015b). By focusing on RNA-protein and protein-protein regulation (Cella et al. 2018) (i.e., with viral protease-responsive RBPs), it is now possible to have complex circuits completely RNA encoded without further requiring DNA delivery. RNA-only circuits exhibit the further advantage of having faster dynamics when compared to DNA devices. Finally, envisioning a more robust control of circuits operations, mammalian biosensors can be embedded in networks that include feedback loop controls (Siciliano et al. 2013). This would enable more sophisticated control engineering strategies to achieve tighter regulation over gene expression.

References Abel S, Theologis A (1996) Early genes and auxin action. Plant physiology 111(1): 9–17. https:// doi.org/10.1104/pp.111.1.9 An CI, Trinh VB, Yokobayashi Y (2006) Artificial control of gene expression in mammalian cells by modulating RNA interference through aptamer-small molecule interaction. RNA 12: 710–716. https://doi.org/10.1261/rna.2299306 Ausländer S, Fussenegger M (2013) From gene switches to mammalian designer cells: present and future prospects. Trends Biotechnol 31:155–168. https://doi.org/10.1016/j.tibtech.2012.11.006 Baaske J, Gonschorek P, Engesser R et al (2018) Dual-controlled optogenetic system for the rapid down-regulation of protein levels in mammalian cells. Sci Rep 8:1–10. https://doi.org/10.1038/ s41598-018-32929-7 Banaszynski LA, Chen L-C, Maynard-Smith LA et al (2006) A rapid, reversible, and tunable method to regulate protein function in living cells using synthetic small molecules. Cell 126:995. https://doi.org/10.1016/j.cell.2006.07.025

452

F. Tedeschi and V. Siciliano

Barnea G, Strapps W, Herrada G et al (2008) The genetic design of signaling cascades to record receptor activation. Proc Natl Acad Sci USA 105:64–69. https://doi.org/10.1073/ pnas.0710487105 Black JB, Perez-Pinera P, Gersbach CA (2017) Mammalian synthetic biology: engineering biological systems. Annu Rev Biomed Eng 19:249–277. https://doi.org/10.1146/annurev-bioeng071516-044649 Boch J, Bonas U (2010) Xanthomonas AvrBs3 family-type III effectors: discovery and function. Annu Rev Phytopathol 48:419–436. https://doi.org/10.1146/annurev-phyto-080508-081936 Bonger KM, Chen L, Liu CW, Wandless TJ (2011) Small-molecule displacement of a cryptic degron causes conditional protein degradation. Nat Chem Biol 7:531–537. https://doi.org/ 10.1038/nchembio.598 Braselmann S, Graninger P, Busslinger M (1993) A selective transcriptional induction system for mammalian cells based on Gal4-estrogen receptor fusion proteins. Proc Natl Acad Sci USA 90:1657–1661. https://doi.org/10.1073/pnas.90.5.1657 Caliendo F, Dukhinova M, Siciliano V (2019) Engineered Cell-Based Therapeutics: Synthetic Biology Meets Immunology. Frontiers in Bioengineering and Biotechnology 7:43. https://doi. org/10.3389/fbioe.2019.00043 Casini A, Christodoulou G, Freemont PS et al (2014) R2oDNA designer: computational design of biologically neutral synthetic DNA sequences. ACS Synth Biol 3:525–528. https://doi.org/10.1 021/sb4001323 Cella F, Siciliano V (2019) Protein-based parts and devices that respond to intracellular and extracellular signals in mammalian cells. Curr Opin Chem Biol 52:47–53. https://doi.org/10.1 016/j.cbpa.2019.04.014 Cella F, Wroblewska L, Weiss R, Siciliano V (2018) Engineering protein-protein devices for multilayered regulation of mRNA translation using orthogonal proteases in mammalian cells. Nat Commun 9:4392. https://doi.org/10.1038/s41467-018-06825-7 Chavez A, Scheiman J, Vora S et al (2015) Highly efficient Cas9-mediated transcriptional programming. Nat Methods 12:326–328. https://doi.org/10.1038/nmeth.3312 Chen YY, Jensen MC, Smolke CD (2010) Genetic control of mammalian T-cell proliferation with synthetic RNA regulatory systems. Proc Natl Acad Sci USA 107:8531–8536. https://doi.org/ 10.1073/pnas.1001721107 Chen Z, Boyken SE, Jia M et al (2018) Programmable design of orthogonal protein heterodimers. Nature 1:106. https://doi.org/10.1038/s41586-018-0802-y Collins JJ, Khalil AS (2010) Synthetic biology applications come of age. Nat Rev Genet 11:367–379. https://doi.org/10.1038/nrg2775.Synthetic Di Bernardo D, Marucci L, Menolascina F, Siciliano V (2012) Predicting synthetic gene networks. Methods Mol Biol 813:57–81. https://doi.org/10.1007/978-1-61779-412-4_4 Elbashir SM, Lendeckel W, Tuschl T (2001) RNA interference is mediated by 21- and 22nucleotide RNAs. Genes Dev 15:188–200. https://doi.org/10.1101/gad.862301 Hammer K, Mijakovic I, Jensen PR (2006) Synthetic promoter libraries – tuning of gene expression. Trends Biotechnol 24:53–55. https://doi.org/10.1016/j.tibtech.2005.12.003 Hirosawa M, Fujita Y, Parr CJC et al (2017) Cell-type-specific genome editing with a microRNAresponsive CRISPR-Cas9 switch. Nucleic Acids Res 45:e118. https://doi.org/10.1093/nar/ gkx309 Hirosawa M, Fujita Y, Saito H (2019) Cell-type-specific CRISPR activation with MicroRNAResponsive AcrllA4 switch. ACS Synth Biol. https://doi.org/10.1021/acssynbio.9b00073 Iwamoto M, Björklund T, Lundberg C et al (2010) A general chemical method to regulate protein stability in the mammalian central nervous system. Chem Biol. https://doi.org/10.1016/j. chembiol.2010.07.009 Jinek M, Chylinski K, Fonfara I et al (2012) A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity. Science 337:816–821. https://doi.org/10.1126/science. 1225829 Katzenellenbogen BS, Katzenellenbogen JA, Medicine C (2000) Estrogen receptor transcription and transactivation Estrogen receptor alpha and estrogen receptor beta: regulation by selective estrogen receptor modulators and importance in breast cancer. Breast Cancer Res 2:33

18

Mammalian Synbio Sensors

453

Kemmer C, Gitzinger M, Daoud-El Baba M et al (2010) Self-sufficient control of urate homeostasis in mice by a synthetic circuit. Nat Biotechnol 28:355–360. https://doi.org/10.1038/nbt.1617 Kim M-S, Kini AG (2017) Engineering and application of zinc finger proteins and TALEs for biomedical research. Mol Cells 40:533–541. https://doi.org/10.14348/molcells.2017.0139 Kim D-S, Gusti V, Pillai SG, Gaur RK (2005) An artificial riboswitch for controlling pre-mRNA splicing. RNA 11:1667–1677. https://doi.org/10.1261/rna.2162205 Kumar D, Il AC, Yokobayashi Y (2009) Conditional RNA interference mediated by allosteric ribozyme. J Am Chem Soc 131:13906–13907. https://doi.org/10.1021/ja905596t Los GV, Encell LP, Mcdougall MG et al (2008) HaloTag: a novel protein labeling technology for cell imaging and protein analysis. ACS Chem Biol 3:373–382 MacDonald JT, Siciliano V (2017) Computational sequence design with R2oDNA designer. In: Humana Press (ed) Methods in molecular biology. Humana Press, New York, pp 249–262 Maeder ML, Linder SJ, Cascio VM et al (2013) CRISPR RNA-guided activation of endogenous human genes. Nat Methods 10:977–979. https://doi.org/10.1038/nmeth.2598 Mandal M, Breaker RR (2004) Gene regulation by riboswitches. Nat Rev Mol Cell Biol 5:451–463. https://doi.org/10.1038/nrm1403 Meister G, Tuschl T (2004) Mechanisms of gene silencing by double-stranded RNA. Nature 431:343–349. https://doi.org/10.1038/nature02873 Miki K, Endo K, Saito H et al (2015a) Efficient detection and purification of cell populations using synthetic MicroRNA switches cell stem cell resource efficient detection and purification of cell populations using synthetic MicroRNA switches. Stem Cell 16:699–711. https://doi.org/10.10 16/j.stem.2015.04.005 Mittal V (2004) Improving the efficiency of RNA interference in mammals. Nat Rev Genet 5:355–365. https://doi.org/10.1038/nrg1323 Miyazaki Y, Imoto H, Chen L-c, Wandless TJ (2012) Destabilizing domains derived from the human estrogen receptor. J Am Chem Soc 23:1–7. https://doi.org/10.1038/jid.2014.371 Moscou MJ, Bogdanove AJ (2009) A simple cipher governs DNA recognition by TAL effectors. Science 326:1501. https://doi.org/10.1126/science.1178817 Neklesa TK, Tae HS, Schneekloth AR et al (2011) Small-molecule hydrophobic tagging-induced degradation of HaloTag fusion proteins. Nat Chem Biol 7:538–543. https://doi.org/10.1038/ nchembio.597 Niopek D, Benzinger D, Roensch J et al (2014) Engineering light-inducible nuclear localization signals for precise spatiotemporal control of protein dynamics in living cells. Nat Commun 5:1–11. https://doi.org/10.1038/ncomms5404 Nishimura K, Fukagawa T, Takisawa H et al (2009) An auxin-based degron system for the rapid depletion of proteins in nonplant cells. Nat Methods 6:917–922. https://doi.org/10.1038/ nmeth.1401 Nudler E, Mironov AS (2004) The riboswitch control of bacterial metabolism. Trends Biochem Sci 29:11–17. https://doi.org/10.1016/j.tibs.2003.11.004 Parr CJC, Katayama S, Miki K et al (2016) MicroRNA-302 switch to identify and eliminate undifferentiated human pluripotent stem cells. Sci Rep 6:32532. https://doi.org/10.1038/srep32532 Qi LS, Larson MH, Gilbert LA et al (2013) Repurposing CRISPR as an RNA-γuided platform for sequence-specific control of gene expression. Cell 152:1173–1183. https://doi.org/10.1016/j. cell.2013.02.022 Repina NA, Rosenbloom A, Mukherjee A et al (2017) At light speed: advances in Optogenetic Systems for Regulating Cell Signaling and Behavior. Annu Rev Chem Biomol Eng 8:13–39. https://doi.org/10.1146/annurev-chembioeng-060816-101254 Roberts ML, Katsoupi P, Tseveleki V, Taoufik E (2017) Bioinformatically informed design of synthetic mammalian promoters. Methods Mol Biol 1651:93–112. https://doi.org/10.1007/9781-4939-7223-4_8 Rosenbaum DM, Rasmussen SGF, Kobilka BK (2009) The structure and function of G-proteincoupled receptors. Nature 459:356-363. https://doi.org/10.1038/nature08144 Rössger K, Charpin-El Hamri G, Fussenegger M (2013) Reward-based hypertension control by a synthetic brain-dopamine interface. Proc Natl Acad Sci USA 110:18150–18155. https://doi.org/ 10.1073/pnas.1312414110

454

F. Tedeschi and V. Siciliano

Rost BR, Schneider-Warme F, Schmitz D, Hegemann P (2017) Optogenetic tools for subcellular applications in neuroscience. Neuron 96:572–603. https://doi.org/10.1016/j.neuron.2017. 09.047 Sauer B, Henderson N (1988) Site-specific DNA recombination in mammalian cells by the Cre recombinase of bacteriophage P1. Proc Natl Acad Sci USA 85:5166–5170. https://doi.org/ 10.1073/pnas.85.14.5166 Saxena P, Bojar D, Fussenegger M (2017) Mammalian synthetic promoters 1651:263–273. https:// doi.org/10.1007/978-1-4939-7223-4 Scheller L, Strittmatter T, Fuchs D et al (2018) Generalized extracellular molecule sensor platform for programming cellular behavior. Nat Chem Biol 14:723–729. https://doi.org/10.1038/ s41589-018-0046-z Siciliano V, Garzilli I, Fracassi C et al (2013) MiRNAs confer phenotypic robustness to gene networks by suppressing biological noise. Nat Commun 4:2364 Siciliano V, Diandreth B, Monel B et al (2018) Engineering modular intracellular protein sensoractuator devices. Nat Commun 9:1–7. 1881. https://doi.org/10.1038/s41467-018-03984-5 Smole A, Lainšček D, Bezeljak U et al (2017) A synthetic mammalian therapeutic gene circuit for sensing and suppressing inflammation. Mol Ther 25:102–119. https://doi.org/10.1016/j. ymthe.2016.10.005 Stanton BC, Siciliano V, Ghodasara A et al (2014) Systematic transfer of prokaryotic sensors and circuits to mammalian cells. ACS Synth Biol 3:880. https://doi.org/10.1021/sb5002856 Suess B, Hanson S, Berens C et al (2003) Conditional gene expression by controlling translation with tetracycline-binding aptamers. Nucleic Acids Res 31:1853–1858. https://doi.org/10.1093/ nar/gkg285 Tucker BJ, Breaker RR (2005) Riboswitches as versatile gene control elements. Curr Opin Struct Biol 15:342–348. https://doi.org/10.1016/j.sbi.2005.05.003 Wan X, Marsafari M, Xu P (2019) Engineering metabolite-responsive transcriptional factors to sense small molecules in eukaryotes: current state and perspectives. Microb Cell Factories 18:1–13. https://doi.org/10.1186/s12934-019-1111-3 Weber W, Fussenegger M (2012) Emerging biomedical applications of synthetic biology. Nat Rev Genet 13:21–35. https://doi.org/10.1038/nrg3094 Werstuck G, Green MR (1998) Controlling gene expression in living cells through small moleculeRNA interactions. Science 282:296–298. https://doi.org/10.1126/science.282.5387.296 Wieland M, Hartig JS (2008) Artificial riboswitches: synthetic mRNA-based regulators of gene expression. Chembiochem 9:1873–1878. https://doi.org/10.1002/cbic.200800154 Win MN, Smolke CD (2007) A modular and extensible RNA-based gene-regulatory platform for engineering cellular function. Proc Natl Acad Sci USA 104:14283–14288. https://doi.org/ 10.1073/pnas.0703961104 Winkler WC, Breaker RR (2005) Regulation of bacterial gene expression by riboswitches. Annu Rev Microbiol 59:487–517. https://doi.org/10.1146/annurev.micro.59.030804.121336 Wroblewska L, Kitada T, Endo K et al (2015a) Mammalian synthetic circuits with RNA binding proteins delivered by RNA. Nat Biotechnol 33:839–841 Wroblewska L, Kitada T, Endo K et al (2015b) Mammalian synthetic circuits with RNA binding proteins for RNA-only delivery. Nat Biotechnol 33:839–841. https://doi.org/10.1038/nbt.3301 Xie Z, Wroblewska L, Prochazka L et al (2011) Multi-input RNAi-based logic circuit for identification of specific cancer cells. Science 333:1307–1311. https://doi.org/10.1126/scie nce.1205527 Yamada M, Suzuki Y, Nagasaki SC et al (2018) Light control of the Tet gene expression system in mammalian cells. Cell Rep 25:487–500.e6. https://doi.org/10.1016/j.celrep.2018.09.026 Ye H, Daoud-El Baba M, Peng R-W, Fussenegger M (2011) A synthetic optogenetic transcription device enhances blood-glucose homeostasis in mice. Science 332:1565–1568. https://doi.org/ 10.1126/science.1203535 Yen L, Svendsen J, Lee JS et al (2004) Exogenous control of mammalian gene expression through modulation of RNA self-cleavage. Nature 431:471–476. https://doi.org/10.1038/nature02844 Zhang J, Lau MW, Ferré-D’Amaré AR (2010) Ribozymes and riboswitches: modulation of RNA function by small molecules. Biochemistry 49:9123–9131. https://doi.org/10.1021/bi1012645

Environmental Biosensors: A Microbiological View

19

Juan-Carlos Gutiérrez, Francisco Amaro, Silvia Díaz, and Ana Martín-González

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Environmental Biosensors: Why Use Microorganisms? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Metal(loid) Biosensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Bacteria-Based Metal(loid) Biosensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Bacterial-WCBs Based on Riboswitches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Eukaryotic Microorganism-Based Metal(loid) Biosensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Xenobiotic Biosensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Bacteria-Based Xenobiotic Biosensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions and Future Perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

456 458 460 460 464 465 468 468 470 471

Abstract

In this mini-review, the potential of using microorganisms to design biosensors for detecting environmental pollutants is analyzed and discussed. A distinction is made between a classical biosensor (CB) and a whole-cell biosensor (WCB), emphasizing their structural components and the possibility of using whole microorganisms as their bioreceptor elements. The advantages and disadvantages of using prokaryotic microorganisms as opposed to eukaryotic microorganisms are described. Likewise, the advantages of using protozoa (ciliates) over other eukaryotic microorganisms are also shown. We analyze the current bibliography on biosensors built on microorganisms as bioreceptors of pollutant molecules, such as inorganic (metal (loid)s) or organic (xenobiotics). New trends, such as the prokaryotic riboswitches, microbial two-component systems where the pollutant can be simultaneously detected and bioremediated, along with advances in synthetic biology, are shown as promising tools in the design of environmental biosensors. J.-C. Gutiérrez (*) · F. Amaro · S. Díaz · A. Martín-González Dpto. Genética, Fisiología y Microbiología. Facultad de Biología, Universidad Complutense (UCM), Madrid, Spain e-mail: [email protected] © Springer Nature Switzerland AG 2022 G. Thouand (ed.), Handbook of Cell Biosensors, https://doi.org/10.1007/978-3-030-23217-7_191

455

456

J.-C. Gutiérrez et al.

Keywords

Environmental pollutants · Microbial biosensors · Metal(loid)s · Riboswitches · Two-component systems · Xenobiotics

Introduction The Anthropocene is the geological epoch that has been proposed to succeed the Holocene (the current epoch of the Quaternary Period), due to the tremendous global impact that human activity has on all the ecosystems of our planet. During the last two centuries, the anthropogenic activity has not only caused a significant accumulation of greenhouse gases (such as CO2) but has also increased levels of both inorganic (metal(loid)s) and organic (xenobiotics) pollutants. This activity has serious impacts on climate (climate change we are already undergoing), biodiversity, environment, and human health. Certain metal(loid)s, mainly those misnamed “heavy metal(loid)s,” are among the most abundant, toxic, and persistent inorganic environmental pollutants (Hill 2004). Although one-third of the elements from the periodic table are essential to life, they can be toxic at high concentrations, and other nonessential elements are very toxic to living beings at very low concentrations. Diverse anthropogenic sources, such as mining and other industrial activities, have substantially increased the metal(loid) content present in the atmosphere and in many terrestrial or aquatic ecosystems (Peñuelas and Fillela 2002). Metal(loid)s can, directly or indirectly, originate reactive oxygen species (ROS) with significant alterations in proteins, nucleic acids, and lipids (Leonard et al. 2004; Valko et al. 2005). A new confirmation of the pollution of our planet is a recent report on Antarctica, in which atmospheric aerosols have been analyzed during the austral summer (2016–2017) (Caceres et al. 2019). The average mass concentration of particulate matter in some places of the coast of the Antarctica region was as high as 28.2 μg/m3. About 100 times more lead and about 600 times more chromium have been found in the air than in the soil in this region, showing a remote anthropogenic origin (Caceres et al. 2019). Organic compounds that are not part of living beings (xenobiotics) can also be an important source of environmental pollution. However, unlike metal(loid)s, many of them can be completely degraded by microorganisms to their final components (CO2 + H2O). But, like metal(loid)s, they can be very toxic too. Because of the ecological, sanitary, and economic consequences of these two big groups of environmental pollutants, they are considered to be priority in ecotoxicology, with the aim of minimizing the exposure to animals or humans. It is difficult to predict the global effects of increasing the different types of environmental pollutants, so there is an overriding need to develop screening methods for environmental monitoring. This need is both for the detection of pollutant compounds and monitoring of bioremediation processes of ecosystems contaminated by inorganic or organic compounds. These chemical pollutants can be measured using molecular recognition or chemical analysis, such as absorption spectroscopy, mass spectroscopy, gas chromatography, polarography, and others. However, these techniques

19

Environmental Biosensors: A Microbiological View

457

require qualified personnel, present a high cost, and it is not possible to carry out in situ analysis. In addition, important ecotoxicological parameters such as bioavailability, toxicity, and genotoxicity can only be assayed using living cells (Gutierrez et al. 2015, 2017). Likewise, the most sensitive screening methods to detect pollutants are those incorporating biological components, which are used as targets for an active substance or pollutant. These detection tools that incorporate biological material or are living cells are known as biosensors or bioreporters. Therefore, we can distinguish two types of biosensors: the classical or conventional biosensors (CBs) and the whole-cell biosensors (WCBs) (Gutierrez et al. 2017). We can define CBs as integrated bioreceptor-physicochemical transducer devices, which consist of three different parts: a bioreceptor or biological recognition element, which interacts with the pollutant molecules; a physicochemical transducer, which converts the biological response into a measurable physicochemical signal; and a microelectronic processor of this signal, which amplifies and converts it into a numeric record. The biological component can be macromolecules (enzymes, antibodies, nucleic acids, etc.) or whole cells (microorganisms or cells from multicellular organisms). We can consider four main different types of transducers: electrochemical (potentiometric or amperometric), optical (spectrophotometric or fluorometric), piezoelectric, or thermometric. To construct these CBs, we need to have both biological and physicochemical knowledge, which frequently involves an interdisciplinary cooperation among different specialists. The second type of biosensors, WCBs, was introduced as an alternative to CBs (Belkin 2003; Van der Meer and Belkin 2010). WCBs use prokaryotic or eukaryotic whole cells as single reporters, which incorporate both bioreceptor and transducer elements into the same cell. In general, this involves that organisms used as WCBs are experimentally modified to incorporate transducer capacity or increase their sensitivity against the pollutant. Unlike CBs, these WCBs have the advantageous feature of carrying out both in situ and ex situ analyses. When using WCBs, two different types of bioassays can be distinguished; turn off or turn on assays (Belkin 2003). Turn off bioassays are similar to standard toxicological bioassays, so the sample toxicity is evaluated from the inhibition degree of a specific cellular activity, such as growth inhibition, respiration rate, motility depletion, etc., or an unspecific cell viability. In these bioassays, the toxic concentration is proportional to the measurement of the cellular function inhibition or the cell mortality percentage. The molecular reporters used in turn off bioassays are under a constitutive gene expression (Gutierrez et al. 2017); therefore, by increasing the toxicity of the sample, cell viability is affected, and the expression of the reporter gene decreases. An example of this type of bioassay is the one marketed by the company NCIMB (UK) and called MARA (Jouanneau et al. 2017). This bioassay is based on the growth inhibition of 11 microbial strains (including 10 bacteria and 1 yeast), and cell viability is measured in microtiter plates (96 wells) by assessing of the intensity of a redox red dye which acts as a marker of cellular metabolic activity. Another possibility is to use natural bioluminescent bacteria strains (LumiMARA bioassay), such as Aliivibrio fischeri or Photobacterium phosphoreum among others, so the pollutant toxic effect is assessed by the bioluminescence inhibition emitted by the cells (Jung et al. 2015).

458

J.-C. Gutiérrez et al.

On the other hand, in turn on bioassays, a quantifiable molecular reporter is fused to an inducible gene promoter to be activated by a specific or group of environmental pollutants. In these WCBs, the sample toxicity or pollutant concentration is proportional to the reporter molecule gene expression. This reporter signal may reach a maximum value (critical pollutant concentration); after reaching this value, the reporter signal decreases due to the greater toxicity of the sample that can lead to cell mortality. This critical pollutant concentration value will depend on the degree of cellular resistance to the pollutant. In general, turn off bioassays are usually quite unspecific because the reporter signal decreases as a result of a broad range of cytotoxic effects originated by very different pollutants, while turn on bioassays or CBs (using specific molecules as bioreceptors) are habitually more specific, as induction of the gene reporter or interaction with the molecular bioreceptor only takes place when the specific pollutant is present. Therefore, the WCB specificity will depend on the degree of the gene promoter specificity to be induced by an exclusive pollutant or a chemically related group of pollutants. CB specificity will depend on the specificity degree of the interaction between the bioreceptor and the pollutant. With regard to specificity, both turn on WCBs and CBs have been classified into effect- and compound-specific biosensors (Yagi 2007). Effect-specific biosensors respond to physicochemical environmental changes (e.g., pH, temperature, or osmotic changes) or to a chemically diverse group of pollutants that induce a type of cellular stress response (e.g., oxidative stress or protein toxicity). Compound-specific biosensors respond to only one type of pollutant or compounds with similar chemical features (e.g., metal(loid)s). For some other specialists, the specificity-based classification of CBs or WCBs may be divided into three classes: class-I biosensors which only respond to a specific or exclusive pollutant by increasing the reporter signal; class-II that responds to a specific cellular stress, like oxidative stress, by increasing the reporter signal; and class-III responding unspecifically to different pollutants or environmental stressors (Gutierrez et al. 2017). All these types of biosensors (CBs and WCBs) can be useful to detect the presence of organic or inorganic pollutant molecules in the environment. In addition, they can be also used for biomonitoring or testing the progress of a bioremediation process after detecting the chemical nature of the pollutant.

Environmental Biosensors: Why Use Microorganisms? Microorganisms are the most abundant living beings on our planet. The estimated number of microbial genomes in the biosphere is around 1029 to 1030 (Huse et al. 2010; Kallmeyer et al. 2012), which exceeds the estimated number of galaxies (1011) present in our observable universe. It is between 2 and 3 orders of magnitude of the total number of animal and plant cells together on our planet. In the oceans, they represent 90% of the weight of all organisms. Therefore, they are not only quantitatively important but also qualitatively important, since they occupy all known ecosystems. This adaptation to different ecosystems has contributed to the origin of many different

19

Environmental Biosensors: A Microbiological View

459

metabolisms and physiologies. Microorganisms participate in many biogeochemical cycles, with some exclusive physiologies, such as the atmospheric nitrogen fixation or the anaerobic ammonium oxidation. Microorganisms interact with both biotic and abiotic components of their ecosystems, and these interactions are essential for ecosystem functions (Shahsavari et al. 2017). They have developed different mechanisms to counteract the toxic action of inorganic and organic contaminants, and they can eat many organic pollutants of anthropogenic origin (biodegradation). This wide adaptation to different habitats makes them very useful tools when we want to design biosensors for the detection of environmental pollutants. The complexity of the biotic and abiotic elements that make up an ecosystem makes the design of an environmental biosensor more complex than those biosensors designed for industry or clinical use. The possible interactions of the microbial biosensor (CB or WCB) with unknown elements existing in a specific ecosystem, affecting the interpretation of the detection, are the main handicap of this type of biosensors. However, in spite of that, the majority of reported CB or WCB are based on prokaryotic or eukaryotic microorganisms (Verma and Singh 2005; Gutierrez et al. 2017; Metha et al. 2016). The main reason lies in the greater facility to obtain the sufficient quantity of a specific purified macromolecule (enzyme, antibody, etc.) or whole microorganisms for generating sufficiently quantifiable signal to design the biosensor. Therefore, it is easy to address using microorganisms which have, in general, a high growth rate. Microbial cultures are more manageable and cheaper than isolated enzymes, and the same enzyme, used as bioreceptor in a CB, presents more activity in the microbial cell due to the optimal microenvironment supplied by the cell itself (Verma and Singh 2005). Many microorganisms can be easily manipulated and grown on a wide variety of different media or culture types. Likewise, many of them can be genetically modified to facilitate the biosensor design improving the sensibility (▶ Chap. 13, “Engineering Prokaryote Synthetic Biology Biosensors”; Wang et al. 2014, 2015) or specificity (Wang and Buck 2014; Wang et al. 2013) to a specific environmental pollutant. In the case of WCBs, this technological capacity is essential due to necessity to incorporate the transduction capacity into the cell. Also, as stated above, microorganisms are distributed all over the planet, occupying all known ecosystems, which constitutes a great advantage if the biosensor designer is looking for a particular microbial capability to detect a specific environmental pollutant. An example of this is offered by Ralstonia metallidurans, a bacterium adapted to toxic metals (Mergeay et al. 2003), colonizing industrial sediments, soils, or wastes with a high content of metal(loid)s. From the knowledge on metal-resistant mechanisms and their regulation obtained from this bacterium and other metal-resistant microorganisms, several types of biosensors detecting metals have been designed (Diels et al. 2009; Leth et al. 2002; Tseng et al. 2014). The existence of cellular resistance mechanisms against metals or xenobiotics is really important when we want to design a WCB to detect these pollutants, because the regulatory genes and promoters involved in these mechanisms can be used for the design of the biosensor. Among the microorganisms that can be used to design biosensors are both prokaryotes and eukaryotes. With regard to the microbial type used in WCBs to detect

460

J.-C. Gutiérrez et al.

metal(loid)s, about 85% of these are based on genetically modified bacteria (Magrisso et al. 2008), while a 15% are based on eukaryotes, and the majority of them are yeasts. However, among eukaryotic microorganisms, there is the possibility of using microbes from the three different taxonomic groups: fungi, microalgae, and protozoa (Gutierrez et al. 2017). The “eukaryotic” character is particularly important because, in general, environmental biosensors are geared toward detection of potential toxic pollutants affecting mainly eukaryotic organisms (including humans). The existence of a greater similarity of metabolism, genome, and cell organization of eukaryotic cell-based biosensors with potential organisms (plants and animals) undergoing chemical pollution makes the comparison of detection level and toxicity safer and more effective. A similar situation is found among the number of biosensors for detecting xenobiotics; prokaryotic-based biosensors exceed eukaryotic one in a ratio of approximately 10:1.

Metal(loid) Biosensors To design metal(loid) CBs or WCBs, we can consider two basic types of microorganisms: the naturally existing or wild type and those genetically modified (anthropogenic origin). The wild-type strains usually present a peculiar feature which can be used to design the metal(loid) biosensor, such as bioluminescence, color, pigmentation, or any other characteristic that can be altered by metal(loid) toxicity. As previously indicated, these are the turn off bioassays. On the other hand, genetically modified microorganisms could be used in both turn off or turn on bioassays to design CBs or WCBs. After an analysis of the published works, in the last 25 years, on CBs and WCBs to detect metal(loid)s, using complete microorganisms as bioreceptors (CBs) or bioreceptor-transducers (WCBs), we have seen that the number of WCBs is more than twice (78%) as much as CBs (22%) (Fig. 1a). It is probably due to the greater complexity of the multidisciplinary work involved in designing CBs. Regardless of the type of biosensor (CB or WCB), the frequency of using prokaryotes exceeds that of using eukaryotes. In CBs, the prokaryote/eukaryote ratio is 59% : 41%, while in WCBs, this ratio is 79%:21% (Fig. 1b). Although eukaryotes have some advantages as WCBs (Gutierrez et al. 2015), many specialists prefer to continue using bacteria because they are easier to grow and manipulate. Among WCBs, turn on bioassays are more abundant (87%) than turn off bioassays (13%) (Fig. 1c). This indicates that much more specific biosensors are being searched for the detection of a given compound than those that respond to general stress caused by any environmental pollutant.

Bacteria-Based Metal(loid) Biosensors In this section, we will review the metal(loid) biosensors designed using bacterial elements or complete bacteria. Electrochemical (amperometric or potentiometric)

19

Environmental Biosensors: A Microbiological View

461

Fig. 1 Statistical analysis of metal(loid) biosensors. CB classical or conventional biosensor, WCB whole-cell biosensor

transducers are the majority (60%) among bacterial CBs, in second place are optical transducers (30%), and in the last position are the acoustic transducers (10%). Among bacterial WCBs, bioluminescence-based transducers have been widely used (75%), while fluorescence and colorimetric-based transducers have been less used (about 17% and 8%, respectively). The following are some examples of CBs designed using bacteria. Immobilized whole bacteria, without genetic modifications, have been used in metal CBs, such as the bioluminescent bacterium Photobacterium phosphoreum on a cellulose nitrate membrane to detect chromium (Lee et al. 1992) or the cyanobacterium Anabaena torulosa embedded in a cellulose membrane to detect Cu(II), Pb(II), and Cd(II) (Wong et al. 2013). The presence of these toxic cations reduces the photosynthetic (cyanobacterium) or bioluminescent activity by changing the fluorescence or bioluminescence quenching of these cells. The release of photosynthetic oxygen is also inhibited under the metal(loid) presence, and the oxygen reduction can be detected by an oxygen electrode (Shing et al. 2008). Bacterial enzymes have been also used to design metal CBs, under the assumption that toxic metals can inhibit the enzyme activity and show a direct correlation between the enzyme inhibition rate and metal toxicity. Enzymes, such as alkaline phosphatase, glucose oxidase, or urease, among others, have been used to detect Cd(II), Pb(II), Zn(II), Ni(II), or Co(II) (Berezhetskyy et al. 2008), in addition to Hg(II), Ag(I), Cu(II), and Fe(III) (Guascito et al. 2008; Ilangovan et al. 2006; Samphao et al. 2012).

462

J.-C. Gutiérrez et al.

Genetically engineered bacteria have been mainly used to design metal WCBs. Here we describe some of the most relevant. Some WCBs presumably specific to one or two metals have been reported. To detect As(III), a sensing construct formed by the arsR gene promoter (arsenic-sensitive promoter) from Escherichia coli fused to mtrB gene (involved in metal reduction pathway) from Shewanella oneidensis, a bacterium with great capacity to reduce metal ions and living in environments with and without oxygen. Accordingly, the output signal is bioelectrochemical (Webster et al. 2014). Another WCB with a high specificity to detect As(III) is also based on the promoter of the operon ars from E. coli, with a Pars::arsR construct fused with the reporter gene gfp, so the signal measurement is fluorescent (Li et al. 2015). To improve the sensitivity of this WCB, the authors carried out directed evolution experiments, where mutants were selected after three generations obtaining strains with a 12-fold increase of their response to As. Likewise, this WCB was tested with environmental samples contaminated with As, which means a positive validation for this environmental biosensor. Other similar constructs based on the ars promoter and the arsR regulator gene, fused with different fluorescent or luminescent reporter genes, have been designed (Huang et al. 2015; Hu et al. 2010; Jia et al. 2019; Merulla and van der Meer 2016; Preveral et al. 2017; Sharma et al. 2013; Wan et al. 2019; Wang et al. 2014, 2015). Several recombinant strains from both Gram-positive (Staphylococcus aureus and Bacillus subtilis) and Gram-negative (E. coli and Pseudomonas fluorescens) bacteria have been constructed to express bioluminescence reporter genes (luciferase system) to be used as metal WCBs (Gutierrez et al. 2017; Ivask et al. 2009). Both turn off and turn on bioassays have been carried out using different bacterial strains of which five for detecting Cu(II) or Hg(II) were target metal specific, whereas eight other strains showed a lower specificity level, responding to diverse metals (Cd(II), Hg(II), Zn(II), and Pb(II)) (Gutierrez et al. 2017). Other multi-metal(loid) response WCBs using E. coli as bacterial recipient are those reported by Branco et al. (2013), Hou et al. (2015), and Kim et al. (2016). Another E. coli-based WCB for detecting chromate in environmental samples use the plasmid pCHRGFP1 containing the construct PchrB::GFP (fluorescence signal) and presumably do not respond to other metal(loid)s (Branco et al. 2013). The promoter PchrB was isolated from the Gram-negative aerobic bacterium Ochrobactrum tritici, which has been also used as WCB by these authors. Other nonclassical bacterial species used as WCB are the Gram-negative Salmonella enterica, which, like other E. coli-based WCBs, hosting the construct PgolB::GFP, has been used to detect Au (Cerminati et al. 2011). The soil bacterium R. metallidurans has been used as a WCB (turn on bioassay) to detect Cr (II) by using the lux reporter gene system (Corbisier et al. 1999). Another strain of R. metallidurans has been used as WCB to detect Ni(II) and Co(II) in soil samples, after transformation with the megaplasmid pMOL1550 containing the cnr operon promoter (Ni(II) and Co(II) resistance system) (Tibazarwa et al. 2000) fused to lux reporter gene (Tibazarwa et al. 2001). A WCB, to detect Cd(II) and Zn(II), has been designed using the cadA gene promoter from S. aureus (Yoon et al. 1991) fused to the firefly luciferase reporter gene into the plasmid pT0024, using cell chassis of both S. aureus and B. subtilis

19

Environmental Biosensors: A Microbiological View

463

(Tauriainen et al. 1998). In the cyanobacterium Synechococcus, the smt locus contains smtB (a trans-acting repressor) and the metal-regulated smtA (encoding a metallothionein). Depending on the environmental bioavailability of metal(loid)s, smtB gene controls the gene expression of smtA (Erbe et al. 1996). Based on this system and using the same cyanobacterium, a multi-metal(loid) response WCB has been designed, harboring the plasmid construct Smt:luxCDABE. This WCB presents a linear response to several metals, such as Cd(II), Cu(II), Zn(II), Co(II), Hg(II), and Ag(I), with sensitivity ranking of Hg > Cu > Ag > Co > Zn > Cd (Martin-Betancor et al. 2015). Other bacterial genetic constructs have improved both specificity and metal detection and quantification (Bernard and Wang 2017; Wan et al. 2019; Wang et al. 2013). For instance, a specific Cu-WCB based on an E. coli strain containing a plasmid with the construct PcopA::GFP has been reported (Kang et al. 2018). This bacterial strain has a deletion affecting the copA gene, which is involved in cellular copper export, so this mutation makes this strain more sensitive to Cu(II) than the wild type. Under small Cu amounts (μM), this WCB responds specifically to Cu, no responding to other metals. Dual sensing WCBs are another very useful monitoring possibility as it can detect two different metals with different sensitivity. A dual sensing WCB based on E. coli has been designed to detect bioavailable As(III) or As(V) and Cd(II) in polluted soil samples (Yoon et al. 2016a). It carries two different genetic constructs: Pars:: mCherry (a monomeric red fluorescent protein) and Pznt::eGFP (enhanced GFP version) or the cross-combination fusions Pars::eGFP and Pznt::mCherry. The first construct responds to As(III) mainly, but also to As(V), while the second one responds to Cd(II). This dual WCB was seen to respond mainly and simultaneously As and Cd, but no against other metal(loid)s (Yoon et al. 2016a). In ecotoxicology, it is important to monitor pollutants both before and during its remediation. A novel strategy to test arsenic bioavailability in soil samples by direct (in vivo) or indirect (in vitro) measurement using an E. coli-based WCB has been reported by Yoon et al. (2016b). As previously reported, the plasmid construct has Pars::eGFP, which shows a high specificity to As. This WCB was applied to detect As(III) and As(V) in both laboratory and polluted soil samples. The authors (Yoon et al. 2016b) also described a standardized protocol to measure bioavailable As in soil samples. Hence this WCB was seen to be a useful tool to evaluate the efficiency of soil arsenic remediation processes. Likewise, metal(loid) environmental pollution monitoring is as important as bioremediation of the pollutant. An ideal situation would be to have a biosensor to detect a specific pollutant but also to bioremediate that same pollutant. These are WCBs based on two-component regulatory systems. The bacterial two-component systems have been engineered as synthetic biotechnological platforms for both monitoring and bioremediation purposes (Ravikumar et al. 2017). These WCBs simultaneously sense and remove (by bioadsorption) metals from the environment. For instance, detection and removal of Cu2+ ions in the E. coli surface have been achieved using the two-component system CusSR which allows for exogenous copper detection via membrane-associated kinase (Ravikumar et al. 2011). The CusSR two-component is a regulatory system for copper homeostasis in E. coli.

464

J.-C. Gutiérrez et al.

The gene ompC codes an outer membrane pore protein (porin), which is induced by high osmolarity and temperature. CusSR system activates the expression of CusCFBA operon, and CusC gene encodes a copper-binding peptide (CBP). Authors (Ravikumar et al. 2011) obtained a bacterial adsorption system constructed by the integration of chimeric OmpC::cusC under the PcusC promoter located in the plasmid pCC1056, to respond to and adsorb exogenous copper. Other two-component regulatory system-based microbial biosensors coupled with bioadsorption have been designed for metal sensing and bioremediation or biorefinery of some organic compounds (Ravikumar et al. 2017).

Bacterial-WCBs Based on Riboswitches Riboswitches are regulatory elements (noncoding sequences) within a mRNA molecule, able to bind different metabolites (purines and derivatives, cofactors, amino acids) and metal ions (Mg(II), Ni(II), Co(II)) as ligands. They regulate mRNA expression by transcription termination/anti-termination or translation inhibition or activation. The mRNA regulatory or switching sequence is most often located in the 5´-UTR, as a stretch of RNA preceding the translation starting site, although in some eukaryotic mRNAs, the riboswitch regulates splicing at the 3´-UTR (Edwards and Batey 2010). Riboswitches have two domains; the aptamer domain (acting as a receptor that specifically binds a ligand) and the expression platform or coding region (ORF: open reading frame). A riboswitch can adopt different secondary structures effecting gene expression depending on whether a target ligand is bound. When ligand is not bound, the expression platform incorporates the switching sequence into an anti-terminator stemloop, and transcription starts through the mRNA coding region. But, when ligand binds, the switching sequence is incorporated into the aptamer domain, and the expression platform folds into a terminator stem-loop, inducing transcription stop. In addition, mRNA translation activated by riboswitches (Serganov and Nudler 2013). The expression of ORFs can be repressed by sequestration of the ribosome entry site, blocking the translation initiation. When a ligand is attached to a riboswitch, the formation of an anti-sequester hairpin can be facilitated, opening the entrance to the ribosome and initiating the translation of the mRNA. Riboswitches are very common in Gram-positive bacteria, where they control the expression of around 4% of the genes (Lünse et al. 2011). Recently (Machtel et al. 2016), riboswitches have been considered as a new tool for WCB design. We can select a riboswitch responding to a specific metabolite or ligand through the in vitro selection process known as SELEX (systematic evolution of ligands by exponential enrichment). This experimental methodology originates the aptamer domain of a riboswitch responding to almost any metabolite (including metals). These riboswitches bind ligands even in very low concentrations in a dose-dependent manner (Fig. 2). Although it is known that riboswitches binding metal cations and anions exist (McCown et al. 2017; Wedekind et al. 2017), a WCB has not yet been built to

19

Environmental Biosensors: A Microbiological View

465

Fig. 2 Schematic of a riboswitch-based cell biosensor and its response to an increasing ligand concentration. GFP green fluorescent protein, AUG start codon, UAA stop codon

detect metal(loid) based on these molecular sensing systems. However, we know the existence of WCBs to detect organic compounds (drugs or xenobiotics), which will be discussed in the corresponding section. Riboswitches as molecular tools to design WCBs for detecting both metal(loid)s or xenobiotic compounds appear as a promising area in biosensorization of environmental pollutants (Findei et al. 2017).

Eukaryotic Microorganism-Based Metal(loid) Biosensors As previously indicated, metal WCBs based on eukaryotic microorganisms are more scarce than those using prokaryotic ones. The main eukaryotic microorganisms used to design both CBs or WCBs to detect metal(loid)s in polluted environments have been selected among microalgae, fungi (including yeasts), and ciliated protozoa (Gutierrez et al. 2017). To detect metal(loid)s, all reported CBs using eukaryotic microorganisms are electrochemical, while an 83% of WCBs are based on bioluminescence and the rest (17%) use fluorescence. The great majority of eukaryotic microbial biosensors use yeasts as the cellular component (yeast-based WCBs are reviewed by Adeniran et al. 2015). Second are microalgae (mainly genera Chlorella

466

J.-C. Gutiérrez et al.

and Chlamydomonas), followed by protozoa which is represented only by the ciliate Tetrahymena thermophila (Amaro et al. 2011, 2014). The yeast-model Saccharomyces cerevisiae is the most widely used eukaryotic microorganism in diverse biological areas, especially in genetic and bioengineering, which leads certain researchers (Walmsley and Keenan 2000) to consider that they can be good candidates to design biosensors for natural polluted environmental samples. However, like bacteria, yeasts have a cell wall protecting the cell and acting as a selective barrier for very different molecules (including substrates used by the biosensor transducer system in WCBs), which makes the transducer signal difficult to be efficiently emitted. For this reason, in some occasions, it is necessary to increase the cell permeability before using them as WCBs or bioreceptor in CBs, which constitutes an additional disadvantage. Mutants with enhanced cell permeability could be used for this purpose (Terziyska et al. 2000; Walmsley and Keenan 2000). The majority of yeast-based biosensors use the CUP1 promoter of the cup1 gene, encoding a copper-binding metallothionein (CUP1), which are used as biosensors for detecting copper. Some examples are given below. A recombinant S. cerevisiae strain has been used to design an amperometric CB (Lehmann et al. 2000) to detect Cu(II). Another amperometric CB using other different recombinant S. cerevisiae strain to detect Cu(II) has been constructed using the lacZ reporter gene (Tag et al. 2007). A biosensor to detect Cu(II) ions in water samples using genetically modified S. cerevisiae within immobilized alginate beads and based on the CUP1 promoter has been elaborated by Vopálenská et al. (2015). To validate this WCB, authors compare the Cu(II) measurements by the biosensor from urban water samples with those results obtained by standard laboratory assays using the same water samples. The transcriptional activator protein AceI present in S. cerevisiae has been used to control the expression of the reporter gene gfp (Shetty et al. 2004). When Cu(II) ions are present, the AceI protein activates the cup1 gene promoter fused with the gfp gene. This WCB is not selective for Cu(II) because it also responds to Ag(II) (Shetty et al. 2004). Another similar WCB, also for Cu(II) detection, has been constructed using the same CUP1 promoter but with the luciferase reporter gene, showing a similar detection level for this metal (Roda et al. 2011). This yeast has been also used to design biosensor for detecting other metals. For instance, the Cd-inducible SEO1 promoter from the yeast Hansenula polymorpha was fused to gfp gene, which was introduced in S. cerevisiae cells to detect Cd(II) ions (Park et al. 2007). This reporter construct is not specific to Cd(II) because it is also inducible by As(III). Likewise, the SEO1 promoter from S. cerevisiae revealed that it is inducible by multiple metal ions with sensitivity rank of As(III) > Cd(II) > Hg(II), hence also unspecific for Cd(II). A WCB based on an engineered S. cerevisiae strain responds to several metal(loid)s (Radhika et al. 2005). This yeast strain harbors a plasmid with the human CREBBP gene (encoding a binding protein to CREB gene, a cAMP response element binding that acts as a transcription factor) and the construct of Pcreb::gfp. When the yeast sensor is under the toxic metal exposure, a stress pathway is activated and cAMP levels increase. cAMP activates the CREBBP gene expression, and, in turn, the CREB protein activates the CREB promoter to express the GFP reporter gene to generate the output signal.

19

Environmental Biosensors: A Microbiological View

467

Microalgae are also significant microorganisms to design biosensors for aquatic ecosystem (Kröger and Law 2005). Chlorella vulgaris immobilized cells have been used to design a conductive CB based on the alkaline phosphatase activity inhibition to detect Cd(II) ions (Chouteau et al. 2004). This same microalga was used as a WCB for detecting Cd(II) in a water suspension or immobilized cells in a translucent silica matrix. The Cd(II) toxicity affected the photosynthetic activity (turn off bioassay) resulting a quenching of cell fluorescence (Nguyen-Ngoc et al. 2009). For monitoring Cu(II) in water supplies, the Chlorophyta Dictyosphaerium chlorelloides has been used with an optical fiber coupled to the cellular flow or a microtiter plate reader (PeñaVazquez et al. 2010). The flagellar motility of Chlamydomonas reinhardtii has been used to design an electrochemical biosensor to detect Cu(II) or Ni(II) (Shitanda et al. 2005). Unlike yeasts, the lack of usable genetic tools for bioengineering many microalgae has likely limited the construction of WCBs, perhaps the main reason to use wild-type strains as the bioreceptor elements in both CBs and WCBs. Although C. reinhardtii is an exception, this microalga has not yet been genetically modified to design biosensors for environmental metal(loid) monitoring, even though diverse studies have already been carried out to assess metal toxicity with this microorganism (Aksmann et al. 2014; De Schamphelaere et al. 2014). Considering microalgae present sufficient qualities to be selected as good potential metal(loid) biosensors, we can conclude that this micro-algal biotechnological aspect has not yet been adequately exploited. Ciliated protozoa have been extensively used in ecotoxicological studies (Gutierrez et al. 2008). With respect to the cell sensitivity level and biosensing toxicity for humans, ciliates have, at least, two additional advantages with regard to other microorganisms. In first place, unlike bacteria, yeasts, or microalgae, ciliates do not have a cell wall in their growth vegetative stage. As it has been indicated in previous reviews (Gutierrez et al. 2015, 2017), microorganisms with a cell wall to be used as WCB or bioreceptor elements of CBs have an important limitation since the selective diffusion of molecules through their cell walls results in a lower emission or less effective cell response to pollutants. In several cases, to address this barrier, cells have had to be permeabilized by physicochemical or enzymatic methods. In addition to this difficulty, the presence of a cell wall may involve a non-specific, uncontrolled metal(loid) biosorption process, which may affect the real cellular response to the external metal concentration, when cells are not used as a biosorption-based biosensor. This problem might be solved using ciliates in the biosensor designing, because the absence of cell walls leads to a high sensitivity and rapid response to a variety of environmental pollutants (Martin-Gonzalez et al. 1999; Gutierrez et al. 2003). In second place, ciliates are eukaryotic cells with an animal biology, which present a series of metabolic traits more similar to those of human cells than bacteria, microalgae, or even yeasts. The genome sequencing projects of two ciliate models, Tetrahymena thermophila and Paramecium tetraurelia (Aury et al. 2006; Eisen et al. 2006), have shown that they share a higher degree of functional conservation with human genes than do other eukaryotic microbial models. For instance, humans and T. thermophila share more ortholog genes with each other (about 2280) than are

468

J.-C. Gutiérrez et al.

shared between humans and the yeast S. cerevisiae (Eisen et al. 2006). Likewise, the scores of P. tetraurelia proteins against human proteins are the highest with regard to the scores of yeast proteins to human proteins, suggesting that Paramecium proteins are most similar to human proteins (Sperling et al. 2002). These similarities with human biology makes it more reasonable to use these eukaryotic microorganisms in ecotoxicological studies (Gutierrez et al. 2008, 2011) or to design biosensors for monitoring toxicity of metal(loid)s or xenobiotic compounds for humans, in aquatic or terrestrial ecosystems. Unfortunately, there are few currently published ciliate-based biosensors. Only the ciliate T. thermophila has been used to design WCBs for metal(loid)s, which have been validated using natural (soil and aquatic) metal polluted environmental samples (Amaro et al. 2011). These turn on WCBs have been designed using promoters of MTT1 or MTT5 metallothionein genes from this ciliate and the firefly luciferase as the reporter gene. These lineal constructs were then introduced into macronuclear genome by biolistic transformation (Amaro et al. 2011). A second type of T. thermophila WCB has been designed using MTT1 gene promoter and the gfp reporter gene fused to the MTT1 or MTT5 complete ORF into a plasmid (Amaro et al. 2014). A comparative analysis of both metal WCBs is reported in Gutierrez et al. (2017). Although the three types of eukaryotic microorganisms (yeasts, microalgae, or protozoa) can be used to design both CBs or WCBs for metal(loid) environmental monitoring, they all have certain advantages and disadvantages, which have been previously discussed (Gutierrez et al. 2017).

Xenobiotic Biosensors Like metal(loid) biosensors, xenobiotic biosensors can be designed using both wildtype and genetically modified microorganisms. The great diversity of microbial capabilities to biodegrade or biotransform xenobiotics is the basis to design biosensors to detect these toxic compounds. In the design of these biosensors, prokaryotes predominate over eukaryotes probably due to the greater capacity of bacteria to degrade very different organic compounds compared to eukaryotes. Unlike what happens with metal biosensors, CBs to detect xenobiotics are the majority (88%) against WCBs (12%). The common physicochemical transducer used in these xenobiotic CBs is electrochemical. Due to the scarcity of xenobiotic biosensors based on eukaryotic microorganisms, in this section, we will explore exclusively those biosensors based on prokaryotes.

Bacteria-Based Xenobiotic Biosensors Like metal(loid) biosensors, bacterial two-component systems have been used to design xenobiotic biosensors. Aromatic compounds are very abundant organic environmental pollutants. Many bacteria can feed on these organic compounds as a carbon source. Several two-component regulatory systems are involved in the

19

Environmental Biosensors: A Microbiological View

469

catabolism of aromatic compounds, by inducing and activating the corresponding aromatic metabolizing pathway. Pseudomonas putida degrades toluene and other benzene derivatives; the first four steps of the catabolic pathway involve the sequential action of seven tod genes, which form part of the tod operon (Zylstra et al. 1988). This operon can be induced by different aromatic compounds (toluene, xylene, benzene, ethylbenzene) and is modulated by a two-component regulatory system. Engineered microbial biosensors based on bacterial two-component systems, like P. putida among others, can be very useful to construct synthetic biotechnology platforms for monitoring, bioremediation, and biorefinery (Ravikumar et al. 2017). Recently, microbial fuel cell-based biosensors have attracted great interest due to their sustainability, low cost, and applications including anaerobic digestion process monitoring or water quality detection (Zhou et al. 2017). The voltage originated by microbial fuel cells is correlated with the amount of a specific substrate, presenting a linear correlation. Several studies have reported that these biosensors can detect organic compounds, such as p-nitrophenol, formaldehyde, or the antibiotic levofloxacin. An important limitation of these biosensors is their low sensitivity, as the detection range is above the currently allowed contaminant level. Therefore, an improvement in sensitivity will be necessary for their widespread application. Riboswitch-based WCB for bisphenol A (BPA), a known endocrine disruptor and potential carcinogen present in plastics, food packaging, and drinking water supplies, has been tried in E. coli (Zorawski et al. 2016). TetA is a transporter protein that pumps tetracycline out of the cell, making the cell resistant to this antibiotic. In addition, this protein allows Ni(II) to enter the cell, inhibiting growth. tetA gene expression is regulated by a riboswitch. The construct PTetA::tetA::gfp was introduced into a plasmid, and then the aptamer domain of the riboswitch was replaced, by PCR, with 40 random bases to generate plasmid libraries. E. coli was transformed using these plasmids such that each bacterium harbored only a copy of the plasmid library. A selection was carried out to identify riboswitches expressing TetA-GFP only in response to BPA. Unfortunately, the authors failed to produce a significant increase in output fluorescence under the presence of BPA compared to control (without ligand) (Zorawski et al. 2016). However, this pioneering work could be a first step in the future construction of WCBs using riboswitches to detect xenobiotic molecules. A selection of research studies reporting xenobiotic biosensors is summarized as follows. CBs using different bacterial species (Pseudomonas, Sphingomonas, Ralstonia, and Rhodococcus) immobilized onto a Clark-type oxygen electrode were tested for monitoring the cellular degradation capacity and detection of several xenobiotic compounds especially their chlorinated derivatives (chlorophenols, chlorobenzoates, PCBs, among others) (Beyersdorf-Radeck et al. 1998). This study showed that each bacterium prefers selectively a type of substrate, suggesting for each xenobiotic biosensor a specific type of bacteria should be selected in advance. A bioluminescent E. coli strain carrying the lux operon and the promoter region of the pgi gene (encoding the glucose-6-phosphate isomerase), which responds to oxidative stress, was used for the construction of a bioluminescent WCB (Niazi et al. 2008). This biosensor is designed on the basis that xenobiotics can induce cell

470

J.-C. Gutiérrez et al.

damage by oxidative stress, such as the herbicide paraquat (Diaz et al. 2016), causing oxygen free radicals. The linA gene encoding the enzyme γ-hexaclorocyclohexane dehydrochlorinase, involved in biodegradation of the pesticide lindane, was cloned and overexpressed in E. coli. Subsequently, cells from this lindane biodegradating strain were immobilized on polyaniline film. The hydrochloric acid generation from the lindane biodegradation by the recombinant E. coli cells led to a change in the conductivity of the polyaniline film, which was monitored amperometrically (Prathap et al. 2012). This CB detects lindane at a concentration of one part per trillion, and was selective to all lindane isomers, but did not respond to other aliphatic, aromatic chlorides or the end product of lindane degradation. Another possible technical approach for designing WCBs to detect xenobiotics is the use of immobilized cells in microplates. Whole cells of the bacterium Sphingomonas sp., which hydrolyzes the methyl parathion (an organophosphate pesticide and insecticide) into p-nitrophenol (yellow-colored product), were immobilized onto the surface of the polystyrene microplate wells. Para-nitrophenol was detected by a colorimetric method (microplate reader) (Kumar and D’souza 2010). Microplate-based biosensors constitute a convenient system to detect multiple numbers of samples in a single platform. Organophosphate pesticides are very serious environmental pollutants. Amperometric, potentiometric, and optical CBs based on engineered microorganisms expressing organophosphate-hydrolyzing enzymes, either located intracellularly or anchored to the cell surface, have been designed to detect these dangerous pesticides (Lei et al. 2007). The advantages of using microorganisms over purified enzymes are lower cost and simpler construction. There are many other examples of potential biosensors for detecting xenobiotics, although most are based on the ability of bacteria to degrade or biotransform these molecules.

Conclusions and Future Perspectives 1. In general, microorganisms used as whole cells or cell factories to design biosensors (CBs or WCBs) for detecting environmental pollutants present greater advantages than using molecules (such as enzymes). In addition, eukaryotic microorganisms have certain advantages over prokaryotic microorganisms, because the comparative analysis with multicellular organisms (including humans) is more reliable than using bacteria. 2. Likewise, protozoa, and mainly ciliates, present a great potential to design both WCBs and CBs, mainly due to its greater sensitivity to environmental pollutants without a cell wall and also due to their greater resemblance to human cells. 3. In almost all prokaryotic and many eukaryotic microorganisms, the presence of a cell wall is a hurdle or inconvenience in the design of a biosensor, because it could hinder permeability or retain (biosorption) pollutant molecules. In addition, using substrate-dependent reporters, substrates must cross the cell wall to reach the

19

Environmental Biosensors: A Microbiological View

471

cytoplasm, where the enzymatic reaction takes place. Therefore, it is sometimes necessary to pre-treat the cell to make it permeable to the molecule. 4. With respect to metal(loid) contaminants, the sensitivity of the biosensor is more important than its level of specificity to the target metal. This is because, in the real world, anthropogenic contamination by metal(loid)s is complex with the presence of several metal(loid)s. 5. In general, many CBs or WCBs are exclusively tested in the laboratory, under strictly controlled conditions, but very few of those published are validated using natural samples from metal(loid) or xenobiotic polluted environments. Due to this lack of essential experimentation, to evaluate a biosensor’s usefulness for detecting an environmental pollutant, many of the biosensors considered as specific for detecting a certain pollutant molecule are not really so. It is due to the presence of other unknown organic or inorganic molecules, from the soil or aquatic ecosystem, which can interact with the bioreceptor element and interfere with the response. This point is really important for designing functional biosensors to be used in polluted environments in the real world. 6. The future development of biosensors (CBs or WCBs) for the monitoring of environmental pollutants could be developed on the basis of three, still little explored, experimental routes: (a) A greater and more intense exploration of prokaryotic riboswitches, to experimentally select those that interact with metals or xenobiotics. This exploration should also extend to eukaryotic microorganisms. (b) Enabling the same biosensor with both capabilities: the monitoring of the environmental pollutant and its bioremediation. For this purpose, two-component systems could be essential parts in the design of these bifunctional systems (biosensor-bioremediator). (c) A synthetic biology approach. This could facilitate the design of environmental biosensor (mainly WCBs) with multi-input systems based on two or more regulatory gene promoters in the same genetic construct, thereby increasing the capacity of the biosensor to detect simultaneously several different organic or inorganic pollutants in the same environmental sample (Wang et al. 2013, 2014).

References Adeniran A, Sherer M, Tyo KEJ (2015) Yeast-based biosensors: design and applications. FEMS Yeast Res 15:1–15 Aksmann A, Pokora W, Bascik-Remisiewicz A et al (2014) Time-dependent changes in antioxidative enzyme expression and photosynthetic activity of Chlamydomonas reinhardtii cells under acute exposure to cadmium and anthracene. Ecotoxicol Environ Saf 110C:31–40 Amaro F, Turkewitz AP, Martin-Gonzalez A et al (2011) Whole-cell biosensors for detection of heavy metal ions in environmental samples based on metallothionein promoters from Tetrahymena thermophila. Microb Biotechnol 4:513–522 Amaro F, Turkewitz AP, Martin-Gonzalez A et al (2014) Functional GFP-metallothionein fusion protein from Tetrahymena thermophila: a potential whole-cell biosensor for monitoring heavy

472

J.-C. Gutiérrez et al.

metal pollution and a cell model to study metallothionein overproduction effects. Biometals 27:195–205 Aury JM, Jaillon O, Duret L et al (2006) Global trends of whole-genome duplications revealed by the ciliate Paramecium tetraurelia. Nature 444:171–178 Belkin S (2003) Microbial whole-cell sensing systems of environmental pollutants. Curr Opin Microbiol 6:206–212 Berezhetskyy AL, Sosovska OF, Durrieu C et al (2008) Alkaline phosphatase conductometric biosensor for heavy metal ions determination. IRBM 29:136–140 Bernard E, Wang B (2017) Synthetic cell-based sensors with programmed selectivity and sensitivity. In: Rasooly A, Prickril B (eds) Biosensors and biodetection: methods and protocols. Methods in molecular biology, vol 1572. Springer, New York, pp 343–363 Beyersdorf-Radeck B, Karlson KR, Bachmann TT et al (1998) Screening of xenobiotic compounds degrading microorganisms using biosensor techniques. Microbiol Res 153:239–245 Branco R, Cristovao A, Morais PV (2013) Highly sensitive, highly specific whole-cell bioreporters for the detection of chromate in environmental samples. PLoS One 8:e54005 Caceres JO, Sanz-Mangas D, Manzoor S et al (2019) Quantification of particulate matter, tracking the origin and relationship between elements for the environmental monitoring of the Antarctic region. Sci Tot Env 665:125–132 Cerminati S, Soncini FC, Checa SK (2011) Selective detection of gold using genetically engineered bacterial reporters. Biotechnol Bioeng 108:2553–2560 Chouteau C, Dzyadevych S, Chovelon JM et al (2004) Development of novel conductometric biosensors based on immobilized whole cell Chlorella vulgaris microalgae. Biosens Bioelectron 19:1089–1096 Corbisier P, van der Lelie D, Borremans B et al (1999) Whole cell-and protein-based biosensors for the detection of bioavailable heavy metals in environmental samples. Anal Chim Acta 387:235–244 De Schamphelaere KAC, Nys C, Janssen CR (2014) Toxicity of lead (Pb) to freshwater green algae: development and validation of a bioavailability model and inter-species sensitivity comparison. Aquat Toxicol 155:348–359 Diaz S, Martin-Gonzalez A, Cubas L et al (2016) High resistance of Tetrahymena thermophila to paraquat: mitochondrial alterations, oxidative stress and antioxidant genes expression. Chemosphere 144:909–917 Diels L, Van Roy S, Taghavi S et al (2009) From industrial sites to environmental applications with Cupriavidus metallidurans. A Leeuwenhoek 96:247–258 Edwards AL, Batey RT (2010) Riboswitches: a common RNA regulatory element. Nat Educ 3(9):9 Eisen JA, Coyne RS, Wu M et al (2006) Macronuclear genome sequence of the ciliate Tetrahymena thermophila, a model eukaryote. PLoS Biol 4:e286 Erbe JL, Adams AC, Taylor KB et al (1996) Cyanobacteria carrying an smt-lux transcriptional fusion as biosensors for the detection of heavy metal cations. J Ind Microbiol 17:80–83 Findei BS, Etzel M, Will S et al (2017) Design of artificial riboswitches as biosensors. Sensors 17:1–28. https://doi.org/10.3390/s17091990 Guascito MR, Malitesta C, Mazzotta E et al (2008) Inhibitive determination of metal ions by an amperometric glucose oxidase biosensor: study of the effect of hydrogen peroxide decomposition. Sens Actuators B Chem 131:394–402 Gutierrez JC, Martin-Gonzalez A, Diaz S et al (2003) Ciliate as potential source of cellular and molecular biomarker/biosensors for heavy metal pollution. Eur J Protistol 39:461–467 Gutierrez JC, Martin-Gonzalez A, Diaz S et al (2008) Ciliates as cellular tools to study the eukaryotic cell-heavy metal interactions. In: Brown SE, Welton WC (eds) Heavy metal pollution. Nova Science Publishers, New York, pp 1–44 Gutierrez JC, Amaro F, Diaz S et al (2011) Ciliate metallothioneins: unique microbial eukaryotic heavy-metal-binder molecules. J Biol Inorg Chem 16:1025–1034 Gutierrez JC, Amaro F, Martin-Gonzalez A (2015) Heavy metal whole-cell biosensors using eukaryotic microorganisms: an updated critical review. Front Microbiol 6:1–8

19

Environmental Biosensors: A Microbiological View

473

Gutierrez JC, Amaro F, Martin-Gonzalez A (2017) Microbial biosensors for metal(loid)s. In: CravoLaureau C et al (eds) Microbial ecotoxicology. Springer International Publishing AG, Cham, pp 313–336 Hill MK (2004) Understanding environmental pollution. A primer. Cambridge University Press, Cambridge Hou Q, Ma A, Wang T et al (2015) Detection of bioavailable cadmium, lead and arsenic in polluted soil by tailored multiple Escherichia coli whole-cell sensor set. Anal Bioanal Chem 407:6865–6871 Hu Q, Li L, Wang Y et al (2010) Construction of WCB-11: a novel phiYFP arsenic-resistant wholecell biosensor. J Environ Sci 22:1469–1474 Huang CW, Yang SH, Sun MW et al (2015) Development of a set of bacterial biosensors for simultaneously detecting arsenic and mercury in groundwater. Environ Sci Pollut Res Int 22:10206–10213 Huse SM, Welch DM, Morrison HG et al (2010) Ironing out the wrinkles in the rare biosphere through improved OUT clustering. Environ Microbiol 12:1889–1898 Ilangovan R, Daniel D, Krastanov A et al (2006) Enzyme based biosensor for heavy metal ions determination. Biotechnol Equip 20:184–189 Ivask A, Rolova T, Kahru A (2009) A suite of recombinant luminescent bacterial strains for the quantification of bioavailable heavy metals and toxicity testing. BMC Biotechnol 9:41. https:// doi.org/10.1186/1472-6750-9-41 Jia X, Bur R, Zhao T et al (2019) Development of a sensitive and specific whole-cell biosensor for arsenic detection. Appl Environ Microbiol 85:e00694 Jouanneau S, Durand MJ, Assaf A et al (2017) Bacterial bioreporter applications in ecotoxicology: concepts and practical approach. In: Cravo-Laureau C et al (eds) Microbial ecotoxicology. Springer International Publishing AG, Cham, pp 283–311 Jung Y, Park C-B, Kim Y et al (2015) Application of multi-species microbial bioassay to assess the effects of engineered nanoparticles in the aquatic environment; potential of a luminous microbial array for toxicity risk assessment (LumiRAMA) on testing for surface-coated silver nanoparticles. Int J Environ Res Public Health 12:8172–8186 Kallmeyer J, Pockalny R, Adhikari RR et al (2012) Global distribution of microbial abundance and biomass in subseafloor sediment. PNAS 109:16213–16216 Kang Y, Lee W, Kim S (2018) Enhancing the Cu-sensing capability of E. coli-based WCB by genetic engineering. Appl Microbiol Biotechnol 102:1513–1521 Kim HJ, Lim JW, Jeong H et al (2016) Development of a highly specific and sensitive cadmium and lead microbial biosensor using synthetic CadC-T7 genetic circuitry. Biosens Bioelectron 79:701–708 Kröger S, Law RJ (2005) Biosensors for marine applications. We all need the sea, but does the sea need biosensors? Biosens Bioelectron 20:1903–1913 Kumar J, D’souza SF (2010) An optical microbial for detection of methyl parathion using Sphingomonas sp. immobilized on microplate as a reusable biocomponent. Biosens Bioelectron 26:1292–1296 Lee S, Sode K, Nakanishi K et al (1992) A novel microbial sensor using luminous bacteria. Biosens Bioelectron 7:273–277 Lehmann M, Riedel K, Adler K et al (2000) Amperometric measurement of copper ions with a deputy substrate using a novel Saccharomyces cerevisiae sensor. Biosens Bioelectron 15:211–219 Lei Y, Mulchandani P, Chen W et al (2007) Biosensor for direct determination of Fenitrothion and EPN using recombinant Pseudomonas putida js444 with surface-expressed organophosphorous hydrolase. 2. Modified carbon paste electrode. Appl Biochem Biotechnol 136:243–250 Leonard SS, Harris GK, Shi X (2004) Metal-induced oxidative stress and signal transduction. Free Radic Biol Med 37(12):1921–1942 Leth S, Maltoni S, Simkus R et al (2002) Engineered bacteria based biosensors for monitoring bioavailable heavy metals. Electroanalysis 14:35–42

474

J.-C. Gutiérrez et al.

Li L, Liang J, Hong W et al (2015) Evolved bacterial biosensor for arsenite detection in environmental water. Environ Sci Technol 49:6149–6155 Lünse CE, Schmidt MS, Wittmann V (2011) Carba-sugars activate the glmS-riboswitch of Staphylococcus aureus. ACS Chem Biol 6:675–678 Machtel P, Bakowska-Zywicka K, Zywicki M (2016) Emerging applications of riboswitches-from antibacterial targets to molecular tools. J Appl Genet 57:531–541 Magrisso S, Erel Y, Belkin S (2008) Microbial reporter of metal bioavailability. Microb Biotechnol 1:320–330 Martin-Betancor K, Rodea-Palomares I, Muñoz-Martín MA (2015) Construction of a self-luminescent cyanobacterial bioreporter that detects a broad range of bioavailable heavy metals in aquatic environment. Front Microbiol 6:186. https://doi.org/10.3389/fmicb.2015.00186 Martin-Gonzalez A, Diaz S, Jareño C et al (1999) The use of protists in ecotoxicology. Recent Res Dev Microbiol 3:93–111 McCown PJ, Corbino KA, Stav S (2017) Riboswitch diversity and distribution. RNA 23:995–1011 Mergeay M, Monchy S, Vallaeys T et al (2003) Ralstonia metallidurans, a bacterium specifically adapted to toxic metals: towards a catalogue of metal-responsive genes. FEMS Microbiol Rev 27:385–410 Merulla D, van der Meer JR (2016) Regulatable and modulable background expression control in prokaryotic synthetic circuits by auxiliary repressor binding sites. ACS Synth Biol 5:36–45 Metha J, Bhardwaj SK, Bhardwaj N et al (2016) Progress in the biosensing techniques for trace – level heavy metals. Biotechnol Adv 34:47–60 Nguyen-Ngoc H, Durrieu C, Tran-Minh C (2009) Synchronous-scan fluorescence of algal cells for toxicology of heavy metals and herbicides. Ecotoxicol Environ Saf 72:316–320 Niazi JH, Kim BC, Ahn J-M et al (2008) A novel bioluminescent bacterial biosensor using the highly specific oxidative stress-inducible pgi gene. Biosens Bioelectron 24:670–675 Park JN, Sohn MJ, Oh DB (2007) Identification of the cadmium-inducible Hansenula polymorpha SEO1 gene promoter by transcriptome analysis and its application to whole-cell heavy-metal detection systems. Appl Environ Microbiol 73:5990–6000 Peña-Vazquez E, Perez-Conde C, Costas E et al (2010) Development of a microalgal PAM test method for Cu(II) in waters: comparison of using spectrofluorometry. Ecotoxicology 19:1059–1065 Peñuelas J, Fillela I (2002) Metal pollution in Spanish terrestrial ecosystems during the twentieth century. Chemosphere 46:501–505 Prathap MUA, Chaurasia AK, Sawant SN et al (2012) Polyaniline-based highly sensitive biosensor for selective detection of lindane. Anal Chem 15:6672–6678 Preveral S, Brutesco C, Descamps EC et al (2017) A bioluminescent arsenite biosensor designed for inline water analyzer. Environ Sci Pollut Res Int 24:25–32 Radhika V, Milkevitch M, Audigé V et al (2005) Engineered Saccharomyces cerevisiae strain BioS1, for the detection of water-borne toxic metal contaminants. Biotechnol Bioeng 90:29–35 Ravikumar S, Yoo IK, Lee SY et al (2011) Construction of cooper removing bacteria through the integration of two-component system and cell surface display. Appl Biochem Biotechnol 165:1674–1681 Ravikumar S, Baylon MG, Park SJ (2017) Engineered microbial biosensors based on bacterial twocomponent systems as synthetic biotechnology platforms in bioremediation and biorefinery. Microb Cell Factories 16:62. https://doi.org/10.1186/s12934-017-0675-z Roda A, Roda B, Cevenini L (2011) Analytical strategies for improving the robustness and reproducibility of bioluminescent microbial bioreporters. Anal Bioanal Chem 401:201–211 Samphao A, Rerkchai H, Jitcharoen J et al (2012) Indirect determination of mercury by inhibition of glucose oxidase immobilizated on carbon paste electrode. Int J Electrochem 7:1001–1010 Serganov A, Nudler E (2013) A decade of riboswitches. Cell 152:17–24 Shahsavari E, Aburto-Medina A, Khudur LD et al (2017) From microbial ecology to microbial ecotoxicology. In: Cravo-Laureau C et al (eds) Microbial ecotoxicology. Springer International Publishing AG, Cham, pp 17–38

19

Environmental Biosensors: A Microbiological View

475

Sharma P, Asad S, Ali A (2013) Bioluminescent bioreporter for assessment of arsenic contamination in water samples of India. J Biosci 38:251–258 Shetty RS, Deo SK, Liu Y et al (2004) Fluorescence-based sensing system for copper using genetically engineered living yeast cells. Biotechnol Bioeng 88:664–670 Shing WL, Surif S, Heng LY (2008) Toxicity biosensor for the evaluation of cadmium toxicity based on photosynthetic behavior of cyanobacteria Anabaena torulosa. Asian J Biochem 3:162–168 Shitanda I, Takada K, Sakai Y et al (2005) Amperometric biosensing systems based on motility and gravitaxis of flagellate algae for aquatic risk assessment. Anal Chem 77:6715–6718 Sperling L, Dessen P, Zagulski M et al (2002) Random sequencing of paramecium somatic DNA. Eukaryot Cell 1:341–352 Tag K, Riedel K, Bauer HJ et al (2007) Amperometric detection of Cu2+ by yeast biosensors using flow injection analysis (FIA). Sens Act B Chem 122:403–409 Tauriainen S, Karp M, Chang W et al (1998) Luminescent bacterial sensor for cadmium and lead. Biosens Bioelectron 13:931–938 Terziyska A, Waltschewa L, Venkov P (2000) A new sensitive test based on yeast cells for studying environmental pollution. Environ Pollut 109:43–52 Tibazarwa C, Wuertz S, Mergeay M et al (2000) Regulation of the cnr cobalt and nickel resistance determinant of Ralstonia eutropha (Alcaligenes eutrophus) CH34. J Bacteriol 182:1399–1409 Tibazarwa C, Corbisier P, Mench M et al (2001) A microbial biosensor to predict bioavailable nickel in soil and its transfer to plants. Environ Pollut 113:19–26 Tseng H-W, Tsai Y-J, Yen J-H, Chen P-H et al (2014) A fluorescence-based microbial sensor for the selective detection of gold. Chem Commun 50:1735–1737 Valko M, Morris H, Cronin MT (2005) Metals, toxicity and oxidative stress. Curr Med Chem 12:1161–1208 Van der Meer JR, Belkin S (2010) Where microbiology meets microengineering: design and application of reporter bacteria. Nat Rev Microbiol 8:511–522 Verma N, Singh M (2005) Biosensors for heavy metals. Biometals 18:121–129 Vopálenská I, Váchová L, Palková Z (2015) New biosensor for detection of copper ions in water based on immobilized genetically modified yeast cells. Biosens Bioelectron 72:160–167 Walmsley RM, Keenan P (2000) The eukaryotic alternative: advantages of using yeasts in place of bacteria in microbial biosensor development. Biotechnol Bioprocess Eng 5:387–394 Wan X, Volpetti F, Petrova E, French C, Maerkl SJ, Wang B (2019) Cascaded amplifying circuits enable ultrasensitive cellular sensors for toxic metals. Nat Chem Biol 15:540–548 Wang B, Buck M (2014) Rapid engineering of versatile molecular logic gates using heterologous genetic transcriptional modules. Chem Commun 50:11642–11644 Wang B, Barahona N, Buck M (2013) A modular cell-based biosensor using engineered genetic logic circuits to detect and integrate multiple environmental signals. Biosens Bioelectron 40:368–376 Wang B, Barahona N, Buck M (2014) Engineering modular and tunable genetic amplifiers for scaling transcriptional signals in cascaded gene networks. Nucleic Acids Res 42:9484–9492 Wang B, Barahona N, Buck M (2015) Amplification of small molecule-inducible gene expression via tuning of intracellular receptor densities. Nucleic Acids Res 43:1955–1964 Webster DP, TerAvest MA, Doud DF et al (2014) An arsenic-specific biosensor with genetically engineered Shewanella oneidensis in a bioelectrochemical system. Biosens Bioelectron 62:320–324 Wedekind JE, Dutta D, Belashov IA et al (2017) Metalloriboswitches: RNA-based inorganic ion sensors that regulate genes. J Biol Chem 292:9441–9450 Wong LS, Lee YH, Surif S (2013) Whole cell biosensor using Anabaena torulosa with optical transduction for environmental toxicity evaluation. J Sens 567272:1. https://doi.org/10.1155/ 2013/567272 Yagi K (2007) Applications of whole-cell bacterial sensors in biotechnology and environmental science. Appl Microbiol Biotechnol 73:1251–1258

476

J.-C. Gutiérrez et al.

Yoon KP, Misra TK, Silver S (1991) Regulation of the cadA cadmium resistance determinant of Staphylococcus aureus plasmid pI258. J Bacteriol 173:7643–7649 Yoon Y, Kim S, Chae Y (2016a) Simultaneous detection of bioavailable As and Cd in contamined soils using dual-sensing bioreporters. Appl Microbiol Biotechnol 100:3713–3722 Yoon Y, Kim S, Chae Y (2016b) Evaluation of bioavailable Ar and remediation performance using a WCB. Sci Total Environ 547:125–131 Zhou T, Han H, Liu P et al (2017) Microbial fuels cell-based biosensor for toxicity detection: a review. Sensors 17:2230. https://doi.org/10.3390/s17102230 Zorawski M, Shaffer J, Velasquez E et al (2016) Creating a riboswitch-based whole-cell biosensor for bisphenol A. FASEB. https://doi.org/10.1096/fasebj.30.1.supplement.805.3 Zylstra GJ, McCombie WR, Gibson DT et al (1988) Toluene degradation by Pseudomonas putida F1: genetic organization of the tod operon. Appl Environ Microbiol 54:1498–1503

Part III Transducers, Materials, and Systems

Live Cell Immobilization

20

Antonia Lopreside, Maria Maddalena Calabretta, Laura Montali, Aldo Roda, and Elisa Michelini

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Covalent Attachment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Adsorption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Encapsulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Entrapment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cell Immobilization on 3D-Printed Matrices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Whole-Cell Biosensors Based on Immobilized Microbial Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Whole-Cell Biosensors Based on Immobilized Mammalian Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

480 484 485 485 486 487 487 490 492 492

Abstract

Maintenance of cell viability and physiological cellular activity is crucial in several applications, spanning from microbial fermentation to produce bulk and fine chemicals to whole-cell biosensing. To this end a plethora of cell immobilization approaches and techniques were explored to entrap living cells and A. Lopreside · M. M. Calabretta · L. Montali Department of Chemistry “G. Ciamician”, University of Bologna, Bologna, Italy A. Roda Department of Chemistry “G. Ciamician”, University of Bologna, Bologna, Italy INBB, Istituto Nazionale di Biostrutture e Biosistemi, Rome, Italy E. Michelini (*) Department of Chemistry “G. Ciamician”, University of Bologna, Bologna, Italy INBB, Istituto Nazionale di Biostrutture e Biosistemi, Rome, Italy Health Sciences and Technologies-Interdepartmental Center for Industrial Research (HST-ICIR), University of Bologna, Bologna, Italy e-mail: [email protected] © Springer Nature Switzerland AG 2022 G. Thouand (ed.), Handbook of Cell Biosensors, https://doi.org/10.1007/978-3-030-23217-7_146

479

480

A. Lopreside et al.

maintain their functionality for long periods of time, each one having peculiar advantages and limitations. Several matrices have been developed and optimized for each application and cell type. Thanks to new available technologies, such as affordable 3D printing and micro- and nano-fabrication, new supports and scaffolds are now available with advanced functionalities. Although it is far from being exhaustive, this chapter aims at providing a glance on this challenging field with an overview on the main approaches and technologies, highlighting the most promising methods which will presumably play major roles in the next years. Keywords

Cell immobilization · Cell entrapment · Cell encapsulation · Cell biosensor · 3D printing

Introduction A major issue in modern biotechnology is the development of methods to immobilize living cells without losing their physiological or functional activity. Immobilization occurs in nature, Radwan et al. have proved that microalgal samples collected in Gulf coast were covered by a biofilm of oil and bacteria able to degrade hydrocarbons found in seawater (Radwan et al. 2002). The past decade has witnessed an increasing interest in long-term storage of living cells, including microorganisms and mammalian cell lines, with tremendous potential in several fields such as tissue engineering, biosensing, and industrial bioproductions. Live cell immobilization techniques were first developed in the biofermentation industry for recombinant production of bulk and fine chemicals. In this context immobilization of single enzymes provided suitable tool for single-step reactions. However, when more complex reactions and the presence of co-factors are required, other approaches are needed in which either whole cells or cellular organelles are immobilized to serve as multienzyme factories. In addition, due to the high cost for enzyme purification, immobilized cells rather than pure enzymes are generally preferred even for recombinant productions involving single enzyme reactions. Another advantage derives from the fact that the cell microenvironment provides optimal conditions for enzyme activity, thus providing higher yields within shorter fermentation times. Several processes, technologies, materials, and surfaces have been explored to improve cell immobilization strategies in different fields such as cell biosensors, industrial process for bioconversion, bioproduction, de novo synthesis and synthesis from precursors, and bioengineering for cell regeneration and regenerative medicine. The use of immobilization strategies, together with drying and sporulation methods, appears to be highly promising strategies as cell preservation methods (Table 1). Since the early 1980s, bacterial and yeast cells were immobilized for biofermentation purposes as they are easier to manage and robust. Nevertheless, the immobilization of mammalian cells represents a challenge in modern medicine

20

Live Cell Immobilization

481

Table 1 Advantages and disadvantages of different techniques used to preserve cell viability Preservation method Freeze drying (lyophilization)

Advantages Proven industrial performance record Easily rehydrated product

Vacuum drying

Yields potentially high survival rates and long-term stability Relatively low production costs Suitable alternative for freezesensitive microorganisms Provides physical shielding and isolation Allows solute diffusion

Encapsulation in organic polymers (e.g., hydrogels)

Encapsulation in inorganic polymers (e.g., sol-gel)

Provides physical shielding and isolation Mechanical rigidity and good optical properties Allows solute diffusion Limits bacterial growth

Disadvantages Costly and complex technique Product sensitive to moisture Less well-proven performance record Harsher drying conditions than freeze drying Biodegradable Bacterial growth may occur Opacity may hinder optical signal detection Tested for a limited variety of microorganisms

and industrial process. For example, for the production of complex proteins (e.g., immunoglobulins) and for proteins that undergo posttranslational modifications, the use of eukaryotic cells is required. The possibility to maintain cells alive for long periods of time without the need to add nutrients is still an open issue, and several efforts are driven in this direction. Several advantages have been shown as result of cell immobilization and reported to have higher production rates than freely suspended cells. For example, immobilized Capsicum frutescens Mill cultures treated with precursors of vanillin and capsaicin were able to accumulate more quantities of biotransformed compounds than freely suspended cultures (Ramachandra and Ravishankar 2000; Ranjitha and Oberoi 2018). Similar observations were reported by other authors (Lindsey and Yeoman 1984) in cell cultures of C. frutescens. Woenderbag and Pras also have shown that freely suspended cells accumulate their secondary metabolites in the stationary phase of their growth cycle. Conversely, thanks to a permanent nongrowing phase due to the immobilization process, encapsulated living cells were able to improve secondary metabolites production (Woerdenbag et al. 1990). It has been also reported that intracellular enzymatic activity changes when cells are immobilized, with a general increased yield of recombinant protein production in comparison to that obtained with suspended cell cultures (Mevy et al. 1999; Bhatia et al. 2015a, b). Bhatia et al. in 2015, studied cadaverine production from lysine decarboxylase (CadA) overexpressed in Escherichia coli cells. Barium alginate was selected as a matrix for immobilization of E. coli YH91. Free cells and immobilized cells were characterized, and immobilized cells turned out three times more thermally stable compared to free cells, showing a half-life of 131 h at 37  C. While free

482

A. Lopreside et al.

cells lost most of lysine decarboxylase activity after nine cycles, immobilized cells retained 56% of their residual activity even after the 18th cycle (Bathia et al. 2015). Increased resistance to external factors like toxic second metabolite was also observed in immobilized cells. Also Jirků reported that immobilized yeast cells obtained by incorporating epichlorohydrin into the hydroxyalkyl methacrylate gel Separon H-1000 were more resistant to ethanol. Changes in the membrane composition were found after one hour cell growth in the immobilized state (Jirků 1999). In general, the main advantages of whole-cell immobilization for biocatalysis applications may be summarized as follows: (i) increase in productivity; (ii) easy separation that allows use in both repeated and continuous processes; (iii) reduction of operational costs; (iv) increased cell resistance to external physical or chemical stressors (i.e., presence of toxic compounds); and (v) prolonged stability for extended periods of time (Wilkinson et al. 1992; Jianlong et al. 2002; Woodward 2004). When cell immobilization is employed for cell biosensing purposes, an additional advantage can be identified, i.e., portability and easy integration into compact analytical devices for on-field use. It must be however pointed out that when living cell biosensors are immobilized, a crucial aspect is not only the maintenance of their viability but also the restoring of a physiologically active state at the time of sample exposure. This means that cells have to be responsive (and inducible) when exposed to the analyte. Microbial species, such as E. coli, are characterized by very rapid growth rates (generation rate of about 15–20 min) and have improved survival in stationary phase; however, their responsiveness in this state is very poor; it comes out that immobilization of microbial cells is not trivial. As concerns eukaryotic cells, several issues must be considered, also in terms of time and cost (Vorlop and Klein 1985; Cassidy et al. 1996). Another issue is related to the fact that immobilized cells can be used only for products that are excreted into culture media. Several efforts have also been made to induce the release of intracellular products by increasing, either in a permanent or transient way, cell permeability using dimethyl sulfoxide (DMSO) or with genetically modified cells. Thanks to these approaches, the intracellular uptake of precursors and nutrients is also favored, and cellular metabolism is triggered. In general, the main limitations of whole-cell immobilizing processes may be related to (i) relatively high cost and time; (ii) difficulty in immobilizing highly demanding cell types, such as mammalian cells; and (iii) need for extracellular products for adequate recovery in bioproduction and bioconversion processes. Table 2 summarizes main desirable requirements of matrices and supports for living cell immobilization depending on the applications. Another consideration is related to oxygen bioavailability and light irradiation, which play important roles in certain cell pathways and enzyme activities. When cells are immobilized into a matrix, only the outer cell layers are exposed to light. Increased dark conditions may be advantageous in the case of some compounds that are light sensitive; on the other hand, the supplement of light in the interior part can be provided with optical fibers. Most used immobilization matrices can be classified into artificial and natural ones (Table 3). Artificial matrices are the most diffused as they are more robust,

20

Live Cell Immobilization

483

Table 2 Optimal requirements for cell immobilization supports and matrices and their relevance according to the desired application: ( ) not relevant, (+) important, (++) very important, (+++) highly important Requirement Easy immobilization procedure under mild conditions Mechanical and chemical stability Regeneration capability High cell-loading capacity Accessibility of nutrients Sterilization capability Low cost Easy scale-up Suitable for implementation into standard reactor systems Retention of immobilized cell viability Maintenance of biological and metabolic activity of immobilized cells Easy separation (of carrier and cells) from media Easy handling Non-toxic Controlled cell growth and oxygenation

Bioproduction ++

Cell biosensing +

Regenerative medicine +

Bioremediation +

++

++

++

++

++ +++ +++ ++ ++ +++ +++

+ + ++ + +++ ++

++ ++ ++ +

+++ +++ +++ ++ +++ +++

+++

+++

+++

+++

+++

+++

+++

+++

+++ + ++ +++

+++ ++ + ++

+ +++ ++

++ + ++

Table 3 Examples of support materials with potential for cell immobilization Inorganic Aluminum oxide Nickel oxide Stainless steel Porous glass Porous silica Iron oxide Titanium oxide Pumice stone Zirconium oxide Silicon dioxide Activated carbon

Organic Polyethylene polystyrene Polyacrylate Nylon Polyacrylamide Polymethacrylate Polyaniline Polyphenol Polyester Poly(vinyl alcohol)

Bio-based Cellulose Dextran Agarose Starch Alginate Carrageenan Chitin Chitosan Collagen, gelatin

Nanomaterials Gold nanoparticles quantum dots Liponanoparticles nanofibers Nanoporous silica Nanotubes Nanoparticles

484

A. Lopreside et al.

Fig. 1 A schematic view of four main different live cell immobilization techniques: covalent binding, adsorption, entrapment, and encapsulation (Borin et al. 2018)

cheaper, and faster to obtain. Several techniques have been developed with these supports both for viable and nonviable cell. Above all polysaccharide gel matrices (i.e., Ca-alginate hydrogels or gelatin) are widely used materials for cell entrapment, while cross-linked cells with glutaraldehyde or carbodiimide are generally preferred for nonviable cells (Junter and Jouenne 2017). Nature-inspired matrices are generally formed by a first step in which cells are weakly adsorbed by electrostatic interaction over a surface followed by colonization of the support and biofilm formation. The naturally developed biofilm is subsequentially entrapped into an extracellular matrix composed by self-produced polymers. This process works efficiently for microbial cells as bacterial cells that naturally develop biofilm to promote cell aggregation and second metabolite production as universal strategy for survival. Also yeast cell flocculation and biofilm formation naturally occur for sugar depletion or stress. Yeast cells capable of flocculation appear to be more resistant to a number of stresses, including ethanol, peroxide, high temperature, or antibiotic exposure. In this chapter we will discuss main immobilization techniques and alternative approaches to immobilize or encapsulate living cells for different purposes. Four main categories will be addressed: (i) covalent attachment, (ii) adsorption, (iii) encapsulation, and (iv) entrapment (Fig. 1).

Covalent Attachment Covalent attachment is an irreversible process that links cells to a surface or to an insoluble matrix. Binding stability is the main advantage of this technique, and it is widely employed in enzyme immobilization. The main drawback is the loss of activity which could be as high as 60% for some enzymes. Since the formation of

20

Live Cell Immobilization

485

covalent bonds generally induces physical and chemical stress to the cells, this type of immobilization is seldomly used for living cells but rather for attaching nonviable cells. Covalent linking requires the binding of a reactive group on the cell, e.g., an amine or carboxyl group, to a reactive group on the surface. To immobilize cells onto porous glass beads, homo- or hetero-bifunctional cross-linkers are generally employed with aminosilane/glutaraldehyde method being one of the most used approaches to covalently bind microbial cells. Other approaches, such as aminosilane/carbodiimide and mercaptosilane/N-succinimidyl 4-maleimidobutyrate (GMBS), showed similar performance (Shriver-Lake et al. 2002). Reagents like glutaraldehyde and proteinic supports such as gelatin, albumin, and hen egg white have been successfully used for viable microbial cell immobilization (D’ Souza 2001).

Adsorption Adsorption is based on a noncovalent interaction between the cell and an organic or inorganic surface. This interaction is mostly provided by van der Waals, electrostatic, and hydrophobic forces between the cell and either porous or nonporous matrices. This approach is highly favored due to the fact that many cells have the capability to naturally interact, as during biofilm formation. This process depends on the chemical nature of the supports and cell age. For example, Basile et al. investigated the interactions between bacterial cells and charged surface such as calcium alginate, genipin-cross-linked chitosan, hydroxyethyl-methacrylate, and cationic charged methacryloxyethyl)trimethylammonium chloride (HM92), on hydrogen production in batch reactor (Kumar et al. 2016; Basile et al. 2010). HM92 enhanced cell adhesion and reaction kinetics leading to a 3.6 times higher hydrogen production in comparison to all other tested matrices. Liu et al. also tested carbon nanotubes as supports and reported a faster and enhanced performance of hydrogen fermentative production (Liu et al. 2012). In principle, adsorption is a reversible process, and the surface could be theoretically reused after the process. Nevertheless, during cell adsorption, strong interactions occur between cell and surface, rendering the surface not reusable. Nedović et al. showed examples of aroma production in beer, cider, wine, and fruit wine with yeast immobilized onto different types of carrier materials (Nedović et al. 2015), demonstrating the high industrial potential of adsorption in industrial processes.

Encapsulation Encapsulation is an immobilization technique that involves the formation of a continuous membrane around a core composed of living cells with the formation of spheres having controlled size (Kailasapathy 2002). The presence of an

486

A. Lopreside et al.

enveloping membrane with selective and tunable permeability is the key factor (Verma et al. 2006) to enable control of nutrient transfer, oxygen permeability, and, when applied to bioproduction, precursors’ uptake and products’ release. Unlike other immobilization methods, cells are tightly retained inside the core without any material leakiness and product contamination. However, it is generally more expensive and time-consuming compared to adsorption and entrapment approaches. Akinbomi et al. (2015) compared the hydrogen production obtained with polyvinylidene fluoride (PVDF) membrane-encapsulated microbial cells with free cells in both semicontinuous and batch mode-operated systems. Encapsulated cells provided an improved biohydrogen generation performance for octanol, myrcene, and hexanal (Akinbomi et al. 2015; Kumar et al. 2016).

Entrapment Thanks to its relative simplicity and low cost, cell entrapment is the most commonly used method for cell immobilization, especially for biotechnological processes such as ethanol production. Several matrices have been exploited including alginate, agar, chitosan, cellulose, and its derivatives, collagen, gelatin, epoxy resins, photo crosslinkable resins, polyester, polystyrene, polyurethane, and acrylic polymers (Lopez et al. 1997; Bayat et al. 2015). Alginate beads are widely used for long-term cell storage especially for bacteria and yeast (Cevenini et al. 2018; Belkin et al. 2017). Alginate beads or slices are formed by simple mixing of cell suspension in alginate with a calcium chloride solution. Alginate in the presence of cations forms a stable gel with entrapped cells. Several variants of the standard protocol have been reported, including the use of κ-carrageenan and potassium to replace alginate and calcium. This approach has been applied to the biosensor field, to immobilize wholecell biosensors; however, several reports highlight issues related to the porous nature of the matrix, which may lead to material leaks (Verma et al. 2006; Trelles and Rivero 2013). On the other hand, the porous structure can be considered an advantage to enable diffusion of analytes to be determined (such as pollutants) and nutrients. Other materials, selected for their high density and improved mechanical stability, have been tested to overcome this limitation, including cotton fibers, fiberglass mats, and reticulate polyurethane foam. Wu et al. used a novel polymer ethylene vinyl acetate copolymer (EVA) for hydrogen production obtaining very high yields (Wu et al. 2005; Kumar et al. 2016). Also the use of polyethylene glycol (PEG) and polydimethylsiloxane (PDMS) provided similar results when used to immobilize a microbial consortium for biohydrogen production (Singh et al. 2013; Ismail et al. 2011). Han et al. compared hydrogen generation by Enterobacter aerogenes immobilized in various matrices like alginate, agar, gelatin, carrageenan, and glass beads. They reported that entrapment in agar gels seems to be the best immobilization strategy compared to adsorption in glass beads and other entrapment techniques (Han et al. 2011).

20

Live Cell Immobilization

487

Cell Immobilization on 3D-Printed Matrices During the last years, thanks to the development of additive manufacturing, also called 3D printing technology, new supports and immobilization methods are becoming available. 3D printing, based on a computer-aided design (CAD) for layer-upon-layer fabrication, enables rapid prototyping with a number of different materials, spanning from polymers to metals, with applications in several fields including regenerative medicine and biosensing (Bishop et al. 2016; Sharafeldin et al. 2018; Lopreside et al. 2019). Dos Santos et al. reported that microbial cells bound to 15 mm beads obtained by 3D printing showed a reduced fermentation time and increased productivity of propionic acid in comparison with free cell suspensions. An improved cell viability was also reported even after long storage with an operational stability of up to 150 days (Dos Santos et al. 2018). Besides applications with microbial cells, the interest in 3D printing for mammalian cell immobilization (Wüst et al. 2011) is exponentially growing due to its potential impact in the clinical field. Boland et al. were the first to report that mammalian cells could be successfully deposited using a standard commercial thermal inkjet printer (Xu et al. 2005). They also demonstrated that this technique allows the creation not only of simple inert scaffolds but also functional material with tunable advanced properties and able to interact with the cells. As proof of concept, they printed primal neural cells onto a collagen scaffold and confirmed that the normal electrophysiology was preserved (Xu et al. 2006). Recently, many groups reported the fabrication of 3D-printed scaffolds with immobilized cells for in vivo implants (Liu et al. 2018). A 3D fully interconnected porous architecture is an ideal requirement of a scaffold to ensure cell differentiation, nutrient exchange, and removal of metabolic waste. Cell migration is followed by proliferation, maturation, and differentiation to generate the tissue of interest (i.e., bone tissue formation). Very interestingly, a 3D printing platform that enables additive manufacturing of complex 3D living architectures was reported. Two “living materials” capable of degrading pollutants and of producing cellulose were obtained by embedding two bacterial strains, P. putida and A. xylinum, in a hydrogel ink (Schaffner et al. 2017). The possibility to combine living cells with 3D printing technology (Fig. 2) has tremendous potential in bioremediation, biosensing, and biomedical fields; however, many challenges still need to be addressed, including reusability, scalability, and environmental impact of these systems (Kyle 2018).

Whole-Cell Biosensors Based on Immobilized Microbial Cells Whole-cell biosensors, relying on living cells as sensing elements, possess unique advantages for on-site monitoring, for instance, the ability to measure the actual bioavailable concentration (Durand et al. 2015; Roda et al. 2011). The well-known unsolved problems of whole-cell biosensors are related to their limited portability

488

A. Lopreside et al.

Fig. 2 3D printing of bacteria to embed multifunctional microbes into bioink hydrogels and create “living material” with advanced properties. (Reprinted with permission from Reference Kyle 2018)

and scarce shelf-life due to the difficulty of keeping cells alive and responsive for long periods of time. The goal is to obtain cells in a ready-to-use format, in which cells can be stored for a long period of time (at least months) under controlled conditions, and once activated (i.e., by a defined temperature change or addition of nutrients), a constant and reproducible number of viable cells can be revitalized (Michelini and Roda 2012). For the development of portable biosensing applications, microbial cells have been immobilized using a variety of approaches employed in other fields such fermentation processes for industrial bioproduction. When dealing with whole-cell biosensors, several factors should be taken into consideration during cell immobilization, for example, opacity of some matrices, such as PVA, is not an issue for electrochemical biosensors while could interfere with optical signal detection (Houbertz et al. 2003; Charrier et al. 2011). Among the most employed matrices for cell biosensors’ immobilization, three groups can be identified: ionic hydrogels, thermogels, and synthetic polymers. Hydrogel entrapment, i.e., using agarose, acrylamide, and alginate, represents one of the most suitable approaches for cell biosensor immobilization (Barin et al. 2009; Ruan et al. 2018; Mbeunkui et al. 2002). Bacteria and yeast entrapped into alginate beads or slices have been implemented into portable biosensing devices with intended use in remote area, showing possibility of long-term storage, low toxicity, and good reproducibility. As an example, Belkin et al. implemented genetically modified bacteria for on-field detection of land mine with remote control (Fig. 3). A scanning laser system was used for detection of immobilized bacterial cells in alginate beads scattered over a contaminated area (Belkin et al. 2017). Also S. cerevisiae cells were successfully immobilized in alginate slices for on-site

20

Live Cell Immobilization

489

Fig. 3 Remote detection system of buried landmines based on alginate-encapsulated bacterial sensors dispersed on soil (a); schematic view of biosensors principle (b); photo of telescope setting for fluorescent detection (b). (Reprinted with permission from Reference Belkin et al. 2017)

monitoring of water pollutants. In this case ready-to-use 3D-printed cartridges embedding whole-cell biosensors were obtained, demonstrating good reproducibility when stored at +4  C, maintaining the 70% of the responsiveness even after 2 weeks (Cevenini et al. 2018). Thouand et al. showed the possibility to obtain disposable cards with immobilized different bacterial strains within 2% agarose (Horry et al. 2007; Affi et al. 2009; Jouanneau et al. 2016). Since ionotropic gelation in alginate, pectate, and κ-carrageenan matrices occurs in the presence of Ca2+ or K+ ions, the presence of chelating agents can represent an issue. Hence, new synthetic materials like latex could be selected to replace natural polymers. This approach could be highly valuable for developing whole-cell biosensors for heavy metals. Yoetz-Kopelman et al. proved the feasibility of physically adsorbed cells to the outer surface of polyacrylamide beads deposited on an electrode’s surface. The developed biochip could be stored at 4  C for at least 6 months without losing cell viability (Yoetz-Kopelman et al. 2016). Current trends in bacterial biosensors are focused on the achievement of fine control over cells, to enable their manipulation, thus providing a new strategy to concentrate cells and separate them from complex media before measurement of the analytical signal. This could lead to a reduction of matrix effect an increase in the analytical performance. Naturally magnetotactic bacteria have been already

490

A. Lopreside et al.

exploited as portable biosensors, but their genetic manipulation is not as easy as that of other bacteria like E. coli (Roda et al. 2013), and therefore, their implementation is not trivial. For this reason, immobilization methods with magnetic iron oxide nanoparticles are coming under the spotlight. Ranmadugala et al. have recently summarized the main advantages and limitations of this approach (Ranmadugala et al. 2018). Iron oxide nanoparticles have unique features such as large surface area to volume ratio, superparamagnetism, and good biocompatibility, and they are easy to separate. Nanoparticle size is considered one of the main features for a successful immobilization process, in which small particles with high surface area to volume ratio seem to possess a higher binding capability along with increased superparamagnetic properties (Gupta and Gupta 2005). However, this increases molecular crowding interactions between bacteria and nanoparticles. In order to reduce repulsion between the negatively charged bacteria and nanoparticles, several classes of positively charged surface coatings have been employed such as polymers, polyelectrolytes, and surfactants. Controlling the size, shape, stability, and dispersibility still represents challenges that need to be solved for further implementation. When comparing the shelf-life of immobilized bacteria, one of the longest storage times for bacterial cell was achieved by Daunert and coworkers, thanks to sporeforming bacteria Bacillus subtilis and Bacillus megaterium. Luminescent sporous systems were developed for arsenic and zinc detection. These miniaturized portable biosensors were integrated into portable devices and stored for 6 and 8 months, respectively, for Bacillus subtilis and Bacillus megaterium. For spore-based biosensing, the cells are grown, spore formation is induced, and the spores are then preserved and loaded in the device. Before sample exposure, the spores are germinated to vegetative cells, returning in their “responsive form.” One of the main advantages in the use of spores is their amazing capability to survive to extreme conditions such as dry and wet heat, freezing temperatures, and desiccation, allowing easier storage and shipment even in remote area (Date et al. 2007). More recently wild-type B. subtilis spores were exposed to simulated Mars surface conditions, with a good survival that mostly depended on UV exposure and was not affected by environmental conditions such as atmospheric composition, low pressure, and low temperature (Cortesão et al. 2019).

Whole-Cell Biosensors Based on Immobilized Mammalian Cells Mammalian cell lines have several disadvantages compared with yeast and bacterial cell; however, their implementation in immobilized conditions is highly valuable when predictive information about bioactivity, such as toxicity to humans, is required. Mammalian cell lines are characterized by slow growth, demanding cell culture facilities, and their viability being highly affected by external factors such as exposure to physicochemical stressors. For these reasons, a suitable support is

20

Live Cell Immobilization

491

required not just to provide the cells an inert matrix for cell growth but to confer them protection from external agents and to create a suitable microenvironment. Indeed, local concentration of cytokine and other soluble factors for paracrine signaling improves cell-to-cell communication and survival. Lin et al. developed a bioactive hydrogel scaffold for pancreatic β-cell culture by conjugating specific receptors and ligands into PEG hydrogels with a synergic effect on cell survival and cell-cell communication (Lin and Anseth 2011). Over the years different materials have been exploited as matrices to optimize the immobilization of mammalian cells, and in some cases, approaches were similar to those explored for microbial cells. Constitutively bioluminescent Hek293 were immobilized in a matrix of 2% agarose to create ready-to-use cartridges and used as “sentinel cells” by Cevenini et al. These cell-agarose mixtures were then transferred into 3D-printed cartridges and stored at 4  C for 6 days. They evaluated the feasibility of immobilizing mammalian cells to obtain ready-to-use cartridges that can be stored for short periods of time, until needed. Results suggest that cells rapidly lose their capability to act as “living sentinels”; in fact the lack of optimal culture conditions during storage (i.e., temperature, humidity, and CO2) negatively affects cell metabolism and responsiveness (Cevenini et al. 2016a, b). Recently, Xu et al. outlined the potential of hypothermic storage of mammalian cells. They employed a cell-membrane-mimetic polymer hydrogel to keep mouse cells alive for more than 1 week at 4  C and 4 days at 25  C (Xu et al. 2015). Even more promising is the work reported by Jack et al. (2006), who reported the possibility to obtain spheroids and exploit the long-term metabolic arrest induced by air-drying. They have obtained a 6-week viability during storage in air, at room temperature of HEK293 cells using partially dried 3D multicellular spheroids. Such approaches could represent a significant step toward the development of robust cell-based biosensors and for their facile transport and storage. Very recently a smartphone-based bioluminescence 3D cell biosensor platform was reported based on immobilized HEK293 spheroids genetically engineered with powerful red- and green-emitting luciferases used as inflammation and viability reporters. Spheroids were immobilized with a medium solution containing 5% v/v gelatin from porcine skin type A to increase the stiffness of the spheroid microenvironment (Michelini et al. 2019; Baraniak et al. 2012). For more demanding cell lines, Cruz-Acuña et al. (2018) developed a protocol for the synthesis of a fully defined, synthetic hydrogel that supports the generation and culture of human pluripotent stem cell (hPSC)-derived organoids (HOs). This new modular, welldefined cell-encapsulating hydrogel is formed of a four-armed poly-(ethylene glycol) macromer that has maleimide groups at each terminus (PEG-4MAL) and is conjugated to cysteine-containing adhesive peptides and cross-linked via protease-degradable peptides (Fig. 4). The PEG-4MAL hydrogel supports the engraftment of the HOs and accelerates colonic wound repair. This strategy can be exploited for the formation of cell-cell and spheroid-spheroid connections, thus ensuring greater communication and prompt supply of nutrients. Thanks to this new technique, tissue preparation and subsequent encapsulation can be performed within few hours.

492

A. Lopreside et al.

Fig. 4 Obtainment of PEG-4MAL hydrogels for human organoid generation. (Reprinted with permission from Reference Cruz-Acuña et al. 2018)

Conclusion During the last years, both microbial and mammalian cells were successfully immobilized on different surfaces or entrapped into suitable matrices for bioproduction, bioremediation, biosensing, and regenerative medicine applications. Depending on the goal of the immobilization and the type of cell, different approaches have been explored. The recent availability of low-cost 3D printing and affordable micro- and nano-fabrication technologies provides new tools in the cell immobilization arsenal. Nature-inspired approaches are also very promising, with the use of spores being the most elegant example with unbeatable performance in cell preservation, as demonstrated by their survival to Mars surface conditions.

References Affi M, Solliec C, Legentillomme P, Comiti J, Legrand J, Thouand G (2009) Numerical design of a card and related physicochemical phenomena occurring inside agarose-immobilized bacteria: a valuable tool for increasing our knowledge of biosensors. Sens Act B Chem 138(1):310–317 Akinbomi J, Wikandari R, Taherzadeh M (2015) Enhanced fermentative hydrogen and methane production from an inhibitory fruit-flavored medium with membrane-encapsulated cells. Membranes 5:616–631 Baraniak PR, Cooke MT, Saeed R, Kinney MA, Fridley KM, McDevitta TC (2012) Stiffening of human mesenchymal stem cell spheroid microenvironments induced by incorporation of gelatin microparticles. J Mech Behav Biomed Mater 11:63–71

20

Live Cell Immobilization

493

Barin M, Otadi M, Khorasheh F, Kheirolomoom A (2009) Effect of cell concentration on the acylation of penicillin G enzymatic reaction in immobilized cells. Sci Iran Trans C Chem Chem Eng 16:69–73 Basile MA, Dipasquale L, Gambacorta A, Vella MF, Calarco A, Cerruti P, d’Ayala GG (2010) The effect of the surface charge of hydrogel supports on thermophilic biohydrogen production. Bioresour Technol 101(12):4386–4394 Bayat Z, Hassanshahian M, Cappello S (2015) Immobilization of microbes for bioremediation of crude oil polluted environments: a mini review. Open Microbiol J 9:48–54 Belkin S, Yagur-Kroll S, Kabessa Y, Korouma V, Septon T, Anati Y, Zohar-Perez C, Rabinovitz Z, Nussinovitch A, Agranat AJ (2017) Remote detection of buried landmines using a bacterial sensor. Nat Biotechnol 35:308 Bhatia SK, Kim KYH, Kim HJ, Seo HM (2015a) Biotransformation of lysine into cadaverine using barium alginate-immobilized Escherichia coli overexpressing CadA. Bioprocess Biosyst Eng 38:2315–2322 Bhatia S, Sharma K, Dahiya R, Bera T (2015b) Modern applications of plant biotechnology in pharmaceutical sciences. Academic, San Diego Bishop GW, Satterwhite-Warden JE, Kadimisetty K, Rusling JF (2016) 3D-printed bioanalytical devices. Nanotechnology 27:284002 Borin GP, de Melo RR, Crespim E, Sato HH, Contesini FJ (2018) An overview on polymer gels applied to enzyme and cell immobilization. In: Polymer Gels. Springer, Singapore Cassidy MB, Lee H, Trevors JT (1996) Environmental applications of immobilized microbial cells: a review. J Ind Microbiol 16:79–101 Cevenini L, Calabretta MM, Lopreside A, Tarantino G, Tassoni A, Ferri M, Roda A, Michelini E (2016a) Exploiting NanoLuc luciferase for smartphone-based bioluminescence cell biosensor for (anti)-inflammatory activity and toxicity. Anal Bioanal Chem 408:8859–8868 Cevenini L, Calabretta MM, Tarantino G, Michelini E, Roda A (2016b) Smartphone-interfaced 3D printed toxicity biosensor integrating bioluminescent “sentinel cells”. Sensors Actuators B 225:249–257 Cevenini L, Lopreside A, Calabretta MM, D’Elia M, Simoni P, Michelini E, Roda A (2018) A novel bioluminescent NanoLuc yeast-estrogen screen biosensor (nanoYES) with a compact wireless camera for effect-based detection of endocrine-disrupting chemicals. Anal Bioanal Chem 410:1237–1246 Charrier T, Durand MJ, Jouanneau S, Dion M, Pernetti M, Poncelet D, Thouand G (2011) A multichannel bioluminescent bacterial biosensor for the on-line detection of metals and toxicity. Part I: design and optimization of bioluminescent bacterial strains. Anal Bioanal Chem 400 (4):1051–1060 Cortesão M, Fuchs FM, Commichau FM, Eichenberger P, Schuerger AC, Nicholson WL, Setlow P, Moeller R (2019) Bacillus subtilis Spore Resistance to Simulated Mars Surface Conditions. Front Microbiol 26(10):333 Cruz-Acuña R, Quirós M, Huang S, Siuda D, Spence JR, Nusrat A, García AJ (2018) PEG-4MAL hydrogels for human organoid generation, culture, and in vivo delivery. Nat Protoc 13:2102–2119 D’ Souza SF (2001) Microbial biosensors. Biosens Bioelectron 16:337–353 Date A, Pasini P, Daunert S (2007) Construction of spores for portable bacterial whole-cell biosensing systems. Anal Chem 79(24):9391–9397 Dos Santos Belgrano F, Diegel O, Pereira JN, Hatti-Kaul R (2018) Cell immobilization on 3D-printed matrices: a model study on propionic acid fermentation. Bioresour Technol 249:777–782 Durand MJ, Hua A, Jouanneau S, Cregut M, Thouand G (2015) Detection of metal and organometallic compounds with bioluminescent bacterial bioassays. Adv Biochem Eng Biotechnol 3:77–99 Gupta AK, Gupta M (2005) Synthesis and surface engineering of iron oxide nanoparticles for biomedical applications. Biomaterials 26:3995–4021

494

A. Lopreside et al.

Han J, Lee D, Cho J, Lee J, Kim S (2011) Hydrogen production from biodiesel byproduct by immobilized Enterobacter aerogenes. Bioprocess Biosyst Eng 35:151–157 Horry H, Charrier T, Durand MJ, Vrignaud B, Picart P, Daniel P, Thouand G (2007) Technological conception of an optical biosensor with a disposable card for use with bioluminescent bacteria. Sens Act B Chem 122(2):527–534 Houbertz R, Domann G, Cronauer C, Schmitt A, Martin H, Park J-U, Fröhlich L, Buestrich R, Popall M, Streppel U, Dannberg P, Wächter C, Bräuer A (2003) Inorganic-organic hybrid materials for application in optical devices. Thin Solid Films 442:194–200 Ismail I, Hassan MA, Soon CS (2011) Effect of retention time on biohydrogen production by microbial consortia immobilized in polydimethylsiloxane. Afr J Biotechnol 10:601 Jack GD, Garst JF, Cabrera MC, DeSantis AM, Slaughter SM, Jervis J, Brooks AI, Potts M, Helm RF (2006) Long term metabolic arrest and recovery of HEK293 spheroids involves NF-kappaB signaling and sustained JNK activation. J Cell Physiol 206:526–536 Jianlong W, Xiangchun Q, Liping H, Yi Q, Hegemann W (2002) Microbial degradation of quinoline by immobilized cells of Burkholderia pickettii. Water Res 36:2288–2296 Jirků V (1999) Whole cell immobilization as a means of enhancing ethanol tolerance. J Ind Microbiol Biotechnol 22:147–151 Jouanneau S, Durand-Thouand MJ, Thouand G (2016) Design of a toxicity biosensor based on Aliivibrio fischeri entrapped in a disposable card. Environ Sci Pollut Res 23(5):4340–4345 Junter GA, Jouenne T (2017) Immobilized viable cell biocatalysts: a paradoxical development. Elsevier, Mont-Saint-Aignan Kailasapathy K (2002) Microencapsulation of probiotic Bacteria: technology and potential applications. Curr Issues Intest Microbiol 3:39–48 Kumar G, Mudhoo A, Sivagurunathan P, Nagarajan D, Ghimire A, Lay CH, Chang JS (2016) Recent insights into the cell immobilization technology applied for dark fermentative hydrogen production. Bioresour Technol 219:725–737 Kyle S (2018) 3D printing of Bacteria: the next frontier in biofabrication. Trends Biotechnol 36:340–341 Lin CC, Anseth KS (2011) Cell–cell communication mimicry with poly (ethylene glycol) hydrogels for enhancing β-cell function. Proc Natl Acad Sci 108(16):6380–6385 Lindsey K, Yeoman MM (1984) The synthetic potential of immobilized cells of Capsicum frutescens mill. cv. Annuum. Planta 162:495–501 Liu Z, Lv F, Zheng H, Zhang C, Wei F, Xing XH (2012) Enhanced hydrogen production in a UASB reactor by retaining microbial consortium onto carbon nanotubes (CNTs). Int J Hydrog Energy 37:10619–10626 Liu Y, Li T, Ma H, Zhai D, Deng C, Wang J, Zhuo S, Chang J, Wu C (2018) 3D-printed scaffolds with bioactive elements-induced photothermal effect for bone tumor therapy. Acta Biomater 73:531–546 Lopez A, Lazaro N, Marques AM (1997) The interphase technique: a simple method of cell immobilization in gel-beads. J Microbiol Methods 30:231–234 Lopreside A, Calabretta MM, Montali L, Ferri M, Tassoni A, Branchini BR, Southworth T, D’Elia M, Roda A, Michelini E (2019) Prêt-à-porter nanoYESα and nanoYESβ bioluminescent cell biosensors for ultrarapid and sensitive screening of endocrine disrupting chemicals. Anal Bioanal Chem 411(19):4937–4949. Mbeunkui F, Richaud C, Etienne AL, Schmid RD, Bachmann TT (2002) Bioavailable nitrate detection in water by an immobilized luminescent cyanobacterial reporter strain. Appl Microbiol Biotechnol 60:306–312 Mevy JP, Rabier J, Quinsac A, Ribaillier D (1999) Sucrose metabolism and indoleglucosinolate production of immobilized horseradish cells. Plant cell, tissue and organ culture 57(3):163–171 Michelini E, Roda A (2012) Staying alive: new perspectives on cell immobilization for biosensing purposes. Anal Bioanal Chem 402:1785–1797

20

Live Cell Immobilization

495

Michelini E, Calabretta MM, Cevenini L, Lopreside A, Southworth T, Fontaine DM, Simoni P, Branchini BR, Roda A (2019) Smartphone-based multicolor bioluminescent 3D spheroid biosensors for monitoring inflammatory activity. Biosens Bioelectron 123:269–277 Nedović V, Gibson B, Mantzouridou TF, Bugarski B, Djordjević V, Kalušević A, Paraskevopoulou A, Sandell M, Šmogrovičová D, Yilmaztekin M (2015) Aroma formation by immobilized yeast cells in fermentation processes. Yeast 32:173–216 Radwan SS, Al-Hasan RH, Salamah S, Al-Dabbous S (2002) Bioremediation of oily sea water by bacteria immobilized in biofilms coating microalgae. Int Biodet Biodeg 50:55–59 Ramachandra R, Ravishankar GA (2000) Vanilla flavour: production by conventional and biotechnological routes. J Sci Food Agric 80(3):289–304 Ranjitha K, Oberoi HS (2018) Food Flavours from yeasts: improved productivity through biotechnological interventions and process optimization. Curr Biotechnol 7(3):199–213 Ranmadugala D, Ebrahiminezhad A, Manley-Harris M, Ghasemi Y, Berenjian A (2018) Magnetic immobilization of bacteria using iron oxide nanoparticles. Biotechnol Lett 40:237–248 Roda A, Cevenini L, Michelini E, Branchini BR (2011) A portable bioluminescence engineered cell-based biosensor for on-site applications. Biosens Bioelectron 26:3647–3653 Roda A, Cevenini L, Borg S, Michelini E, Calabretta MM, Schüler D (2013) Bioengineered bioluminescent magnetotactic bacteria as a powerful tool for chip-based whole-cell biosensors. Lab Chip 13:4881–4889 Ruan B, Wu P, Chen M, Lai X, Chen L, Yu L, Gong B, Kang C, Dang Z, Shi Z, Liu Z (2018) Immobilization of Sphingomonas sp. GY2B in polyvinyl alcohol–alginate–kaolin beads for efficient degradation of phenol against unfavorable environmental factors. Ecotoxicol Environ Saf 162:103–111 Schaffner M, Rühs PA, Coulter F, Kilcher S, Studart AR (2017) 3D printing of bacteria into functional complex materials. Sci Adv 3(12):eaao6804 Sharafeldin M, Jones A, Rusling JF (2018) 3D-printed biosensor arrays for medical diagnostics. Micromachines (Basel) 9(8):E394 Shriver-Lake AC, Gammeter WB, Bang SS, Pazirandeh M (2002) Covalent binding of genetically engineered microorganisms to porous glass beads. Anal Chim Acta 470:71–78 Singh L, Wahid ZA, Siddiqui MF, Ahmad A, Rahim MHA, Sakinah M (2013) Biohydrogen production from palm oil mill effluent using immobilized Clostridium butyricum EB6 in polyethylene glycol. Process Biochem 48:294–298 Trelles JA, Rivero CW (2013) Whole cell entrapment techniques. Methods Mol Biol 1051:365–374 Verma M, Brar SK, Blais JF, Tyagi RD (2006) Aerobic biofiltration processes advances in wastewater treatment. Pract Period Hazard Toxic Radioact Waste Manag 10:264–276 Vorlop KD, Klein J (1985) Immobilization of whole cells. In: Finn RK, Prave P, Schlingmenn M, Crueger W, Esser K, Thauer R, Wagner F (eds) Biotechnotogy focus. Fundamentals, applications, information. Hanser Publishers, New York Wilkinson AK, Williams PD, Mavituna OF (1992) The effect of oxygen stress on secondary metabolite production by immobilised plant cells in bioreactors. In: Fowler MW, Warren GS, Moo-Young M (eds) Plant cell biotechnology. Elsevier Woerdenbag HJ, Van Uden W, Frijlink HW, Lerk CF, Pras N, Malingré TM (1990) Increased podophyllotoxin production in Podophyllum hexandrum cell suspension cultures after feeding coniferyl alcohol as a β-cyclodextrin complex. Plant Cell Rep 9:97–100 Woodward J (2004) Methods of immobilization of microbial cells. J Microbiol Methods 8:91–102 Wu SY, Lin CN, Chang JS, Chang JS (2005) Biohydrogen production with anaerobic sludge immobilized by ethylene-vinyl acetate copolymer. Int J Hydrog Energy 30:1375–1381 Wüst S, Müller R, Hofmann S (2011) Controlled positioning of cells in biomaterials—approaches towards 3D tissue printing. J Funct Biomat 2:119–154 Xu T, Jin J, Gregory C, Hickman JJ, Boland T (2005) Inkjet printing of viable mammalian cells. Biomaterials 26:93–99

496

A. Lopreside et al.

Xu T, Gregory CA, Molnar P, Cui X, Jalota S, Bhaduri SB, Boland T (2006) Viability and electrophysiology of neural cell structures generated by the inkjet printing method. Biomaterials 27:3580–3588 Xu Y, Mawatari K, Konno T, Kitamori T, Ishihara K (2015) Spontaneous packaging and hypothermic storage of mammalian cells with a cell-membrane-mimetic polymer hydrogel in a microchip. ACS Appl Mater Interfaces 7:23089–23097 Yoetz-Kopelman T, Dror Y, Shacham-Diamand Y, Freeman A (2016) “Cells-on-beads”: a novel immobilization approach for the construction of whole-cell amperometric biosensors. Sensors Actuators B Chem 232:758–764

Sol-Gel Process, Structure, and Properties

21

Thibaud Coradin

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Chemistry of the Sol-Gel Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Inorganic Polymerization: The Example of Silica . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Controlling Inorganic Polymerization: The Alkoxide Precursors . . . . . . . . . . . . . . . . . . . . . . . . . . Organically Modified Metal Oxides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Structural Properties of Sol-Gel Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Structural Evolution upon Aging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . From Gels to Ceramics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Controlling Porous Structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Properties of Sol-Gel Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chemical Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Physical Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Biological Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions and Future Directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

498 498 498 500 501 504 504 505 507 508 508 510 511 513 514

Abstract

The sol-gel technology provides a highly versatile route to metal oxide materials. It now belongs to the toolbox of many academic and industrial researchers. It is based on a polymerization process that starts from ions or molecules and ends with gels, powders, thin films, and ceramics, among others. It is compatible with the formation of hybrid materials where organic and biological species are intimately associated with an inorganic backbone. However, a deep understanding of the underlying chemical and processing parameters is required to fully control the structure and properties of the final materials. T. Coradin (*) Chimie de la Matière Condensée de Paris, Sorbonne Université, CNRS, Collège de France, Paris, France e-mail: [email protected]; [email protected] © Springer Nature Switzerland AG 2022 G. Thouand (ed.), Handbook of Cell Biosensors, https://doi.org/10.1007/978-3-030-23217-7_141

497

498

T. Coradin

Keywords

Sol-gel chemistry · Organosilanes · Ceramics · Porous materials · Hybrid materials · Encapsulation

Introduction The sol-gel process is a chemical route that transforms solvated molecules or ions into solid-state materials. Although established from earlier works in the mid-nineteenth century, the basics and applications of the sol-gel process have emerged in the 1950s and have become part of the material scientists’ toolbox for more than 30 years. Over the last 5 years, WoS database identifies more than 30,000 publications corresponding to the “sol-gel” topic. The reasons behind this popularity are manifold, including the availability of many easy-to-implement protocols, the huge number of materials that can be prepared, and the wide range of processing and shaping techniques to which it can be applied. In addition, the possibility to form materials in solution and mild conditions, especially at moderate temperatures, opens the route to the association of inorganic phases with organic or even biological species. Altogether the sol-gel process can be applied to all fields of material science, including energy, health, catalysis, optics, etc. The amount of books and review papers dedicated to the sol-gel process and its applications is already overwhelming (Brinker and Scherer 1990; Hench and West 1990; Sakka 2005; Ciriminna et al. 2013; Levy and Zayat 2015; Danks et al. 2016). This chapter aims at providing the fundamentals of sol-gel chemistry and at describing the various steps by which materials with well-defined structures and properties can be obtained using this technology.

The Chemistry of the Sol-Gel Process Inorganic Polymerization: The Example of Silica The concept of polymerization is very-well established in organic chemistry (Harris 1981). It starts from molecules, termed monomers (M), that can react with each other, either spontaneously or after activation, to form larger molecules. Among possible polymerization mechanism, the so-called polycondensation reaction involves monomers bearing two reactive end groups. The reaction involves the formation of a new bond between the monomers accompanied by the departure of small molecules, i.e., a condensation reaction. This is typically the case of peptide link formation by amino acids. These reactions commonly occur by the simultaneous formation of short chains and reaction of these chains one with another. It is important to point out that the repeated unit in the chains is not the initial monomer because some atoms or groups are released during condensation.

21

Sol-Gel Process, Structure, and Properties

499

Let us now consider a very common inorganic material, glass (Kolb and Kolb 1979). There are indeed many kinds of glasses, but the simple form of these is an amorphous solid of chemical composition SiO2, i.e., silicon dioxide or more simply silica. A simplified view of glass is shown on Fig. 1a, illustrating that it consists of SiO4 tetrahedra linked together by Si-O-Si siloxane bonds. In some way, glass can be considered as a polymer whose repeating units are these SiO4 units. However, because each oxygen atom is shared by two silicon atoms, the chemical composition is SiO2. Many other forms of silicon dioxide exist in nature (Ehrlich et al. 2010), such as quartz where silica is in a crystalline state. Basically, glasses were initially obtained by heating sand with alkaline salts until melting and then cooling the mixture at controlled rate, i.e., fast enough for solidification to occur before SiO4 units can form an ordered structure. Silica is also formed by some living organisms, such as diatoms algae, certain sponges, and higher plants. However, in this case, only about three of four of oxygens of a SiO4 tetrahedra are linked to another Si, while the fourth is linked to a hydrogen, forming Si-OH silanol groups. These silanol groups have acid/ base properties and can form hydrogen bonds so that the resulting phase has a general formula SiO2-x.nH2O, i.e., hydrated silica. The reaction responsible for biogenic silica starts from silicic acid Si(OH)4, corresponding to the silicon species stable in water at low concentration (such as in seawater, for instance). If the living organisms are able to accumulate silicic acid in some way, for instance, within specific vesicles, a condensation reaction can occur, forming a dimer and releasing a water molecule. The next step is the formation of a trimer. However, because Si atoms are tetracoordinated, there is a possibility for the next condensation step to take place not in a linear way but forming a ring. This possibility is in fact favored for electronic reasons so that

Fig. 1 (a) Schematic structure of amorphous silica in glass (2D view); (b) Condensation reaction of silicic acid. (Reprinted with permission from Coradin and Lopez (2003). Copyright (2001) John Wiley and Sons)

500

T. Coradin

further condensation steps lead to the formation of 3D clusters that ultimately yield to silica nanoparticles (Fig. 1b) (Coradin and Lopez 2003). Indeed, silicon atoms located at the surface of the particles possess some pending Si-OH groups. This population of primary particles in suspension in an aqueous medium is called a sol. The fate of these particles is determined by the principles of colloidal chemistry: they can grow by further polymerization or aggregate. Depending on the concentration, particle size, and interparticle interactions, precipitation (phase separation between the solid and the liquid phase) or gelation (formation of two continuous phases) can then take place. Reaction pH has a major influence on the overall process (Iler 1979). On the one hand, the pKa for the first deprotonation of Si(OH)4 is ca. 9.8. Thus, at pH 8 or above, it exists in a non-negligible amount as Si(OH)3O-. These negative species are more reactive than the neutral silicic acid, speeding up the condensation process. However at pH 11–12, when the negative species are in majority, their condensation is not possible due to electrostatic repulsion. Moreover, as polymerization proceeds, the pKa value of the silanol groups decrease so that the point of zero charge of hydrated silica is ca. 2–3. Therefore, in slightly acidic conditions, silica nanoparticles will bear a small negative charge, with a strong tendency to aggregate, while in alkaline conditions, repulsive electrostatic interactions between highly charged colloids will favor their stability in the sol state. Altogether, between pH 2 and 5, small particles will be slowly formed that have a strong tendency to aggregate; between 5 and 8, larger particles are formed that have a lower tendency to aggregate (due to charge repulsion) but will show stronger interparticle interactions (due to the coexistence of silanols and silanolates Si-O ). These are the optimal conditions for gel formation. Above pH 8, large particles can grow and remain stable, but at pH >11, silica dissolution rapidly occurs. Basically, most metal oxides can in principle be formed in a similar manner (Jolivet 2000). When a metal ion is placed in an aqueous medium, it is solvated by water molecules. The positive charge bore by most of these ions interacts with oxygen atoms of H2O, leading to metal-OH or metal-O groups. These precursors are quite similar to Si(OH)4 and can therefore undergo polycondensation. However, the electronic properties of the metal species as well as their coordination number impact on the kinetics of the process as well as the nature of the particles that can range from oxides to hydroxides. Again the pH conditions will strongly impact on the reaction course and products. It is nevertheless important to mention that, in contrast to hydrated silica that can remain amorphous over a relatively broad range of temperature, hydrated metal oxides or hydroxides tend to crystallize by water expulsion so that the gels rapidly evolve into precipitated particles.

Controlling Inorganic Polymerization: The Alkoxide Precursors To obtain some control over the polymerization reaction in aqueous medium, it is possible to add organic ligands to the metal salts (Livage et al. 1988). Coordination makes the oxygen atoms less reactive and therefore slow down the condensation

21

Sol-Gel Process, Structure, and Properties

501

reaction. Moreover, they can remain adsorbed on the surface of the growing clusters or particles, limiting their size but also preventing their interactions. Finally, a suitable selection of the ligand can allow to tune the morphology of the nanoparticles. However, in many cases, these approaches cannot provide a fine control of the sol-gel reaction due to high reactivity of the inorganic ions in water. Moreover, for many applications, the possibility to run the process in organic solvents or using gas-phase reagents is highly desirable. For this reason, the most popular precursors for sol-gel reactions are metal alkoxides M(OR)n where M is the metal, R an organic group, and n the number of alkoxy groups around the metal center (Turova et al. 2002). The key principle behind the use of these alkoxides is that no reaction is possible between two M-OR functions. It is necessary that at least one M-OH group is present to form M-O-M bonds. In most cases, the conversion of M-OR into M-OH can be achieved by addition of water. This reaction is called hydrolysis and produces the ROH alcohol as a by-product. Importantly, the amount of reactive M-OH groups that can then undergo condensation depends on the amount of water molecules added to the medium. Thus, the hydrolysis ratio, as defined by [H2O]/[M-OR], is a key parameter in the synthesis. The nature of the OR group also plays a role on the kinetics on the hydrolysis process as larger R groups induce steric constraints that slow down the reaction. The reaction pH is also of paramount importance (Fig. 2) (Brinker and Scherer 1990). In acidic conditions, the substitution of R groups by protons is favored, speeding up hydrolysis. However, as explained above, the condensation reaction is slow. Therefore, many small particles will be formed that can aggregate to form dense gels. On the contrary, in basic conditions, hydrolysis is slow but formed M-OH groups will readily react. This will lead to the formation of large particles that can aggregate to either form porous gels or, if pH is high enough and therefore their surface charge is highly negative, form a stable suspension. Many other parameters such as ionic strength, solvents, temperature, and pressure can be used to control the sol-gel reaction. As an alternative to these processes that involve water, so-called non-hydrolytic sol-gel routes have been developed (Vioux 1997). In this case, alkoxides are reacted with other organometallic precursors such as metal halides in organic solvents. A particularly interesting advantage of the sol-gel process is the possibility to obtain multimetallic oxides by mixing precursors of different inorganic elements (Alifanti et al. 2003). However, the major issue to be faced is to take into account the difference in reactivity of each precursor to avoid the faster formation of one phase over the other which can lead to phase separation rather than yield to a continuous mixed network.

Organically Modified Metal Oxides The possibility to obtain inorganic materials from solutions at moderate temperature offers the advantage of performing the sol-gel reaction in the presence of organic or

502

T. Coradin

Fig. 2 Influence of pH on the kinetics of silicon alkoxide hydrolysis and condensation and corresponding reaction products. (Redrawn from Brinker and Scherer 1990)

even biological species. One approach consists in adding these species in the reaction medium which usually results in their physical entrapment within the metal oxide phase. Alternatively, it is possible to use so-called organically modified silanes where one or several organic groups are linked to the Si atom via a covalent Si-C bond, resulting in organically modified ceramics. Note that because very few metals form covalent M-C bonds and even less water-stable ones, this latter strategy is mostly limited to silica-based materials. These two kinds of materials are commonly denominated as organic-inorganic hybrid materials (or shortly hybrid materials) and, when the added species is from biological origin, biohybrid materials (Faustini et al. 2018). However, in terms of chemistry, they represent very distinct situations. In the first case, a simplified view would be that the inorganic network grows as it would do in the absence of the additive and trap it in its intrinsic porosity (Fig. 3a). However, this is often not the case because this additive can interact with the precursors and modify the course of the sol-gel reaction. For instance, it is well-known that cationic polymers can interact with silanes and activate their condensation (Coradin et al. 2002). The resulting materials consist of particles or gels where the polymers are incorporated inside the silica network (and not in its porosity) (Fig. 3b). The dimensions of the additive have

21

Sol-Gel Process, Structure, and Properties

503

Fig. 3 Hybrid materials obtained by encapsulation: (a) small additives are trapped in the intrinsic porosity of the gel; (b) additives are incorporated within the inorganic walls; (c) large additives induce the formation of larger pores; (d) the inorganic network forms at the surface of the large additives

also a strong influence on the structure of these hybrid materials. As a general guideline, when the additive exhibits dimensions that are larger than the size of the growing metal oxide nanoparticles, it can act as a template, i.e., the sol-gel reaction occurs around this object (Mann et al. 1997). If no specific interactions exist between the silanes and the additive, surface, a gel is formed that entraps it similarly to the previous situation except that it creates its own pore (Fig. 3c). On the contrary, if strong attractive interactions exist, the sol-gel reaction can be limited to the surface of the additive, resulting in its coating by a metal oxide layer (Fig. 3d). It is also important to mention the specific case of self-assembling systems. These molecules can exist as single molecules or assemble as larger objects depending on their concentration and/or variation of physicochemical conditions. It has been shown that, in some cases, interactions of these molecules, such as surfactants, with silanes modify their selfassembly properties so that the structure that imprints the sol-gel material is different from the one existing in the solution without precursors (Patarin et al. 2002). The use of organosilanes follows a completely different strategy as the organic group is integrated within the molecular structure of the precursor (Kickelbick 2007). Three types of organosilanes may be used: monofunctional R1M(OR)3, bifunctional R1R2M(OR)2, and trifunctional R1R2R3M(OR), where R1,R2,R3 are the organic groups. When placed in hydrolysis conditions, only the OR groups will be converted into OH functions and can condense. Thus, the highest the functionality, the lower is the connectivity of the organosilane and therefore the dimensionality of the resulting material. For instance, bifunctional silanes can connect by two Si-OH only and will therefore only form chains (polymers), as found in silicones. For this reason, organosilanes are usually added as dopants to M(OR)4 precursor solutions, and the hybrid network is formed via a co-condensation reaction. However, the presence of the organic group can have a strong influence on the reactivity of the silane toward hydrolysis and condensation, mainly for electronic and steric reasons but also considering their solubility, charge, etc. (Loy et al. 2000). For instance, methyltriethoxysilane (MTES,CH3-Si(OC2H5)3) hydrolyzes slower than tetraethoxysilane (TEOS, Si(OC2H5)4) which itself hydrolyzes slower than

504

T. Coradin

Fig. 4 Hybrid materials from bridged organosilanes. (Reprinted with permission from Shea and Loy (2001). Copyright (2001) American Chemical Society)

aminopropyltriethoxysilane (APTES, NH2-C3H7-Si(OC2H5)3). Taking these differences into account is required to achieve a homogeneous repartition of the organic functions within the sol-gel materials. Because the condensation reaction occurs only with Si-OH groups, the organic groups are usually located at the surface of the primary nanoparticles that are formed during the polymerization reaction. As a result, they will also be found at the surface of the particles in the sols or on the internal surface of the pores in the gels. However, it is possible to introduce an organic group within the silica network by using bis-silane (OR)3-R1-(OR)3 as condensation will proceed at both extremities of the molecules, trapping the R1 function in between (Fig. 4) (Shea and Loy 2001). Combination of these two strategies is indeed possible. For instance, preparing a silica gel bearing covalently linked hydrophobic organic groups allows to tune the environment and therefore the reactivity of entrapped (bio)molecules (Reetz et al. 1996). It must be emphasized that it is also possible to associate an organic molecule with the sol-gel inorganic particle or gel after its formation, by post-grafting of organosilanes (Calvo et al. 2009). One should note that several sol-gel metal oxides can be functionalized with organosilanes, such as iron oxides or zirconia, while, for others, such as titania or alumina, phosphonate derivatives may be more efficient (Silverman et al. 2005).

Structural Properties of Sol-Gel Materials Structural Evolution upon Aging Until now, we have presented the chemical pathways that ground the sol-gel process. We have enlightened that the condensation reaction creates objects of increasing sizes by formation of M-O-M bonds. The reaction conditions define the stage at which this

21

Sol-Gel Process, Structure, and Properties

505

growth process stops and whether the resulting sol remains stable or forms a gel. We also emphasized that because the process occurs most of the time in water, some M-OH groups remain at the surface of the particles and/or gels. A first process that happens spontaneously in the mother liquor (i.e., without separation or washing) originates from the tendency of the metal oxides to increase their stability by creating more and more M-O-M bonds through the condensation of these remaining M-OH groups. As pointed out before, the condensation reaction forms a water molecule that will therefore be expelled from the inorganic network. In the case of amorphous materials, such as silica, this process is called syneresis and results in a densification of the gel and its shrinkage (Hench and West 1990). For many metal oxides, it will lead to the crystallization of the particles and destabilization of the gel structure. In both cases, varying pH, temperature, and humidity level or adding specific additives can allow to control these structural transformations (Gawel et al. 2010).

From Gels to Ceramics Although the term hydrogel usually corresponds to highly hydrated organic or bioorganic gels, it can also be applied to sol-gel materials prepared in aqueous media. As such, these gels face the same issues as (bio)polymer upon drying. If they are left to dry under uncontrolled conditions, they not only undergo syneresis as indicated above, but water evaporation also creates capillary stress at the interface between liquid and solid phases that perturb the gel structure. These so-called xerogels become more rigid (higher Young modulus) and more fragile (lower deformation at break), often resulting in monoliths that easily crack (Abramoff and Klein 1991). There are several ways to address this point. A first one is to work under controlled humidity to regulate the course of the water evaporation. Another possibility is to use drying control chemical additive (DCCA) such as dimethylformamide that has lower surface tension than water and therefore decreases the capillary stresses generated during their evaporation (Adachi and Sakka 1988). Limiting the syneresis process, i.e., further condensation reactions of the surface silanol, is another strategy that can be achieved by modifying the pore surface with organosilanes and especially hydrophobic ones that also favor water expulsion (Sarawade et al. 2011). However, the most popular method to obtain such aerogels is through the supercritical drying process (Cansell et al. 2003). In this technique, the solvent present within the gel pores is brought to its supercritical state, meaning that it can be withdrawn from the inorganic network with minimal impact on its structure. However, because the conditions to bring water to its critical state are very difficult to reach, a first process of solvent exchange from water to a suitable alcohol is performed. Often, a second exchange is usually made with CO2 that is less reactive with the oxide surface and whose critical point conditions are easier to implement experimentally. The more effective are these techniques, the better the initial structure of the hydrogel is preserved in the dry state, and therefore the more porous is the final dry material. For instance, a dense xerogel can reach density of 1.2 g.cm 3, while that of an aerogel can be smaller than 0.2 g. cm 3. However, the presence of reactive organic groups can allow for bridging the gap between these two dried forms of sol-gel materials (Shimizu et al. 2016) (Fig. 5).

506

T. Coradin

Fig. 5 Comparison of aerogels and xerogels obtained before (left) and after (right) polymerization of organosilanes. (Reprinted with permission from Shimizu et al. (2016). Copyright (2016) American Chemical Society)

Again, for other metal oxide systems, most drying procedures will favor the transformation of amorphous (or at least poorly crystalline) particles that could be stable in a gel state, thanks to pending M-OH groups into better crystallized particles that will be recovered as powders (Ponthieu et al. 1997). It is worth noticing that freeze-drying (lyophilization) is well-adapted to control the dispersion state of the particles in the powder. In contrast, its application to hydrogel drying is far less common, as the water solidification step has a large impact on the structure stability and often leads to the recovery of powders rather than monoliths. Although drying steps described above may involve heating, but at a moderate temperature, to favor solvent evaporation, much higher temperatures can be used to convert products of the sol-gel reactions into ceramics. The main process to achieve this conversion is called sintering and involves both high temperature and high pressure to achieve the coalescence (fusion) of the particles (Pierre 1998). This is a very efficient process to obtain dense materials, but it also has a great impact on the surface reactivity of the resulting grains. Recently, the spark plasma sintering method, which use an electric discharge instead of heating, has shown great promise to improve this drawback (Garay 2010).

21

Sol-Gel Process, Structure, and Properties

507

Controlling Porous Structures As explained before, the porous structure of gels is mainly related to the chemical and physicochemical conditions of their synthesis (concentration, pH, ionic strength, etc.), while that of solvent-free materials depend on the drying/sintering process. While it is possible to control the size domain of the porosity (microporous 90% organic, and structures, from very dense to highly porous, and therefore a wide diversity of optical and mechanical properties. The works by Wei et al. nicely illustrate this diversity (Wei et al. 1998a, b). In a first step, they prepared poly[methyl methacrylate-co-3-(trimethoxysilyl)propyl methacrylate polymers by reacting 3-(trimethoxysilyl)propyl methacrylate with methyl methacrylate. This polymer was mixed with an acidic pre-hydrolyzed tetraethoxysilane in THF with varying ratios and left to gel and dry at room temperature. Resulting materials contained from ca. 25 wt% to 100 wt% silica, with density varying from 1.32 to 1.85 g.cm 3, and were all transparent in the visible range. The specific surface area of the hybrid materials was negligible, whereas that of silica was ca. 75 m2 g 1. The condensation extent of the silica network (i.e., the percentage of Si atoms that are linked to four other Si atoms by siloxane bonds) increased with its content. In terms of mechanical properties, Vickers hardness varied from 25 to 180 kg.mm2 with increasing silica content within the hybrid systems, to be compared with 200 kg.mm2 for pure silica. Using dynamic mechanical analysis, it was shown that the Young modulus increased with silica content (from less than 2000 MPa up to 4500 MPa), whereas tan δ became negligible. At the same time, the thermal stability of the polymer could be increased by more than 50  C. Taken together, these data show that highly homogeneous hybrid materials can be obtained ranging from polymer matrices incorporating small silica domains to silica networks whose porosity is filled by small polymer domains with, in all cases, strong interactions between the organic and inorganic phases.

21

Sol-Gel Process, Structure, and Properties

511

Specific Properties Indeed, all specific properties of oxide phases are also found in sol-gel materials. They can correspond to intrinsic features of the materials, such as conductivity in tin oxide (Park and Mackenzie 1995) or photochromism in vanadium oxide (Partlow et al. 1991), or obtained by incorporation of functional molecules, polymers, or particles within the sol-gel network. Again, the main benefit of the sol-gel process lies in the possibility to process these materials in different shapes, especially as thin films which are of large interest for practical application. Well-controlled doping of the main phase with inorganic or organic species is another advantage. The access to porous systems with high specific surface area and therefore large interfaces with gas or liquid is also crucial for reactive oxides, such as catalysts (Debecker et al. 2013). However, it must be emphasized that unless special treatments are undertaken, sol-gel-based materials will be used as polycrystalline samples, which indeed impact on their properties compared to monocrystals grown by traditional ceramics technologies. Their chemistry and surface reactivity can also decrease their performance, for instance, for photocatalysis (Cano-Casanova et al. 2018) (Fig. 8).

Biological Properties To complete this overview of the properties of sol-gel materials, it is worth providing a short insight about their specific interactions with living organisms. These can be examined at two levels, first toward cells (cytocompatibility) and second toward the human body (biocompatibility).

Cytocompatibility Cytocompatibility must be considered at three stages of the sol-gel process: the sol synthesis, the gel formation, and the final material. The first stage includes the possible toxicity of the precursors and the products of the sol-gel reaction but also the physicochemical conditions (pH, ionic strength, temperature) of this reaction. The second one is mainly related to the transition from a liquid to a solid phase that may be stressful for the cells. The last one includes both interactions between gel surface and cell membrane as well as network porosity that must allow diffusion of oxygen/CO2 as well as nutrients and metabolites. Despite these many challenges, and thanks to the versatility of the sol-gel process, it has been possible to design silica gels that are compatible with the long-term survival of many different living organisms (Depagne et al. 2011). However, very few animal cells could be successfully preserved within silica gels, not because of the sol-gel conditions but due to the fact that they could not adhere on the internal surface of the inorganic network. Alumina- and titania-based hosts have also been reported (Blondeau and Coradin 2012). Several hybrid materials incorporating biological polymers were also found particularly well-suited for cell preservation.

512

T. Coradin

a

80

SBET(m2/g)

Propene conversion (%) 30 mL/min

Anatase (%) Brookite (%)

70

140 120

Rutile (%)

60

100

50 80 40 60 30 40

20

0

0 -S 2M 2 -1

2 -5

Ti

Ti

Ti

O

O

O Ti

O

G M

M 2 -1

M .8 2 -0

-S

-S

G -S

G -S M 2 -0

O Ti

b 100

G

20

G

10

2 SBET(m /g)

Propene conversion (%) 30 mL/min

90

Anatase (%) Brookite (%)

80

Rutile (%)

70

160 140 120 100

60 50

80

40

60

30 40

20

0

0 T

2M

M -H Ti

O

2 -1

2 -5

O Ti

2 -1

Ti O

.8 M 2 -0

O Ti

M -H

-H T

T M -H 2 -0

Ti O

-H T

20

T

10

Fig. 8 Comparison of the catalytic performances of TiO2 particles prepared by (a) sol-gel (SG) and (b) hydrothermal (HT) routes. Higher efficiency found for HT particles is attributed to the larger amounts of the anatase phase as well as larger specific surface area. Test reaction: propene oxidation. (Figure reproduced from Cano-Casanova et al. (2018) licensed under CC BY 4.0)

Biocompatibility Biocompatibility can be defined as a property of a material that has a beneficial impact on our body. Not only the chemical composition of the material will affect this property but its shape and structure. Among sol-gel materials, silica-based ones have been the most studied, especially as so-called bioglasses, that associate silica

21

Sol-Gel Process, Structure, and Properties

513

X103

(c)

70

Interleukin 6(IL-6) (mm2)

60 50 40 30 20 10 0 Stainless steel TiO2/316L/1 316L

Fig. 9 Sol-gel coating of stainless steel (SS) implants by titania favors bone formation (see arrows on (b) compared to (a) and decrease inflammation, as shown by the decrease of IL-6 production (c). (Figure reproduced from Marycz et al. (2014) licensed under CC BY 3.0)

with calcium, sodium, and phosphorus, for bone repair (Jones 2013). When implanted, the bioglass matrix dissolves, releasing silicic acid and the incorporated ions. However, silicic acid tends to recondense on the surface of the materials, binding calcium and favoring the formation of hydroxyapatite, the mineral phase of bone. This in turn favors the adhesion of osteoblast cells and therefore the formation of a new bone. Besides this, there is accumulating evidence that the released silicic acid can also have a direct beneficial biological effect on osteoblasts. Sol-gel processes also allow to prepare titanium oxide coatings that improve the interface between titanium implants, such as the one used for hip replacement, and the body (Marycz et al. 2014) (Fig. 9). Again the benefits of sol-gel chemistry are to control the thickness and morphology of the deposited coating to optimize its interactions with the surrounding tissue. Finally, it is important to mention that sol-gel chemistry can also be used to prepare hydroxyapatite coatings.

Conclusions and Future Directions The sol-gel process provides access to an almost unlimited range of materials, in terms of chemical composition, structure, processing techniques, shapes, and therefore properties and applications. However, while the literature offers a very high number of recipe-like protocols that would first appear relatively easy to apply, the reality of the experimental work is that the course of the sol-gel reaction is sensitive

514

T. Coradin

to a wide range of parameters, which has sometimes raised question of its reproducibility. Understanding the elementary reactions that drive the sol-gel synthesis, from the liquid to the solid state, is therefore of a prerequisite for its successful achievement. This explains why, in parallel to application-oriented research, there is still a great deal of interest for more fundamental studies, especially related to the interplay between organic and silicon moieties in organosilanes. Meanwhile, considering that many living organisms have learned to tackle the main principles of sol-gel chemistry to form mineralized tissues, it is very likely that this process will be very useful for the development of bio-based functional materials.

References Abramoff B, Klein LC (1991) Mechanical properties of silica xerogels. J Am Ceram Soc 74:1469–1471 Adachi T, Sakka S (1988) The role of N,N-dimethylformamide, a DCCA, in the formation of silica gel monoliths by sol-gel method. J Non-Cryst Solids 99:118–128 Alifanti M, Baps B, Blangenois N, Naud J, Grange P, Delmon B (2003) Characterization of CeO2ZrO2 mixed oxides. Comparison of the citrate and Sol-Gel preparation methods. Chem Mater 15:395–403 Asgari S, Bagheri H, Es-haghi A (2018) Imprinted silica nanofiber formation via sol–gel-electrospinning for selective micro solid phase extraction. New J Chem 42:13864–13872 Avnir D, Coradin T, Lev O, Livage J (2006) Recent bio-applications of sol-gel materials. J Mater Chem 16:1013–1030 Barczak M (2018) Template removal from mesoporous silicas using different methods as a tool for adjusting their properties. New J Chem 42:4181–4191 Ben Ahmed M, Masse S, Laurent G, Piquemal JY, Yéprémian C, Brayner R, Coradin T (2018) Optical microalgal biosensors for aqueous contaminants using organically doped silica as cellular hosts. Anal Bioanal Chem 410:1205–1216 Blondeau M, Coradin T (2012) Living materials from sol–gel chemistry: current challenges and perspectives. J Mater Chem 22:22335–22343 Brinker CJ, Scherer GW (1990) Sol-gel science: the physics and chemistry of sol-gel processing. Academic Press, Boston Brinker CJ, Hurd AJ, Schunk PR et al (1992) Review of sol-gel thin film formation. J Non-Cryst Solids 147–148:424–436 Calvo A, Joselevich M, Soler-Illia GJAA et al (2009) Chemical reactivity of amino-functionalized mesoporous silica thin films obtained by co-condensation and post-grafting routes. Microporous Mesoporous Mater 121:67–72 Cano-Casanova L, Amoros-Perez A, Lillo-Rodenas MA et al (2018) Effect of the preparation method (sol-gel or hydrothermal) and conditions on the TiO2 properties and activity for propene oxidation. Materials 11:2227 Cansell F, Aymonier C, Loppinet-Serani A (2003) Review on materials science and supercritical fluids. Curr Opin Solid State Mater Sci 7:331–340 Ciriminna R, Fidalgo A, Pandarus V et al (2013) The sol-gel route to advanced silica-based materials and recent applications. Chem Rev 113:6592–6620 Coradin T, Lopez PJ (2003) Biogenic silica patterning: simple chemistry or subtle biology? ChemBioChem 4:251–259 Coradin T, Durupthy O, Livage J (2002) Interactions of amino-containing peptides with sodium silicate and colloidal silica: a biomimetic approach of Silicification. Langmuir 18:2331–2336 Danks AE, Hall SR, Schnepp Z (2016) The evolution of ‘sol-gel’ chemistry as a technique for materials synthesis. Mater Horiz 3:91–112

21

Sol-Gel Process, Structure, and Properties

515

Debecker DP, Hulea C, Mutin PH (2013) Mesoporous mixed oxide catalysts via non-hydrolytic sol–gel: a review. Appl Catal A 451:192–206 Depagne C, Roux C, Coradin T (2011) How to design cell-based biosensors using the sol–gel process. Anal Bioanal Chem 400:965–976 Destino JF, Dudukovic NA, Johnson MA et al (2018) 3D printed optical quality silica and silicatitania glasses from sol-gel feedstocks. Adv Mater Technol 3:1700323 Mann S, Burkett SL, Davis SA et al (1997) Sol–gel synthesis of organized matter. Chem Mater 9:2300–2310 Ehrlich H, Demadis KD, Pokrosky OS et al (2010) Modern views on desilicification: biosilica and abiotic silica dissolution in natural and artificial environments. Chem Rev 110:4656–4689 Faustini M, Nicole L, Ruiz-Hitzky E et al (2018) History of organic-inorganic hybrid materials: prehistory, art, science, and advanced applications. Adv Funct Mater 28:1704158 Garay JE (2010) Current-activated, pressure-assisted densification of materials. Annu Rev Mater Sci 40:445–468 Gawel B, Gawel K, Oye G (2010) Sol-gel synthesis of non-silica monolithic materials. Materials 3:2815–2833 Harris FW (1981) Introduction to polymer chemistry. J Chem Educ 58:837–843 Hench LL, West JK (1990) The sol-gel process. Chem Rev 90:33–72 Iler RK (1979) The chemistry of silica. John Wiley, New York Jolivet JP (2000) Metal oxide chemistry and synthesis. Wiley, New York Jones JR (2013) Review of bioactive glass: from Hench to hybrids. Acta Biomater 9:4457–4486 Kickelbick G (ed) (2007) Hybrid materials: synthesis, characterization, and applications. Weinheim, Wiley-VCH Verlag GmbH Kolb D, Kolb KE (1979) The chemistry of glass. J Chem Educ 56:604–608 Levy D, Zayat M (eds) (2015) The sol-gel handbook. Weinheim, Wiley-VCH Verlag GmbH Livage J, Henry M, Sanchez C (1988) Sol-gel chemistry of transition metal oxides. Prog Solid St Chem 18:259–341 Loy DA, Baugher BM, Baugher CR et al (2000) Substituent effects on the sol-gel chemistry of Organotrialkoxysilanes. Chem Mater 12:3624–3632 Marycz K, Krzak J, Urbanski W et al (2014) In vitro and in vivo evaluation of sol-gel derived TiO2 coatings based on a variety of precursors and synthesis conditions. J Nanomater 2014:350579 Papirer E (ed) (2000) Adsorption on silica surfaces. Marcel Dekker, New York Park SS, Mackenzie JD (1995) Sol-gel-derived tin oxide thin films. Thin Solid Films 258:268–273 Partlow DP, Gurkovich SR, Radford KC et al (1991) Switchable vanadium oxide films by a sol-gel process. J Appl Phys 70:443–452 Patarin J, Lebeau B, Zana R (2002) Recent advances in the formation mechanisms of organized mesoporous materials. Curr Opin Colloid Interface Sci 7:107–115 Pierre AC (ed) (1998) Introduction to sol-gel processing. Springer, New York Ponthieu E, Payen E, Pajonk GM et al (1997) Comparison of drying procedures for the preparation of alumina powders with the system Al-Alkoxide/tertiary Butanol/water. J Sol-Gel Sci Technol 8:201–206 Reetz MT, Zonta A, Simpelkamp J (1996) Efficient immobilization of lipases by entrapment in hydrophobic sol-gel materials. Biotechnol Bioeng 49:527–534 Sakka S (ed) (2005) Handbook of sol-gel science. Springer Science+Business Media, New York Sarawade PB, Kim JK, Hilonga A et al (2011) Synthesis of hydrophilic and hydrophobic xerogels with superior properties using sodium silicate. Microp Mesop Mater 139:138–147 Shea KJ, Loy DA (2001) Bridged polysilsesquioxanes. Molecular-engineered hybrid organicinorganic materials. Chem Mater 13:3306–3319 Shimizu T, Kanamori K, Maeno A et al (2016) Transparent, highly insulating polyethyl- and polyvinylsilsesquioxane aerogels: mechanical improvements by vulcanization for ambient pressure drying. Chem Mater 28:6860–6868 Silverman BM, Wieghaus KA, Schwartz J (2005) Comparative properties of Siloxane vs phosphonate monolayers on a Kay titanium alloy. Langmuir 21:225–228

516

T. Coradin

Soler-Illia GJAA, Azzaroni O (2011) Multifunctional hybrids by combining ordered mesoporous materials and macromolecular building blocks. Chem Soc Rev 40:1107–1150 Spitz RN, Barton JE, Barteau MA et al (1986) Characterization of the surface acid-base properties of metal oxides by titration/displacement reactions. J Phys Chem 90:4067–4075 Turova NY, Turevskaya EP, Kessler VG et al (2002) The chemistry of metal alkoxides. Kluwer Academic Publishers, Dordrecht Vioux A (1997) Nonhydrolytic sol-gel routes to oxides. Chem Mater 9:2292–2299 Wei Y, Jin D, Brennan DJ et al (1998a) Atomic force microscopy study of organic-inorganic hybrid materials. Chem Mater 10:769–772 Wei Y, Jin D, Yang C et al (1998b) Organic-inorganic hybrid materials: relations of thermal and mechanical properties with structures. Mater Sci Eng C 6:91–98 Yang Y, Li LH, Li Y et al (2017) Hierarchically porous materials: synthesis strategies and structure design. Chem Soc Rev 45:481–558

Acoustic Transducer and Its Applications in Biosensors

22

Junyu Zhang, Qian Wu, Xi Zhang, Hao Wan, and Ping Wang

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acoustic Transducer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acoustic Wave Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Quartz Crystal Microbalance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Film Bulk Acoustic Resonator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Rayleigh Wave . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Shear-Horizontal Surface Acoustic Wave . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Surface Transverse Wave . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Love Wave . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Shear-Horizontal Acoustic Plate Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Flexural Plate Wave . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions and Future Directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

518 519 520 523 525 526 527 528 529 530 531 533 533

Abstract

An acoustic transducer is a device that converts acoustic energy and electrical energy into each other. When acoustic waves propagate through a medium, its physical properties can be affected by many factors. According to these phenomenon, a wide variety of acoustic wave sensors can be manufactured by means of the acoustic transducer, including quartz crystal microbalance, film bulk acoustic resonator, Rayleigh wave sensor, shear-horizontal surface acoustic wave sensor, surface transverse wave sensor, Love wave sensor, shear-horizontal acoustic plate mode sensor, and flexural plate wave sensor. Nowadays, acoustic wave devices

J. Zhang · Q. Wu · X. Zhang · H. Wan · P. Wang (*) Biosensor National Special Laboratory, Key Lab for Biomedical Engineering of Ministry of Education, Department of Biomedical Engineering, Zhejiang University, Hangzhou, China e-mail: [email protected] © Springer Nature Switzerland AG 2022 G. Thouand (ed.), Handbook of Cell Biosensors, https://doi.org/10.1007/978-3-030-23217-7_65

517

518

J. Zhang et al.

have been widely used for the detection of mass, viscosity, conductivity, and density. They have advantages of high precision, high sensitivity, easy integration, good reliability, small size, light weight, and low power consumption. Keywords

Acoustic transducer · Interdigital electrodes · Acoustic wave biosensor

Introduction More and more sensing technologies are currently available for biochemical sensing. Among many sensing technologies, electrochemical, optical, and acoustic waves are widely regarded as the most promising biochemical sensor technologies. Most optical and electrochemical detection processes require labeling of the test sample, such as ELISA, which increases the cost of the sample analysis and the operational complexity. Electrochemical sensing is a very mature technology. The technique generally utilizes the electrochemical properties of the analyte itself or the electrochemical signal generated by the reaction between the sensitive component and the analyte, and then converts the electrochemical signals into identifiable by means of an electrochemical sensor. The electrical signals are used for processing and analysis later. Nowadays, most of the common glucose and lactic acid detection devices on the market are based on electrochemical sensing technology (Coté et al. 2003). Optical sensing is currently the most common and direct technology in biochemical detection applications. The mechanisms of optical sensing include interference, infrared absorption, scattering, fluorescence, and polarization measurements. Although optical sensors are very sensitive, they are more expensive than other technologies (Sant et al. 2003; Webb 2002). Among these optical sensing technologies, fluorescence and surface plasmon resonance (SPR) are two types that are used more. Fluorescence detection is one of the most sensitive detection techniques (Petryayeva et al. 2013), but the labeling operation is required for the sample to be tested, and the detection system is also expensive and complicated. This is also the biggest limiting factor for this technology. The SPR sensor detects the variation of the refractive index of the surface layer after the biochemical reaction. SPR sensors can be measured without marking the sample. But the disadvantage of this method is the high demand for the light source and the inability to achieve high-throughput detection. Acoustic sensing is primarily based on the development of piezoelectric resonators used in RF communication technology over the past few decades. It is often used as filters and resonators (Lec 2014). In 1959, Sauerbrey first reported the use of acoustic sensors to detect mass deposition techniques (Sauerbrey 1959), which was achieved by detecting changes in the resonant frequency of the resonator caused by the mass attached to the surface of the sensor. This discovery set up a new sensing method for biochemical detection in gas and liquid media.

22

Acoustic Transducer and Its Applications in Biosensors

519

Acoustic Transducer An acoustic transducer is a device that receives an electrical (or acoustic) input signal and converts it into an acoustic (or electrical) output signal such that certain desired characteristics of the input signal are reflected in the output signal. It is usually interdigital electrodes (IDTs) structure. IDTs were originally proposed by White and Voltmer in 1965, and they use this structure to effectively excite surface acoustic waves on piezoelectric substrates (White and Voltmer 1965a). The most simplified structure of IDTs consists of two similar comb-shaped metal electrode fingers that periodically overlap each other. Figure 1a shows the basic structure of a single IDTs. The length of the two electrode fingers and the gap between them is one acoustic wavelength (p). The width of the finger and the gap are both p/4. The width of the overlapping portion of the two electrodes is the acoustic aperture (W). Two electrodes form a single IDTs, one earthed and the other connected to the RF signal. Figure 1b shows the standard frequency response curve for IDTs, where A(f) is the amplitude of the RF signal. When the wavelength of the acoustic wave excited by the RF signal is equal to the period p of the IDTs, the amplitude of the response A(f) is the maximum. In addition, the bandwidth B of the response curve in the figure can be narrowed by increasing the number of fingers of the interdigital electrodes properly. Research has shown that the number of the fingers is at most 100 pairs. Because when more than 100, the signal scattering and mass deposition loss of the electrode will increase significantly, affecting the signal quality. In order to improve the performance and stability of the sensor, it is important to reduce the insertion loss. In addition to using piezoelectric substrates with high electromechanical coupling coefficients, the acoustic loss of the sensor can be improved by optimizing the type of IDTs, IDTs length, acoustic aperture, and delay line length. Among them, the type of IDTs is the most important parameter to achieve low insertion loss while improving stability. Figure 2 shows the structure of two types of IDTs. The biggest advantage of the traditional IDTs structure (Fig. 2a) is that the IDTs have wider fingers and the processing is easy to implement. However, this structure also has its obvious disadvantages. For example, Bragg reflection results in higher insertion loss and poor frequency stability. Using the

Fig. 1 (a) Basic structure of individual IDTs; (b) Standard frequency response curve of IDTs

520

J. Zhang et al.

Fig. 2 Two types of IDTs: (a) IDTs of traditional structures; (b) IDTs of split-finger structures

split-finger structure (Fig. 2b) can significantly reduce the acoustic reflection effect. The capacitance of the split-finger IDTs structure is about 1.4 times that of the traditional IDTs, and the electrical anti-interference ability is also stronger than the traditional IDTs (Hartmann and Abbott 1989). With different kinds of acoustic transducers and related materials, plenty of acoustic devices can be manufactured.

Acoustic Wave Devices According to the acoustic wave conduction process and propagation mode, acoustic wave devices can be classified into the following three categories: Bulk Acoustic Wave (BAW) devices, Surface Acoustic Wave (SAW) devices, and Acoustic Plate Mode (APM) devices. In BAW devices, acoustic waves propagate inside the substrate without guiding. In SAW devices, the sound waves propagate along one surface of the substrate, both guided and unguided. While in APM devices, the sound waves are propagated under the guidance of the reflection on multiple surfaces. Both SAW and APM devices can be classified as Surface Generated Acoustic Wave (SGAW) devices, as shown in Fig. 3, since both devices generate and detect acoustic waves on the surface of the piezoelectric substrate. So SGWA devices have many similar working principles.

22

Acoustic Transducer and Its Applications in Biosensors

521

Fig. 3 Classification of acoustic wave devices

Besides, the acoustic wave device can be classified according to the vibration direction (transverse or longitudinal) of the particle relative to the wave propagation direction, or according to the vibration direction (horizontal or vertical) of the particle relative to the device surface. The longitudinal vibration (or compression wave) particle vibration direction is parallel to the wave propagation direction, and the transverse wave (or shear wave) particle vibration direction is perpendicular to the wave propagation direction, as shown in Fig. 4. The direction of particle vibration of the vertical wave is perpendicular to the surface of the device, while the direction of particle vibration of the horizontal wave is parallel to the surface of the device. When the acoustic wave device is used in a liquid environment, the surface of the substrate is usually in contact with the liquid medium, and if the substrate is excited by a vertical wave, the vertical wave transmits to the liquid closed to the substrate, causing severe signal attenuation. In liquid environment applications, it is of great necessity to minimize the attenuation of acoustic energy into the liquid medium. Therefore, shear-horizontal waves are widely used in this application to avoid the above problems and have good performance. In general, the most common acoustic biosensors are based on Thickness Shear Mode (TSM) devices. The TSM device is one of the BAW devices, the most typical of which is the quartz crystal microbalance (QCM). As the most commonly used biosensor, QCM has been developed in detail for more than 50 years and has become a mature commercial product with low price and good stability (Janshoff et al. 2000). However, in recent years, more and more studies have reported that acoustic wave devices can still work effectively under liquid medium contact, and they even have better sensitivity than traditional QCM sensors. These devices include Thin Film Bulk Acoustic Resonators (FBAR), Shear-Horizontal Surface Acoustic Wave (SH-SAW), Rayleigh Wave (RW), Love Wave (LW), Surface Transverse Wave (STW), Shear-Horizontal Acoustic Plate Mode (SH-APM), and Flexural Plate Wave (FPW). Figure 5 shows a top and cross-sectional view of some of those devices, and the arrows in the cross-section indicate the direction of vibration of the wave particles.

522 Fig. 4 Two modes of motion of particles relative to the direction of wave propagation. The figure above is a longitudinal wave or a compression wave, the figure below is a shear wave or a transverse wave. The black arrow represents the direction of wave propagation, the red represents the direction of particle motion

Fig. 5 Schematic diagram of various types of acoustic wave devices

J. Zhang et al.

22

Acoustic Transducer and Its Applications in Biosensors

523

The biggest limitation in developing BAW devices as sensors is the sensitivity. The sensitivity of such devices depends primarily on the thickness of the piezoelectric substrate layer. The reduced thickness of the piezoelectric substrate can raise the operating frequency of the BAW device. According to the working principle of the BAW device, the higher the frequency, the higher the detection sensitivity of the device, because the penetration depth of the acoustic energy into the adjacent liquid medium is reduced (Smith and Hinson-Smith 2006). It makes the device very sensitive to any subtle changes in the surface of the substrate, such as mass deposition. In addition, such devices will become more sensitive to other factors, such as the viscoelastic properties of the coating, electrical conductivity, liquid viscosity, temperature, and mechanical stress (Grate et al. 1993). However, thinning the BAW substrate can make the device fragile and more difficult to operate. APM devices also face the same problems as BAW. To increase sensitivity, the thickness of the substrate must be reduced. However, APM devices are more robust than general BAW devices. SAW devices were originally used in the electronics and communications fields as filters and resonators. In the last two decades, SAW devices have gradually been applied to the field of sensing. Currently, SAW devices have been widely used as chemical sensors in the field of gas and liquid detection. Generally, SAW devices operate at higher frequencies than QCM (Länge et al. 2008), and the acoustic energy of such devices is limited to the surface of the device (Gronewold 2007). The following sections elaborate the acoustic sensors, which are commonly used in biochemical detection applications, including their structure, characteristics, operating principles, operating frequencies, and their strengths and weaknesses.

Quartz Crystal Microbalance The most classic QCM devices are fabricated from AT-cut quartz crystal sheets, as shown in Fig. 6. The sensor electrodes are distributed at the front and back ends of the quartz wafer, and the sound waves are excited by applying a radio frequency signal to the electrodes. The QCM excites shear-horizontal waves, so the QCM sensor can operate in a liquid environment (Kanazawa and Gordon 1985).

Fig. 6 (a) quartz crystal microbalance; (b) commercial QCM inspection platform

524

J. Zhang et al.

In 1959, Sauerbrey proposed the relationship between QCM resonance frequency variation and sensor surface deposition mass density, which led to the increasing use of QCM devices in sensing applications. The QCM theoretical mass sensitivity Sσ is proportional to the square of the resonant frequency f0, expressed as (Sauerbrey 1991): Sσ ¼


Δf 2 ¼ f 2 ¼ 2:267  1015 f 0 2 Hz∙cm2 ∙ng1 Δσ ρq ∙vq 0

ð1Þ

where Δf is the frequency offset due to surface mass deposition, Δσ is the variation of the deposited mass density above the sensor surface, and ρq is the density of the quartz. vq is the horizontal wave propagation velocity of the AT-cut quartz crystal, and f0 is the center frequency of the sensor. However, the linear relationship of the Sauerbrey formula can only be established under a certain condition, that is, the surface layer of the sensor must be very thin. Under this condition, the substance attached to the surface of the quartz maintains its rigid characteristics and undergoes a non-negligible deformation as the acoustic wave passes. This makes the elastic properties of the substance not affect the resonant frequency of the sensor, so the resonant frequency of the sensor is entirely related to the surface quality. This assumption is inaccurate when the thickness of the surface deposit becomes thicker. For thicker deposits, the sensor is more sensitive to viscoelasticity, so the Sauerbrey formula does not work in this case. It can be seen from Eq. 1 that the theoretical mass sensitivity of the QCM sensor is completely dependent on the inherent material properties and resonant frequency of the sensor. This makes QCM sensors ideal for sensing applications. Another important evaluation parameter for sensors is the Limit of Detection (LOD). Unlike sensitivity, LOD depends not only on the characteristics of the sensor material, but also on the sensor’s detection system. The detection system limits the minimum signal that can be distinguished and detected from system noise. QCM technology has great application prospects in biochemical detection. At present, QCM sensors have expanded many detection applications, including interactions between different types of molecules, as well as detection of peptides (Furtado et al. 1999), proteins (Ben-Dov et al. 1997), nucleotides (Höök et al. 2001), phage (Hengerer et al. 1999), viruses (Zhou et al. 2002), bacteria (Fung and Wong 2001), and cells (Richert et al. 2002). It has recently been reported that QCM sensors can be applied to the detection of DNA strands and transgenic organisms (Stobiecka et al. 2007). Although QCM technology has been widely used in various sensing fields, its sensitivity and LOD still need to be further improved. The 30 MHz QCM has a sensitivity of 66.6 cm2/g and a mass resolution of approximately 10 ng/ cm2 (Lin et al. 1993). Currently, the sensitivity and LOD are more and more higher requested by many sensing applications. As can be seen from the Sauerbrey formula, if you want to increase the mass sensitivity of the sensor, you must increase the center frequency of the QCM. But increasing the frequency will bring new problems that do not exist at low frequencies. The most urgent problem to be solved now is the sensor characterization detection system and the batch processing of the device.

22

Acoustic Transducer and Its Applications in Biosensors

525

When the thickness of the substrate is reduced, the device is very fragile. So this is why the commercial QCM system on the market has a frequency of only about 10 MHz or less, and there are no commercial sensors with hundreds of MHz.

Film Bulk Acoustic Resonator A typical film bulk acoustic resonator includes two metal layers and a piezoelectric film (such as ZnO or AlN) sandwiched between metal layers. Figure 7 shows the structure of FBAR (Link 2006). In the past few years, silicon-based FBARs have been used as filters in RF device (Vale et al. 1990). Gabl et al. first proposed the use of FBAR for biochemical sensing (Gabl et al. 2011). FBAR works like QCM, but FBAR devices with GHz operating frequency can be achieved with piezoelectric film thicknesses from 100 nm to several μm. The main advantage of FBAR technology is that it can be directly processed on a silicon substrate using micro-manufacturing technology, so it is compatible with CMOS technology. This is a necessary condition for the integrated electronic process to manufacture the sensor and its system. It indicates that the miniature sensor system can be manufactured on a large scale and low cost. However, the miniaturization of the sensor device should be synchronized with the optimized design of the microfluidic system, as the flow control system is critical to reducing system noise and improving system stability. Unfortunately, the main problems of the current microfluidic system are the complex integration and high cost. According to Eq. 1, the sensitivity of the FBAR device is much higher than that of the conventional QCM due to its higher resonant frequency. Thin film acoustic wave sensing technology has been able to achieve a film bulk acoustic resonator with good performance in liquid media environments and operating frequencies up to 2 GHz (Gabl et al. 2004; Bjurstrom et al. 2006). Weber et al. reported a ZnO-based film bulk acoustic resonator used for biosensor detection applications with a sensitivity of 740.5 cm2/g and a minimum LOD of 2.3 ng/cm2. Although the high operating frequency of the FBAR device will increase the sensitivity of the device, it also Fig. 7 The structure of FBAR

526

J. Zhang et al.

increases the noise level of the detection system. In addition, because of the smaller size of the FBAR device, the material’s boundary effects have a greater impact on device performance than QCM. These problems can affect the noise of the system, which deteriorates the detection resolution of the sensor. At present, it is reported that only the detection resolution of the FBAR sensor detection system based on network analyzer is optimal, but it is only very close to the traditional 10 MHz QCM sensor (Weber et al. 2006; Wingqvist et al. 2007).

Rayleigh Wave In 1885, Lord Rayleigh first proposed the concept of surface acoustic waves when studying seismic waves. He analyzed the properties of this elastic surface wave propagating in an isotropic medium and pointed out that it plays an important role in seismic and elastic solid collisions (Francis 2000). This type of wave propagates along one side of the infinite space, and the amplitude decays exponentially as the depth of the substrate increases. The energy of such waves is mainly concentrated between the surface and one wavelength below the surface. In 1965, White and Voltmer reported a new type of electrode structure, Interdigital Transducers (IDTs) (White and Voltmer 1965b). They used this electrode to excite surface acoustic waves on the piezoelectric material, which promoted the development of SAW devices. The working principle of the Rayleigh wave device is as follows: a sinusoidal voltage Vin (radio frequency signal) is applied to the input port of the device. The input IDTs deposited on the piezoelectric substrate are used to convert the mechanical acoustic wave into the electrical signals due to the inverse piezoelectric effect. The excited acoustic waves are transmitted along the surface of the substrate. When the acoustic waves reach the output IDTs, they are converted into electrical signals Vout. On one hand, any physical change near the surface of the device will cause a change in the velocity and amplitude of the wave that propagates through it. Then it can be converted to the variation of the amplitude and phase of the electrical signal, which can be detected using an instrument, such as network analyzers. On the other hand, the velocity of acoustic wave can be characterized by the resonant frequency of the device. Therefore, by detecting the electrical characteristics of the device, subtle changes on the surface of the device can be obtained, so that the Rayleigh wave device can be used as a sensor with this principle. Since the vibration direction of the surface particles of the Rayleigh wave has a component perpendicular to the surface of the device, when the surface contacts the liquid medium, a considerable portion of the acoustic energy penetrates into the liquid, resulting in severe attenuation of the sensor signal, so the Rayleigh wave Devices are generally used only when detecting gases. The size and physical properties of the piezoelectric substrate, as well as the dimensional parameters of the IDTs, affect the optimal resonant frequency of the acoustic waves produced by the device. The most common materials used as piezoelectric substrates are quartz, lithium niobate, zinc oxide, and aluminum

22

Acoustic Transducer and Its Applications in Biosensors

527

nitride. The general Rayleigh wave device operates at 40–400 MHz and has a sensitivity of 100–200 cm2/g in gas detection.

Shear-Horizontal Surface Acoustic Wave As mentioned above, Rayleigh SAW devices have severe signal attenuation in liquid environments. In order to improve the SAW device to be used in a liquid environment, SH-SAW has been developed in recent years. It works like Rayleigh wave device. The surface acoustic wave is generated by the input IDTs on the piezoelectric substrate, and then the signal can be detected at the output IDTs (Fig. 8a). The main difference between Rayleigh wave devices and this device is the vibrational direction of the acoustic particles, and the piezoelectric substrate material is the most important factor. The first application of the SH-SAW sensor is the leaky-wave mode SAW, which is only partially confined to the device surface. Leaky-wave mode SAW is primarily a shear-horizontal wave, but it is not a pure shear-horizontal wave. So the attenuation of the acoustic signal is increased in a liquid environment (Baer and Flory, n.d.). Leaky wave penetrates into the surface of the device (Fig. 8b), so the sensitivity to changes in the surface of the device is not high. In addition, the most important drawback of existing SH-SAW devices on the market is that IDTs must be isolated from the liquid to function properly. Therefore, an additional protective layer is required at the top of the IDTs, such as polyimide (Wessa et al. 1998), parylene (Barié et al. 1998), or polystyrene (Deobagkar et al. 2005a). In addition, another method is to use a flow cell that allows the SH-SAW device to work more efficiently in a liquid environment and to avoid IDTs being corroded by liquid. SH-SAW devices can be used for gas and liquid detection. The device typically operates at 30–500 MHz with a sensitivity of 100–180 cm2/g. The most commonly used substrate material for the device is 36 YX lithium niobate, other materials such as ST-cut quartz (Deobagkar et al. 2005b), 41 YX lithium niobate (Hechner and Soluch 2005), 64 YX lithium niobate (Chivukula et al. 2007). Potassium citrate has also been reported as a piezoelectric substrate for SH-SAW devices. The dielectric constant ε is an important parameter for selecting a piezoelectric substrate material. If the sensor is operated in an aqueous environment, the dielectric constant of

Fig. 8 (a) Structure of SH-SAW device; (b) SH-SAW or leaky wave propagation mode

528

J. Zhang et al.

the material should be close to the dielectric constant of water (εr ≈ 80) to reduce the capacitive effect at the IDTs, for example, using lithium niobate (εr ¼ 47). If the piezoelectric substrate is made of quartz material (εr ¼ 4.7), it can be found that the acoustic-electrical coupling drops sharply and the electrical impedance is mismatched, which causes the IDTs to be short-circuited in the water environment (Rapp et al. 1995).

Surface Transverse Wave The surface shear wave is derived from the surface skimming bulk wave (SSBW). This acoustic wave travels very close to the surface of the substrate, but does not propagate completely along the surface. The orientation of the substrate material and IDTs determines whether the substrate excites SSBW or leaky waves. The difference between the SSBW and the leaky wave is the angle of acoustic wave propagation, and the angle of the leaky wave γl is larger than that of the SSBW γs (Figs. 8b and 9b). Figure 9a is a schematic diagram of the structure of the STW device. The metal gate on the surface of the device between the input and output IDTs reduces the speed of acoustic wave and captures the acoustic energy on the surface of the substrate, thus improving the sensitivity of the device. Therefore, STW can be considered as the SSBW with the metal gate. In 1977, Milson and Lewis first reported the SSBW devices (Milsom et al. 1977; Lewis 1977). However, the signal of the device generates a huge loss after being propagated into the substrate. Until 1987, Bagwell and Bray proposed a new dualport resonator structure STW device with the metal gate on the surface that captures acoustic energy on the surface of the substrate. The device can have a Q value of up to 5600 without load (Bagwell and Bray 1987). Its surface particle vibration is shearhorizontal, therefore it is suitable for sensing in gas and liquid media. Metal gate has certain advantages over slab in waveguide. The metal gate can be matched to the transducer to prevent acoustic reflection and provide higher waveguide capability (Tom-Moy et al. 1995a). Substrate materials with metal gates also

Fig. 9 (a) Structure of the STW device; (b) SSBW and STW propagation modes

22

Acoustic Transducer and Its Applications in Biosensors

529

have higher sensitivity (Baer et al. 1992). STW has proven to outperform traditional Rayleigh SAW devices in many parameters. STW devices have some advantages, such as low propagation losses, are fast-responding, and have high sensitivity to external influences. According to some reports, its general operating frequency is 30–300 MHz, and the surface quality sensitivity is 100–200 cm2/g (Ferrari and Lucklum 2009). Although there has been a lot of research on the STW resonant structure, there are few documents that give the optimal parameter design and satisfactory test results. This is due to the huge loss of sound waves propagating inside the substrate, which causes great difficulties for analysis and limits its use in sensing detection.

Love Wave The physical phenomenon of Love wave was originally discovered by mathematician Augustus Love (Love 2015). He observed that although this wave is far from the earthquake center, it is the most destructive (Voinova 2009). This is because the wave velocity is lower when the seismic wave propagates along the geological layer, and the wave energy decays slowly (Fig. 10). The Love wave device is a layered structure, which includes a piezoelectric substrate that excites SSBW, IDTs that excite and receive acoustic waves, and a waveguide layer on them. Therefore, the Love wave device can be considered as a

Fig. 10 (a) Schematic diagram of Love wave device structure; (b) Love wave surface particle vibration mode

530

J. Zhang et al.

SSBW containing a waveguide layer. Under this structure, IDTs are well isolated from the external environment such as liquids to reduce their interference from the outside world. The key premise of Love wave formation is that the shear rate of the waveguide layer material is smaller than that of the substrate (Du et al. 1996). The different mechanical properties between the waveguide layer and the substrate cause the acoustic energy to be concentrated in the waveguide layer and slow the wave propagation speed. The acoustic energy gathering effect of this structure reduces the penetration of sound waves on the substrate. The acoustic waves collected above the waveguide layer make the device very sensitive to any changes on the surface, such as mass load, viscosity, and conductivity change. The sensitivity of the device is related to the acoustic wave aggregation effect. The stronger the acoustic energy aggregation, the higher the sensitivity (Rocha-Gaso et al. 2009). The surface particle vibration of the Love wave device is horizontally sheared, so it can be used for sensing in gas and liquid media. The parameters that determine the resonant frequency of the device are the material of the device substrate and waveguide layer, the length of the IDTs period as well as the thickness of the waveguide layer. According to some research, the device usually operates at 80– 300 MHz and has a surface quality sensitivity of 150–950 cm2/g (Arnau et al. 2008). In 1992, Gizeli et al. first reported a Love wave biosensor based on PMMA/ST-cut quartz (PMMA is the waveguide layer material and ST-cut quartz is the piezoelectric base material) (Gizeli et al. 1992a). However, the biggest drawback of this device is its poor temperature stability. Therefore, temperature compensation devices based on various types of quartz and lithium niobate substrates have been studied to solve the temperature stability problem. In addition, the waveguide layer material of the Love wave device may be selected from the polymer (Gizeli et al. 1992b), SiO2 (Herrmann et al. 2001), ZnO (Powell et al. 2002) etc.

Shear-Horizontal Acoustic Plate Mode The concept of the SH-APM was proposed in the 1980s as an acoustic plate wave device with shear-horizontal particle vibration. Therefore, SH-APM devices can be used in both gas and liquid environments. The basic structure of this type of device is shown in Fig. 11. The SH-APM works like the previous ones, using IDTs to excite acoustic waves on the substrate. Since the thickness of the substrate is thinner than the wavelength of the acoustic wave, the acoustic wave excited by the substrate is a plate acoustic wave rather than a surface wave. The advantage of SH-APM devices is that IDTs can be located on the back side of the device, away from sensitive areas of detection, to avoid contact with liquids and the signal degradation. As discussed before, the parameters that determine the resonant frequency of sensor are the substrate material, the thickness of the substrate, and the period length of the IDTs. It is reported that the typical operating frequency of SH-APM sensor is between 20–200 MHz, and the surface quality sensitivity is 20–50 cm2/g (Ballantine et al. 1996). There have been many studies using SH-APM devices to measure surface quality deposition in liquid phase environments, as well as applications for

22

Acoustic Transducer and Its Applications in Biosensors

531

Fig. 11 Schematic diagram of SH-APM device structure

the detection of biomolecules (Andle et al. 1991). However, the application of this type of APM is still hindered by related design techniques. The main problem in the design process is to excite and detect single-mode acoustic waves on a plate substrate. In fact, the substrate usually excites multiple modes of acoustic waves with close frequency at the same time, which results in the frequency hopping phenomenon when using the oscillation circuit. This is why it is difficult to implement signal detection for the SH-APM with a standard oscillating circuit. Another important drawback is that the mechanical and electrical loads on the surface can affect the SH-APM’s response, especially when using materials with higher electromechanical coupling coefficients such as lithium niobate, and the acoustic-electric effect is aggravated (Bender et al. 1997).

Flexural Plate Wave FPW sensors are typically implemented using a plate with a thickness less than a wavelength of the acoustic wave. Limiting the sonic energy to a very thin film layer gives the sensor a high sensitivity of mass. As shown in Fig. 12, the FPW sensor is typically a composite structure fabricated on a silicon wafer, including a silicon nitride layer, an aluminum ground plane, and a zinc oxide piezoelectric layer formed by sputtering. The acoustic waves excited by the device are vibrating in the vertical direction. However, its FPW wave velocity is smaller than the propagation velocity of the compression wave in the liquid, so it can be used for detection applications in liquid phase environments. Therefore, the acoustic wave of the device is not coupled into the liquid, so that no dissipation of large acoustic energy is generated. The theoretical mass sensitivity of the FPW device can be calculated from the formula Sσ ¼ 1/2ρd, where ρ is the plate density and d is the plate thickness. Therefore, reducing the thickness of the plate can increase the sensitivity of the device. According to the literature, the mass sensitivity of FPW devices is about

532

J. Zhang et al.

Fig. 12 Schematic diagram of FPW device

200–1000 cm2/g (White et al. 1987). However, if applied in a liquid phase environment, the sensitivity of the device will decrease from 1000 cm2/g to 200 cm2/g (Tom-Moy et al. 1995b; Wenzel and White 1988). The operating frequency of FPW devices is typically 2–20 MHz, depending on the device material, the length of the IDTs, and the thickness of the plate. The device also provides high quality sensitivity based on low operating frequency, which will reduce the difficulty of integration of the detection system. However, the main disadvantage of FPW is that the thickness of the plate will be very thin in order to achieve higher sensitivity, resulting in its frangibility and increasing the difficulty of micromachining. In general, the sensitivity of the QCM is primarily dependent on the center frequency of the device. In order to increase the center frequency, the thickness of the wafer has to be reduced, which makes the QCM device very fragile. FBAR has excellent mass sensitivity and offers great promise for device miniaturization and CMOS-compatible. However, the main disadvantages of such devices are high research cost, high complexity of micro-process, and high operating frequency of the device, which will increase the difficulty of designing the circuit detection system. Since the Rayleigh SAW does not work properly in a liquid environment, it is not suitable as a biosensor. The other two STWs and Lebo waves can be considered as derivatives of the SH-SAW device. Because these two devices concentrate sound energy on the sensor surface, they have excellent sensitivity. In addition, the Love Wave device has better stability than QCM and can be integrated to form a sensor array. Compared with QCM, SH-APM has no significant improvement in mass sensitivity. In addition, SH-APM devices are prone to frequency hopping when detected by an oscillating circuit, and the signal output is unstable and accurate. FPW devices have good mass sensitivity but low operating frequency, which is the biggest advantage of this device. However, in a liquid medium environment, the mass sensitivity of the device is greatly reduced. After these comparisons, SGAW devices seem to be more suitable for use in biometric applications. Among SGAW devices, Love Wave seems to be more popular with researchers. This is because the Love Wave surface acoustic wave sensor is the most sensitive one in the liquid phase environment of the SGAW device.

22

Acoustic Transducer and Its Applications in Biosensors

533

Conclusions and Future Directions The acoustic transducer is the most important part of an acoustic wave biosensor. The IDTs have promoted the further development of acoustic wave biosensors. So far, acoustic transducer acoustic wave biosensors have been developed and widely used in many applications such as disease diagnosis, drug efficacy evaluation, environmental monitoring, and food analysis. Great efforts will be made for developing novel materials, designing sensor structure, optimizing surface modification, and exploiting new signal amplification strategies which result in better performance of acoustic wave biosensors with high sensitivity, low LOD, small size, high throughput, and good anti-interference ability. Besides, acoustic wave biosensors are integrated with other technologies such as gas chromatography (GC), surface plasmon resonance (SPR), and microfluidic chips to design multiparameter systems for some multi-analyte applications. In this regard, artificial intelligence are also important to achieve more reliable detection results by training suitable models. It is firmly believed that the acoustic wave biosensors will play important roles in biosensing applications in the future.

References Andle JC, Vetelino JF, Lade MW, McAllister DJ (1991) An acoustic plate mode device for biosensor applications. IEEE:483–485 Arnau A, Ferrari V, Soares D, Perrot H (2008) Interface electronic systems for AT-cut QCM sensors: a comprehensive review. In: Vives A (ed) Piezoelectric transducers and applications. Springer, Berlin/Heidelberg, pp 117–186 Baer RL, Flory CA (n.d.) Some limitations on the use of leaky SAW mode sensors in liquids. IEEE:279–284 Baer RL, Flory CA, Tom-Moy M, Solomon D (1992) STW chemical sensors. IEEE:293–298 Bagwell TL, Bray RC (1987) Novel surface transverse wave resonators with low loss and high Q. IEEE:319–324 Ballantine DS Jr, White RM, Martin SJ, Ricco AJ, Zellers ET, Frye GC et al (1996) Acoustic wave sensors: theory, design and physico-chemical applications. Academic Press Barié N, Rapp M, Sigrist H, Ache HJ (1998) Covalent photolinker-mediated immobilization of an intermediate dextran layer to polymer-coated surfaces for biosensing applications. Biosens Bioelectron 13:855–860 Bender F, Meimeth F, Dahint R, Grunze M, Josse F (1997) Mechanisms of interaction in acoustic plate mode immunosensors. Sensors Actuators B Chem 40:105–110 Ben-Dov I, Willner I, Zisman E (1997) Piezoelectric immunosensors for urine specimens of Chlamydia trachomatis employing quartz crystal microbalance microgravimetric analyses. Anal Chem 69:3506–3512 Bjurstrom J, Wingqvist G, Katardjiev I (2006) Synthesis of textured thin piezoelectric AlN films with a nonzero c-axis mean tilt for the fabrication of shear mode resonators. IEEE Trans Ultrason Ferroelectr Freq Control:53 Chivukula VS, Shur MS, Čiplys D (2007) Recent advances in application of acoustic, acousto-optic and photoacoustic methods in biology and medicine. Phys Status Solidi A 204:3209–3236 Coté GL, Lec RM, Pishko MV (2003) Emerging biomedical sensing technologies and their applications. IEEE Sensors J 3:251–266

534

J. Zhang et al.

Deobagkar DD, Limaye V, Sinha S, Yadava RDS (2005a) Acoustic wave immunosensing of Escherichia coli in water. Sensors Actuators B Chem 104:85–89 Deobagkar DD, Limaye V, Sinha S, Yadava RDS (2005b) Acoustic wave immunosensing of Escherichia coli in water. Sensors Actuators B Chem 104:85–89 Du J, Harding GL, Ogilvy JA, Dencher PR, Lake M (1996) A study of Love-wave acoustic sensors. Sensors Actuators A Phys 56:211–219 Ferrari V, Lucklum R (2009) Overview of acoustic-wave microsensors, Piezoelectric transducers and applications. Springer, pp 39–62 Francis, LA (2000) SAW sensors, bachelor’s thesis, Université caholique de Louvain Fung YS, Wong YY (2001) Self-assembled monolayers as the coating in a quartz piezoelectric crystal immunosensor to detect Salmonella in aqueous solution. Anal Chem 73:5302–5309 Furtado LM, Su H, Thompson M, Mack DP, Hayward GL (1999) Interactions of HIV-1 TAR RNA with Tat-derived peptides discriminated by on-line acoustic wave detector. Anal Chem 71: 1167–1175 Gabl R, Feucht HD, Zeininger H, Eckstein G, Schreiter M, Primig R et al (2004) First results on label-free detection of DNA and protein molecules using a novel integrated sensor technology based on gravimetric detection principles. Biosens Bioelectron 19:615–620 Gabl R, Green E, Schreiter M, Feucht HD, Zeininger H, Primig R et al (2011) Novel integrated FBAR sensors: a universal technology platform for bio-and gasdetection. IEEE:1184–1188 Gizeli E, Goddard NJ, Lowe CR, Stevenson AC (1992a) A Love plate biosensor utilising a polymer layer. Sensors Actuators B Chem 6:131–137 Gizeli E, Stevenson AC, Goddard NJ, Lowe CR (1992b) A novel Love-plate acoustic sensor utilizing polymer overlayers. IEEE Trans Ultrason Ferroelectr Freq Control 39:657–659 Grate JW, Martin SJ, White RM (1993) Acoustic wave microsensors-Part I. Anal Chem (Washington), (United States) 65 Gronewold TMA (2007) Surface acoustic wave sensors in the bioanalytical field: recent trends and challenges. Anal Chim Acta 603:119–128 Hartmann CS, Abbott BP (1989) Overview of design challenges for single phase unidirectional SAW filters, Ultrasonics Symposium, 1989 Proceedings, IEEE, 1989, vol 1, pp 79–89 Hechner J, Soluch W (2005) Pseudo surface acoustic wave dual delay line on 41 YX LiNbO 3 for liquid sensors. Sensors Actuators B Chem 111:436–440 Hengerer A, Kösslinger C, Decker J, Hauck S, Queitsch I, Wolf H et al (1999) Determination of phage antibody affinities to antigen by a microbalance sensor system. BioTechniques 26: 956–965 Herrmann F, Weihnacht M, Buttgenbach S (2001) Properties of sensors based on shearhorizontal surface acoustic waves in LiTaO/sub 3//SiO/sub 2/and quartz/SiO/sub 2/structures. IEEE Trans Ultrason Ferroelectr Freq Control 48:268–273 Höök F, Ray A, Nordén B, Kasemo B (2001) Characterization of PNA and DNA immobilization and subsequent hybridization with DNA using acoustic-shearwave attenuation measurements. Langmuir 17:8305–8312 Janshoff A, Galla HJ, Steinem C (2000) Piezoelectric mass-sensing devices as biosensors – an alternative to optical biosensors? Angew Chem Int Ed 39:4004–4032 Kanazawa KK, Gordon JG (1985) Frequency of a quartz microbalance in contact with liquid. Anal Chem 57:1770–1771 Länge K, Rapp BE, Rapp M (2008) Surface acoustic wave biosensors: a review. Anal Bioanal Chem 391:1509–1519 Lec RM (2014) Acoustic wave sensors. Arch Acoust 21:179194 Lewis M (1977) Surface skimming bulk wave, SSBW. IEEE:744–752 Lin Z, Yip CM, Joseph IS, Ward MD (1993) Operation of an ultrasensitive 30-MHz quartz crystal microbalance in liquids. Anal Chem 65:1546–1551 Link M (2006) Study and realization of shear wave mode solidly mounted film bulk acoustic resonators (FBAR) made of c-axis inclined zinc oxide (ZnO) thin films: application as gravimetric sensors in liquid environments

22

Acoustic Transducer and Its Applications in Biosensors

535

Love AEH (2015) Some problems of geodynamics. Cambridge University Press Milsom RF, Reilly NHC, Redwood M (1977) Analysis of generation and detection of surface and bulk acoustic waves by interdigital transducers. IEEE Trans Sonics Ultrason 24:147–166 Petryayeva E, Algar WR, Medintz IL (2013) Quantum dots in bioanalysis: a review of applications across various platforms for fluorescence spectroscopy and imaging. Appl Spectrosc 67: 215–252 Powell DA, Kalantar-zadeh K, Ippolito S, Wlodarski W (2002) A layered SAW device based on ZnO/LiTaO/sub 3/for liquid media sensing applications. IEEE:493–496 Rapp M, Wessa T, Ache HJ (1995) Modification of commercially available low-loss SAW devices towards an immunosensor for in-situ measurements in water. IEEE:433–436 Richert L, Lavalle P, Vautier D, Senger B, Stoltz JF, Schaaf P et al (2002) Cell interactions with polyelectrolyte multilayer films. Biomacromolecules 3:1170–1178 Rocha-Gaso M-I, March-Iborra C, Montoya-Baides Á, Arnau-Vives A (2009) Surface generated acoustic wave biosensors for the detection of pathogens: a review. Sensors 9:5740–5769 Sant W, Pourciel ML, Launay J, Do Conto T, Martinez A, Temple-Boyer P (2003) Development of chemical field effect transistors for the detection of urea. Sensors Actuators B Chem 95:309–314 Sauerbrey GZ (1959) Use of quartz vibration for weighing thin films on a microbalance. Z Phys 155:206–212 Sauerbrey GZ (1991) The use of quartz crystal oscillators for weighing thin layers and for microweighing applications Smith JP, Hinson-Smith V (2006) Commercial SAW sensors move beyond military and security applications. Anal Chem 78:3505–3507 Stobiecka M, Cieśla JM, Janowska B, Tudek B, Radecka H (2007) Piezoelectric sensor for determination of genetically modified soybean roundup ready (R) in samples not amplified by PCR. Sensors 7:1462–1479 Tom-Moy M, Baer RL, Spira-Solomon D, Doherty TP (1995a) Atrazine measurements using surface transverse wave devices. Anal Chem 67:1510–1516 Tom-Moy M, Baer RL, Spira-Solomon D, Doherty TP (1995b) Atrazine measurements using surface transverse wave devices. Anal Chem 67:1510–1516 Vale C, Rosenbaum J, Horwitz S, Krishnaswamy S, Moore R (1990) FBAR filters at GHz frequencies. IEEE:332–336 Voinova MV (2009) On mass loading and dissipation measured with acoustic wave sensors: a review. J Sensors Webb DJ (2002) Optical-fiber sensors: an overview. MRS Bull 27:365–369 Weber J, Albers WM, Tuppurainen J, Link M, Gabl R, Wersing W et al (2006) Shear mode FBARs as highly sensitive liquid biosensors. Sensors Actuators A Phys 128:84–88 Wenzel SW, White RM (1988) A multisensor employing an ultrasonic Lamb-wave oscillator. IEEE Trans Electron Devices 35:735–743 Wessa T, Barié N, Rapp M, Ache HJ (1998) Polyimide, a new shielding layer for sensor applications. Sensors Actuators B Chem 53:63–68 White RM, Voltmer FW (1965a) Direct piezoelectric coupling to surface elastic waves. Appl Phys Lett 7:314–316 White RM, Voltmer FW (1965b) Direct piezoelectric coupling to surface elastic waves. Appl Phys Lett 7:314–316 White RM, Wicher PJ, Wenzel SW, Zellers ET (1987) Plate-mode ultrasonic oscillator sensors. IEEE Trans Ultrason Ferroelectr Freq Control 34:162–171 Wingqvist G, Bjurström J, Liljeholm L, Yantchev V, Katardjiev I (2007) Shear mode AlN thin film electro-acoustic resonant sensor operation in viscous media. Sensors Actuators B Chem 123: 466–473 Zhou X, Liu L, Hu M, Wang L, Hu J (2002) Detection of hepatitis B virus by piezoelectric biosensor. J Pharm Biomed Anal 27:341–345

Acoustic Biosensors for Cell Research

23

Samar Damiati

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Overview of Acoustic Wave Biosensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Basic Parts of Acoustic Wave Biosensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Fundamentals and Categories of Acoustic Wave Biosensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Assembly of Puzzle Pieces to Design Biosensor Architectures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Applications of Acoustic Biosensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acoustic Biosensors for Cell Mimicry: Lipid Membrane-Based Biosensor . . . . . . . . . . . . . . . Acoustic Biosensors for Cell Behavior: Whole Cell-Based Biosensors . . . . . . . . . . . . . . . . . . . . Acoustic Biosensors for Cell Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cells as Sensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

538 540 540 541 547 550 550 553 557 560 562 563

Abstract

Drawing inspiration from nature and applying natural principles can support the continuous improvement of sensing technologies in various fields, such as medicine, pharmacy, and environmental applications. It is difficult to directly connect a sensing system to a complex biological system. Thus, finding a suitable technique that simplifies and interprets complicated biological information to generate readable signals is in high demand. Acoustic technology appears to be S. Damiati (*) Department of Biochemistry, Faculty of Science, King Abdulaziz University (KAU), Jeddah, Saudi Arabia Institute for Synthetic Bioarchitectures, Department of Nanobiotechnology, University of Natural Resources and Life Sciences (BOKU), Vienna, Austria Division of Nanobiotechnology, Department of Protein Science, Science for Life Laboratory, School of Engineering Sciences in Chemistry, Biotechnology and Health, KTH Royal Institute of Technology, Stockholm, Sweden e-mail: [email protected]; [email protected] © Springer Nature Switzerland AG 2022 G. Thouand (ed.), Handbook of Cell Biosensors, https://doi.org/10.1007/978-3-030-23217-7_150

537

538

S. Damiati

a promising sensing model. The monitoring of the biochemical processes or the quantification of a captured analyte can be performed utilizing acoustic wave devices that rely on gravimetric sensing of materials adsorbed onto the sensor surface. Considering nature as a toolkit that provides individual puzzle pieces that can be assembled carefully into a sensory system offers a rich source to build selective and sensitive biosensors. The natural toolbox includes biological components such as DNA, RNA, sugar, amino acids, proteins, and lipids, in addition to nonbiological components such as graphene, carbon nanotubes, and metals. These molecules can be assembled together onto piezoelectric substrates to enhance the functionality of fabricated acoustic devices. This chapter has classified acoustic biosensors into four classes for various cell applications. First, lipid membrane-based biosensors are biomimetic models constructed by natural biological materials to simplify the complexity of biological cell membranes and enable investigations of membrane proteins in a native-like environment. These bioarchitectures also offer a good opportunity to investigate the interactions of lipids and proteins under controlled conditions. Second, whole cell-based biosensors are fabricated to enable investigations of cellular behaviors such as cell adhesion and cell-substrate interactions. Third, detection biosensors are also attracting attention due to their high sensitivity, ability to track cells in real time without labeling, and ability to differentiate between viable and nonviable cells. Finally, recent advancements in the fabrication of acoustic biosensors have enabled cells themselves to act as biosensors to detect analytes. All designed acoustic platforms are aimed at studying the cell, the basic unit of life, from different perspectives. The facts discussed in this chapter are based on phenomena that cannot be visualized by the eye, such as cellular interactions, or factors present in such small quantities, but they can be heard by tracking their acoustic sounds.

Introduction Drawing inspiration from nature is an ancient strategy for innovating new designs or solving existing design problems. Biomimicry is a multidisciplinary approach that involves biology, chemistry, medicine, engineering, electronics, architecture, etc. Biomimicking relies on copying, emulating, understanding, and learning from nature to design biosensors in various fields, such as medicine, pharmacology, industry, agriculture, food, and environmental applications (Damiati 2018; Schuster 2018). For cell research, combining biology and biosensor technology leads to the development of four biosensing platforms: (i) biomimetic sensors that mimic the structural and functional features of biological cell membranes and can be used further as a platform for drug screening and protein investigations, (ii) whole cellbased biosensors to track various cell interactions, (iii) detection biosensors that enable qualification and/or quantification of specific analytes and subsequent exploitation to develop diagnostic tools, and (iv) cells as sensors for analyte detection and cellular interaction investigations. Biosensors can be further classified based on the

23

Acoustic Biosensors for Cell Research

539

sensing film architecture and sensing methods. Biosensors are analytical devices that integrate physiochemical transducers such as acoustic, optical, electrochemical, magnetic, mechanical, and radiation transducers that respond to chemical or biological compounds or respond to biochemical interactions and are converted to quantified signals (Grieshaber et al. 2008; Hasan et al. 2014; Bhalla et al. 2016). Despite the availability of many sensing techniques, the development of small, reliable, inexpensive biosensors is still a challenging task in clinical and healthcare applications due to the complexity and variability of biomolecules. Among the available biosensing technologies, acoustic wave biosensors offer powerful tools with high selectivity and sensitivity to detect pathogens, metabolites, toxins, and whole cells. In addition to rapid detection, the real-time tracking ability of acoustic sensors and their suitability for label-free measurements make this technology a competitive technology for clinical applications and for investigations of biomolecule interactions. Acoustic wave biosensors utilize sound waves (type of mechanical waves) with particular frequencies generated by sequential physical compression/expansion (oscillation or vibration) cycles of gases, liquids, or solid media (Stroble et al. 2009). As a monitoring mechanism to gain biophysical or biochemical information about the molecule of interest, acoustic wave devices monitor changes in the physical characteristics of an acoustic wave, such as mass, elasticity, temperature, pressure, dielectric, and conductivity properties. Most acoustic biosensors depend mainly on gravimetric (mass) properties rather than optical properties, which make the acoustic sensing method appropriate for direct measurements on unpurified samples and enables biochemical investigations over various surfaces. Furthermore, since acoustic wave sensors are highly sensitive devices, they respond to small environmental perturbations. The principles of some acoustic wave biosensors were inspired by nature. For example, a cochlear amplifier design mimics the morphology of the human inner ear cochlea. In this system, the outer hair cells act as a piezoelectric device that responds to stimulation as a sound-sensing element. The function of the cochlear amplifier is analogous to that of the surface acoustic wave (SAW) resonator (Spector et al. 2003; Bell 2006). Another example is a sonar receiver developed by direct mimicry of dolphin echolocation based on the morphological and functional basics of the dolphin lower jaw (Dobbins 2007). Moreover, most human body sounds, such as the heart, lung, gurgling intestine, and snoring sounds, generate acoustic biosignals. The acoustical path of body sounds starts at the sound source, vibrates the blood volume, and oscillates the biological structures. The generated acoustic biosignals can be monitored to disclose the state of health and the body’s vital functions using a stethoscope (Fig. 1) (Kaniusas 2015; Dey et al. 2018). The present chapter focuses on the assembly of biological puzzle pieces with the help of nonbiological pieces to build acoustic biosensors for various cell applications. First, in this chapter, an overview of the basic components and principles of acoustic technique is discussed. After reviewing the different modes of acoustic devices, the importance of surface architecture for fabricating an acoustic sensor is highlighted. Examples of biological and nonbiological molecules used to create effective biosensing systems are also included. The last section outlines the four

540

S. Damiati

Fig. 1 The acoustic wave propagates in the stethoscope and reflects the heart sounds [Created with BioRender.com]

categories of biosensors for cell applications: lipid membrane-based biosensors, cellbased biosensors, detection biosensors, and cells as sensors. Successful strategies based on these platforms, which are tracked by acoustic technology and cell research, are presented.

Overview of Acoustic Wave Biosensors Basic Parts of Acoustic Wave Biosensors A typical acoustic wave biosensor generally consists of the following parts (Fig. 2): (i) bioreceptors that specifically recognize and capture the analyte (e.g., cells), which is particularly important for detection biosensors; (ii) an interface architecture where reactions occur that requires careful design to improve signal detection; (iii) a transducer element, which is always piezoelectric crystal to convert the transducer signals generated due to mass changes to an electrical signal and whose surface attaches to a metallic electrode to bring the electrical field to the active sensing film; (iv) an oscillator circuit to generate electric currents; (v) a frequency counter that converts data into a frequency value; and (vi) software to analyze and process the output signals to a meaningful format (Grieshaber et al. 2008; Saad and Zaaba 2014). Indeed, the basic components of acoustic wave biosensors can be pieced together

23

Acoustic Biosensors for Cell Research

541

Fig. 2 Components of an acoustic biosensor to capture whole cells [Created with BioRender.com]

with additional parts, such as a temperature controller, flow injection system, or impedance analyzer.

Fundamentals and Categories of Acoustic Wave Biosensors All acoustic wave platforms are functional in gaseous or vacuum environments, but not all are functional in liquids, and they depend on the direction of particle displacement at the surface of the device. Acoustic wave sensors, as a gravimetric technique, are based on the propagation of acoustic waves through or on the substrate surface. Any disruption in the propagation path due to mass attachment on the sensor surface leads to changes in the output signals: wave velocity and/or amplitude (Rocha-Gaso et al. 2009; Fogel et al. 2016). The generated changes are monitored by recording the resonance frequency of the sensor and further correspond to the physical quantity on the surface (Sauerbrey 1959). The acoustic waves in the piezoelectric substrate are based on mechanical wave propagation through the substrate. These waves are generated by applying electric fields and transformed

542

S. Damiati

back to electrical fields for measurement. Many crystals exhibit piezoelectric effects, such as quartz (SiO2), lithium tantalate (LiTaO3), and lithium niobate (LiNbO3). Quartz is the choice for most acoustic sensors due to its abundance in nature, insolubility in water, resistance to high temperature, good chemical stability, excellent mechanical properties, and low-cost manufacturing (Durmus et al. 2008; Dey et al. 2018; Fogel et al. 2016). Physical properties, such as cut, size, density, thickness, and shear modulus, affect the resonant frequency of the quartz crystal vibration (Morgan 1991; Lec and Lewin 1999; Şeker and Elc¸in 2017). Natural or synthetic quartz crystals can cut precisely slab by different quartz cut types that have different properties and vibration modes (Czanderna and Lu 1984). At-cut and St-cut crystals are commonly used as acoustic sensors (Fogel et al. 2016). Each cut type leads to different piezoelectric deformations, allowing efficient propagation of acoustic waves in two different directions: along the surface of the sensor or away from the sensor surface toward the bulk substrate. Acoustic wave sensors are based on wave velocity and particle movement directions within the substrate. Many variants are possible depending on the materials and boundary conditions of the substrate. Hence, depending on the wave propagation mode, the piezoelectric crystal resonator in the sensor system is classified mainly as a bulk acoustic wave (BAW) or a SAW sensor. Acoustic waves propagate unguided through the volume of the substrate in BAW devices and propagate guided or unguided along one side of the substrate in SAW devices (Rocha-Gaso et al. 2009; Durmuş et al. 2008; Fogel et al. 2016).

Bulk Acoustic Wave (BAW) Sensors BAW sensors employ one of the two types of waves: longitudinal (also called a pressure wave) or transverse (also called a shear wave) (Fig. 3). However, longitudinal waves have higher velocities than transverse waves, while transverse waves reduce acoustic radiation in the media. The most common BAW sensors are quartz crystal microbalance (QCM) and shear horizontal acoustic plate mode (SH-APM) systems (Kaspar et al. 2000; Durmuş et al. 2008). Both systems generate waves in a shear horizontal motion that allows the mechanical oscillation of waves across the bulk of the quartz substrate. The shear horizontal waves do not radiate noticeable energy into liquids, which enables functionality without excessive attenuation (Grainger and Castner 2011).

Fig. 3 Longitudinal and transverse waves. In a longitudinal wave, the oscillation occurs in the direction of wave propagation or the longitudinal direction. Particle motion in the medium is parallel to the wave direction. In a transverse wave, the particles of the medium are transverse to the direction of propagation or oscillate at a right angle. Particle motion travels perpendicular to the wave direction

23

Acoustic Biosensors for Cell Research

543

Quartz Crystal Microbalance (QCM) Bulk Acoustic Wave (BAW) Sensors

QCM is the oldest, simplest, most well-established, and most widely used acoustic wave device today, and its name comes from its piezoelectric resonator properties prepared from quartz. Furthermore, QCM can be exploited as a resonator in a pure thickness shear mode (TSM); hence, it is also known as the TSM sensor. The transducer for the QCM sensor is usually in the form of a piezoelectric thin circular disk of an AT-cut quartz plate sandwiched between two metallic circular electrodes, e.g., silver or gold (Fig. 4) (Grate and Frye 1996). Due to the piezoelectric nature of the quartz disk, applying an alternating current (AC) voltage between the electrodes results in a shear deformation of the quartz crystal. The oscillation of voltage frequency generates a mechanical resonance, and the optimal amplitude of displacement occurs at the quartz surface and reaches certain acoustic (resonant) frequencies (Rodahl and Kasemo 1996). The following formula can be used to calculate the fundamental resonant frequency ( f0): f 0¼

vq 2 tq

where vq is the shear wave speed in the crystal and tq is the crystal thickness. The quartz substrate can be found in different dimensions and frequencies. The resonant frequencies for commercial QCM devices are in the range of 5–30 MHz. The resonant parameters of oscillation of the crystal are sensitive to attached mass at the quartz crystal surface; resonant frequency decreases, while the overall weight of the quartz increases due to mass added to the surface in the form of a film (King

Fig. 4 Schematic view of the QCM sensor. The quartz resonator is sandwiched between two electrodes, and the sensing layer is positioned on top to receive materials and subsequently responds to mass changes on the QCM sensor. Applying an alternating current voltage between the two electrodes through the AT-cut quartz crystal generates a resonant shear acoustic wave in the sensor crystal. The acoustic waves travel toward the bulk fluid in which the sensor is immersed, while the quartz atoms oscillate side-to-side. In response to the loaded mass and viscoelastic changes, the shear oscillation accompanies to the attached materials on the sensing layer and changes the resonant frequency and energy dissipation, respectively

544

S. Damiati

1964; Czanderna and Lu 1984; Konradi et al. 2012). Thus, molecules loaded onto the surface of the piezoelectric resonator can be quantified by measuring the changes in the resonant frequency using the Sauerbrey equation: Δm ¼ C  Δf where Δm is the mass change (g/cm2), C is the mass sensitivity constant of the resonator (17.7 ng/Hz  cm2 for a 5 MHz AT-cut quartz crystal), and Δf is the frequency shift (Hz) (Konradi et al. 2012; Huang et al. 2017). QCM is suitable for real-time label-free measurements due to its ability to determine the resonant frequency by the entire oscillating mass, which involves all mass adsorbed onto the surface. The Sauerbrey equation is valid for oscillation in dry conditions and applicable to thin and rigid films attached to sensor surfaces in a vacuum. Small changes caused by mass adsorbed from gas or air onto a quartz sensor can be monitored in situ for mass deposition, which appears as changes in resonance frequencies in a linear manner. In contrast, the linear Sauerbrey relation may not be validated in wet conditions. In a liquid environment, acoustic sensors lose the mechanical energy stored in the oscillation at a higher rate than they do in a gas or vacuum because molecules in liquids travel with the oscillating surface (Konradi et al. 2012). The propagation of shear waves radiated from the resonator is affected by the density and viscosity of the liquid, which subsequently affects the frequency of the QCM sensor that is in direct contact with the liquid at one surface. A simple relationship was derived by Kanazawa and Gordon (1985) to express the variation in the oscillation frequency of a quartz crystal in contact with a liquid: Δf ¼

3=2 f0 :



ηL ρL π μ Q ρQ

1=2

where f0 is the oscillation frequency of free, dry quartz crystal, ηL and ρL are the absolute viscosity and density of the liquid in contact with the crystal, respectively, and μQ and ρQ are the elastic modulus and density of the quartz crystal, respectively. Despite the importance of water as a biomaterial molecule, its value in biomolecule adsorption investigations is often neglected. The hydration of biomolecules and polymeric surfaces either in solution or at interfaces plays significant roles in their functionality. Furthermore, coupled water affects the adsorption and recognition of the interacting biomolecules (Konradi et al. 2012). Compared to other techniques, such as surface plasmon resonance (SPR), QCM has the ability to quantify the amount of hydrodynamically coupled water in an adsorbed film. Hence, the resonance frequency of the quartz sensor depends on the total mass including water coupled to the oscillations. These water molecules involve water bound to or trapped in the adsorbed film (Konradi et al. 2012; Walters et al. 2012). The amount of coupled water depends on the type of molecule and surface. For example, a small amount of water is coupled to elongated molecules that adsorb flat on a surface. In contrast, a significant amount of water is coupled to elongated molecules that stand up on a surface. However, the QCM technique is considered a very sensitive

23

Acoustic Biosensors for Cell Research

545

balance due to its ability to detect the oscillating mass of the sensor and adhering layers on the surface, including water. The frequency signal decreases when a thin layer attaches to the QCM sensor. For a thin and rigid adsorbed film, the decrease in the frequency signal is proportional to the mass attached to the surface of the sensor using the Sauerbrey relation. In the case of adsorbing biomolecules such as proteins, a nonrigid layer is often generated, which affects the oscillation of the crystal and results in energy dissipation. Furthermore, the Sauerbrey relation underestimates the adsorbed soft film on the sensor. Taking into account the viscoelastic property of biological molecules has supported the development of the QCM with dissipation (QCM-D) monitoring technique. QCM-D enables the simultaneous recording of shifts in resonance frequency and energy dissipation to provide quantitative information on attachment kinetics and qualitative information on the mechanical properties of the internal structure of the adhered film on the sensor (Rodahl et al. 1995; Rodahl and Kasemo 1996). Dissipation of the quartz crystal can be recorded in real time, either by recording the decay constant of the shear amplitude of the freely oscillating sensor or by recording the width and position of the resonance peak in the frequency domain. QCM-D allows rapid sampling within a few milliseconds due to the dampened oscillations. In contrast to traditional biosensing technologies, the relationships between mass and frequency and between viscoelasticity and dissipation make QCM-D a pioneering technique that improves the development of quantitative and qualitative biosensing platforms (Konradi et al. 2012; Walters et al. 2012). Shear Horizontal Acoustic Plate Mode (SH-APM) SH-APM sensors use thin quartz plates and generate shear horizontal waves that propagate between the upper and lower sides of the plate to enable sensing on either side (Fig. 5). The piezoelectric substrate guides the acoustic wave and reserves the energy within the upper and lower surfaces. The typical frequencies operated by SHAPM sensors are 100–150 MHz, and a decrease in plate thickness leads to a decrease in frequencies. SH-APM devices share many principles with QCM sensors; the main

Fig. 5 Schematic sketches of the shear horizontal acoustic plate mode (SH-APM). On the top surface (left), the electrode-free face hosts molecules and their interactions. On the bottom surface (right), the two IDTs are located without access to or contact with the samples loaded on the top surface

546

S. Damiati

difference is that SH-APM sensors use interdigital transducers (IDT) instead of electrode plates. Two IDTs are positioned on one side of the plate, without access to the conducting gases or fluids. By applying an oscillating voltage, one IDT generates SH waves possessing particle displacement principally parallel to the plate surface perpendicular to the propagation direction along the separation path between the two IDTs, whereas the other side of the plate receives the generated displacement waves and acts as a sensing surface. This free surface without IDT can be biofunctionalized and immersed in the target sample, and analysis can be conducted with a complete separation between the electric and liquid sides, which overcomes corrosion problems that are observed when electrode plates are immersed in biological solutions (Drafts 2001; Durmus et al. 2008; Ferrari and Lucklum 2004).

Surface Acoustic Wave (SAW) Sensors Most SAW sensors employ Rayleigh waves, which are 2D waves, and a combination of longitudinal and transverse waves (Ferrari and Lucklum 2004). SAW devices allow propagation of acoustic waves along or near the free surface of the piezoelectric substrate’s surface (Fig. 6). Due to energy confinement near the surface, SAW sensors are highly sensitive to surface adsorption and can be used to measure concentrations, viscosity, acceleration, temperature, and pressure (Dey et al. 2018; Durmus et al. 2008). Furthermore, due to the ability of SAW devices to apply high frequencies above several gigahertz (GHz), the mass sensitivity is improved. The accumulated mass on the piezoelectric crystal results in a shift in the frequency signal of the generated acoustic wave and can be exploited for biosensing. SAW devices can effectively perform fluid manipulation and biosensing assessments. Furthermore, SAWs generated by microelectrodes are routinely used to focus, align, separate, and manipulate fluids in a microchannel (Wang and Zhe 2011). The basic components of SAW sensors are a piezoelectric plate and an IDT composed of two identical comb-like structures with fingers that are patterned on a thin piezoelectric film and arranged in an alternating pattern (Ferrari and Lucklum 2004). The generation of effective surface waves occurs when the wavelength of SAW equals the spacing between the transducer fingers. SAW devices are operated in Fig. 6 Schematic sketches of the SAW sensor. The IDT and the sensing area are located on the same side. SAW acoustic wave propagation parallel to the sensor surface. Applying electrical power across the STcut quartz between the two sets of IDT drives the quartz atoms to oscillate up and down

23

Acoustic Biosensors for Cell Research

547

either (i) a resonator configuration that is similar to QCM sensors with the traditional outputs of frequency and dissipation shifts, with additional reflectors to confine the acoustic energy within the resonant cavity to reduce energy loss, or (ii) a delay line configuration that considers the time needed for the acoustic wave to travel between two sets of transducers as the measured output. Furthermore, the velocity of the SAW is regulated by the surface adsorbates (Fogel et al. 2016). Other surface waves have shown important applications as acoustic sensors, such as surface transverse waves (STW), Love waves (LW), and Lamb waves. Although SAW devices in general have exhibited high sensitivity in gravimetric applications and Rayleigh mode SAW sensors in particular achieved good detection limits in response to gas or vapor exposure in air, Rayleigh waves are not suitable for liquid applications because of radiation losses and rapid energy dissipation into the liquid on the surface, resulting in excessive damping. As alternatives, STW and SH-SAW modes are highly sensitive and convenient for real-time sensing in liquids by allowing acoustic wave propagation along the surface without significant coupling of acoustic energy into the liquid (Durmus et al. 2008; Fu et al. 2017; Ferrari and Lucklum 2004).

Assembly of Puzzle Pieces to Design Biosensor Architectures The efficiency of biosensor design largely relies on the successful assembly of several pieces to build an efficient interface matrix. Although extensive studies have developed biosensors, only a few sensors are available on the market for limited applications and do not fully suit users’ needs. One example is glucose sensors as a detection tool to quantify blood sugar levels. The glucose sensor is a well-known, inexpensive, and handheld device that can be easily used by diabetic patients at home and provides a sensitive point-of-care test. The main obstacle to developing highly selective and sensitive detection biosensors is the lack of a sensing architecture that enables the capture of biomolecules, especially those with small molecular weights at low concentrations and in crowded biological media. Furthermore, sensors with biomimetic designs to investigate biomolecule interactions are still limited due to the complexity of biological systems. Accordingly, borrowing ideas from nature may enhance the development of efficient sensors at different levels and for several purposes. To address sensor fabrication limitations, natural solutions can be found by looking for missing pieces of the puzzle in nature to improve the functionality of the sensing system. There is a trending approach to using natural materials that are biocompatible, such as polymers, hydrogels, graphene, carbon nanotubes, and quantum dots, in the biosensor industry, not only because they can improve the efficiency of the sensor but also because they are environmentally friendly. Biomaterials are flexible, biocompatible, biodegradable, and disposable. Furthermore, they are convenient, with several detection techniques, such as optical, fluorescence, electrochemical, and acoustic measurements. In biosensing applications, biomaterials need to be stable, inexpensive, selective, and sensitive and to overcome the limitations related to nonspecific binding, possess

548

S. Damiati

strong affinity, have no interactions with biological molecules, and minimize any physical-chemical interactions to avoid interfering with the measured signal. For biosensor bases, silicon, glass, and petroleum-derived plastics are usually used, while for electrodes, several metals can be exploited (Perez et al. 2019). For the surface architecture, several biomaterials are used either as tether molecules or as recognition/capture molecules. The surface patterns improve sensor functionality when biomaterials are present at a good density and in a well-organized arrangement. Furthermore, biosensor efficiency performance is enhanced, and stability improves when spacer molecules that act as an intermediate are immobilized between the active sensing film, which contains the recognition/capture molecules or lipid membrane, and the solid substrate (Fig. 1). Carbon and carbonaceous materials such as graphene and carbon nanotubes are in high demand for biosensing applications. Their physicochemical properties offer easy functionalization and good electrical conductivity (Damiati et al. 2018a, 2019; Haslam et al. 2018; Islam et al. 2018). Noble metals, such as gold, silver, and platinum, and noble metal oxides, such as cerium, copper, nickel, iron, zinc, and titanium, are widely used in biosensors with nanostructure geometries in several forms, including nanoparticles, nanorods, and nonporous thin films. The properties of pure noble metals, such as stability, conductivity, and low toxicity, make them favorable for use in sensor development. Furthermore, their optical, electric, and magnetic characteristics make them compatible with many investigation techniques (Maduraiveeran et al. 2018; Perez et al. 2019). Indeed, the gold surface can also be modified with self-assembled monolayers (SAMs) of sulfides (thiols) and disulfides. Thiol molecules can be immobilized in well-organized structures and in high order onto gold surfaces and subsequently exploited for coupling with biological elements (Wink et al. 1997; Grieshaber et al. 2008). Natural or synthetic polymers (e.g., chitosan, cellulose, and collagen) are often used as a supporting structure for the sensor that can be easily combined with other materials and in different geometries. Polymers can be obtained from different natural sources, such as animal bones and skin, plants, exoskeletons of marine crustaceans and insects, and fungal cell walls. Multiple functional groups that can be chemically modified present polymers as an effective matrix for biosensing. The adsorption of polymers on sensor surfaces leads to good film-forming ability with high stability and easy handling (Grieshaber et al. 2008; Reimhult and Höök 2015; Damiati et al. 2018a; Perez et al. 2019). A main component of detection biosensors is the recognition/capture biomolecule, such as antibodies, antibody fragments, aptamers, and affimers. These biomolecules are used due to their high affinity and specificity for particular targets. A binding recognition molecule with a target antigen leads to the formation of a complex that gives a signal that can be quantitatively measured. Antibodies are one of the most widely used biorecognition molecules. In a sandwich assay, the primary antibody is usually a monoclonal antibody that is exploited to capture the antigen or cell of interest, while the secondary antibody is usually a labeled polyclonal antibody used to enhance the detection signal. The immobilization of antibodies is an issue because their orientation and density on the sensing surface affect their biological activity. Thus, it is important to minimize random orientation and

23

Acoustic Biosensors for Cell Research

549

find the optimal concentration to prevent denaturation and minimize the nonspecific interactions that occur when recognition molecules immobilized at very high density onto the sensor surface (Lu et al. 1996; Grieshaber et al. 2008). Antibody fragments are similar to whole antibodies, with high specificity but a smaller size, which is considered an advantage in biosensor architecture design (Holliger and Hudson 2005). Aptamers are also widely used because they possess advantages that overcome the limitations of antibody immobilization. Aptamers are highly reproducible, easily deposit on the sensor surface, are highly resistant to denaturation, and have a long half-life (Tombelli et al. 2005). Affimers are relatively new molecules that have the advantages of antibodies and aptamers in addition to several more advantages. Affimers are either oligonucleotides or peptides, ~10 times smaller than antibodies, and are easily genetically or chemically modified, with rapid development (Ko Ferrigno 2016). However, alternative recognition molecules, such as peptides, vitamins, and synthetic antibodies, have also been exploited in biosensor development. To create lipid membrane-based biosensors, the choice of lipid composition is a critical step as it may affect protein reconstitution and membrane activity. Since natural cell membranes are negatively charged, negatively charged lipids such as phosphatidylserine (PS), phosphatidylglycerol (PG), and phosphatidylinositol (PI) are usually used in the generation of artificial lipid membranes. The importance of lipid charge has been noticed naturally in the cell membrane. The inner monolayer of the mammalian plasma membrane is composed of phosphatidylethanolamine (PE) and PS. Due to the presence of negatively charged lipids, there is a difference between the two halves of the bilayer membrane. This negativity is important for the functionality of several cytosolic proteins that bind specifically to negative lipid head groups in response to extracellular signals (Alberts et al. 2002). It is common to form artificial lipid membranes on negatively charged substrates (e.g., mica and silica), which causes repulsive electrostatic interactions. Thus, high concentrations of negatively charged lipids disrupt lipid bilayer generation (Hardy et al. 2013; Cremer and Boxer 1999). Two strategies can be followed to minimize electrostatic interactions with negatively charged substrates. Positively charged components such as positively charged lipids, peptides/proteins, or fluorescent probes are incorporated into the membrane, or the pH is changed, which affects the degree of ionization, and hence changes in lipid pKa values change the lipid charge (Hardy et al. 2013; Cho et al. 2010). For example, PE is a zwitterionic lipid with pKa ¼ ~1.0 that is neutral at any pH > ~3 (Marsh 1990). Lateral membrane components, such as cholesterol and sphingolipids, also play important roles in membrane modeling. Cholesterol, for example, is important for lipid organization, regulates membrane fluidity and permeability, and improves mechanical resistance. Although cholesterol is a fundamental component of plasma membranes, biomimetic modes generated from vesicle fusion are usually composed of simple lipid compositions with no/little cholesterol because they may hinder vesicle fusion due to high rigidity (Dopico and Tigyi 2007; Hardy et al. 2012, 2013). Since protein is a main component in the cell, finding productive strategies to express membrane proteins needs to be optimized at the gene and protein levels. There are two methods to express proteins: cell-free (in vitro) and

550

S. Damiati

cell-based (in vivo) protein expression systems. Cell-free protein expression systems provide many advantages to address the limitations associated with cell-based systems, such as the lack of need for cloning, the possibility of synthesizing toxic proteins that living cells cannot tolerate, and the generation of functional and correctly folded proteins, rapid production, and high protein yield. Cell-free systems involve simple mixing of genetic templates (mRNA or DNA) that are sequenced carefully with a cell extract that contains cellular factors and enzymes important for the transcription-translation machinery (Jackson et al. 2004; Endo and Sawasaki 2006; Damiati et al. 2018b). Hence, all of the abovementioned elements can be considered puzzle pieces that need to be assembled in a defined manner onto acoustic sensor surfaces. These molecules enhance biosensor functionality for cell research.

Applications of Acoustic Biosensors Biosensors for cell applications can be classified into four platforms: lipid membrane-based biosensors, whole cell-based biosensors, detection biosensors, and cells as sensors. Fundamental information and several successful fabricated acoustic models for each platform are highlighted in this section. All designed models are based on the careful assembly of biological and nonbiological materials and may contain living elements, i.e., cells, to functionalize the acoustic sensor surface.

Acoustic Biosensors for Cell Mimicry: Lipid Membrane-Based Biosensor The basic structure of biological membranes is provided by lipid molecules, while the functional characteristics are predominantly conferred by peptides and proteins. In addition to its role as a gatekeeper that regulates the movement of materials in and out of the cell, many biological reactions take place on the cellular membrane surface, such as interactions with the lipid layer, membrane affinity, and adsorption or penetration of peptides, proteins, or drugs molecules (Pomorski et al. 2014; Ágnes et al. 2017). Synthetic biomimetic models (lipid membrane-based biosensors) are developed to mimic the natural architecture of biological membranes to simplify the complexity of the highly organized structure of membrane molecules and their complex interactions (Eeman and Deleu 2010; Damiati 2018, 2019). These types of biomimetic models are usually based on lipid membranes and are used mainly to investigate the functions of membrane-bound or transmembrane proteins as well as membrane-mediated processes such as signal transduction and cell-cell interactions. The importance of synthetic membranes arises from the fact that more than 60% of drugs on the market target membrane proteins. It is still challenging to study these proteins due to their low abundance and their tendency to denature when isolated from the lipid membrane. Hence, there is a requirement to develop a seminatural environment that mimics the natural environment to study the interactions between

23

Acoustic Biosensors for Cell Research

551

drugs and membrane proteins or cell receptors. The construction of membrane-based biosensors is based on the de novo assembly of nonliving elements from biological origins such as DNA, lipids, and proteins following a bottom-up strategy and takes advantage of molecular self-assembly processes (Castellana and Cremer 2006; Schrems et al. 2011; Damiati et al. 2015a; Guidelli and Becucci (2012); Misawa et al. (2018)). Supported lipid bilayer (SLB) membranes, tethered bilayer lipid membranes (tBLMs), and cushioned bilayer lipid membranes are common cell membrane mimicking models (Fig. 7). These models can be developed and characterized as acoustic biosensors that offer robust platforms that enable in vitro investigation of biomolecular interactions, including lipid-lipid, lipid-protein, and protein-protein interactions. Several successful attempts have been presented concerning the use of membrane-based sensors for biosensing applications or for pharmacological research related to drug screening. The SLB membrane was exploited in an immunological study to monitor complement convertase assembly into a synthetic lipid platform. Complement convertases are biomacromolecular complexes that play a significant role in the innate immune system by amplifying the innate response against different pathogens, e.g., apoptotic cells and microbes. Complement proteins are inactive in the blood circulation but activated when interacting with a pathogen and adsorbed onto the target surface. QCM-D measurements were used to track the self-assembly of C3 convertase onto maleimide-functionalized SLB starting from the attachment of C3b protein, followed by the addition of a mixture of complement factors B and D. Subsequently, C3 protein was added and immobilized covalently to the SLB, leading to functional convertase assembly. Shifts in frequency and energy dissipation recorded the real-time kinetics of the sequential assembly of the molecular components (C3b, B and D factors, and C3). The developed model enabled the assessment of membrane-bound biomacromolecular complexes at the lipid membrane interface and hence could be exploited further to measure protein-membrane and proteinprotein interactions in seminatural environments (Avsar et al. 2017). The integration or anchoring of arrays of molecules has also been studied by applying in-plane SAWs on SLBs. Lateral standing waves are generated by SAWs on a piezoelectric

Fig. 7 Schematic of the biological cell membrane and models of biomimetic architectures: supported lipid bilayer (SLBs) and tethered or cushioned bilayer lipid membranes

552

S. Damiati

substrate, which lead to a lateral modulation of lipid concentration and the accumulation of DNA and proteins in periodic patterns (Neumann et al. 2010; Hennig et al. 2011). The generated waves create local traps in the lipid bilayer that are responsible for the redistribution of DNA or proteins, and the created patterns are reversible without any effect on the lipid bilayer integrity. Hence, the external forces that may affect the assembly of biomolecules or membranes in a natural environment can be mimicked by electroacoustic fields and further enable investigation of binding or membrane integrity (Hennig et al. 2009; Reusch et al. 2014). tBLMs and cushioned bilayer lipid membranes offer more advantages than SLBs because they have higher stability, which is important for long-term experiments, and they decouple lipid membranes from solid substrates, which minimize the direct contact between lipid bilayers and integrated proteins with sensor substrates (Tanaka 2006; Damiati 2018). Thus, the water reservoir increases while maintaining the structural and dynamic integrity of the biomimetic membrane. A tBLM platform was developed by Inci et al. (2015) to allow the investigation of drug-membrane protein interactions. The constructed model was composed of a modified lipid (DSPEPEG) that was used as a tether molecule to provide a reservoir layer, a lipid bilayer resulting from liposome fusion, and multidrug resistance protein 1 (MDR1), which is a membrane protein with a large extramembrane domain. Incorporation of MDR1 into tBLM was tracked in real time using SPR and QCM-D, and the obtained results showed the successful incorporation of MDR1 without requirements for any modification or enzymatic cleavage steps. Further, direct analysis of the drug-membrane protein interaction between MDR1 and pravastatin (a drug used to lower cholesterol levels) was subsequently investigated with the developed tBLM model. Another membrane model that enabled the study of biomolecular interactions was developed by Andrä et al. (2008) to mimic the outer bacterial and cytoplasmic membranes. The model was composed of phospholipids, lipopolysaccharides, and the antimicrobial peptide human cathelicidin-derived peptide LL32. The fabricated LW device used to analyze the surface binding of LL32 to the immobilized phosphatidylserine membrane showed that the peptide-membrane interaction increases bilayer rigidity. Cushioned bilayer lipid membranes can be used to probe the intracellular and intercellular movement of different molecules, such as ions and drugs. A para-crystalline protein surface layer (S-layer, SbpA) has been exploited as a cushion layer to generate a 2D array on a QCM-D sensor combined with electrochemical impedance spectroscopy (EIS). An S-layer-supported lipid bilayer membrane (SsLBM) was subsequently used to monitor the spontaneous incorporation of gramicidin (channel-forming peptide), αhemolysin (a pore-forming toxin), and the voltage-dependent anion channel (VDAC) protein (the main protein in the outer mitochondrial membrane). Functional incorporation and the formation of active channels that enable ion flux through the membrane were confirmed for gramicidin and the VDAC protein but not for α-hemolysin, which may be attributed to the protein shape and its impacts on assembly into a biomimetic model (Damiati et al. 2015a, b). As an equivalent type of ion channel that exhibits gating behavior without protein channels, ultrasound of GHz frequency (known as hypersound) and submicrometer wavelength has been proposed to affect membrane permeability. However,

23

Acoustic Biosensors for Cell Research

553

increasing the ultrasonic frequency accelerates the oscillation of cavitation bubbles and enhances the acoustic pressure, thus causing membrane disruption and allowing ion conduction across the SLB. Lu et al. (2019) fabricated a microchip to analyze the acoustic behavior of an SLB and monitor the currents across the SLB induced by hypersound on the same platform. The generated hypersonic nanopores enabled realtime monitoring of the changes in membrane permeability, which can be used to understand the mechanism of hypersonic poration and can be further exploited to enhance drug and gene delivery and control molecule encapsulation and release. Biomimetic sensors are also exploited in cell behavior studies that take into account the advantage of phosphocholine-based SLBs, which has antifouling properties that minimize the nonspecific adsorption of water-soluble proteins and cell attachment. Several factors prevent cell adhesion onto pure phosphocholine-based SLBs, such as surface hydration, surface electrostatics, and lateral mobility of lipid molecules (Glasmastar et al. 2002; Kilic and Kok 2018). However, phosphocholine bilayers can be modified with bioactive agents to promote cell-surface interactions and provide a promising approach for cell culture substrates to better mimic in vivo systems. Peptide-functionalized SLBs are a good model that allows for cell attachment and cell-surface interaction tracking in real time by QCM-D. SLBs were functionalized either with RGD, a hydrophilic tripeptide (Arg-Gly-Asp), or osteocalcin mimetic (OSN), a hydrophilic negatively charged peptide. In the RGD-SLB model, human fetal osteoblast cells underwent rapid protein-mediated specific interactions, enabling cell spread and adhesion, while OSN-SLB osteoblast cells did not spread well or specifically interact, promoting biomineralization rather than cell adhesion (Kilic and Kok 2018). In another study, the RGD-SLB model was also generated to evaluate the adhesion of bone mesenchymal stem cells (BMSCs). The results obtained from changes in frequency and dissipation signals confirmed that RGD offers an anchoring site on the SLB for cell adhesion and thus leads to changes in cell morphology due to rearrangement of the cytoskeleton (Zhu et al. 2014). Figure 8 shows various applications that use synthetic membranes for biomolecule (lipid or protein) interactions or for whole cell interactions.

Acoustic Biosensors for Cell Behavior: Whole Cell-Based Biosensors Whole cell-based biosensors are innovative approaches that enable the study of cell behaviors in their native states. Acoustic devices are able to monitor various cellular behaviors, such as cell morphology, signal transduction, division, and motility, noninvasively and in real time by tracking cell-substrate interactions. These investigations are important for medical diagnosis and pharmaceutical research because abnormalities in cellular processes have a direct impact on human health and can be used as markers for disease diagnosis (Chen et al. 2018). Cell adhesion is a fundamental approach for various cellular processes, such as cell proliferation, differentiation, migration, and survival. The importance of probing cell adhesion is that any abnormality in this process can cause many diseases, such as atherosclerosis (Galkina and Ley 2007) and cancer (Bendas and Borsig 2012). To remodel cellular

554

S. Damiati

Fig. 8 Schematic of lipid bilayer platforms for membrane protein-based biosensors and cell-based biosensors. (a) On the protein-based biosensor, different types of transmembrane proteins can be associated in various ways with the lipid membrane; (i) and (ii) integral monotopic proteins covalently attached to functionalized lipid bilayer or noncovalent adsorbed onto the lipid bilayer, respectively; (iii) polytopic transmembrane α-helical protein and (iv) ion channel protein reconstituted into the lipid bilayer. (b) On the cell-based biosensor, the functionalized lipid membrane promotes cell adhesion, mimicking the cell-surface interface

Fig. 9 The in vitro cell adhesion process. In the initial step, the cell attaches loosely onto the substrate surface. This contact is established by an integrin-mediated mechanism. Then, the cell membrane starts to spread over the substrate surface. Finally, the transmembrane integrin receptor forms focal adhesion complexes (FAs) with the ECM, enabling cells to strongly adhere to the substrate surface [Created with BioRender.com]

adhesion and spreading, precoating the sensor surface with an extracellular matrix (ECM) is a commonly constructed platform. Controlling the underlying substrate and the preadsorbed ECM improves adhesion and promotes different phenotypes (Lord et al. 2008); Khalili and Ahmad (2015). The adhesion process is established by the mechanical interactions between cells and the ECM via the cell membrane receptor integrin (Fig. 9). Such interactions affect and control cell behaviors and functions, in addition to providing structural and functional connections between intracellular and extracellular environments. Manipulation of cell-substrate adhesion

23

Acoustic Biosensors for Cell Research

555

can be accomplished by modifying the sensor surface with biological or nonbiological biocompatible materials to offer a convenient environment for adhesion with high selectivity and appropriate strength. Several factors need to be optimized to enhance cell adhesion on the acoustic surface, such as hydrophilicity, surface charge, and surface roughness. QCM can qualitatively monitor the dynamics of the adhesion process of varied cell types by relying on an acoustic wave which penetrated into the basal plane region of the cell. Hence, cell adhesion kinetics are attributed to the signals generated by molecular interactions and adhesion processes. Qualitative data about the dynamics of the adhesion process can be provided by the frequency. However, in cell adhesion studies, mass loaded on the acoustic sensor and changes in frequency do not obey the Sauerbrey equation because cell behaves as a soft material rather than a rigid mass on a QCM sensor. On the other hand, the dissipation response is a fluctuation that may increase or decrease during cell adhesion depending on the cell type. The fingerprint of cell adhesion is the ratio of dissipation to frequency (ΔD/ Δf) that relates to the structural and/or mechanical characteristics of adhered cells rather than the quantity and spatial distribution of adhered cells (Westas et al. 2015). A QCM cell composed of a cell culture incubator and a detection system was developed to monitor the growth of epithelial colorectal adenocarcinoma cells (Caco-2). Direct attachment and spreading of Caco-2 cells on the quartz sensor were monitored in the long term by measuring the frequency, amplitude, electrical resistance, and impedance of the QCM system. The fabricated system allowed investigations of cell adhesion, proliferation, and death (Lee et al. 2012). Another QCM platform was developed to study the variations in the acoustic sensor response to epithelial and fibroblast cell adhesion, in addition to tracking real-time formation and organization of the cytoskeleton (Da-Silva et al. 2012). The cell adhesion process was investigated by monitoring changes in frequency and resistance for three types of cells: epithelial, cervical cancer, and fibroblast cells on gold electrodes. Cell growth occurred on the QCM chip, and subsequently, the generated fingerprints of the cell adhesion process were distinctive for each cell type. Furthermore, it was possible to differentiate among the different phases of the cell adhesion process by combining this approach with microscopy, which enabled the characterization of several cell morphologies and physiological activities. Another QCM sensor was developed by applying ECM and separate cells from solid substrates to monitor the adhesion and growth of liver epithelial and lung melanoma cells. The gold electrode was modified with ECM proteins, either vitronectin or laminin. The cell adhesion and spreading processes exhibited variations in the frequency and resistance of the piezoelectric substrate. The two types of cells showed different behavior, which may be explained by cell motion, and various cellular metabolic activities near the sensing film (Fohlerová et al. 2007; Braunhut et al. 2005) adapted a QCM sensor to detect cell apoptosis and to predict the tumor response to taxanes, chemotherapeutic agents for many human cancers. Prediction of the resistance or hypersensitivity of breast cancer cells to the treatment was achieved by measuring real-time frequency and resistance. Changes in mass distribution and

556

S. Damiati

viscoelastic properties lead to alterations in cell morphology and loss of adhesion. These results present QCM as an effective tool for apoptosis detection that may be used prior treatment to predict therapeutic outcomes. QCM was also exploited to differentiate the metastatic potential of high- and low-metastatic breast cancer cells based on cell affinities toward transferrin molecules attached to gold sensor surfaces. The changes in the QCM signal were correlated with the number of transferrin receptors on the cells, and hence, transferrin receptor interactions and binding kinetics on the QCM sensor surface can potentially detect highly metastatic breast cancer cells (Atay et al. 2016). Simple acoustic systems were developed to evaluate the interaction of cell membrane receptors captured with immobilized ligands on LW and QCM-D sensors to provide complementary information about cell-substrate interactions using whole cells (Saitakis et al. 2010). Antihuman leukocyte antigen-A2 (HLA-A2) antibody was immobilized onto the sensor devices to enable attachment of LG2 Blymphoblastoid cells that express the human class I histocompatibility complex (HMC) molecule HLA-A2 on the cell membrane. Both LW and QCM-D sensors detected the number of cells in solution and the cells captured on the surface based on the HLA/anti-HLA interaction. However, the LW sensor differentiates the different types of LG2 cells based on the number of cells attached to the surface according to the specific cellular binding event between HLA and anti-HLA molecules. The QCM-D sensor was more sensitive for detecting cytoskeletal perturbations and responded more effectively to different treatments that impact cell membrane rigidity. A TSM acoustic sensor combined with an electrical impedance analyzer was also exploited to study the process of living cell adhesion in real time on a fibronectin-coated substrate (Li et al. 2007). Attachment, spreading, and the formation of FAs by human skin fibroblasts were assessed by tracking shifts in frequency and resistance. Changes in cell-substrate interactions and morphology are altered based on the cell density on the acoustic surface. Furthermore, these changes can result from the disruption of the actin cytoskeleton, which is important for cell adhesion. Furthermore, nonbiological materials such as tantalum (Ta), chromium (Cr), lithium tantalate (LiTaO3), and hydroxyapatite were used as surface coating materials to manipulate cell-substrate adhesion for different types of cells (Modin et al. 2006; Tagaya 2015). A QCM-D model was developed to assess the dynamic cell adhesion behavior in normal and cancerous human thyroid cells. Three surfaces were used: bare titanium, bare gold, and fibrinogen-coated gold. The adhesion pattern for normal and cancerous cells was different, with the acoustic signal decreasing in the order of titanium, gold, and fibrinogen-coated gold (Chronaki et al. 2016). A 36 YX-LiTaO3-based LW sensor with a parylene-C layer was developed to assess the adhesion process of tendon stem/progenitor cells (TSCs). The deposited parylene-C film on the acoustic sensor is used as a biocompatible interface and as a wave guiding layer. The propagation of LW occurs along the parylene-C layer, adherent TSC layer, and liquid medium. The LW sensor exhibited high sensitivity to surface perturbations and good response to the adhesion process of TSCs as a result of cell-substrate interactions (Wu et al. 2019).

23

Acoustic Biosensors for Cell Research

557

Acoustic Biosensors for Cell Detection Acoustic detection biosensors are noninvasive, and label-free tools offer many advantages, including compatibility, easy fabrication, and fast fluid actuation for cell detection (Ding et al. 2013). In general, biosensors for detection applications can mainly target three types of analytes: genetic biomarkers, protein biomarkers, and whole cells. Piezoelectric biosensors are considered to be very sensitive analytical techniques due to their ability to detect analytes in the picogram range (Kumar 2000). To achieve high sensitivity, several parameters should be verified for a biosensor system: (i) the biological molecules must preserve their biological activities when immobilized onto the sensor surface, (ii) the functional and structural properties of the biological film must be preserved while the film is associated with the sensor surface, (iii) the designed bioarchitecture on the sensor surface must be stable with long-term durability, (iv) the recognition biomolecules adsorbed onto the sensor surface need to be optimized to improve their distribution and orientation, and (v) a high degree of interaction specificity between the biological molecules must be validated (Tigli et al. 2010). However, due to high sensitivity, several piezoelectric biosensors were developed to detect pathogenic cells, including viral, bacterial, and cancer cells (Fig. 1). The ideal feature of detecting a whole cell is the possibility of eliminating preprocessing steps that make it applicable for bodily fluids (blood, saliva, sputum, and urine), water resources (natural, e.g., seas, rivers, or synthetic water such as pools), and food samples. A QCM-based immunosensor for the recognition of influenza A and B viruses in nasal wash samples was fabricated by Hewa et al. (2009). Conjugated complexes between gold nanoparticles and anti-influenza A or B antibodies were prepared to enhance the mass sensitivity of the sensor, which reached a detection limit of 1  103 pfu/mL. Compared to standard methods such as PCR, ELISA, standard cell culture, and shell vials, the developed QCM sensor offered several advantages, including high sensitivity and specificity, label-free measurements, and short time measurements. Another QCM immunobiosensor was designed for the rapid and sensitive detection of avian influenza virus (AIV) H5N1 (Wang and Li 2013). They developed an aptamer-ssDNA cross-linked polymeric hydrogel that formed an SAM on the gold sensor. Due to the cross-linking between the aptamer and ssDNA in the hydrogel, the polymer remains shrink when no cells are captured onto the sensor. The change of the hydrogel from shrink to abrupt swelling occurs when AIV H5N1 virus binds to the aptamer, leading to dissociation of the binding between the aptamer and ssDNA, which decreases the resonance frequency attributed to the swelling of the hydrogel on the QCM sensor and changes the viscosity. They also followed the same strategy to functionalize the hydrogel with anti-H5 antibody and developed a QCM immunosensor. The obtained results showed that the hydrogel QCM aptasensor has a lower detection limit and a shorter detection time. Bisoffi et al. (2008) fabricated an SH-SAW biosensor that allowed detection of coxsackie virus B4 and Sin Nombre virus (SNV), a hantavirus in the Bunyaviridae family of RNA viruses, within seconds. The viral agents were recognized and captured by specific antibody-target virus interactions and verified quantitatively by monitoring the phase

558

S. Damiati

differential mass shift (Δφ). The fabricated SAW device exhibited good efficiency in detecting viral particles in complex solutions. There is a high demand to produce rapid and sensitive sensors to detect pathogenic bacteria that do not require culture enrichment. Thus, several acoustic immunosensors have been fabricated either to diagnose bacterial infections or to prevent foodborne illness by detecting bacteria in food samples. Wang et al. 2018, reported on QCM immunosensors to detect Campylobacter jejuni (a leading cause of foodborne human gastrointestinal diseases) in poultry products. Campylobacter cells are first captured by an anti-C. jejuni antibody functionalized onto a magnetic nanobead surface and then bind to another anti-C. jejuni antibody immobilized onto the piezoelectric sensor. The magnetic nanobeads are used to improve the capture and separation efficiency of C. jejuni cells from the poultry samples and to enhance the sensor signal by increasing the loaded mass on the QCM surface. To further enhance the sensitivity of detection, a secondary antibody conjugated with gold nanoparticles was used to amplify the QCM signals of C. jejuni. The fabricated sensor enabled recognition of a small number of bacterial pathogens (20–30 CFU/mL) in less than 30 min. A dual channel SAW device has been fabricated to sense two different types of bacteria, Legionella and E. coli, simultaneously (Howe and Harding 2000). In contrast to the conventional biosensor, where the antibody was immobilized onto the sensor surface, the proposed sensor relied on the absorption of bacterial cells onto the SAW surface, and then the antibodies specifically bound to the cells. Changes in frequency related to the bound antibodies and observation of a dose-dependent effect depended on the number of cells coated on the sensor surface. This strategy achieved a better limit of detection than applying the antibody to the sensor first. As an alternative to classical recognition biomolecules (e.g., antibodies and aptamers) for bacterial cells on biosensors, whole cell imprinting-based sensors have also been developed by exploiting the imprinting process. The generation of cavities in a polymeric structure fits the shape of target cells and has chemical recognition memory that can be considered synthetic receptors similar to natural ones. Yilmaz et al. (2015) developed a whole cell imprinting-based QCM sensor for E. coli detection using a monomeric amino acid recognition element, N-methacryloyl L-histidine methylester (MAH), which is a polymerizable form of histidine. The E. coli imprinted polymeric film on the QCM surface showed the capability to distinguish E. coli with high affinity in a short time period. Acoustic sensors also developed for point-of-care (PoC) diagnostics to detect cancer cells with a hope that this approach will be a promising alternative to conventional invasive techniques such as biopsy. The importance of developing highly sensitive biosensors for cancer cell detection comes from the fact that cancer cells circulate at low numbers in bodily fluids, including blood, lymph, and amniotic fluid. Two strategies have been established to fabricate biosensors targeting cancer cells: ligand-dependent and ligand-independent approaches. In the first approach, there are several natural recognition biomolecules such as antibodies, aptamers, enzymes, receptors, and vitamins that have high affinity for targeting cancer cells due the recognition of specific receptors that are usually overexpressed on the

23

Acoustic Biosensors for Cell Research

559

surface of the cell membrane. In the second approach, cancer cells are separated from samples based on their physical properties, such as size, density, deformability, and electrical characteristics. Both strategies have exploited the acoustic technique to capture cancer cells. Circulating tumor cells (CTCs) were found in the range of 1–100 cells/mL of blood (Alix-Panabieres and Pantel 2014). Bröker et al. (2012) fabricated a nanostructured SAW chip-based immunosensor to detect two types of cancer cells: T-cell lymphoblastic leukemia and placental choriocarcinoma. The LW chip was composed of gold nanospots and self-assembled antibodies that clustered at a fixed spacing on the nanospot surface. The proposed SAW device offers a simple procedure for antibody self-assembly and prevents any interference from toxic agents in the surrounding fluid. Furthermore, it showed high sensitivity, as the sensor responded to 40  C for 2 h), the GUS enzyme is generated which can then be detected by using an enzyme-specific substrate (p nitrophenyl β-glucuronide (cells)/phenolphthalein β-glucuronide (plant)). The substrate-enzyme generates an electroactive product. The chip is applied with the redox potential of the electroactive product, and the effect of the heat is reflected in an increase in the current due to oxidation or reduction of the product. Measurements were conducted with a fixed quantity cells with induced with a heat shock and then tested using the chip for different concentration of the substrates (Fig. 16c). For the heat shock (HSP-) induced cells delineates the efficiency of promoter-based GUS synthesis in plant cells and its profiling. The cells induced with heat shock showed an increase in the current signal.

24

Electrodes for Cell Sensors Interfacing

593

Fig. 16 (a) Schematic of plant sensor. (b) Electrochemical chip – chip holder connected to the plant leaf. (c) Chronoamperogram of HSP/GUS cells at different concentration of substrate with and without heat shock induction. (d) Chronoamperogram of HSP/GUS plants with injected substrate

The measurement with the plants involves a manual injection of the substrate from the abaxial side of the leaf. This side is chosen due to presence of stomata, which are known to be responsible exchange of gases or chemical. The heat-induced (HSP+) plant shows an increase in the current when compared to the non-heat induced plant (Fig. 16d). This approach can be still improved by adding an automated microfluidics-based injection system and a more flexible polymer based or 3D printed electrode-microfluidics system.

ZnO Nanoparticle Electrodes In this section, we describe a recently investigated version of high band gap semiconductor electrodes for whole cell sensors. Specifically, we describe the use of ZnO nanoplates and nanorods and their possible modification by metal nanoparticles. The ease of fabrication using low cost processes, which can yield a wide range of nanostructures, makes ZnO-based matrices a promising platform for low cost biosensors. A great number of chemical and electrochemical approaches have been documented in the last few years (Rajesh and Kumar 2009). Zinc oxide (ZnO) is, indeed, a topic of fundamental and technical interest for many biochemical applications and semiconductor device applications, due to its wide range of properties. It crystallizes in the Wurtzite structure, has a wide direct band gap (3.44 eV) in the near-UV spectral region, and a large free-exciton binding

594

H. Ben-Yoav et al.

energy (60 meV). It also shows high transparency and piezoelectricity effects. Without much effort, it can be grown in different nanoscale forms, allowing various novel devices to be achieved. With its unique properties could provide a suitable microenvironment for immobilization of enzymes leading to an expanded use of this material for the construction of biosensors with enhanced analytical performance (Fiaschi 2018). Hence, different biosensors based on ZnO nanomaterials have been developed for the detection of glucose (Yogeswaran and Chen 2008; You et al. 2003a, b; Yoetz-Kopelman et al. 2016a), tyrosinase (Urmann et al. 2016), and uric acid (Couniot et al. 2016). In particular, the great research interest gained by this kind of nanostructures is derived from the fact that they can be prepared by a variety of methods and in a range of different morphologies. ZnO has one of the richest family of nanostructures among all materials, both in structures and properties. In fact, researchers have studied numerous ZnO nanostructures for biosensor applications, synthesized in many physical and chemical ways. These nanostructures include wires (Wang et al. 2007; Ahmad et al. 2009; Weber et al. 2008), rods (Liu et al. 2009; Wei et al. 2006; Kim et al. 2006), walls (Maiolo et al. 2014; Israr et al. 2011), particles (Umar et al. 2009; Kahn et al. 2008; Ren et al. 2009), tubes (Yang et al. 2009; Fulati et al. 2010), combs (Wang et al. 2006), as well as a number of decorated nanostructures (Chang et al. 2010; Yu 2009; Rahman et al. 2010; Dai et al. 2009). Furthermore, nanostructures in metal oxides have recently become important, because they can provide a bigger and effective surface for biomolecule immobilization with desired orientation, better conformation, and high biological activity resulting in enhanced sensing characteristics (Rahman et al. 2010). Moreover, the use of nanostructured ZnO allows for the possibility of many new signal transduction technologies in biosensors, arising from the sub-micrometer dimensions that can be utilized for simple and rapid analysis (Greene et al. 2006). The high surface-to-volume ratio of ZnO nanostructures provides greater enzyme loading and provides a favorable microenvironment, which can preserve the activity of the immobilized biomolecules. In addition, due to their excellent electron transfer rate, the ZnO nanostructures can facilitate the direct electrochemistry of the enzymes (Dai et al. 2009; Willander et al. 2008) (Fig. 17). Electrochemical techniques have emerged as new methods to synthesize ZnO nanostructures (Ahmad and Zhu 2011; Trano et al. 2014), allowing the simple synthesis of ZnO nanostructures. The most common method for the electrodeposition of ZnO involves the reduction of oxygen or nitrate on the substrate in the presence of a Zn salt at neutral pH and elevated temperature. With these techniques, it is possible to control the morphological and structural characteristics of the ZnO nanostructures by adjusting the growth process parameters, such as the reagent of interest, concentration, temperature, and pH (Pal and Santiago 2005; Tong et al. 2006; Cosentino et al. 2017). Moreover, the performances of ZnO nanorods-based electrodes can be greatly improved through decoration with metal nanoclusters. As an example, we have decorated ZnO nanostructured working electrode with different bimetallic (AuxPt1-x) nanoparticles.

24

Electrodes for Cell Sensors Interfacing

595

Fig. 17 (a) SEM pictures of bare ZnO electrode and (b) after the deposition of 5ML (Mono Layer equivalent) of Au clusters on top. Similar results are observed for Pt or AuPt clusters

Fig. 18 Nyquist diagrams recorded in the presence of pure PBS (black line) and of the mixture of pAP in PBS (red line) for bare ZnO (a) and nanocluster modified Au/ZnO (b), Pt/ZnO (c), and AuPt/ZnO (d). The injected product solution includes pAP (0.1 mg/ml) in PBS (0.1 M). The insets present the enlarged plots for high frequencies

596

H. Ben-Yoav et al.

The metal nanoparticles can electrically interface the redox centers in biomolecules with the electrode surface, enhancing the electron transfer efficiency between the electrode and the electrolyte, thus acting as a catalyst in the activation of electrochemical reaction (Luo et al. 2006; Yoetz-Kopelman et al. 2016b). Those results are shown in Fig. 18. The integration of catalytic metal nanoparticles on the wide band gap metal oxide semiconductor nanorods dramatically improves the detection capability of specific electroactive products such as p-aminophenol, a typical product to many enzyme sensors, especially in whole cell biosensors (Fiaschi et al. 2018; Pandey et al. 2018a).

Summary and Conclusions In this chapter, we present a brief overview of common electrodes, their material, architecture, structure, and preparation methods for biosensing in general and for whole cell biosensing in particular. There is a great diversity in such electrodes due to the various needs, analytes, and the plurality of analytical methods. A great improvement emerged by using nanostructured and nonplanar electrodes. Additionally, 3D printing also added another option of printing “application specific” and “target specific” electrodes that can be customized according to the need. These electrodes have been used for various whole cell sensors: microbes, yeast, mammalian cells, and plant. Each application requires a unique design from the mechanical, electrochemical, and electrical aspects. There is a lot of room for new design and applications employing new methods and approaches. Acknowledgment The research leading to these results has received funding from the European Union’s Seventh Framework Programme (FP7/2007–2013) under grant no. 607417 (CATSENSE). This research was also supported by the Israel Science Foundation (grant no. 1616/17). We would also like to acknowledge the Boris Mints Institute for Strategic Policy Solutions to Global Challenges, the Department of Public Policy and the Manna Centre for Food Security, Tel Aviv University for their generous support under the program “Plant based heat stress whole-cellbiosensor” (grant no. 590351) 2017.

References Ahmad M, Zhu AJ (2011) ZnO bnased advanced functional nanostructures: synthesis, properties and applications. J Mater Chem 21(3):599–614 Ahmad M, Pan C, Iqbal J, Gan L, Zhu AJ (2009) Bulk synthesis route of the oriented arrays of tipshape ZnO nanowires and an investigation of their sensing capabilities. Chem Phys Lett 480 (1):105–109 Almog RO, Sverdlov Y, Fishelson N, Shmilovich T, Rabinovich E, Shacham-Diamand YY (2010) Electrochemical micro technologies for polymeric MEMS and biochip applications. ECS Trans 25(16):17–21 Bai Q, Wise KD, Anderson DJ (2000) A high-yield microassembly structure for three-dimensional microelectrode arrays. IEEE Trans Biomed Eng 47(3):281–289

24

Electrodes for Cell Sensors Interfacing

597

Bard AJ, Faulkner LR (2001) Electrochemical methods: fundamentals and applications, 2nd edn. Wiley, New York, p 864 Beica R, Sharbono C, Ritzdorf T (2008) Through silicon via copper electrodeposition for 3D integration. Proc Electron Compon Technol Conf 406:577–583 Belkin S (2003) Microbial whole-cell sensing systems of environmental pollutants. Curr Opin Microbiol 6(3):206–212 Ben-Yoav H, Ofek Almog R, Sverdlov Y, Sternheim M, Belkin S, Freeman A, Shacham-Diamand Y (2012) Modified working electrodes for electrochemical whole-cell microchips. Electrochim Acta 82:109–114 Bousse L (1996) Whole cell biosensors. Sens Actuators B 34(1–3):270–275 Bouwen W, Thoen P, Vanhoutte F, Bouckaert S, Despa F, Weidele H, . . ., Lievens P (2000) Production of bimetallic clusters by a dual-target dual-laser vaporization source. Rev Scient Instr 71(1):54–58 Chang C, Chiu N, Lin D, Chu-Su Y, Liang Y, Lin AC (2010) High sensitivity detection of carbohydrate antigen 15-3 using a gold/zinc oxide thin films surface plasmon resonance-based biosensor. Anal Chem 82(4):1207–1212 Chaubey A, Malhotra BD (2002) Mediated biosensors. Biosens Bioelectron 17(6–7):441–456 Choi J-W, Rosset S, Niklaus M, Adleman JR, Shea H, Psaltis D (2010) 3-Dimensional electrode patterning within a microfluidic channel using metal ion implantation. Lab Chip 10(6):783–788 Clark LC, Lyons C (1962) Electrode systems for continuous monitoring in cardiovascular surgery. Ann N Y Acad Sci 102:29–45 Cosentino S, Fiaschi G, Strano V, Hu KJ, Liao TW, Mintz Hemed N et al (2017) Role of Au x Pt1–x clusters in the enhancement of the electrochemical activity of ZnO nanorod electrodes. J Phys Chem C 121(29):15644–15652 Couniot N, Francis LA, Flandre D (2016) A 16$\times $16 CMOS capacitive biosensor array towards detection of single bacterial cell. IEEE Trans Biomed Circuits Syst 10(2):364–374 Dai Z, Shao G, Hong J, Bao J, Shen AJ (2009) Immobilizatoin and direct electrochemistry of glucose oxidase on a tetragonal pyramid-shaped porous ZnO nanostructure for a glucose biosensor. Biosens Bioelectron 24(5):1286–1291 Dara RO, Jenkinson N, Banks CE (2009) Characterisation of commercially available electrochemical sensing platforms. Sens Actuators B 138(2):556–562 Dixit P, Miao J (2006) Aspect-ratio-dependent copper electrodeposition technique for very high aspect-ratio through-hole plating. J Electrochem Soc 153(6):G552 Eller A (1996) Amperometric biosensors. Curr Opin Biotechnol 7(1):50–54 Fiaschi G (2018) FET based whole cell sensor, PhD thesis, Tel Aviv University Fiaschi G, Cosentino S, Pandey R, Mirabella S, Strano V, Maiolo L, Grandjean D, Lievens P, Shacham-Diamand Y (2018) J Electroanal Chem 811:89–95 Fulati A, Ali SU, Riaz M, Amin G, Nur O, Willander AM (2010) Miniaturized pH sensor based on zinc oxide nanotubes/nanorods. Anal Chem 82(4):1207–1212 Greene BY, Law M, Zitoun D, Yang AP (2006) Solution-grown zinc oxide nanowires. Inorg Chem 42(26):3031–3034 Grieshaber D, MacKenziel R, Voros J, Reimhult E (2008) Electrochemical biosensors – sensor principles and architectures. Sensors 8:1400–1458 Hahm JI (2011) Functional polymers in protein detection platforms: optical, electrochemical, electrical, mass-sensitive, and magnetic biosensors. Sensors 11(3):3327–3355 Han J, Tan Z, Sato K, Shikida M (2004) Three-dimensional interconnect technology on a flexible polyimide film. J Micromech Microeng 14(1):38–48 Harnett CK, Satyalakshmi KM, Craighead HG (2001) Bioactive templates fabricated by low-energy electron beam lithography of self-assembled monolayers. Langmuir 17(1):178–182 Hemed NM, Convertino A, Shacham-Diamand Y (2016a) Investigation of functionalized silicon nanowires by self-assembled monolayer. Appl Surf Sci 367:231–236 Hemed NM, Yoetz-Kopelman T, Convertino A, Shacham-Diamand Y (2016b) Performance of whole-cell electrochemical biosensor using integrated microbes/Si Nano-Forest structure. ECS Trans 75(16):157–164

598

H. Ben-Yoav et al.

Hillman T, Badi AN, Normann RA, Kertesz T, Shelton C (2003) Cochlear nerve stimulation with a 3-dimensional penetrating electrode array. Otol Neurotol 24(5):764–768 Israr M, Sadaf J, Nur O, Willander M, Salman S, Danielsson AB (2011) Chemically fashioned ZnO nanoealls and their potential application for potentiometric cholesterol biosensor. Appl Phys Lett 98(25):253705 Jo BH, Van Lerberghe LM, Motsegood KM, Beebe DJ (2000) Three-dimensional microchannel fabrication in polydimethylsiloxane (PDMS) elastomer. J Microelectromech Syst 9(1):76–81 Johansson T, Abbasi M, Huber RJ, Normann RA (1992) A three-dimensional architecture for a parallel processing photosensing array. IEEE Trans Biomed Eng 39(12):1292–1297 Kahn R, Kaushik A, Solanki P, Ansari A, Pandey M, Malhotra AB (2008) Zinc-oxide nanoparticlechitosan composite film for cholesterol biosensor. Anal Chim Acta 616(2):207–213 Kang MG, Joon Park H, Hyun Ahn S, Jay GL (2010) Transparent Cu nanowire mesh electrode on flexible substrates fabricated by transfer printing and its application in organic solar cells. Sol Energy Mater Sol Cells 94(6):1179–1184 Kim G, Park J, Lee W, Yi C (2006) ZnO nanorod biosensor for highly sensitive detection of specific protein binding. J Korean Phys Soc 49(4):1635–1639 Kim ET, Seo J-M, Woo SJ, Zhou JA, Chung H, Kim SJ (2008) Fabrication of pillar shaped electrode arrays for artificial retinal implants. Sensors 8(9):5845–5856 Liu CC (1993) Applications of microfabrication techniques in electrochemical sensor development. Appl Biochem Biotechnol 41(1–2):99–107 Liu X, Hu Q, Wu Q, Zhang W, Fang Z, Xie AQ (2009) Aligned ZnO nanorods: a useful film to fabricate amperometric glucose biosensor. Colloids Surf B Biointerfaces 74(1):154–158 Liu F, Piao Y, Choi JS, Seo TS (2013) Three-dimensional graphene micropillar based electrochemical sensor for phenol detection. Biosens Bioelectron 50:387–392 Luo XL, Morrin A, Killard AJ, Smyth MR (2006) Electroanalysis 18:319 Luong JHT, Male KB, Glennon JD (2008) Biosensor technology: technology push versus market pull. Biotechnol Adv 26(5):492–500 Maiolo L, Mirabella S, Maita F, Alberti A, Minotti A, Strano V, Pecora A, Shacham-Diamand Y (2014) Flexible pH sensors based on polysilicon thin film transistors and ZnO nanowalls. Appl Phys Lett 105(9):93501 McLamore ES, Porterfield DM (2011) Non-invasive tools for measuring metabolism and biophysical analyte transport: self-referencing physiological sensing. Chem Soc Rev 40:5308 Meadows D (1996) Recent developments with biosensing technology and applications in the pharmaceutical industry. Adv Drug Deliv Rev 21(3):179–189 Mehrotra P (2016a) Biosensors and their applications – a review. J Oral Biol Craniofacial Res 6 (2):153–159 Mehrotra P (2016b) Biosensors and their applications – a review. J Oral Biol Craniofacial Res 6 (2):153–159 Melamed S, Elad T, Belkin S (2012) Microbial sensor cell arrays. Curr Opin Biotechnol 23(1):2–8 Mintz-Hemed N, Yoetz-Kopelman T, Convertino A, Freeman A, Shacham-Diamand Y (2017) Whole-cell electrochemical biosensor integrating microbes with Si nanowire-Forest. J Electrochem Soc 164(6):B253–B257 Motoyoshi M, Koyanagi M (2009) 3D-LSI technology for image sensor. J Instrum 4(03):23–26 Pai RS, Walsh KM, Crain MM et al (2009) Fully integrated three-dimensional electrodes for electrochemical detection in microchips: fabrication, characterization, and applications. Anal Chem 81(12):4762–4769 Pal U, Santiago AP (2005) Controlling the morphology of ZnO nanostructures in a low-temperature hydrothermal process. J Phys Chem B 109(32):15317–15321 Pandey R, Almog RO, Sverdlov Y, Shacham-Diamand Y (2017) Self-aligned electrochemical fabrication of gold nanoparticle decorated polypyrrole electrode for alkaline phosphatase enzyme biosensing. J Electrochem Soc 164(4):B168–B175 Pandey R, Friedberg S, Beggiato M, Sverdlov Y, Lishnevsky K, Demarchi D, Shacham-Diamand Y (2018a) Highly conductive copper film on inkjet-printed porous silver seed for flexible electronics. J Electrochem Soc 165(5):D236–D242

24

Electrodes for Cell Sensors Interfacing

599

Pandey R, Teig-Sussholz O, Schuster S, Avni A, Shacham-Diamanda Y (2018b) Integrated electrochemical chip-on-plant functional sensor for monitoring gene expression under stress. Accepted Publ J Biosens Bioelectron 117:493–500 Park J, Kim B (2012) 3D micro patterning on a concave substrate for creating the replica of a cylindrical PDMS stamp. Microelectron Eng 98:540–543 Park TH, Shuler ML (2003) Integration of cell culture and microfabrication technology. Biotechnol Prog 19(607):243–253 Pavesi A, Piraino F, Fiore GB, Farino KM, Moretti M, Rasponi M (2011) How to embed threedimensional flexible electrodes in microfluidic devices for cell culture applications. Lab Chip 11 (9):1593–1595 Ragones H, Schreiber D, Inberg A, Berkh O, Kósa G, Shacham-Diamand Y (2015a) Processing issues and the characterization of soft electrochemical 3D sensor. Electrochim Acta 183:125–129 Ragones H, Schreiber D, Inberg A, Berkh O, Kósa G, Freeman A, Shacham-Diamand Y (2015b) Disposable electrochemical sensor prepared using 3D printing for cell and tissue diagnostics. Sens Actuators B 216:434–442 Rahman M, Ahammad A, Jin J, Ahn S, Lee AJ (2010) A comprehensive review of glucose biosensors based on nanostructured metal-oxides. Sensors 10(5):4855–4886 Rajaraman S, Choi SO, McClain MA, Ross JD, Laplaca MC, Allen MG (2011) Metal-transfermicromolded three-dimensional microelectrode arrays for in-vitro brain-slice recordings. J Microelectromech Syst 20(2):396–409 Rajesh AT, Kumar D (2009) Recent progress in the development of nano-structured conducting polymers/nanocomposites for sensor applications. Sens Actuators B 136:275–286 Ren X, Chen D, Meng X, Tang F, Hou X, Han D, Zhang AL (2009) Zinc oxide nanoparticles/ glucose oxidase photoelectrochemical system for the fabrication of biosensor. J Colloid Interface Sci 334(2):183–187 Rnkainen NJ, Halsall HB, Heineman WR (2010) Electrochemical biosensors. Chem Soc Rev 39 (5):1747–1763 Sato H, Yoshimine K, Otsuka T, Shoji S (2007) Interdigitated array 3D micromesh electrodes for electrochemical sensors. J Micromech Microeng 17(5):909–914 Shacham-Diamand Y, Osaka T, Datta M, Ohba T (eds) (2009) Advanced nanoscale ULSI interconnects: fundamentals and applications. Springer, New York Singh P (2016) SPR biosensors: historical perspectives and current challenges. Sens Actuators B 229:110–130 Soper SA et al (2006) Point-of-care biosensor systems for cancer diagnostics/prognostics. Biosens Bioelectron 21:1932–1942 Su L, Jia W, Hou C, Lei Y (2011) Microbial biosensors: a review. Biosens Bioelectron 26(5):1788–1799 Timko BP, Cohen-Karni T, Qing Q, Tian B, Lieber CM (2010) Design and implementation of functional nanoelectronic interfaces with biomolecules, cells, and tissue using nanowire device arrays. IEEE Trans Nanotechnol 9(3):269–280 Tong Y, Liu Y, Dong L, Zhao D, Zhang J, Lu Y, Shen D, Fan AX (2006) Growth of ZnO nanostructures with different morphologies by using hydrothermal technique. J Phys Chem B 110(41):20263–20267 Thevenot DR, Toth K, Durst R, Wilson GS (2001) Electrochemical biosensors: recommended definitions and classification. Biosens Bioelectron 16:121–131 Trano VS, Urso R, Scuderi M, Iwu K, Simone F, Ciliberto E, Spinella C, Mirabella AS (2014) Double role of HMTA in ZnO nanorods grown by chemical bath deposition. J Phys Chem C 118 (48):28189–28195 Turner APF (2013) Biosensors: sense and sensibility. Chem Soc Rev 42(8):3184–3196 Turner APF, Karube I, Wilson GS (1987) Biosensors: fundamentals and applications. Oxford University Press, New York Umar A, Rahman M, Vaseem M, Hahn AY (2009) Ultra-sensitive cholesterol biosensor based on low-temperature grown ZnO nanoparticles. Electrochem Commun 11(1):118–121 Urmann K, Arshavsky-Graham S, Walter JG, Scheper T, Segal E (2016) Whole-cell detection of live lactobacillus acidophilus on aptamer-decorated porous silicon biosensors. Analyst 141(18):5432–5440 Verma A (2006) Oxygen-sensing in tumors. Curr Opin Clin Nutr Metab Care 9(4):366–378

600

H. Ben-Yoav et al.

Wang J (2001) Analytical electrochemistry, 2nd edn. Wiley, Hoboken Wang J (2006) Electrochemical biosensors: towards point-of-care cancer diagnostics. Biosens Bioelectron 21(10):1887–1892 Wang J (2008) Electrochemical glucose biosensors. Electrochem Sensors Biosens Biomed Appl 57–69 Wang W-C (2014) Polymeric micro sensors and actuators. Sensors 14(8):15065–15066 Wang J, Sun X, Wei A, Lei Y, Cai X, Li C, Dong AZL (2006) Zinc oxide nanocomb biosensor for glucose detection. Appl Phys Lett 88(23):233106 Wang X, Wang F, Yang W, Yang AX (2007) A study on the antibacterial activity of one-dimensional ZnO nanowire arrays: effects of the orientation and plane surface. Chem Commun 42:4419–4421 Weber J, Jeedigunta S, Kumar AA (2008) Fabrication and characterization of ZnO nanowire arrays with an investigation into electrochemical sensing capabilities. J Nanomater 2008:75 Wei A, Sun X, Wang J, Lei Y, Cai X, Li C, Dong Z, Huang AW (2006) Enzymatic glucose biosensor based on ZnO nanorod array grown by hydrothermal decomposition. Appl Phys Lett 89 (12):123902 Willander M, Klason P, Yang L, Al-Hilli S, Zhao Q, Nur AO (2008) ZnO nanowires chemical growth, electrodeposition, and application to intracellular nano-sensors. Phys Status Solidi C 5 (9):3076–3083 Xiao T, Wu F, Hao J, Zhang M, Yu P, Mao L (2017) In vivo analysis with electrochemical sensors and biosensors. Anal Chem 89(1):300–313 Yang K, Shhe G, Wang H, Ou X, Zang X, Lee C, Lee AS (2009) ZnO nanotube arrays as biosensors for glucose. J Phys Chem C 113(47):20169–20172 Yoetz-Kopelman T, Porat-Ophir C, Shacham-Diamand Y, Freeman A (2016a) Whole-cell amperometric biosensor for screening of cytochrome P450 inhibitors. Sens Actuators B 223:392–399 Yoetz-Kopelman T, Porat-Ophir C, Shacham-Diamand Y, Freeman A (2016b) Whole-cell amperometric biosensor for screening of cytochrome P450 inhibitors. Sens Actuators B 223:932–399 Yogeswaran U, Chen SM (2008) A review on the electrochemical sensors and biosensors composed of nanowires as sensing material. Sensors 8(1):290–313 You T, Niwa O, Tomita M, Hirono S (2003a) Characterization of platinum nanoparticle-embedded carbon film electrode and its detection of hydrogen peroxide. Anal Chem 75:2080–2085 You T et al (2003b) An Amperometric detector formed of highly dispersed Ni nanoparticles embedded in a graphite-like carbon film electrode for sugar determination. Anal Chem 75:5191–5196 Yu T (2009) Carbon-decorated ZnO nanowire array: a novel platform for direct electrochemistry of enzymes and biosensing applications. Electrochem Commun 11(1):202–205 Ziegler C (1998) and W. Göpel. Biosensor development. Curr Opin Chem Biol 2(5):585–591 Zou Y, Pisciotta J, Baskakov IV (2010) Nanostructured polypyrrole-coated anode for sun-powered microbial fuel cells. Bioelectrochemistry 79(1):50–56

New Materials for the Construction of Electrochemical Cell-Based Biosensors

25

Andreea Cernat, Bianca Ciui, Luminita Fritea, Mihaela Tertis, and Cecilia Cristea

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Generalities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Electrochemical Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cell Viability Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Improvement in the Properties of the Whole-Cells Biosensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . Carbon-Based Nanomaterials as Platform for Cell Biosensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Carbon Nanotubes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Graphene . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Other Carbon-Based Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Metallic-Based Nanomaterial as Platform for Cell Biosensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Wearable Sensing Devices Future Trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . New Materials for Flexible Wearable Sensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . New Materials for Stretchable Wearable Sensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . New Materials for Self-healing Wearable Sensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

602 604 604 605 607 608 609 612 613 614 616 618 620 620 622 622

A. Cernat · B. Ciui · M. Tertis · C. Cristea (*) Department of Analytical Chemistry, Faculty of Pharmacy, University of Medicine and Pharmacy Iuliu Hatieganu Cluj-Napoca, Cluj-Napoca, Romania e-mail: [email protected]; [email protected]; [email protected]; [email protected] L. Fritea Department of Preclinical Disciplines, Faculty of Medicine and Pharmacy, University of Oradea, Oradea, Romania e-mail: [email protected] © Springer Nature Switzerland AG 2022 G. Thouand (ed.), Handbook of Cell Biosensors, https://doi.org/10.1007/978-3-030-23217-7_138

601

602

A. Cernat et al.

Abstract

The development of new biosensors with applications in various domains represents a contemporary challenge that it is extensively studied. The elaboration of cell-based biosensors has been a major outbreak in the sensing field due to their high sensitivity and specificity doubled by high stability and catalytic activity of the enzymatic systems included in the immobilized living cells. The major drawback of the conventional enzymatic biosensors represented by the loss of the enzymatic activity was eliminated by the immobilization of whole-cells on sensing platforms while maintaining their stability. The selection of the wholecell, either mammalian or microorganism, orientates the detection towards heterogeneous compounds such as heavy metals, pollutants, foodborne pathogens, and biomedical biomarkers. By keeping the enzymes in a cellular environment, the enzymatic turnover was facilitated and their catalytic activity loss was greatly reduced. A key point is represented by the selection of suitable platforms that can ensure the stability of the cell and allow the monitoring of the metabolic transfer with the extracellular environment. Depending on the end-application, the cells could be coated with a protective polymeric layer or they can be immobilized in different biocompatible polymers. Moreover, adding carbon-based and/or metallic nanoparticles, their stability and catalytic properties are highly improved. The future trends in sensing strategies involve the association between different nanomaterials, miniaturization and the development of out-of the box sensing devices with improved analytical performances. The present chapter discusses multiple approaches for the elaboration of nanohybrid platforms, along with their advantages and limitations and it also underlines the materials used for designing wearable sensing devices. Keywords

Whole-cell · Carbon-based nanomaterials · Metallic nanoparticles · Coating · Wearable sensors · Surface modification

Introduction Cell-based biosensors are sensing devices that employ living cells, such as bacteria or yeast strains, as biorecognition elements. Coupled with a specific transducer, they are able to detect the physiological exchanges with the close environment, while proving an outstanding capacity of regeneration of their enzymatic systems. Starting with the first cell-based biosensor developed by Diviès in 1975, their development registered a continuous progress linked to new immobilization techniques, nanohybrid platforms, and a wide range of applications across biomedical, food safety and toxicology, environmental analysis, and biofuels development. The conventional biosensors employ enzymes or antibodies immobilized on a sensing platform, but their major drawbacks are related to their limited renewability and the inactivation risk in the synthesis protocol and during the analysis. Therefore,

25

New Materials for the Construction of Electrochemical Cell-Based Biosensors

603

their elaboration requires controlled protocols of immobilization and physiological conditions suitable for the biological systems. The whole cells include different biomolecules such as enzymes, coenzymes, and cofactors that are programmed to follow metabolic pathways that regulate the homeostasis of the cell. Moreover, they have self-recycling process and their enzymatic systems are employed for the detection of target analytes. Thus, the major advancement in the progress of biosensing devices could be attributed to the complexity of these systems that are capable to catalyze the simultaneous electrochemical oxidation/reduction of various analytes. The selection of the cell that could be included in a biosensor depends on its characteristics extensively documented in a controlled environment. An example is represented by the immobilization of strains that grow in harsh environments and require specific compounds to yield metabolic pathways and could be used for the analysis of heavy metals and pollutants. Also, the strains cultured in a friendly environment respond to specific stimuli that could be used as possible candidates for biosensor target. The challenge to develop cell-based biosensors consists in the incorporation of the whole cell in the configuration of the transducer relying then on a high specificity and selectivity. The analyses can be performed either in situ or ex situ. The two types of bioassays are represented by turn on and turn off systems. In the first case, the enzymatic systems are specifically activated by the established target and the electrochemical response is proportional with the enzymatic process. In the second case, the target inhibits the metabolic processes (growth, respiration rate, motility depletion of the whole-cell), and its concentration is proportional with the degree of inhibition (Belkin 2003). Currently, there are several types of microorganisms and cells that are incorporated into biosensors depending on the applications. For instance, the highly sensitive and specific sensing microbial biosensors for the heavy metal detection are well established. Several cells developed the ability to respond to changes in their environment and starting with this ability, the incorporation of one or more microorganisms capable of producing a measurable signal represented the start of those analytical devices (Belkin 2003). The advantages of using the whole cells instead of enzymes or other bioreceptors are based on few important features of these living systems. While enzyme biosensors require special attention when attaching the bioelement on the surface of transducers and a control environment, the whole cell biosensors being aggregates of enzymes, cofactors and coenzymes behave like a device capable of assuring chemical reactions. Another big advantage is the presence of several enzymes which could work simultaneously performing a well-defined succession of reactions on different substrates. An important goal of whole cell biosensors is to obtain a sensitive and selective response. While the selectivity of the living systems is assured by the performance of the whole cell, with the innovation in nanotechnology and nanomaterials fields, it is offered interesting solutions for the enhancement of the sensitivity. From carbonbased materials able to increase the active surface of the transducers by hundreds of time, to biocompatible nanoparticles, there is a large plethora of nanomaterials that could be integrated into the whole-cell biosensors.

604

A. Cernat et al.

Generalities Electrochemical Approach Electrochemical sensors consists of three electrodes, namely, reference electrode, working electrode and auxiliary electrode, being all simultaneously placed in contact with the sample. The working electrode has different geometries and configurations and can have a conventional or a miniaturized feature and it is used by itself or included in a screen-printed electrochemical cell together with the reference and auxiliary electrodes. The working electrode is made of large plethora of materials such as noble metals (gold and platinum), carbon-based materials, and indium tin oxide (ITO) (Freitas et al. 2018). Furthermore, the surface of the working electrode can be modified with different nanocompounds such as carbon-based nanomaterials (carbon nanotubes, graphene, fullerene, carbon nanodots, etc.), gold nanoparticles, other metallic and/or magnetic particles, etc. These nano-entities are simply deposited onto the electrode surface or embedded in different polymeric films or membranes, various nanocomposite materials being, thus, obtained and used for cellbased biosensor development. Promising strategies have been applied for the design and elaboration of electrochemical cell-based biosensors for various applications across multidisciplinary domains. Amperometric, voltammetric, and impedimetric methods were especially used as electrochemical techniques for the detection and quantification of biomedical, environmental or food and alimentary markers, since they provide low limits of detection (ngmL
1 to fgml
1). A major concern of enzyme-based biosensors is related to the stability of the enzymatic compound after the immobilization into a compatible matrix through covalent or non-covalent bonds. Despite the good analytical performances and low detection limits, their elaboration requires time consuming and expensive protocols, doubled with the decrease/loss of their catalytic activity and turnover. A novel and alternative strategy is represented by the immobilization of a microbial cell that displays intracellular enzymatic systems, achieving, thus, the preservation of the enzymatic activity and unlimited regeneration under controlled environment. As a consequence, the cellbased biosensors have a high stability, specificity, reproducibility but are also benefit from low-cost preparation when compared with enzyme-based biosensors that use highly purified enzymes (Liang et al. 2013a). The whole cells that are used as sensing elements can offer in real-time valuable information which is related to their physiological status and microenvironment, and depending on their affinity could be used for the detection of disease biomarkers, pollutants, and micronutrients. The detection of foodborne pathogens, quantification of toxins, analysis of allergens, contamination of food with heavy metals and/or antibiotics, represents another side of the applications of this type of biosensors in food safety and quality control (Ye et al. 2018). The types of cells used for the development of whole cells biosensors are represented by: (i) Mammalian cells that allow various biochemical mechanisms and signaling pathways relevant to humans and animals. Mainly, they are used in the

25

New Materials for the Construction of Electrochemical Cell-Based Biosensors

605

development of food-control sensing device, when they become the living biocompounds that are the real target of food ingredients. The major drawback is that they are difficult to culture, grow and they must be genetically engineered when comparing with bacteria (Banerjee and Bhunia 2009). (ii) Microorganism cells have the advantage of facile culture, rapid growth, ease of genetic manipulation and the ability to metabolize various compounds (Kintzios and Banerjee 2015). The majority is represented by bacteria and yeast strains. Recently, bacteriophages have been used as recognition elements for bacterial biosensing due to some advantages such as high specificity, simple production, good stability, accuracy, reduced assay times, and low cost. Phage-based biosensors have been employed for sensitive, specific, rapid, cheap, and easy identification of a wide range of bacteria. Various strategies have been developed for the phage immobilization such as adsorption, covalent bonding, and entrapment in polymeric matrix. Their controlled expression and orientation on the electrode surface is also possible due to the presence of various functional groups (with positive and negative charges) on their heads and tail fibers (Farooq et al. 2018).

Cell Viability Strategies Immobilization of a biocatalyst on the surface of a working electrode could enhance its maximum potential by easy recovery, reusability, and stability. The development of cell-based biosensors with living cell as recognition elements is mainly focused on immunological, optical, and electrochemical techniques. A major concern is focused on the viability of the cells immobilized on different platforms. The conventional immobilization involves different techniques (chemical crosslinking, adsorption, matrix embedment) can have as drawback a decrease in viability of the cells linked to a poorer analytical performances of the sensor (Fakhrullin et al. 2012). This issue could be addressed by coating the bacterial or yeast cell with different materials that are able to maintain their integrity and physiological processes. The encapsulation with inorganic layers can enhance the functionalities and protection against harsh environments, suitable for the development of cell-based sensors and reactors (Liu et al. 2018b). A suitable option is represented by the charged polymers (e.g., cationic poly (allylamine)hydrochloride and anionic poly(styrene) sulfonate) that ensure the electrostatic interactions necessary for the adsorption on substrates and facilitate the electron transfer rate. Also the coating polymer needs to be biocompatible in order to maintain the integrity of the cell. Thus, different materials such as chitosan, hyaluronic acid, polyglutamic acid, polylysine, albumin, lysozyme, gelatin A, protamine sulfate, chondroitin sulfate, cellulose, b-lactoglobulin, alginate, and plantderived pyrogallol were used as polyelectrolytes (Dai et al. 2018). The coating conductive nanomaterials could be employed following different strategies: the direct assembly of the nanoparticles on the cell surface, the in situ production, or different metabolic process used for signaling the coating process. Despite the biocompatibility of these materials, the viability of the cells could be affected and

606

A. Cernat et al.

many studies based on Escherichia coli, as a model molecule, were focused on increasing the cell stability and the sensitivity of the device. Furthermore, it was observed that coating cell membranes with different films using layer-by-layer technique helps to preserve cell integrity during the multiple processing and handling steps (cycles of centrifugation, washing, etc.), as graphically illustrated in Fig. 1 (Matsuzawa et al. 2012). 3D nanomaterials with higher strength-to-weight and surface-area-to-volume ratios were studied as suitable for the bio-recognition element immobilization. Moreover, the supplementary addition of metallic nanoparticles as nanocoating for biological cells increased the viability during the elaboration and the testing protocol, while maintaining the metabolic functions. This technique could achieve controlled immobilization, a critical step in the development of electrochemical sensors. Magnetic nanoparticles stabilized with poly(allylamine)hydrochloride were used for coating living GreenScreenTM yeast and chlorella cells and a magnetic field was used for the accumulation on screen printed electrodes. The sensor was employed for the detection of atrazine and propazine, used as herbicide agents. The advantage of this approach is defined by the reusability of the electrodes, the accumulation process being controlled by a magnetic field (Zamaleeva et al. 2011). Nanoporous gold with a high active surface area was used for coating Escherichia coli cells and deposited on a glassy electrode surface. The biomaterial was used as a bioreceptor for the detection of sulfide (Liu et al. 2017a). Calcium ions linked to citrate gold nanoparticles (AuNPs) were functionalized with graphene oxide sheets and the

Fig. 1 Schematic illustration of the effect of physical stress on (a) uncoated cell or (b) cell with films by layer-by-layer assembly. (Reprinted with permission from (Matsuzawa et al. 2012). Copyright (2018) American Chemical Society)

25

New Materials for the Construction of Electrochemical Cell-Based Biosensors

607

nanohybrid material was used as coating for Saccharomyces cerevisiae yeast cells. The Ca+2 ions on the surface of AuNPs were used as binding between the graphene oxide sheets and yeast cells (Kempaiah et al. 2011a). Also, chemically reduced graphene form an electrical and thermal conducting layer on the surface of Saccharomyces cerevisiae that has a 70–90 nm cell wall mainly composed of polysaccharides (carbohydrates), and hence it is an electrical insulator (Kempaiah et al. 2011b). As it can be seen, the carbon-based nanomaterials have the advantage of excellent electrical conductivity when coating the whole cell and allow the physiological processes between the cell and the environment, essential features for the development of whole-cells biosensors.

Improvement in the Properties of the Whole-Cells Biosensors The bacterial cells immobilized on the working electrode, such as Pseudomonas syringae, Escherichia coli, Bacillus thuringiensis (Shao et al. 2009), could be employed as stable platforms for anchoring other proteins. Thus, the microorganism membranes display on their surface various biomolecules such as enzymes (xylose dehydrogenase, glucose dehydrogenase) acting as biocatalysts for the sensitive and selective detection of saccharides such as xylose or glucose. Two microbial biosensors were developed based on Escherichia coli cells which were immobilized at the electrode via a composite material of Nafion and multiwalled carbon nanotubes (MWCNTs) in the first example or via MWCNTs followed by the immobilization of the nanomaterial modified with bacteria with a Nafion layer in the second example. This procedure allows a stable, fast, sensitive, and selective detection of xylose and glucose respectively (Han et al. 2018; Li et al. 2012; Liang et al. 2013a). A bacterial laccase was linked to the surface of Escherichia coli immobilized by adsorption directly onto a glassy carbon electrode and employed for the detection of catechol. Despite the simplicity of the immobilization method, the biosensor retained 89.3% of the initial enzymatic activity when kept 1 month at room temperature, while being tested daily for the detection of catechol (Zhang et al. 2018). Another similar approach describes the modification of Vero cells from African Green monkey kidney with antiaflatoxin B1 antibodies by electroinsertion and osmotic insertion. The cells were immobilized onto the working surface of screen-printed electrodes modified with AuNPs and coated with poly-L-lysine. The use of this polymer ensured the adherence of the cell on different substrates, such as glass or plastic (Mavrikou et al. 2017). Another material that ensured a high enzymatic activity is represented by agar that was used as a supporting material for whole cell cyanide dihydratase of Flavobacterium indicum physically adsorbed onto nylon membrane. The ammonium ions released as a result of cyanide hydrolysis by the enzyme were quantified with an ammonium ion selective electrode (Mavrikou et al. 2017). Also, the selection of the electron-shuttle mediator for the elaboration of the whole-based biosensors cells is of high importance in order to facilitate the electron transportation through the bacterial wall. Both hydrophilic (potassium ferricyanide) and lipophilic (benzoquinone, menadione, dichloroindophenol, 2,3,5,6-tetramethylphenylenediamine, neutral red) mediators

608

A. Cernat et al.

were studied and different conclusions were drawn: in the first case despite the high water solubility and diffusion, the permeability through the lipophilic microbial cell is highly limited, allowing only the interaction with the proteins from the periplasm; in the case of lipophilic ones, the permeability through the membrane was not an issue, but their low solubility in aqueous media reduces the detection signal and thus the performances of the sensor. The strategy of using a combined mediator system, with both types solved the problem of crossing the lipophilic membrane and diffusion in aqueous media. A commonly used double-system is represented by menadioneferricyanide for the investigation of the redox activity of several enzymes (Majdinasab et al. 2018). The surfaces of the electrodes modified with nanostructured and composite materials can increase the analytical performance of the cell-based biosensor and opens the way for miniaturization and portability for these biosensors. These modifiers also contribute to the stability and improvement of the immobilization as well as the load in the whole cell entities that enhance the sensitivity of the sensor and catalyzes the electrochemical process involved in detection. Electrochemical impedance spectroscopy (EIS) was very often applied as electrochemical technique in order to detect Escherichia coli and Pseudomonas aeruginosa using magnetic beads, AuNPs, carbon nanotubes and graphene as nanomaterials which improved not only the sensitivity/conductivity of the biosensor, but also the phage immobilization (Farooq et al. 2018).

Carbon-Based Nanomaterials as Platform for Cell Biosensors Carbon and carbon-based materials and nanocomposites have become attractive materials for electrochemical biosensors development and are widely used as modifiers for electrode surface in order to increase the active surface area and implicitly the sensitivity, mainly due to the increase in the number of biologically immobilized entities. Besides this, their unique physical, chemical, electrical, optical, mechanical, and thermal properties, including their excellent electrical conductivity and biocompatibility, qualify them for use in the construction of electrochemical biosensors with amazing environmental and biomedical applications where the biorecognition elements were immobilized by covalent and noncovalent methods. A wide variety of nanomaterials including carbon based nanomaterials and nanoparticles has been successfully incorporated in electrochemical biosensors with sensitive, selective, and fast response taking advantage of their nanometer size and high surface area (Majdinasab et al. 2018; Malekzad et al. 2016). Different type of carbon-based nanomaterials such as fullerene, carbon dots (CDs), carbon nanohorns (CNHs), carbon nanotubes: single-walled carbon nanotubes (SWCNTs), MWCNTs, graphene nanosheets, graphene, graphene oxide, reduced graphene oxide as well as graphene doped with different elements (nitrogen-doped graphene sheets), graphene-based hybrid nanomaterials and boron-doped nanocrystalline diamond were used in biosensors development. Furthermore, synthesis of conjugated nanomaterials composed from gold, polymers, ionic liquid, and carbon nanomaterials are starting

25

New Materials for the Construction of Electrochemical Cell-Based Biosensors

609

blocks for the development of electrochemical cell-based biosensors with superior analytical performance and good stability of the biomolecules (Yang et al. 2018). Some examples of cell-based biosensors will be presented below, highlighting the type of nanomaterial used to modify the electrode surface and its role in the electrochemical detection process.

Carbon Nanotubes Carbon nanotubes represent carbon based nanomaterials composed by one (SWCNTs) or multiple layers (MWCNTs) of carbon sheets giving dimensions (length and diameter) in micro and nanometers range. An enzymatic assay for the direct D-glucose electrochemical sensing was constructed starting from a glassy carbon electrode modified with MWCNTs, yeast (as whole-cell entity), a-agglutinin as an anchor and glucose oxidase (GOx) as biosensing element. The immobilization of MWCNTs at the electrode surface was obtained via Nafion after the evaporation of the solvent used for the suspension preparation. The GOx displayed on the yeast cell surface had a good stability within a pH range from 3.5 to 11.5, excellent thermostability, as well as good specificity for the target molecule compared with conventional GOx enzymatic system. This biosensor was successfully applied for D-glucose sensing with a detection limit of 0.05 mM and proved to be stable, specific, reproducible, simple, and cost-effective, when applied for D-glucose detection in real samples. The photographs of unmodified yeast cell and of GOx surface-displayed yeast are comparatively presented in Fig. 2 (left), while in the right side of Fig. 2, cyclic voltammograms of the cell-based biosensor is compared with the other steps involved in the elaboration protocol (Wang et al. 2013).

Fig. 2 (Left) Photograph of (a) yeast cell control and (b) GOx surface-displayed yeast (Right) CVs of (a) Nafion/GOx-yeast/GCE, (b) Nafion/MWNTs/GCE, and (c) Nafion/GOx-yeast/MWNTs/GCE in N2-saturated PBS solution (pH 7.4). (Reprinted with permission from (Wang et al. 2013). Copyright (2018) American Chemical Society)

610

A. Cernat et al.

Li et al. fabricated an innovative cell-based electrochemical biosensor for the sensitive and selective determination of d-xylose (INS 967). A MWCNTs and Nafion composite film was used for this biosensor construction, together with xylose dehydrogenase (XDH). The XDH-displayed bacteria were anchored using an ice nucleation protein from Pseudomonas borealis DL7 and used without any further extraction or purification steps. This sensor based on MWCNTs, bacteria-XDH and Nafion presented long-term stability, wide dynamic range, a low detection limit of 0.5 μM for D-xylose and no interferences from D-glucose, D-fructose, D-maltose, D-galactose, D-mannose, D-sucrose, D-cellobiose, and L-arabinose. The nanobiocomposite sensor elaboration was fast and simple and it was successfully applied for real samples analysis with excellent recoveries revealing good accuracy of the proposed detection method (Li et al. 2012). A bacterial impedimetric biosensor was elaborated for trichloroethylene detection, based on the immobilization of Pseudomonas putida F1 strain on gold microelectrodes modified with SWCNTs modified with carboxylic moieties. The SWCNTs units were covalently linked to anti-Pseudomonas antibodies, and this sensing system allowed the sensing of the target with a limit of detection of 20 g/L and a good stability in time up to 4 weeks. This microbial biosensor was tested for the determination of trichloroethylene in real water samples, very good recoveries being obtained. The immobilization of Pseudomonas putida was documented by atomic force microscopy (AFM) in tapping mode (Fig. 3) (Hnaien et al. 2011). The use of mammalian cells as biocompound for the sensors design can provide useful information about human physiology. An interesting approach is related with the use of screen-printed electrodes modified with MWCNTs for the immobilization

Fig. 3 Tapping mode AFM images of anti-Pseudomonas antibodies fixed on SWCNT-COOH functionalized electrode (a) and of bacterial cells bound to the electrode (b) (Hnaien et al. 2011). (Reprinted with permission from (Hnaien et al. 2011). Copyright (2018) Elsevier)

25

New Materials for the Construction of Electrochemical Cell-Based Biosensors

611

of the living mast cells and its use for the detection and quantification of the N-acylhomoserinelactones, as spoilage bacterial quorum signaling molecules in freshwater fish. A correlation of the biosensor signal with cell concentration was obtained using EIS, the results being also confirmed with the flow cytometry analysis, by the detection of cellular calcium concentration, and observing the cell morphology. This electrochemical cell-based biosensor is simple, fast, low-cost and allows renewal of the initial sensing surface, thus, being suitable for multiple use to evaluate the growth and propagation of spoilage bacteria in the fresh fish. In this case, the use of MWCNTs increases the electrochemical signal and enhances the sensitivity towards the target analyte. The electrochemical biosensor stepwise fabrication is shown in Fig. 4. Bare screen-printed electrodes were stabilized using cyclic voltammetry, then MWCNTs were drop-casted from aqueous suspension and dried in nitrogen stream. A cell suspension was deposited on the surface of the screenprinted electrodes modified with MWCNTs and covered with a hydrogel based on sodium alginate and graphene oxide, at 37  C for 5 min and conditioned in CaCl2buffered DMEM medium for 5 min in 37  C. The morphology of the cells was revealed using scanning electron microscopy tests performed for native RBL cells (Fig. 4 down, left image) and RBL cells modified with AHLs (Fig. 4 down, right image) (Jiang et al. 2018).

Fig. 4 (Up) Schematic illustration for the preparation of cell sensor and process of AHLs detection. (Down) The morphology of naive RBL cells (left) and RBL cells treated with AHLs (right). (Reprinted with permission from (Jiang et al. 2018). Copyright (2018) Elsevier)

612

A. Cernat et al.

Another similar configuration based on MWCNTs and Nafion was employed for the elaboration of a microbial sensor based on the resting cells of Methylobacterium extorquens AM1. The biosensor proved good analytical performance for formate electrochemical oxidation and carbon dioxide reduction, this being due to formate dehydrogenase contained in the resting cells and to methyl viologen used as mediator (Xia et al. 2018). Carboxylated MWCNTs in association with polyaniline, glutaraldehyde, and Bacillus subtilis were used for the elaboration of a new amperometric microbial biosensor. The configuration was elaborated and optimized for sensitive and selective detection of paracetamol in drug samples. The use of the composite structure provided an increase of the sensitivity towards paracetamol detection, but also enhanced the stability and conductivity of the sensing platform. Furthermore, the microorganism has proven to be economically convenient for routine analysis of the selected drug (Bayram and Akyilmaz 2016).

Graphene Graphene contains 1 atom layer with 2D structure and is an allotrope of sp2hybridized carbon like carbon nanotubes and fullerenes. Its first isolation was performed by exfoliation of graphite, then other synthesis methods have been developed, such as: micromechanical exfoliation, chemical vapor deposition, epitaxial growth, colloidal suspension, unzipping of carbon nanotubes, and reduction of graphene oxide, each of them with particular advantages and limitations. Due to its outstanding electrical, optical, thermal and mechanical properties, graphene has quickly become very popular in many areas including nanotechnology manly with applications in (bio)sensors design. The noncovalent and covalent functionalization of graphene increased its potential for biological applications. The electrochemical sensing of a wide range of target compounds has been investigated by using graphene, starting from small and simple molecules, to complex biotargets such as proteins, DNA, and also cells (Biju 2014; Atta et al. 2015; Cernat et al. 2016; Rowley-Neale et al. 2018). A miniaturized platform based on a pencil graphite electrode modified with graphene oxide quantum dots (GOQDs)/carboxylate MWCNTs hybrid nanomaterial was used for the development of a cell-based biosensor employing the human hepatoma cells (HepG2) as the biological recognition element. GOQDs exhibit excellent features such as edge and quantum confinement effects, chemical inertness, excellent biocompatibility, and good hydrophilicity. Carboxylate MWCNTs were uniformly wrapped with GOQDs having a diameter around 15 nm due to π-π interaction. The electrochemical behavior of HepG2 cell suspension recorded with this platform revealed three well-defined anodic peaks corresponding to guanine/ xanthine, adenine, and hypoxanthine. The cytotoxicity of six pollutants such as Cd, Hg, Pb, 2,4-dinitrophenol, 2,4,6-trichlorophenol, and pentachlorophenol was also assessed in a 96-well plate by measuring the changes of the three electrochemical

25

New Materials for the Construction of Electrochemical Cell-Based Biosensors

613

signals of purine bases (Zhu et al. 2017). Graphene exhibits a large interfacing area providing a stable interface for microbial and mammalian cells to detect intracellular and extracellular phenomena. The electrostatic deposition of Bacillus cereus (a Gram-positive bacterium with highly negatively charged surface) on positively charged graphene-amine led to the increase of graphene conductivity. A single bacterium resolution interfacial device was thus developed based on chemically modified graphene proving the nanomaterial potential in building valuable tools for biomedicine (Nguyen and Berry 2012; Mohanty and Berry 2008). An impedance cell device was elaborated based on positively charged poly-L-lysine electrostatically assembled on negatively charged graphene oxide, biocompatible film which favored the adhesion of leukemia K562 cells and also preserved the cells activity. The oxygen-containing groups from graphene oxide surface and the amino groups from poly-L-lysine are suitable for building of nano/bio interface through adhesion (Zhang et al. 2013). The bacteria-mediated bioimprinting method was used for the selective detection of marine pathogen sulfate-reducing bacteria (SRB). The fabrication of the bioimprinted film consisted in a few steps: reduced graphene sheets/chitosan hybrid film was electrodeposited on indium tin oxide electrode, then the bacteria solution was incubated followed by another layer of chitosan. The bioimprinted film showed good selectivity towards SRB detection based on its size and shape (Qi et al. 2013). Gluconobacter oxydans was immobilized in glassy carbon paste modified with graphene oxide and graphene-platinum NPs leading to two microbial biosensors. The GOx from the bacteria membrane converted glucose to glucuronic acid by using Fe(CN)63
/4- as mediator; therefore, these biosensors were applied for glucose determination using cyclic voltammetry (Aslan and Anik 2016).

Other Carbon-Based Materials A yeast strain, Saccharomyces cerevisiae, was immobilized by adsorption coupled with a double system of mediators: menadione-ferricyanide onto a chitosan hydrogel polymer film with boron-doped nanocrystalline diamond nanoparticles. The menadione could permeate inside the yeast cell and it is reduced by NADPH, an intracellular enzyme system, to menadiol, that is further oxidized by ferrocyanide after diffusion in the extracellular environment. Boron-doped nanoparticles have excellent conductivity, stability, and most important, biocompatibility enabling biological applications. Chitosan is an amino-polysaccharide that generates an electrochemically inactive film which can trap the negative charged mediator, facilitating the electron transfer rate. The biosensor was used to assess the biotoxicity of water in terms of heavy metals (Cu2+, Cd2+, Ni2+, and Pb2+) and three phenols (3,5-dichlorophenol, 4-chlorophenol and phenol) with promising results. This double-mediator based whole cell electrochemical biosensor proves to be suitable for the assessment of the acute toxicity of wastewater and has promising characteristics for portability and online detection (Gao et al. 2017).

614

A. Cernat et al.

Metallic-Based Nanomaterial as Platform for Cell Biosensors Metallic nanoparticles have attracted increased attention in biosensing area due to their characteristic advantageous properties. AuNPs are maybe the most used NPs in biological and sensing applications. Their excellent conductivity, good biocompatibility, and large surface area lead to higher loading of biological elements on the sensing platform. There are two major synthesis procedures of AuNPs composites: the in situ technique involving the formation and direct growth of NPs onto a certain surface/material and the ex situ method consisting in a preliminary synthesis of NPs followed by a subsequent attachment to a certain surface/material (Khalil et al. 2016). AuNPs were embedded into chitosan polymer by in situ reduction, leading to a rough and porous nanocomposite gel which was a suitable medium for cell immobilization and redox probe diffusion. This bio-inspired gel combined the advantages of both polymer and AuNPs and enabled an increased capacity for cell immobilization due to its spatial structure and good biocompatibility for K562 leukemia cells which were used as model. An impedimetric cell-based biosensor was, thus, developed not only for cells detection, but also for their proliferation and apoptosis (Ding et al. 2007). Another fast, low cost, highly sensitive, and selective recognition of drug-sensitive (K562/B.W.) and drug-resistant leukemia K562 cells (K562/ADM) was achieved by using a relatively hydrophilic interface of AuNPs/ polylactide nanofibers cast on indium tin oxide electrode and tested by EIS (Wu et al. 2011). The same two types of leukemia K562 cells were distinguished using a platform based on indium tin oxide electrode modified with TiO2 NPs (25 nm diameter). This platform presented a rougher and more hydrophilic surface, which facilitated the cells adhesion (Wu et al. 2010). An inhibitor microbe-biosensor was elaborated based on silk derived carbon fiber mat modified with Au@Pt urchilike NPs on which E. coli was attached. The amino group from the silk allowed an electrostatic self-assembly technique of the NPs. This hybrid nanomaterial provided a biocompatible microenvironment for E. coli with high conductivity and electrocatalytic activity. The presence of bacteria at the electrode surface was indicated by a pair of redox waves in cyclic voltammogram ascribed to the endogenous redox mediator. Direct oxidation of glucose was observed at a low potential (
570 mV vs Ag/AgCl) leading to a sensitive amperometric detection of glucose with a limit of detection of 0.3 μM. When exposed to an organophosphorus pesticide, fenamiphos, as a toxic substrate, the glucose current decreased due to the inhibition of E. coli activity. Therefore, this biosensor can also be applied for toxic compound detection (Deng et al. 2010). An eggshell membrane was covalently modified with mercaptopropionic acid, then AuNPs were attached by Au-S bonds. Subsequently, another layer of mercaptopropionic acid was deposited in order to covalently immobilize a layer of bacterial cells Methylobacterium organophilium. This M. organophilium/AuNPs immobilized eggshell membrane was positioned on a commercial O2 sensor leading to a microbial biosensor tested for methanol determination. The exposure to methanol decreased the dissolved O2 concentration because O2 was consumed during bacterial oxidation of methanol. The biosensor presented a linear response to methanol from 0.05 to 2.5 mM

25

New Materials for the Construction of Electrochemical Cell-Based Biosensors

615

and a limit of detection of 0.047 mM. A phosphate buffer of pH 7 and a temperature of 20–25  C were set as optimal parameters (Wen et al. 2014). Another example of a mediated microbial sensor was designed using a platform based on gold electrode modified with AuNPs by electrodeposition and selfassembled monolayers of 6-(ferrocenyl) hexanethiol. A bio-active layer of Gluconobacter oxydans cells was dropped onto the modified electrode being covered with a dialysis membrane. The glucose dehydrogenase from the bacterium membrane catalyzed the glucose oxidation by ferrocene (Yildirim et al. 2015). A mixture of synthesized AuNPs, bovine serum albumin and Pseudomonas sp. (GSN23) was drop-casted on gold interdigitated microelectrodes using the cross-linking method with glutaraldehyde for bacteria immobilization. This bacterium isolated from oil contaminated soil can be employed for phenol detection because it can use phenol as source of energy. This biosensor was applied for the sensitive detection of phenol in spiked river samples with a limit of detection of 2 μM (Kolahchi et al. 2018). Nanoporous gold (NPG) was used as metallic nanomaterial for the development of a whole cell based electrochemical biosensor for highly sensitive detection of catechol. The carE gene of Sphingobium yanoikuyae XLDN2–5 encoding catechol 2,3-dioxygenase (C23O) is a key enzyme in the biodegradation of aromatic compounds. This gene was cloned and over-expressed in E. coli BL21 that proved to have higher catalytic activity towards catechol electrochemical detection, the whole cells presence providing a better stability for C23O when compared with simple enzyme. NPG has unique structural properties and was successfully selected as a support for the immobilization of the recombinant E. coli BL21 over-expressed C23O and revealed a good stability (Liu et al. 2019). A different approach is represented by the selection of a specific cell that could enhance the signal of the target molecule. Shewanella oneidensis bacteria are able to achieve bidirectional extracellular electron transfer controlled by the redox potential shift of flavin when it interacts with outer-membrane cytochromes. These properties transformed this system in the perfect candidate for enhancing riboflavin signal. The bacteria were not immobilized on the carbon cloth working electrode, but it was inoculated directly in the electrochemical cell enhancing the detection of riboflavin. In contrast to the conventional designs that involve the use of nanomaterials, noble metal catalysts and protocols with multiple procedures, the use of a microbial biocatalyst reduces the cost, simplifies the protocol, and increases the sensitivity of the method, while using only a carbon based electrode (Yu et al. 2017). Also the bacteria could be embedded in a membrane sandwich in order to ensure the adherence to the working electrode and to guarantee the viability of the cells. For instance, Pseudomonas sp. strain ASA86 was immobilized in a porous cellulose nitrate membrane and covered with another cellulose membrane. Strain ASA86 had been found to be a suitable bioelement for the determination of trichloroethylene, a common pollutant in soil and groundwater. The target was degraded by the bacteria from the biofilm and the concentration of the released chloride ions was determined with a chloride ion electrode. Moreover, the stability of the biosensor was over 5 days at 30  C (Chee 2016).

616

A. Cernat et al.

Molecular imprinting using whole bacteria as a stamp represents a new approach for the rapid and highly sensitive detection of different bacterial strains. After the removal of the stamp, the polymer platform with cavities allows the detection of the target with low detection limit. The reaction mixture containing 2-hydroxyethyl methacrylate as monomer, ethyleneglycol dimethacrylate as cross-linker, N-methacryloyl L-histidine methylester-Cu (II) complex as preorganized monomer was employed for the synthesis of a molecular imprinting polymer. The polymerization was successfully carried out in the presence of E. coli cells on a gold electrode under UV light. The interaction of the bacterial stamp with the polymeric platform was based on affinity interactions and steric matching. The results indicated that E. coli was detected in the presence of other bacterial strains like Staphylococcus aureus, Bacillus subtilis, and Salmonella paratyphi (Idil et al. 2017).

Wearable Sensing Devices Future Trends Skin is the largest organ of the human body, offering a high diagnostic interface linked to vital biological signals from the dermis, epidermis, blood vessels, or either inner organs (Liu et al. 2017c). For this reason, skin can generate and transmit biological indicators that offer significant health metrics of an individual. As a consequence, the trend of developing soft, flexible, and stretchable electronic devices received more attention in last few years, especially for the continuous health monitoring (Son and Bao 2018). So-called “lab-on-skin” devices need to address few requirements and challenges such as proper thickness and elastic modulus, which resemble those of the human skin. These wearable devices need to cover the human body and to diminish the irritation of the skin and human cavities, which is usually caused by the conventional, rigid electronics (Son and Bao 2018). Equally important, these devices need to be able to deliver an accurate, non- (or minimum) invasive, long-term, and constant health status of the wearer. The development of wearables has, thus, ushered a new era of personalized diagnosis devices for various healthcare, environment, entertainment, and sports or fitness applications. The primary fabrication technologies used in designing of wearable devices are represented by the lithographic and printing methods. The lithographic methodologies (e.g., photolithography, e-beam, and ion-beam lithography) have the advantage of reproducible fabrication which led to high performance devices. However, this method comes with few drawback aspects linked to high financial cost, the need for clean-room facilities, expensive chemicals, and lengthy processes. In majority of the situations, the cost of the device fabrication dictates the researchers’ preferences. For this reason, inexpensive printing techniques are often employed since have significantly lower costs while assuring large scale manufacturing process with desired precision and accuracy. Even though the printing technology was implemented centuries ago, it has only lately captured the imagination and interest of scientists for developing high-end sensing devices (Kim et al. 2017).

25

New Materials for the Construction of Electrochemical Cell-Based Biosensors

617

The typical printing technologies that are used in developing (bio)sensors (including electrochemical ones) are generally divided into template and non-template based methodologies. Further, the template-based printing techniques can be divided into screen printing, gravure printing, flexography, and imprinting methodologies (Kim et al. 2017; Mattana and Briand 2016). In the case of screen printing method, it is required printing the conductive inks at a low pressure. This step can be performed using a screen stencil which is characterized by a designed pattern and uniform thickness. A squeegee blade made of metal or rubber materials is used in order to squeeze the conductive ink through the patterned stencil onto the substrate. The inks utilized in screen printing technique have a high viscosity, but when a force is applied through the screen stencil by the squeegee blade, the inks undergo thinning which facilitates penetration through the stencil. By doing so, it is achieved the final conductive pattern on the desirable substrate. In addition to screen printing, gravure and flexography are other types of template-based printing technologies that are frequently utilized. These two techniques involve transferring the ink to the substrate from an engraved (gravure) or raised (flexography) patterns on a roll. This approach is extremely suitable for high throughput manufacture of flexible devices which require a large sensing area (Mattana and Briand 2016). On the other hand, nontemplate printing techniques include inkjet and 3D printing. These types of methodologies rely on dispensing the ink in a controlled manner on the desirable substrate, without the use of a template. The recent progress in the field of engineering and advanced materials has led to complex inkjet and 3D printing solutions, that comprise advanced dispensing technologies linked to pneumatic, piezoelectric, aerosol, electrohydrodynamic, and thermal approaches (Kim et al. 2017; Liu et al. 2018b). Among the subfields of wearable electronics and sensors, printed wearable systems are of significant importance for opening new avenues for body-integrated electronics which can be applied in various fields, and that were earlier impossible to achieve (Kim et al. 2017). As a matter of fact, screenprinted devices can be easily incorporated on a variety of substrates characterized by different shapes and sizes. Based on the advantages, screen printing has been used to develop inexpensive wearable sensors on platforms like PET (Nyein et al. 2016), rings (Sempionatto et al. 2017), gloves (Ciui et al. 2018; Mishra et al. 2017), temporary tattoo (Kim et al. 2016), textiles (Castano and Flatau 2014), or mouthguards (Ciui et al. 2019) for noninvasively detecting biomarkers in various biofluids (such as saliva, tears, or sweat), or from environment. For the realization of these types of wearable applications, the printed electrochemical sensors are fabricated to match the non-planarity and mechanical properties of the human body (Kim et al. 2017). Innovative platforms consist often in the combination of conductive inks, novel nanomaterials, polymers, or diverse composites with biological elements (including living cells) (Kim et al. 2017). Because the term of “wearable sensors” is linked to skin/cavities-mounted devices, the physical properties (such as thicknesses, mechanical properties, thermal masses) of these devices are usually required to be close to those of the human skin, and implicit to enable compliant and robust contact upon attaching on the human body (Bandodkar et al. 2016a; Bandodkar and Wang 2014). Therefore, wearable

618

A. Cernat et al.

sensing devices must be capable of not only bending to conform to the structure of the skin, but also stretching to overcome the strains resulted from the natural body motion (Liu et al. 2017c; Kim et al. 2017). On the other hand, such anatomically compliant printed devices are required to preserve their conductive properties under severe range of ambient conditions (such as temperature, humidity) and are preferable to be made of harmless, biocompatible materials. Meeting all these conditions and requirements is certainly challenging and necessitates innovations in materials science for realizing high-performance wearable sensing platforms. In the continuation of this subchapter, it will be discussed the key advances made in the development of new ink materials or different formulations that meet the demands of high performance bio-integrated wearable devices. Since the applications of these sensors depend on different challenges: flexibility, stretchability, mechanical/electrical properties, self-healing capabilities, it will also be offered examples of materials/combination of materials used to overcome each of these challenges. It will also be briefly discussed representative studies of wearable systems incorporating cell as biocomponents on the sensing platforms. Generally, the printing inks are composed of fillers, binders, additives, and solvents. The selection and the quantities of these components depend on the category of printing technique which is employed. The fillers, which can be metallic, ceramic, organic- based materials (or a combination of these), constitute the active element of the ink. With fast advances in nanotechnology, researchers have been able to mix these fillers with nanomaterials (such as nanosheets, nanowires, nanoparticles) in order to increase the conductivity of inks. The binder component is a polymeric material that assures the regular dispersion of the fillers into the ink matrix and guarantees the binding of the printing layers onto the desirable substrates. In addition to that, the binders are holding the ink components together when the solvent is evaporating.

New Materials for Flexible Wearable Sensors Upon placing the wearable device on the skin, the sensor is conforming the curvature of the body. As a result, a strain (which is a function of the radius of bending) is formed within the printed materials, but also within the substrate. Thus, the device should be flexed enough to resist against mechanical deformation (Wang et al. 2016). For this reason, it is important that the materials are carefully synthesized for developing sensors that own the desired degree of flexibility without affect their performance. A rich variability of material formulations has been established over the past years for designing printed wearable sensors that endure high flexural stress with insignificant negative impact on their analytical performance (Son and Bao 2018). These materials comprise of a wide range of materials, including carbon nanotubes, graphene, noble metal-based nanoparticles, quantum dots, or a combination of these materials. The binders and adhesion promoters/surfactants represent also key components that have a vital role in deciding the flexible and homogeneity nature of the inks (Liu et al. 2017c; Son and Bao 2018; Kim et al. 2017).

25

New Materials for the Construction of Electrochemical Cell-Based Biosensors

619

Polymers-based binders/additives represent promising solutions due to low mechanical stiffness, flexibility, and softness texture offered to the wearable prototypes. Indeed, the advanced strategies in device design coupled with polymeric materials selection have led to many electronics capable of digitizing biological signals for healthcare monitoring (Liu et al. 2017c; Wang et al. 2016). However, the addition of nonconductive binders/additives comes to a cost: the higher their amount is, the more affected the conductivity of the final sensing is. To overcome these issues, several researchers started utilizing conductive polymer-based binders (such as PEDOT: PSS, polyaniline, polypyrrole, polyacetylene, and poly indole). In addition, these formulations were also mixed with carbon nanotubes, graphene, gold, or silver. These conductive binders can improve the suspension of the carbonaceous nanomaterials in the inks, but also offer flexibility to the ink matrix (Kim et al. 2017). In case it is required a low viscosity material, few solutions relied on the usage of surfactants (sodium dodecyl sulfate or sodium dodecylbenzene sulfonate) or by generating functional groups on the carbon-based (carbon nanotubes/graphene) surface via acidic treatment. Functional polar groups augment the suspension of these carbon-based nanomaterials in polar solvents, but also led to a stable suspension due to electrostatic repulsion between similarly charged functional groups (Kim et al. 2017). An innovative flexile wearable biosensor consisted of graphene-based interdigitated electrodes printed onto very flexible silk thin-film was attached on tooth enamel (Mannoor et al. 2012). Although the biocomponent was not incorporated on the transducer (such as in the classical cell-based sensors), the biotransferrable device offered sensitive, specific, and label-free detection of Helicobacter pylori in saliva, which is worth mentioning in the present book chapter. The capture of bacteria by antimicrobial peptides on the sensor surface led in conductivity change of the graphene film, being monitored with an inductively coupled radio frequency readout unit (Mannoor et al. 2012) (Fig. 5).

Fig. 5 Biotransferrable graphene wireless nanosensor. (a) Graphene is printed on the bioresorbable silk. (b) Transfer of the innovative material platform onto the surface of a tooth. (c) Schematic illustrating wireless readout. (d) Binding of pathogenic bacteria by peptides self-assembled on the graphene nanotransducer (Reprinted with permission from (Mannoor et al. 2012). Copyright (2018) Springer Nature)

620

A. Cernat et al.

New Materials for Stretchable Wearable Sensors Another challenge in designing robust wearable sensors consists in employing materials able to stretch, since the human body is also soft and stretchable in nature. Usually, materials that are significantly thin are able to withstand high bending strains (ε ¼ d/2r). However, just because a material can bend (is flexible), this does not guarantee its inherent stretchability (Heikenfeld et al. 2018; Bandodkar et al. 2016a). This failure may occur due to the weak adhesion between the thin printed film and substrate, or because of fracturing, slipping, or delamination of the thin film (Heikenfeld et al. 2018). Knowing the importance of sensor stretchability, the trend now is also focused on improving the stretchability of the materials. Compared with fabrication of flexible materials, the synthesis of stretchable materials represents a more challenging goal, because the device needs to experience higher strain levels. Such in the case of flexible devices, the selection of binders and surfactants is of significant importance for creating stretchable ones. Incorporating highly stretchable elastomeric binders (such as silicone, fluorine rubber, poly-urethane isoprene block co-polymers) represent few examples of binders that researchers have used to develop stretchable materials for sensing platforms. Besides stretchable binders, it can also improve the stretchability by including a suitable surfactant to avoid the issue of delamination of the printed film upon stretching (Kim et al. 2017; Bandodkar et al. 2016a). Importantly, it was also demonstrated that although the resistance of carbon nanotubes-printed electrochemical device (that uses polyurethane as a binder) increased after repeated stretching cycles, its electrochemical properties remained stable till 100% strain (Bandodkar et al. 2016a). X. Liu et al. demonstrated the design of a set of living materials and devices based on stretchable, robust, and biocompatible hydrogel–elastomer hybrids that host different genetically engineered bacterial cells (such as Escherichia coli) (Liu et al. 2017c). While the hydrogel had the role to supply the water and nutrients, the elastomer which is air-permeable provided long-term viability and functionality of the encapsulated cells. The innovative platform facilitated the communication between the bacterial strains and the environment via diffusion of molecules in the hydrogel. Additionally, the high stretchability and robustness of the hydrogel– elastomer combination prevented the leakage of cells from the living materials and the proposed device, even under large deformations. In this work it was proved the efficiency and application of stretchable living sensors that were responsive to multiple chemicals in a variety of wearable platforms, including skin patches and gloves-based sensors (Liu et al. 2017c) (Fig. 6).

New Materials for Self-healing Wearable Sensors Advancement in materials science has also led to the progress of self-healing systems for wearable applications, being inspired by the biological systems. Selfhealing is a remarkable characteristic which allows the sensing systems to recover after mechanical deformations, and implicit extending their lifespan (Son and Bao

25

New Materials for the Construction of Electrochemical Cell-Based Biosensors

621

Fig. 6 Living wearable devices. (a) Schematic illustration of a living patch. Engineered bacteria inside can detect signaling molecules (b–d). (e) Illustration of a glove with chemical detectors robustly integrated at the fingertips. Different cell strains were encapsulated in the chambers. (f–h) When the living glove was used to grab a wet cotton ball containing the inducers, fluorescence was shown in the cognate sensors. (Reprinted with permission from (Liu et al. 2017c). Copyright (2018) National Academy of Science)

2018; Tee et al. 2012). Because of the prolonged and harsh bending or starching phenomena, wearable systems often face mechanical deformations that can result in device’s damage and lead to device failure (Liu et al. 2017c). As a consequence, researchers developed new repair strategies based on self-healing materials. These innovative materials can be generally divided into three groups: vascular, capsule, and intrinsic-based materials. For example, vascular-based self-healing materials are linked to the incorporation of channels filled with self-healing material. With a

622

A. Cernat et al.

different mechanism, capsule-based materials rely on hollow microcapsules loaded with self-healing agents (such as organic solvent). In this case, when the printed sensor is mechanically damaged, the microcapsules rupture and release the encapsulated healing agent. The healing material has the role to dissolve the binder, which further causes the rearrangement of the conductive layers of the sensing platform (Patrick et al. 2014; Bandodkar et al. 2015).

Conclusions The requirements of nowadays analytical analysis such as rapidity, low cost, high specificity and sensitivity, stability, and long shelf-life are connected with the development of cell-based sensing device. The engineering of cells with a high resistance and genes with faster and stronger expression, the multidetection based on array sensors with different targets, and continuous monitoring of the physiological/biochemical processes are only a few of the possible improvements in this narrow field. The main advantages of whole cell biosensors are represented by the fast and real time detection from harsh environments where enzymatic biosensors could fail (extremely acidic or basic medium, complex matrices, extremes temperatures, or toxic environments). However, in the presence of toxic compounds and organic pollutants, the choice of the cells must be carefully done to microbes or bacteria resistant to them. The lifetime of whole cell biosensors is also limited since the cell could diffuse or even became inactive under harsh conditions. With several analytical devices already on the market, especially for toxicology and environmental quality allowing the on-line and continuing monitoring of various compounds, the future of whole cell biosensors is assured even that some disadvantages must be overcome like the response time and attachment of the cells on the electrodes. Early screening and drug testing represents future areas of interest besides toxicological studies; therefore, the need of engineered cells with high affinity for newly synthesized drugs could be necessary. With the advances in material sciences, the easy integration of whole cells on printable devices and self-healing materials will facilitate the spreading of those biosensors in other field of interest besides monitoring pollutants and biomedical analysis. Acknowledgments This work was supported by the Romanian National Authority for Scientific Research and Innovation, CNCS/CCCDI-UEFISCDI, project number PN-III-P1-1.2-PCCDI20170407 (INTELMAT).

References Aslan S, Anik U (2016) Microbial glucose biosensors based on glassy carbon paste electrodes modified with Gluconobacter Oxydans and graphene oxide or graphene-platinum hybrid nanoparticles. Microchim Acta 183:73–81 Atta NF, Galal A, El-Ads EH (2015) Graphene -a platform for sensor and biosensor applications. In: Rinken T (ed) Biosensors – micro and nanoscale applications. InTechOpen, London

25

New Materials for the Construction of Electrochemical Cell-Based Biosensors

623

Bandodkar AJ, Wang J (2014) Non-invasive wearable electrochemical sensors: a review. Trends Biotechnol 32:363–371 Bandodkar AJ, Nuñez-Flores R, Jia W et al (2015) All-printed stretchable electrochemical devices. Adv Mater 27:3060–3065 Bandodkar AJ, Jeerapan I, Wang J (2016a) Wearable chemical sensors: present challenges and future prospects. ACS Sens 1:464–482 Bandodkar AJ, Jeerapan I, You JM et al (2016b) Highly stretchable fully-printed CNT-based electrochemical sensors and biofuel cells: combining intrinsic and design-induced stretchability. Nano Lett 16:721–727 Banerjee P, Bhunia AK (2009) Mammalian cell-based biosensors for pathogens and toxins. Trends Biotechnol 27:179–188 Bayram E, Akyilmaz E (2016) Chemical development of a new microbial biosensor based on conductive polymer/multiwalled carbon nanotube and its application to paracetamol determination. Sensors Actuators B Chem 233:409–418 Belkin S (2003) Microbial whole-cell sensing systems of environmental pollutants. Curr Opin Microbiol 6:206–212 Biju V (2014) Chemical modifications and bioconjugate reactions of nanomaterials for sensing, imaging, drug delivery and therapy. Chem Soc Rev 43:744–764 Castano LM, Flatau AB (2014) Smart fabric sensors and e-textile technologies: a review. Smart Mater Struct 23:053001 Cernat A, Tertiş M, Fritea L et al (2016) Graphene in sensors design. In: Tiwari A, Syväjärvi M (eds) Advanced 2D materials. Wiley, New York, pp 387–431 Chee GJ (2016) A novel whole-cell biosensor for the determination of trichloroethylene. Sensors Actuators B Chem 237:836–840 Ciui B, Tertiş M, Cernat A et al (2018) Finger-based printed sensors integrated on a glove for on-site screening of Pseudomonas aeruginosa virulence factors. Anal Chem 90:7761–7768 Ciui B, Tertis M, Feurdean CN et al (2019) Cavitas electrochemical sensor toward detection of Nepsilon (carboxymethyl)lysine in oral cavity. Sensors Actuators B Chem 281:399–407 Clayton J et al (2012) Graphene-based wireless bacteria detection on tooth enamel. Nat Commun 3:763 Dai B, Wang L, Wang Y et al (2018) Single-cell nanometric coating towards whole-cell-based biodevices and biosensors. Chemistry Select 3:7208–7221 Deng L, Guo S, Zhou M et al (2010) A silk derived carbon fiber mat modified with au@Pt urchilike nanoparticles: a new platform as electrochemical microbial biosensor. Biosens Bioelectron 25:2189–2193 Ding L, Hao C, Xue Y et al (2007) A bio-inspired support of gold nanoparticles - chitosan nanocomposites gel for immobilization and electrochemical study of K562 leukemia cells. Biomacromolecules 8:1341–1346 Fakhrullin RF, Zamaleeva AI, Minullina RT et al (2012) Cyborg cells: functionalisation of living cells with polymers and nanomaterials. Chem Soc Rev 41:4189–4206 Farooq U, Yang Q, Wajid M et al (2018) Biosensors and bioelectronics bacterial biosensing: recent advances in phage-based bioassays and biosensors. Biosens Bioelectron 118:204–216 Freitas M, Nouws HPA, Delerue-matos C (2018) Electrochemical biosensing in cancer diagnostics and follow-up. Electronalysis 8:1584–1603 Gao G, Fang D, Yu Y et al (2017) A double-mediator based whole cell electrochemical biosensor for acute biotoxicity assessment of wastewater. Talanta 167:208–216 Han L, Zhao Y, Cui S et al (2018) Redesigning of microbial cell surface and its application to whole-cell biocatalysis and biosensors. Appl Biochem Biotechnol 185:396–418 Heikenfeld J, Jajack A, Rogers J et al (2018) Wearable sensors: modalities, challenges, and prospects. Lab Chip 18:217–248 Hnaien M, Bourigua S, Bessueille F et al (2011) Impedimetric microbial biosensor based on single wall carbon nanotube modified microelectrodes for trichloroethylene detection. Electrochim Acta 56:10353–10358 Idil N, Hedström M, Denizli A et al (2017) Whole cell based microcontact imprinted capacitive biosensor for the detection of Escherichia coli. Biosens Bioelectron 87:807–815

624

A. Cernat et al.

Jiang D, Liu Y, Jiang H et al (2018) A novel screen-printed mast cell-based electrochemical sensor for detecting spoilage bacterial quorum signaling molecules (N-acyl-homoserine-lactones) in freshwater fish. Biosens Bioelectron 102:396–402 Kempaiah R, Chung A, Maheshwari V (2011a) Graphene as cellular interface: electromechanical coupling with cells. ACS Nano 5:6025–6031 Kempaiah R, Salgado S, Chung WL et al (2011b) Graphene as membrane for encapsulation of yeast cells: protective and electrically conducting. Chem Commun 47:11480–11482 Khalil I, Julkapli NM, Yehye WA et al (2016) Graphene-gold nanoparticles hybrid-synthesis, functionalization, and application in a electrochemical and surface-enhanced raman scattering biosensor. Materials (Basel) 9:406–444 Kim JI, Imani S et al (2016) Noninvasive alcohol monitoring using a wearable tattoo-based iontophoretic-biosensing system. ACS Sensors 1:1011–1019 Kim J, Kumar R, Bandodkar AJ et al (2017) Advanced materials for printed wearable electrochemical devices: a review. Adv Electron Mater 3:1–15 Kintzios S, Banerjee P (2015) Mammalian cell-based sensors for high throughput screening for detecting chemical residues, pathogens, and toxins in food. In: Bhunia AK, Kim MS, Taitt CR (eds) High throughput screening for food safety assessment. Woodhead Publishing, Amsterdam, pp 123–146 Kolahchi N, Braiek M, Ebrahimipour G et al (2018) Direct detection of phenol using a new bacterial strain-based conductometric biosensor. J Environ Chem Eng 6:478–484 Li L, Liang B, Shi J et al (2012) A selective and sensitive d-xylose electrochemical biosensor based on xylose dehydrogenase displayed on the surface of bacteria and multi-walled carbon nanotubes modified electrode. Biosens Bioelectron 33:100–105 Liang B et al (2013a) Microbial surface display of glucose dehydrogenase for amperometric glucose biosensor. Biosens Bioelectron 45:19–24 Liang B, Lang Q, Tang X et al (2013b) Simultaneously improving stability and specificity of cell surface displayed glucose dehydrogenase mutants to construct whole-cell biocatalyst for glucose biosensor application. Bioresour Technol 147:492–498 Liu Z, Ma H, Sun H et al (2017a) Nanoporous gold-based microbial biosensor for direct determination of sulfide. Biosens Bioelectron 98:29–35 Liu Y, Pharr M, Salvatore GA (2017b) Lab-on-skin: a review of flexible and stretchable electronics for wearable health monitoring. ACS Nano 11:9614–9635 Liu X, Tang T-C, Tham E et al (2017c) Stretchable living materials and devices with hydrogel– elastomer hybrids hosting programmed cells. Proc Natl Acad Sci 114:2200–2205 Liu K, Li Y, Zhang H et al (2018a) Synthesis of the polypyrrole encapsulated copper nanowires with excellent oxidation resistance and temporal stability. Appl Surf Sci 439:226–231 Liu X, Yuk H, Lin S et al (2018b) 3D printing of living responsive materials and devices. Adv Mater 30:1–9 Liu Z, Zhang Y, Bian C et al (2019) Highly sensitive microbial biosensor based on recombinant Escherichia coli overexpressing catechol 2 , 3-dioxygenase for reliable detection of catechol. Biosens Bioelectron 126:51–58 Majdinasab M, Hayat A, Marty JL (2018) Aptamer-based assays and aptasensors for detection of pathogenic bacteria in food samples. TrAC-Trends Anal Chem 107:60–77 Malekzad H, Zangabad PS, Mirshekari H et al (2016) Noble metal nanoparticles in biosensors: recent studies and applications. Nanotechnol Rev 6:301–329 Mannoor M, Tao H, Clayton J et al (2012) Graphene-based wireless bacteria detection on tooth enamel. Nat Commun 3:763 Mattana G, Briand D (2016) Recent advances in printed sensors on foil. Mater Today 19:88–99 Mavrikou S, Flampouri E, Iconomou D et al (2017) Development of a cellular biosensor for the detection of aflatoxin B1, based on the interaction of membrane engineered Vero cells with antiAFB1antibodies on the surface of gold nanoparticle screen printed electrodes. Food Control 73:64–70

25

New Materials for the Construction of Electrochemical Cell-Based Biosensors

625

Mishra RK, Hubble LJ, Martin A et al (2017) Wearable flexible and stretchable glove biosensor for on-site detection of organophosphorus chemical threats. ACS Sens 2:553–561 Mohanty N, Berry V (2008) Graphene-based single-bacterium resolution biodevice and DNA transistor: interfacing graphene derivatives with nanoscale and microscale biocomponents. Nano Lett 8:4469–4476 Nguyen P, Berry V (2012) Graphene interfaced with biological cells: opportunities and challenges. J Phys Chem Lett 3:1024–1029 Nyein HYY, Gao W, Shahpar Z et al (2016) A wearable electrochemical platform for noninvasive simultaneous monitoring of Ca2+ and pH. ACS Nano 10:7216–7224 Patrick JF, Hart KR, Krull BP et al (2014) Continuous self-healing life cycle in vascularized structural composites. Adv Mater 26:4302–4308 Qi P, Wan Y, Zhang D (2013) Impedimetric biosensor based on cell-mediated bioimprinted films for bacterial detection. Biosens Bioelectron 39:282–288 Rowley-Neale SJ, Randviir EP, Abo Dena AS et al (2018) An overview of recent applications of reduced graphene oxide as a basis of electroanalytical sensing platforms. Appl Mater Today 10:218–226 Sempionatto JR, Mishra RK, Martín A et al (2017) Wearable ring-based sensing platform for detecting chemical threats. ACS Sens 2:1531–1538 Shao X, Jiang M, Yu Z et al (2009) Surface display of heterologous proteins in Bacillus thuringiensis using a peptidoglycan hydrolase anchor. Microb Cell Factories 8:48–65 Son D, Bao Z (2018) Nanomaterials in skin-inspired electronics: toward soft and robust skin-like electronic nanosystems. ACS Nano 12:11731–11739 Tee BCK, Wang C, Allen R et al (2012) An electrically and mechanically self-healing composite with pressure- and flexion-sensitive properties for electronic skin applications. Nat Nanotechnol 7:825–832 Wang H, Lang Q, Li L et al (2013) Yeast surface displaying glucose oxidase as whole-cell biocatalyst: construction, characterization, and its electrochemical glucose sensing application. Anal Chem 85:6107–6112 Wang SQ, Chinnasamy T, Lifson MA et al (2016) Flexible substrate-based devices for point-of-care diagnostics. Trends Biotechnol 34:909–921 Wen G, Wen X, Shuang S et al (2014) Whole-cell biosensor for determination of methanol. Sensors Actuators B Chem 201:586–591 Wu X, Jiang H, Zhou Y et al (2010) Selective determination of drug resistant cancer cells on indium tin oxide electrode modified with nano titanium dioxide. Electrochem Commun 12:962–965 Wu X, Jiang H, Zheng J et al (2011) Highly sensitive recognition of cancer cells by electrochemical biosensor based on the interface of gold nanoparticles/polylactide nanocomposites. J Electroanal Chem 656:174–178 Xia H, Sakai K, Kitazumi Y et al (2018) Enzyme and microbial technology carbon-nanotube-caged microbial electrodes for bioelectrocatalysis. Enzym Microb Technol 117:41–44 Yang Y, Yang X, Yang Y et al (2018) Aptamer-functionalized carbon nanomaterials electrochemical sensors for detecting cancer relevant biomolecules. Carbon 129:380–395 Ye Y, Guo H, Sun X (2018) Recent progress on cell-based biosensors for analysis of food safety and quality control. Biosens Bioelectron 26:389–404 Yildirim N, Demirkol DO, Timur S (2015) Modified gold surfaces with gold nanoparticles and 6-(ferrocenyl)hexanethiol: design of a mediated microbial sensor. Electroanalysis 27:52–57 Yu YY, Wang JX, Si RW et al (2017) Sensitive amperometric detection of riboflavin with a wholecell electrochemical sensor. Anal Chim Acta 985:148–154 Zamaleeva AI, Sharipova IR, Shamagsumova RV et al (2011) A whole-cell amperometric herbicide biosensor based on magnetically functionalised microalgae and screen-printed electrodes. Anal Methods 3:509–513

626

A. Cernat et al.

Zhang D, Zhang Y, Zheng L et al (2013) Graphene oxide/poly-l-lysine assembled layer for adhesion and electrochemical impedance detection of leukemia K562 cancer cells. Biosens Bioelectron 42:112–118 Zhang Z, Liu J, Fan J et al (2018) Detection of catechol using an electrochemical biosensor based on engineered Escherichia coli cells that surface-display laccase. Anal Chim Acta 1009:65–72 Zhu X, Wu G, Lu N et al (2017) A miniaturized electrochemical toxicity biosensor based on graphene oxide quantum dots/carboxylated carbon nanotubes for assessment of priority pollutants. J Hazard Mater 324:272–280

Biosensors and Bioelectronics on Smartphone

26

Shuang Li, Daizong Ji, Gang Xu, Jinglong Liu, Yanli Lu, Sze Shin Low, and Qingjun Liu

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Smartphone-Based Electrochemistry System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Amperometry Sensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Potentiometry Sensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Impedimetry Sensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Smartphone-Based Spectroscopy System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Optical Sensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Electrochemical-LSPR Sensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Electrochemiluminescence Sensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NFC on Smartphone for Biosensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NFC-Based Wearable Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NFC-Based Implanted Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NFC-Based Gas Sensing Tags . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusion and Future Direction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

628 630 631 634 636 638 639 639 642 644 645 647 647 649 650

Abstract

Biosensors and bioelectronics have been developing rapidly, playing an increasingly important role in the fields of medicine, food, environment, and public safety. It is of utmost importance to develop fast, real-time, accurate, and portable biosensors and detection systems. Smartphone, owing to the advantages of high processing speed, high-definition image analysis, and excellent human–computer interaction interface, has been widely integrated with sensors, such as sensor chips and handheld detectors for biochemical detections. In general, the smartphone-based detection system used the built-in function modules as S. Li · D. Ji · G. Xu · J. Liu · Y. Lu · S. S. Low · Q. Liu (*) Biosensor National Special Laboratory, Key Laboratory for Biomedical Engineering of Education Ministry, Department of Biomedical Engineering, Zhejiang University, Hangzhou, P. R. China e-mail: [email protected] © Springer Nature Switzerland AG 2022 G. Thouand (ed.), Handbook of Cell Biosensors, https://doi.org/10.1007/978-3-030-23217-7_157

627

628

S. Li et al.

controller, analyzer, and displayer, which significantly simplified the design and reduced the cost of the system. This chapter presents smartphone-based biosensing with bioelectronics in electrochemistry, spectroscopy, and near-field communication. Detector attachments, sensor strategies, and communication modes were introduced in detail to provide clear designs of smartphone-based systems. In electrochemistry systems, amperometry sensing, potentiometry sensing, and impedimetry sensing were introduced with specific technical circuit implementation and biochemical applications. In spectroscopy systems, we reviewed the optical sensing on smartphone, firstly. Then electrochemical-localized surface plasmon resonance and electrochemiluminescence sensing were comprehensively discussed on smartphone. Finally, near-field communication technique on smartphone was applied for wearable and implanted biosensing on flexible devices. Along with their applications in point-of-care testing, it can be concluded that biosensors and bioelectronics based on smartphone would be the promising developing directions in biochemical sensing. Keywords

Smartphone · Biosensor · Bioelectronics · Electrochemistry · Spectroscopy · Near-field communication (NFC) · Point-of-care testing (POCT)

Introduction Biosensor is a device that combines biological sensing elements with physical and chemical transducers for analysis and detection. In the past few decades, the development of biosensors was rapid due to their high sensitivity and selectivity (Goode et al. 2014; Velasco-Garcia and Mottram 2003; Fan et al. 2008). Various biological materials, such as cells, enzymes, antigen-antibodies, peptides, and nucleic acids, have been used as sensing elements, while combining with different types of transducers like electrochemical and optical detectors (Védrine et al. 2003; Gorton et al. 1999; Conroy et al. 2009; Wang 2002; Brandt and Hoheisel 2004). Via signal readout equipment and signal processing system, these biosensors and bioelectronics have been successfully applied for clinical diagnosis, drug screening, and environmental monitoring (Singh et al. 2014; Joseph et al. 2003; Wang et al. 1997). Nowadays, the miniaturization of biosensor system has become the research hotspot in order to realize convenient and real-time detection. With the development of micro/nano processing technology, biosensors can be integrated into “lab-on-achip” for biochemical detection (Balakrishnan et al. 2014; Wang 2000; Abgrall and Gue 2007). Although these micro/nano biosensors have made great progress in sensitivity and automation, they usually need bulky readout devices to implement sensing processes, detect response signals, analyze data, and display results. Therefore, the large volume and high cost of readout devices became one of the obstacles for miniaturized biosensors to be used for portable and efficient detection.

26

Biosensors and Bioelectronics on Smartphone

629

Since the advent of the first smartphone in 1983, smartphone has brought tremendous reformation to the market, in which the number of worldwide users will reach 3.5 billion by 2019 (Sood et al. 2018). Due to its high popularity, convenient operation, open-source operation systems, high imaging, and computing capabilities, smartphone has played a pivotal role in electrochemical detection and optical detection (Zhang and Liu 2016; Vashist et al. 2014; Xu et al. 2018; Bisetty 2018). Using the high resolution camera equipped in smartphone, it is easy to combine the imaging system in the phone with the biosensor system to achieve optical detection, such as microscopic imaging, fluorescence, colorimetry, and surface plasmon resonance (McCracken and Yoon 2016; Kanchi et al. 2017; Geng et al. 2017). One of the most prominent advantages of smartphone-based optical biosensors is that phone cameras can be used as optical detectors or as a link between phones and sensors, which can greatly simplified the system and reduced the use of optical accessories. Nevertheless, simple image capturing is susceptible to the influence of external environment, which hinders the accurate quantification for detection. Although many attempts have been made, such as using black boxes, matching algorithms, and parallel detection, the problem has not been thoroughly solved so far. Therefore, smartphone-based optical biosensors are more widely used for qualitative analysis. As an excellent quantitative detection technology, electrochemical detection with the characteristics of direct feedback has the advantages of high reliability, easy operation, low detection limit, and cost (Lisdat and Schäfer 2008; Biran et al. 2000; Kerman et al. 2003). These features made it conducive for the miniaturization and integration of the electrochemical detection system. Smartphone with many functional modules, such as Bluetooth, universal serial bus (USB), and audio jack, have been applied in the combination with portable devices to form an integrated platform for controlling, recording, and displaying electrochemical detection. According to the different sensing technology, electrochemical sensing on smartphone can be divided into three different types, namely, amperometry (Guo 2016; Nemiroski et al. 2014; Ji et al. 2017, 2018a, b), potentiometry (Nemiroski et al. 2014; Sun et al. 2014; Bai and Lin 2014; Zhang et al. 2015a; Yao et al. 2019), and impedimetry (Zhang et al. 2015b, 2016; Liu et al. 2017). In addition to the single electrochemical sensing detection, the electrochemical-optical combination technology has attracted the attention of researchers due to its mutual enhancement and complementarity. It has shown advantages in electron photocatalysis (Wang et al. 2017a; Clavero 2014), dielectric tuning (Ma et al. 2017; Llorente et al. 2017; Kawawaki et al. 2017; Di Martino et al. 2017), and light-matter interactions control (Kato et al. 2018). Usually, the electrochemical-optical detection is carried out on nanochips, and therefore, it is also called the electrochemical-localized surface plasma resonance (LSPR) (Li et al. 2018a; Zhang et al. 2015c; Li et al. 2016a, b, 2017). The nanochip with high sensitivity could make it convenient to design smartphone-based real-time, accurate, and portable biosensor. Besides, electrochemiluminescence is an electrochemistry controlled optical radiation process. The luminescence may be triggered by voltage signal without

630

S. Li et al.

peripheral excitation unit. This made electrochemiluminescence show unique superiorities over other electrochemical and optical methods, which could not only invalidate interferences of background lights but also simplify detection apparatuses (Forster et al. 2009). Researches showed that the interfaces on smartphone could be used to provide the excitation energy for luminescence to occur (Delaney et al. 2013a; Doeven et al. 2015; Li et al. 2018b). In addition, the luminous images could be captured with luminous intensities be analyzed by the high-resolution cameras and applications equipped in smartphones (Lebiga et al. 2015; Yao et al. 2017). Thus, electrochemiluminescence on smartphone would be a potential candidate for the design of “all in one phone” detection systems, which could realize the selfexcitation and luminescence analysis, benefiting for point-of-care testing (POCT). For most smartphone-based biosensors and bioelectronics, power supply and data transmission are inevitable issues (Zhang and Liu 2016; Gao et al. 2016a). Usually, they require wired connections with smartphones through USB/audio port for simultaneous power and data transmission (Sun et al. 2014; Doeven et al. 2015; Yao et al. 2015), or communicate wirelessly with smartphone through Bluetooth and powered by external batteries (Parrilla et al. 2016; Nyein et al. 2016; Kim et al. 2016a). Although these solutions worked well in many application scenarios such as portable devices, they were constrained for wearable and implantable applications. Recent years, wearable sensors and electronics have aroused considerable interests owing to their great promise for real-time monitoring of personal activity and health status, without interrupting or limiting the user’s motions (Bandodkar et al. 2016; Bandodkar and Wang 2014). With the connectivity of ubiquitous smartphones, these devices could be used for POCT and mobile health. Therefore, this chapter described the biosensors and bioelectronics based on smartphone from three aspects: (1) electrochemistry system, (2) spectroscopy system, (3) wireless and passive system. In the field of electrochemistry detection, amperometry sensing, potentiometry sensing, and impedimetry sensing were introduced with specific technical circuit implementation and their application in POCT. In the field of spectroscopy detection, optical sensing, electrochemical-LSPR sensing, and electrochemiluminescence sensing were introduced with nanochip design and imaging analysis on smartphone. Lastly, near-field communication (NFC) on smartphone was discussed for wearable and implanted biosensing.

Smartphone-Based Electrochemistry System With the advantages of high reliability, easy operation, and high sensitivity, electrochemical sensing technique was employed as a significant quantitative method in biosensor. In spite of this, electrochemical detection was still limited to the wellresourced environment. Smartphone, with many excellent capacities, provided electrochemical analysis a promising choice for in-situ and real-time detection. Based on advanced sensor techniques used in electrochemical devices, smartphone-based electrochemical system not only completed the biosensing requirements with wide detection range and low limit detection but also provided the convenience to POCT.

26

Biosensors and Bioelectronics on Smartphone

631

Table 1 Electrochemical sensing on smartphone (LoD: Limit of detection, CA: Chronoamperometric, CV: Cyclic voltammetry, DPV: Differential pulse voltammetry, DPA: Differential pulse amperometry, CP: Chronopotentiometry, EIS: Electrochemical impedance spectroscopy) Technique Mechanism Analyte Amperometry CA Glucose

LoD 0.18 mM

10 cfu/mL

Reference Steinberg et al. (2015) Kassal et al. (2015) Ji et al. (2017) Wang et al. (2015) Ji et al. (2018a) Wang et al. (2017b) Ji et al. (2018b) Zhang et al. (2015a) Nemiroski et al. (2014) Sempionatto et al. (2017) Zhang et al. (2015b) Zhang et al. (2016) Jiang et al. (2014)

1.56 ppm

Liu et al. (2017)

CA

Uric acid

0.1–0.8 nM

0.1 nM

CV CV DPV DPV

0.1–10 mM 0.2–70 ppm 20–375 μM 10–106 cells

0.026 mM 0.2 ppm 1.04 μM 60 cells

0.5–200 μM 30–1000 U/mL

0.1 μM 0.12 U/mL

CP

Glucose Nitrate Ascorbic acid White blood cell Levodopa Salivary αamylase Na+

10
4–10
1 M

10
4 M

CP

K+

0.1–100 mM

10–3.9 M

EIS

TNT

10
5–10
3 M

10
6 M

EIS

Thrombin

2.97 ng/ml

EIS

Bacteria

EIS

Acetone

10 ng/ml–1 μg/ ml 102–105 cfu/ mL 1–10 ppm

DPA Potentiometry CP

Impedimetry

Range 0.1–10 mM

Table 1 showed the summarization of electrochemical sensing on smartphone with different techniques, such as amperometry, potentiometry, and impedimetry.

Amperometry Sensing Amperometry sensing was reported as smartphone-based electrochemical sensor system for biosensors at the earliest (Zhang and Liu 2016; Guo 2017). Until now, amperometry is still the most widely used electrochemical technique combined with smartphone for biosensing (Xu et al. 2018; Aymerich et al. 2018). In the amperometry, the variation of current resulted from the oxidation or reduction of the electrical mediating species when the potential was applied, which is proportional to the concentration of analytes (Huang et al. 2018). Hence, the various amperometry sensing methods, such as chronoamperometry, cyclic voltammetry, differential pulse amperometry, and differential pulse voltammetry, were combined with smartphone for different analyte detections (Nemiroski et al. 2014; Ji et al. 2017, 2018a, b; Wang et al. 2017b). For smartphone-based amperometry system, these extraordinary methods could significantly enhance the sensitivity of sensor system for analyte confirmation.

632

S. Li et al.

Lillehoj et al. reported an exquisite smartphone-based chronoamperometry platform for rapid and quantitative detection of Plasmodium falciparum histidine-rich protein 2 (PfHRP2) firstly, which was an important biomarker for malaria (Lillehoj et al. 2013). The platform consisted of smartphone, the embedded circuit, and the disposable microfluidic chips. The smartphone was used to transfer the data with the embedded circuit by audio jack, control the fluid and biosensing on the microfluidic chip, and display the results on the screen for immediate assessment. By recoding the electrochemical current from bio-interactions with constant potential, this system can successfully complete the biomarker measurement. After that, a series of further work about amperometry system were combined with smartphone for biosensing. Cyclic voltammetry has shown its great practicability for quantitative detection and electrode modification. While smartphone is an integrated platform to receive, analyze, and display data, it plays an irreplaceable role in the portable device and is considered as the hotspot testing device in the field of biosensor (Wang et al. 2015; Quesada-González and Merkoc¸i 2017; Aronoff-Spencer et al. 2016; Sun et al. 2016). Therefore, Ji et al. developed a smartphone-based cyclic voltammetry system, which could perform electrodes modification and cyclic voltammetry detections (Ji et al. 2017). Compared with the traditional electrochemical workstation, this system featured the lower price, smaller volume, and obtaining the real-time monitoring data. There were three main portions in the system: modified electrodes, portable electrochemical detector, and smartphone (Fig. 1a). The detector consisted of an energy transformation module applying the stimuli signals and a low-cost potentiostat module for cyclic voltammetry measurements with a Bluetooth module for transmitting data and commands (Fig. 1b). The experimental results showed that the system was linear, sensitive, and specific responses to glucose at different doses, even in blood serum as low as about 0.026 mM. Furthermore, the test error of this system was less than 3.8% compared to that of commercial electrochemical workstation (Fig. 1c). As a result, this smartphone-based cyclic voltammetry system could show great potentials of detection and modification of electrodes in various fields, such as public health, water monitoring, and food quality. Except for the cyclic voltammetry, differential pulse voltammetry is also a popular electrochemical method that could be used for trace detection especially at low concentration of supporting electrolyte (Wang et al. 2017b; Fan et al. 2017). Combing these two methods could be useful to perform simultaneous modification and detection in POCT. Hence, Ji et al. designed a smartphone-based integrated voltammetry system using modified electrode for simultaneous detection of biomolecules (Fig. 1d) (Ji et al. 2018a). In his study, ascorbic acid, dopamine, and uric acid were chosen as the determination substances, since they are all important electroactive biomolecules for health monitoring and coexist in serum or urine. Therefore, their quantitative determination by electrochemistry could provide the accurate reference for diseases diagnosis and treatment. Besides, differential pulse amperometry was a preeminent electrochemical method, which could be used for real-time monitoring of biochemical substances and drug (Moreira et al. 2018). For example, Parkinson’s disease caused by lack of dopamine in brain is a common neurodegenerative disorder. The traditional treatment is to replenish levodopa since

Biosensors and Bioelectronics on Smartphone

Fig. 1 Amperometry sensing on smartphone. (a) The image of the smartphone-based cyclic voltammetry system. (Reproduced from Reference (Ji et al. 2017). Copyright 2017 Elsevier). (b) The schematic diagram of the smartphone-based cyclic voltammetry system. (Reproduced from Reference (Ji et al. 2017). Copyright 2017 Elsevier). (c) The cyclic voltammogram of the smartphone-based system (solid line) and the commercial electrochemical workstation (dotted line) at different scan rate in redox couple solution. (Reproduced from Reference (Ji et al. 2018a). Copyright 2018 Elsevier). (d) The simultaneous detection for ascorbic acid, dopamine, and uric acid using smartphone-based voltammetry system. (Reproduced from Reference (Ji et al. 2018a). Copyright 2018 Elsevier). (e) The diagram of differential pulse amperometry measurement using the system for levodopa at different concentrations. (Reproduced from Reference (Ji et al. 2018b). Copyright 2018 Elsevier)

26 633

634

S. Li et al.

it could pass through blood brain barrier and form dopamine. However, its accumulation can cause patients’ movement disorders and uncontrollable emotion. Therefore, it is critical to control the levodopa dosage accuracy to improve the curative effect in clinical. As shown in Fig. 1e, Ji et al. constructed a smartphone-based electrochemical detection system was developed for rapid monitoring of levodopa. First, single-wall carbon nanotubes and gold nanoparticles modified screen-printed electrodes were used to convert and amplify the electrochemical current signals upon presence of levodopa molecules. Then the electrochemical detectors were used to generate electrochemical excitation signals and detect the resultant currents. What’s more, smartphone was connected to the detector, which was used to control the detector, calculate data, and plot graph in real-time. The experimental data showed that this smartphone-based differential pulse amperometry system was demonstrated to monitor levodopa at concentrations of 0.5 μM in human serum. Furthermore, it has also been verified to be able to distinguish levodopa from other representative substances in the body. Therefore, its performance was more sensitive and rapid than electrochemical workstation. This system can be used in the field of POCT to detect levodopa and provide the possibility to solve clinical demand for levodopa detection with these advantages.

Potentiometry Sensing As important as amperometry, potentiometry was also applied to smartphone-based electrochemical system for biosensing (Zhang and Liu 2016; Yao et al. 2019). As an electro-readout electrochemical technique, potentiometry was appropriate for potential sensing applications in which the cumulative electrical charges result in the differences in electrical potential on top of the dielectric layer (Zhang et al. 2015a; Huang et al. 2018). Hence, potentiometry was usually employed to measure ions with ion-selective electrodes. Nemiroski and his co-workers presented a mobile electrochemical detector with audio jack for the detection of sodium and potassium ions (Fig. 2a) (Nemiroski et al. 2014). They used ion-selective electrodes to evaluate the performance of potentiometry on their system and compared the system with commercial instrument (Fig. 2b). Moreover, they also detected a series of urine samples with different levels of sodium from standard urine samples, since sodium content in urine is often used to evaluate the amount of fluid within the blood vessels or the overall balance of electrolytes within the body. The results showed that the systematic error of the measured concentrations (~8%) falls within the certified range of the assayed urine samples (14%). As shown in Fig. 2c, Sempionatto and his companions present a fully integrated wireless multiplexed chemical sensing platform on eyeglasses, which could be used for real-time monitoring of sweat electrolytes and metabolites. An amperometric lactate biosensor and a potentiometric potassium ion-selective electrode were integrated on the two nose bridge pads of the glasses and connected with a wireless electronic backbone placed on the glasses arms. Then, the platform could transmit data through Bluetooth to the host device, such as smartphone. The sensor responds

26

Biosensors and Bioelectronics on Smartphone

635

in 20 s in a near-Nernstian fashion to varying potassium concentrations over the 0.1– 100 mM range (R2 ¼ 0.99), with a Nernstian slope of 58.0  4.3 mV∙decade
1 (n ¼ 15) (Fig. 2d). The results shown demonstrate that the potentiometric platform could be used for potassium detection in human sweat, given that the average levels of potassium in human sweat varies from 6.7 μM to 38 mM. The glasses prototype was also used to detect sweat potassium and lactate on body, which showed good performance of sensing in real time during exercise activity. Gao and his co-workers reported a flexible and fully integrated sensor array for multiplexed in situ perspiration analysis with smartphone (Fig. 2e), which simultaneously and selectively measures sweat metabolites (such as glucose and lactate) and electrolytes (such as sodium and potassium ions), as well as the skin temperature (to calibrate the response of the sensors). Figure 2f showed the open circuit potentials of Na+ sensors in the electrolyte solutions with physiologically relevant concentrations

Fig. 2 Potentiometry sensing on smartphone. (a) Photograph of the sensor system for universal electrochemical detections. (Reproduced from Reference (Nemiroski et al. 2014). Copyright 2014 National Academy of Sciences). (b) Detection of [K+] and [Na+] with potentiometry in an ionic strength adjuster. (Reproduced from Reference (Nemiroski et al. 2014). Copyright 2014 National Academy of Sciences). (c) Multi-Analyte Glasses biosensor device. (Reproduced from Reference (Sempionatto et al. 2017). Copyright 2017 The Royal Society of Chemistry). (d) Response of the potassium sensor to 0.1, 0.1, 1, and 100 mM KCl in DI water. (Reproduced from Reference (Sempionatto et al. 2017). Copyright 2017 The Royal Society of Chemistry). (e) Photograph of a wearable flexible integrated sensing array on a subject’s wrist. (Reproduced from Reference (Gao et al. 2016b). Copyright 2016 Macmillan). (f) The open circuit potential responses of the sodium sensors in NaCl solutions. (Reproduced from Reference (Gao et al. 2016b). Copyright 2016 Macmillan)

636

S. Li et al.

of 10–160-mM Na+. The Na+ sensors showed a near-Nerstian behavior with sensitivities of 64.2 mV per decade of concentration. The sensors also showed excellent performance, such as repeatability and long-term stability. The platform bridges the existing technological gap between signal transduction, conditioning, processing, and wireless transmission in wearable biosensors. It merged commercially available integrated-circuit components, consolidated on a flexible printed circuit board (FPCB), with flexible and conforming sensor technologies fabricated on plastic substrates. Hence, the platform could be employed for in situ analyses of other biomarkers within sweat or other fluid samples.

Impedimetry Sensing Electrochemical impedance spectroscopy is a kind of conductivity sensing mode, which shows the changes of electrochemical properties such as surface capacitance of electrode and interface electron transfer resistance by measuring the complex impedance of electrochemical system (Chang and Park 2010; Macdonald and Barsoukov 2005). By applying a small disturbance signal to the electrochemical system, and then observing the electrochemical properties of the feedback analysis system, the corresponding electrochemical signal detection can be realized. It took the sinusoidal potential/current wave as the excitation signal, which made the response in the system approximately linear. Usually, it measured the impedance spectrum of the electrode system in a certain frequency range. This is also called the alternating current (AC) impedance spectroscopy. In biosensor applications, electrochemical impedance spectroscopy detection is generally used to measure the impedance characteristics of the electrode interface, which was modified with biomolecules such as antibody, peptide, and aptamer (Zhang et al. 2015b, 2016; Li et al. 2008; Fan et al. 2013). Therefore, it could realize the analysis and detection of biological binding and reaction process. The design of smartphone-based AC impedance spectroscopy system mainly included two parts. On the one hand was the real-time monitoring of complex impedance information, such as real impedance value, imaginary impedance value, complex impedance value, and phase angle. On the other hand was the real-time monitoring of multiple different frequencies of sensors within a certain frequency range. Zhang et al. designed a handheld detector for impedance spectroscopy analysis (Zhang et al. 2016). Among them, the hardware and software of the detection circuit were mainly designed to measure the impedance-time information and impedance-frequency information of the sensor, and the corresponding smartphone application program was developed to receive information, analyze data, and display results. As shown in Fig. 3a, complex impedance of the parallel circuit of 10 kΩ resistance and 1 nF capacitance was tested by the smartphone-based system, which held good consistency with the workstation. Then, AC impedance was measured on two electrode system in presence of the redox couple, in which gold electrode and platinum electrode were used as working electrode and reference electrode, respectively. The smartphone-controlled system showed similar Nyquist

26

Biosensors and Bioelectronics on Smartphone

637

Fig. 3 Impedimetry sensing on smartphone. (a) The smartphone screen showing the AC impedance spectroscopy measurement. (Reproduced from Reference (Zhang et al. 2016). Copyright 2016 Elsevier). (b) Picture of the impedance monitoring device with Arduino board and expansive board. The expansive board consists of Bluetooth module (1), AD5933chip (2), LM358chip (3), and sensor interface (4) connecting to the working and reference electrodes. (Reproduced from Reference (Zhang et al. 2015b). Copyright 2015 Elsevier). (c) The screen-printed electrode modified by anti-BSA with nitrocellulose membrane for BSA detection. (Reproduced from Reference (Zhang et al. 2016). Copyright 2016 Elsevier). (d) Binding of peptides on the surface of the electrodes for TNT detection. (Reproduced from Reference (Zhang et al. 2015b). Copyright 2015 Elsevier). (f) The interdigital electrode with immobilization of graphene, ZnO, and nitrocellulose membrane for VOC detection. (Reproduced from Reference (Liu et al. 2017). Copyright 2017 Elsevier)

plot to the workstation in the frequency scanning from 10 Hz to 10 kHz. The handheld device had impedance converter network analyzer (AD5933 and LM358 circuit), microcontroller (Arduino board), and Bluetooth module. The analyzer gave out sinusoidal signals on the working electrode as electric stimuli and monitored feedback signals from the counter electrodes. In presence of analytes, impedance

638

S. Li et al.

signals could be sent back to the smartphone through Arduino board and Bluetooth module, on which AD5933 circuit and Bluetooth module was integrated into an expansion printed circuit board (Fig. 3b). The Arduino board received commands from smartphone by Bluetooth module and controlled the impedance parameter of AD5933 circuit, such as starting, finishing point, and AC frequency of the stimuli signals. For biosensing, anti-BSA was immobilized on the electrode surface by nitrocellulose membrane, on which conjunctions of BSA and anti-BSA could block the electron transferring on the electrodes. As shown in Fig. 3c, the interface impedance on the electrode was increased with the increase of the BSA concentration. Besides, the interdigital gold electrode was also a kind of sensitive electrodes for highthroughput measurement, which was fabricated using microscale electrodes on silicon and glass substrate by micro-electro-mechanics. Besides, Zhang et al. also designed the TNT-peptide sensor on the smartphone-based impedimetry sensing system (Zhang et al. 2015b). Here, peptides were immobilized on surface of the electrodes and specifically bind to TNT molecules, which inhibited electron transfer on surface of the electrode (Fig. 3d). The whole impedance detection lasted for 120 s, during which TNT was added at the point of 50 s after beginning of the recording. After a rapid decrease, the binding of TNT to peptides on the electrodes increased the impedance dramatically. Ultimately, significant impedance increase could be observed with the concentration increase. Except for the solid-phase or liquidphase detection, Liu et al. proposed the volatile organic compounds (VOCs) detection by the smartphone-based system (Liu et al. 2017). By adding a gas sample device, the system was demonstrated to detect acetone, even carried out the measurements of acetone exhalations from human in exhaled breath before and after exercise. To effectively capture VOCs, the surface of the electrode was modified with graphene, ZnO, and nitrocellulose membrane (Fig. 3e). During the detection, acetone was added into the sensing container at 90 s from the beginning. The resistance increased rapidly into a high plateau phase and kept stable for a period of time. After 150 s, the resistance decreased returned to the initial base line, because VOCs were removed, and pure air was added into the sensing container again. Therefore, the modified electrode showed good reversibility in the detection of VOCs. Meanwhile, the peak of the response increased along with the increasing concentration of acetone. The results demonstrated that the resistance change had dose-dependent characteristics. The smartphone-based VOCs detection system provided a convenient, portable, and efficient approach in exhaled breath detection, which benefitted for early diagnosis of some diseases.

Smartphone-Based Spectroscopy System Smartphone is a kind of built-in high-resolution image acquisition device, which is widely used in biosensing system for optical detection (McCracken and Yoon 2016; Kanchi et al. 2017; Geng et al. 2017). In this section, single optical sensing was

26

Biosensors and Bioelectronics on Smartphone

639

reviewed firstly. Then, electrochemical-LSPR sensing and electrochemiluminescence sensing were introduced with nanochip design and imaging analysis on smartphone.

Optical Sensing Smartphone-based microscopy is one of the most widely used devices, which can be realized by integrating mobile phones with high-resolution optical accessories. These microscopic imaging results could be calculated and analyzed in real time. Then the acquired microscopic image features (such as the size, color, and brightness of the target) could be quickly quantified. These image features were applied as signal labels to diagnose disease and detect food safety (Breslauer et al. 2009; Zhu et al. 2011; Smith et al. 2011). In addition to microscopic imaging, the progresses of light intensity detection technology have also promoted the development of fluorescence, colorimetric, and surface plasmon resonance detection on smartphone (Barbosa et al. 2015; Shen et al. 2012; Dutta et al. 2016). Among them, fluorescence was introduced as an effective sample marker to improve microscopic imaging resolution. However, smartphone-based fluorescence microscopes typically required large optical attachments and long analysis, which invisibly increased the complexity of imaging operations. To capture and quantify nanoscale targets and overcome these limitations, the quantitative assay of fluorescence intensity was proposed by Michigan State University’s Stedtfeld et al. (Stedtfeld et al. 2012) In addition, colorimetry is also easy to implement method on smartphone, which can be used to analyze image pixel intensity for target detection (Shen et al. 2012). In order to overcome external interferences and individual differences, Li et al. reported a smartphone-based system with multimode analysis of red-green-blue (RGB), huesaturation-value (HSV), hue-saturation-lightness (HSL), and cyan-magenta-yellowblack (CMYK) (Li et al. 2018c). Overall, smartphone can be used as both an image acquisition device and an image analysis device, which greatly simplified the detection system and reduced the use of optical accessories. Therefore, it can meet the needs of portability and rapid detection in POCT field, especially in rural areas where medical equipment is lacking.

Electrochemical-LSPR Sensing Traditionally, electrochemical sensing happens on the electrodes, and optical sensing happens on materials with spectral effects. Finding a device that that can be used for both electrochemical and optical detection was the top priority. LSPR is a special optical phenomenon of noble metal nanoparticles. At the same time, the metal film used to excite plasma resonance is used as the working electrode in the electrochemical detection. Li et al. designed a nanochip for electrochemical spectroscopy detection (Li et al. 2018a). The nanochip consisted of a cone array structure with gold and silver nanoparticles, fabricated by green and nontoxic physical control and electrochemical reduction. The nanochip showed excellent performances in LSPR,

640

S. Li et al.

electrochemistry, and electrochemical-LSPR detections (Fig. 4a). The dynamic electro-optical spectroscopy was based on a unique combination of LSPR and linear sweep voltammetry. During the detection, plasmon-induced charge separation were concentrated on the metal nanoparticles surface, which resulted in excited energetic charges of the hot-electrons and hot-holes. Following light absorption and LSPR excitation on the nanochip, electromagnetic decay took place on a femtosecond timescale and transferred the energy non-radiatively to hot-electrons. With the electrochemical driving, hot-electrons became more active leading to an enhanced propagation of surface plasmon resonance. This process could also effectively inhibit the recombination of hot-electrons with hot-holes to stabilize LSPR reaction. Therefore, electrochemical-LSPR realized the mutual amplification of the electrooptical response and also attained high efficiency and good sensitivity during biochemical analysis. Besides, Zhang et al. proposed the combination of LSPR and cyclic voltammetry for BSA detection (Zhang et al. 2015c). The redox current on the nanochip brought extra dip shift in wavelength in transmission spectrum and enhanced LSPR signals. Further, Li et al. used the electrophoresis to enhance LSPR sensing on the nanochip for thrombin detection (Li et al. 2016a). The refractive index on the sensor surface decreased with the cleavage of the peptide. When direct current was added to the nanochip, the peptide was freed from the sensing surface, which caused the offset enhanced. Besides, the nanochip-based electrochemical-LSPR was also applied by Li et al. for neurotransmitters detection (Li et al. 2017). As shown in Fig. 4b, arrows allocated the location of dopamine’s electrooxidation, in which first anodic peak appeared at 55 s, with the following peaks separated by 280 s. To better observe the intensity changes and peak position of the transmission spectrum during cyclic voltammetry scanning, the color contour plot was presented by aligning every transmission spectrum as a function of time (Fig. 4c). From the contour plot, the intensity of both characteristic peaks of the nanochip fluctuated periodically as the wave form developed. When anodic current of dopamine appeared, the transmission intensity reached its maximum. Xu’s et al. study combined this nanochip with handheld device for biochemical detection. They demonstrated a highly sensitive, wafer-scale, highly uniform plasmonic nano-mushroom substrate based on plastic for naked-eye plasmonic colorimetry detection (Xu et al. 2016). With the increases of refractive index from 1.333 to 1.518, the color changes from green to yellow and then to red. RGB and HSV analysis of the images indicated that the red (R) channel increased, the green (G) and blue (B) decreased as the refractive index of the liquid increased. The handheld Raman spectrometer can be used for easy and fast detection of trace amounts of the narcotic drug methamphetamine in drinking water. Next, Wang et al. proposed a self-referenced smartphone-based nanoplasmonic imaging platform (Fig. 4d) (Yao et al. 2019). Liquid refractive index sensing and optical absorbance enhancement sensing on the nanostructured plasmonic sensor enabled the colorimetric biochemical sensing. Refractive indices of colorless liquids were measured by smartphone-based imaging and color analysis. The red-green-blue channels were normalized according to the standard RGB color patterns captured on the same

Biosensors and Bioelectronics on Smartphone

Fig. 4 Electrochemical-LSPR detection. (a) The apparatus of the system and the schematic diagram of electrochemical-LSPR. (Reproduced from Reference (Li et al. 2018a). Copyright 2018 Elsevier). (b) Current responses of the nanosensor. (Reproduced from Reference (Li et al. 2017). Copyright 2017 Elsevier). (c) Spectral responses of the nanosensor. (Reproduced from Reference (Li et al. 2017). Copyright 2017 Elsevier). (d) Smartphone-based nanoplasmonic imaging platform. (Reproduced from Reference (Yao et al. 2019). Copyright 2016 American Chemical Society). (e) Normalized spectrum of white LED and light source from Princeton Instrument. (Reproduced from Reference (Yao et al. 2019). Copyright 2016 American Chemical Society)

26 641

642

S. Li et al.

image. This normalization step overcame the imperfectness of the reference color patterns from the plastic reflection and ink absorption. Among them, the green channel showed a decreasing trend in both spectral sensitivity and light source, which made it the most sensitive channel for characterization (Fig. 4e). By the matching of plasmon resonance wavelength and chromophore’s absorbance peak wavelength, optical absorbance enhancement in the colorimetric biochemical assay was achieved on the nanochip. Compared with a traditional colorimetric assay, this smartphone plasmon enhanced colorimetric sensing system provided improvement in the limit of detection.

Electrochemiluminescence Sensing Electrochemiluminescence is a kind of electrochemistry triggered luminescence technique, which produces luminescence on the surface of electrodes due to electrochemical reactions. It has the advantages of wide dynamic concentration response range, voltage controllability, and high sensitivity and rapidity (Blackburn et al. 1991; Muzyka 2014; Rizwan et al. 2018). These made it develop rapidly and gradually form a set of perfect system applications in POCT. Especially, the integration of electrochemiluminescence into “all-in-one device” is one attractive direction to design portable, user-friendly, and cost-effective detector (Delaney et al. 2013b; Gao et al. 2017). These portable systems often connected sensor devices to smartphones for sensing different types of information from optical spectra or electrochemical currents. By using the built-in hardware and sensors of smartphone, emission excitation and photography could be realized, which would replace the excitation device in traditional electrochemiluminescence detection. Li et al. developed a smartphone-based electrochemiluminescence system by using universal serial bus on-the-go (USB-OTG) as voltage excitation voltage and camera as luminescence collector (Fig. 5a) (Li et al. 2018b). USB-OTG on the smartphone had five lines including GND, ID, D+, D-, and Vcc. Among them, ID was used to identify the host device or slave device during USB-OTG communication. Vcc was the power line with GND as the reference potential. Luminescence images were captured by phone camera, which were analyzed by red-green-blue (RGB), hue-saturation-brightness (HSB), and gray. Results showed that HSB_S was the most principal parameters in luminescence images during trinitrotoluene detection with the detection limit of 2.3  10
9 mg/mL. Besides, Delaney et al. used the audio jack of a mobile phone for potentiostatic control (Delaney et al. 2013b). Smartphone with suitable software installed can serve as a potentiostat in controlling an applied potential to oxidize electrochemiluminescence-active molecules, while the resultant photonic signal was monitored using the camera in video mode (Fig. 5b). Due to the consumption of coreactant, the photonic signal raised and fell with the excitation signal and gradually decreased over time. Similarity responses demonstrated that even under the conditions used, the smartphone can also

26

Biosensors and Bioelectronics on Smartphone

643

Fig. 5 Sensing on smartphone. (a) Smartphone-based electrochemiluminescence system with USB-OTG as the electrical stimulation and camera as luminescence images detector for trinitrotoluene detection. (Reproduced from Reference (Li et al. 2018b). Copyright 2018 Elsevier). (b) Audio jack supplied the potentiostatic control of electrochemiluminescence. (Reproduced from Reference (Delaney et al. 2013b). Copyright 2013 Elsevier). (c) Handheld electrochemiluminescence system with luminescence images on the driving voltage from 3.0 V to 9.0 V. (Reproduced from Reference (Chen et al. 2016). Copyright 2016 Elsevier)

644

S. Li et al.

adequately perform the function of potentiostatic control necessary for electrochemiluminescence detection. Both the excitation and detection processes were controlled by a software application which can also transmit the results via e-mail. Chen et al. designed a peripheral system to stimulate electrochemiluminescence (Chen et al. 2016). The peripheral system was composed of eight parts: (1) smartphone, (2) instrument container, (3) charging cable, (4) rechargeable lithium battery, (5) electronic circuit model for voltage control and display, (6) paper device, (7) device support, and (8) wireless transmission of electrochemiluminescence signal to PC (Fig. 5c). It was used as the power supply and could last about 146 h with operating current at 30 mA. Thus, such a battery would be highly suitable in POCT. In order to maintain, adjust, and display the driving voltage, an electronic circuit model for voltage control was applied, including the xl4015 chip, nonpolar capacitor, polar capacitor, potentiometer, inductance coil, Zener diode, and digital tube. Electrochemiluminescence intensity varies with voltages. The response was increased by increasing the driving voltage from 3.0 V to 7.0 V. There was no electrochemiluminescence signal when the driving voltage was lower than 3.0 V. This was due to insufficient voltage to initiate the oxidation of luminol. A high driving voltage initiated background reactions, such as the oxidation of water. In turn, the formation of oxygen interfered both chemically and physically with electrochemiluminescence emission. Therefore, when driving voltage was above 7.0 V, the electrochemiluminescence signal decreased. In the case of an adequate driving voltage, electrochemiluminescence signal can be collected by the complementary metal-oxide semiconductor (CMOS) camera of the smartphone. Through fully developed the functions of smartphone, potentiostatic control and luminescence analysis were realized on smartphone for self-exciting and imagingfriendly electrochemiluminescence. Especially, the handheld system showed good stability and sensitivity in biosensing, which would greatly benefit for mobile biochemical testing.

NFC on Smartphone for Biosensing Recently, NFC technology has come into the sight of researchers and spurring extensive interest (Morak et al. 2012; Leikanger et al. 2017). Compared to common smartphone-based interfaces (e.g., USB port, audio port, and Bluetooth), wireless power and data transmission through NFC is an effective alternative solution for wearable and implanted devices, which could obviate the need for long wires or onboard batteries (Steinberg et al. 2016). According to HIS technology, there would be 64% of the cellphones integrated with NFC modules worldwide by 2018 (IHS T 2014). By integrating NFC module in the biosensors and bioelectronics, NFCenabled smartphones can wireless power the devices and get the detection results through the inductive coupling between antennas (Coskun et al. 2013). Thus, the devices can be designed totally flexible, miniaturized, and more integrated, which could be comfortably adhered on human epidermis or implanted subcutaneously. Until now, NFC-enabled biosensors and electronics have been widely applied in

26

Biosensors and Bioelectronics on Smartphone

645

biomedical detections, environment monitoring, and implantable diagnostics/stimulations (Xu et al. 2017; Jung et al. 2015; Kim et al. 2015). Overall, NFC-enabled biosensors and bioelectronics can be divided into three categories, including wearable devices for on-body health monitoring, implanted devices for in vivo biosensing and excitations, and some flexible tags for health-related gas sensing.

NFC-Based Wearable Devices An ultrathin, lightweight, flexible near-field communication device with ultraminiaturized format, which was developed by Roger’s group in 2015 (Fig. 6a) (Kim et al. 2015). Benefit from the NFC technology, the device obviated the need for batteries and external wires. This millimeter-sized NFC device could be integrated onto teeth, fingernails, and ears to continuously detect the skin temperature, which could be wirelessly transmitted to smartphones. Then in 2016, an epidermal, stretchable optoelectronic system with NFC capabilities was developed by the same group (Kim et al. 2016b). As illustrated in Fig. 6b, this optoelectronic system consisted of several chip-scale components (e.g., LEDs, photodetectors, amplifier,

Fig. 6 NFC-based biosensors and bioelectronics. (a) Miniaturized flexible electronics system with NFC capability for temperature monitoring. (Reproduced from Reference (Kim et al. 2015). Copyright 2015 Wiley). (b) Battery-free and stretchable optoelectronic system for wireless optical characterization of skin. (Reproduced from Reference (Kim et al. 2016a). Copyright 2016 American Association for the Advancement of Science). (c) NFC-based adhesive patch for sweat electrolytes monitoring. (Reproduced from Reference (Rose et al. 2015). Copyright 2015 IEEE). (d) Soft wearable microfluidic device for colorimetric sweat analysis. (Reproduced from Reference (Koh et al. 2016). Copyright 2016 American Association for the Advancement of Science). (e) NFCenabled bioresorbable electronic system for implantable sensing of brain. (Reproduced from Reference (Kang et al. 2016). Copyright 2016 American Association for the Advancement of Science). (f) Flexible near-field wireless optoelectronics as subdermal implants for optogenetics excitations. (Reproduced from Reference (Shin et al. 2017). Copyright 2017 Cell Press)

646

S. Li et al.

and NFC die), antenna coils, and stretchable substrates. Through NFC schemes and magnetic inductive coupling between the antennas of the device and standard NFCenabled smartphone, the power was wirelessly delivered to multicolored diodes and other chip-scale components, and the digital data from integrated photodetectors was extracted simultaneously. Thus, this system could function without batteries and in an entirely wireless mode. Besides, carefully optimized materials and mechanics designs contributed to form factors capable of soft and conformal lamination onto the skin, with the ability to function properly under large strain deformation. This wearable device can be applied in monitoring heart rate and temporal dynamics of arterial blood flow, quantifying tissue oxygenation and ultraviolet dosimetry, and performing four-color spectroscopic evaluation of the skin, respectively. The detection results demonstrated the good performance of this sensing device, which proved that the device could be used both in hospital care and at-home diagnostics. Figure 6c illustrated an NFC-enabled electrochemical patch developed in 2015 (Rose et al. 2015). This flexible, battery-less, and fully integrated patch could be adhered to skin for monitoring of sweat electrolytes, such as sodium, and wireless read by a smartphone. The multilayer design of the patch was clearly demonstrated, including the porous adhesive layer, microfluidic paper pad, and the antenna/sensor layer. The microfluidics wicked sweat from the sweat porous adhesive to the ion selective electrodes (ISE). When exposed to different concentrations, the potential of ISE would have a linear response. With the energy harvesting and regulation of the NFC antenna and relevant electronics, the power could be wirelessly transmitted from the smartphone to the circuits of the patch for signal acquisition, data processing, and transmission. Finally, we could read the concentration of target analytes on the smartphone. As for wearable colorimetric devices, Roger’s group developed a soft, wearable microfluidic device for the capture, storage, and colorimetric sensing of sweat in 2016, as shown in Fig. 6d (Koh et al. 2016). The device was composed of multilayer stack of three subsystems. The bottom layer contacting with the epidermis was a skin-compatible adhesive layer with micromachined openings that defined the areas of sweat inlet/harvesting. This medical-grade acrylic adhesive film could tightly and seamlessly contact to skin without any irritation, even in the presence of sweat. Besides, because of the thin geometry and low modulus, this layer provided stress release during the deformation of the skin and ensured long-term wearability. In the middle of the device was a polydimethylsiloxane (PDMS) layer, embossed with elaborate microfluidic channels and reservoirs filled with color-responsive materials for quantitative analysis of sweat volume and chemical analytes. Particularly, the layout included four independent circular chambers and a surrounding orbicular serpentine channel, without any crosstalk. Each channel was connected with the corresponding openings on the adhesive layer by separate guiding channels. The top layer of the device was a magnetic loop antenna and associated near-field communication (NFC) electronics for interfacing to external devices, such as NFC-enabled smartphones. When the epidermal colorimetric sensing device pasted on the sweaty skin, the openings and microchannels would capture and route sweat to five different reservoirs and channels filled with quantitative colorimetric assay reagents, with the

26

Biosensors and Bioelectronics on Smartphone

647

help of capillarity and action of the natural pressure. Through enzymatic or chromogenic reactions, the colorimetric detection reservoirs enabled determinations of sweat loss, pH, glucose, lactate, and chloride. Here, with NFC system embedded on the device, the image capture and analysis software on the smartphone could be launched automatically when they were close enough, and the temperature information from the colorimetric sensor could be wirelessly transmitted to the smartphone to correct the detection results.

NFC-Based Implanted Devices NFC-enabled devices are suitable for implantable applications, because they could be designed miniaturized, without the need for replacing batteries. Fig. 6e illustrated an NFC-based bioresorbable silicon sensing system which could be implanted in the body’s abdomen, extremities, and deep brain for detecting fluid flow, motion, pH, and thermal characteristics (Kang et al. 2016). The power could be wirelessly transmitted from NFC-enabled readers, such as smartphones, to the devices in vivo. Also, the sensing data could be wirelessly transmitted outside of body. Anabtawi et al. reported an integrated system on chip (SoC) that forms the core of a long-term, battery assisted, passive continuous glucose monitor (Anabtawi et al. 2016). This device could be fully implanted subcutaneously to detect blood glucose concentration. Compared to most wearable devices that detect glucose in sweat, tear, and saliva, implanted glucose sensors could get more accurate and convincing sensing results of blood glucose concentration. Except for physiological or biochemical sensing, implanted devices could also be used for excitations, such as the applications in optogenetics. As illustrated in Fig. 6f, a flexible, thin, and millimeter-scale device was developed by Roger’s group in 2017, for wireless, programmed delivery of light into biological tissues for optogenetic experiments (Shin et al. 2017). In this study, a large loop antenna was wrapped round a box as the NFC reader for power delivery, a mouse with the device in deep brain was put inside the box. This device could be powered and then illuminate surrounding cells with lights of different wave lengths. This kind of implantable device could be widely used in postoperative treatment and health recovery.

NFC-Based Gas Sensing Tags In addition to wearable and implanted devices, NFC-based devices also performed well in environmental monitoring. For example, in 2014, researchers combined NFC with sensing technique and developed the first NFC-enabled flexible tag sensor for chemical gas detections (Azzarelli et al. 2014). In this study, single-walled carbon nanotubes (SWCNTs) were integrated into the circuitry of commercial NFC tags, to achieve portable, non-line-of-sight, and inexpensive detection of gas-phase chemicals, such as ammonia and cyclohexanone (Fig. 7a). Without batteries or redundant wires, this tag sensor was designed ultrathin and flexible with long

648

S. Li et al.

lifetime, which could be pasted on various surfaces. In the detections, an NFCenabled smartphone was closed to the tag to transmit power and receive the data. When exposed to target gases, the absorption of gas molecules would increase the resistance of SWCNTs and further influence the readability of the tag sensor. According to the change from readable to unreadable of tag sensor, the concentration range of target gases could be approximately determined. On the basis of the same principle, Liu’s group made some improvements and developed another NFCenabled tag sensor for biochemical sensing (Xu et al. 2017). As seen in Fig. 7b, the commercial NFC tag was refabricated by cutting off the oscillator circuit and connecting the breakpoints with carbonic interdigitated electrodes, which were covered with monolayer chemical vapor deposition (CVD) grown graphene for gas detection. For the NFC tag sensors, there is an On/Off threshold (Rt). Once the resistance of the electrodes increased and exceeded Rt with the exposure of target gas, the tag could not communicate normally with the smartphone, then the gas was detected. However, this approach could just detect the absence/presence of target gas by the On/Off change but unable to distinguish different concentrations. Herein, a rheostat R0 was added on the electrodes branch, which could change the On/Off

Fig. 7 NFC-based flexible tags for health-related gas sensing. (a) Wireless tag sensor for gas sensing. (Reproduced from Reference (Azzarelli et al. 2014). Copyright 2014 National Academy of Sciences). (b) Passive and wireless near-field communication tag sensor for semiquantitative biochemical detections with smartphone. (Reproduced from Reference (Xu et al. 2017). Copyright 2017 Elsevier). (c) Flexible passive NFC tag for multigas sensing. (Reproduced from Reference (Escobedo et al. 2017). Copyright 2016 American Chemical Society)

26

Biosensors and Bioelectronics on Smartphone

649

threshold. Tags with different thresholds would show different readabilities in the presence of target gas. Using an array containing five NFC tag sensors with different thresholds, the researchers achieved the accurate discrimination of ethanol at partper-million concentrations. Except for electrochemical approaches, optoelectronic technique was also used in gas sensing. Figure 7c presented a full-passive flexible multigas sensing tag for the determination of oxygen, carbon dioxide, ammonia, and relative humidity readable by a smartphone. The white LED mounted on the tag was used as common optical excitation source for all the sensors. Under the exposure of target gases, the four different color membranes around the LED would show optical response which could be read by high-resolution digital color detectors. Based on NFC technology, the antenna harvested energy to power LED, detectors, and other electronics, and enabled data transmission with a smartphone. Above all, NFC technology gives us a solution for wireless power and data transmission with the device and smartphones. It is just like a bridge, which broadens the application of smartphone-based biosensors and bioelectronics from conventional portable and wearable devices to some new areas, such as epidermal electronics, implantable devices, and flexible tags for environment monitoring.

Conclusion and Future Direction This chapter mainly presented the summary of biosensors and bioelectronics on smartphone for biochemical detections. The development of biosensor technology on smartphone can be classified into two parts: optical sensing and electrochemical sensing. Among them, smartphone-based optical biosensors usually used the integrated cameras for imaging and optical intensity measurements. These designs had the characteristics of simplicity and high miniaturization, but the reliability and accuracy were not as high as electrochemical detection. Here, we mainly reviewed the bioelectronics-based detections on smartphone, which included the amperometry sensing, potentiometry sensing, impedimetry sensing, electrochemical-LSPR sensing, and electrochemiluminescence sensing. Besides, NFC-based wireless and passive devices, especially wearable and implanted biosensors and bioelectronics, are also reviewed in detail. With powerful computational ability and easy-to-operate interface, smartphonebased biosensors and bioelectronics showed extensive application prospects in practical mobile testing, especially in areas with complex terrain and inconvenient traffic. In addition, smartphone had abundant built-in sensors and multiple signal transmission capabilities. These characteristics make it suitable for POCT, which required the shortest time to produce as many test results as possible. According to our review, the devices in electrochemical sensing system can be miniaturized to the palm size. Combining with smartphone, the whole process of electrochemical detection can be easily completed. However, these electrochemical detections still needed to be carried out with corresponding devices outside the phone. Here, smartphone is only used as a control and display analysis device. With the ultra-

650

S. Li et al.

high-speed development of smartphone, the integration of biosensor modules on smartphone is also developing rapidly. In the future, electrochemical systems based on smartphone can be realized by embedding corresponding detection modules in smartphone with universal sensor interfaces. Customizing smartphone with electrochemical sensing functions and building an integrated smartphone will greatly meet the growing individual requirements. Researches about wearable and implanted sensing on smartphone have been widely developed for biomedical applications. However, only a few of them have been commercialized, for that these devices are still in the early stage of research and development. With the market demand for specific products, such as diabetes monitoring products, multifunctional wearable and implanted equipment still need to be developed and commercialized urgently. At the same time, a large number of clinical samples should be used to prove the accuracy, repeatability, and antiinterference performances of these sensors, before they are put into clinical and commercial production. Today, modern medicine is gradually transforming from public health care to personalized diagnosis. In the initial stage of transformation, development of wearable and implanted equipment is a promising work. Furthermore, with the development of flexible electronics, smartphones can be produced totally flexible and miniaturized. They would be curved like a pencil and unfolded like a pad. Various biomedical sensors, such as electrochemical and optical sensors, can be integrated in the flexible smartphones. By then, the whole smartphone-based sensing system could be attached on human body for continuous health monitoring, with minimum limit of regular activities. Acknowledgments This work was supported by the National Natural Science Foundation of China (Grant No. 31671007, 81801793), the Zhejiang Provincial Natural Science Foundation of China (Grant No. LZ18C100001), the China Postdoctoral Science Foundation (Grant No. 2018 M630677), and the Collaborative Innovation Center of Traditional Chinese Medicine Health Management of Fujian province of China.

References Abgrall P, Gue A (2007) Lab-on-chip technologies: making a microfluidic network and coupling it into a complete microsystem – a review. J Micromech Microeng 17:R15–R49 Anabtawi N, Freeman S, Ferzli R (2016) A fully implantable, NFC enabled, continuous interstitial glucose monitor. In: Biomedical and Health Informatics (BHI), 2016 IEEE-EMBS International Conference on, IEEE 2016, pp 612–615 Aronoff-Spencer E, Venkatesh A, Sun A et al (2016) Detection of Hepatitis C core antibody by dual-affinity yeast chimera and smartphone-based electrochemical sensing. Biosens Bioelectron 86:690–696 Aymerich J, Márquez A, Terés L et al (2018) Cost-effective smartphone-based reconfigurable electrochemical instrument for alcohol determination in whole blood samples. Biosens Bioelectron 117:736–742 Azzarelli JM, Mirica KA, Ravnsbæk JB et al (2014) Wireless gas detection with a smartphone via rf communication. Proc Natl Acad Sci 111:18162–18166 Bai Y-W, Lin C-H (2014) A portable oxygen concentration detection and monitor system using a smartphone and a portable sensor module. In: Consumer Electronics-Taiwan (ICCE-TW), 2014 IEEE International Conference on, IEEE 2014, pp 129–130

26

Biosensors and Bioelectronics on Smartphone

651

Balakrishnan SR, Hashim U, Letchumanan G et al (2014) Development of highly sensitive polysilicon nanogap with APTES/GOx based lab-on-chip biosensor to determine low levels of salivary glucose. Sensors Actuators A Phys 220:101–111 Bandodkar AJ, Wang J (2014) Non-invasive wearable electrochemical sensors: a review. Trends Biotechnol 32:363–371 Bandodkar AJ, Jeerapan I, Wang J (2016) Wearable chemical sensors: present challenges and future prospects. ACS Sens 1:464–482 Barbosa AI, Gehlot P, Sidapra K et al (2015) Portable smartphone quantitation of prostate specific antigen (PSA) in a fluoropolymer microfluidic device. Biosens Bioelectron 70:5–14 Biran I, Babai R, Levcov K et al (2000) Online and in situ monitoring of environmental pollutants: electrochemical biosensing of cadmium. Environ Microbiol 2:285–290 Bisetty K (2018) Smartphone based bioanalytical and diagnosis applications: a review. Biosens Bioelectron 102:136–149 Blackburn GF, Shah HP, Kenten JH et al (1991) Electrochemiluminescence detection for development of immunoassays and DNA probe assays for clinical diagnostics. Clin Chem 37:1534–1539 Brandt O, Hoheisel JD (2004) Peptide nucleic acids on microarrays and other biosensors. Trends Biotechnol 22:617–622 Breslauer DN, Maamari RN, Switz NA et al (2009) Mobile phone based clinical microscopy for global health applications. PLoS One 4:e6320 Chang B-Y, Park S-M (2010) Electrochemical impedance spectroscopy. Annu Rev Anal Chem 3:207–229 Chen L, Zhang C, Xing D (2016) Based bipolar electrode-electrochemiluminescence (BPE-ECL) device with battery energy supply and smartphone read-out: a handheld ECL system for biochemical analysis at the point-of-care level. Sensors Actuators B Chem 237:308–317 Clavero C (2014) Plasmon-induced hot-electron generation at nanoparticle/metal-oxide interfaces for photovoltaic and photocatalytic devices. Nat Photonics 8:95–103 Conroy PJ, Hearty S, Leonard P et al (2009) Antibody production, design and use for biosensorbased applications. Semin Cell Dev Biol 20:10–26 Coskun V, Ozdenizci B, Ok K (2013) A survey on near field communication (NFC) technology. Wirel Pers Commun 71:2259–2294 Delaney JL, Doeven EH, Harsant AJ et al (2013a) Use of a mobile phone for potentiostatic control with low cost paper-based microfluidic sensors. Anal Chim Acta 790:56–60 Delaney JL, Doeven EH, Harsant AJ et al (2013b) Reprint of: use of a mobile phone for potentiostatic control with low cost paper-based microfluidic sensors. Anal Chim Acta 803:123–127 Di Martino G, Turek VA, Lombardi A et al (2017) Tracking nanoelectrochemistry using individual plasmonic nanocavities. Nano Lett 17:4840–4845 Doeven EH, Barbante GJ, Harsant AJ et al (2015) Mobile phone-based electrochemiluminescence sensing exploiting the ‘USB On-The-Go’ protocol. Sensors Actuators B Chem 216:608–613 Dutta S, Saikia K, Nath P (2016) Smartphone based LSPR sensing platform for bio-conjugation detection and quantification. RSC Adv 6:21871–21880 Escobedo P, Erenas M, Lopez-Ruiz N et al (2017) Flexible passive near field communication tag for multigas sensing. Anal Chem 89:1697–1703 Fan X, White IM, Shopova SI et al (2008) Sensitive optical biosensors for unlabeled targets: a review. Anal Chim Acta 620:8–26 Fan L, Zhao G, Shi H et al (2013) A highly selective electrochemical impedance spectroscopybased aptasensor for sensitive detection of acetamiprid. Biosens Bioelectron 43:12–18 Fan Y, Liu J, Wang Y et al (2017) A wireless point-of-care testing system for the detection of neuron-specific enolase with microfluidic paper-based analytical devices. Biosens Bioelectron 95:60–66 Forster RJ, Bertoncello P, Keyes TE (2009) Electrogenerated chemiluminescence. Annu Rev Anal Chem 2:359–385 Gao W, Nyein HY, Shahpar Z et al (2016a) Wearable sweat biosensors. In: Electron Devices Meeting (IEDM), 2016 IEEE International, IEEE 2016, pp 6.6.1–6.6.4

652

S. Li et al.

Gao W, Emaminejad S, Nyein HYY et al (2016b) Fully integrated wearable sensor arrays for multiplexed in situ perspiration analysis. Nature 529:509–514 Gao W, Saqib M, Qi L et al (2017) Recent advances in electrochemiluminescence devices for pointof-care testing. Curr Opin Electrochem 3:4–10 Geng Z, Zhang X, Fan Z et al (2017) Recent progress in optical biosensors based on smartphone platforms. Sensors 17:2449 Goode J, Rushworth J, Millner P (2014) Biosensor regeneration: a review of common techniques and outcomes. Langmuir 31:6267–6276 Gorton L, Lindgren A, Larsson T et al (1999) Direct electron transfer between heme-containing enzymes and electrodes as basis for third generation biosensors. Anal Chim Acta 400:91–108 Guo J (2016) Uric acid monitoring with a smartphone as the electrochemical analyzer. Anal Chem 88:11986–11989 Guo J (2017) Smartphone-powered electrochemical dongle for point-of-care monitoring of blood βketone. Anal Chem 89:8609–8613 Huang X, Xu D, Chen J et al (2018) Smartphone-based analytical biosensors. Analyst 143:5339– 5351 IHS T (2014) NFC-enabled cellphone shipments to soar fourfold in next five years. Antenna Systs & Technol 17:27 Ji D, Liu L, Li S et al (2017) Smartphone-based cyclic voltammetry system with graphene modified screen printed electrodes for glucose detection. Biosens Bioelectron 98:449–456 Ji D, Liu Z, Liu L et al (2018a) Smartphone-based integrated voltammetry system for simultaneous detection of ascorbic acid, dopamine, and uric acid with graphene and gold nanoparticles modified screen-printed electrodes. Biosens Bioelectron 119:55–62 Ji D, Xu N, Liu Z et al (2018b) Smartphone-based differential pulse amperometry system for realtime monitoring of levodopa with carbon nanotubes and gold nanoparticles modified screenprinting electrodes. Biosens Bioelectron 129:216–223 Jiang J, Wang X, Chao R et al (2014) Smartphone based portable bacteria pre-concentrating microfluidic sensor and impedance sensing system. Sensors Actuators B Chem 193:653–659 Joseph S, Rusling JF, Lvov YM et al (2003) An amperometric biosensor with human CYP3A4 as a novel drug screening tool. Biochem Pharmacol 65:1817–1826 Jung Y, Park H, Park J-A et al (2015) Fully printed flexible and disposable wireless cyclic voltammetry tag. Sci Rep 5:8105 Kanchi S, Sabela MI, Mdluli PS et al (2017) Smartphone based bioanalytical and diagnosis applications: a review. Biosens Bioelectron 102:136–149 Kang S-K, Murphy RK, Hwang S-W et al (2016) Bioresorbable silicon electronic sensors for the brain. Nature 530:71 Kassal P, Kim J, Kumar R et al (2015) Smart bandage with wireless connectivity for uric acid biosensing as an indicator of wound status. Electrochem Commun 56:6–10 Kato F, Minamimoto H, Nagasawa F et al (2018) Active tuning of strong coupling states between dye Excitons and localized surface plasmons via electrochemical potential control. ACS Photon 5:788 Kawawaki T, Zhang H, Nishi H et al (2017) Potential-scanning localized plasmon sensing with single and coupled gold Nanorods. J Phys Chem Lett 8:3637–3641 Kerman K, Kobayashi M, Tamiya E (2003) Recent trends in electrochemical DNA biosensor technology. Meas Sci Technol 15:R1–R11 Kim J, Banks A, Xie Z et al (2015) Miniaturized flexible electronic systems with wireless power and near-field communication capabilities. Adv Funct Mater 25:4761–4767 Kim J, Jeerapan I, Imani S et al (2016a) Noninvasive alcohol monitoring using a wearable tattoobased Iontophoretic-biosensing system. ACS Sens 1:1011–1019 Kim J, Salvatore GA, Araki H et al (2016b) Battery-free, stretchable optoelectronic systems for wireless optical characterization of the skin. Sci Adv 2:10 Koh A, Kang D, Xue Y et al (2016) A soft, wearable microfluidic device for the capture, storage, and colorimetric sensing of sweat. Sci Transl Med 8:366ra165

26

Biosensors and Bioelectronics on Smartphone

653

Lebiga E, Fernandez RE, Beskok A (2015) Confined chemiluminescence detection of nanomolar levels of H2O2 in a paper–plastic disposable microfluidic device using a smartphone. Analyst 140:5006–5011 Leikanger T, Häkkinen J, Schuss C (2017) Interfacing external sensors with Android smartphones through near field communication. Meas Sci Technol 28:044006 Li X, Shen L, Zhang D et al (2008) Electrochemical impedance spectroscopy for study of aptamer– thrombin interfacial interactions. Biosens Bioelectron 23:1624–1630 Li S, Zhang D, Zhang Q et al (2016a) Electrophoresis-enhanced localized surface plasmon resonance sensing based on nanocup array for thrombin detection. Sensors Actuators B Chem 232:219–225 Li N, Zhang D, Zhang Q et al (2016b) Combining localized surface plasmon resonance with anodic stripping voltammetry for heavy metal ion detection. Sensors Actuators B Chem 231:349–356 Li N, Lu Y, Li S et al (2017) Monitoring the electrochemical responses of neurotransmitters through localized surface plasmon resonance using nanohole array. Biosens Bioelectron 93:241–249 Li S, Liu J, Lu Y et al (2018a) Mutual promotion of electrochemical-localized surface plasmon resonance on nanochip for sensitive sialic acid detection. Biosens Bioelectron 117:32–39 Li S, Zhang D, Liu J et al (2018b) Electrochemiluminescence on smartphone with silica nanopores membrane modified electrodes for nitroaromatic explosives detection. Biosens Bioelectron 129:284–291 Li S, Zhang Q, Lu Y et al (2018c) Gold nanoparticles on graphene oxide substrate as sensitive nanoprobes for rapid L-cysteine detection through smartphone-based multimode analysis. ChemistrySelect 3:10002–10009 Lillehoj PB, Huang M-C, Truong N et al (2013) Rapid electrochemical detection on a mobile phone. Lab Chip 13:2950–2955 Lisdat F, Schäfer D (2008) The use of electrochemical impedance spectroscopy for biosensing. Anal Bioanal Chem 391:1555–1567 Liu L, Zhang D, Zhang Q et al (2017) Smartphone-based sensing system using ZnO and graphene modified electrodes for VOCs detection. Biosens Bioelectron 93:94–101 Llorente VB, Dzhagan VM, Gaponik N et al (2017) Electrochemical tuning of localized surface plasmon resonance in copper Chalcogenide Nanocrystals. J Phys Chem C 121:18244–18253 Ma L, Xu S, Wang C et al (2017) Electrically modulated localized surface plasmon around selfassembled-monolayer-covered nanoparticles. Langmuir 33:1437–1441 Macdonald JR, Barsoukov E (2005) Impedance spectroscopy: theory, experiment, and applications. History 1:1–13 McCracken KE, Yoon J-Y (2016) Recent approaches for optical smartphone sensing in resourcelimited settings: a brief review. Anal Methods 8:6591–6601 Morak J, Kumpusch H, Hayn D et al (2012) Design and evaluation of a telemonitoring concept based on NFC-enabled mobile phones and sensor devices. IEEE Trans Inf Technol Biomed 16:17–23 Moreira CM, Pereira SV, Raba J et al (2018) Based enzymatic platform coupled to screen printed graphene-modified electrode for the fast neonatal screening of phenylketonuria. Clin Chim Acta 486:59–65 Muzyka K (2014) Current trends in the development of the electrochemiluminescent immunosensors. Biosens Bioelectron 54:393–407 Nemiroski A, Christodouleas DC, Hennek JW et al (2014) Universal mobile electrochemical detector designed for use in resource-limited applications. Proc Natl Acad Sci 111:11984–11989 Nyein HYY, Gao W, Shahpar Z et al (2016) A wearable electrochemical platform for noninvasive simultaneous monitoring of Ca2+ and pH. ACS Nano 10:7216–7224 Parrilla M, Cánovas R, Jeerapan I et al (2016) A textile-based stretchable multi-ion potentiometric sensor. Adv Healthc Mater 5:996–1001 Quesada-González D, Merkoc¸i A (2017) Mobile phone-based biosensing: an emerging “diagnostic and communication” technology. Biosens Bioelectron 92:549–562

654

S. Li et al.

Rizwan M, Mohd-Naim NF, Ahmed MU (2018) Trends and advances in electrochemiluminescence nanobiosensors. Sensors 18:166 Rose DP, Ratterman ME, Griffin DK et al (2015) Adhesive RFID sensor patch for monitoring of sweat electrolytes. IEEE Trans Biomed Eng 62:1457–1465 Sempionatto JR, Nakagawa T, Pavinatto A et al (2017) Eyeglasses based wireless electrolyte and metabolite sensor platform. Lab Chip 17:1834–1842 Shen L, Hagen JA, Papautsky I (2012) Point-of-care colorimetric detection with a smartphone. Lab Chip 12:4240–4243 Shin G, Gomez AM, Al-Hasani R et al (2017) Flexible near-field wireless optoelectronics as subdermal implants for broad applications in optogenetics. Neuron 93:509–521. e503 Singh R, Mukherjee MD, Sumana G et al (2014) Biosensors for pathogen detection: a smart approach towards clinical diagnosis. Sensors Actuators B Chem 197:385–404 Smith ZJ, Chu K, Espenson AR et al (2011) Cell-phone-based platform for biomedical device development and education applications. PLoS One 6:e17150 Sood VR, Gururajan R, Hafeez-Baig A et al (2018) Adoption of mobile devices in the Australian healthcare: A conceptual framework approach. IGI Global, pp 954–977 Stedtfeld RD, Tourlousse DM, Seyrig G et al (2012) Gene-Z: a device for point of care genetic testing using a smartphone. Lab Chip 12:1454–1462 Steinberg MD, Kassal P, Kereković I et al (2015) A wireless potentiostat for mobile chemical sensing and biosensing. Talanta 143:178–183 Steinberg MD, Tkalcec B, Steinberg IM (2016) Towards a passive contactless sensor for monitoring resistivity in porous materials. Sens Actuator B Chem 234:294–299 Sun A, Wambach T, Venkatesh A et al (2014) A low-cost smartphone-based electrochemical biosensor for point-of-care diagnostics. IEEE Biomed Circuits Syst Conf 2014:312–315 Sun AC, Yao C, Venkatesh A et al (2016) An efficient power harvesting mobile phone-based electrochemical biosensor for point-of-care health monitoring. Sensors Actuators B Chem 235:126–135 Vashist SK, Mudanyali O, Schneider EM et al (2014) Cellphone-based devices for bioanalytical sciences. Anal Bioanal Chem 406:3263–3277 Védrine C, Leclerc J-C, Durrieu C et al (2003) Optical whole-cell biosensor using Chlorella vulgaris designed for monitoring herbicides. Biosens Bioelectron 18:457–463 Velasco-Garcia MN, Mottram T (2003) Biosensor technology addressing agricultural problems. Biosyst Eng 84:1–12 Wang J (2000) Survey and summary: from DNA biosensors to gene chips. Nucleic Acids Res 28:3011–3016 Wang J (2002) Electrochemical nucleic acid biosensors. Anal Chim Acta 469:63–71 Wang J, Rivas G, Cai X et al (1997) DNA electrochemical biosensors for environmental monitoring. A review. Anal Chim Acta 347:1–8 Wang X, Gartia MR, Jiang J et al (2015) Audio jack based miniaturized mobile phone electrochemical sensing platform. Sensors Actuators B Chem 209:677–685 Wang C, Nie X-G, Shi Y et al (2017a) Direct plasmon-accelerated electrochemical reaction on gold nanoparticles. ACS Nano 11:5897–5905 Wang X, Lin G, Cui G et al (2017b) White blood cell counting on smartphone paper electrochemical sensor. Biosens Bioelectron 90:549–557 Xu Z, Jiang J, Wang X et al (2016) Large-area, uniform and low-cost dual-mode plasmonic nakedeye colorimetry and SERS sensor with handheld Raman spectrometer. Nanoscale 8:6162–6172 Xu G, Zhang Q, Lu Y et al (2017) Passive and wireless near field communication tag sensors for biochemical sensing with smartphone. Sensors Actuators B Chem 246:748–755 Xu D, Huang X, Guo J et al (2018) Automatic smartphone-based microfluidic biosensor system at the point of care. Biosens Bioelectron 110:78–88 Yao C, Sun A, Hall D A (2015) Efficient power harvesting from the mobile phone audio jack for mHealth peripherals. Global Humanitarian Technology Conference (GHTC), 2015 IEEE, pp 219–225

26

Biosensors and Bioelectronics on Smartphone

655

Yao Y, Li H, Wang D et al (2017) An electrochemiluminescence cloth-based biosensor with smartphone-based imaging for detection of lactate in saliva. Analyst 142:3715–3724 Yao Y, Jiang C, Ping J (2019) Flexible freestanding graphene paper-based potentiometric enzymatic aptasensor for ultrasensitive wireless detection of kanamycin. Biosens Bioelectron 123:178–184 Zhang D, Liu Q (2016) Biosensors and bioelectronics on smartphone for portable biochemical detection. Biosens Bioelectron 75:273–284 Zhang L, Yang W, Yang Y et al (2015a) Smartphone-based point-of-care testing of salivary αamylase for personal psychological measurement. Analyst 140:7399–7406 Zhang D, Jiang J, Chen J et al (2015b) Smartphone-based portable biosensing system using impedance measurement with printed electrodes for 2, 4, 6-trinitrotoluene (TNT) detection. Biosens Bioelectron 70:81–88 Zhang D, Lu Y, Jiang J et al (2015c) Nanoplasmonic biosensor: coupling electrochemistry to localized surface plasmon resonance spectroscopy on nanocup arrays. Biosens Bioelectron 67:237–242 Zhang D, Lu Y, Zhang Q et al (2016) Protein detecting with smartphone-controlled electrochemical impedance spectroscopy for point-of-care applications. Sensors Actuators B Chem 222:994–1002 Zhu H, Yaglidere O, Su T-W et al (2011) Cost-effective and compact wide-field fluorescent imaging on a cell-phone. Lab Chip 11:315–322

Part IV Biosensor, Market, and Innovation

Business Models for Biosensors in the Food Industry

27

Rosa Caiazza and Barbara Bigliardi

Contents The Food Industry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Innovation in the Food Industry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Technological Innovations: Product and Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Non-technological Innovations: Marketing and Networking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Business Model and Business Model Innovations: What Do They Mean? . . . . . . . . . . . . . . . . . . . . Business Model Innovation in the Food Industry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . New Trends and Future Challenges in Food Industry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Business Model Innovation Enabled by Biosensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

660 660 662 663 664 667 668 670 674 675

Abstract

The development of biosensor technology has made available online quality control of food production, reducing the cost of food production and improving safety and food quality. Consequently, the introduction of biosensors in the agrofood industry has changed business models of firms operating in both the agriculture and agro-food industry. This chapter aims to evidence existing trends in both technological and non-technological innovations and new trends and future challenges of the food industry. Keywords

Food industry · Business models · Firms’ value chain · Innovation · biosensors R. Caiazza (*) Parthenope University, Naples, Italy e-mail: [email protected] B. Bigliardi Department of Engineering and Architecture, University of Parma, Parma, Italy e-mail: [email protected] © Springer Nature Switzerland AG 2022 G. Thouand (ed.), Handbook of Cell Biosensors, https://doi.org/10.1007/978-3-030-23217-7_17

659

660

R. Caiazza and B. Bigliardi

The Food Industry Food microbiologists are constantly seeking rapid and reliable automated systems for the detection of biological activity. Biosensors provide sensitive systems that can be used to detect unwanted microbial activity or the presence of a biologically active compound, such as glucose or a pesticide. The development of biosensor technology has made available online quality control of food production, reducing cost of food production and improving safety and food quality. Consequently, introduction of biosensors can modify business models of firms operating in both the agriculture and agro-food industry. Agriculture is highly dependent on natural resources such as land, soil, and water. These conditions affect global production, investment, and trade of each region of the world. Thus the production of agricultural commodities requires investment in technology and policies for managing natural resources. Important agricultural inputs include seeds, chemicals, fertilizers, machinery, and tools. For some of these agricultural inputs, microbiologists play an important role in supporting technological improvements and knowledge about the use of biosensors. Given the important role that research and development (R&D) play for agricultural productivity, both private and public organizations have to invest in it. Technological development and basic R&D encompass the application of biotechnologies, improvements in agricultural resource management, reductions in the use of pesticides and fertilizers, and support measures for sustainable farming (Audretsch and Caiazza 2016; Brockhoff 1992; Caiazza 2015; Teece 1998; Caiazza 2016b). Public institutions play a crucial role in providing research found for firms involved in agro-food industry. Specifically, supporting innovations in biosensors, public institutions can modify business models of firms operating in many industries involved in the production, processing, and inspection of solely food products made from agricultural commodities. Such firms operate along a value chain that goes from suppliers of inputs to farmers and other agricultural producers and service providers, to processors of agricultural goods, to trading companies dealing with agricultural commodities and retailers. Agricultural value chains involve many firms operating in both agriculture and agriculture-related industries (i.e., food processors/manufacturers, retailers, traders, and suppliers of inputs) that add value to the whole chain. This could range from being an input supplier to farmers, engaging in harvesting operations, transportation, processing, marketing, and retailing. Consequently, the introduction of biosensors can change the business model of each firm of the value chain and their reciprocal interaction in creating the whole value.

Innovation in the Food Industry Innovation in the food industry comes in different forms. Different approaches have been proposed in the literature based on the type of innovation and on the degree of novelty of innovation (Baregheh et al. 2012).

27

Business Models for Biosensors in the Food Industry

661

A general way to classify innovation that is according to the type of innovation distinguishes innovation according to the outcome of the innovation process. Specifically, Clarysse and Utterhaegen (1998) distinguished between technological and non-technological innovations. Furthermore, they divided these two categories into product and process innovations and organizational and market innovations (Bigliardi and Dormio 2009). Classifications based on the type of innovation have been proposed or applied also in the food context. As stressed by Bigliardi and Galati (2013b), food innovations may occur throughout all parts of the food system. They thus proposed an even too precise classification of food innovations: (i) new food ingredients and materials (from the early purification of vegetable oils and raw sugar to the present-day development of low-calorie fat substitutes and sweetener); (ii) innovations in fresh foods (including new types or new varieties of fresh foods, as well as innovation in process, e.g., atmospheric and temperature control during storage and transport), new methods of food preservation (from the traditional, heat, cold, chemical, dehydration, and mechanical methods to the more innovative methods including the use of electric fields, magnetic fields, ultrasound, pulsed light, and irradiation), or new types of packaging; (iii) new food process techniques (such as the realization of complex food formulations combining ingredients that are designed to give specific flavors, structures, and nutrition); (iv) innovations in food quality (e.g., the ready-to-eat poultry products by Moira Mac’s – CSIRO); (v) new packaging methods (e.g., from the early cellophane packs to the PET bottle); and (vi) new distribution or retailing methods (e.g., McDonald’s fast foods). Most of the research follows the classification in product and process innovation (Avermaete et al. 2004; Brewin et al. 2009; Capitanio et al. 2010). A product innovation is any good that is perceived by someone as new that results in a new or improved product and that offers consumer an advantage and a higher utility compared to already existing products. Product innovations usually focus on the market or demand-oriented issues, and in the majority of the case, they focus in application fields known to the food company that innovates. The improvement of useful properties of the product, the increasing of quality, and the changing of design are only some examples of important product innovation attributes (Pleschak and Sabisch 1996). Functional foods are a typical and successful example of product innovation in the food industry (Bigliardi and Galati 2013a). In a similar way, a process innovation is an adaptation of existing production lines as well the installation of an entirely new infrastructure and the implementation of new technologies that allow the creation of new products (Bigliardi and Dormio 2009). Process innovations are usually more wide-ranging (i.e., they do not refer necessarily to a specific field), and their main aims are the improvement of product quality and the improvement of the production process in terms of time, cost, and flexibility. To cite only some, high-pressure technology and ultrafiltration techniques are typical examples of process innovations introduced in the food industry (Menrad and Feigl 2007). However, Brewin et al. (2009) highlighted that often product and process innovations go hand in hand, thus making difficult a clear distinction between them. Potato chips provide an example of the interdependence of product and process innovations: the development of the extruder technique (the process innovation) was a prerequisite to produce potato chips (the product innovation). As far as non-

662

R. Caiazza and B. Bigliardi

technological innovation is concerned, organizational innovation refers to changes in marketing, purchases, sales, administration, management, and staff policy, while market innovation includes the exploitation of new territorial markets and the penetration of new market segments within existing markets. Grunert et al. (1997) added a third category of innovation to the product/process orientation, considering also the market orientation. A market innovation can be defined as the “exploitation of new territorial markets and the penetration of new market segments within existing markets” (Avermaete et al. 2004). High levels of market innovation usually characterize R&D intense food products, such as functional foods (Bigliardi and Galati 2013a). In addition to product, process, and market innovations, Gellynck and Vermeire (2009) proposed also packaging innovation: food packaging was simply a container to hold food, while in recent years it can play an active role in food quality. An example of packaging innovation is represented by active packages (Risch 2009). More recently, Baregheh et al. (2012) adopted in their study the classification originally proposed by Francis and Bessant (2005) identifying position and paradigm innovation in addition to the classic product and processed ones. Specifically, position innovation refers to changes in how a specific product or process is perceived symbolically and how they are used, while they refer to paradigm innovation as any “changes in the underlying mental models which frame what organization does.” To understand what is a position innovation in the food industry, the glucose-based drink Lucozade can be considered: it originally aimed at helping children and invalids in convalescence and was thus associated with sickness. SmithKline Beecham abandoned the original concept of sickness and relaunched the product as a health drink aimed at the growing fitness market. In such a way, its new aim was the enhancement of performance aid to healthy exercise. Recent examples of paradigm innovation in the food industry include the repositioning of drinks like coffee and fruit as premium designer products. Moreover, even if usually it is not considered within a specific classification, another type of innovation to be considered also within the food context is business model innovation (BMI). BMI is a different type of innovation that differentiates from the technological ones (Zott and Amit 2002; Comes and Berniker 2008) because while product and process innovations can easily be copied, business model innovations are difficult for competitors to follow. This is due to two main reasons: as first, they require time and effort and as second a new business model is not a general one but has to fit a company’s long-term strategy. However, business models are fundamentally linked to technological innovation; indeed, technology development can facilitate new business models (Baden-Fuller and Haefliger 2013). In the following, a brief description of these types of innovations is provided.

Technological Innovations: Product and Process The primary objective of the food processing industry is to offer good and healthy food. To this end, microbiologists can be involved in realizing biosensors aimed at creating healthier products that are suitable for different needs and lifestyles.

27

Business Models for Biosensors in the Food Industry

663

Biosensors can help to improve the quality of protected denomination products. They can improve the level of both raw and processed goods. At the same time, biosensors can support the development of functional food market enhancing products’ quality and innovating the value of territorial food traditions to meet the changing needs and consumer lifestyles. Investments in research, development, and innovation in biosensors can change the business model of firms producing traditional foods (pasta, cheese, wine, oil), evolved foods (frozen foods, ready sauces, fresh condiments), and new products (foods with a high health and service content). Both radical and incremental innovations can improve food safety, sustainable production, social responsibility, food consumption, and lifestyles addressing main consumer needs. The main consumption trends concern the naturalness and freshness of the products, the texture and the organoleptic contents, the nutritional and health values, the functionality, the opportunity, and the place of consumption (Caiazza et al. 2015; Caiazza and Stanton 2016). Innovative companies have to change their business model according to such a trend for developing new products, adopting more advanced production systems, and improving technologies that guarantee greater productivity. The new business has to be based on the ability to integrate technologies incorporated into advanced machinery to its production processes.

Non-technological Innovations: Marketing and Networking Biosensors can also support activities of companies that aim to put in place marketing innovations, helping consumers to make choices based on detailed information on products, raw materials, and conservation methods. These innovations must relate to packaging that plays a key role in guaranteeing food quality and safety; protecting the integrity of the product during transport, distribution, and consumption; and conveying brand values but also providing nutritional and nutritional information (Battcock and Azam-Ali 1998; Caiazza and Volpe 2017). The improvement of the nutritional characteristics of food products is achieved through a change in their composition, within as much as technologically possible and accepted by the consumer and trying to maintain the organoleptic characteristics of the product (flavor, consistency, and shelf life). Agro-food firms have a greater propensity to invest in product design and packaging. Some of them choose as a strategy of diversification and improvement of the production the development of innovations in design and the adoption of new solutions in the field of packaging (Haard et al. 1999; Caiazza and Volpe 2015). The food industry, however, is not self-sufficient for the raw materials of some strategic supply chains for which it must put in place organizational innovations aimed at optimizing relations with foreign suppliers through integrated management systems that involve participation in the upstream activities of the supply chain. In addition, companies operating in the agro-food industry tend to internationalize through export processes that involve mainly foreign buyers, as the phenomenon of outsourcing of production activities abroad is almost inexistent. The growth

664

R. Caiazza and B. Bigliardi

strategy of many companies must, therefore, be oriented toward strengthening the markets in which they already operate and entering new markets through organizational innovations that provide for the establishment of their own sales structures abroad.

Business Model and Business Model Innovations: What Do They Mean? Before discussing in the next section on food business model innovation, a more concrete definition of what is, generally speaking, a business model (BM) and a business model innovation (BMI) is worthwhile. A BM can be defined as “a unit of analysis to describe how the business of a firm works” (Frankenberger et al. 2013, p. 6). More in detail, Magretta (2002, p. 91) stated: “business models describe, as a system, how the pieces of a business fit together.” Historically, Bellman et al. (1957) first used the term BM in their paper about business games for training purposes, reflecting a simplification of reality aimed at educating future managers on technology. Three years later, Jones (1960) used for the first time the term in the title of his paper (even if revealing an arbitrary use because the term was never repeated in the text of the same paper). In the late 1990s, it emerged as a buzzword in the popular press, and since then it has gained growing attention from both practitioners and academics, as clearly emerge from Fig. 1. A research on Scopus with the keyword “business model” within article title, abstract, and keywords provides more than 22.100 results in different subject areas from “computer science” to “business, management and accounting” and “engineering” to “agricultural and biological sciences.” BM may be the tools managers use to design, implement, operate, change, and control their businesses (Wirtz et al. 2010).

Fig. 1 Trends of publications dealing with BM since the 1990s. (Data retrieved on July 2018, 22)

27

Business Models for Biosensors in the Food Industry

665

Notwithstanding the growing interest toward BMs in the literature, a general definition of BM does not still exist. Indeed, Morris et al. (2005) in their study found more than 30 definitions of BMs and propose the following definition “A business model is a concise representation of how an interrelated set of decision variables in the areas of venture strategy, architecture, and economics are addressed to create sustainable competitive advantage in defined markets” (Morris et al. 2005, p. 727). As a matter of the fact, while some researchers see BM as part of the strategy lexicon (e.g., Zott and Amit 2007; Osterwalder and Pigneur 2010), others see BM as separable from technology and strategy (Baden-Fuller and Morgan 2010; Teece 2010). Similarly, also different classifications of BMs exist. Among others, Sayan Chatterjee (2013) proposed three types of BMs, namely, efficiency-based BM, value-based BM, and network efficiency BM. Efficiency-based BMs are usually adopted by price takers in a highly competitive market and rely on human or capital resources to produce commodities. Conversely, value-based BM refers to those companies that position their output as a “want” item driven by objective or subjective/perceived (such as, usually, the food case) values. The third type of BM (i.e., the network efficiency one) extends the efficiency across the entire value-added chain by considering all the actors involved in the supply chain. Various BMs have been proposed in the literature. For instance, Osterwalder et al. (2005) proposed the BM canvas that consists of nine building blocks, namely: (i) customer segments (i.e., who are the most important customers for whom the company creates value); (ii) the value proposition (i.e., what product or service is offered and what value is delivered to the customer); (iii) value channels (i.e., the activities needed to develop, produce, and deliver the firm’s products/services); (iv) customer relationships (i.e., the relationships to be established with the customer); (v) revenue streams (i.e., for what value to the customer pays and how much does he pay); (vi) resources (i.e., which key resources are required); (vii) activities (i.e., which are the key activities to be carried out); (viii) partnership (i.e., which are the key partners and which key activities do they perform); and (ix) cost structure (i.e., what are the costs associated with the BM). Richardson (2008) suggested a concentrated framework, based on the one proposed by Osterwalder et al. (2005), composed by three elements, namely, (i) the value proposition; (ii) the value creation and delivery system; and (iii) the value capture system. More recently, Frankenberger et al. (2013) adopted a four-dimensional conceptualization, namely, the who (“who is the consumer?”), the what (“what is offered to the target customer?”), the how (“which processes and activities are carried out?”), and the why (“why the BM is financially viable?”). A similar approach was proposed by Mandour and Brees (2013) with the BM wheel based on four primary drivers: (i) customer segment and relation (to whom do we deliver value), (ii) proposition and channels (what is the offer and how is it delivered), (iii) revenue model (how do we capture value), and (iv) resources and partners (what and who do we need). According to this wheel, to develop a new successful and competitive BM, a company has to choose one of these primary drivers while thinking critically about the other ones. As far as BMI is concerned, since the mid-1990s, interest in it has increased from both a practical and a theoretical point of view (Osterwalder and Pigneur 2010).

666

R. Caiazza and B. Bigliardi

In general, “new ways of organizing businesses are often referred to as BMIs” (Teece 2010). Paraphrasing Matzler et al. (2013) “business model innovation occurs when a company increases simultaneously customer value and creates a new value creation and revenue model that allows the company to capture some of the value created in a new way.” Such a definition emphasizes the system-level nature of BMI, which contrarily to previous types of innovation involves the firm and its environment as a whole. Most of BMIs are consequent to the disruptive changes motivated by new technology, such as ICT in general and the Internet in particular, but they can also occur without technology development. Scholars agree in acknowledging that BMI is a key source of competitive advantage (e.g., Baden-Fuller and Morgan 2010; Björkdahl 2009; Chesbrough and Rosenbloom 2002; McGrath 2010), more (in terms of profitability) than pure product or process innovators (BCG 2008). However, despite the perceived importance of BMI, the research base in that field is thinner than that in the field of BM. The same search on Scopus with the keyword “business model innovation” returns about 950 results. Most of these studies deal with the strategic change antecedents (Doz and Kosonen 2010), the barriers to the adoption of BMIs (Bouchikhi and Kimberly 2003), or the risks that companies have to face when adopting a BMI (Girotra and Netessine 2013). Other issues remain still un- or under-explored (Schneider and Spieth 2013), such as the stages that companies go through to implement a BMI. As for this latter issue, a step ahead has been done by Frankenberger et al. (2013). Specifically, they developed a framework describing the phases of BMI and identified the key challenges in each stage. The four phases identified by the authors are (i) initiation (i.e., the analysis of the ecosystem), (ii) ideation (i.e., the generation of new ideas), (iii) integration (i.e., the building of a new BM), and (iv) implementation (i.e., the realization of the new BM). A framework incorporating the main frameworks proposed in the literature about BMs and BMIs, and previously described, is depicted in Fig. 2. Specifically, it depicts BMIs as a process made of different phases (the four phases proposed by Frankenberger et al. (2013)). A correspondence can be found between the nine building blocks identified by Osterwalder and Pigneur (2002) and the four dimensions identified by Frankenberger et al. (2013). Indeed, in answering the question “who is the customer?”, a BM takes into consideration customer segments as well as relationships to be established with them (customer relationship). In a similar way, the “what” dimension includes the value proposition, while the “why” dimension includes the revenue streams and the cost structure. Finally, the remaining building blocks (i.e., value channels, resources, activities, and partnership) all refer to the “how” dimension. Moreover, a similar correspondence can be proposed also between the four phases and the four dimensions: typically, the idea initially determines the “what” and/or “who” dimension of the future BM, whereas the revenue model (i.e., the “why” dimension) and value chain architecture (i.e., the “how” dimension) are added during integration phase. The double arrow between phases 3 (integration) and 4 (implementation) indicates that iterations between these phases can be observed in order to adjust the BM. Finally, the figure also shows the main challenges for each phase. Challenges exist because, usually, a new BM

27

Business Models for Biosensors in the Food Industry

667

1. Initiation

• activities: understanding and monitoring of the value chain • challenges: understanding the needs of the players, identification of change drivers • dimensions: who? (customer segments, customer relationship) 2. Ideation

• activities: transforming of oppurtunities into concrete ideas for new BMs • challenges:overcoming the current business logic, thinking in BMs and not solely in NPD • dimensions: who? (customer segments, customer relationship) what? (value proposition) 3. Integration

• activities: transforming ideas into a complete and viable BM • challenges: integrating all pieces of the new BM, involving and managing the partners • dimensions:why? (revenue streams, cost structure) how? (value channels, resources, activities, partnership) 4. Implementation

• activities: implementing the new BM • challenges: overcoming internal resistance, managing organizational change, managing the chosen implementation approach • dimensions: who? what? why? how? Fig. 2 The main phases of BMI. (Authors own creation)

requires changes to different areas within the firm. Surely, the most obvious obstacle is the management of organizational change, and it is thus important to be able to explain how the new BM can help the company.

Business Model Innovation in the Food Industry When the first research appeared on BMs and BMIs, they have been mainly focused on the industry such as the information technology or the biotechnology one. Conversely, research about BMs or BMI in the food sector has generally received little attention (Ulvenblad et al. 2014). However, because of the increased worldwide competition and advanced technological developments that occurred also in the food sector, also food entrepreneurs have started moving toward a more strategic and innovative perspective. Searching for new growth opportunities, innovative entrepreneurial activities in the food context have emerged in recent years. These

668

R. Caiazza and B. Bigliardi

companies, realizing that to be competitive mere food product or process innovations are no more sufficient, but they must move into more radical food BMIs, have changed their BMs. Consequently, BMIs are becoming increasingly critical in practice (Bucherer et al. 2012). In the food context, the main BMIs are referred to food insecurity, obesity, food distribution, sustainability, and improvements to nutritional qualities (Ulvenblad et al. 2014). In the context investigated in the chapter, the first research available in the literature relates to developmental areas in the world (Beuchelt and Zeller 2016). Then, examples of the nineteenth century appeared, such as the case of a Swift Company in America which, by means of local slaughters, railways, and the new refrigerated solutions, decreased the shipping time for live cattle (Teece 2010). Furthermore, researchers such as Markowska et al. (2011) proposed a research about entrepreneurship in the rural environment focusing on BMs in the food production value chain such as restaurants. More recently, the attention of researchers has moved toward studies of BMs addressing sustainable development and taking a value-added approach (e.g., Short et al. 2014; Stubbs and Cocklin 2008). Consequently, the new BMs proposed in the literature are different, ranging from smartphone applications for consumers to rate sources and nutrition of food products, to food “hubs” increasing opportunities for producer aggregation, through initiatives for alternative food distribution or waste reduction. In the last couple of decades, food industry faces various challenges, and its credibility was heavily challenged by a number of food crises, among others, the bovine spongiform encephalopathy (BSE), known also as “mad cow disease,” dioxin in chicken feed, food-and-mouth disease (FMD), the use of genetically modified (GM) crops in foods, and so on (Aung and Chang 2014). One of the main challenges the food industry has to face with refers to the detection, with quick and cost-effective methods, and the presence of allergenic components and pathogens in the food products. The modern consumer requires food products to fit a certain level of expected quality, more than other any products. Consequently, food companies were thus led to develop new principles providing the consumer with real quality guarantees. Generally, chemical and microbiological analyses are realized periodically by trained operators, which are expensive and require time for the analysis. Biosensors can be used to overcome these disadvantages: they offer rapid, nondestructive, and affordable methods for quality control and have in general the potential to produce an analytical revolution to resolve the challenges in the food industry.

New Trends and Future Challenges in Food Industry The agro-food industry is characterized by some structural gaps that restrain firms’ ability to compete globally. The main factor in limiting the development of firms operating in agriculture and agro-food industry is the excessive fragmentation of the production, which is added to infrastructural, logistical, and distributive deficiencies, the excessive production costs starting from the energy, and the poor quality services

27

Business Models for Biosensors in the Food Industry

669

they offer for businesses, finance, and credit. A strong impulse to innovations would help to improve firms’ competitiveness (Gassmann et al. 2010; Caiazza 2016a). However, R&D of a single firm is not sufficient if not supported by a regulatory environment conducive to business development. Harmonization with the everchanging national regulatory framework and a homogeneous implementation of the rules on the territory are fundamental concepts to guarantee the correct development of efficient industrial policies. Consequently, the issues of security, the rationalization of controls, the lightening of bureaucracy, the sustainability of food production, and the need for common rules on supply conditions in international commodity markets are crucial for the development of main sectors of excellence. Most of the innovative food companies do not consider the contribution of external actors decisive for the management of innovation processes, even if the propensity toward the outside is relatively higher than in the rest of manufacturing. In particular, in the innovation processes the formal relationships of collaboration with the scientific community are still infrequent, while informal cooperation agreements with universities and public research organizations are more frequent. Moreover, a stable collaboration of research and innovation between food companies and external parties, able to satisfy the technological transfer needs of the sector, would be desirable. On the other hand, informal relations are more widespread, especially along the supply chain between suppliers and customers. Through collaboration, the realization of technological and non-technological innovations would favor the process of strategic growth of small- and mediumsized companies on the markets. The implementation of the innovations necessary to strengthen the companies in the agro-food sector cannot be left open to the single small and medium firm. A systemic growth objective can only be reached through the support of local and national institutions. Policy-makers have to reinforce policies aimed at reduce barriers for exports and support the fight against counterfeiting. Policy-makers have also to encourage the development of business models focused on sustainable production and consumption that would make agrofood firms able to face future needs of population. With this in mind, the aim is to ensure quality products and guarantee supplies of raw materials respecting the environment and the health of future generations (Teece 1986; Caiazza 2017). The sustainable development of agro-food companies must also consider the possibility of extending the reuse of waste by-products in the production of animal feed, in the production of bioenergy, in the cosmetic and pharmaceutical industry, and in the production of fertilizers. Significant progress must also be made on the more efficient use of energy and water in all stages of production of value chain. Energy efficiency is an important driver of industrial competitiveness and environmental protection. Generally, the food industry is characterized by a relatively low energy impact compared to other sectors (i.e., mechanical). In industrialized countries, an increasing number of people have health problems because of their lifestyle, which mainly includes diet, physical activity, and a set of mental and emotional stresses. Lifestyle has undergone profound changes in a relatively short period of time. In fact, in the last decade, social and economic changes due to globalization have had a strong impact on the organization and

670

R. Caiazza and B. Bigliardi

functioning of our society, leading to pervasive and consistent changes in eating habits and behaviors. At the same time, there has been a change in the epidemiological scenario characterized by a dramatic decline in acute and infectious diseases which corresponds to an equally dramatic increase in chronic diseases and in particular to diseases associated with diet (such as obesity, diabetes, hypertension, and some types of cancer). To address their complexity, it is therefore necessary to use multidisciplinary strategies combining advanced technologies capable of integrating different parameters to improve knowledge on the main biological mechanisms responsible for diseases. This may encourage further segmentation of the market in response to the continued growth of niches such as the elderly, vegans, malnourished, athletes, or population subgroups with specific nutritional needs. To guarantee food safety and safeguard the agro-food industry from recurrent crises, policy-makers have adopted the global strategy of intervention from field to table security aimed at ensuring a high level of health protection on human rights and the protection of consumer interests while ensuring the effective functioning of the internal market. The legal framework on food has been gradually developing, following some fundamental principles, in order to achieve the freedom of circulating food products in compliance with the general principles and safety requirements and to achieve the objective of being ranked first in the world as regards public health and consumer protection. The dangers of microbiological nature, globally and for the rapid consequences generated, are a priority, both in view of their immediate impact on the health of consumers and for the food crises they are able to cause (Ming-Yeu 2012; Caiazza and Ferrara 2016). Other recognized diseases of food origin include those caused by toxins or harmful chemicals that have contaminated food. The chemical risk, although of less immediate and general impact on health and on the choices of the consumers, represents a problem of crucial importance for the agro-food sector and for long-term health problems. Not all chemicals that contaminate food are residues of active ingredients used in production processes or come from industrial machinery and technologies used. Some toxic substances can transfer from the environment to the food chain; others are products of the metabolism of particular fungi and algae. Other toxic chemicals can be formed during food processing. Food security opens up important challenges to the world of research and to the world of food production and marketing. Companies must be able to maintain their competitiveness and allow technological innovation in full compliance with the requirements, directing production systems toward technologies and practices able to prevent problems, also taking into account the protection of the health and well-being of animals, plant health, and the environment.

Business Model Innovation Enabled by Biosensors The challenges and crises described above represented the starting point for a new approach regarding the management of food products. Indeed, companies’ objectives moved toward safety, legality, consistency, and consumer acceptability

27

Business Models for Biosensors in the Food Industry

671

(Hennink et al. 2011). Different approaches were developed to achieve these objectives, such as traceability systems, known as “the ability to track any food, feed, food-producing animal or substance that will be used for consumption, through all stages of production, processing and distribution, applied both upstream (where does this product come from?) as well as downstream (where did this product go to?)” (Regulation EC n 178/2002, European Commission 2002). More recently, biosensors were introduced for the rapid detection of pathogens, allergens, as well as pesticide residues in food (Murugaboopathi et al. 2013). The term “biosensor” was coined by Cammann (1977), and its definition was introduced by IUPAC as follows: “Biosensors are devices which use a biological recognition element retained in direct spatial contact with transduction system.” Biosensors are analytical devices that can detect, record, and/or transmit data or information about biochemical reaction and that convert a biological response into an electrical signal proportional to the concentration of a specific analyte or analytes. They usually comprise two main components: a bioreceptor (i.e., biological materials such as enzymes, metabolites, hormones, aiming at recognizing of the target analyte) and a transducer (that can be in several forms such as optical, electrochemical, and acoustic or voice and that convert biochemical signals into electrical response). Research on biosensors is booming around the world: a search on Scopus with the keyword “biosensor*” returns more than 4.000 results only in 2018. Their area of applications ranges from medical to agricultural and food. In the food industry, biosensors can be divided into large multi-analyzers, benchtop portable instruments, and one-shot disposable sensors, and they are used for mainly two purposes. As first, enzyme biosensors are mainly used in liquor and beverage industry for the detection or measurement of carbohydrates from alcohol, amino acids, amines, amides, phenol, etc. As second, biosensors are used in food industry for the direct and indirect detection of microorganisms. In general, while traditional techniques present some limitations, mainly related to human fatigue and to the fact that they are expensive and time-consuming, biosensors are recognized as a cost-effective alternative (Scognamiglio et al. 2014). Different examples of biosensors’ application in the food industry have been presented in the literature (Adley 2014; Mehrotra 2016). For example, enzymatic biosensors based on cobalt phthalocyanine are used to monitor the aging of beer during storage, highlighting a good capability to monitor the aging of beer. Similarly, the presence of Escherichia coli in vegetables is measured by detecting variation in pH caused by ammonia by means of potentiometric alternating biosensing systems. Enzymatic biosensors, based on a screen-printed carbon electrode, are in the dairy industry to quantify pesticides in milk. Biosensors are also adopted in fermentation industries, where process safety and product quality are fundamental, to monitor the presence of products, biomass, enzyme, antibody, or by-products of the process and to indirectly measure the process conditions. Biosensors (such as glutamate biosensor) are also employed in ion exchange retrieval, in order to carry out detection of change of biochemical composition. In addition to these examples of biosensors’ adoption, also BMIs enabled by biosensors can be identified in the literature. Among all, an example of BMI involves

672

R. Caiazza and B. Bigliardi

the packaging. Packaging has an essential role in the food industry because it deals with the safety and quality of the food (Yam et al. 2005). Specifically, its main aim is twofold: to contain and to secure the food in order to satisfy both the industrial economic requirements and consumer’s preferences, by in the meantime keeping up quality and well-being and limiting negative effects on the environment. Traditional packaging usually performs its primary functions, that is, containment, protection, convenience, and communication. Traditional packaging has contributed greatly to the early development of the food distribution systems; however today it is no more sufficient because of the new consumer’s preferences and requirements toward quality and safe food and the increasing demands by retailers for cost-effective extensions to product shelf lives, as well as the increased demand by food companies in terms of hygiene and safety issues associated with fresh and processed meat products. These requirements lead companies to develop new packaging systems. Consequently, new BMs such as concepts of active packaging, intelligent packaging, and biodegradable/edible packaging have been developed and successfully applied in food packaging, thus improving the primary functions of a traditional packaging. Specifically, intelligent packaging uses the communication function of the package to facilitate decision-making and to achieve the benefits of enhanced food quality and safety, communication, shelf life prolongation, and freshness regulation (Yam et al. 2005). Intelligent packaging refers to the monitoring of the internal and external environment of the package in order to predict the quality and remaining shelf life of the food product better than best before date (De Jong et al. 2005). The term “intelligent packaging” is appearing more and more frequently in the literature: a growing number of papers, indeed, have been published in conferences’ proceedings, journals, and magazines. Notwithstanding, an unequivocal definition of intelligent packaging is not yet available. Ahvenainen 2003, for instance, defined intelligent packaging as “packaging ‘systems which monitor the condition of packaged foods to give information about the quality of the packaged food during transport and storage’” (p. 3). Similarly, Clarke (2001) defined it as a packaging with logic capability, while according to, intelligent packaging is a packaging that allows the monitoring of packaged food conditions to give information about the quality of the food during transport and storage. Yam et al. (2005) proposed a more precise and useful definition of intelligent packaging, that is, “a packaging system that is capable of carrying out intelligent functions (such as detecting, sensing, recording, tracing, communicating, and applying scientific logic) to facilitate decision making to extend shelf life, enhance safety, improve quality, provide information, and warn about possible problems” (Yam et al. 2005). Out of the four basic functions of the package previously mentioned, intelligent packaging performs communication part in a more efficient way. The transformation from traditional packaging into intelligent ones is due to the growing importance of recent practices such as, in particular, sustainability and biocompatibility of biodegradable polymers, their monitoring and tracking ability, as well as their interactive and dynamic performance. Upcoming changes in the packaging industry have determined the development of this innovative packaging.

27

Business Models for Biosensors in the Food Industry

673

Specifically, the most important factors leading to rapid innovations in the packaging sector are the following (Appendini and Hotchkiss 2002; Vanderroost et al. 2014): 1. The increasing demand of communicating packages that indicate damage and remaining shelf life of the food product. 2. The problem of food-borne microbial outbreaks that lead to introduce antimicrobial effects in the packaging system without compromising the quality of the food. 3. The development of minimally processed foods. 4. Strict laws in matter of food safety and quality. 5. Modern distribution practices (e.g., online shopping). 6. Pollution-related issues. Intelligent packaging facilitates both materials and information flows in the food supply chain. Consequently, the partners involved in the BMI’s development are all the actors of the food supply chain: consumers, raw materials’ suppliers, food companies, companies manufacturing packages, and distribution. In particular, intelligent packaging provides accurate information to the consumers about safety, quality, and integrity of the package. Figure 3 summarizes the main differences

AIMS AND FUNCTIONS

Traditional food packaging • Containment • Protection • Convenience • Communication

DRIVERS → Sustainability → Biocompatibility

CHALLENGES

Increasing demand in terms of hygiene and safety issues

Intelligent food packaging • Containment • Protection • Convenience • Communication • Enhanced food quality and safety • Shelf life prolongation • Freshness regulation

Increasing demand for cost-effective extensions to product shelf-lives

CONSEQUENCES → Facilitation of material and information flows in the food supply chain → Improved shelf life → Less losses of products

New consumers’ preferences and requirements

Raw Food Food Retailers, materials’ machinery Consumers suppliers suppliers companies distributors

Laws in matter of food safety and quality

Pollution-related issues Food supply chain

Fig. 3 From traditional to intelligent packaging. (Authors own creation)

674

R. Caiazza and B. Bigliardi

among traditional and intelligent food packaging, as well as the main drivers, challenges, and consequences of their introduction. In order to develop an intelligent packaging, various microbial, biochemical, and enzymatic reactions have to be eliminated by adopting biosensors and strategies of temperature control, moisture regulations, addition of chemicals, and implementation of an effective packaging system. In terms of benefits, this BMI will help strengthen the economy because of improved shelf life and less losses of the products. As stressed by Yam et al. (2005), the introduction of such BMIs requires researchers to continue to think “outside the box” and use nontraditional packaging approaches to meet new challenges. The literature proposed two main types of intelligent package devices: data carriers (e.g., barcode labels and radio-frequency identification tags) used to store and transmit data and package indicators (such as biosensors, time temperature, or gas indicators) used to monitor the external environment and communicate warnings. An example of biosensors’ application in food packaging is the prototype known as Food Sentinel System developed by Sira Technologies (Pasadena, CA, USA). In this packaging, the membrane-forming part of the barcode is attached to a specific antibody, which allows the detection of the contaminating bacteria by developing localized dark bar that makes the barcode unreadable upon scanning (Yam et al. 2005). Also Toxin Alert (Ontario, Calif., USA) has developed a diagnostic system known as Toxin Guard for the detection of gross contamination. In this system, which incorporates antibodies into plastic packaging films, when the antibodies encounter a target pathogen, the packaging material displays a visual signal, thus alerting the consumer, retailer, or inspector (Bodenhamer et al. 2004).

Conclusions The introduction of biosensors in the agro-food industry has changed BMs of firms operating in both the agriculture and agro-food industry. As highlighted in the chapter, BMs have been defined as “a unit of analysis to describe how the business of a firm works” (Frankenberger et al. 2013), “describing, as a system, how the pieces of a business fit together” (Magretta 2002). Consequently, BMI involves the firm and its environment as a whole. BMI is a key source of competitive advantage, even more than pure product or process innovators. In the food context, the main BMIs are referred to food insecurity, obesity, food distribution, sustainability, and improvements to nutritional qualities. The food industry is facing various challenges; one of the most important is the need for quick and cost-effective detection methods of allergenic components and pathogens in the food. Generally, chemical and microbiological analyses are expensive, time-consuming, and labor intensive, being carried out by trained operators. Biosensors may help companies operating in the food industry in activities such as contaminant detection, product content verification, product freshness monitoring, and so on. Specifically, they allow companies to overcome all the disadvantages related to the traditional analyses methods and, in particular, they allow the

27

Business Models for Biosensors in the Food Industry

675

application of rapid and affordable methods for quality control. Thus, their potential in the aforementioned industry is growing rapidly. The literature recognizes the potential of biosensors to produce an analytical revolution to resolve the abovementioned challenges that the agro-food industry is facing. In recent years, the development of biosensor technology has improved safety and quality of agro-food products. This chapter has described different examples of biosensors’ application in the food industry (e.g., enzymatic biosensors used to monitor the aging of beer during storage, potentiometric alternating biosensing systems used to detect the presence of Escherichia coli in vegetables, and so on). In addition, it has described also BMIs enabled by biosensors and in particular the example of intelligent packaging. The examples provided showed the benefits that such a method may provide. In addition they confirm that biosensors have nowadays a great positive impact on our daily life. The application of biosensors in the food industry still has successfully competed with the standard analytical techniques in terms of cost, performance, and reliability. To further develop these BMIs, a strong interaction between different complementary areas and disciplines. In addition, also some barriers merit attention when introducing new BMs, mainly referred to human factors (such as, for instance, individuals’ attitudes, histories, and traditions) as well as to external factors (such as, in the case of the agro-food industry, government regulations, value chain position, consumers’ demand, and so on).

References Adley C (2014) Past, present and future of sensors in food production. Foods 3(3):491–510 Ahvenainen R (2003) Active and intelligent packaging: An introduction. In: Ahvenainen R (ed) Novel food packaging techniques. Woodhead Publishing Ltd., Cambridge, UK, pp 5–21 Appendini P, Hotchkiss JH (2002) Review of antimicrobial food packaging. Innov Food Sci Emerg Technol 3(2):113–126 Audretsch DB, Caiazza R (2016) Technology transfer and entrepreneurship: cross-national analysis. J Technol Transf 41(6):129–148 Aung MM, Chang YS (2014) Traceability in a food supply chain: safety and quality perspectives. Food Control 39:172–184 Avermaete T, Viaene J, Morgan EJ (2004) Determinants of product and process innovation in small food manufacturing firms. Trends Food Sci Technol 15(10):474–483 Baden-Fuller C, Haefliger S (2013) Business models and technological innovation. Long Range Plan 46(6):419–426 Baden-Fuller C, Morgan MS (2010) Business models as models. Long Range Plan 43(2–3): 156–171 Baregheh A, Rowley J, Sambrook S (2012) Innovation in food sector SMEs. J Small Bus Enterp Dev 19(2):300–321 Battcock M, Azam-Ali S (eds) (1998) Fermented fruits and vegetables: a global perspective. FAO Agricultural Services Bulletin No. 134, Rome. Also available at www.fao.org/docrep/x0560E/ x0560E00.htm BCG (2008) BusinessWeek/BCG Innovation Survey Bellman R, Clark CE, Malcolm DG (1957) On the construction of a multi-stage, multi-person business game. Oper Res 5(4):469–503

676

R. Caiazza and B. Bigliardi

Beuchelt TD, Zeller M (2016) The role of cooperative business models for the success of smallholder coffee certification in Nicaragua: a comparison of conventional, organic and organic-Fairtrade certified cooperatives. Renew Agric Food Syst 28(3):195–211 Bigliardi B, Dormio AI (2009) An empirical investigation of innovation determinants in food machinery enterprises. Eur J Innov Manag 12(2):223–242 Bigliardi B, Galati F (2013a) Innovation trends in the food industry: the case of functional foods. Trends Food Sci Technol 31(2):118–129 Bigliardi B, Galati F (2013b) Models of adoption of open innovation within the food industry. Trends Food Sci Technol 30(1):16–26 Björkdahl J (2009) Technology cross-fertilization and the business model: the case of integrating ICTs in mechanical engineering products. Res Policy 38(9):1468–1477 Bodenhamer WT, Jackowski G, Davies E (2004) Toxin Alert, Inc. Feb 17, 2004. Surface binding of an immunoglobulin to a flexible polymer using a water soluble varnish matrix. United States Patent 6,692,973 Bouchikhi H, Kimberly JR (2003) Escaping the identity trap. MIT Sloan Manag Rev 44(3):20–26 Brewin DG, Monchuk DC, Partridge MD (2009) Examining the adoption of product and process innovations in the Canadian food processing industry. Can J Agr Econ 57(1):75–97 Brockhoff K (1992) Instruments for patent data analysis in business firms. Technovation 12:42–58 Bucherer E, Eisert U, Gassmann O (2012) Towards systematic business model innovation: lessons from product innovation management. Creat Innov Manag 21(2):183–198 Caiazza R (2015) Explaining innovation in mature industries: evidences from Italian SMEs. Technol Anal Strat 27(8):975–985 Caiazza R (2016a) A cross-national analysis of policies effecting innovation diffusion. J Technol Transf 41(6):1406–1419 Caiazza R (2016b) Internationalization of SMEs in high potential markets. Trends Food Sci Tech 58:127–132 Caiazza R (2017) Innovation for sustainability: a conceptual framework. J Manag Dev 36(1):37–47 Caiazza R, Ferrara G (2016) Economic geography and multipolar strategies: an empirical analysis. J Manag Dev 35(3):394–405 Caiazza R, Stanton J (2016) The effect of strategic partnership on innovation: an empirical analysis. Trends Food Sci Technol 54(1):208–212 Caiazza R, Volpe T (2015) M&A process: a literature review and research agenda. Bus Process Manag J 21(1):205–220 Caiazza R, Volpe T (2017) Innovation and its diffusion: process, actors and actions. Technol Anal Strat 29(2):181–189 Caiazza R, Richardson A, Audretsch DB (2015) Knowledge effects on competitiveness: from firms to regional advantage. J Technol Transf 40(6):899–909 Cammann K (1977) Biosensors based on ion-selective electrodes. Fresen Z Anal Chem 287:1–9 Capitanio F, Coppola A, Pascucci S (2010) Product and process innovation in the Italian food industry. Agribusiness 26(4):503–518 Chatterjee S (2013) Simple rules for designing business models. Calif Manag Rev 55(2):97–124 Chesbrough H, Rosenbloom RS (2002) The role of the business model in capturing value from innovation: evidence from Xerox Corporation’s technology spin-off companies. Ind Corp Chang 11(3):529–555 Clarke R (2001) Radio frequency identification: smart or intelligent packaging? J Packaging Sci Technol Jpn 10(5):233–247 Clarysse B, Utterhaegen M (1998) Inside the black box of innovation: strategic differences between SMEs. University of Ghent Comes S, Berniker L (2008) Business model innovation. In: Pantaleo D, Pal N (eds) From strategy to execution. Springer, Berlin, Heidelberg, pp 65–86 De Jong AR, Boumans H, Slaghek T (2005) Active and intelligent packaging for food: is it the future? Food Addit Contam 22(10):975–979

27

Business Models for Biosensors in the Food Industry

677

Doz YL, Kosonen M (2010) Embedding strategic agility: a leadership agenda for accelerating business model renewal. Long Range Plann 43(2–3):370–382 European Commission (2002) Regulation (EC) no 178/2002 of the European Parliament and of the council of 28 January 2002 laying down the general principles and requirements of food law, establishing the European food safety authority and laying down procedures in matters of food safety. Off J Eur Communities 31(01/02/2002):1–24 Francis D, Bessant J (2005) Targeting innovation and implications for capability development. Technovation 25(3):171–183 Frankenberger K, Weiblen T, Csik M (2013) The 4I-framework of business model innovation: a structured view on process phases and challenges. Int J Prod Dev 18(3/4):249–273 Gassmann O, Enkel E, Chesbrough H (2010) The future of open innovation. R&D Manag 40:213–221 Gellynck X, Vermeire B (2009) The contribution of regional networks to innovation and challenges for regional policy. Int J Urban Regional 33(3):719–737 Girotra K, Netessine S (2013) OM forum – business model innovation for sustainability. M&SOMManuf Serv Oper Manag 15(4):537–544 Grunert KG, Harmsen H, Meulenberg M (1997) A framework for analysing innovation in the food sector. In: Trail B, Grunert KG (eds) Products and process innovation in the food industry. Springer, Boston, pp 1–37 Haard NF, Odunfa SA, Lee CH (1999) Fermented cereals: a global perspective. FAO Agricultural Services Bulletin No. 138, Rome. Also Available at www.fao.org/docrep/x2184e/x2184e00.htm Hennink M, Hutter I, Bailey A (eds) (2011) In-depth interviews. Qualitative research methods. Sage, London, pp 108–134 Jones GM (1960) Educators, electrons, and business models: a problem in synthesis. Account Rev 35(4):619–626 Magretta J (2002) Why business models matter. Harv Bus Rev 80:3–8 Mandour Y, Brees K (eds) (2013) Groeimodellen. Van Duuren Management Markowska M, Saemundsson RJ, Wiklund J (2011) Contextualizing business model developments in Nordic rural gourmet restaurants. In: Alsos GA, Carter S, Ljunggren E (eds) The handbook of research on entrepreneurship in agriculture and rural development. Edward Elgar, Cheltenham, pp 162–179 Matzler K, Bailom F, Friedrich von den Eichen Set (2013) Business model innovation: coffee triumphs for Nespresso. J Bus Strategy 34(2):30–37 McGrath RG (2010) Business models: a discovery driven approach. Long Range Plan 43(2–3): 247–261 Mehrotra P (2016) Biosensors and their applications–a review. J Oral Biol Craniofac Res 6(2): 153–159 Menrad K, Feigl S (2007) Innovations in traditional food products in small and medium-sized companies of the food industry. Available at: http://www.wz-straubing.de/fachhochschulewei henstephan/download/literature%20review_truefood.pdf Ming-Yeu W (2012) Exploring potential R&D collaborators with complementary technologies: the case of biosensors. Technol Forecast Soc Chang 79:862–874 Morris M, Schindehutte M, Allen J (2005) The entrepreneur’s business model: toward a unified perspective. J Bus Res 58:726–735 Murugaboopathi G, Parthasarathy V, Chellaram C (2013) Application of biosensors in food industry. Biosci Biotechnol Res Asia 10(2):711–714 Osterwalder A, Pigneur Y (2002) An eBusiness model ontology for modeling eBusiness. In: Bled 2002 proceedings Osterwalder A, Pigneur Y (2010) Business model generation. A handbook for visionaries, game changers, and challengers. Wiley, New York Osterwalder A, Pigneur Y, Tucci CL (2005) Clarifying business models: origins, present, and future of the concept. Commun Assoc Inf Syst 16(1):1 Pleschak F, Sabisch H (1996) Innovations management. Schäffer-Pöschel, Stuttgart

678

R. Caiazza and B. Bigliardi

Richardson J (2008) The business model: an integrative framework for strategy execution. Strateg Chang 17(5–6):133–144 Risch SJ (2009) Food packaging history and innovations. J Agric Food Chem 57(18):8089–8092 Schneider S, Spieth P (2013) Business model innovation: towards an integrated future research agenda. Int J Innov Manag 17(01):1340001 Scognamiglio V, Arduini F, Palleschi G (2014) Bio sensing technology for sustainable food safety. Trends Anal Chem 62:1–10 Short SW, Bocken NM, Barlow CY (2014) From refining sugar to growing tomatoes. J Ind Ecol 18:603–618 Stubbs W, Cocklin C (2008) Conceptualizing a “sustainability business model”. Organ Environ 21:103–127 Teece DJ (1986) Profiting from technological innovation: implications for integration, collaboration, licensing and public policy. Res Policy 15:285–305 Teece DJ (1998) Capturing value from knowledge assets: the new economy, markets for know-how, and intangible assets. Calif Manag Rev 40:55–79 Teece DJ (2010) Business models, business strategy and innovation. Long Range Plan 43(2–3): 172–194 Ulvenblad P, Hoveskog M, Jet T (2014) Agricultural business model innovation in Swedish food production: the influence of selfleadership and lean innovation. In: Proceedings of the DRUID society conference 2014 on entrepreneurship-organization-innovation 2014. Copenhagen Business School (CBS), Copenhagen, Denmark, 16–18 June 2014 Vanderroost M, Ragaert P, Devlieghere F (2014) Intelligent food packaging: the next generation. Trends Food Sci Technol 39(1):47–62 Wirtz BW, Schilke O, Ullrich S (2010) Strategic development of business models: implications of the Web 2.0 for creating value on the internet. Long Range Plan 43:272–290 Yam KL, Takhistov PT, Miltz J (2005) Intelligent packaging: concepts and applications. J Food Sci 70(1):R1–R10 Zott C, Amit R (eds) (2002) Measuring the performance implications of business model design: evidence from emerging growth public firms. INSEAD, Fontainebleau Zott C, Amit R (2007) Business model design and the performance of entrepreneurial firms. Organ Sci 18(2):181–199

Biosensors: Ethical, Regulatory, and Legal Issues

28

Mohammad Ershadul Karim

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Biosensor: Definition, Emergence, and Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Biosensors: Ethics, Regulation, and Law . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ethical Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Legal Regulations: An Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Legal Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Biosensors, Ethical, Legal, and Regulatory Responses: An Evaluation . . . . . . . . . . . . . . . . . . . . . . . Last Few Words . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

680 681 683 684 689 690 699 701 703

Abstract

The highest attainable standard of health is a fundamental human right as envisaged in important international human rights instruments. Biosensors have the prospect to introduce revolutionary changes in the promotion, preservation, and protection of human health, which are fundamental to enjoy other human rights. The biosensors are also promising in the monitoring of various components of the environment and biodiversity, and thus, can be used to attain some of the 2030 United Nations Sustainable Development Goals. With all the promises, there remain some legal, regulatory, and ethical challenges and concerns as well. This is because, in the different stages of the cradle-to-grave life cycle of scientific research and product development, many stakeholders are involved, and various legal and ethical issues evolve. While the ongoing R&D activities should be continued for the betterment of mankind, consideration of relevant ethical and legal issues should not be ignored. Various international and national bodies, having the authority to guide effective regulation and development of scientific M. E. Karim (*) Faculty of Law, University of Malaya, Kuala Lumpur, Malaysia Bangladesh Supreme Court, Dhaka, Bangladesh e-mail: [email protected] © Springer Nature Switzerland AG 2022 G. Thouand (ed.), Handbook of Cell Biosensors, https://doi.org/10.1007/978-3-030-23217-7_23

679

680

M. E. Karim

and technological applications, have already developed some general regulatory tools, which are either process or product based and can be used in the context of biosensors also, though there is a dearth of such instrument exclusively in the context of biosensors. As this may not be possible to cover every single legal and ethical issues relating to biosensors precisely, this chapter highlights some of the important legal and ethical issues relating to the biosensor.

Introduction Effectively regulated scientific innovations and breakthroughs have been successfully addressing many challenges of global concerns apart from their business and commercial prospects (Donders 2011). For all human beings, the enjoyment of scientific advancements is recognized as an important human right through the core international human rights instruments and non-binding soft law instruments. (The right to enjoy scientific developments has understandably been included in the core international human rights instruments such as the Universal Declaration of Human Rights, 1948 [article 27 (1)] and the International Covenant on Economic, Social and Cultural Rights, 1966 [article 15], and in non-binding soft law instruments such as the Charter of Economic Rights and Duties of States, 1974 [art. 13], the Declaration on the Use of Scientific and Technological Progress in the Interests of Peace and for the Benefit of Mankind, 1975, the UNESCO’s Universal Declaration on the Human Genome and Human Rights, 1997, the International Declaration on Human Genetic Data, 2003, and the Universal Declaration on Bioethics and Human Rights, 2005.) Scholars have identified at least five connections between science, technology, and human rights, i.e.: (a) scientists and engineers have human rights, (b) science and technology can be applied for human rights purposes, (c) the conduct of science and the scientific and technological applications can have negative human rights applications, (d) science and technology can be a constituency for human rights, and (e) right to science is recognized by international human rights law (Wyndham and Harris 2014). The highest attainable standard of health is a fundamental human right as envisaged in the Constitution of the World Health Organization, 1946, and a few other international instruments. The applications of biosensors, analytical devices capable to convert biological responses into electrical signals, are very promising in various areas, including the promotion, preservation, and protection of human health, which is fundamental to enjoy other human rights (Karim 2010). The biosensors are also promising in the monitoring of various components of the environment and biodiversity and thus, can be used to attain some of the United Nations Sustainable Development goals. With all the promises, there remain some legal, regulatory, and ethical challenges and concerns as well. Various international and national bodies, having the authority to guide effective regulation and development of scientific and technological applications, have already developed some regulatory tools, which are either process or product based. The understanding of “regulation” in the context of scientific innovations and breakthroughs is different and extremely

28

Biosensors: Ethical, Regulatory, and Legal Issues

681

complex. This is because, in the different stages of the cradle-to-grave life cycle of scientific research and product development, many stakeholders are involved, and various legal and ethical issues evolve. Important discussions on the ethical and legal issues of ingestible electronic sensors have recently been made (Gerke et al. 2019). Various chapters of this handbook have already covered different aspects and prospects of biosensors. As this may not be possible to cover every single legal and ethical issues relating to biosensors precisely, this chapter will be an attempt to highlight some of the important issues in this regard. Using the desk-based research methodology while reviewing the very limited available literature in the field, this chapter aims to highlight some of the important legal and ethical issues relating to the biosensor. Part two of this chapter highlights the definition, historical development, and some applications of biosensors to contextualize the issue, followed by a brief discussion on possible ethical and legal issues along with various types of available regulatory tools and measures concerning biosensors in Parts three. Finally, the overall ethical and legal issues will be evaluated, and practices of some countries and the European Union in this context will be shared next in Part four. The discussion in this chapter will assist the stakeholders in maintaining their on-going biosensors related activities while guiding the newcomers in setting their policy framework in this regard.

Biosensor: Definition, Emergence, and Applications For the scientific community, the scientific and technical discussion on biosensor may not be warranted, but this may be very relevant to have such a discussion in an easy language for the general readers, including the legal community. A biosensor represents a synergistic combination of biotechnology and microelectronics (Verma and Singh 2005). Generally speaking, the prefix “bio” means anything relating to life or living things, and “sensor” denotes a device that responds to a physical movement. Thus, sensors can detect or measure a physical property, e.g., heat, light, sound, pressure, magnetism, or any motion, and thereafter, record, indicate, respond, or transmit a result. The term “biosensor” is often used to cover sensor devices used to determine the concentration of substances and other parameters of biological interest, and this can be done even without utilizing a biological system directly. Biosensors devices are used to detect the presence or concentration of a biological analyte, e.g., a biomolecule, a biological structure, or a microorganism. The International Union of Pure and Applied Chemistry defines “biosensors” as “[a] device that uses specific biochemical reactions mediated by isolated enzymes, immunosystems, tissues, organelles or whole cells to detect chemical compounds usually by electrical, thermal or optical signals.” Biosensors have two main components: biological (enzyme, antibody, and hormone) and physical (transducer and amplifier). (IUPAC. Compendium of Chemical Terminology, second ed. (the “Gold Book”). Compiled by A. D. McNaught and A. Wilkinson. Blackwell Scientific Publications, Oxford (1997). Online version (2019–) created by S. J. Chalk. ISBN 0–9,678,550–9-8. doi:▶ https:// doi.org/10.1351/goldbook) There are three main parts of biosensors, i.e., (a) a

682

M. E. Karim

component that recognizes the analyte and produces a signal, (b) a signal transducer, and (3) a reader device. Some scholars identify the whole activities of biosensors into five parts, i.e., analyte, bioreceptor, transducer, electronics, and display (Bhalla et al. 2016). From the legal and regulatory point of view, it is crucial to have a specific definition of anything for the parties involved to understand, comply with, and implement the provisions of relevant laws. Unfortunately, no such universally accepted regulatory definition of “biosensor” is available. Nevertheless, this may not create a serious problem as the biosensors and related applications can be regulated using the existing product-based laws, e.g., medical devices, which is a well-regulated sector and sectoral law such as environmental law. Differences of opinions can be identified in the published literature regarding the first use of the word “biosensor.” With its development since 1906, the first manmade biosensor, i.e., enzyme electrode was coined by Clark and Lyons in 1962, while the first commercial biosensor capable to measure glucose in whole blood was developed in 1974 (Griffiths and Hall 1993; Fatoyinbo and Hughes 2012). The interest in the research community and entrepreneurs has been proliferating exponentially because of the prospect of huge and projected commercial outcomes of biosensors. Though the scholars found that in 1995, there were only a very few commercially successful biosensors. Biosensor technologies have some advantages over analytical methods and animal bioassays that include low cost, ease-of-use, and speed. Besides, not many highly trained lab personnel and automation are required in the development stage. Moreover, it is believed that the biosensor applications do not need to consider laboratory animals use related legal or ethical issues. Additionally, the evaluation of the toxicity of a sample with biosensor-based techniques does not require the use of a toxin standard of every compound of a toxin group; just a representative member would be sufficient (Vilariño et al. 2009). Thus, because of its projected prospects, scholars have identified more than 84,000 reports on biosensor indexed in the “Web of Science” database between 2005 and 2015 (Bhalla et al. 2016). Biosensors are used in different applications such as disease monitoring, diagnosis, prevention, drug discovery, and detection of pollutants, disease-causing microorganisms, and markers that are indicators of disease in bodily fluids (blood, urine, saliva, and sweat), food production, pesticide, and blood glucose level identification, etc. (Bhalla et al. 2016; Olson and Bae 2019). Biosensors can give early warnings of some deadly diseases and assist a patient to get the right treatment before there arises any real health-related danger. Biosensors, assisting to biomonitoring to comply with various laws, are prospective and can assist the stakeholders (Farré et al. 2009). Biosensors can be used in the forensic analysis (Yáñez-Sedeño et al. 2014). Researchers have further been experimenting to develop biosensors to detect driving under the influence of marijuana (Stevenson et al. 2019). BEEP-WATER biosensor technology can detect and quantify on-site a wide range of chemical contaminants including pesticides, heavy metals, and emerging contaminants in water, saving the cost associated with current laboratory analysis (https://cordis.europa.eu/project/id/ 673710).

28

Biosensors: Ethical, Regulatory, and Legal Issues

683

In the context of sustainable development, access to science and technology has already been identified as crucial to implement and revitalize the global partnership for sustainable development (goal 17 of the UN Sustainable Development Goals). Biosensors can be used to attain universal access to health care, minimize global health threats, an important vision of UN Sustainable Development Goals. Biosensors can be used to attain a number of other goals as included in the UN Sustainable Development Goals such as ensuring of healthy lives and promotion of well-being for all at all ages [Goal 3], ensuring of sustainable consumption and production patterns [Goal 12], and end of hunger, the achievement of food security, improvement of nutrition level, and promotion of sustainable agriculture [Goal 2].

Biosensors: Ethics, Regulation, and Law The existing provisions on ethics, regulation, and law are sometimes overlapping in the context of scientific and technological innovations. For an easy but general understanding and reference, “law” can be defined as a set of selected provisions containing formal rules, usually documented and specific. These are generally encapsulated in regulatory instruments, be it soft or hard in terms of consequence, with provisions prescribing the behavior that someone intended must or must not do. For the violation of such legal provisions, some provisions on punishments or sanctions of various natures are also prescribed. On the other hand, ethical issues can be seen as the law or regulation of an individual’s mind. Ethical issues guide standard human conduct shaped by the beliefs, principles, norms, culture, etc. of a society or community. Moreover, ethics deal with the issues which law can or does not address. Ethical rules, having links with morality, are generally undocumented though these can sometimes be documented in the form of Code of Conduct, Guidelines, or Guidance documents, etc. which are relatively soft but wide in nature comparing to laws and are generally seen to be observed by the professional bodies mainly. For the violation of ethical principles or rules, unless there is a parallel violation of legal provisions, depending on the nature of the violation, the punishment may not be severe, and the matters of ethical violations are generally handled utilizing administrative or institutional mechanism without the involvement of the law enforcement agency. Generally, ethical evaluations are performed by the experts, laypersons, and panels with both experts and laypersons. It will be pertinent to share that some of these issues are generally common in all kind of scientific and technological innovations and applications, and thus, will be applicable to biosensors as well, whereas some issues may be very specific to biosensors. For example, biosafety or whole-cell biosensors pose the risk or possibility that genetically engineered microorganisms may be released into the environmental components, and therefore, they need to undergo higher levels of legal and ethical scrutiny (▶ Chap. 13, “Engineering Prokaryote Synthetic Biology Biosensors”; Dana et al. 2012). Some ethical, legal, and regulatory issues relating to biosensors are highlighted below.

684

M. E. Karim

Ethical Issues Consideration of various ethical issues relating to the research and development (R&D), scientific and technological innovations, and applications play a pivotal role in its ultimate projected commercial or noncommercial success. Within the R&D and commercialization life cycle of any scientific innovation, ethical issues are considered to evaluate the prospective possible consequences having diversified implications on the stakeholders and environmental components, engage and inform the public in the development process of technological innovations, and use of products developed exploiting this technology, etc. Before any attempt to delve into the relevant ethical issues, it will be pertinent to share that in the area of scientific and technological knowledge, innovations, and applications, ethical rules and principles have been being practiced since time immemorial. Nevertheless, different international, intergovernmental, and non-governmental organizations and bodies have been enjoying the authority to develop relevant ethical standards and norms within their mandated areas. Corresponding national bodies have also been developing ethical standards and norms by taking into their socio-economic situation, which are more or less similar to the initiatives of their counterpart international and inter-governmental organizations. With the understanding of ethics shared in the previous section, one of the very first ethical issues to consider in this regard will be the issue of fraud and misconduct in scientific research which started to appear in the popular news media since the 1970s (Steneck 1994; Gold 1993). In the United States of America (USA), the issue of the fake skin transplants in white mice incident at the Sloan-Kettering Institute in New York in 1974 was reported (Wells and Farthing 2008). Similar issues were also reported in the United Kingdom (UK) context almost in the same period (Smith 2016). The extent of different fraudulent practices and misconduct in scientific research has started to get official attention since then (Kumar 2008; Kansu and Ruacan 2002) both nationally and internationally. In the area of ethical regulations of scientific research and applications, within the auspices of the United Nations, the United Nations Educational, Scientific and Cultural Organization (UNESCO) has the mandate to act as the global forum on ethics since the 1970s. As early as in 1974, the UNESCO framed the UNESCO Recommendations on the Status of Scientific Researchers, which were recently updated in 2017. The World Commission on the Ethics of Scientific Knowledge and Technology (COMEST), a body of 18 scholars with expertise in the area of science, law, philosophy, and politics, and appointed by the UNESCO DirectorGeneral together with 11 ex officio members of UNESCO, was established in 1998 to formulate ethical principles to assist the decision-makers with criteria. The International Bioethics Committee, which was formed to follow the progress in life sciences and applications to ensure respect for human dignity and freedom, together with COMEST, make the UNESCO the leading global organization in this sector. Besides, UNESCO has also developed a very resourceful repository, Global Ethics Observatory, containing information on, inter alia, ethics legislation, codes of conduct, etc. of different countries in the world. Even if these ethical codes are

28

Biosensors: Ethical, Regulatory, and Legal Issues

685

mere recommendations or guidelines, which may influence someone to interpret these strictly as non-binding, this repository can be consulted for direction as a ready reference since there are materials available from different jurisdictions having different legal systems. Apart from the international move, Europe, as a region, appears to be the most progressive in terms of dealing with ethical issues regarding scientific research and technological applications in a systematic and coordinated manner. Since the beginning of this decade, the European policymakers have initiated a Responsible Research and Innovation (RRI) approach “for more responsible, acceptable, and ethical science and technology development in Europe - in the pursuit of a better, more sustainable and more equitable world.” RRI, a pillar of the European Union (EU) Framework Programme for Research and Innovation – Horizon 2020, becomes decisive towards attaining a smart, sustainable, and inclusive growth. A “Practical Guide to Responsible Research and Innovation: Key Lessons from RRI Tools” has also been developed containing toolkit, implementation techniques, and five golden rules to achieve RRI. RRI needs to anticipate, reflect, deliberate, and be responsive. The UK Engineering and Physical Research Council proposed the anticipate, reflection, engagement, and action (AREA) framework to ensure RRI (Owen 2014). Besides, the European policymakers have been issuing some non-binding documents. For example, earlier in 2009, the European Commission recommended a code of conduct for responsible nanosciences and nanotechnologies research. Anyone from Europe involved in R&D of biosensors should take these into account and for others, these can be considered as international best practices. Similar to Europe, the Heads of Arab States have approved the Charter of Ethics of Science and Technology in the Arab Region in 2019 as the guiding document to be implemented voluntarily within the national jurisdictions of the Arab states. The Charter embodied some general ethical principles, ethics of productions, transfer and localization, and harnessing and use of science and technology. In the USA, a similar initiative like the European RRI is known as the Responsible Conduct of Research (Steneck and Bulger 2007). In order to promote integrity in Public Health Service (PHS) funded research, the “Introduction to the Responsible Conduct of Research” was earlier adopted and subsequently revised in August 2007 by the Office of Research Integrity which was established as a consequence to stop research fraud and misconduct. It will be pertinent to mention here that even though this is neither an official policy statement nor a guideline, this has a huge impact on PHS funded research. Since it is neither possible nor desirable for the purpose of this Handbook to cover the exhaustive list of relevant legal and ethical issues regarding biosensors, only some of the important ones will be discussed here. In the discussion of biosensorbased applications, some of the relevant ethically challenging areas, which are mostly common in all sorts of scientific and technological applications, may, inter alia, include: (a) access to health care, nondiscrimination, and equal opportunities for all, (b) animal testing and medical research with a human being, (c) autonomy and best interest of the human being including children, (d) citizen’s right to information, (e) informed consent, and (f) privacy and personal data protection

686

M. E. Karim

issues. As some of these issues are regulated through legal instruments, the legal aspects of these will be discussed later on. The next segment will provide an overview of these from the ethical point of view.

Access to Health Care, Nondiscrimination, and Equal Opportunities for all The right to health, which includes both freedoms (i.e., control on and non-interference with one’s health and body) and entitlements of equal opportunities, must be enjoyed without discrimination on the grounds of race, age, ethnicity, or any other status. The human right to health requires that the states should work towards progressive realization using maximum available resources without adopting a retrogressive measure. The core components of the right to health include availability, accessibility, acceptability, and quality (safe, effective, people-centered, timely, equitable, integrated, and efficient). (General Comment 14, the Committee on Economic, Social and Cultural Rights, 2000.) At the international level, the human rights commitments of the member states of the United Nations are monitored by UN Charter-based and treaty-based mechanisms. The obligations of any state, which is a member of any human rights treaty, to support the right to health – through the allocation of “maximum available resources” to progressively realize this goal – is reviewed through various international human rights mechanisms, such as the Universal Periodic Review, or the Committee on Economic, Social and Cultural Rights along with the mechanisms established within the core international human rights treaties. Some of the relevant ethical issues are also simultaneously covered in international and regional human rights instruments. The list of such ethical issues includes: nondiscrimination in health care (The International Covenant on Civil and Political Rights (1966) [art. 26], the International Covenant on Economic, Social and Cultural Rights (1966) [arts. 2 & 12], the International Convention on the Elimination of All Forms of Racial Discrimination (1965) [art. 5], the Convention on the Elimination of All Forms of Discrimination Against Women (1979) [art. 12], and the Convention on the Rights of the Child (1989) [art. 24]), access to health care and essential medicines (The Convention on the Rights of the Child (1989), the International Covenant on Civil and Political Rights (1966) [art. 24], and the International Covenant on Economic, Social and Cultural Rights (1966) [art. 12]), and medical research with human beings, etc. (The UN International Covenant on Civil and Political Rights (1966) [art. 7].) Some of these core human rights instruments are already signed and ratified by most of the member states of the United Nations and therefore, they are under an obligation to implement these provisions within their domestic jurisdictions to give effect to their pledge and commitment made at the international level unless any reservation is made. Unfortunately, it seems that the ethical issues are not that seriously considered in their agenda and there are news reported and broadcasted almost regularly in the mass and social media. The Council of Europe, the human rights organization, adopted the Convention for the Protection of Human Rights and Dignity of the Human Being with regard to the Application of Biology and Medicine in 1997. This Convention, as a hard law

28

Biosensors: Ethical, Regulatory, and Legal Issues

687

and entered into force in 1999, together with its Additional Protocols contains provisions on the access to health care and essential medicines, medical research with human beings, nondiscrimination in health care, etc. Thus, the signatories to this Convention are responsible to implement these ethical issues as their treaty obligations. It has already been shared that access to health care, nondiscrimination, universal health care, and equality in getting health care services are recognized as fundamental human rights by core international human rights instruments and domestic laws of most of the countries in the world. Nevertheless, the people of the least developed countries have been suffering from basic necessities including access to health care-related needs and services. Like the digital divide, there are numerous instances of health divide and discrimination in the use and application of health care facilities even in developed regions such as Europe (Marmot et al. 2012), let alone the real situation in the least developed countries of Asia and Africa. A similar situation of health divide may arise in relation to the use of biosensor technology and it may also happen that only the wealthy people may afford to get the benefit of it, which will frustrate the real purpose of universal health care. Besides, there are general ethical concerns that these Asian and African countries have already been or can be used as a testing ground of these technological applications before their long-term human health and environmental implications are confirmed in a completely convincing manner. Therefore, to avoid such disparity, the human rights-based approach, which advocates for accountability, equality, and nondiscrimination, that stemmed from the international human rights law regime should be implemented (Marmot et al. 2012). In the absence of any effective legal arrangements, the exercise of ethical practices will play an instrumental role in this regard.

Animal Testing and Medical Research with Human Being The issue of care or welfare of animals used in scientific research is in consideration for more than six decades since the publication of Guide for the Care and Use of Laboratory Animals in 1963, which was revised several times subsequently. Fortunately, in relation to biosensors, it is believed that legal and ethical issues relating to the laboratory use of animals are not serious initially (Vilariño et al. 2009). However, during their clinical trials on human beings, internationally recognized guidance documents should be followed. Besides, at the time of application of such technological innovations, it is an ethical responsibility to ensure the autonomy and best interest of the human being, including children and this is within the very basic ethical norms of medical practice derived from Hippocrates oath, which was subsequently endorsed by World Medical Association through the Geneva Declaration, 1948, and was recently revised at Chicago in 2017 in the form of the Physician’s Pledge. Besides, some soft and non-binding international legal instruments such as the World Medical Association Deceleration on Ethical Principles for Medical Research involving Human Subjects, 1964 (Helsinki Declaration), and the Universal Declaration on Bioethics and Human Rights, 2005, containing important provisions are in place. Health professionals simultaneously need to take into account and

688

M. E. Karim

ensure privileged communication and patient’s expectation of medical care and advice.

Informed Consent In the development and application of scientific innovations, the issue of consent, one of the most debated issues, will remain a great concern though this has been in practice since the ancient Roman, Greek, and Egyptian civilization (Mallardi 2005). Even though this issue is nowadays considered as a legal matter, some of its general and open-ended provisions have ethical implications. There are five elements of informed consent, i.e., voluntariness, the capacity of the party, sufficient disclosure of the implications and consequences, proper understanding and ability to decide any time after the consent is given. There are some ethical dilemmas regarding the true nature of informed consent, i.e., whether this ensures the patient’s autonomy, or it is being used as a precaution to avoid medical malpractice cases. Regarding consent in biomedical or medical research, the service providers have the general duty to inform the patient about the consequences so that this can be used in prospective future medical malpractice cases (Faden and Beauchamp 1986). As the issue of consent is contentious and there are misunderstandings among the stakeholders regarding the standard of valid consent, the World Health Organization (WHO) has already developed some templates that can be used to obtain informed consent. The list of such templates includes Informed consent for clinical studies, Consent for storage and future use of unused samples, Informed consent for qualitative studies, Informed assent for children/minors, Informed parental consent for research involving children (qualitative), Informed parental consent for research involving children (clinical), etc. It may be argued that these template-based consents may not be sufficient as with the change of circumstances, the status of consent may be changed. Besides, there are concerns about consent in the case of a vulnerable patient group and persons in a critical emergency condition. The issue of a properly recorded consent will be even serious and complicated after the introduction of the European General Data Protection Regulation (GDPR) which has introduced strict requirements on consent when personal information such as health-related data and information are collected, processed, and used for commercial purposes. Furthermore, some software or electronic means are regularly used to maintain the contractual terms and conditions and to obtain consent. Studies revealed that people rarely read such terms and conditions, which are also full of legalese. A strict interpretation of GDPR will require to obtain fresh consent every time a new purpose or use arises. For scientific innovations such as biosensors, it may not be possible to communicate every single consequence precisely and that’s why the ethical norms and standards should be practiced. Nevertheless, since these are mere templates, the stakeholders can adjust these based on their needs and demands. Privacy and Personal Data Protection Issues In the development and commercialization process of scientific and technological innovations, there arise a number of privacy and personal data protection concerns.

28

Biosensors: Ethical, Regulatory, and Legal Issues

689

A universally accepted definition of “privacy” is impossible to share as the understanding of private matters or privacy vary on the basis of culture, race, sex, religion, nationality, age, etc (Munir et al. 2018). However, the human right to privacy is recognized through the provisions of some of the core international human rights instruments and many countries in the world have included the qualified constitutional right to privacy in their national Constitutions (Karim 2005), while around 150 countries have laws on personal data protection and these laws have included a definition of personal data. While this is a trend that the issues relating to the collection, processing and use of personal data are considered through the binding provisions of law, in case of privacy matters, the understanding and application of ethical principles will play an important role as a legal right to privacy, especially in countries with common law legal system, is difficult to establish in the court of law.

Citizen’s Right to Information In democratic societies, citizens have the right to know most of the thing that ultimately affects them collectively as the government is a representative government and the government is primarily run on the taxpayers’ contributions. This is also ethically very significant and relevant when any scientific research is conducted using the public fund. There started a global movement of “open data” recently which advocates to share the research data in a platform available for all so that some kind of accountability can be ensured and the public fund can be managed properly by reducing the duplication of similar research. Europe has been leading the movement and other parts of the world are following suit. Regarding biosensors, most of the ethical concern may revolve with the health professionals as a caregiver or people involved in the application of biosensors in the environment. The applications and consequences of biosensors should be made clear to the society in an effective way for its ultimate success so that similar fate of GM food in Europe can be avoided.

Legal Regulations: An Overview The word “regulation” is difficult to define as it presents different meanings to different people in a different context (Levi-Faur 2011). The word, however, is generally used to denote the mechanisms used to control the business or market activities. Regulations are broadly defined as the rules imposed mainly by the government, except in case of self-regulation, “backed by the use of penalties that are intended specifically to modify the economic behavior of individuals and firms in the private sector.” (Glossary of Industrial Organisation Economics and Competition Law, compiled by R. S. Khemani and D. M. Shapiro, commissioned by the Directorate for Financial, Fiscal and Enterprise Affairs, OECD, 1993, available at http:// www.oecd.org/regreform/sectors/2376087.pdf) Regulations can be evidence-based developed using the risk assessment profile or process-based which can be done through cradle-to-grave product life cycle assessment (LCA). Again, regulations can be either process or outcome-based.

690

M. E. Karim

Various types of legal regulatory tools are available and in use to address different needs in a different context. These, in terms of effect, can be hard and soft in nature. From the jurisprudential point of view, at the domestic level, hard regulations also referred as “command and control regulation,” can take the form of Acts of Parliament; or subsidiary or delegated legislation framed through the authority derived from the Acts of Parliaments which generally come in the form of Rules or Regulation. Hard laws are again either general or specific/special in nature. Besides, there are industry self-regulation tools too which are binding on and mandatory to follow by the members of the said industry. On the other hand, some regulations are soft in nature and come usually in the form of Guidelines or Codes. At the international level, some internationally accepted market authorization hard laws and standards are followed to ensure the safety and quality standards of a product. Some of the ethical standards developed by authoritative bodies to cover relevant issues are mandatory to follow by the stakeholders. These are issued by international organizations such as Organisation for Economic Co-operation and Development (OECD), WHO, and standard-setting organizations such as the International Organization for Standardization (ISO), etc. A list of such instruments includes: Good Laboratory Practice (GLP), Good Manufacturing Practice (GMP), and Good Clinical Practice (GCP), etc. Many developed countries have adopted their national versions of these instruments to ensure their binding legal effects. It will be pertinent to mention here that any decision taken on the nature and effect of a legal or regulatory instrument only through a glance at the title of the instruments, i.e., rule, regulation, code, guidance, etc. can be misleading. Instead, such a definition should be taken by looking at the relevant provisions of the instruments to avoid confusion in this regard. For example, guidelines, when strictly interpreted, are seen as a soft law instrument. However, some Guidelines have binding force and can be enforced through the court of law (Karim et al. 2015). Of the various types of regulation available, e.g., economic, financial, social, administrative, and so on, in the legal regulation of emerging scientific and technological applications, it is internationally recognized and widely practiced that the cradle-to-grave LCA is initially conducted to decide the suitable form of regulation. Even though this type of LCA exercise is primarily intended to understand the environmental implications of any product, project, or initiative, such an exercise, properly designed following an international standard such as ISO 14040, can easily assist the stakeholders to visualize and get some preliminary ideas about various legal issues involved regarding any scientific and technological applications.

Legal Issues In the development process of scientific and technological innovations and applications, including biosensors, myriad legal issues can arise. The presence of some overlapping legal and ethical issues in the cradle-to-grave product lifecycle of biosensors can be identified and some can be foreseen. Though cradle-to-grave product LCA is employed to understand the environmental impact of any product in its

28

Biosensors: Ethical, Regulatory, and Legal Issues

691

different phases of the life cycle, i.e., from materials processing, manufacturing, distribution, use and end-use disposal, this was found to be very effective and has been used in the context of various emerging technologies (Salieri et al. 2018; Guinée et al. 2017). A quality cradle-to-grave product LCA, developed, for example, using ISO Standard 14040 methods as explained (Muralikrishna et al. 2017), can be taken into consideration. Of the two types of regulations, i.e., process based and outcome based, the LCA can be seen as a process-based regulation. On outcomebased regulation, for some products, there needs pre-market approval and/or premarket notification and post-market reporting. It may be pertinent to share here that since the product review process usually requires a huge amount of time, the regulators usually ignore the regulation of low-risk products. To understand the various legal and regulatory issues involved in anything including biosensors, it is crucial to have a specific definition. It has already been shared that there is no such definition of biosensors is generally available in legal text. Nevertheless, such a gap regarding the legal definition of bioseonsor should not cause a real problem for policymakers. In the absence of such a definition, since biosensors are and can be used both as a device/product or chip integrated into a product, the existing legal provisions on medical devices, product liability, consumer protection, environment, etc. will obviously be relevant. The application of international environmental law which is applicable to nanomaterials (Karim and Munir 2016) and product liability law which is applicable to polymer nanocomposite (Karim et al. 2018) will equally be applicable in the case of biosensors. From the regulatory point of view, “biosensors” can again be considered within the definition of “digital health and care,” which is defined by the USA Food and Drug Administration (FDA) and EU to mean the “tools and services that use information and communication technologies (ICTs) to improve prevention, diagnosis, treatment, monitoring and management of health and lifestyle.” (European Commission, eHealth: Digital health and care, https://ec.europa.eu/health/ehealth/ home_en) If that is the case, then the various legal issues relating to “digital health and care,” e.g., medical device, personal data protection, cybersecurity, etc. (Gerke et al. 2019), shall also be applicable to biosensors. Based on the process-based and outcome-based regulation, various legal issues may be relevant to discuss. In process-based regulation, the product life cycle analysis can be considered to understand various legal issues, whereas, for outcome-based regulation, the discussion on product liability and ancillary matters will be relevant. Thus, it can be anticipated that various legal issues may arise based on the point of view or context considered. Some other issues such as occupational health and safety issues or legal issues on intellectual property rights may be relevant also, but will not be considered in this chapter. (For example, in a very recent judgment, while clarifying when and how much compensation an employee inventor should receive under the Patent Act 1977, the UK Supreme Court has awarded Professor Shanks, who developed biosensors for monitoring diabetes, GBP two million. The findings and decision of this case are very relevant for the entrepreneurs and should be taken care of. See, Shanks v Unilever Plc and others [2019] UKSC 45.) This part will be an attempt to highlight some selected legal issues.

692

M. E. Karim

Medical Device The discussion on the legal issues in the medical device industry, which integrates technologies such as biomedical, materials and electronics, seems to be crucial in the discussion on legal issues in biosensors (Chen et al. 2018). Nevertheless, before concentrating on such a discussion, it has to be decided, whether biosensors can be treated as a “device” within the definition of “medical device.” Leading international organizations and the regulators of the developed countries have already introduced a regulatory definition of “device” in the context of health care. The WHO has accepted the definition of “medical device” as proposed by the Global Harmonization Task Force on Medical Device (GHTF), which was subsequently renamed as the International Medical Device Regulators Forum (IMDRF). (The Global Harmonization Task Force on Medical Device (GHTF) is a voluntary group founded in 1992 through the representatives from national medical device regulatory authorities and industry from Europe, Asia-Pacific, and North America. The International Medical Device Regulators Forum (IMDRF), which replaced GHTF, is now a coalition of members from Australia, Brazil, Canada, China, Europe, Japan, Russia, Singapore, South Korea, and the USA along with WHO as an Official Observer. IMDRF created three regional initiatives in Asia, Asia Pacific, and Latin America.) Accordingly, “medical device” is defined as “any instrument, apparatus, implement, machine, appliance, implant, reagent for in vitro use, software, material or other similar or related article, intended by the manufacturer to be used, alone or in combination, for human beings, for one or more of the specific medical purpose(s) of: – diagnosis, prevention, monitoring, treatment or alleviation of disease; – diagnosis, monitoring, treatment, alleviation of or compensation for an injury; – investigation, replacement, modification or support of the anatomy or of a physiological process; – supporting or sustaining life; – control of conception; – disinfection of medical devices; – providing information by means of in vitro examination of specimens derived from the human body, and does not achieve its primary intended action by pharmacological, immunological or metabolic means, in or on the human body, but which may be assisted in its intended function by such means.” (World Health Organization, Medical Device – Full Definition, available at https://www.who.int/medical_devices/full_deffinition/en/) To be a medical device, the purpose of use should be established. It is important to understand that even if a device, traditionally used for medical purposes, is used for a different purpose, it will not be treated as a “medical device.” (Brain Products GmbH v BioSemi VOF, Antonius Pieter Kuiper, Robert Jan Gerard Honsbeek, Alexander Coenraad Metting van Rijn (Case C-219/11), 22 November 2012.) Therefore, the manufacturers should be careful about the intended purpose of a device as that will define the legal consequences. It is evident that the above definition of “medical

28

Biosensors: Ethical, Regulatory, and Legal Issues

693

device” is very wide and can be presumed that if the biosensors are incorporated in or used for medical treatment, then it will come within the definition of medical device and thus, will strictly be regulated since medical devices, a domain which is strictly regulated, are regulated following very high international standards such as ISO 13485:2016 Medical devices – Quality management systems – Requirements for regulatory purposes, ISO 14155:2011 Clinical investigation of medical devices for human subjects – Good clinical practice, ISO 14971:2019 Medical devices – Application of risk management to medical devices, and so on. In Europe, “CE Marking” is used to ensure the genuineness of a medical device. A manufacturer will strictly be liable for a defective product. In Europe, the highest judicial authority has declared that even a suspicion of defect is sufficient to classify all products defective resulting in replacement. (Boston Scientific Medizintechnik GmbH v. AOK Sachsen-Anhalt – Die Gesundheitskasse (C-503/13), and Boston Scientific Medizintechnik GmbH v. Betriebskrankenkasse RWE (C-504/13), 5 March 2015.) With the intended use of the medical device, i.e., domestic use and/or use in a different jurisdiction, the entrepreneurs should consider different legal and regulatory requirements as because of the legal system, i.e., civil law or common law or a hybrid legal system, as well as the socio-legal condition of a country, the regulatory requirements may be quite different. Generally, countries around the world classify medical devices into categories such as I, II, III, IV or A, B, C, D and include different regulatory requirements, which must be adhered to strictly. There may have “de novo” classification too depending on the exclusive features of any device. Compliance to Mutual Acceptance of Data (MAD), GMP, and GLP may assist to some extent and it may be presumed that a medical device that has received approval in the developed jurisdiction will be treated positively in other jurisdictions though there is no guarantee to this end. Hence, the decision of the respective regulators seems to be the last word to consider. The entrepreneurs should also consider the issues on pre-market, during the market, and post-market regulatory requirements and comply with these issues. Sometimes as the pre-market stage, the regulator may require clinical data developed using the literature, clinical experience, and trial to demonstrate that the intended device is safe and effective. In some instance, the regulator can issue open public letter too to get further clarification about a product which has already received approval. For example, the US FDA has issued a letter to Biosense Technologies Private Limited in 2017. (Food and Drug Administration, Letter to Biosense Technologies Private Limited concerning the uChek Urine Analyzer, 12/14/2017, https:// www.fda.gov/medical-devices/industry-medical-devices/letter-biosense-technolo gies-private-limited-concerning-uchek-urine-analyzer) Some regulators also provide the opportunity to discuss with them even before the introduction of products in the market. The entrepreneurs should explore such opportunities and utilize these, if available, actively. After the issue of regulatory permission is solved, then the issue of agency/partnership, employment contract, etc. will be relevant. Responsible authoritative bodies have already released some guidance documents which the stakeholder should consider also in the course of their activities. For example, the WHO framed that Global Model Regulatory Framework for Medical

694

M. E. Karim

Devices including in vitro diagnostic medical devices, Asian Harmonisation Working Party (AHWR) developed a Playbook for Implementation of a Medical Device Regulatory Framework.

Informed Consent Even though apparently the issue of informed consent was historically originated as an ethical matter, it is now treated as a legal issue and more carefully followed in the health care service ecosystem to avoid various legal consequences. Generally speaking, in common law legal systems, any consent will not be considered as a free one if it is obtained from anyone minor or does not have the capacity to give free consent, and when consent is obtained by way of fraud, mistake, coercion, misrepresentation, etc. A few uncertainties regarding consent, in case of personal information collection and processing, have been clarified through the provisions of the European GDPR. Under articles 4(11) and 7, the GDPR, consent must be freely given, specific, informed, and unambiguous, and should be given through a statement or clear affirmative action. Moreover, such consent once given can be withdrawn at any time. It will be relevant to share here that though these are suggested in personal data protection scenario, the GDPR consent framework can be considered for guidance in every instance where “consent” is an issue. Hence, if biosensors are used in any product or device involving human being or environmental components which may have an adverse impact on human society, the elements of consent discussed here should be considered seriously to avoid future adverse legal consequences. Consumer Protection Biosensors can be integrated into various devices to be used in different sectors to be enjoyed by the natural persons in the form of goods or services, who will be treated as “consumers,” and thus, a discussion on the provisions of the consumer protection law shall also be relevant for the purpose of this chapter. The definition of the term “consumer” is very wide, and no uniformity or consistency can be found in the available definitions. The presence of the term can be found in almost every commercial sector. In 2015, a research of the European Parliament revealed that at least 90 community legislation has provisions on consumer protection. (European Parliament, ‘Consumer Protection in the EU: Policy Overview’, European Parliamentary Research Service, September 2015, p. 5.) The term is generally used to mean a natural person who purchases, acquires, or uses any goods or services for personal, domestic, or household use or consumptions. Different international and national instruments have bestowed some rights to consumers. A list of such rights includes: the right to satisfaction of basic needs, right to safety, right to be informed, right to choose, right to be heard, right to redress, right to consumer education, right to a healthy environment, etc. When these are the rights of the consumers, these are, in converse, obligations on the manufacturers or service providers and therefore, they should take initiatives that will enable the consumer to get these rights. Even after having these provisions, since the consumers have very weak bargaining power, these rights are not respected always generally in the developing countries. This situation becomes even worse in this hyper-connected

28

Biosensors: Ethical, Regulatory, and Legal Issues

695

world when the stakeholders in the consumer ecosystem can be based in different jurisdictions. Albeit, the manufacturer or supplier of biosensors should consider consumer protection related legal rights and obligations of the respective jurisdiction. Under the existing international legal regime, the Economic and Social Council of the United Nations through the United Nations Conference on Trade and Development (UNCTAD) has been playing a pivotal role in promoting the consumers’ rights. The UNCTAD has developed the United Nations Guidelines for Consumer Protection which are “a valuable set of principles that set out the main characteristics of effective consumer protection legislation, enforcement institutions and redress systems.” These Guidelines were subsequently revised and the revised version was then adopted by the UN General Assembly in 2015. An innovation of the revised Guidelines is the scope to establish an Intergovernmental Group of Experts on Consumer Protection Law and Policy (IGE), to be held under the authority of UNCTAD. This forum can play a leading role in promoting consumer rights in this very connected but complicated world with regard to any scientific and technological innovation, including biosensor enabled innovation, having direct impacts on the consumer.

Product Liability The issue of product liability, an important purpose of which is to ensure safe and desirable moveable consumer products, is considered through general and product specific legal provisions. For example, in the context of the Europe, a general legal framework on product liability can be found in the provisions of the Liability for Defective Products Directive 1994 (Directive 85/374/EEC), Directive on Certain aspects of the Sale of Consumer Goods and Associated Guarantees, 1999 (Directive 1999/44/EC), and the European General Product Safety Directive, 2001 (Directive 2001/95/EC). Whereas, product liability-related provisions can also be found in the product-specific European community law on, e.g., medical device, agricultural products, cosmetics, pharmaceuticals, etc. Product liability laws contain provisions ranging from general liability to strict liability, individual liability to joint liability for a defective product. A product can be declared as “defective” if it does not meet the expectations of the consumer in its design, manufacturing process or outcome and misuse. (See, for example, articles 6, the Liability for Defective Products Directive 1994 (Directive 85/374/EEC).) To be successful in getting the proper legal remedy regarding a defective product, an injured person, however, has to prove the damage, the defect, and the causal relationship between defect and damage. (See, for example, Article 4, the Liability for Defective Products Directive 1994 (Directive 85/374/EEC).) While these may not be difficult in case of many other products, these issues will really be challenging for the consumers to prove with respect to biosensor enabled products both in the developed and developing countries due to the scientific uncertainties with regard to their long term human health and environmental implications. (See, for example, the judgment of the high profile product liability group action case Gee v DePuy International Ltd. ((2018) EWHC 1208 (QB)), the so-called ‘Depuy Pinnacle

696

M. E. Karim

Metal on Metal Hip Litigation’, where the court gave the strict legal interpretation and ruled in favour of the manufacturers.) For the manufacturers, the provisions of the product liability law should not be seen as an obstacle to innovation and commercialization. The existing legal framework also contains a variety of provisions using which the producer/manufacturer and supplier of the product can avoid his liability if it can be proved that he did not put the product into circulation, the damage did not exist at the time when the product was put into circulation, the defect was due to compliance of the product with mandatory regulations issued by the public authorities, the state of scientific and technical knowledge at the time when the product was put into circulation was not such as to enable the existence of the defect to be discovered, etc. On the issue of the misuse of the products, the liability can be avoided if precise instructions as to the product use or label containing safety instructions are included. Besides, the liability may be reduced or disallowed in case there is any fault on the part of the injured person. Moreover, the manufacturers can put the “limited liability clause”; however, the law does not permit the producer to insert provisions that will limit or exclude or exempt him from his liabilities. With all these systems available, one of the main concerns shall be the burden of proof as the product liability law imposes the obligation on the injured party who has to prove the actual damage, a defect in the product and a causal link between the damage and the defect. These can only be possible if the precise consequences as to human health and/or environmental implications can be understood. Besides, an injured party needs to bring the action within a statutory period of limitation, which is usually 3 years. From our experience on the regulation of tobacco or asbestos, which were initially projected as medicine or miracle fiber, it can be presumed that it may take a few decades to realize and understand the precise human health and environmental implications of emerging technological applications, including biosensors, if any. The idea or apprehension about the chance of repetition in the case of all emerging technological innovations, including biosensors, should not be ignored. Therefore, the regulators need to come up with the regulatory information requirements and the manufacturers need to share the information, as far and precise as possible, about their products. It is a matter of hope that at the European level, the regulators have started the initiative to understand the reactions of the common people if the community product liability legal framework requires an update and the European Court of Justice has clarified some of the uncertainties.

Personal Information Protection Personal information or data, frequently dubbed as the oil or the currency of the Internet, generally means to include any information relating to a natural person using which that person can be identified directly or indirectly by reference to an identifier and one or more factor such as physical, physiological, genetic, mental condition, etc. and used for commercial purpose or in exchange of consideration collected for the provision of goods or services. Even though the issue is so crucial, there is no internationally binding single and comprehensive legal instrument to regulate this. However, some regional or organization-based, principle-oriented instruments are available. (For example, the

28

Biosensors: Ethical, Regulatory, and Legal Issues

697

OECD Guidelines on the Protection of Privacy and Transborder Flows of Personal Data, 1980 (revised in 2013), African Union Convention on Cyber Security and Personal Data Protection, 2014.) In this legal vacuum at the international level, the recently enforced the European General Data Protection Regulation (GDPR) can be seen as the international standard and the most comprehensive legal instrument on personal data protection which was introduced to ensure a balance between the proportionate collection, use, and processing of personal information while imposing some strict responsibilities on the personal data user towards protection. In the European context, personal information collected and processed, and subsequently shared with others need to strictly comply with the GDPR requirements, and similarly in the context of other countries having data protection laws. Recently, in a case, the England and Wales High Court (Queen’s Bench Division) held that the oral disclosure of personal information does not constitute personal information (David Scott v LGBT Foundation Ltd [2020] EWHC 483 (QB) (3 March 2020)), but such a practice can be protected under the law on breach of confidence. Besides, the more personal data will be shared, the more ethical issues will arise. Patient’s confidentiality may be sacrificed, and information may be shared with insurance companies, relatives, and so on to settle inheritance claims, or for employment purposes, etc. Service providers may need to use some of the personal information for obvious reasons, but the patient may not be in favor of sharing these with others without their consent. Therefore, personal information collected through biosensors should not be shared with third parties such as insurance companies, financial institutions, or family members having adverse or conflicting interests and employers unless there is a legal and genuine need, “necessary” to protect the “vital interests” of the data subject, and proper legal due process is followed. European GDPR, the global standard of personal data protection in the world, has set some clear guidelines for anyone who will be collecting, using, and processing personal information of any data subject. GDPR provisions have some far-reaching legal consequences as this European law contain provisions on its extraterritorial applications and the European authority can apply the provisions of GDPR if goods and services are provided to European citizens and their behavior is monitored. There are technical solutions and best practices available which may be used to comply with the legal provisions. Medical data is sensitive personal information which is treated very seriously by most of the countries. For example, in the USA, though there is no single comprehensive domestic legislation on personal data protection like European GDPR, the Health Information Portability and Accountability Act (HIPAA) was enacted with the aim to create national standards to protect the disclosure of sensitive patient health information without the patient’s consent or knowledge. In addition to this, another very significant leap from the USA state of California, which has enacted the Consumer Privacy Act, 2018, a law frequently referred to as “Californian GDPR” due to its strict provisions, can be shared. Thus, the manufacturers and users of biosensor based products capable to collect personal information should consider such provisions carefully to avoid future legal challenges.

698

M. E. Karim

Cybersecurity With the burgeoning introduction of smart technological applications and transformation powered by big data, machine learning, artificial intelligence, 4-5G technologies, internet of things, etc. when smart features are embedded in all devices which enable them to track and monitor, the issue of privacy and personal information protection and cybersecurity should be taken into consideration seriously. These concerns will equally be present in the case of biosensors and related applications when these will be introduced in the market with digital features. These devices can collect and track the health conditions of human beings, which are sensitive in nature and thus, requires additional attention. Globally, human health data is one of the most targeted and desirable things the cybercriminals have been searching for (Luna et al. 2016). Because of the serious consequences of the leaked health data, various authoritative organizations have already come up with some guiding documents. In the USA, the FDA, with the aim to assist the industry and FDA Staff, has issued a Guidance Document on Postmarket Management of Cybersecurity in Medical Devices in December, 2016. The newly adopted European Regulation no. 745/2017 on Medical Devices and Regulation no. 746/2017 on in vitro diagnostic medical devices (collectively referred to as Medical Device Regulations) contain, in Annex I, provisions with the aim to ensure that the medical devices used in Europe, both pre-market and post-market period, are fit to meet the technological challenges linked to cybersecurity risks. A Guidance Document is also endorsed by the Medical Device Coordination Group, established under the European Regulation No. 745/2017, in December 2019. In the UK, after the Wannacry ransomware attack, the National Health Service published a range of guidance materials on protection against a cybersecurity attack. The International Medical Device Regulators Forum (IMDRF), the successor of GHTF, have also realized the importance of cybersecurity and has developed the “Principles and Practices for Medical Device Cybersecurity” in March 2020. Some countries, for example, Malaysia have included the issue within their Guideline on the Internet of Things. Concerned prospective entrepreneurs can consider to develop similar kinds of guiding documents following international standards and guidelines which are already adopted with regard to biosensor enabled products and services. Environmental Law Principles Findings of scientific studies revealed that the biosensors can be used in agriculture (Dar et al. 2020; Mufamadi and Sekhejane 2017) and to monitor different things in various components of the environment (Sha et al. 2017; Gupta and Kakkar 2020; Bidmanova et al. 2016; Saini et al. 2019). Biosensors can also be used to prevent environmental damages and thus, it assists the stakeholders to comply with environment related legal provisions. Researchers claim that the biosensors can help to comply with the European Union Water Framework Directive and Marine Strategy Framework Directive (Farré et al. 2009). It has been already shared that the legal provisions will follow the use and application, i.e., outcome of any product. Biosensors can be used as genetically modified whole-cell biosensors, i.e., bioreporter to be used in the environment, and can be used within or outside the laboratory as a contained device. It can again be

28

Biosensors: Ethical, Regulatory, and Legal Issues

699

designed to directly expose to the environment for a particular reason or survive in the environment for a longer period. Hence, the stakeholders need to consider the relevant legal provisions. For example, in such circumstances, the UNECE existing laws on Genetically Modified Organisms (Contained Use) Regulations 2000, EU Directives 2001/18/EC shall be applicable (French et al. 2011). Simultaneously, in the discussion of the legal consequences of biosensors applications, the two principles of environmental law, i.e., the precautionary principle and polluter pay principle shall be relevant. Some important aspects of these principles are highlighted below. Rooted in the German environmental law legal regime from the principle of foresight, and subsequently encapsulated in the historic Rio Declaration on Environment and Development, 1992, the “precautionary principle” is one of the most important options, sometimes confusing coupled with some criticism though (Foster et al. 2000), available in the hands of the stakeholders to thrive towards sustainable development while maintaining research and innovation related activities (Karim and Munir 2016; Gilbert 2020). In simple language, this principle stipulates that if an action can possibly cause harm, then the actor shall be ready to accept the liability as to the logical conclusion of the act even though the cause and effect relationship of the act and the consequence cannot be established precisely. This principle can be exercised to protect the environment, human, animal, and plant health for the possible adverse consequences of scientific and technological innovations including biosensors. While the precautionary principle is anticipatory in nature, the polluter pay principle makes the wrongdoers liable for causing damages to environmental components because of his actions or inactions. However, even though these principles are available in the national and international legal instruments, the regulators around the world have been facing some obvious challenges in implementing these principles. Some pragmatic steps have already been taken by the regulators towards the effective implementation of these environmental law principles. In the case of regulatory uncertainties, some regulators in developed countries now allow the entrepreneurs to consult them even before any initiatives are taken regarding the research, innovation, and commercialization of any goods or products. Some countries have also developed product registries to monitor the effects of the products, sometimes covering their pre-market to end-use environmental effects stages. While these are the case of the developed and industrialised countries, the proper implementation of these provisions in developing and least developed countries will require a holistic effort from the stakeholders.

Biosensors, Ethical, Legal, and Regulatory Responses: An Evaluation Biosensors have the prospect to introduce revolutionary changes in getting human right to universal health care and in achieving some of the goals included in the 2030 UN Sustainable Development Goals. While the ongoing R&D activities should be

700

M. E. Karim

continued for the betterment of mankind, consideration of relevant ethical and legal issues should not be ignored. It is believed that the manufacturers from only a few developed countries have the capacity to manufacture popular life-saving medical devices and the rest of the world uses them. To utilize the optimum benefits derived from the scientific and technological developments, different stakeholders have been working to take the lead in the market share. Here comes the issue of harnessing the ethics culture in the organization and also during the different stages of product development such as quality system, clinical investigation, pre-market approval, post-market management, etc. and the strict implementation of the law. An understanding of the importance of the teaching of ethical principles relating to science and technology in all stages of the academic curriculum, training and knowledge sharing with the stakeholders in the different stages of the life cycle of any scientific innovation and application is of paramount importance. Training of the employees and staff, incorporating the ethical issues, should get priority to avoid possible future legal issues. It can be seen that there are some broad legal provisions and principles available using which some of the projected and possible challenges with regard to biosensorbased products and applications can be theoretically addressed. At the international level also, there are some hard and soft legal instruments. Nevertheless, the implementation of these provisions will remain a challenge for the stakeholders. If we consider the applications of biosensors, then the different relevant international organization can be identified, and these organizations have been working to shape the regulatory framework of biosensors. For example, OECD has developed the Biosecurity Code, FAO has been working on the agricultural, UNEP has been considering environmental, and WHO has been looking at the public health aspects. These organizations have different mandates and have adopted some legal instruments too. A well-tailored synergy of these organizations can guide the stakeholders to avoid future legal and regulatory challenges and consequences. The legal regulation of health and safety concerns is rather a recent practice, even in the USA, such a move was started in the mid-1960s (Rabin and Picard 2018). If any initiative is taken to regulate biosensors legally, the regulatory definition of “biosensor” may be considered first and then the issue of regulation should get the priority. Presumably, it can be anticipated that it will take time to reach to the global consensus on the issue of regulation and its form. For example, it took 6 years for the FDA to develop the Guidance Documents for Glucose Meters. While everyone is working to reach to such a consensus, precautionary approach, which is principle based and anticipatory in nature, should be considered during every stage of cradleto-grave life cycle of biosensors. Already there are so many challenges, both man-made and historically culturally driven, for the manufactures embedded due to the differences in legal systems and regulatory cultures, government policies, regulatory burdens, and favorable market structure. The harnessing of an ethical culture throughout the product life cycle can ease some of these burdens. Regarding the issue of product liability, the implementation of legal provisions and fulfillment of legal requirements will be challenging in

28

Biosensors: Ethical, Regulatory, and Legal Issues

701

developing countries where the corruption practices are deep-rooted. Therefore, it will always be wise to use the devices prescribed or approved by the regulators e.g. the medical devices with ‘CE’ mark as available in Europe. Taking the opportunity of improper implementation of personal data protection laws, all personal information collected can be stored in various forms and can be used subsequently. Here, only ethical practices can guide to tackle the issue. There are technical solutions available too to monitor every single stage of any product. Privacy by design and default, privacy impact assessment, and encryption technology are some of the regulatory compliance-related options widely practiced in developed countries. In order to prevent the consequences of cybersecurity attacks, some organizations have been offering Cyber insurance facilities. It is crucial to include the stakeholders and involve the mass people in the regulatory process. Involving the public in regulatory decision-making is an important development. In the Europe, the UNECE Convention on Access to Information, Public Participation in Decision-Making and Access to Justice in Environmental Matters, popularly known as the Aarhus Convention on Public Participation, signed on 25 June 1998 was adopted. Though the effectiveness and their implementation will remain a concern, some countries have also adopted national mechanisms in this regard too. This is even crucial after the GMO incident in Europe, where such a wonderful technology could not be successful as it failed to garner public support. Traditionally, tort law principles may be viewed as an available option in ensuring and improving the safety of the product including medical devices, and the regulation of malpractices. Therefore, an understanding of the tort law principles and provisions on negligence, vicarious liability, and hospital corporate negligence, etc. will also be beneficial for the stakeholders. Unfortunately, tort law is not a developed one in most jurisdictions and even in a developed country with a civil law legal system, where legal provisions are generally codified, such as Japan, there are very few case laws relating to medical devices (Altenstetter 2017). In a developed country such as the USA, it was found that tort law principles play a limited and complementary role (Rabin and Picard 2018). It can be anticipated that suits on software liability and medical malpractice will be increased in the future because of the wrong reading of biosensors and wrong detection of disease. Therefore, the FDA’s recently proposed framework for regulating prescription drug-use-related software, the new EU Medical Device Regulation, ISO adopted international standard such as ISO 35.240.80 – IT applications in health care technology including computer tomography can be considered to minimize liabilities.

Last Few Words With the existing serious challenges and problems triggered by, e.g., climate change, the world community has been experiencing a difficult and unprecedented time since the outbreak of the novel COVID-19 virus, the perceived consequences of which will have to be borne by the global community and the coming generations for

702

M. E. Karim

decades. A properly coordinated and effectively regulated scientific and technological innovations promise to generate some comforts in solving some of these challenges and problems. Therefore, research and innovation activities must be going on but within the strict but preferably principle-based anticipatory regulatory framework. Biosensors and biosensor powered applications are very promising and can be used as a catalyst in achieving some of the UN Sustainable Development Goals, though there are some ethical, regulatory, and legal challenges, some of which are highlighted in this chapter. An understanding of the relevant legal and ethical issues is important for the safe, sustainable, and responsible development of scientific and technological innovations, including biosensors as we have the experiences of many technological applications that were introduced with a huge prospect but could not be successful eventually. Existing published literature projects that various means, tools, ways, and techniques are used by the regulators across the world to promote the scientific and technological research, innovations, and commercialization-related activities, some of which can be tested in case of biosensors. The legal regulations are either process based, or product based, general and specific. Based on applications, the legal, ethical, and regulatory issues may take a different route. Theoretically, one may find either general or specific legal and regulatory provisions on relevant scientific and technological innovations and applications. While these may be present in the legal and regulatory framework of the developed countries, these may not be adequate in the context of developing or the least developed countries. There is some very general legal framework on the protection of environmental components, consumer protection, product liability, etc. which should be observed at all the time, while there are product specific legal provisions also. Taking into account the findings of the product LCA, legal issues can also be discussed in pre-product, production, and postproduction stage. Moreover, there are tort law provisions too. Nevertheless, the interpretation and proper implementation of these provisions will be a challenge as regulators respond after and usually follow the innovations. The regulators, however, should take a cautious move while considering the adoption of a suitable regulatory option since overregulation may give a wrong message to the market and prevent the entrepreneurs to take the risks of innovations. To do so, the regulators can consider the “regulatory impact assessment” which may assist them to ensure the balance between the interest of the entrepreneurs, human health, and environmental components. A series of dialogues among the stakeholders exploiting the secured state-of-the-art technologies should be considered in this process. After that, following the broad principle-based anticipatory and soft law mechanisms, some regulatory tools such as guidelines and code of conduct, code of ethics or code of practice instilling the provisions of the precautionary principle and polluter pay principle can be considered. The industries should also continuously thrive to gain and maintain the trust and confidence of the consumers in an ethical and transparent manner. In the whole process, the issue of personal data protection and cybersecurity, ethical and technical training of the employees should be a normal consideration to avoid possible “digital pandemics,” the consequences of which will

28

Biosensors: Ethical, Regulatory, and Legal Issues

703

be even disastrous than this recent natural one i.e. Covid-19 due to some inherent limitations and challenges in cybercrime investigation and trial. The developed countries need to continue to their ongoing support and perform some additional moral responsibilities in the development of legal, ethical, and regulatory frameworks of scientific applications as we all have the common but differentiated responsibility to attain sustainable development which will shape our common future. Already there are some international forums available, a coordinated movement of these will be the key to success in this regard. The United Nations Interagency Task Team on Science, Technology and Innovation for the SDGs, as launched as a means to implement the UN Sustainable Development Goals and the Global Partnership, should be utilized in this regard and the issue of biosensors can be included within their agenda.

References Altenstetter C (2017) Medical technology in Japan: the politics of regulation. Routledge Bhalla N et al (2016) Introduction to biosensors. Essays Biochem 60(1):1–8 Bidmanova S et al (2016) Fluorescence-based biosensor for monitoring of environmental pollutants: from concept to field application. Biosens Bioelectron 84:97–105 Chen Y-J et al (2018) A comparative study of medical device regulations: US, Europe, Canada, and Taiwan. Ther Innov Regul Sci 52(1):62–69 Dana GV et al (2012) Four steps to avoid a synthetic-biology disaster. Nature 483(7387):29–29 Dar FA, Qazi G, Pirzadah TB (2020) Nano-biosensors: NextGen diagnostic tools in agriculture. In: Nanobiotechnology in agriculture. Springer, pp 129–144 Donders Y (2011) The right to enjoy the benefits of scientific progress: in search of state obligations in relation to health. Med Health Care Philos 14(4):371–381 Faden RR, Beauchamp TL (1986) A history and theory of informed consent. Oxford University Press, New York Farré M et al (2009) Sensors and biosensors in support of EU directives. TrAC Trends Anal Chem 28(2):170–185 Fatoyinbo HO, Hughes MP (2012) Biosensors. In: Bhushan B (ed) Encyclopedia of nanotechnology. Springer Netherlands, Dordrecht, pp 329–345 Foster KR, Vecchia P, Repacholi MH (2000) Science and the precautionary principle. Science 288 (5468):979–981 French CE et al (2011) Synthetic biology and the art of biosensor design. In: The science and applications of synthetic and systems biology: workshop summary. National Academies Press, Washington, DC Gerke S et al (2019) Ethical and legal issues of ingestible electronic sensors. Nat Electron 2(8):329– 334 Gilbert SG (2020) Precautionary principle. In: Information resources in toxicology. Elsevier, pp 489–494 Gold BD (1993) Congressional activities regarding misconduct and integrity in science. In: Responsible science: ensuring the integrity of the research process, committee on science, engineering, and public policy, panel on scientific responsibility and the conduct of research. National Academies Press Griffiths D, Hall G (1993) Biosensors – what real progress is being made? Trends Biotechnol 11 (4):122–130 Guinée JB et al (2017) Setting the stage for debating the roles of risk assessment and life-cycle assessment of engineered nanomaterials. Nat Nanotechnol 12(8):727

704

M. E. Karim

Gupta S, Kakkar V (2020) Development of environmental biosensors for detection, monitoring, and assessment. In: Nanomaterials and environmental biotechnology. Springer, pp 107–125 Kansu E, Ruacan S (2002) Research ethics and scientific misconduct in biomedical research. In: Research and publishing in neurosurgery. Springer, Vienna, pp 11–15 Karim ME (2005) Citizen’s right to privacy: Reflection in the international instruments and national laws. Bangladesh J Law 9(1–2):35–68 Karim M (2010) Health as human rights under national and international legal framework: Bangladesh perspective. JE Asia Int’l L 3:337 Karim ME, Munir AB (2016) Nanotechnology and international environmental law: a preliminary assessment. In: Advanced environmental analysis, pp 348–380 Karim ME et al (2015) Too enthusiastic to care for safety: present status and recent developments of nanosafety in ASEAN countries. Technol Forecast Soc Chang 92:168–181 Karim ME et al (2018) Polymer nanocomposites and related legal issues: an overview. In: New polymer nanocomposites for environmental remediation. Elsevier, pp 679–698 Kumar MN (2008) A review of the types of scientific misconduct in biomedical research. J Acad Ethics 6(3):211–228 Levi-Faur D (2011) Regulation and regulatory governance. In: Levi-Faur D (ed) Handbook on the politics of regulation. Edward Elgar Publishing, Cheltenham, pp 3–21 Luna R et al (2016) Cyber threats to health information systems: a systematic review. Technol Health Care 24(1):1–9 Mallardi V (2005) The origin of informed consent. Acta Otorhinolaryngol Ital 25(5):312–327 Marmot M et al (2012) WHO European review of social determinants of health and the health divide. Lancet 380(9846):1011–1029 Mufamadi M, Sekhejane P (2017) Nanomaterial-based biosensors in agriculture application and accessibility in rural smallholding farms: food security. In: Nanotechnology. Springer, pp 263– 278 Munir AB, Siti HM, Md Ershadul K (2018) Data protection law in Asia. Sweet & Maxwell/ Thomson Reuters Muralikrishna, I.V. and V. Manickam, Chapter five – life cycle assessment, in Environmental management, I.V. Muralikrishna and V. Manickam, Editors. 2017, Butterworth-Heinemann Oxford. p. 57–75 Olson N, Bae J (2019) Biosensors—publication trends and knowledge domain visualization. Sensors 19(11):2615 Owen R (2014) The UK Engineering and Physical Sciences Research Council's commitment to a framework for responsible innovation. J Resp Innov 1(1):113–117 Rabin RL, Picard AJ (2018) Reassessing the regulation of high-risk medical device cases. DePaul L Rev 68:309 Saini R et al (2019) Advances in whole cell-based biosensors in environmental monitoring. In: Tools, techniques and protocols for monitoring environmental contaminants. Elsevier, pp 263– 284 Salieri B et al (2018) Life cycle assessment of manufactured nanomaterials: where are we? Nano Impact 10:108–120 Sha R, Badhulika S, Mulchandani A (2017) Graphene-based biosensors and their applications in biomedical and environmental monitoring. In: Label-free biosensing. Springer, pp 261–290 Smith R (2016) Statutory regulation needed to expose and stop medical fraud. Br Med J Publ Group Steneck NH (1994) Research universities and scientific misconduct: history, policies, and the future. J High Educ 65(3):310–330 Steneck NH, Bulger RE (2007) The history, purpose, and future of instruction in the responsible conduct of research. Acad Med 82(9):829–834 Stevenson H et al (2019) A rapid response electrochemical biosensor for detecting Thc in saliva. Sci Rep 9(1):1–11 Verma N, Singh M (2005) Biosensors for heavy metals. Biometals 18(2):121–129

28

Biosensors: Ethical, Regulatory, and Legal Issues

705

Vilariño N et al (2009) Use of biosensors as alternatives to current regulatory methods for marine biotoxins. Sensors 9(11):9414–9443 Wells F, Farthing M (2008) Fraud and misconduct in biomedical research. CRC Press Wyndham JM, Harris T (2014) Introduction: science, technology and human rights: lessons learned from the right to water and sanitation. Sci Eng Ethics 20(4):827–831 Yáñez-Sedeño P et al (2014) Biosensors in forensic analysis. A review. Anal Chim Acta 823:1–19

University-Industry Relationships for the Development and Commercialization of Biosensors

29

Christina G. Siontorou

Contents The Biosensor Innovation Hub and the Role of Industry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Established University Clusters in Cell-Based Biosensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Scientific Frame and the Orientation of the Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Collaborative Framework . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Innovation Environment and the Role of Industry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Concluding Remarks on Current Aspects and Future Trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

708 710 711 711 713 716 718 720

Abstract

Cell-based biosensing is a significant part of the biosensor research, presenting many advantages for both, university research and industrial development. With some products already in the market, the majority of the university-produced technology remains largely unexploited. The present work uses the huge scientific (academic output) and technical (patent applications) literature on the subject in order to study the cell-based biosensor innovation system. Emphasis is given on the science and the technology base in an attempt to highlight the most prominent and promising research pathways toward the industry. The results provide two possible direct links between the university and the industry: nanobioelectronics technology for environmental whole cell biosensing and synthetic biology tools for clinical detection. A new pathway just emerging may become a significant link in the near future: cell-mimicking artificial cells; future applications may actually include the entire range of biosensor research scope: sitespecific and intended-use optimized detectors for infield and online simultaneous monitoring of a large number of target analytes. C. G. Siontorou (*) Laboratory of Simulation of Industrial Processes, Department of Industrial Management and Technology, School of Maritime and Industry, University of Piraeus, Piraeus, Greece e-mail: [email protected] © Springer Nature Switzerland AG 2022 G. Thouand (ed.), Handbook of Cell Biosensors, https://doi.org/10.1007/978-3-030-23217-7_25

707

708

C. G. Siontorou

Keywords

Cell biosensor · Innovation clusters · Technology transfer · Research network · University-industry relationship · Science mapping · Citation analysis

The Biosensor Innovation Hub and the Role of Industry Industry relies on innovation. It translates ideas into products with a view to fulfilling, or even changing, market needs. These products derive from a system built on intra- and interorganizational networks, some nodes of which critically involve university functions: knowledge generation and repository, education and training center for industry, and instrumentation and infrastructure development (Debackere and Veugelers 2005). The involvement of academia is more pronounced in the science-intensive products: the associations of industry with universities result in the commercial translation of the academic output (front-end of innovation) or serve the problem-solving aspect of technology commercialization (aft-end of innovation). While the former is most common in pharmaceuticals, where, after the university output is patented, the proprietary rights are given to industry (Cockburn and Long 2015), the latter can be mostly seen through university– industry partnerships in funded research programs for energy, food, and biomedical products (Hanberger and Schild 2004). A wide variety of factors determine the strength and pace of these associations, either preserving them as links (i.e., loose associations with minimal university–industry interaction) or turning them into productive relationships within an interactive innovation system. The most critical factors involve research policies, market needs, R&D costs, academic entrepreneurship, human resource transfer, property rights, and innovation environment (Perkmann and Walsh 2007). Many studies have been published on university–industry relations, focused either on the product of innovation and its market base (see e.g., Chang et al. 2009) or the process of innovation and its science base (see e.g., Oliveira and Rozenfeld 2010). Some recent studies on the technology base, that is, the knowledge that is actually transferred from the university to industry (Siontorou and Batzias 2010, 2013), used the biosensors domain as a paradigm in order to examine how the research pathways that dominate university efforts and shape academic clusters affect the rate of commercialization. The biosensor concept was set at early 1970s and involved the reproduction of nature’s sensing processes (i.e., of natural chemoreception) in vitro with a view to developing reliable detectors for medical, industrial, and environmental applications (Siontorou and Batzias 2010). This novel concept can be materialized through the successful interconnection of two components: a biological moiety (bioelement) that reacts with the target chemical compound (analyte) and produces a biochemical information and a chemical transducer that translates the biochemical information into an electric signal. Quite appealing as per all aspects, this approach soon turned into the huge market success of the still thriving glucose sensor (Siontorou and Batzias 2013), dragging along many researchers that were eager to immobilize as

29

University-Industry Relationships for the Development and. . .

709

many biological moieties as available (or as becoming available), on all the transducers they could access, with as many methods as they could invent. While industry was reluctant to adopt any other biosensor format, university research was intensifying, forming clusters (especially in Europe) and shaping a culture. The reproduction of the convoluted natural chemoreception was becoming more efficacious as new knowledge on biological systems was produced and new tools for engineering and handling the biology were made available. Notwithstanding, the interplay between the bioelement and the transducer proved to be quite challenging; while the biochemical signal could be definitely produced, the limits of the transduction technology and the lack of signal processing means rendered the goal of comparable-to-nature detection almost unattainable (Siontorou et al. 2010). The net result was the structuring of a diverse, fast-growing, and dynamic research network that showed a remarkable capacity to absorb exogenous technological change in order to solve fast its scientific and technical problems (Siontorou and Batzias 2013). Without any intend to oversimplify the process, the strong university–industry collaboration that yielded the glucose sensor was mainly the result of three factors: the pressing market need, the absence of comparable alternatives, and the supportive innovation environment (Siontorou and Batzias 2013). As the glucose sensor was intended for a worldwide market, industry was fast to assume the role of coordinator, setting the scopes of research. Academic entrepreneurship and human resource transfer proved inadequate to address the ever-increasing market demands: performance, reliability, convenience, low cost, noninvasive monitoring, and artificial pancreas. Thereby, the revived involvement of academia in this endeavor was indispensable, requiring, however, a significant shifting toward multidisciplinary research and engineering processes (Siontorou and Batzias 2010). The biosensor domain largely failed to position a large part of its diverse research yield within a glucose-like market frame. Research efforts focused on solving problems with fabrication, miniaturization, performance, and functionalization (Siontorou and Batzias 2013). The early adoption of nanotechnology addressed many issues and opened new avenues for research and applications (Siontorou et al. 2017a), without satisfactorily resolving signal amplification and signal processing issues. The use of cells (algae, bacteria, yeasts, protozoa) as bioelements has been long proposed and extensively studied due to the benefit of a built-in optimized and effective intracellular mechanism for signal processing, amplification, and transmission. Further, device fabrication is more simple and operational stability is better than those of other biosensor systems. In effect, cell-based platforms offer many advantages compared to enzymes, antibodies, aptamers, or other moieties, as per both, bench-scale experimentation and scale-ups. Cells can be immobilized on transducers with easier processes than those required for enzymes or DNA (D’Souza 2001). The optimal environment for the function of the bioelement is largely included within the bioelement (cofactors, regulators, stabilizers, etc.); thus, experimentation and operational costs are significantly reduced (Su et al. 2011). More importantly, cells can self-heal and self-produced and, also, they can replenish on their own most reagents consumed during analysis (D’Souza 2001); these properties almost ensure the regeneration and the reusability of the sensor. Cells already

710

C. G. Siontorou

possess multiple sensors and cascade mechanisms; provided that the biosensor developers have the means to discriminate the output signals, multiarray analysis is feasible (Melamed et al. 2012). Alternatively, any cell response (even cumulative or indistinct) might be very informative as to how environmental stressors and factors affect the ecosystem (Stenger et al. 2001). Operation-wise, a critical problem in the transition of biosensors to industry is the inherent fragile nature of the biological moieties. Shelf and operational lifetimes are determined by the stability of the moiety under storage and operation (Siontorou et al. 2017b); in general, lifetimes are prohibitively short. Cells have the ability to adapt and tolerate a wide range of external conditions and even abrupt changes (Stenger et al. 2001). In theory, and, to a certain extent, on the bench, as well, the above benefits have oriented cell-based biosensing toward environmental and medical applications. Some toxicity sensors are commercially available (Mongra and Kaur 2012), from many USA, European, and Japanese companies that manufacture and distribute analytical instruments, but the biological oxygen demand (BOD) biosensor is at present the most successful product, with more than 15 analyzers in the market (Bahadir and Sezgintürk 2015). The present work uses the huge scientific (academic output) and technical (patent applications) literature on the subject in order to study the cell-based biosensor innovation system. Emphasis is given on the science and the technology base, in an attempt to highlight the most prominent and promising research pathways toward the industry.

Methodology This work on cell biosensors covers the period 1990–2017, so that the analysis includes the impact of both, the biotechnology and the nanotechnology eras. Data search strategy was based on the algorithm developed by Mogoutov and Kahane (2007). The Web of Science Core Collection revealed 10,251 papers, whereas the EspaceNet Worldwide Database provided 1,329 patents. In order to avoid terminology limitations, multiple search queries were used (Batzias and Siontorou 2012): cell, bacterial, microbial, algae, yeast, protozoa, etc. The produced data set was categorized according to the type of cell, the transduction strategy, and the field of application to form subject categories in order to highlight the propagation of the research trends. These categories were then grouped into macrocategories: analytical chemistry, applied microbiology, materials science, cell biology, device integration and engineering, etc., in order to draw the frame within which the research trends were evolving. Double entries had to be evaluated and deleted manually. Multi- and cross-disciplinarity and the extent of the collaborative framework was evaluated by identifying coauthorship networks using the Co-auth.exe program (available online at http://www.leydesdorff.net/software/coauth/index.htm). Formal university–industry collaboration was similarly assessed. The study parameters included, also, funding agencies in order to draw conclusions on the supportive innovation environment (at national and international level) and to assess informal or forthcoming industry involvement.

29

University-Industry Relationships for the Development and. . .

711

The technology base has been elucidated using the methodology proposed by Lee and Su (2011) for mapping scientific knowledge. Briefly, this methodology draws the structure of knowledge in qualitative and quantitative terms looking into the degree and extent of the scientific input (cited articles) and the scientific output (article citations). The selection of information to be included in mapping produces different knowledge maps to reflect dominant knowledge structures and structures with a high tendency for dominancy. Especially the latter has been proven efficient in technology improvement and optimization (Siontorou and Batzias 2013), increasing the likelihood of industry involvement (Chang et al. 2009).

Established University Clusters in Cell-Based Biosensing The Scientific Frame and the Orientation of the Research Academic research on cell biosensing, all forms and formats included, represents 21.28% of the biosensor output. The rate of publication shows an increasing trend (Fig. 1). There exist two distinct disruptions, occurring around 2004 and 2015. These signify three distinct phases in research: the first covers the period 1990–2004, the second extends till 2015, and a third is just starting to emerge. Looking into the content of publications, the first phase is biotechnology-driven and follows an almost steady rate of increase (32.87% annually, on average). In a primary subphase, extending till late 1990s, research was mostly exploratory, trying to establish possibilities and prospects. Using electrochemical and optical transduction platforms, many cell types were tested with a view to broadening the range of

Fig. 1 Publications on cell biosensors in the Web of Science Core Collection for the period 1990– 2017. The disruption that occurred around 2004 signifies a shifting toward nano-tools, whereas a new disruption in 2015 accommodates the principles of synthetic biology

712

C. G. Siontorou

bioelements. Setting the proof of concept, preliminary data were used to highlight the potential application domains and intended use; environmental monitoring (see e.g., D’Souza 2001) and food safety (see e.g., Patel 2005) have been posted as almost within reach. The capabilities for cell engineering developed by the microbiology domain optimized bioelements and intensified considerably the environmental applicability that triggered a new subphase and a clustering around the bioluminescent (luxCDABE) gene and fluorescent detection (see e.g., Aisyah et al. 2014). This topic covers almost 66% of the total volume of publications for that period. Regardless, research became more focused toward cell immobilization problems and alternative transduction means. Briefly, two strategies to constrain cells on the surface of the transducer have been proposed, none of which is effective: (i) covalent binding or crosslinking (chemical method) and (ii) physical adsorption or entrapment (physical method). The former uses chemicals and harsh reaction conditions in order to induce the formation of covalent bonds between cell wall components and the transducer (D’Souza 2001). Crosslinking uses milder conditions but the viability of cells may be equally affected (Lei et al. 2006). Physical methods are more cell-friendly but may lead to desorption and poor sensor stability (Lei et al. 2006). Alternative transduction takes advantage of cell respiration in order to detect the metabolic activity via oxygen consumption (Tzoris and Hall 2006) or the inhibition of this activity via carbon dioxide production (Vaiopoulou et al. 2005). Toward the end of the biotechnology phase, the technology problems have been determined: (a) the inherent tolerability of cells toward external sensors may hinder response, thus sensitivity is relatively poor (D’Souza 2001); (b) the cellular mechanisms may produce a single response for a set of stressors, thus selectivity is impaired (Stenger et al. 2001); (c) the rate of diffusion of the analyte through cell walls is slow, thus response times are prolonged (Su et al. 2011); (d) the intrinsic heterogeneity, generated by factors such as cell cycle state and age, is largely uncontrolled, thus the reproducibility and repeatability of the sensor is impacted (González-Cabaleiro et al. 2017); (e) cell manipulation, especially of circuitry, is difficult, thus the signal transmission may be suboptimal (Swain et al. 2002); (f) the immobilization of cells affects adversely their viability, thus sensor durability is limited (Lei et al. 2006); (g) despite facile handling, contamination during storage cannot be overruled (Lim et al. 2015); (h) the growth phase affects genetic circuit behavior, increasing the sensor’s noise levels, thus reproducibility is reduced (González-Cabaleiro et al. 2017); (i) the exploitation of the cells’ sensing machinery is based on the ability of the analyte to cross the membrane, thus a large range of analytes are excluded (Su et al. 2011); (j) some target analytes, especially environmental pollutants, may be toxic to the sensing cells, thus applicability is limited (Jarque et al. 2016). Nano-tools and processes launched a new phase of exponential growth. The technology problems have been significantly addressed through transduction modifications: the traditional transducer platforms could be minimized using nanoparticles and conducting polymers, processes could be semi- or fully automated,

29

University-Industry Relationships for the Development and. . .

713

different types of cells could be co-immobilized, and cell engineering became more effective. Lab-on-chip technology allowed the transducer cell integration and the emergence of nano-bioelectronics (Duan and Lieber 2015), while chip manufacturability was improving (Kou et al. 2016). Microbial fuel cells and their ability to convert the substrate into electricity opened new research paths for strategic sensing (Du et al. 2007). Advances in instrumentation allowed the high-throughput analysis of extracellular metabolites (Wang et al. 2014) and the recognition of genetic elements (Park et al. 2013), putting forward the sensory-regulative biosensor principles. Application opportunities in the field of clinical diagnostics gained many followers (see e.g., Yoo and Lee 2016). The capability to manipulate cells offered cellbased biosensing a clear competitive advantage over other types of biosensors and some conventional techniques, as well; microbes can be engineered to detect and differentiate between different strains of pathogens, giving an accurate result in shorter time than conventional culture isolation and identification (see e.g., Xu et al. 2017). A shifting becomes apparent as regards the analytical aspects of sensors intended for environmental applications. Instead of putting effort in improving precision, accuracy, detectability, sensitivity, and selectivity, new formats have been proposed, investing on a silently implied interplay between assay cost and level of accuracy required. Environmental monitoring refers to wide, inaccessible areas; elaborate testing under strict analytical protocols may be (and usually does) too costly to implement in a long term (Siontorou et al. 2017a). A more realistic scheme that does not affect overall reliability significantly involves preliminary assessments with accurate-enough means to be followed, if necessary, by sampling and lab testing. The development of simple, fast, and disposable (one-shot) sensors, such as colorimetric strips, patches, or probes (see e.g., Struss et al. 2010), could be appealing to the market. This trend has been, also, observed in other types of biosensors, in an effort to gain a competitive advantage over the well-established (and, in some cases, more suitable) conventional technology (Siontorou and Batzias 2013). A new phase is now starting to arise, focusing on the bioelement per se: synthetic biology. Going beyond living cells, cell parts and extracts can be isolated to provide all the necessary machinery for biosensing (Chang et al. 2017). These microbially derived systems seem to be the next-generation biosensing platforms enabling target-specific customizations and advanced signal processing.

The Collaborative Framework Collaborations within and between departments and universities are very important for enhancing the absorptive capacity of the sector, as well as for supporting sectoral transformations that may lead, even in the long run, to the transition of the produced knowledge to industry (Pandza and Holt 2007). Sharing the writing of a publication is a common means to state a collaboration (Newman 2004). Coauthorship analysis has been used herein in order to assess the collaborative network formed (either as an

714

C. G. Siontorou

opportunity or as a need) within the academic clusters and, further, to highlight any university–industry collaboration patterns. On the data set produced by Web of Science, the coauthorship networks have been identified with the program Co-auth.exe. Figure 2 presents the contribution of (mono)disciplinary, multidisciplinary, and university–industry collaboration per year of publication. Industrial involvement remains quite low for the period studied, while a slight increase is observed toward the nanotechnology shifting. During the 1980s, industry affiliations could be mostly seen in food quality topics. Toward late 1990s, environmental monitoring, especially toxicity assessments, received also the attention of industry while biomedical applications appeared quite later. This trend is similar with that observed for other biosensor platforms (Siontorou and Batzias 2013). Interestingly, the patterns on intra-academia collaboration are more prominent, especially after 2003. Looking into the content of the domain, multidisciplinarity in cell-biosensing goes beyond the boundaries set by the biosensor domain. Scientific mapping is illustrated in Fig. 3. Two trails become obvious. The first trail evolves from analytical chemistry (device development) and it gradually incorporates biophysics (analyte-cell and cell-transducer interface), electrochemistry that turned to optics (the monitoring of cellular stress response), materials (cells and transducers), device integration (miniaturization and automations), and mathematical modeling. This trail produced a significant output in biomedical diagnostics and imaging, food technology applications, and chemical engineering (production) improvements.

Fig. 2 Disciplinary (light shaded) and multidisciplinary (blue shaded) publications, as well as the percentage of papers with an industrial partner (dark shaded) on cell biosensors, retrieved from the Web of Science Core Collection for the period 1990–2017. The Co-auth.exe program was used to highlight the intrinsic (intrauniversity) and extrinsic (university–industry) collaborations

29

University-Industry Relationships for the Development and. . .

715

The second, parallel trail has a strong base on applied microbiology, involving research on physiology, plant science, and toxicology that evolved into tissue engineering and cell biology (Fig. 3). The knowledge level gradually deepened toward genetics and synthetic biology with a lesser but distinct dominion in environmental science. A significant yield toward energy (microbial fuel cells), biomedical engineering, and drug testing (pharmaceuticals) as well as agricultural engineering almost drew an independent path. The impact of nanotechnology is especially prominent in the analytical chemistry trail, entirely focused on materials. A direct link between the trails involves mathematical modeling. Modeling comes as a necessity for effective device integration (according to engineering principles) in an attempt to accommodate the intracellular circuitry in the response of the sensor. To advance modeling, information on intracellular functions is retrieved from the microbiology domain. Modeling is expected to impact both trails equally. It could advance the elucidation of cellular processes to support synthetic biology (Tomita 2001) and/or enhance the reliability of biosensors with target-specific cells and cell-relevant immobilization strategies (Ben-Yoav et al. 2013). The endeavor has proved quite challenging. The model should efficiently represent all intracellular and extracellular processes that take place, physical and chemical, in order to justify its usefulness for sensor design and forecasting. All information about the biochemical reactions and the transport of mass within the cell must be reliably incorporated in a comprehensive model (Fig. 4). As a consequence, modeling should incorporate and interrelate information from different disciplines and scales (micro-, meso-, and macroparameter identification). The mathematical complexity increases drastically as the

Fig. 3 Cell biosensor research trends (1990–2017), putting emphasis on the trails with the higher significance, as determined by scientific and citation mapping. The size of bubbles is analogous to the publication volume for each trail per time window

716

C. G. Siontorou

number of processes increase and the size of compartments decrease (Goldberg et al. 2018). The complexity of cell biosensing, as revealed from computational modeling, best demonstrates a critical hurdle in the transition of cell-based technology to industry: too many levels of control are required for effective sensor development and reliable sensing. Remarkably, a second direct link between the analytical base and the microbiology base (Fig. 3) has been formed around environmental engineering. The specific application field does not only demonstrate a multidisciplinary network but actually starts to move toward cross-disciplinarity, significantly transcending traditional domain boundaries. This shifting could be indicative of a strengthening of the university–industry relationship in the near future.

The Innovation Environment and the Role of Industry Patent search in the EspaceNet Worldwide Database (Fig. 5) revealed that the intention to invest on any given technology output in cell biosensing is significant yet quite low. Early patents had to do with a variety of cells immobilized on

Fig. 4 The physical and chemical parameters required for comprehensive modeling. The complexity of cell biosensing is indicative of the levels of control required for effective sensor development and reliable sensing. Blue boxes represent cell functions, gray boxes the chemistry of the sample to be analyzed and the properties of the target analyte, while the red boxes show the sample/analyte-cell interrelation

29

University-Industry Relationships for the Development and. . .

717

Fig. 5 Patents on cell biosensors in the EspaceNet Worldwide Database for the period 1990–2017

electrochemical transducers with some possible applicability. The next stream of patenting is characterized by a distinct shifting toward bioluminescence and optical platforms. The nanotechnology-driven suggestions almost revived the early phase: the cell biosensors of the past became on-chip and integrated platforms of today. The current trend in patenting shifts interest entirely on microfluidic chips and fuel cells. The comparison between scientific publications (Fig. 1) and patents (Fig. 5) indicates that an initially observed lag time in patenting, probably designating the low rate of transition to industry, is currently diminished; knowledge production and patenting could be thought of as concurrent. At any worth given on this supposition, microbial fuel cells could be an attractive sector for industry, although scientific mapping does not justify, at the moment, its long-term viability. In effect, advantages are currently fewer than the disadvantages, especially in the presence of more appropriate alternative technologies (see e.g., Santoro et al. 2017). Nonetheless, the innovation environment, within which any transition to industry will occur, cannot be presently considered as supporting. National funding is usually an indicator of coordinated efforts toward the development of an environment to support the local industry toward innovation (Guerzoni et al. 2014). Figure 6 presents the number of university publications (from the Web of Science database) per country or continent (in order to enhance visualization), as well as the percentage of funded research in each case. At a first glance, the big players in cell biosensing include the USA, China, and EU, indicating the location of the significant university clusters worldwide. At a second glance, Chinese institutions financially supported 70.8% of university research, through a variety of programs, some of which involved industrial partnership. The university clusters in the USA received less funding (34%), 88% of which came from private sources. EU has supported only 7.5% of research, although 21.7% of cell biosensing research is taking place within the continent. The Far East has currently a small contribution in the scientific output, yet a large percentage received national funding.

718

C. G. Siontorou

Fig. 6 Scientific (university-derived) publications per country and the percentage of financial support received

This picture clearly indicates a differentiation of scopes and intends. Western universities act as the regulators of the scientific frame, and Eastern countries assume the role of the mediators to industry. The significance of this differentiation is very difficult to be comprehended at this point. Partial science mapping reveals, more or less, the trends shown in Fig. 3; thus, no reliable conclusions can be drawn on what aspects of cell-based sensing will be first to reach the market or whether the commercialization process will critically involve the university. More data are required before a link between academia and industry would be clear enough to be used to denote a loose or more tight alliance.

Concluding Remarks on Current Aspects and Future Trends The marketability of university research is a subject that is just started to emerge as a research topic. Presently, the links between universities and industry can be only studied retrospectively; the prospective implementation, however, of any mechanisms identified is not a reliable means to make predictions or draw future trajections. New methods and tools are required, more suitable to handle the dynamic nature of academic innovation routes. Even in the cases of already marketed or ready-to-market products, such as sensors, the use of market-base technology foresight has been proven inappropriate to reliably comprehend academic trends (see e.g., the recent work by Andersen et al. 2004).

29

University-Industry Relationships for the Development and. . .

719

University research generally forms a self-regulating frame that supports the production of science. Each domain or subdomain is strongly based on its history and established infrastructure to shape the research scopes and formulate the output. Looking into the evolutionary trends, certain conclusions can be safely drawn as regards where the research is heading in the near future. The industrial involvement requires a more coherent frame as well as obvious benefits over alternative technologies. The way that these two will interact remains vastly uncertain until the interaction actually occurs. Technological change and innovation remain inextricably linked with dense, multidisciplinary networks in academia (see e.g., Lowe 1993), with or without the mediation of knowledge brokers (Siontorou and Batzias 2010). The biosensor domain is an excellent paradigm of such a network, continuously opening up new and excited pathways for innovation. As evident, the research environment becomes very important in supporting the direct transfer of knowledge and its transformation into products; taking as example the US innovation framework that promotes academic spin-offs, licensing, and collaboration or consultancy schemes with the private sector (Mowery et al. 1999), it came to no surprise the huge market success of the glucose sensor. In effect, the university to industry pathways may become more apparent and even more effective, when this interplay between science, technology, and innovation is seen under the cognitive tools of innovation studies (Fagerberg and Verspagen 2009). Nonetheless, the present study has drawn some conclusions for cell biosensing products. The science part still lacks critical knowledge in cellular processes. No one can currently predict exactly how a cell will react in any given stressor within a full of stressors environment. This became apparent at the very first attempts to model the system. Considering every subsystem in isolation, detail modeling and optimization is easy; the harmonized function of all these subsystems is not. The market requires reliable and user-friendly detectors. Toward that end, nanotools are expected to solve many problems in device construction, whereas mathematical modeling will support operation-specific design and signal processing until the limits of the technology are reached. Nano-bioelectronics, for example, deal with the integration of nanowires with cells, opening avenues in noninvasive monitoring and remote sensing (see e.g., Duan and Lieber 2015). Thus, at this period a niche market that cell biosensors could be proven competitive is, by far, that of environmental applications. Looking at the dominant research pathways, the impact of synthetic biology is becoming substantial. It is expected to facilitate microbiology and analytical chemistry, building a cell-derived new frame that might advance biomedical applications. Looking at the impact of cell biosensing on the research of other biosensor types, one particular pathway starts to gain attention: cell mimicking. Being at its infancy, the construction of artificial cells is heavily based on knowledge produced throughout the cell-based biosensing era. Linking the artificial bilayer concept with cellular functionality (communication, reproduction, lysis, fusion, etc.), an entirely new domain is expected to occur (see e.g., Xu et al. 2016). The most obvious merits of such an integration refer to the ability to study evolution by recreating step-by-step

720

C. G. Siontorou

cellular complexity. Future applications may actually include the entire range of biosensor research scope: site-specific and intended-use optimized detectors for infield and online simultaneous monitoring of a large number of target analytes.

References Aisyah WN, Jusoh W, Wong LS (2014) Exploring the potential of whole cell biosensor: a review in environmental applications. Int J Chem Environ Biol Sci 2:52–56 Andersen PD, Jørgensen BH, Lading L, Rasmussen B (2004) Sensor foresight – technology and market. Technovation 24:311–320 Bahadir EB, Sezgintürk MK (2015) Applications of commercial biosensors in clinical, food, environmental, and biothreat/biowarfare analyses. Anal Biochem 478:107–120 Batzias FA, Siontorou CG (2012) Creating a specific domain ontology for supporting R&D in the science-based sector – the case of biosensors. Expert Syst Appl 39:9994–10015 Ben-Yoav H, Biran A, Sternheim M, Belkin S, Freeman A, Shacham-Diaman Y (2013) Functional modeling of electrochemical whole-cell biosensors. Sensors Actuators B Chem 181:479–485 Chang Y-C, Yang PY, Chen M-H (2009) The determinants of academic research commercial performance: towards an organisational ambidexterity perspective. Res Policy 38:936–946 Chang H-J, Voyvodic PL, Zuniga A, Bonnet J (2017) Microbially derived biosensors for diagnosis, monitoring and epidemiology. Microb Biotechnol 10:1031–1035 Cockburn I, Long G (2015) The importance of patents to innovation: updated cross-industry comparisons with biopharmaceuticals. Expert Opin Ther Pat 25:739–742 D’Souza SF (2001) Microbial biosensors. Biosens Bioelectron 16:337–353 Debackere K, Veugelers R (2005) The role of academic technology transfer organizations in improving industry science links. Res Policy 34:321–342 Du ZW, Li HR, Gu TY (2007) A state of the art review on microbial fuel cells: a promising technology for wastewater treatment and bioenergy. Biotechnol Adv 25:464–482 Duan X, Lieber CM (2015) Nanoscience and the nano-bioelectronics frontier. Nano Res 8:1–22 Fagerberg J, Verspagen B (2009) Innovation studies – the emerging structure of a new scientific field. Res Policy 38:218–233 Goldberg AP, Szigeti B, Chew YH, Sekar JAP, Roth YD, Karr JR (2018) Emerging whole-cell modeling principles and methods. Curr Opin Biotechnol 51:97–102 González-Cabaleiro R, Mitchell AM, Smith W, Wipat A, Ofiţeru ID (2017) Heterogeneity in pure microbial systems: experimental measurements and modeling. Front Microbiol 8:1813 Guerzoni M, Aldridge TT, Audretsch DB, Desai S (2014) A new industry creation and originality: insight from the funding sources of university patents. Res Policy 43:1697–1706 Hanberger A, Schild I (2004) Strategies to evaluate a university–industry knowledge-exchange programme. Evaluation 10:475–492 Jarque S, Bittner M, Blaha L, Hilscherova K (2016) Yeast biosensors for detection of environmental pollutants: current state and limitations. Trends Biotechnol 34:408–419 Kou S, Cheng D, Sun F, Hsing I-M (2016) Microfluidics and microbial engineering. Lab Chip 16:432–446 Lee PC, Su HN (2011) Quantitative mapping of scientific research-the case of electrical conducting polymer nanocomposite. Technol Forecast Soc Change 78:132–151 Lei Y, Chen W, Mulchandani A (2006) Microbial biosensors. Anal Chim Acta 568:200–210 Lim JW, Ha D, Lee J, Lee SK, Kim T (2015) Review of micro/nanotechnologies for microbial biosensors. Front Bioeng Biotechnol 3:61 Lowe J (1993) Commercialization of university research: a policy perspective. Tech Anal Strat Manag 5:27–37

29

University-Industry Relationships for the Development and. . .

721

Melamed S, Elad T, Belkin S (2012) Microbial sensor cell arrays. Curr Opin Biotechnol 23:2–8 Mogoutov A, Kahane B (2007) Data search strategy for science and technology emergence: a scalable and evolutionary query for nanotechnology tracking. Res Policy 36:893–903 Mongra AC, Kaur A (2012) Overview of biosensors development around the world. Int J Biomed Adv Res 3:519–530 Mowery DC, Nelson RR, Sampat B, Ziedonis AA (1999) The effects of the Bayh-Dole Act on US university research and technology transfer: an analysis of data from Columbia University, the University of California, and Stanford University. Res Policy 29:729–740 Newman MEJ (2004) Co-authorship networks and patterns of scientific collaboration. Proc Natl Acad Sci USA 101:5200–5205 Oliveira MG, Rozenfeld H (2010) Integrating technology roadmapping and portfolio management at the front-end of new product development. Technol Forecast Soc Change 77:1339–1354 Pandza K, Holt R (2007) Absorptive and transformative capacities in nanotechnology innovation systems. J Eng Technol Manag 24:347–365 Park M, Tsai S-L, Chen W (2013) Microbial biosensors: engineered microorganisms as the sensing machinery. Sensors 13:5777–5795 Patel PD (2005) (Bio)sensors for measurement of analytes implicated in food safety: a review. TRAC Trend Anal Chem 21:96–115 Perkmann M, Walsh K (2007) University–industry relationships and open innovation: towards a research agenda. Int J Manag Rev 9:259–280 Santoro C, Arbizzani C, Erable B, Ieropoulos I (2017) Microbial fuel cells: from fundamentals to applications. A review. J Power Sources 356:225–244 Siontorou CG, Batzias FA (2010) Innovation in biotechnology: moving from academic research to product development – the case of biosensors. Crit Rev Biotechnol 30:79–98 Siontorou CG, Batzias FA (2013) A methodological combined framework for roadmapping biosensor research: a fault tree analysis approach within a strategic technology evaluation frame. Crit Rev Biotechnol 34:31–55 Siontorou CG, Batzias FA, Tsakiri V (2010) A knowledge-based approach to online fault diagnosis of FET biosensors. IEEE Trans Instrum Meas 59:2345–2364 Siontorou CG, Georgopoulos KN, Nalantzi M-M (2017a) Designing biosensor networks for environmental risk assessment of aquatic systems. Crit Rev Environ Sci Technol 47:40–63 Siontorou CG, Keramidas VT, Nikoleli G-P, Nikolelis DP, Karapetis S, Bratakou S, Tzamtzis N (2017b) Nano-enabled medical devices based on biosensing principles: technology basis and new concepts. AIMS Mater Sci 4:250–266 Stenger DA, Gross GW, Keefer EW, Shaffer KM, Andreadis JD, Ma W, Pancrazio JJ (2001) Detection of physiologically active compounds using cell-based biosensors. Trends Biotechnol 19:304–308 Struss A, Pasini P, Ensor CM, Raut N, Daunert S (2010) Paper strip whole cell biosensors: a portable test for the semi quantitative detection of bacterial quorum signaling molecules. Anal Chem 82:4457–4463 Su L, Jia W, Hou C, Lei Y (2011) Microbial biosensors: a review. Biosens Bioelectron 26:1788–1799 Swain PS, Elowitz MB, Siggia ED (2002) Intrinsic and extrinsic contributions to stochasticity in gene expression. Proc Natl Acad Sci USA 99:12795–12800 Tomita M (2001) Whole-cell simulation: a grand challenge of the 21st century. Trends Biotechnol 19:205–210 Tzoris A, Hall EAH (2006) Rapid detection of toxicity in wastewater: recent developments with manometric respirometry. Anal Chim Acta 573–574:147–157 Vaiopoulou E, Melidis P, Kampragou E, Aivasidis A (2005) On-line load monitoring of wastewaters with a respirographic microbial sensor. Biosens Bioelectron 21:365–371 Wang BL, Ghaderi A, Zhou H, Agresti J, Weitz DA, Fink GR, Stephanopoulos G (2014) Microfluidic high-throughput culturing of single cells for selection based on extracellular metabolite production or consumption. Nat Biotechnol 32:473–478

722

C. G. Siontorou

Xu C, Hu S, Chen X (2016) Artificial cells: from basic science to applications. Mater Today 19:516–532 Xu M, Wang R, Li Y (2017) Electrochemical biosensors for rapid detection of Escherichia coli O157:H7. Talanta 162:511–522 Yoo SM, Lee SY (2016) Optical biosensors for the detection of pathogenic microorganisms. Trends Biotechnol 34:7–25

Part V Applications of Cells Biosensors

International Organization of Standards for Measurement Validation: Food Analysis

30

Michèle Lees

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analytical Method Quality Assurance: Where Does Method Validation Fit in . . . . . . . . . . . . Finding One’s Way Around the Maze of Standards and Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . What Is Involved in Validating a Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Different Approaches to Method Validation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Key Stages of a Criteria-Based Approach to Method Validation . . . . . . . . . . . . . . . . . . . . . . Method Performance Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Validating a Method for Analytes with Regulatory or Action Limits . . . . . . . . . . . . . . . . . . . . . . . . . . Validating an Alternative or Proprietary Method Against a Reference Method . . . . . . . . . . . . . . . Validating a Qualitative or Semi-Quantitative Screening Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Further Requirements for Internal Quality Control and Laboratory Accreditation . . . . . . . . . . . . Internal Quality Control (IQC) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proficiency Testing (PT) Schemes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Laboratory Accreditation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions and Future Directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

726 726 727 729 730 730 731 736 736 737 739 740 740 742 743 743

Abstract

The importance of obtaining accurate and reliable data applies to any system that produces measurement data, whether it is chemical, biological, or physical. There are numerous internationally recognized standards and guidelines that help in ensuring valid data when developing a new measuring system. This chapter takes the reader through some of those standards, taking examples from the area of food analysis. Keywords

Analysis · Validation · Accuracy · Quality control

M. Lees (*) Food Integrity Consultancy, Sucé-sur-Erdre, France © Springer Nature Switzerland AG 2022 G. Thouand (ed.), Handbook of Cell Biosensors, https://doi.org/10.1007/978-3-030-23217-7_84

725

726

M. Lees

Introduction No matter what type of analytical system is used, whether in the field of analytical chemistry, biochemistry, biology, microbiology, clinical, or pharmacological testing, ensuring good quality measurement data is of utmost importance. Producing reliable results on which the clinical diagnosis of a patient might depend is self-evident, but, from a wider perspective, demonstrated method performance is imperative whenever a decision is taken based on analytical data. In today’s increasing globalization of trade, for example, foodstuffs crossing international borders need to fulfill certain import and/or export requirements such as absence of organic contaminants (for example, residues of pesticides or veterinary medicines) or specific nutrient composition (such as vitamin concentration, sugar, and/or fat content). Help in settling disputes between food control authorities of different countries is the role of the Codex Alimentarius Commission or CAC which works on behalf of the Joint FAO/ WHO Food Standards Programme, charged with the remit of protecting consumer health and promoting fair practices in food trade. This organization has developed specific guidelines to be used when an assessment on the status of a food consignment that is based on test results differs between the importing and the exporting countries involved in the dispute (Codex Alimentarius 2009a). The key points that are investigated in these guidelines include primarily the use of a “validated” method of analysis, followed by proof of internal quality assurance by the laboratory, its performance in proficiency testing schemes, and its official accreditation status. Thus, analytical method validation is now internationally acknowledged as a means of ensuring that analytical results are both reliable and comparable, regardless of where they have been produced. This chapter is intended to take the reader through the different aspects associated with the quality assurance of analytical measurements against a background of international standards. Although the text takes its examples from the field of food testing, the overall concept of method validation can be applied to other sectors and should serve as a general guide for researchers and students involved in the development of measurement methodology.

Analytical Method Quality Assurance: Where Does Method Validation Fit in ISO/IEC 17025 (International Organization for Standardization 2017a) defines method validation as “the confirmation by examination and the provision of objective evidence that the particular requirements of a specific intended use are fulfilled.” The “objective evidence” in this definition relates to method performance criteria and the “specific intended use” concerns the scope of the analytical method and the purpose for which it is to be used. This is generally summarized as demonstrating that an analytical method is “fit-for-purpose,” that is the “validation applies to a defined protocol, for the determination of a specified analyte and range of concentrations in a particular type of test material, used for a specified purpose” (Rambla-Alegre et al. 2012). Validation is required for any new method or “analytical system” under development, or when an existing method is used for a new application. The validation

30

International Organization of Standards for Measurement Validation:. . .

727

Fig. 1 Different levels of Quality Assurance measurements. (Adapted from TrAC Trends in Analytical Chemistry, Vol. 23, No. 8, Isabel Taverniers, Marc De Loose, Erik Van Bockstaele, Trends in quality in the analytical laboratory. II. Analytical method validation and quality assurance, 535–552, (2004) with permission from Elsevier)

procedure, as described below, must include the scope of the method (purpose and type of method, type and concentration range of the analyte(s) being measured) and the matrices to which the method is to be applied. However, it is important to point out that method validation is only one of the main requirements that a laboratory must establish in order to provide confidence in the quality of the analytical data being produced. Indeed, the measuring process is one part of the laboratory’s overall Quality Assurance (QA) system. This, according to ISO 9000 (International Organization for Standardization 2015) should take into account the other activities that the laboratory puts in place to ensure that not only the method but also the measuring equipment and its operator(s) produce consistently reliable results on a regular basis over a period of time. The scheme shown in Fig. 1 shows how method validation fits into a comprehensive QA system (Taverniers et al. 2004). Note that Quality Control (QC) differs from Quality Assurance (QA), as it refers to individual measures taken to comply with quality requirements in the overall QA scheme (International Organization for Standardization 2015). Further details of the concepts shown in Fig. 1 are provided in this chapter.

Finding One’s Way Around the Maze of Standards and Guidelines Any scientists setting out to develop a new analytical method can quite easily find themselves lost in the complex maze of standards, guidelines, and regulatory requirements that have been published on quality assurance and method validation.

728

M. Lees

Within the European Union, there is a clear regulatory framework for the use of analytical methods for food and feed testing. European Commission Decision 2002/ 657/EC (European Commission 2002) sets out performance criteria and other requirements for analytical methods. It provides generally accepted definitions for the majority of terms encountered in a validation procedure and includes guidelines for sample handling. This performance criteria approach is also endorsed in the EC Regulation EC/882/2004 (European Commission 2004) on official food and feed controls which stipulates that only officially accepted methods or those that have been validated according to internationally recognized protocols can be used. It also Table 1 A nonexhaustive list of organizations and published guidelines and standards relating to method validation Type Regulatory requirements

Guidelines

Organization or regulatory body European Commission

US FDA (Food and Drug Administration) Codex Alimentarius CCMAS

Document 2002/657/ EC EC/882/ 2004 EC/2073/ 2005 Guidance for industry CAC/GL 64-1995

CAC/GL 72-2009 CAC/GL 54-2004 CAC/GL 70-2009 Eurachem

Guide Guide

IUPAC (International Union of Pure and Applied Chemistry) AOAC International (Association of Official Analytical Chemists)

Article

Appendix F

Appendix D

Description Performance criteria of analytical methods Official controls on food and feed Microbiological criteria of foodstuffs Analytical procedures and method validation for drugs and biologics Protocol for the design, conduct, and interpretation of method performance studies Guidelines on analytical terminology Guidelines on measurement uncertainty Guidelines for settling disputes on analytical (test) results The fitness for purpose of analytical methods Terminology in analytical measurement Harmonized guidelines for single laboratory validation of methods of analysis Standard method performance requirements Collaborative study procedures to validate characteristics of a method of analysis

References

(continued)

30

International Organization of Standards for Measurement Validation:. . .

729

Table 1 (continued) Type Standards

Organization or regulatory body ISO (International Standardization for Organization)

Document ISO IEC 17025 ISO 5275 Parts 1, 2, 3, 4, 5, 6 ISO/TR 13843:2000

CEN (European Committee for Standardization)

CEN Guide 13:2008 Various

Description General requirements for the competence of laboratories Accuracy (trueness and precision) of measurement methods and results Water quality – Guidance on validation of microbiological methods Validation of environmental test methods Standard methods of sampling and analysis for food and feedstuffs

References

further requires laboratories undertaking official control work to have been assessed and accredited according to ISO standards. In the United States, the Food and Drug Administration (Food and Drug Administration 2015) has published non-binding guidelines on method validation for drugs and biologicals. The recommendations specifically apply to drug substances and products covered in new drug and/or biologics license applications. On the international level, the Codex Committee on Methods of Analysis and Sampling (CCMAS), part of the Codex Alimentarius Commission with the United Nations FAO/WHO Food Standards Programme, also takes a criteria-based approach. CCMAS has published a number of documents on how to evaluate the acceptability of an analytical method through a validation process (Codex Alimentarius 2009b). In addition to regulatory requirements, a number of both international and European bodies have published harmonized guidelines on how to validate methods and on the terminology involved. A list of the main documents in this area, primarily targeted at food and feedstuffs, is given in Table 1. The main standardization bodies such as ISO and CEN have also published standard methods that have been through the process of validation including collaborative studies. These “reference” methods may be used to compare with a newly developed method as mentioned later.

What Is Involved in Validating a Method There are a number of published protocols that are internationally accepted for achieving what is known as the “full” validation of an analytical method (Horwitz 1995). The reader is also encouraged to look up some of the very accessible and

730

M. Lees

more detailed scientific publications on the subject (EURACHEM 1998). What follows here is a general review of the main steps involved in method validation, the purpose of which is to guide the non-specialist in charge of developing a new analytical method through the validation process.

Different Approaches to Method Validation There are two main approaches to method validation; the most commonly used being the criteria-based approach which is endorsed by both EU legislation and the Codex Alimentarius, as explained above. The other approach is one in which fitness-forpurpose is assessed using measurement uncertainty (MU) as a key indicator of how the method has performed (Gustavo González and Ángeles Herrador 2007). MU is defined as the “parameter associated with the result of the measurement, that characterizes the dispersion of the values that could reasonable be attributed to the measurand” (EURACHEM 1998). In this approach to validation, the analytical result is expressed as a range of values lying between predefined levels of acceptability. Further details on this approach to validation are available in the scientific literature. The information provided in this chapter will be limited to the criteria-based approach.

The Key Stages of a Criteria-Based Approach to Method Validation The first step in any method validation is to determine what is required of the method, in other words, what is its target performance. This could be according to specific customer requirements, for example, a producer of infant formula will require particularly stringent limits. Or the method’s characteristics may already be laid down in the legislation, as for example, limits of mycotoxins in food and feed samples. A further consideration is what type of answer is required from the method; will it be quantitative or qualitative? Generally speaking a qualitative method is one which, starting from a specified quantity of sample, can identify the presence of an analyte and provides a response in terms of “presence/absence.” A quantitative method, on the other hand, is one that measures the quantity or mass fraction of an analyte so that it can be expressed as a numerical value in the appropriate units. Detailed definitions of these two types of methods are also given in 2002/657/EC (European Commission 2002). Once the purpose of the method has been identified, the specific technical characteristics such as its specificity, sensitivity, etc. can be determined. At this point, it is important not to neglect aspects linked to the implementation of the method, such as its cost, its ease of use, rapidity, and analytical turnaround time (time taken from reception of a sample to reporting the results). This will provide an indication of whether the proposed method can be easily used routinely and/or transferred to other laboratories. The extent of the validation exercise will also depend on the type of method to be validated. If the method has already been fully validated (standard or normalized method) but is being applied for the first time in the laboratory, the analyst will only

30

International Organization of Standards for Measurement Validation:. . .

731

Fig. 2 An overview of the main steps in the method validation process

have to demonstrate that the previously established performance characteristics can be achieved. Similarly, a standard method used for a new type of sample, or under different circumstances, also only requires a limited validation. On the other hand, a new method under development in-house will require a full validation. However, since method development is closely linked to method validation, it is usually possible to evaluate a number of the method performance parameters during this early stage. An overall view of the main steps in the method validation process is shown in Fig. 2. At this stage, most guides recommend following a written validation plan, with details of the method itself (its scope, the type and concentration range of the analyte(s), and the matrices to which the method is being applied) as well as the selection of validation experiments to be carried out, the characteristics of the analytical equipment, any standards, and/or reagents being used. This written document will help later on if further validation experiments are required or when writing up the overall procedure for the validated method. The key step of the validation process is determining the technical characteristics of the method, or method performance criteria. The full list is given below, with definitions and some explanations. However, depending on the method, it may not be necessary to measure all the parameters, some of which may also be estimated as the method is being developed.

Method Performance Criteria Measuring the following parameters will provide the necessary evidence that the method produces results that are fit-for-purpose. These will show that the measuring system has the potential to work in a chosen environment, which might be in the laboratory or in a factory and will provide reliable and consistent analytical results

732

M. Lees

on target analytes, in selected matrices in a specified time. Official guidelines do not generally indicate in which order the validation experiments should be conducted to obtain the required information. This will depend on the method itself.

Accuracy (Trueness and Precision) In order to show that a method is fit for purpose, it is important that the results it produces are as accurate as possible, that is the quantity measured is as close as possible to the true value of that quantity. Various definitions are available for both the general term “accuracy” and its components which are “trueness” and “precision.” In the Eurachem Terminology in Analytical Measurement (Barwick and Prichard 2011), accuracy includes the effects of both trueness and precision. These terms are also defined in the IUPAC Harmonized Guidelines for Single Laboratory Validation of Methods of Analysis (Thompson et al. 2002) as follows: • Trueness is the closeness of agreement between a test result and the accepted reference value of the property being measured. Trueness is stated quantitatively in terms of “bias,” with smaller bias indicating greater trueness. • Precision is the closeness of agreement between independent test results obtained under stipulated conditions. It is usually specified in terms of standard deviation or relative standard deviation. Both terms are illustrated in Fig. 3 which shows individual measurement results falling on or around a target which represents the true or accepted reference value of the measurement. Case (b) is clearly the best situation with all results clustered together (close agreement between individual results) and in the center of the target (close agreement with the reference value). The worst situation is case (c) where the results are scattered and not close to the reference value. Comparing cases (a) and (d)

Fig. 3 The difference between trueness and precision represented by four different cases in which individual results are positioned on a target representing the reference value. (Figure reproduced by permission of Eurachem from V J Barwick and E Prichard (Eds), Eurachem Guide: Terminology in Analytical Measurement – Introduction to VIM 3 (2011). ISBN 978-0-948926-29-7. Available from www.eurachem.org)

30

International Organization of Standards for Measurement Validation:. . .

733

also highlights the difference between trueness and precision. In case (a), the results are close to the target value (high trueness) but fairly widely scattered (low precision). In case (d) the results are clustered together (high precision) but far from the target value (low trueness). In this case, the difference between the group of results and the reference value is the measurement bias which provides a means of expressing trueness.

Measurement Bias, Recovery Measurement bias is the difference between the test results and an accepted reference value that arises due to a combination of one or more systematic errors. It can be estimated by comparing the mean of several measurements to a reference value, using for example a Certified Reference Material (CRM) which has been assigned a known value as shown in Fig. 4. If a suitable reference material is not available, measurement bias can be estimated using a spiking and recovery experiment. This involves analyzing a test material both on its own and after adding (spiking) a known amount of the analyte being measured. The difference between the two results is known as the recovery. If the measurement bias can be estimated, it can subsequently be used as a correction factor in future measurements. It should be noted that any measurement uncertainty due to the estimation and correction should be taken into account. Repeatability, Intermediate Precision, Reproducibility Precision is usually expressed as the standard deviation, s (or SD) or relative standard deviation, RSD. Precision measures can be considered at three different levels: • Repeatability. This is the measurement of precision under repeatability conditions, i.e. conditions where independent test results are obtained with the same method on identical test items in the same laboratory by the same operator using the same equipment within a short interval of time. According to the FDA’s Guidance for Industry for the Validation of Analytical Procedures which

Fig. 4 Estimation of measurement bias. (Figure reproduced by permission of Eurachem from V J Barwick and E Prichard (Eds), Eurachem Guide: Terminology in Analytical Measurement – Introduction to VIM 3 (2011). ISBN 978-0-948926-29-7. Available from www.eurachem.org)

734

M. Lees

incorporates ICH methodology, repeatability should be assessed using a minimum of 9 determinations covering the specified range for the procedure (e.g., 3 concentrations/3 replicates each); or a minimum of 6 determinations at 100% of the test concentration (ICH International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use 1996). • Intermediate measurement precision. This is the measurement of precision in which measurements are made on portions of the same test material using the same method but over a longer period of time than the measurements carried out under repeatability conditions. In this case, the operator may be different as may certain pieces of equipment. This measurement is also known as within-laboratory reproducibility. • Measurement reproducibility. This is the measurement of precision under reproducibility conditions, i.e. conditions where test results are obtained with the same method on identical test items in different laboratories with different operators using different equipment. Reproducibility is usually assessed as part of an interlaboratory trial and should also be measured if the analytical procedure is to become a standard. Figure 5 shows the relationship between the repeatability, intermediate measurement precision, and reproducibility in terms of the estimated standard deviation.

Working Range (Limits of Detection, Limits of Quantification, Linearity) During the method validation process, it is important to assess at which concentrations of analyte the method operates correctly and with an acceptable precision. This

Fig. 5 Schematic illustrating the relationship between repeatability, intermediate precision and reproducibility in terms of standard deviation. (Figure reproduced by permission of Eurachem from V J Barwick and E Prichard (Eds), Eurachem Guide: Terminology in Analytical Measurement – Introduction to VIM 3 (2011). ISBN 978-0-948926-29-7. Available from www.eurachem.org)

30

International Organization of Standards for Measurement Validation:. . .

735

is known as the measuring interval or “working range,” defined by the Eurachem Guide to Method Validation (EURACHEM 1998) as “the interval over which the method provides results with an acceptable uncertainty.” The lower end of the working range is bounded by the limit of quantitation, LOQ. The upper end of the working range is defined by concentrations at which significant anomalies in analytical sensitivity are observed (Fig. 6). This method working range can be evaluated by using samples of known concentrations which cover the whole range of interest. In general terms, the limit of quantitation, LOQ, is the minimum signal (concentration or amount) that can be quantified. In contrast, the limit of detection, LOD, is the lowest content or concentration that can be detected but not necessarily quantified in the given conditions of the test. A practical LOD is obtained by analyzing samples with decreasing concentration. The minimum concentration which fulfills the identification criteria is the practical LOD.

Ruggedness/Robustness The method validation process must also demonstrate that the results produced by the analytical method will not change significantly if the experimental conditions undergo minor modifications. This is known as the “ruggedness” (sometimes called “robustness”) of the analytical method and it can be tested by deliberately varying external factors such as the operator, the reagents, and the period when the analysis is carried out. Any subsequent effect on the performance of the method, either on the tests results or on a specific experimental parameter such as a peak in a chromatogram, can then be investigated and further improvements made.

Fig. 6 Diagram illustrating the measuring interval and limits of detection and quantitation. (Figure reproduced by permission of Eurachem from V J Barwick and E Prichard (Eds), Eurachem Guide: Terminology in Analytical Measurement – Introduction to VIM 3 (2011). ISBN 978-0-948926-297. Available from www.eurachem.org)

736

M. Lees

Selectivity Finally, the validated method must demonstrate that it can provide accurate results in the presence of interferences. IUPAC defines “selectivity” as “the extent to which the method can be used to determine particular analytes in mixtures or matrices without interferences from other components of similar behaviour” (EURACHEM 1998). The literature provides several guidelines for assessing method selectivity, for example, by analyzing test samples that contain potential interferences in the presence of the analytes of interest and examining whether their presence has an effect on the detection or quantitation of the analytes (EURACHEM 1998).

Validating a Method for Analytes with Regulatory or Action Limits Increasingly in the field of analytical measurements, and particularly in the case of trace contaminants, measurements are made at extremely low concentrations. In such cases, it may be necessary to establish a value of the result which is considered to indicate an analyte level that is significantly different from zero and that some action is required. This level is known as the “critical value” or “decision limit.” In the EU Decision 2002/657/EC, this decision limit, CCα, is defined as “the limit at and above which it can be concluded with an error probability of α that a sample is non-compliant.” Similarly it is important to know the lowest concentration of the analyte that can be detected by the method at a specified level of confidence. This is the “detection capability” (CCβ) of the method and defined in EU Decision 2002/657/EC as “the smallest content of the substance that may be detected, identified, and/or quantified in a sample with an error probability of β.” The European Commission has also published practical guidelines for implementing 2002/657/EC and for validating a method for substances for which an MRL (Maximum Residue Limit) has been established (European Commission DG SANCO 2004).

Validating an Alternative or Proprietary Method Against a Reference Method With the great strides being made in analytical instrumentation, the field of analytical testing in the food sector as well as other fields is rapidly evolving. New methods are constantly being developed with the aim of improving accuracy and shortening analytical turnaround time. Several documents are available in order to guide the laboratory through the different steps needed to validate their newly developed method. A good example is the “Practical guide for the validation, quality control, and uncertainty assessment of an alternative oenological analysis method” published by the OIV (International Organisation of Wine and Vine). This document sets out

30

International Organization of Standards for Measurement Validation:. . .

737

clearly the different steps an oenological laboratory must follow to compare an alternative in-house method with an official OIV method. The area of food microbiology is a particularly good example where there is an increasing move away from traditional microscopy, culture, and biochemical testing to the use of novel diagnostic techniques based on nucleic acid-based assays and mass spectrometry. While the traditional techniques remain the gold standard for the official detection of microorganisms, most routine laboratories have a preference for a more rapid enumeration and characterization of microorganisms. Standard ISO 16140 (International Organization for Standardization 2016) provides the requirements for validating a newly developed microbiological method against a standardized method. This includes comparing the performances of the alternative method against those of the reference method by testing the following parameters: – Relative accuracy, specificity, and sensitivity – Relative detection level And to determine the following parameters for the alternative method: – Inclusivity: This refers to ‘the strains or isolates or variants of the target agent(s) that the method can detect’ (AOAC 2012) similar to “selectivity”’ defined above. – Exclusivity: This refers to ‘the strains or isolates or variants of the target agent(s) that the method must not detect’ (AOAC 2012). – Practicability Even after the method has been validated, in the case of pathogenic microorganisms such as Salmonella or Listeria, all samples identified as positive by the alternative method must still be systematically confirmed by an established reference method.

Validating a Qualitative or Semi-Quantitative Screening Method As mentioned above, when developing a new method, it is important to ascertain in advance both the purpose of the method and how it will be eventually used. Reviewing these “end-user” needs at an early stage may indicate that a qualitative or semi-quantitative method providing a yes/no or presence/absence answer would be suitable. For example, the newly developed method could be used as a rapid screening technique, followed up if necessary, by confirmation using an official reference method. Screening methods are defined in Commission Decision 2002/657 (European Commission 2002) as “methods used to detect the presence of an analyte or class of analytes at the level of interest. These methods have the capability for a high sample throughput and are used to sift large numbers of samples for potential non-compliant results. They are specially designed to avoid false compliant results.” Table 2 gives the requirements for validating screening methods whether they are to be used for

738

M. Lees

Table 2 Requirements for the validation of qualitative and quantitative methods according to 2002/657/EC Performance criteria CCß (detection capability) CCα (decision limit) Trueness/recovery Precision Selectivity Ruggedness

Qualitative methods S C + + − + − − − − + + + +

Quantitative methods S C + + − + − + + + + + + +

S = screening, C = confirmation, + = mandatory determination

qualitative or quantitative measurements. Thus if a newly developed measuring system is to be used as a qualitative (or semi-quantitative) method then the performance criteria required to demonstrate fitness for purpose can be limited to the analytical system’s detection capability (CCβ), its selectivity, and its ruggedness. In the case of a screening method the error β, or false compliant rate, should be Co3O3 > NiO
TiO2, Fe2O3 while in the plant seeds was CuO > ZnO > NiO > Co3O3, TiO2, Fe2O3. These differences indicate that a battery of bioassays could constitute a better tool for assessing environmental pollutants than any single one. Subsequently, Kong et al. (2018) also used E. coli RB1436, seeds of the plant Lactuca sativa, and also Chlorella vulgaris in order to evaluate the effect of the size of distinct ZnO NMs. The different organisms had different responses where ZnO NMs were more toxic for E. coli RB1436. Interestingly, the larger size of the NMs affected seed germination and algal growth but not the bioluminescence of E. coli RB1436, supporting the need of a set of bioassays for an accurate valuation of environmental toxicity. P. putida BS566::luxCDABE (Winson et al. 1998) was used to test the toxic effects of Ag NMs on the activated sludge process (Dams et al. 2011) where Ag+ and micro Ag were included for comparison purposes. The most toxic sample was Ag+ followed by Ag NMs and then Ag microparticles. Mallevre et al. (2014) brought to light the relevance of NMs ecotoxicological studies in real matrices; Ag, ZnO, and TiO2 NMs were tested with P. putida BS566::luxCDABE in culture media and in artificial wastewater, and the Ec50 was different in each case, being in both cases Ag NMs the most toxic in the artificial wastewater. The same group evaluated the ecotoxicity of pristine and aged silver nanoparticles in real and artificial wastewater samples (Mallevre et al. 2016). Aged Ag NMs were less toxic in artificial wastewater, but not in real wastewaters where the toxicity remained despite aging. Deryabin et al. (2016) compared the sensitivity of four different natural and recombinant bioreporters after the exposure to carbon nanotubes (CBNs) and

32

Luminescent Microbial Bioassays and Microalgal Biosensors as Tools. . .

793

metallic NMs. The bioreporters were the natural and commercial Microbiosensor B17-677F (previously described); the commercial and recombinant Ecolum (Deryabin et al. 2012); the recombinant bioreporter E. coli K12 MG1655 (Zarubina et al. 2009) for testing the toxicity of CBNs; and B. subtilis EG168-1 harboring the luxCDABE genes (Deryabin et al. 2016). All these bioreporters were exposed to CBNs and metallic NMs, and B. subtilis EG168-1 was the most sensitive for both kinds of NMs. Another commercial recombinant bioreporter is Salmonella typhimurium TApr1 which harbors the lux genes under a constitutive promoter. This strain has been included in the VITOTOX ® test to normalize the response with respect to possible cytotoxic effects, improving the reproducibility of assays in different laboratories (Fernández-Piñas et al. 2014; Verschaeve et al. 1999). Freshwater recombinant bioreporters such us Janthinobacterium lividum YH9RC are ecologically relevant microorganism from inland water. This strain harbors the luxAB genes integrated in the genome and was patented under the commercial name of BactoTox®. Basically, it consists of freeze-dried cells in 348-multiwell plate for continuous monitoring using a special software. This method was 7.8–8.6 times more sensitive to heavy metals and organic pollutants than Microtox ® (Cho et al. 2004). Another group of freshwater recombinant bioreporters are based on cyanobacteria. Cyanobacteria are ubiquitous in freshwater environments, and as primary producers, they are at the base of food chain being particularly relevant because they are representative of the health of the environment they live in. Two cyanobacterial transgenic bioreporters have been constructed until date: Synechocystis sp. strain PCC6803 and Anabaena CPB4337 (A. CPB4337) (Fernández-Piñas et al. 2014). Synechocystis sp. strain PCC6803 marked with the luciferase gene luc from P. pyralis. This bioreporter was sensitive to herbicides as well as Cu, Zn, and DCP (Shao et al. 2002). The cyanobacterium Anabaena sp. PCC7120 (A. sp. PCC 7120) has been transformed with luxCDABE genes from P. luminescens and was named A. CPB4337 (Fernández-Pinas and Wolk 1994). This bioreporter has a high and stable luminescent signal (Fig. 3); thus, it has been used in different environmental matrices and in combination with several organisms such as A. fischeri, D. magna, or Pseudokirchneriella subcapitata in toxicity studies of priority and emerging pollutants (Fernández-Piñas et al. 2014). A. CPB4337 has shown high sensitivity to emerging pollutants such as fibrates (Rosal et al. 2010a), perfluorinated surfactants and chlorinated by-products (Rosal et al. 2010b), antibiotics (González-Pleiter et al. 2013), NMs (Pulido-Reyes et al. 2017; RodeaPalomares et al. 2010, 2012), and NMs in wastewater (Martín-de-Lucía et al. 2017). A. CPB4337 has been also used to evaluate mixture toxicity by RodeaPalomares and co-workers, where combination-index-isobologram equation (Chou 2006; Chou and Talalay 1984) was applied in order to study the toxicological interactions. This research group also tested heavy metals as well as emerging pollutants (González-Pleiter et al. 2013; Rodea-Palomares et al. 2009a, b, 2016; Rosal et al. 2010b). In a more recent study, Rodea-Palomares et al. (2016) used A. CPB4337 to allow the identification of hidden drivers of toxicity in environmental mixtures in a high-throughput study.

794

J. Hurtado-Gallego et al.

Fig. 3 Recombinant cyanobacterial toxicity bioreporter Anabaena CPB4337 (panel a is a brightfield image of a filament; panel b is an image of constitutive bioluminescence from the same filament with a Hamamatsu Photonics Systems, model C1966-20, mounted on a Zeiss Universal Research Microscope). Panel c shows a multiwell plate where the bioreporter is deployed to be used in a medium-to-high-throughput configuration (Mixes 1–3 refer to different mixtures of pollutants) (Fernandez-Piñas F., 2019)

Nitrosomonas europaea ATCC19781 is an autotrophic ammonia-oxidizing bacterium isolated from soil and transformed with luxAB genes directed by hydroxylamine oxidoreductase promoter, giving constitutive light emission which is correlated with respiration (Iizumi et al. 1998). The bioreporter has been used in linear alkylbenzene sulfonate surfactant (LAS) toxicity in soil solution and in a soilphase contact, obtaining in both cases similar toxicity (Brandt et al. 2002). Later, Ore et al. (2010) used the same strain and tested the Cu toxicity in soils concluding that toxicity depends on the metal free ion activity in solution, the ions competing for metal sorption, and the biotic ligand. P. fluorescens DF57-40E7 has been used as a global toxicity bioreporter in conjunction with P. fluorescens-based lights-on bioreporters in the presence of Cu in soils and in solid-phase contact (Brandt et al. 2006, 2008; Tom-Petersen et al. 2004). Burkholderia sp. RASC c2 is a 2,4-DCP-mineralizating bacterium that harbors the luxCDABE genes fused to tet promoter obtaining constitutive light emission, so this strain can be used at the same time as bioremediation tool and to monitor the bioremediation process in DCP-contaminated soils (Shaw et al. 1999). Furthermore, it has been used to detect toxicity of mono-, di-, and trichlorophenols (Boyd et al. 2001) and of heavy metals (Chinalia et al. 2008) in soils. P. fluorescens strain, isolated from activated sludge, was transformed by Kelly et al. (1999) carrying the luxCDABE genes and named P. fluorescens Shk1. This bioreporter was sensitive to DCP, Cd, and hydroquinone in wastewater (Kelly et al. 1999), narcotic chemicals (Ren et al. 2004), and metal mixtures (Ren and Frymier 2005). Lajoie et al. (2002) used P. fluorescens Shk1 to evaluate the influent and effluent of wastewater treatment plants. Furthermore, it has been used in conjunction

32

Luminescent Microbial Bioassays and Microalgal Biosensors as Tools. . .

795

with another strain: P. fluorescens PM6 expressing the lux genes (Ren and Frymier 2003). Toxicity of 7 metals and 25 organic compounds was found for both bioreporters, and no general pattern for the sensitivity between both strains was observed. The authors concluded that the strains were appropriate for evaluating wastewater toxicity (Ren and Frymier 2005). Another ecologically relevant microorganism is Acinetobacter sp. DF4, isolated from industrial wastewater and transformed with luxCDABE genes (in pUTK2 plasmid) fused to a constitutive promoter by Abd-El-Haleem et al. (2006) and named Acinetobacter sp. DF4/pUTK2. Toxicity of heavy metals has been tested in water and wastewater samples, and the sensitivity order was Zn, Cd, Fe, Co, Cr, and Cu (Abd-El-Haleem et al. 2006). Later, the strain was immobilized in a matrix of Caalginate, and the toxicity of phenolic compounds was tested (Zaki et al. 2008). Other recombinant bioreporters with the pUTK2 (harboring the luxCDABE genes behind a constitutively expressed promoter) plasmid are Stenotrophomonas 664 (pUTK2) and Alcaligenes eutrophus BR6020 (pUTK2). Both strains along with A. fischeri were tested with the nonionic surfactant polyoxyethylene 10 lauryl ether. Both recombinant strains were 400 times more resistant than A. fischeri, highlighting their usefulness to evaluate toxicity of remediation processes (Layton et al. 1999). Recently, Cui et al. (2018) have constructed the bioreporter Acinetobacter baylyi Tox2 from the soil bacterium Acinetobacter baylyi ADP1 harboring the luxCDABE genes controlled by the constitutively expressed tet promoter. In this study, heavy metal toxicity was tested with the bioreporter and with the marine fish Mugilogobius chulae showing correlation between these two toxicity test methods. The suitability of Acinetobacter baylyi Tox2 was proved as this strain was able to evaluate toxicity of real contaminated seawater. Although the bioreporters described so far are prokaryotic bioreporters, eukaryotic bioreporters have also been used in environmental applications. The main eukaryotic organism used to construct recombinant bioreporters is the yeast Saccharomyces cerevisiae (S. cerevisiae) whose cell wall protects the organism from extreme pH, solvent exposure, or osmotic changes (Fernández-Piñas et al. 2014). S. cerevisiae W303-1B lucΔ (without the peroxisome target sequence) was constructed by Hollis et al. (2000) and harbors the lucΔ gene from P. pyralis integrated into the genome, but the substrate D-luciferin had to be added for bioluminescence production. The strain responded to diuron and mecoprop herbicides in a pH range of 3 to 10; however, E. coli HB10 (pUCD697), as a model recombinant bioreporter, did not respond to these analytes. Moreover, S. cerevisiae W303-1B lucΔ detected Cu in different solvents, but not in extremes pH due to changes in Cu speciation (Hollis et al. 2000). The marine green alga Ostreococcus tauri isolated from a Mediterranean coastal lagoon was modified by Sanchez-Ferandin (2015). Four strains were constructed: the translational bioreporter TOC1-Luc (chlorophyll a binding protein expression (CAB)), the transcriptional bioreporter pCAB::luc (promoter of chlorophyll a binding protein expression), and the translational bioreporters cyclin A-Luc and CDKALuc (cyclin A and cyclin dependent kinase, respectively). The antifouling biocides

796

J. Hurtado-Gallego et al.

diuron and Irgarol 1015 were tested as compound toxic models. CDKA-Luc translational bioreporter was the most effective O. tauri bioreporter, demonstrating that it eukaryotic recombinant microorganism could be used as bioreporter of contaminated seawaters.

Oxidative Stress Bioreporters Reactive oxygen species (ROS) are formed by aerobic organisms as a by-product of respiration. They have developed defenses against these toxic species maintaining a balance between them (Latifi et al. 2009). In the presence of a toxic compound, ROS level increases producing an unbalance called oxidative stress. Several evidences on mechanism studies denote that oxidative stress is involved in most environmental toxicity processes (Livingstone 2001; Lushchak 2011; Stone and Donaldson 2006; Valavanidis et al. 2006). At present, it is complicated to measure ROS directly, so attempts at the evaluation of free radical production and the pro-oxidant/antioxidant balance are based on markers of oxidative reactions often involving in vitro experiments (Betteridge 2000). As described by Monaghan et al. (2009), there are a lot of methods for oxidative stress measurement estimating the ROS production, antioxidant defenses, oxidative damage, and repair mechanisms. The enzymatic antioxidant defenses are superoxide dismutase, peroxidase, catalase, and glutathione reductase (Gagne 2014). However, measuring oxidative stress is complex, and a combination of different assays is required. A few oxidative stress lights-on bioreporters have been constructed to know more about the mechanisms of action of different pollutants in the environment. Among the transgenic bacterial bioreporters for oxidative stress, E. coli still takes a prominent place (Robbens et al. 2010). The most used E. coli bioreporter for oxidative stress in environmental toxicity is the strain DPD2511 constructed by Belkin et al. (1996). This strain harbors the fusion of the promoter of katG that encodes catalase to luxCDABE genes (PkatG::luxCDABE), and the bioluminescence induction was checked with H2O2, ethanol, methyl viologen (MV), or xanthine. In addition, this bioreporter was used with other model environmental pollutants such as chromium, cadmium, or ethidium bromide in order to elucidate the different toxicological mechanisms of these pollutants. Belkin et al. (1997) constructed E. coli DPD2515 bearing PmicF::luxCDABE (micF gene encodes an outer membrane porin regulator) and was checked with the previously commented pollutants and, particularly, detected superoxide anion. With the same purpose, Lee and Gu (2003) constructed another E. coli bioreporter bearing the promoter of sodA that encodes superoxide dismutase fused to luxCDABE (PsodA::luxCDABE) named as EBHJ. Comparing the data of EBHJ with those of DPD2511 strain after the exposure with the pollutants (MV, potassium dichromate, cadmium chloride, bisphenol A, ethidium bromide, and H2O2), different responses were found indicating that the mechanisms of toxicity leading to oxidative stress for different pollutants can be classified as a PsodAdependent response or a PkatG-dependent response in E. coli. Although no real matrices were used, the authors concluded that they may be potentially used as tools

32

Luminescent Microbial Bioassays and Microalgal Biosensors as Tools. . .

797

for monitoring and assessing the hazards in environmental samples in different media such as drinking water, wastewater, soil, and air. Both strains EHBJ and DPD2515 were used with other strains in a panel of bioreporters to test potential environmental toxicity (Belkin et al. 1997). Furthermore, this DPD2515 strain was also used in a different panel for identification and quantification of different chemicals (Ben-Israel et al. 1998). EHBJ strain has been used in wastewater samples (Bazin et al. 2017). E. coli EBHJ has been used in an integrated mini biosensor system for continuous water toxicity monitoring (Lee and Gu 2005). This strain with other three E. coli bioreporters was used in a biosensor configuration, generating four mini bioreactors (each one with one strain). The experiments were carried out with three chemicals, MV, H2O2, and mitomycin C, in an experiment where the water sample was contaminated at various times during the system operation. Another strain of this mini biosensor was DK1 constructed by Mitchell and Gu (2004) bearing the fusion PkatG::luxCDABE, and the tested chemicals were H2O2, cadmium chloride, and isopropanol, showing induction of bioluminescence for all of them. More E. coli bioreporters to detect oxidative stress have been constructed harboring the fusions of zwf (encodes a flavodoxin and ferredoxin reductase) and fpr (encodes a glucose-6-phosphate dehydrogenase) promoters to luxCDABE, namely, E. coli ZWF RFM443 and E. coli FPR RFM443, respectively (Niazi et al. 2007). Both bioreporters were exposed to MV and its derivatives showing bioluminescence induction, but no response was seen after the exposure to H2O2, denoting their specificity to superoxide anion. E. coli PGRFM was constructed by the same group Niazi et al. (2008) to detect oxidative stress bearing the fusion Ppgi::luxCDABE, where pgi encodes an glucose-6-phosphate isomerase. This strain was checked for MV and analogues, showing to be sensitive to superoxide anion. Furthermore, this bioreporter showed bioluminescence induction in the presence of H2O2 and OH. In this case, it is worth nothing that the authors concluded that growth in minimal medium increased the sensitivity. Different NMs such as fullerenes, CBNs, or metal oxides have been shown to induce oxidative stress. As well as from surface-dependent properties, metals and chemical compounds on the NMs surface accelerate the production of ROS (Manke et al. 2013). Nowadays, a few bioreporters have been used to detect the oxidative stress caused by nanoparticles profiling their ecotoxicological properties. The first one was E. coli K12::soxRSsodAlux that harbors the fusion of sodA promoter to luxCDABE genes. The induction of this strain is specific for superoxide as in the presence of H2O2 no bioluminescence was shown. After the exposure to ZnO, CuO, and Ag NMs, their bulk, and ionics references, bioluminescence induction was observed for all of them except for ZnO compounds (Ivask et al. 2010). E. coli K12::katGlux was constructed to study the oxidative stress induction of CuO NMs (Bondarenko et al. 2012) and also to know the mechanisms of action at sub-toxic concentrations. The authors observed biological adverse effects even at very low concentrations of CuO NMs and concluded that these effects were triggered by the solubilized copper ions.

798

J. Hurtado-Gallego et al.

All these aforementioned E. coli bioreporters harbor the fusion of a specific promoter to luxDCABE genes. However, E. coli ro-GFP2 constitutively expressed redox-sensitive green fluorescent protein roGFP2 (Arias-Barreiro et al. 2010). This strain has been tested with a variety of environmental oxidants such as H2O2, menadione, heavy metals, or naphthalene, denoting a very rapid response and a broad specificity, because it detects both H2O2 and superoxide anion. As previously described, although E. coli is the most used microorganism to construct bioreporters for environmental toxicity, there are more ecologically relevant microorganisms used as bioreporters. In the case of oxidative stress bioreporters, the cyanobacteria Nostoc sp. PCC7120 (also known as Anabaena sp. PCC7120) has been used to construct several strains that specifically detect superoxide anion (Hurtado-Gallego et al. 2018). Both strains harbor the fusion of sodA and sodB promoters to luxCDABE operon, and their names are Nostoc sp. PCC7120 pBG2154 and Nostoc sp. PCC7120 pBG2165, respectively. MV and triclosan were tested with these bioreporters showing bioluminescence induction, but not with H2O2 denoting their specificity. Furthermore, real water matrices were used in MV spiking experiments, showing that they could be used in real water matrices. They were the first bioreporters constructed to date which detect superoxide anion based on cyanobacteria and, also, the most superoxide anion-sensitive bioreporter strains (Hurtado-Gallego et al. 2018). Another ecologically relevant microorganism for soil environments which has been used as oxidative stress bioreporter is Pseudomonas putida mt-2 (Svenningsen et al. 2015). In this work, four bioreporters bearing the katA (encodes catalase), ahpC (encodes alkylhydroperoxide reductase), sodB (encodes superoxide dismutase), and osmC (hydroperoxide resistance gene) promoters fused to monomeric superfolder gfp gene were constructed. These bioreporters were used to detect the oxidative stress caused under desiccation condition in soils. As described along this section, most bioreporters have been constructed using bacteria due to their small size, adaptability, easy driving, and the great information about their genome sequences. However, there are also other bioreporters based on eukaryotic organisms. Both prokaryotic and eukaryotic bioreporters contribute with valuable information regarding the mechanisms of action and environmental toxicity of a great range of organic and inorganic pollutants. Therefore, the use of a battery of bioreporters could be a key strategy to assess the ecotoxicity of different contaminants in a more realistic scenario.

Microalgal-Based Biosensors Microalgae-based biosensors have emerged as a valuable tool to test any pollutant that might target photosynthesis or enzymatic activities such as esterase and alkaline phosphatase activities. Within those pollutants, mostly different classes of herbicides as well as heavy metals have been assessed. Within microalgae, cyanobacteria and green algae have mostly been used; rarely, diatoms such as Phaeodactylum tricornutum have also been used; both groups are main components of

32

Luminescent Microbial Bioassays and Microalgal Biosensors as Tools. . .

799

phytoplankton in aquatic ecosystems, and being phototrophic, they are critical for the health of those ecosystems, and any detrimental effect on them may significantly alter the trophic chain. Green alga Chlorella vulgaris has been usually selected for the construction of the biosensors; its easy cultivation and immobilization in a variety of matrices make it the organism of choice. Table 3 lists the biosensors that will be discussed in this section, although not exhaustive, the table shows relevant microalgal-based biosensors constructed to date in chronological order. We have tried to complement and update the excellent review on this subject by Brayner et al. (2011). Most microalgal biosensors are based on the measurement of the effect of pollutants on photosynthesis either by chlorophyll fluorescence analysis (optical biosensors) or directly by measuring photosynthetic O2 evolution (photoelectrochemical biosensors, mostly photo-amperometric). There are also a few biosensors based on the inhibition of enzymatic activities such as esterase and alkaline phosphatase activities (conductometric biosensors) (Table 3). Most microalgal biosensors reported to date are able to detect global toxicity as any toxicant that may affect algal metabolism may impair photosynthesis, which is the main metabolic function of these organisms. As nonselective biosensors, they might be able to detect the toxicity of a variety of compounds in an environmental sample; they are adequate for testing samples from aquatic environments, soil abstract, and even aerosols (see Table 3). They have been proposed by most authors as early warning systems of water pollution. Some biosensors based on algae are Algae Toximeter II, Fluotox, or Robot Luminotox (please see the review on commercially available bioassays by Kokkali and van Delft (2014)). As stated before, most biosensors have been exposed to herbicides, which are widely used in agriculture for crops, but they may remain in the soil and leach into ground and surface waters. The European Water Act of 1980 document limits the concentration of herbicides in water to less than 0.1 μg/L for individual herbicides and to 0.5 μg/L for a total herbicide class; most algal biosensors are able to detect the toxicity of herbicides in the low μg/L, so that no preconcentration of the sample should be necessary.

Electrochemical Microalgal Biosensors Pollutants may alter different cell metabolic functions resulting in the consumption or generation of electroactive species that may be monitored by different electrochemical methods. Electrochemical biosensors are usually highly sensitive and maybe miniaturized without losing their properties. Depending on the detected species and the transducer type, electrochemical biosensors may be, among others, amperometric, potentiometric, or conductometric. Most microalgal biosensors discussed in this section are either amperometric (usually used to monitor photosynthesis) or conductometric (used to monitor several enzyme activities). In the 1980s last century, Rawson et al. (1987) reported the construction of a mediator-assisted amperometric biosensor based on the cyanobacterium

800

J. Hurtado-Gallego et al.

Table 3 Main features and applications of algal bioassays

Microorganism/type Synechococcus sp. PCC6301/ prokaryotic Chlorella vulgaris, Scenedesmus subspicatus, and Selenastrum capricornutum/ eukaryotic Chlorella vulgaris, Scenedesmus subspicatus, and Selenastrum capricornutum/ eukaryotic Scenedesmus subspicatus/ eukaryotic

Transducer type Amperometric (photosynthetic activity) Amperometric (photosynthetic activity)

Synechococcus sp. PCC7942/ prokaryotic Synechocystis sp. PCC6803/ prokaryotic Chlorella vulgaris/ eukaryotic

Amperometric (photosynthetic activity) Amperometric (photosynthetic activity) Amperometric (photosynthetic O2 evolution) Amperometric (photosynthetic O2 evolution)

Marine Spirulina subsalsa/prokaryotic

Amperometric (photosynthetic O2 evolution)

Optical (chlorophyll fluorescence)

Chlorella vulgaris/ eukaryotic

Optical (chlorophyll fluorescence)

Chlorella vulgaris/ eukaryotic Nostoc commune/ prokaryotic

Optical (fluorescence; photosystem II photochemical efficiency)

Environmental applications/tested pollutants Herbicides (DCMU, chlortoluron, linuron) Herbicides (isoproturon, chlortoluron, atrazine, propanil) Heavy metals (Cu2+, Hg2+) Herbicides (isoproturon, chlortoluron, atrazine, propanil) Heavy metals (Cu2+, Hg2+) Synthetic wastewater spiked with herbicides (atrazine, endocrine) Herbicides (diuron) and mercuric chloride Herbicides (diuron) and mercuric chloride Volatile organic compounds (VOCs; perchloroethylene) Estuarine natural waters spiked with heavy metals (copper and mercury), triazine herbicides (atrazine), and carbamate insecticide (carbaryl) Herbicides (diuron, simazine, atrazine, alachlor, glyphosate) Airborne warfare agents (tabun, sarin, mustard agent, tributylamine, dibutyl sulfide)

Biosensor configuration References Yes (Rawson et al. 1987) Yes

(Pandard et al. 1993)

Yes

Yes

(Frense et al. 1998)

Yes

(Rouillona et al. 1999)

Yes

(Avramescu et al. 1999)

Yes

(Naessens and Tran-Minh 1999) (Campanella et al. 2001)

Yes

Yes

(Naessens et al. 2000)

Yes

(Sanders et al. 2001)

(continued)

32

Luminescent Microbial Bioassays and Microalgal Biosensors as Tools. . .

801

Table 3 (continued)

Microorganism/type Chlorella vulgaris/ eukaryotic

Chlorella vulgaris/ eukaryotic

Chlorella vulgaris/ eukaryotic

Klebsormidium nitens/eukaryotic Klebsormidium sp. strain M1939/ eukaryotic Chlorella vulgaris/ eukaryotic Nine microalgal strains Immobilized on an array biochip/both eukaryotic and prokaryotic Chlamydomonas reinhardtii/ eukaryotic Chlamydomonas reinhardtii/ eukaryotic Chlorella vulgaris/ eukaryotic

Chlorella vulgaris/ eukaryotic Chlorella vulgaris/ eukaryotic

Transducer type Optical (fluorescence; photosystem II photochemical efficiency) Optical (chlorophyll fluorescence)

Environmental applications/tested pollutants Freshwater drinking samples spiked with KCN, DCMU, methyl parathion, and paraquat Herbicides (atrazine, DNOC, simazine, isoproturon, diuron) Cd2+

Biosensor configuration References Yes (Rodriguez Jr et al. 2002)

Yes

(Védrine et al. 2003)

Yes

(Chouteau et al. 2004)

VOCs (formaldehyde, methanol vapors)

Yes

(Podola et al. 2004)

Optical (chlorophyll fluorescence)

Herbicides (atrazine, diuron, isoproturon, paraquat, simazine)

Yes

(Podola and Melkonian 2005)

Amperometric (changes in flagellar movement) Amperometric (changes in gravitaxis) Amperometric (alkaline phosphatase activity) Optical (chlorophyll fluorescence) Amperometric (alkaline phosphatase activity)

Toluene, copper sulfate, and nickel chloride

Yes

(Shitanda et al. 2005)

Toluene, copper sulfate, and nickel chloride No mention

Yes

Yes

(Ionescu et al. 2006)

Herbicides (diuron)

Yes

Zn2+; Cd2+

Yes

(Nguyen-Ngoc and Tran-Minh 2007) (Chong et al. 2008)

Conductometric (alkaline phosphatase activity) Optical (fluorescence; photosystem II photochemical efficiency)

(continued)

802

J. Hurtado-Gallego et al.

Table 3 (continued)

Microorganism/type Chlorella vulgaris/ eukaryotic

Chlamydomonas reinhardtii photosystem II D1 protein mutants/ eukaryotic Chlamydomonas reinhardtii photosystem II D1 protein mutants/ eukaryotic Dictyosphaerium chlorelloides, Scenedesmus intermedius, and Scenedesmus sp./ eukaryotic Marine Dunaliella tertiolecta and Phaeodactylum tricornutum/ eukaryotic Marine Dunaliella tertiolecta and Phaeodactylum tricornutum/ eukaryotic Chlorella vulgaris, Pseudokirchneriella subcapitata, and Chlamydomonas reinhardtii/ eukaryotic Dictyosphaerium chlorelloides/ eukaryotic Chlamydomonas reinhardtii/ eukaryotic

Transducer type Conductometric (alkaline phosphatase activity) Optical (fluorescence; photosystem II photochemical efficiency) Optical (fluorescence; photosystem II photochemical efficiency) Optical (fluorescence; photosystem II photochemical efficiency)

Environmental applications/tested pollutants Cd2+

Biosensor configuration References Yes (Guedri and Durrieu 2008)

Herbicides (atrazine, prometryn, diuron)

Yes

(Scognamiglio et al. 2009)

Triazines and ureabased herbicides

Yes

(Giardi et al. 2009)

Triazines and ureabased herbicides

Yes

(Peña-Vázquez et al. 2009)

Conductometric (membrane sterase activity)

Herbicides (diuron and glyphosate and degradation products)

Yes

(Durrieu et al. 2011)

Optical (chlorophyll fluorescence)

Herbicides (diuron and glyphosate and degradation products)

Yes

Optical (chlorophyll fluorescence)

Herbicides (DCMU and atrazine)

Yes

(Ferro et al. 2012)

Optical (bioluminescencebased O2 transduction) Amperometric (photosynthetic O2 evolution)

Herbicides (simazine)

Yes

(Haigh-Flórez et al. 2014)

Urea-based herbicides (diuron)

No (bioassay with dense algal solutions)

(Tsopela et al. 2014, 2016)

(continued)

32

Luminescent Microbial Bioassays and Microalgal Biosensors as Tools. . .

803

Table 3 (continued)

Microorganism/type Scenedesmus quadricauda, strain GREIFSWALD/15/ eukaryotic

Transducer type Amperometric (photosynthetic O2 evolution)

Chlamydomonas reinhardtii/ eukaryotic

Optical luminescent O2 sensor and chlorophyll fluorescence

Environmental applications/tested pollutants – Aqueous extracts of soils collected from roadsides – Waste sample from old dried-up metal processing industrial tailing pond enriched with insecticide Drevosan Profi 058/ 14/506 Herbicides (simazine, atrazine, diuron)

Biosensor configuration References No (Buckova et al. (bioassay 2017) with algal solutions)

Yes (microfluidic device)

(Tahirbegi et al. 2017)

Synechococcus PCC6301. The authors tested two other cyanobacteria and a green alga, but the Synechococcus strain proved to be the most successful. The biosensor measured the photosynthetic electron transfer, so that the redox mediator is reduced and then reoxidized at the working electrode resulting in a flow of current. Ferricyanide was used as the most appropriate redox mediator as it did not reduce the sensor life, as did p-benzoquinone. The cyanobacterial biosensor responded to herbicides (DCMU, chlortoluron, and linuron) in the μg/L range in less than 10 min. Two amperometric microalgal biosensors were developed by Pandard et al. (1993). The authors immobilized green algae Chlorella vulgaris, Scenedesmus subspicatus, and Selenastrum capricornutum (at present denoted as Raphidocelis subcapitata) onto an alumina filter disc. Both types of sensors measured photosynthetic activity: one used a redox mediator ( p-benzoquinone proved to give the best results) as a terminal electron acceptor of the photosynthetic electron transfer chain which was subsequently reoxidized at the biosensor electrode surface resulting in a flow of measurable current and another used an oxygen electrode to measure photosynthetic oxygen evolution. Biosensors were exposed to herbicides isoproturon, chlortoluron, atrazine, propanil, Cu2+, and Hg2+. In general, the oxygen proved to give higher sensitivity and longer operational life than the redox-mediator system. Rouillona et al. (1999) and Avramescu et al. (1999) constructed two biosensors based on unicellular cyanobacteria: Synechococcus PCC7942 and Synechocystis PCC6803; both strains were immobilized in poly(vinylalcohol) bearing styrylpyridinium. Photosynthetic activity was measured amperometrically by the use of the redox mediator 2,5-dichlorobenzoquinone. Biosensors were exposed to

804

J. Hurtado-Gallego et al.

diuron and HgCl2, and both were more sensitive to diuron than to HgCl2; detection limits were lower for the Synechocystis biosensor. Naessens and Tran-Minh (1999) developed a biosensor based on the green alga Chlorella vulgaris immobilized on a glass microfiber Whatman filter. Immobilized cells were associated with and oxygen Clark electrode and photosynthetic oxygen evolution measured. The biosensor device was designed to use analytes in the form of aerosols. The biosensor was exposed to perchloroethylene sprayed into the chamber, and an increase in oxygen evolution was found; the authors concluded that perchloroethylene was an activator of photosynthesis with limits of detection below limits values of workers exposure, so that authors suggested the biosensor might be used as an early warning device to determine volatile organic compounds in workplace environments. Marine filamentous cyanobacterium Spirulina subsalsa was inserted into the electrode cap of a Clark-type oxygen electrode and used to assess the effects on photosynthesis (oxygen evolution) of standardized estuarine waters spiked with heavy metals (Hg2+ and Cu2+), the herbicide atrazine or the carbamate insecticide carbaryl (Campanella et al. 2001). The biosensor was more sensitive to the herbicide than to the carbamate insecticide or the heavy metals; issues of metal speciation in saline environments might account with the decrease of metal toxicity. Alkaline phosphatase activity (APA) located on the cell wall has been chosen as a parameter suitable to test the toxicity of heavy metals to microalgal-based biosensors. Chlorella vulgaris has been the alga of choice. Chouteau et al. (2004) constructed a conductometric biosensors with the green alga immobilized inside bovine serum albumin membranes cross-linked with glutaraldehyde vapors. The biosensor was tested in the presence of Cd2+, resulting in inhibition of APA activity, with a detection limit of just 1 μg/L. Ionescu et al. (2006) entrapped Chlorella vulgaris cells either in alginate gel or a newly synthesized pyrrole-alginate matrix. APA was measured by adding p-nitrophenol phosphate as the substrate to generate pnitrophenol, which is electroactive and can be measured amperometrically. The authors demonstrated the higher stability of the pyrrole-alginate gel for immobilization, but no test with pollutants was reported. Guedri and Durrieu (2008) constructed a novel conductometric algal biosensor to measure APA activity in the presence of metals, namely, Cd2+. Chlorella vulgaris was immobilized on self-assembled monolayers (SAMs) of alkanethiolate with the advantage that there is no physical barrier between the algae and the constituents of the reaction medium. The detection limits for the metal were in the μg/L range. Chong et al. (2008) entrapped Chlorella vulgaris in a bovine serum albumin membrane and immobilized directly onto the surface of a diamond electrode. APA was measured amperometrically by adding pnitrophenol phosphate as the substrate. The biosensor was exposed to Zn2+ and Cd2+ with detection limits of 0.1 μg/L for both metals. Shitanda et al. (2005) constructed two novel amperometric biosensors based on motility and gravitaxis of the flagellate algae Chlamydomonas reinhardtii. Both biosensors were exposed to toluene, copper (II) sulfate, and nickel (II) chloride. The negative gravitaxis biosensors proved to more sensitive to toluene than the motility-based biosensor; to increase the sensitivity further, a thin-layer gravitaxis

32

Luminescent Microbial Bioassays and Microalgal Biosensors as Tools. . .

805

biosensor was developed. The authors claimed that this biosensor was much more sensitive than conventional algal biosensors, which measure photosynthetic activity. Durrieu et al. (2011) constructed conductometric algal biosensors based on measuring the esterase activity localized on the external membrane of the algal cells. Two unicellular marine algae were used: green alga Dunaliella tertiolecta and diatom Phaeodactylum tricornutum. Algal cells were immobilized using the SAM technique already discussed above. The Dunaliella tertiolecta biosensor was tested with herbicides diuron and glyphosate and several of their degradation products. Results were highly variable, so that authors conclude that it would be necessary to test a larger number of sensors and lower concentrations of the tested compounds. Tsopela et al. (2014) fabricated an amperometric biosensor with three electrodes integrated on a silicon chip for the detection of herbicides in water. The authors used the electrochemical microsensors device in dense suspensions of the green microalga Chlamydomonas reinhardtii. The device could detect O2 (photosynthetic activity of the algae), H2O2, and H3O+/OH ions, ionic species taking part in other metabolic activities of algae (reactive oxygen species, ROS, formation by the action of pollutants). The photosynthetic activity (O2 monitoring) was tested in the presence of the herbicide diuron, reaching a limit of detection of 0.2 μM. In a subsequent article, Tsopela et al. (2016) developed a lab-on-chip integrated device composed of three fluidic chambers integrating electrochemical sensors and other three chambers dedicated to optical-fluorescent base detection. Algal photosynthetic activity was measured. The device was tested also with diuron achieving a limit of detection of 0.1 μM. Recently, Buckova et al. (2017) developed an amperometric device denoted as AlgaTox which measures oxygen evolution during photosynthesis; it has been patented in the Czech Republic (CZ 305687). A suspension of cells of the green algae Scenedesmus quadricauda is placed in the reaction vessel, and the AlgaTox device measures alterations of photosynthetic oxygen production because of environmental pollution. The device has been tested with aqueous extracts of soils collected from roadsides and with waste sample from old dried-up metal processing industrial tailing pond enriched with insecticide Drevosan Profi 058/14/506. The authors claimed that the values of oxygen production alterations recorded by the device were up to six times higher than the corresponding values of alterations in growth rates determined by standard procedures, implying a higher sensitivity.

Optical Microalgal Biosensors In photosynthesis, electromagnetic energy is converted into chemical energy by the absorption of light by the photosynthetic antenna systems, mainly chlorophylls that transfer that excitation energy to the reaction centers of the two photosystems, photosystems I and II (PSI and PSII). There, the energy drives the primary photochemical reactions that initiate the photosynthetic energy conversion. Some of the absorbed energy is dissipated as heat or emitted as fluorescence at a wavelength

806

J. Hurtado-Gallego et al.

longer than the excitation energy. Due to the high yield of the primary photochemical energy conversion (it is calculated that more than 90% of absorbed light quanta are utilized by photosynthesis), only a small portion (around 3% of absorbed light) is released as fluorescence (Krause and Weis 1991). At room temperature, most fluorescence is emitted by chlorophyll a of PS II, which is the main site highly susceptible to many environmental stresses in photosynthetic organisms (Geoffroy et al. 2007; Papageorgiou et al. 2007). Around 30% herbicides inhibit PSII; the target site is usually the QB site of the D1 protein. These herbicides, including triazines, triazinones, phenylureas, biscarbamates, and phenolic herbicides (Giardi and Pace 2005), inhibit photosynthetic electron flow from the primary acceptor QA to the secondary quinone QB and, thus, increase the chlorophyll a fluorescence intensity. Figure 4 shows some examples of optical microalgal biosensors. Frense et al. (1998) reported the construction of an optical biosensor based on the green alga Scenedesmus subspicatus which was immobilized in filter paper and covered with alginate and fixed on the surface of the optical electrode. Synthetic wastewater was spiked with increasing concentrations of the herbicides atrazine and endrin, detection was possibly down to the μg/L, and the response time was short (10 min); the authors claimed that reproducible results could be obtained after regeneration in the nutrient medium for 1 h. Immobilized algae could be stored for about 6 months retaining chlorophyll fluorescence. In 2000, Naessens et al. (2000) constructed a new biosensor based on green alga Chlorella vulgaris. The algae were immobilized on glass microfiber Whatman filters placed in front of an optical fiber bundle inside a homemade microcell. The biosensor was exposed to herbicides, and chlorophyll fluorescence intensity was monitored. An increase of fluorescence was observed after exposure to herbicides that target PSII: diuron, simazine, and atrazine; alachlor, whose site of action in photosynthesis was not so clear, also increased chlorophyll fluorescence, while glyphosate, an inhibitor of an enzyme involved in amino acid synthesis, decreased it. Detection limits differed with the herbicide, in the order of nM for diuron, simazine, and atrazine and μM-mM for alachlor and glyphosate, respectively. The algal biosensor could be reused with atrazine, simazine, and diuron. Sanders et al. (2001) developed two optical biosensors, one based on the cyanobacterium Nostoc commune and another based on the green alga Chlorella vulgaris. They used a pulse-amplitude modulated (PAM) fluorometer to measure maximum PSII photochemical efficiency in dark-adapted cells. This parameter was calculated as Fv/Fm where the variable fluorescence (Fv) was calculated as the difference between the maximum fluorescence (Fm) and the minimum fluorescence (Fo). Fo and Fm are determined by pre- and post-saturating flash fluorescence measurement on the entrapped biosensor samples. Both biosensors were exposed to chemical warfare agents in a gas stream: tabun, sarin, mustard agent, tributylamine (a sarin stabilizer), and the mustard agent analogue dibutyl sulfide. The same group reported in 2002 (Rodriguez Jr et al. 2002) the use of Chlorella vulgaris and naturally occurring populations of algae in river field samples to calculate the maximum PSII photochemical efficiency when exposed to freshwater drinking samples spiked

32

Luminescent Microbial Bioassays and Microalgal Biosensors as Tools. . . cross-section

a

807

front-view

30 mm

80 mm outlet sample solution. suction by pump

bold connection

front lid transparent plastic panel flow cell connector water thermostat immobilised microalgae biochip membrane CCD camera of imaging fluorometer

back panel nose-piece supporting chip membrane separate flow cells

flow direction of sample solution

b Flow inlet (open)

inlet sample solution

c

FLUORESCENCE SENSOR OFF

Flow inlet (closed)

Flow outlet (open)

Herbicide Alga strain

Filter

Emitting Excitation Fluorescence Light LED

LED

Photodiode

Filter

LED

Photodiode

e

f

LED

i

Biomediator immobilized on a silicon septum that fits the cell size

Flow outlet (closed)

Polycarbonate Window

LED

d

FLUORESCENCE SENSOR ON

g

h

b

g

j k l

c

m n

d a

f

o p

OPTOSEN®

q

e

Fig. 4 Examples of optical microalgal-based bioassays (see Table 3). (a) Schematic optical section of a microalgal biosensor for detection of herbicides inhibiting photosynthesis – cross-section and front view. (Reprinted with permission from Podola and Melkonian (2005), Copyright (2005), Springer Nature). b and c show a scheme of a fluorescent microalgal biosensor based on Chlamydomonas reinhardtii. The biological container and the optical compartment are represented at the top and the bottom, respectively. (b) The sample solution containing the herbicide under test flows into the biological cell; (c) the alga strain biomediator is photoactivated and the emitting fluorescence captured by the photodetector (in static condition). (d) Two biological arrays made up by six containers (10 mm ∅int  10 mm H, 785 μl each) for static and flow measurement mode, respectively. (Reprinted with permission from Scognamiglio et al. (2009), Copyright (2009), Springer Nature). e and f depict a microalgal dual-head biosensor for herbicide detection based on Dictyosphaerium chlorelloides. (e) Schematic diagram of the microalgal dual-head fiber-optic

808

J. Hurtado-Gallego et al.

with KCN, the organophosphorus insecticide methyl parathion and the herbicides DCMU (diuron), and MV finding a significant decrease in photochemical yields after 24-min exposure. In 2003, Védrine et al. (2003) reported a Chlorella vulgaris-based biosensor entrapping the microalgae on a quartz microfiber Whatman filter and placed in a five-membrane homemade flow cell. The biosensor was exposed to a range of herbicides: atrazine, DNOC, simazine, isoproturon, and diuron, and alterations in chlorophyll fluorescence were analyzed. Limits of detection were in the μg/L range of the herbicides; the lower detection limit was found with isoproturon and diuron (0.025 μg/L). The response of the biosensor to 1 μg/L of atrazine was claimed by the authors to be stable for over 1 month. Podola et al. (2004) constructed the so-called algal sensor chip (ASC) using strains of the green algae Klebsormidium sp. (filamentous) and Chlorella vulgaris. Algae were immobilized by filtration onto a membrane filter; the multiple-strain algal biosensor was located on a glass fiber and continuously supplied with culture medium. Biosensors were exposed to vapors of formaldehyde and methanol. The toxicity end point was the calculation of the effective quantum yield of photochemical energy conversion in PS II (ΔF/Fm0 ) (Genty et al. 1989). Formaldehyde vapors were detected down to the μg/L range, while they were less sensitive to methanol; the biosensor response was recorded within minutes, and it was reversible. In 2005, the same group reported an array-chip biosensor configuration containing nine microalgal strains (Podola and Melkonian 2005). Alterations in chlorophyll fluorescence were chosen as toxicity end point; the biosensor was exposed to the herbicides atrazine, simazine, diuron, isoproturon, and paraquat and was detectable within minutes at concentrations in the μg/L range. Herbicide specificity was encoded in the response pattern of the algal strains to each of the five herbicides. Nguyen-Ngoc and Tran-Minh (2007) reported the development of a Chlorella vulgaris-based biosensor; cells were entrapped in an inorganic translucent matrix so that the excitation light can penetrate the membrane in contrast with immobilization on an opaque matrix. The biosensor was exposed to the herbicide diuron; changes in chlorophyll fluorescence were the measured parameter with a detection limit of 1 μg/L; the immobilized algal cells could keep over 95% of their initial activity after a period of 5 weeks. A rather different approach was undertaken by Scognamiglio et al. (2009) who used several Chlamydomonas reinhardtii mutants of the D1 protein of PSII as biosensors for the detection of pollutants targeting the photosystem. Mutant cells were immobilized on silicon septa attached to a portable instrument denoted as ä Fig. 4 (continued) biosensor: a, water sample/culture medium; b, degasser; c, dual chamber measuring cell; d, peristaltic pump; e, optoelectronic O2 transducer; f, temperature probe; g, optical fibers for actinic illumination; h, blue LED source; i, laptop computer. (f) Exploded view of a flow chamber: g, optical fiber for actinic illumination; j, sapphire window; k, flow-through chamber; l, cellulose membrane; m, microalgal membrane; n, gray silicone layer; o, O2-sensitive film; p, polymer window; q, optical fiber to the optoelectronic O2 transducer. (Reprinted with permission from Haigh-Flórez et al. (2014), Copyright (2014), Elsevier)

32

Luminescent Microbial Bioassays and Microalgal Biosensors as Tools. . .

809

OPTICBIO-Multicell. Chlorophyll fluorescence parameters linked to PSII photochemical efficiency were calculated. The mutant strains showed different sensitivities to the different tested herbicides (atrazine, prometryn, and diuron) with limits of detection (LODs) in the low nM range. The different sensitivities were useful for the development of an integrated biosensor for the detection of complex environmental samples containing mixtures of pollutants that may affect PSII (Giardi et al. 2009). Peña-Vázquez et al. (2009) encapsulated three microalgal species, Dictyosphaerium chlorelloides, Scenedesmus intermedius, and Scenedesmus sp., in silicate sol-gel matrices and exposed them to triazines (simazine, atrazine, propazine, and terbuthylazine), the urea-based herbicide linuron, the hormonal herbicide 2,4D, and Cu2+. Chlorophyll fluorescence alterations were measured; the biosensors were sensitive to the triazines and linuron which target PSII but did not respond to 2,4 D or Cu2+. The limits of detection were in the low μg/L range, and the biosensors showed reversibility in their performance. Interestingly, the authors obtained a simazine-resistant mutant that was key for improving selectivity toward this herbicide. Most of the microalgal biosensors are based on freshwater species that might not be adequate for testing lagoons and coastal waters; in this context, it would be necessary to develop biosensors of environmental relevance for those ecosystems. Durrieu et al. (2011) constructed biosensors based on marine green alga Dunaliella tertiolecta and diatom Phaeodactylum tricornutum. Two biosensors were made: one based on esterase activity (conductometric, already discussed in the previous section) and an optical biosensor. The latter consisted of Dunaliella cells immobilized in a quartz Millipore fiber membrane; chlorophyll fluorescence was the measured parameter, and the biosensor was shown to be responsive to diuron. Ferro et al. (2012) constructed biosensors based on three species of green algae: Chlorella vulgaris, Pseudokirchneriella subcapitata (at present denoted as Raphidocelis subcapitata), and Chlamydomonas reinhardtii. Cells of each microalga were immobilized in alginate and translucent silica hydrogels following a two-step procedure. Biosensors were exposed to herbicides DCMU and atrazine; chlorophyll fluorescence was the measured parameter. Chlamydomonas reinhardtii was shown to be the most sensitive species with a limit of detection of 0.1 μM for atrazine after 40 min exposure. Later works of the same group (Durrieu et al. 2016; Perullini et al. 2014) reported enhancements in immobilization techniques of the three algae to improve the optical quality of the biosensors. Haigh-Flórez et al. (2014) developed a microalgal dual-head biosensor consisting in two strains of the green alga Dictyosphaerium chlorelloides, one sensitive and another resistant to the herbicide simazine. Cells were immobilized into porous silicone films. Cells were coupled to an optoelectronic device with an O2 transducer based on the oxygen-sensitive luminescent dye RD3; thus, photosynthetic activity was the chosen parameter. These dual biosensors may allow for selectivity toward target waterborne pollutants. Recently, a microfluidic device with integrated optical pH, oxygen sensors, and algal fluorescence was reported (Tahirbegi et al. 2017). Chlamydomonas reinhardtii was suspended in buffer solution and injected into the microfluidic device to form a biofilm at the solid-liquid interface on the bottom of the microfluidic device. The

810

J. Hurtado-Gallego et al.

biosensor was exposed to the herbicides simazine, atrazine, and diuron, and algal photosynthetic/respiratory activity was measured with the optical pH and oxygen sensors and intrinsic chlorophyll fluorescence. The authors claimed that this miniaturized system could determine the pesticide concentration in the nanomolar concentration range and in less than 10 min of exposure.

Conclusions and Future Directions Microorganisms-based toxicity bioassays and biosensors have emerged as a promising tool for environmental toxicity monitoring. Their ease of culture, low cost, and potential of mass production have made them a good complement to chemical analyses; they report on global toxicity and, most importantly, on pollutant bioavailability. There is a variety of bioassays using microorganisms of ecological relevance, which may monitor toxicity in different environmental compartments: water, soil, or the atmosphere. Some of them have been already validated and standardized by international agencies, have been commercialized, and are used worldwide. Most microorganisms may be immobilized in a number of inert matrices, so that biosensors can be fabricated. There is the possibility of biosensor miniaturization and integration in microfluidics to construct lab-on-a-chip devices, so that multiplex biosensors may be made. These biosensors may allow for in situ, online, and continuous toxicity monitoring. Despite their advantages, some aspects should be improved before this technology can get out of the laboratory and be implemented by regulatory agencies and stakeholders: • Validation and standardization by international agencies of microorganismsbased toxicity bioassays and biosensors should be extended. • Most microorganisms-based toxicity bioassays and biosensors are intended to assess acute toxicity effects, which are relevant when many samples have to be tested, as early warning systems, but chronic toxicity is much more realistic, and efforts in that direction should be made. • New immobilization procedures should be developed so that many toxicity bioassays may be transformed into biosensor configuration. Nanotechnology may help in this direction and in further miniaturization of the biosensor devices. • Ecologically relevant bioassays and biosensors should be used depending on the environmental sample, e.g., bioassays based on marine microorganisms may be not completely adequate for testing freshwater samples. • Most microorganisms-based toxicity bioassays and biosensors have not been tested with complex mixtures or real environmental samples, only with a selection of pollutants applied singly in the laboratory. This should be necessary for their implementation. • In the case of bioluminescent recombinant bioreporters, it should be advisable the expression of the genes for aldehyde synthesis when using lux genes and to find

32

Luminescent Microbial Bioassays and Microalgal Biosensors as Tools. . .

811

the way for intracellular luciferin production when using eukaryotic luciferase reporter genes. • Long-term storage and reusability of biosensors versus single-use, disposable inexpensive biosensors is an important decision to be made which may be critical for commercialization and extended use of the biosensors. Acknowledgments This research was supported by the Spanish Ministry of Economy and Competitiveness (MINECO), grant CTM2016-74927-C2-2-R. JHG is working under FPI contract (MINECO-EU). MGP acknowledges a postdoctoral contract from Comunidad de Madrid (CMEU).

References Abbas M et al (2018) Vibrio fischeri bioluminescence inhibition assay for ecotoxicity assessment: a review. Sci Total Environ 626:1295–1309 Abd-El-Haleem D, Zaki S, Abulhamd A, Elbery H, Abu-Elreesh G (2006) Acinetobacter bioreporter assessing heavy metals toxicity. J Basic Microbiol 46:339–347 Ahn J-M, Gu MB (2012) Geno-Tox: cell array biochip for genotoxicity monitoring and classification. Appl Biochem Biotechnol 168:752–760 Aichberger H, Loibner AP, Celis R, Braun R, Ottner F, Rost H (2006) Assessment of factors governing biodegradability of PAHs in three soils aged under field conditions. Soil Sediment Contamin 15:73–85 Amin-Hanjani S, Meikle A, Glover L, Prosser J, Killham K (1993) Plasmid and chromosomally encoded luminescence marker systems for detection of Pseudomonas fluorescens in soil. Mol Ecol 2:47–54 Andreani T et al (2017) Influence of the stabilizers on the toxicity of metallic nanomaterials in aquatic organisms and human cell lines. Sci Total Environ 607:1264–1277 Arias-Barreiro CR et al (2010) A bacterial biosensor for oxidative stress using the constitutively expressed redox-sensitive protein roGFP2. Sensors 10:6290–6306 Aruoja V, Pokhrel S, Sihtmäe M, Mortimer M, Mädler L, Kahru A (2015) Toxicity of 12 metalbased nanoparticles to algae, bacteria and protozoa. Environ Sci Nano 2:630–644 Avramescu A, Rouillon R, Carpentier R (1999) Potential for use of a cyanobacterium Synechocystis sp. immobilized in poly (vinylalcohol): application to the detection of pollutants. Biotechnol Tech 13:559–562 Azimi Y, Thompson IP (2017) Phosphorus depletion as a green alternative to biocides for controlling biodegradation of metalworking fluids. Environ Sci Technol 51:5695–5702 Bao VW, Koutsaftis A, Leung KM (2008) Temperature-dependent toxicities of chlorothalonil and copper pyrithione to the marine copepod ‘Tigriopus japonicus’ and dinoflagellate ‘Pyrocystis lunula’. Australas J Ecotoxicol 14:45 Bao VW, Leung KM, Qiu J-W, Lam MH (2011) Acute toxicities of five commonly used antifouling booster biocides to selected subtropical and cosmopolitan marine species. Mar Pollut Bull 62:1147–1151 Barrena R, Casals E, Colón J, Font X, Sánchez A, Puntes V (2009) Evaluation of the ecotoxicity of model nanoparticles. Chemosphere 75:850–857 Bazin I, Seo HB, Suehs CM, Ramuz M, De Waard M, Gu MB (2017) Profiling the biological effects of wastewater samples via bioluminescent bacterial biosensors combined with estrogenic assays. Environ Sci Pollut Res 24:33–41 Belkin S (2003) Microbial whole-cell sensing systems of environmental pollutants. Curr Opin Microbiol 6:206–212

812

J. Hurtado-Gallego et al.

Belkin S, Smulski DR, Vollmer AC, Van Dyk TK, LaRossa RA (1996) Oxidative stress detection with Escherichia coli harboring a katG’:: lux fusion. Appl Environ Microbiol 62:2252–2256 Belkin S, Smulski DR, Dadon S, Vollmer AC, Van Dyk TK, Larossa RA (1997) A panel of stressresponsive luminous bacteria for the detection of selected classes of toxicants. Water Res 31:3009–3016 Ben-Israel O, Ben-Israel H, Ulitzur S (1998) Identification and quantification of toxic chemicals by use of Escherichia coli carryinglux genes fused to stress promoters. Appl Environ Microbiol 64:4346–4352 Bentley A, Atkinson A, Jezek J, Rawson D (2001) Whole cell biosensors – electrochemical and optical approaches to ecotoxicity testing. Toxicol In Vitro 15:469–475 Betteridge DJ (2000) What is oxidative stress? Metabolism 49:3–8 Bhattacharyya J, Read D, Amos S, Dooley S, Killham K, Paton GI (2005) Biosensor-based diagnostics of contaminated groundwater: assessment and remediation strategy. Environ Pollut 134:485–492 Bispo A, Jourdain MJ, Jauzein M (1999) Toxicity and genotoxicity of industrial soils polluted by polycyclic aromatic hydrocarbons (PAHs). Org Geochem 30:947–952 Bondarenko O, Rahman PK, Rahman TJ, Kahru A, Ivask A (2010) Effects of rhamnolipids from Pseudomonas aeruginosa DS10-129 on luminescent bacteria: toxicity and modulation of cadmium bioavailability. Microb Ecol 59:588–600 Bondarenko O, Ivask A, Käkinen A, Kahru A (2012) Sub-toxic effects of CuO nanoparticles on bacteria: kinetics, role of Cu ions and possible mechanisms of action. Environ Pollut 169:81–89 Bondarenko OM et al (2016) Multilaboratory evaluation of 15 bioassays for (eco) toxicity screening and hazard ranking of engineered nanomaterials: FP7 project NANOVALID. Nanotoxicology 10:1229–1242 Bondarenko O, Sihtmäe MM, Kuzmičiova J, Ragelien_e L, Kahru A, Daugelavičius R (2018) Bacterial plasma membrane is the main cellular target of silver nanoparticles in Escherichia coli and Pseudomonas aeruginosa. bioRxiv. https://doi.org/10.1101/322727 Booth AM, Hansen BH, Frenzel M, Johnsen H, Altin D (2016) Uptake and toxicity of methylmethacrylate-based nanoplastic particles in aquatic organisms. Environ Toxicol Chem 35:1641–1649 Boyd EM, Killham K, Meharg AA (2001) Toxicity of mono-, di-and tri-chlorophenols to lux marked terrestrial bacteria, Burkholderia species Rasc c2 and Pseudomonas fluorescens. Chemosphere 43:157–166 Brandt KK, Pedersen A, Sørensen J (2002) Solid-phase contact assay that uses a lux-marked Nitrosomonas europaea reporter strain to estimate toxicity of bioavailable linear alkylbenzene sulfonate in soil. Appl Environ Microbiol 68:3502–3508 Brandt KK, Holm PE, Nybroe O (2006) Bioavailability and toxicity of soil particle-associated copper as determined by two bioluminescent Pseudomonas fluorescens biosensor strains. Environ Toxicol Chem 25:1738–1741 Brandt KK, Holm PE, Nybroe O (2008) Evidence for bioavailable copper dissolved organic matter complexes and transiently increased copper bioavailability in manure-amended soils as determined by bioluminescent bacterial biosensors. Environ Sci Technol 42:3102–3108 Brayner R, Couté A, Livage J, Perrette C, Sicard C (2011) Micro-algal biosensors. Anal Bioanal Chem 401:581–597 Brouwer H, Murphy T, McArdle L (1990) A sediment-contact bioassay with Photobacterium phosphoreum. Environ Toxicol Chem 9:1353–1358 Brown J, Rattray E, Paton G, Reid G, Caffoor I, Killham K (1996) Comparative assessment of the toxicity of a papermill effluent by respirometry and a luminescence-based bacterial assay. Chemosphere 32:1553–1561 Buckova M, Licbinsky R, Jandova V, Krejci J, Pospichalova J, Huzlik J (2017) Fast ecotoxicity detection using biosensors. Water Air Soil Pollut 228:166 Bulich AA, Isenberg D (1981) Use of the luminescent bacterial system for the rapid assessment of aquatic toxicity. ISA Trans 20:29–33

32

Luminescent Microbial Bioassays and Microalgal Biosensors as Tools. . .

813

Burga-Perez KF, Toumi H, Cotelle S, Ferard J-F, Radetski CM (2013) Sensitivity of different aquatic bioassays in the assessment of a new natural formicide. J Environ Sci Health B 48:57–62 Campanella L, Cubadda F, Sammartino M, Saoncella A (2001) An algal biosensor for the monitoring of water toxicity in estuarine environments. Water Res 35:69–76 Chaudri AM et al (1999) Determination of acute Zn toxicity in pore water from soils previously treated with sewage sludge using bioluminescence assays. Environ Sci Technol 33:1880–1885 Chinalia F, Paton GI, Killham K (2008) Physiological and toxicological characterization of an engineered whole-cell biosensor. Bioresour Technol 99:714–721 Cho J-C et al (2004) A novel continuous toxicity test system using a luminously modified freshwater bacterium. Biosens Bioelectron 20:338–344 Choi SH, Gu MB (2002) A portable toxicity biosensor using freeze-dried recombinant bioluminescent bacteria. Biosens Bioelectron 17:433–440 Chong KF, Loh KP, Ang K, Ting YP (2008) Whole cell environmental biosensor on diamond. Analyst 133:739–743 Chou T-C (2006) Theoretical basis, experimental design, and computerized simulation of synergism and antagonism in drug combination studies. Pharmacol Rev 58:621–681 Chou T-C, Talalay P (1984) Quantitative analysis of dose-effect relationships: the combined effects of multiple drugs or enzyme inhibitors. Adv Enzym Regul 22:27–55 Chouteau C, Dzyadevych S, Chovelon J-M, Durrieu C (2004) Development of novel conductometric biosensors based on immobilised whole cell Chlorella vulgaris microalgae. Biosens Bioelectron 19:1089–1096 Corrêa AX, Tiepo EN, Somensi CA, Sperb RM, Radetski CM (2009) Use of ozone-photocatalytic oxidation (O3/UV/TiO2) and biological remediation for treatment of produced water from petroleum refineries. J Environ Eng 136:40–45 Cui Z et al (2018) Application of a bacterial whole cell biosensor for the rapid detection of cytotoxicity in heavy metal contaminated seawater. Chemosphere 200:322–329 Cukurluoglu S, Muezzinoglu A (2013) Assessment of toxicity in waters due to heavy metals derived from atmospheric deposition using Vibrio fischeri. J Environ Sci Health A 48:57–66 Dams R, Biswas A, Olesiejuk A, Fernandes T, Christofi N (2011) Silver nanotoxicity using a lightemitting biosensor Pseudomonas putida isolated from a wastewater treatment plant. J Hazard Mater 195:68–72 Dawson JJC, Campbell C, Towers W, Cameron C, Paton GI (2006) Linking biosensor responses to Cd, Cu and Zn partitioning in soils. Environ Pollut 142:493–500 Dawson J, Iroegbu C, Maciel H, Paton G (2008) Application of luminescent biosensors for monitoring the degradation and toxicity of BTEX compounds in soils. J Appl Microbiol 104:141–151 de García SO, García-Encina PA, Irusta-Mata R (2016) Dose–response behavior of the bacterium Vibrio fischeri exposed to pharmaceuticals and personal care products. Ecotoxicology 25:141–162 de OF Rossetto AL, Melegari SP, Ouriques LC, Matias WG (2014a) Comparative evaluation of acute and chronic toxicities of CuO nanoparticles and bulk using Daphnia magna and Vibrio fischeri. Sci Total Environ 490:807–814 de OF Rossetto AL, Vicentini DS, Costa CH, Melegari SP, Matias WG (2014b) Synthesis, characterization and toxicological evaluation of a core–shell copper oxide/polyaniline nanocomposite. Chemosphere 108:107–114 Deryabin D, Aleshina E, Efremova L (2012) Application of the inhibition of bacterial bioluminescence test for assessment of toxicity of carbon-based nanomaterials. Microbiology 81:492–497 Deryabin D, Efremova L, Karimov I, Manukhov I, Gnuchikh EY, Miroshnikov S (2016) Comparative sensitivity of the luminescent Photobacterium phosphoreum, Escherichia coli, and Bacillus subtilis strains to toxic effects of carbon-based nanomaterials and metal nanoparticles. Microbiology 85:198–206 Ding K et al (2017) In vitro and in silico investigations of the binary-mixture toxicity of phthalate esters and cadmium (II) to Vibrio qinghaiensis sp.-Q67. Sci Total Environ 580:1078–1084

814

J. Hurtado-Gallego et al.

Durrieu C, Guedri H, Fremion F, Volatier L (2011) Unicellular algae used as biosensors for chemical detection in Mediterranean lagoon and coastal waters. Res Microbiol 162:908–914 Durrieu C, Ferro Y, Perullini M, Gosset A, Jobbágy M, Bilmes SA (2016) Feasibility of using a translucid inorganic hydrogel to build a biosensor using immobilized algal cells. Environ Sci Pollut Res 23:9–13 Dzyadevych SV, Chovelon J-M (2002) A comparative photodegradation studies of methyl parathion by using Lumistox test and conductometric biosensor technique. Mater Sci Eng C 21:55–60 Ejeian F et al (2018) Biosensors for wastewater monitoring: a review. Biosens Bioelectron 118:66 El-Alawi YS, Dixon DG, Greenberg BM (2001) Effects of a pre-incubation period on the photoinduced toxicity of polycyclic aromatic hydrocarbons to the luminescent bacterium Vibrio fischeri. Environ Toxicol Int J 16:277–286 Ellouze M, Saddoud A, Dhouib A, Sayadi S (2009) Assessment of the impact of excessive chemical additions to municipal wastewaters and comparison of three technologies in the removal performance of pathogens and toxicity. Microbiol Res 164:138–148 Farré M, Barceló D (2003) Toxicity testing of wastewater and sewage sludge by biosensors, bioassays and chemical analysis TrAC. Trends Anal Chem 22:299–310 Farré M, Gonçalves C, Lacorte S, Barcelo D, Alpendurada M (2002a) Pesticide toxicity assessment using an electrochemical biosensor with Pseudomonas putida and a bioluminescence inhibition assay with Vibrio fischeri. Anal Bioanal Chem 373:696–703 Farré M, Klöter G, Petrovic M, Alonso MC, de Alda MJL, Barceló D (2002b) Identification of toxic compounds in wastewater treatment plants during a field experiment. Anal Chim Acta 456:19–30 Farré M, Asperger D, Kantiani L, González S, Petrovic M, Barceló D (2008) Assessment of the acute toxicity of triclosan and methyl triclosan in wastewater based on the bioluminescence inhibition of Vibrio fischeri. Anal Bioanal Chem 390:1999–2007 Fekete-Kertész I, Piszmán D, Molnár M (2017) Particle size and concentration dependent ecotoxicity of nano-and microscale TiO2 – comparative study by different aquatic test organisms of different trophic levels. Water Air Soil Pollut 228:245 Fernández-Pinas F, Wolk CP (1994) Expression of luxCD-E in Anabaena sp. can replace the use of exogenous aldehyde for in vivo localization of transcription by luxAB. Gene 150:169–174 Fernández-Piñas F, Rodea-Palomares I, Leganés F, González-Pleiter M, Muñoz-Martín MA (2014) Evaluation of the ecotoxicity of pollutants with bioluminescent microorganisms. In: Bioluminescence: fundamentals and applications in biotechnology, vol 2. Springer, Berlin, pp 65–135 Ferro Y, Perullini M, Jobbagy M, Bilmes SA, Durrieu C (2012) Development of a biosensor for environmental monitoring based on microalgae immobilized in silica hydrogels. Sensors 12:16879–16891 Fetters KJ, Costello DM, Hammerschmidt CR, Burton GA Jr (2016) Toxicological effects of shortterm resuspension of metal-contaminated freshwater and marine sediments. Environ Toxicol Chem 35:676–686 Flynn HC, Mc Mahon V, Diaz GC, Demergasso CS, Corbisier P, Meharg AA, Paton GI (2002) Assessment of bioavailable arsenic and copper in soils and sediments from the Antofagasta region of northern Chile. Sci Total Environ 286:51–59 Frense D, Müller A, Beckmann D (1998) Detection of environmental pollutants using optical biosensor with immobilized algae cells. Sensors Actuators B Chem 51:256–260 Gagne F (2014) Biochemical ecotoxicology: principles and methods. Elsevier, Amsterdam Gao L, Chen J, Tang C, Ke Z, Wang J, Shimizu Y, Zhu A (2015) Distribution, migration and potential risk of heavy metals in the Shima River catchment area, South China. Environ Sci: Processes Impacts 17:1769–1782 Genty B, Briantais JM, Baker NR (1989) The relationship between the quantum yield of photosynthetic electron transport and quenching of chlorophyll fluorescence Biochim. Biophys. Acta 990:87–92

32

Luminescent Microbial Bioassays and Microalgal Biosensors as Tools. . .

815

Geoffroy L et al (2007) Effect of selenate on growth and photosynthesis of Chlamydomonas reinhardtii. Aquat Toxicol 83:149–158 Georgantzopoulou A et al (2012) Ag nanoparticles: size-and surface-dependent effects on model aquatic organisms and uptake evaluation with NanoSIMS. Nanotoxicology 7:1168–1178 Giardi MT, Pace E (2005) Photosynthetic proteins for technological applications. Trends Biotechnol 23:257–263 Giardi MT et al (2009) Optical biosensors for environmental monitoring based on computational and biotechnological tools for engineering the photosynthetic D1 protein of Chlamydomonas reinhardtii. Biosens Bioelectron 25:294–300 Gil GC, Mitchell RJ, Chang ST, Gu MB (2000) A biosensor for the detection of gas toxicity using a recombinant bioluminescent bacterium. Biosens Bioelectron 15:23–30 Gil GC, Kim YJ, Gu MB (2002) Enhancement in the sensitivity of a gas biosensor by using an advanced immobilization of a recombinant bioluminescent bacterium. Biosens Bioelectron 17:427–432 González-Pleiter M et al (2013) Toxicity of five antibiotics and their mixtures towards photosynthetic aquatic organisms: implications for environmental risk assessment. Water Res 47:2050–2064 Gouider M, Feki M, Sayadi S (2010) Bioassay and use in irrigation of untreated and treated wastewaters from phosphate fertilizer industry. Ecotoxicol Environ Saf 73:932–938 Gu MB, Chang ST (2001) Soil biosensor for the detection of PAH toxicity using an immobilized recombinant bacterium and a biosurfactant. Biosens Bioelectron 16:667–674 Guedri H, Durrieu C (2008) A self-assembled monolayers based conductometric algal whole cell biosensor for water monitoring. Microchim Acta 163:179–184 Haigh-Flórez D, de la Hera C, Costas E, Orellana G (2014) Microalgae dual-head biosensors for selective detection of herbicides with fiber-optic luminescent O2 transduction. Biosens Bioelectron 54:484–491 Hassan SH, Oh SE (2010) Improved detection of toxic chemicals by Photobacterium phosphoreum using modified Boss medium. J Photochem Photobiol B Biol 101:16–21 Hassan SH, Van Ginkel SW, Hussein MA, Abskharon R, Oh S-E (2016) Toxicity assessment using different bioassays and microbial biosensors. Environ Int 92:106–118 Hechmi N, Bosso L, El-Bassi L, Scelza R, Testa A, Jedidi N, Rao MA (2016) Depletion of pentachlorophenol in soil microcosms with Byssochlamys nivea and Scopulariopsis brumptii as detoxification agents. Chemosphere 165:547–554 Heimann K, Matuszewski JM, Klerks PL (2002) Effects of metals and organic contaminants on the recovery of bioluminescence in the marine dinoflagellate Pyrocystis lunula (Dinophyceae). J Phycol 38:482–492 Heinlaan M, Ivask A, Blinova I, Dubourguier H-C, Kahru A (2008) Toxicity of nanosized and bulk ZnO, CuO and TiO2 to bacteria Vibrio fischeri and crustaceans Daphnia magna and Thamnocephalus platyurus. Chemosphere 71:1308–1316 Hermens J, Busser F, Leeuwangh P, Musch A (1985) Quantitative structure-activity relationships and mixture toxicity of organic chemicals in Photobacterium phosphoreum: the Microtox test. Ecotoxicol Environ Saf 9:17–25 Hernando M, De Vettori S, Bueno MM, Fernández-Alba A (2007) Toxicity evaluation with Vibrio fischeri test of organic chemicals used in aquaculture. Chemosphere 68:724–730 Hildenbrand ZL et al (2015) Rapid analysis of eukaryotic bioluminescence to assess potential groundwater contamination events. J Chem 2015:1 Hirmann D, Loibner AP, Braun R, Szolar OH (2007) Applicability of the bioluminescence inhibition test in the 96-well microplate format for PAH-solutions and elutriates of PAHcontaminated soils. Chemosphere 67:1236–1242 Hjorth R, Coutris C, Nguyen NH, Sevcu A, Gallego-Urrea JA, Baun A, Joner EJ (2017) Ecotoxicity testing and environmental risk assessment of iron nanomaterials for sub-surface remediation–recommendations from the FP7 project NanoRem. Chemosphere 182:525–531

816

J. Hurtado-Gallego et al.

Hollis R, Killham K, Glover L (2000) Design and application of a biosensor for monitoring toxicity of compounds to eukaryotes. Appl Environ Microbiol 66:1676–1679 Horsburgh AM et al (2002) On-line microbial biosensing and fingerprinting of water pollutants. Biosens Bioelectron 17:495–501 Horswell J, Weitz HJ, Percival HJ, Speir TW (2006) Impact of heavy metal amended sewage sludge on forest soils as assessed by bacterial and fungal biosensors. Biol Fertil Soils 42:569–576 Hsieh C-Y, Tsai M-H, Ryan DK, Pancorbo OC (2004) Toxicity of the 13 priority pollutant metals to Vibrio fisheri in the Microtox® chronic toxicity test. Sci Total Environ 320:37–50 Hurtado-Gallego J, Martín-Betancor K, Rodea-Palomares I, Leganés F, Rosal R, Fernández-Piñas F (2018) Two novel cyanobacterial bioluminescent whole-cell bioreporters based on superoxide dismutases MnSod and FeSod to detect superoxide anion. Chemosphere 201:772–779 Iizumi T, Mizumoto M, Nakamura K (1998) A bioluminescence assay using nitrosomonas europaeafor rapid and sensitive detection of nitrification inhibitors. Appl Environ Microbiol 64:3656–3662 Ionescu RE, Abu-Rabeah K, Cosnier S, Durrieu C, Chovelon JM, Marks RS (2006) Amperometric algal Chlorella vulgaris cell biosensors based on alginate and polypyrrole-alginate gels. Electroanalysis 18:1041–1046 Ivask A, François M, Kahru A, Dubourguier H-C, Virta M, Douay F (2004) Recombinant luminescent bacterial sensors for the measurement of bioavailability of cadmium and lead in soils polluted by metal smelters. Chemosphere 55:147–156 Ivask A, Green T, Polyak B, Mor A, Kahru A, Virta M, Marks R (2007) Fibre-optic bacterial biosensors and their application for the analysis of bioavailable Hg and As in soils and sediments from Aznalcollar mining area in Spain. Biosens Bioelectron 22:1396–1402 Ivask A, Rõlova T, Kahru A (2009) A suite of recombinant luminescent bacterial strains for the quantification of bioavailable heavy metals and toxicity testing. BMC Biotechnol 9:41 Ivask A, Bondarenko O, Jepihhina N, Kahru A (2010) Profiling of the reactive oxygen speciesrelated ecotoxicity of CuO, ZnO, TiO2, silver and fullerene nanoparticles using a set of recombinant luminescent Escherichia coli strains: differentiating the impact of particles and solubilised metals. Anal Bioanal Chem 398:701–716 Jemec A et al (2016) An interlaboratory comparison of nanosilver characterisation and hazard identification: harmonising techniques for high quality data. Environ Int 87:20–32 Jennings VL, Rayner-Brandes MH, Bird DJ (2001) Assessing chemical toxicity with the bioluminescent photobacterium (Vibrio fischeri): a comparison of three commercial systems. Water Res 35:3448–3456 Ji J-Y, Xing Y-J, Ma Z-T, Zhang M, Zheng P (2013) Acute toxicity of pharmaceutical wastewaters containing antibiotics to anaerobic digestion treatment. Chemosphere 91:1094–1098 Joost U et al (2015) Photocatalytic antibacterial activity of nano-TiO2 (anatase)-based thin films: effects on Escherichia coli cells and fatty acids. J Photochem Photobiol B Biol 142:178–185 Jośko I, Oleszczuk P (2013) The influence of ZnO and TiO2 nanoparticles on the toxicity of sewage sludges. Environ Sci: Processes Impacts 15:296–306 Jung Y, Park C-B, Kim Y, Kim S, Pflugmacher S, Baik S (2015) Application of multi-species microbial bioassay to assess the effects of engineered nanoparticles in the aquatic environment: potential of a luminous microbial array for toxicity risk assessment (LumiMARA) on testing for surface-coated silver nanoparticles. Int J Environ Res Public Health 12:8172–8186 Jurado E, Fernández-Serrano M, Núñez J, Lechuga M, Luzón G (2008) Evolution of acute toxicity of non-ionic surfactants over the biodegradation process. Environ Toxicol II 110:11101 Kahru A, Tomson K, Pall T, Külm I (1996) Study of toxicity of pesticides using luminescent bacteria Photobacterium phosphoreum. Water Sci Technol 33:147–154 Kaiser KL, Palabrica VS (1991) Photobacterium phosphoreum toxicity data index. Water Qual Res J 26:361–431 Kang I-M, Kong IC (2016) Effects of properties of metal-contaminated soils on bacterial bioluminescence activity, seed germination, and root and shoot growth. Springerplus 5:272

32

Luminescent Microbial Bioassays and Microalgal Biosensors as Tools. . .

817

Katsoyiannis A, Samara C (2007) Ecotoxicological evaluation of the wastewater treatment process of the sewage treatment plant of Thessaloniki, Greece. J Hazard Mater 141:614–621 Kelly C, Lajoie C, Layton A, Sayler G (1999) Bioluminescent reporter bacterium for toxicity monitoring in biological wastewater treatment systems. Water Environ Res 71:31–35 Kim BC, Gu MB (2005) A multi-channel continuous water toxicity monitoring system: its evaluation and application to water discharged from a power plant. Environ Monit Assess 109:123–133 Kim B, Gu M, Hansen P (2005) Evaluation of water-borne toxicity using bioluminescent bacteria. In: Environmental chemistry. Springer, Berlin, pp 699–705 Knox OG, Weitz HJ, Anderson P, Borlinghaus M, Fountaine J (2018) Improved screening of biochar compounds for potential toxic activity with microbial biosensors. Environ Technol Innov 9:254 Ko K-S, Kong IC (2014) Toxic effects of nanoparticles on bioluminescence activity, seed germination, and gene mutation. Appl Microbiol Biotechnol 98:3295–3303 Kokkali V, van Delft W (2014) Overview of commercially available bioassays for assessing chemical toxicity in aqueous samples. Trends Anal Chem 61:133–155 Kong IC (2008) Microbial characteristics associated with six different organic wastes undergoing anaerobic decomposition in batch vial conditions. Waste Manag Res 26:261–266 Kong I-C (2011) Comparative studies on the acute toxicities of whole solids and solids aqueous extracts based on the inhibition of bacterial bioluminescence production. Environ Eng Res 16:41–45 Kong IC, Raliya R, Ko K-S, Biswas P (2018) ZnO nanoparticles: effect of size on bacterial bioluminescence, seed germination, algal growth, and gene mutation. Environ Eng Sci 35:231–239 Kováts N, Refaey M, Varanka B, Reich K, Ferincz Á, Ács A (2012) Comparison of conventional and Vibrio fischeri bioassays for the assessment of municipal wastewater toxicity. Environ Eng Manag J 11:2073–2076 Krause G, Weis E (1991) Chlorophyll fluorescence and photosynthesis: the basics. Annu Rev Plant Biol 42:313–349 Kungolos A, Hadjispyrou S, Petala M, Tsiridis V, Samaras P, Sakellaropoulos G (2004) Toxic properties of metals and organotin compounds and their interactions on Daphnia magna and Vibrio fischeri. Water Air Soil Pollut: Focus 4:101–110 Kurvet I, Ivask A, Bondarenko O, Sihtmäe M, Kahru A (2011) LuxCDABE – transformed constitutively bioluminescent Escherichia coli for toxicity screening: comparison with naturally luminous Vibrio fischeri. Sensors 11:7865–7878 Kurvet I, Juganson K, Vija H, Sihtmäe M, Blinova I, Syvertsen-Wiig G, Kahru A (2017) Toxicity of nine (doped) rare earth metal oxides and respective individual metals to aquatic microorganisms Vibrio fischeri and Tetrahymena thermophila. Materials 10:754 la Farré M, Garcıa M-J, Tirapu L, Ginebreda A, Barceló D (2001) Wastewater toxicity screening of non-ionic surfactants by Toxalert® and Microtox ® bioluminescence inhibition assays. Anal Chim Acta 427:181–189 Lagarde F, Jaffrezic-Renault N (2011) Cell-based electrochemical biosensors for water quality assessment. Anal Bioanal Chem 400:947 Lajoie CA, Lin S-C, Nguyen H, Kelly CJ (2002) A toxicity testing protocol using a bioluminescent reporter bacterium from activated sludge. J Microbiol Methods 50:273–282 Lapota D, Moskowitz GJ, Rosenberger DE, Grovhoug JG (1993) The use of stimulable bioluminescence from marine dinoflagellates as a means of detecting toxicity in the marine environment. In: Environmental toxicology and risk assessment, vol 2. ASTM International, Philadelphia Lapota D, Rosenberger D, Duckworth D (1994) A bioluminescent dinoflagellate assay for detecting toxicity in coastal waters. In: Bioluminescence and chemiluminescence: fundamentals and applied aspects. Wiley, Chichester, pp 156–159

818

J. Hurtado-Gallego et al.

Lapota D, Osorio AR, Liao C, Bjorndal B (2007) The use of bioluminescent dinoflagellates as an environmental risk assessment tool. Mar Pollut Bull 54:1857–1867 Latifi A, Ruiz M, Zhang C-C (2009) Oxidative stress in cyanobacteria. FEMS Microbiol Rev 33:258–278 Layton A, Gregory B, Schultz T, Sayler G (1999) Validation of genetically engineered bioluminescent surfactant resistant bacteria as toxicity assessment tools. Ecotoxicol Environ Saf 43:222–228 Lechuga M, Fernández-Serrano M, Jurado E, Núñez-Olea J, Ríos F (2016) Acute toxicity of anionic and non-ionic surfactants to aquatic organisms. Ecotoxicol Environ Saf 125:1–8 Lee H, Gu M (2003) Construction of a sodA:: luxCDABE fusion Escherichia coli: comparison with a katG fusion strain through their responses to oxidative stresses. Appl Microbiol Biotechnol 60:577–580 Lee JH, Gu MB (2005) An integrated mini biosensor system for continuous water toxicity monitoring. Biosens Bioelectron 20:1744–1749 Lee HY, Choi SH, Gu MB (2003) Response of bioluminescent bacteria to sixteen azo dyes. Biotechnol Bioprocess Eng 8:101–105 Lee JH, Mitchell RJ, Kim BC, Cullen DC, Gu MB (2005) A cell array biosensor for environmental toxicity analysis. Biosens Bioelectron 21:500–507 Lei Y, Chen W, Mulchandani A (2006) Microbial biosensors. Anal Chim Acta 568:200–210 Li K, Yediler A, Yang M, Schulte-Hostede S, Wong MH (2008) Ozonation of oxytetracycline and toxicological assessment of its oxidation by-products. Chemosphere 72:473–478 Livingstone D (2001) Contaminant-stimulated reactive oxygen species production and oxidative damage in aquatic organisms. Mar Pollut Bull 42:656–666 Lushchak VI (2011) Adaptive response to oxidative stress: bacteria, fungi, plants and animals. Comp Biochem Physiol C Toxicol Pharmacol 153:175–190 Ma M, Wang C, Wang Z (2005) Assessing toxicities of hydrophobic organic pollutants in Huaihe River by using two types of sampling. J Environ Sci Health 40:331–342 Mallevre F, Fernandes TF, Aspray TJ (2014) Silver, zinc oxide and titanium dioxide nanoparticle ecotoxicity to bioluminescent Pseudomonas putida in laboratory medium and artificial wastewater. Environ Pollut 195:218–225 Mallevre F, Alba C, Milne C, Gillespie S, Fernandes TF, Aspray TJ (2016) Toxicity testing of pristine and aged silver nanoparticles in real wastewaters using bioluminescent Pseudomonas putida. Nano 6:49 Manke A, Wang L, Rojanasakul Y (2013) Pulmonary toxicity and fibrogenic response of carbon nanotubes. Toxicol Mech Methods 23:196–206 Martín-de-Lucía I et al (2017) Reverse Trojan-horse effect decreased wastewater toxicity in the presence of inorganic nanoparticles. Environ Sci Nano 4:1273–1282 McConkey BJ, Duxbury CL, Dixon DG, Greenberg BM (1997) Toxicity of a PAH photooxidation product to the bacteria Photobacterium phosphoreum and the duckweed Lemna gibba: effects of phenanthrene and its primary photoproduct, phenanthrenequinone. Environ Toxicol Chem 16:892–899 McGrath SP, Knight B, Killham K, Preston S, Paton GI (1999) Assessment of the toxicity of metals in soils amended with sewage sludge using a chemical speciation technique and a lux-based biosensor. Environ Toxicol Chem 18:659–663 Mendes LF, Stevani CV (2010) Evaluation of metal toxicity by a modified method based on the fungus Gerronema viridilucens bioluminescence in agar medium. Environ Toxicol Chem 29:320–326 Mendes LF, Bastos EL, Stevani CV (2010) Prediction of metal cation toxicity to the bioluminescent fungus Gerronema viridilucens. Environ Toxicol Chem 29:2177–2181 Michelini E, Roda A (2012) Staying alive: new perspectives on cell immobilization for biosensing purposes. Anal Bioanal Chem 402:1785–1797 Min J, Lee CW, Gu MB (2003) Gamma-radiation dose-rate effects on DNA damage and toxicity in bacterial cells. Radiat Environ Biophys 42:189–192

32

Luminescent Microbial Bioassays and Microalgal Biosensors as Tools. . .

819

Mitchell RJ, Gu MB (2004) An Escherichia coli biosensor capable of detecting both genotoxic and oxidative damage. Appl Microbiol Biotechnol 64:46–52 Mitchell RJ, Gu MB (2006) Characterization and optimization of two methods in the immobilization of 12 bioluminescent strains. Biosens Bioelectron 22:192–199 Mo L-Y, Liu J, Qin L-T, Zeng H-H, Liang Y-P (2017) Two-stage prediction on effects of mixtures containing phenolic compounds and heavy metals on Vibrio qinghaiensis sp. Q67. Bull Environ Contam Toxicol 99:17–22 Molins-Delgado D, Gago-Ferrero P, Díaz-Cruz MS, Barceló D (2016) Single and joint ecotoxicity data estimation of organic UV filters and nanomaterials toward selected aquatic organisms. Urban groundwater risk assessment. Environ Res 145:126–134 Mollon L, Norton G, Trakal L, Moreno-Jimenez E, Elouali F, Hough R, Beesley L (2016) Mobility and toxicity of heavy metal (loid) s arising from contaminated wood ash application to a pasture grassland soil. Environ Pollut 218:419–427 Monaghan P, Metcalfe NB, Torres R (2009) Oxidative stress as a mediator of life history trade-offs: mechanisms, measurements and interpretation. Ecol Lett 12:75–92 Mortimer M, Kasemets K, Heinlaan M, Kurvet I, Kahru A (2008) High throughput kinetic Vibrio fischeri bioluminescence inhibition assay for study of toxic effects of nanoparticles. Toxicol In Vitro 22:1412–1417 Muneer M, Theurich J, Bahnemann D (1999) Formation of toxic intermediates upon the photocatalytic degradation of the pesticide diuron. Res Chem Intermed 25:667 Naessens M, Tran-Minh C (1999) Biosensor using immobilized Chlorella microalgae for determination of volatile organic compounds. Sensors Actuators B Chem 59:100–102 Naessens M, Leclerc JC, Tran-Minh C (2000) Fiber optic biosensor using Chlorella vulgaris for determination of toxic compounds. Ecotoxicol Environ Saf 46:181–185 Neamtu M, Yediler A, Siminiceanu I, Macoveanu M, Kettrup A (2004) Decolorization of disperse red 354 azo dye in water by several oxidation processes – a comparative study. Dyes Pigments 60:61–68 Nguyen-Ngoc H, Tran-Minh C (2007) Fluorescent biosensor using whole cells in an inorganic translucent matrix. Anal Chim Acta 583:161–165 Niazi JH, Kim BC, Gu MB (2007) Characterization of superoxide-stress sensing recombinant Escherichia coli constructed using promoters for genes zwf and fpr fused to lux operon. Appl Microbiol Biotechnol 74:1276–1283 Niazi JH, Kim BC, Ahn J-M, Gu MB (2008) A novel bioluminescent bacterial biosensor using the highly specific oxidative stress-inducible pgi gene. Biosens Bioelectron 24:670–675 Nohava M, Vogel WR, Gaugitsch H (1995) Evaluation of the luminescent bacteria bioassay for the estimation of the toxicological potential of effluent water samples – comparison with data from chemical analyses. Environ Int 21:33–37 Nuzzo A, Hosseinkhani B, Boon N, Zanaroli G, Fava F (2017) Impact of bio-palladium nanoparticles (bio-Pd NPs) on the activity and structure of a marine microbial community. Environ Pollut 220:1068–1078 Ocampo-Duque W, Sierra J, Ferré-Huguet N, Schuhmacher M, Domingo JL (2008) Estimating the environmental impact of micro-pollutants in the low Ebro River (Spain): an approach based on screening toxicity with Vibrio fischeri. Chemosphere 72:715–721 Okamoto OK, Shao L, Hastings JW, Colepicolo P (1999) Acute and chronic effects of toxic metals on viability, encystment and bioluminescence in the dinoflagellate Gonyaulax polyedra. Comp Biochem Physiol C Pharmacol Toxicol Endocrinol 123:75–83 Okonokhua BO (2014) Bioavailability of Cu in freshly spiked, leached and field-contaminated soils. Soil Environ 33:79 Oleszczuk P, Kołtowski M (2017) Effect of co-application of nano-zero valent iron and biochar on the total and freely dissolved polycyclic aromatic hydrocarbons removal and toxicity of contaminated soils. Chemosphere 168:1467–1476

820

J. Hurtado-Gallego et al.

Oliveira AG, Desjardin DE, Perry BA, Stevani CV (2012) Evidence that a single bioluminescent system is shared by all known bioluminescent fungal lineages. Photochem Photobiol Sci 11:848–852 Ore S, Mertens J, Brandt KK, Smolders E (2010) Copper toxicity to bioluminescent Nitrosomonas europaea in soil is explained by the free metal ion activity in pore water. Environ Sci Technol 44:9201–9206 Osano O, Admiraal W, Klamer H, Pastor D, Bleeker E (2002) Comparative toxic and genotoxic effects of chloroacetanilides, formamidines and their degradation products on Vibrio fischeri and Chironomus riparius. Environ Pollut 119:195–202 Pandard P, Vasseur P, Rawson DM (1993) Comparison of two types of sensors using eukaryotic algae to monitor pollution of aquatic systems. Water Res 27:427–431 Papadopoulou D, Samara C (2002) Polycyclic aromatic hydrocarbon contamination and LUMIStox ® solvent extract toxicity of marine sediments in the North Aegean Sea, Greece. Environ Toxicol Int J 17:556–566 Papageorgiou GC, Tsimilli-Michael M, Stamatakis K (2007) The fast and slow kinetics of chlorophyll a fluorescence induction in plants, algae and cyanobacteria: a viewpoint. Photosynth Res 94:275–290 Park HJ, Hwang KO, Kim ES (2002) Construction and characterization of a recombinant bioluminescence streptomycetes for potential environmental monitoring. J Microbiol Biotechnol 12:706–709 Paton G, Campbell C, Glover LA, Killham K (1995a) Assessment of bioavailability of heavy metals using lux modified constructs of Pseudomonas fluorescens. Lett Appl Microbiol 20:52–56 Paton G, Palmer G, Kindness A, Campbell C, Glover L, Killham K (1995b) Use of luminescencemarked bacteria to assess copper bioavailability in malt whisky distillery effluent. Chemosphere 31:3217–3224 Paton G, Palmer G, Burton M, Rattray E, McGrath S, Glover L, Killham K (1997) Development of an acute and chronic ecotoxicity assay using lux-marked Rhizobium leguminosarum biovar trifolii. Lett Appl Microbiol 24:296–300 Paton GI, Viventsova E, Kumpene J, Wilson MJ, Weitz HJ, Dawson JJ (2006) An ecotoxicity assessment of contaminated forest soils from the Kola Peninsula. Sci Total Environ 355:106–117 Peña-Vázquez E, Maneiro E, Pérez-Conde C, Moreno-Bondi MC, Costas E (2009) Microalgae fiber optic biosensors for herbicide monitoring using sol–gel technology. Biosens Bioelectron 24:3538–3543 Pereira R et al (2011) Screening evaluation of the ecotoxicity and genotoxicity of soils contaminated with organic and inorganic nanoparticles: the role of ageing. J Hazard Mater 194:345–354 Pérez S, la Farré M, Garcıa MJ, Barceló D (2001) Occurrence of polycyclic aromatic hydrocarbons in sewage sludge and their contribution to its toxicity in the ToxAlert ® 100 bioassay. Chemosphere 45:705–712 Perullini M, Ferro Y, Durrieu C, Jobbágy M, Bilmes SA (2014) Sol–gel silica platforms for microalgae-based optical biosensors. J Biotechnol 179:65–70 Podola B, Melkonian M (2005) Selective real-time herbicide monitoring by an array chip biosensor employing diverse microalgae. J Appl Phycol 17:261–271 Podola B, Nowack EC, Melkonian M (2004) The use of multiple-strain algal sensor chips for the detection and identification of volatile organic compounds. Biosens Bioelectron 19:1253–1260 Pretti C et al (2014) Ecotoxicity of pristine graphene to marine organisms. Ecotoxicol Environ Saf 101:138–145 Prosser J, Anderson C, Horswell J, Speir T (2014) Can manuka (Leptospermum scoparium) antimicrobial properties be utilised in the remediation of pathogen contaminated land? Soil Biol Biochem 75:167–174 Puerari RC et al (2016) Synthesis, characterization and toxicological evaluation of Cr2O3 nanoparticles using Daphnia magna and Aliivibrio fischeri. Ecotoxicol Environ Saf 128:36–43

32

Luminescent Microbial Bioassays and Microalgal Biosensors as Tools. . .

821

Pulido-Reyes G, Martín E, Coronado JG, Leganes F, Rosal R, Fernández-Piñas F (2017) Physicochemical and biological interactions between cerium oxide nanoparticles and a 1, 8naphthalimide derivative. J Photochem Photobiol B Biol 172:61–69 Rattray E, Prosser JI, Killham K, Glover LA (1990) Luminescence-based nonextractive technique for in situ detection of Escherichia coli in soil. Appl Environ Microbiol 56:3368–3374 Rawson DM, Willmer AJ, Cardosi MF (1987) The development of whole cell biosensors for on-line screening of herbicide pollution of surface waters. Toxicity Assess 2:325–340 Ren S, Frymier PD (2003) Use of multidimensional scaling in the selection of wastewater toxicity test battery components. Water Res 37:1655–1661 Ren S, Frymier PD (2005) Toxicity of metals and organic chemicals evaluated with bioluminescence assays. Chemosphere 58:543–550 Ren S, Mee RW, Frymier PD (2004) Using factorial experiments to study the toxicity of metal mixtures. Ecotoxicol Environ Saf 59:38–43 Ríos F, Fernández-Arteaga A, Fernández-Serrano M, Jurado E, Lechuga M (2018) Silica micro-and nanoparticles reduce the toxicity of surfactant solutions. J Hazard Mater 353:436–443 Robbens J, Dardenne F, Devriese L, De Coen W, Blust R (2010) Escherichia coli as a bioreporter in ecotoxicology. Appl Microbiol Biotechnol 88:1007–1025 Rodea-Palomares I, Fernández-Piñas F, González-García C, Leganés F (2009a) Use of lux-marked cyanobacterial bioreporters for assessment of individual and combined toxicities of metals in aqueous samples. In: Handbook on cyanobacteria: biochemistry, biotechnology and applications. Nova Science Publishers, New York, pp 283–304 Rodea-Palomares I, González-García C, Leganés F, Fernández-Piñas F (2009b) Effect of pH, EDTA, and anions on heavy metal toxicity toward a bioluminescent cyanobacterial bioreporter. Arch Environ Contam Toxicol 57:477 Rodea-Palomares I, Boltes K, Fernández-Pinas F, Leganés F, García-Calvo E, Santiago J, Rosal R (2010) Physicochemical characterization and ecotoxicological assessment of CeO2 nanoparticles using two aquatic microorganisms. Toxicol Sci 119:135–145 Rodea-Palomares I, Gonzalo S, Santiago-Morales J, Leganés F, García-Calvo E, Rosal R, Fernández-Pinas F (2012) An insight into the mechanisms of nanoceria toxicity in aquatic photosynthetic organisms. Aquat Toxicol 122:133–143 Rodea-Palomares I et al (2016) Hidden drivers of low-dose pharmaceutical pollutant mixtures revealed by the novel GSA-QHTS screening method. Sci Adv 2:e1601272 Rodriguez M Jr, Sanders CA, Greenbaum E (2002) Biosensors for rapid monitoring of primarysource drinking water using naturally occurring photosynthesis. Biosens Bioelectron 17:843–849 Rosal R, Rodea-Palomares I, Boltes K, Fernández-Piñas F, Leganés F, Gonzalo S, Petre A (2010a) Ecotoxicity assessment of lipid regulators in water and biologically treated wastewater using three aquatic organisms. Environ Sci Pollut Res 17:135–144 Rosal R, Rodea-Palomares I, Boltes K, Fernández-Piñas F, Leganés F, Petre A (2010b) Ecotoxicological assessment of surfactants in the aquatic environment: combined toxicity of docusate sodium with chlorinated pollutants. Chemosphere 81:288–293 Rosen G, Osorio-Robayo A, Rivera-Duarte I, Lapota D (2008) Comparison of bioluminescent dinoflagellate (QwikLite) and bacterial (Microtox) rapid bioassays for the detection of metal and ammonia toxicity. Arch Environ Contam Toxicol 54:606–611 Rouillona R, Tocabens M, Carpentier R (1999) A photoelectrochemical cell for detecting pollutantinduced effects on the activity of immobilized cyanobacterium Synechococcus sp. PCC 7942. Enzym Microb Technol 25:230–235 Saddoud A, Abdelkafi S, Sayadi S (2009) Effects of domestic wastewater toxicity on anaerobic membrane-bioreactor (MBR) performances. Environ Technol 30:1361–1369 Sajayan A, Kiran GS, Priyadharshini S, Poulose N, Selvin J (2017) Revealing the ability of a novel polysaccharide bioflocculant in bioremediation of heavy metals sensed in a vibrio bioluminescence reporter assay. Environ Pollut 228:118–127

822

J. Hurtado-Gallego et al.

Salizzato M, Bertato V, Pavoni B, Ghirardini AV, Ghetti PF (1998a) Sensitivity limits and EC50 values of the Vibrio fischeri test for organic micropollutants in natural and spiked extracts from sediments. Environ Toxicol Chem 17:655–661 Salizzato M, Pavoni B, Ghirardini AV, Ghetti P (1998b) Sediment toxicity measured using Vibrio fischeri as related to the concentrations of organic (PCBs, PAHs) and inorganic (metals, sulphur) pollutants. Chemosphere 36:2949–2968 Sanchez-Ferandin S (2015) Assessing ecotoxicity in marine environment using luminescent microalgae: where are we at? Am J Plant Sci 6:2502 Sanchez-Ferandin S, Leroy F, Bouget F-Y, Joux F (2012) A new, sensitive marine microalgal recombinant biosensor using luminescence monitoring for the toxicity testing of antifouling biocides. Appl Environ Microbiol. https://doi.org/10.1128/AEM.02688-12 Sanchís J, Olmos M, Vincent P, Farré M, Barceló D (2015) New insights on the influence of organic co-contaminants on the aquatic toxicology of carbon nanomaterials. Environ Sci Technol 50:961–969 Sanders CA, Rodriguez M Jr, Greenbaum E (2001) Stand-off tissue-based biosensors for the detection of chemical warfare agents using photosynthetic fluorescence induction. Biosens Bioelectron 16:439–446 Schmitz RP, Kretkowski C, Eisenträger A, Dott W (1999) Ecotoxicological testing with new kinetic Photorhabdus luminescens growth and luminescence inhibition assays in microtitration scale. Chemosphere 38:67–78 Scholz B, Liebezeit G (2012) Screening for biological activities and toxicological effects of 63 phytoplankton species isolated from freshwater, marine and brackish water habitats. Harmful Algae 20:58–70 Scholz B, Ziehe D, Pivetta LA, Pielok N, Liebezeit G (2012) Toxicity of East Sumatra River sediments–bacterial luminescence, brine shrimp and acetylcholinesterase inhibition tests. Asian J Water Environ Pollut 9:1–9 Scognamiglio V et al (2009) Chlamydomonas reinhardtii genetic variants as probes for fluorescence sensing system in detection of pollutants. Anal Bioanal Chem 394:1081 Segura PA, Kaplan P, Erythropel HC, Yargeau V (2012) Comparative rapid toxicity screening of commercial and potential “green” plasticizers using bioluminescent bacteria. Ind Eng Chem Res 51:11555–11560 Serat WF, Budinger FE Jr, Mueller PK (1965) Evaluation of biological effects of air pollutants by use of luminescent bacteria. J Bacteriol 90:832 Serat WF, Budinger FE Jr, Mueller P (1967) Toxicity evaluation of air pollutants by use of luminescent bacteria. Atmos Environ (1967) 1:21–32 Shao C, Howe C, Porter AJR, Glover LA (2002) Novel cyanobacterial biosensor for detection of herbicides. Appl Environ Microbiol 68:5026–5033 Shaw L, Beaton Y, Glover L, Killham K, Meharg A (1999) Development and characterization of a lux-modified 2, 4-dichlorophenol-degrading Burkholderia sp. RASC. Environ Microbiol 1:393–399 Shitanda I, Takada K, Sakai Y, Tatsuma T (2005) Amperometric biosensing systems based on motility and gravitaxis of flagellate algae for aquatic risk assessment. Anal Chem 77:6715–6718 Somensi CA, Simionatto EL, Bertoli SL, Wisniewski A Jr, Radetski CM (2010) Use of ozone in a pilot-scale plant for textile wastewater pre-treatment: physico-chemical efficiency, degradation by-products identification and environmental toxicity of treated wastewater. J Hazard Mater 175:235–240 Stauber JL, Binet MT, Bao VW, Boge J, Zhang AQ, Leung KM, Adams MS (2008) Comparison of the Qwiklite™ algal bioluminescence test with marine algal growth rate inhibition bioassays. Environ Toxicol Int J 23:617–625 Steinberg SM, Poziomek EJ, Engelmann WH, Rogers KR (1995) A review of environmental applications of bioluminescence measurements. Chemosphere 30:2155–2197 Stone V, Donaldson K (2006) Nanotoxicology: signs of stress. Nat Nanotechnol 1:23

32

Luminescent Microbial Bioassays and Microalgal Biosensors as Tools. . .

823

Strigul N, Vaccari L, Galdun C, Wazne M, Liu X, Christodoulatos C, Jasinkiewicz K (2009) Acute toxicity of boron, titanium dioxide, and aluminum nanoparticles to Daphnia magna and Vibrio fischeri. Desalination 248:771–782 Svartz G et al (2017) Monitoring the ecotoxicity of γ-Al2O3 and Ni/γ-Al2O3 nanomaterials by means of a battery of bioassays. Ecotoxicol Environ Saf 144:200–207 Svenningsen NB, Pérez-Pantoja D, Nikel PI, Nicolaisen MH, de Lorenzo V, Nybroe O (2015) Pseudomonas putida mt-2 tolerates reactive oxygen species generated during matric stress by inducing a major oxidative defense response. BMC Microbiol 15:202 Tahirbegi IB et al (2017) Fast pesticide detection inside microfluidic device with integrated optical pH, oxygen sensors and algal fluorescence. Biosens Bioelectron 88:188–195 Tchounwou PB, Yedjou CG, Patlolla AK, Sutton DJ (2012) Heavy metal toxicity and the environment. In: Molecular, clinical and environmental toxicology. Springer, Basel, pp 133–164 Thomulka KW, Schroeder JA, Lange JH (1997) Use of Vibrio harveyi in an aquatic bioluminescent toxicity test to assess the effects of metal toxicity: treatment of sand and water–buffer, with and without EDTA. Environ Toxicol Water Qual Int J 12:343–348 Tom-Petersen A, Hansen HCB, Nybroe O (2004) Time and moisture effects on total and bioavailable copper in soil water extracts. J Environ Qual 33:505–512 Tsiridis V, Petala M, Koukiotis C, Darakas E (2017) Implications of handling practices on the ecotoxic profile of alumina nanoparticles towards the bacteria Vibrio fischeri. J Environ Sci Health A 52:15–22 Tsopela A et al (2014) Integrated electrochemical biosensor based on algal metabolism for water toxicity analysis. Biosens Bioelectron 61:290–297 Tsopela A et al (2016) Development of a lab-on-chip electrochemical biosensor for water quality analysis based on microalgal photosynthesis. Biosens Bioelectron 79:568–573 Turner NL, Horsburgh A, Paton GI, Killham K, Meharg A, Primrose S, Strachan NJ (2001) A novel toxicity fingerprinting method for pollutant identification with lux-marked biosensors. Environ Toxicol Chem 20:2456–2461 Valavanidis A, Vlahogianni T, Dassenakis M, Scoullos M (2006) Molecular biomarkers of oxidative stress in aquatic organisms in relation to toxic environmental pollutants. Ecotoxicol Environ Saf 64:178–189 Védrine C, Leclerc J-C, Durrieu C, Tran-Minh C (2003) Optical whole-cell biosensor using Chlorella vulgaris designed for monitoring herbicides. Biosens Bioelectron 18:457–463 Velzeboer I, Hendriks AJ, Ragas AM, Van de Meent D (2008) Nanomaterials in the environment aquatic ecotoxicity tests of some nanomaterials. Environ Toxicol Chem 27:1942–1947 Verschaeve L, Van Gompel J, Thilemans L, Regniers L, Vanparys P, Van der Lelie D (1999) VITOTOX ® bacterial genotoxicity and toxicity test for the rapid screening of chemicals. Environ Mol Mutagen 33:240–248 Vydryakova G, Gusev A, Medvedeva S (2011) Effect of organic and inorganic toxic compounds on luminescence of luminous fungi. Appl Biochem Microbiol 47:293–297 Wang C, Yediler A, Lienert D, Wang Z, Kettrup A (2002) Toxicity evaluation of reactive dyestuffs, auxiliaries and selected effluents in textile finishing industry to luminescent bacteria Vibrio fischeri. Chemosphere 46:339–344 Wang D, Zhao L, Ma H, Zhang H, Guo L-H (2017) Quantitative analysis of reactive oxygen species photogenerated on metal oxide nanoparticles and their bacteria toxicity: the role of superoxide radicals. Environ Sci Technol 51:10137–10145 Weitz HJ, Ritchie JM, Bailey DA, Horsburgh AM, Killham K, Glover LA (2001) Construction of a modified mini-Tn 5 lux CDABE transposon for the development of bacterial biosensors for ecotoxicity testing. FEMS Microbiol Lett 197:159–165 Weitz HJ, Campbell CD, Killham K (2002) Development of a novel, bioluminescence-based, fungal bioassay for toxicity testing. Environ Microbiol 4:422–429 Winson MK et al (1998) Engineering the luxCDABE genes from Photorhabdus luminescens to provide a bioluminescent reporter for constitutive and promoter probe plasmids and mini-Tn 5 constructs. FEMS Microbiol Lett 163:193–202

824

J. Hurtado-Gallego et al.

Wolska L, Kuczyńska A, Namieśnik J (2006) Quality of groundwater contaminated by leachates from seven Polish landfills–chemical and ecotoxicological classifications. Toxicol Environ Chem 88:501–513 Wu S, Zhang X, Yang P, Li L, Tang S (2018) Rapid detection and toxicity assessment of citreoviridin using luminescent Vibrio qinghaiensis sp.-Q67 in drinking water. Int J Food Sci Technol 53:2141 Yu H, Nie E, Xu J, Yan S, Cooper WJ, Song W (2013) Degradation of diclofenac by advanced oxidation and reduction processes: kinetic studies, degradation pathways and toxicity assessments. Water Res 47:1909–1918 Zabetoglou K, Voutsa D, Samara C (2002) Toxicity and heavy metal contamination of surficial sediments from the Bay of Thessaloniki (Northwestern Aegean Sea) Greece. Chemosphere 49:17–26 Zaki S, Abd-El-Haleem D, Abulhamd A, Elbery H, AbuElreesh G (2008) Influence of phenolics on the sensitivity of free and immobilized bioluminescent Acinetobacter bacterium. Microbiol Res 163:277–285 Zarubina A, Lukashev E, Deev L, Parkhomenko I, Rubin A (2009) Biotesting the biological effects of single-wall carbon nanotubes using bioluminescent bacteria test-system. Nanotechnol Russia 4:871–875 Zhang J, Chen Q (2014) Predicting the long-term toxicity of five-antibiotic mixtures to Vibrio qinghaiensis sp. Q67. Pak J Pharm Sci 27:1605–1609 Zhao JL, Jiang YX, Yan B, Wei C, Zhang LJ, Ying GG (2014) Multispecies acute toxicity evaluation of wastewaters from different treatment stages in a coking wastewater-treatment plant. Environ Toxicol Chem 33:1967–1975

Detection and Effects of Metal and Organometallic Compounds with Microbial Bioluminescence and Raman Spectroscopy

33

Sulivan Jouanneau, Ali Assaf, Marie-Jose´ Durand, and Ge´rald Thouand

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Metals and Organometallic Compounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Toxic Effects and Regulations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Measurement by Microbial Bioluminescence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Toxicity Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Bioavailable/Bioaccessible Fraction Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Field Application: Biosensor Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Advantages, Limits, and Perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Evaluation of Toxicity by Raman Spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Principle of Raman Spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Non-targeted Approach for Measuring the Toxicity of Heavy Metals by Raman Spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Advantages, Limits, and Perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2 2 2 4 6 6 8 10 12 14 14 15 20 22

Abstract

Metals and organometallic compounds are well known for their toxic effects on the biosphere. Despite these toxic effects, organometallic compounds are omnipresent in the various environmental compartments (water, air, soil). In this context, it is crucial to be able to detect these environmental contaminants and to assess their effects so that the environment and the public health can be protected. Physicochemical methods are particularly relevant for evaluating the overall fraction in the environment. However, this view is insufficient to determine the potential effects of pollution on the biosphere. In this chapter, we focus on two metrological approaches based on microbial bioluminescence and Raman S. Jouanneau · A. Assaf (*) · M.-J. Durand · G. Thouand CNRS, GEPEA, UMR 6144, Université de Nantes, La Roche-sur-Yon, France e-mail: [email protected] © Springer Nature Switzerland AG 2022 G. Thouand (ed.), Handbook of Cell Biosensors, https://doi.org/10.1007/978-3-030-23217-7_90

825

826

S. Jouanneau et al.

spectroscopy to characterize pollutants in different ways, namely the toxicological effects induced by pollutants and their bioavailability/bioaccessibility in environmental matrices.

Introduction In this chapter, we will focus on the detection of metals (family of compounds covering notably the heavy metals and metalloids), as well as on the detection of organometallic compounds. The latter are defined as compounds containing at least one chemical bond between a carbon atom of an organic molecule and a metallic atom (e.g., tributyltin – Linear formula: (CH3CH2CH2CH2)3Sn-Sn (CH2CH2CH2CH3)3). The concentrations of these compounds in the different environmental compartments are functions of anthropic supplies, including industrial activities, transport, etc., or are from natural sources, including volcanism, the geological nature of the soils, etc. Some metals and organometallic compounds are necessary for life, such as iron (essential component of hemoglobin), zinc, copper, and selenium, which are essential trace elements but are generally most known for their harmful properties, such as their bioaccumulative capabilities and/or their short- and long-term toxicity on living organisms. The initial step is to be able to detect and assess the effects of these metals and organometallic compounds to limit their consequences on health and prevent their harmful impacts on the environment. Among the wide spectra of available analytical methods (physicochemical methods (US EPA 2019) or biological methods (Jouanneau and Thouand 2019; Aloisi et al. 2019)), we will only focus on microbial bioluminescence-based methods and Raman spectroscopy as means of detecting metals and organometallic compounds in the environment and to understand their effects on the living.

Metals and Organometallic Compounds Definition As expressed in the technical report written by J.H. Duffus (2002), the term “metal” is commonly used in the literature to describe several groups of compounds. The term “metal” is technically defined as elements “which conduct electricity, have a metallic lustre, are malleable and ductile, form cations, and have basic oxides” (Atkins 2010). However, in conventional usage, this term refers to a pure element or an alloy of several pure elements. Among the most recurrent groups in the literature, we can cite “heavy metals,” “metalloids,” “metallic element traces” (MTE), or “essential metals” (Fig. 1). These terms refer to specific properties (physical or chemical properties, effects on living organisms) of the described elements (Table 1).

33

Detection and Effects of Metal and Organometallic Compounds with Microbial. . .

827

Fig. 1 Periodic table of metals, metalloids, and nonmetals (Duffus 2002)

The used term is often linked to a dose notion. Indeed, the same element can be considered by authors as an essential metal or a trace element in a context where its absence is responsible for harmful effects on life (Zoroddu et al. 2019). In another context, this same metal at too high concentrations can induce adverse consequences on the environment (Dameron and International Programme on Chemical Safety 1998) and/or health (Jaiser and Winston 2010). For example, copper deficiency can cause health problems, such as artery weakness, liver disorder, and secondary anemia, but an excess of copper can induce liver injuries. Organometallic compounds are molecules that contain at least one covalent bond between a carbon atom and a metal. The carbon atom can be from an inorganic origin, such as carbon monoxide. In this case, the carbon atom forms a molecule with the metal named “metal carbonyl.” However, most often, the metal element is bonded to a hydrocarbon group, such as methyl, propyl, or butyl groups. These compounds are generally very toxic by comparison with the pure metal that they contain. Due to this property, organometallic compounds have been widely used as biocides (pesticides or antifouling molecules) (Antizar-Ladislao 2008) as well as fuel additives (tetraethyllead) (Herrmann 2008) or in some industrial processes (Cornils and Herrmann 2002).

828

S. Jouanneau et al.

Table 1 Terms used to describe metals and subgroups Term Metal Metalloid

Heavy metal

Metallic Trace Element (MTE)

Essential metal

Nonmetal

Definition Elements that conduct electricity, have a metallic luster, are malleable and ductile, form cations, and have basic oxides Elements that possesses properties of metals and nonmetals. There is no standard definition of this term; nevertheless boron, silicon, germanium, arsenic, antimony, and tellurium are commonly recognized as metalloids This term is, according to authors, used to describe an element with high density (> 5000 kg.m
3) or high atomic weight or is often used with the connotation of toxicity on living organisms. Consequently, some authors recommend removing this term because of its imprecision and to replace it with simply “metal” (or “metalloid” in the case of B, Si, Ge, As, Sb, and Te). In 2000 in the European legislation (applied to all member states), the European community defined a “heavy metal” (undertone “toxic element”) as “any compound of antimony, arsenic, cadmium, chromium(VI), copper, lead, mercury, nickel, selenium, tellurium, thallium and tin, as well as these materials in metallic form, as far as these are classified as dangerous substances” Metallic element found in low concentration in some specified matrices, including soil, plant, water, tissue, etc. According to authors, this term can be used to replace “heavy metal” in publications as a way to describe metallic elements with inherent toxic properties; in other publications, it can be used to depict “essential metals” Elements required for the life cycle and the lack of which can induce harmful effects on organisms (specific deficiency symptoms relieved only by these metals). Nevertheless, it is relevant to mention that too high concentrations can also induce harmful effects. Among these elements, we can cite manganese, iron, copper, nickel, zinc, etc. Elements that usually have poor thermal and electrical conductivities. Metalloids are sometimes counted among nonmetals

References (Atkins 2010) (Atkins 2010)

(Duffus 2002)

(Duffus 2002)

(Duffus 2002)

(Cox 2004)

Toxic Effects and Regulations Metals are naturally present in the environment, but their concentrations can be strongly increased by anthropogenic activities, such as urbanization, industrial activity (Qiao et al. 2013; Christophoridis et al. 2009), or transport. Organometallic compounds may also come from natural sources, such as biological mechanisms of mercury biomethylation by microorganisms, or from anthropic origins, such as when tributyltin is used in some antifouling paints for boats. Metals and organometallic compounds are particularly known for their toxic effects on human health or the environment. Many pathologies are attributed

33

Detection and Effects of Metal and Organometallic Compounds with Microbial. . .

829

Table 2 Regulatory values and detection limits by physicochemical methods WFD

Minimum current detection limit 0.01 μg.L
1(AAS)

Compound Antimony (Sb) Arsenic (As) Cadmium (Cd) Chromium (Cr) Copper (Cu) Iron (Fe)

WHO 20 μg.L
1

USEPA 6 μg.L
1

ECE 5 μg.L
1

10 μg.L
1 3 μg.L
1

10 μg.L
1 5 μg.L
1

10 μg.L
1 5 μg.L
1

50 μg.L
1

100 μg.L
1

50 μg.L
1

0.05 μg.L
1 (AAS)

2 mg.L
1 –

1,3 mg.L
1 300 μg.L
1

0.02 μg.L
1 (ICP-MS)

Lead (Pb) Manganese (Mn) Mercury (Hg)

10 μg.L
1 –

15 μg.L
1 50 μg.L
1

2 mg.L
1 200 μg. L
1 10 μg.L
1 50 μg.L
1

6 μg.L
1

2 μg.L
1

Nickel (Ni) Silver (Ag) Thallium (Tl) Uranium (U) Zinc (Zn) Tributyltin

70 μg.L
1 – – 30 μg.L
1 –

– 100 μg.L
1 2 μg.L
1 30 μg.L
1 5 mg.L
1

X

0.1 μg.L
1 (ICP-MS) 0.01 μg.L
1 (ICP-MS)

X

1 μg.L
1 (AAS)

1 μg.L
1

X

20 μg.L
1 – – – –

X

0.05 μg.L
1 (cold vapor AAS) 0.1 μg.L
1 (ICP-MS)

0.01 μg.L
1 (ICP-MS) X

1 ng.L
1 (GC)

WHO, World Health Organization; USEPA, United States Environmental Protection Agency; ECE, European Commission Environment; WFD, Water Framework Directive; X, Concerned by the WFD; AAS, Atomic Absorption Spectrometry; ICP-MS, Inductively Coupled Plasma Mass Spectrometry; GC, Gas Chromatography

(specifically or as cofactors) to acute or chronic exposures to these compounds by the living (Fernández-Luqueño et al. 2013). Due to their different toxic properties (carcinogenic, neurotoxic, endocrine disrupting, etc.), some metals and organometallic compounds were classified as priority substances by the European Union (2013/39/EU). This regulation concerns lead, cadmium, nickel, mercury, and tributyltin, which must be detected in water as a priority. To this, the more restrictive drinking water regulations must be added, setting the concentration limits of these metals and organometallic compounds in potable water (Fernández-Luqueño et al. 2013) (Table 2). The detection of these compounds in different environmental compartments (water, soil, air) is, by consequence, a major issue for the scientific community around the world. Physicochemical methods are widely used to detect and measure the contamination levels in these different matrices because of their sensitivities and their specificities, but they are not suitable to assess the potential effects of these compounds and their action modes on living organisms. To overcome this metrological deficiency, several approaches have been proposed based on biotechnologies. In this chapter, we will only focus on biotechnologies based on microbial

830

S. Jouanneau et al.

bioluminescence to detect and assess these compounds in different environmental matrices, and Raman spectroscopy will be used to understand their effects on life.

Measurement by Microbial Bioluminescence As mentioned above, physicochemical methods are particularly relevant to assess the concentration of metals in different environmental compartments (air, water, soil). The impacts that these molecules have on the living are closely linked with their distributions within an environment (bioavailable and/or bioaccessible fractions). As this parameter is particularly complex to assess by physicochemical methods, other metrological approaches have been developed. Among them, methods based on bioluminescent microorganisms have been widely proposed (Fig. 2) for the past three decades. Bibliometric analysis (Fig. 2c) of data from the Scopus library highlights three main topics relative to (i) toxicity assessment, (ii) detection, and (iii) biosensors. Assessments of organometallic compounds do not appear in this figure because of their lack of representation in the literature. The first topic (in blue) focuses on the assessment of the toxic effects of metals on life. The authors were notably interested in the intensity (acute toxicity) and the nature of these effects (genotoxicity, reprotoxicity, etc.). The second topic (in red) addresses the detection of compounds by bioluminescent microorganisms. For this process, the used cells were genetically modified to be able to produce light, specifically in the presence of some metals or organometallic compounds. The last axis (in green) has a strong technological aspect through the development of biosensors for the detection of these compounds or the assessment of their effects in the environment.

Toxicity Assessment Biological approaches are able to provide another point of view for the characterization of metals and organometallic compounds; however, first, their effects on the living (not accessible to the physiochemical method) need to be evaluated. For that purpose, the use of bioluminescent microorganisms is relatively widespread.

Overall Effect The most known approach makes use of the naturally bioluminescent strain Aliivibrio fischeri to assess the acute toxicity of chemicals, such as metals and organometallic compounds, and has been in use since 1982 (Bulich 1982). However, a similar concept was proposed in 1965 by Serat et al. dedicated to air pollutants and based on another bioluminescent strain (Serat et al. 1965). When bioluminescent cells are exposed to a toxic sample, their luminescence is reduced, allowing a rapid and direct toxicity assessment. This strategy has been the subject of several international standards (ISO 11348-1; ISO 11348-2; ISO 11348-3), and use of commercial devices for this strategy has been declining (Microtox, Modern Water, UK;

33

Detection and Effects of Metal and Organometallic Compounds with Microbial. . .

831

Fig. 2 Evolution of the publication number over the three last decades: (a) contribution of authors (b) and main topics – VOSviewer 1.6.14 (c)

Lumistox, Hach Lange, Deutschland; BioTox, Aboatox, Finland; ToxMini/ iToxControl, MicroLan, Netherlands; Insitox, Tame Water, France; etc.). On the basis of this concept, other measurement methods were proposed using other bioluminescent strains. The NCIMB Company (UK) proposed an approach (LumiMara) based on 11 strains (9 marine and 2 freshwater species) of naturally occurring bioluminescent bacteria. This multispecies strategy aims to increase the

832

S. Jouanneau et al.

representativity of information provided by the use of strains from a variety of habitats (genetic diversity). In addition, some authors are interested in studying toxic interactions between metals via the bioluminescence inhibition of two genetically modified bacterial strains (Escherichia coli and Pseudomonas fluorescens). This work shows the synergistic or antagonistic interactions between some metals (Preston et al. 2000), inducing toxic effects that are difficult to predict.

Specific Effects Limiting the toxicity of metals and organometallic compounds to their acute effect is very reductionist. Additionally, it is necessary to assess their different medium- and long-term effects. Microbial bioluminescence is used for that purpose. In a work published in 1988 (Ulitzur and Barak 1988), the authors proposed a strategy based on a mutant strain (a dark mutant of A. fischeri) that emitted a light signal under the action of the mutagenic effect. Through this work, the authors show activation of bioluminescence induced by metals while simultaneously demonstrating the intrinsic mutagenic properties of some compounds (HgCl2, AsO42
Co2+, AgNO3, KH2AsO4, etc.). To improve the knowledge of these compounds, some authors used genetic engineering to develop biological probes based on the fusion between some stress-responsive promoters with the reporter genes responsible for bioluminescence (firefly or bacterial luciferases) (Lewis et al. 1998; Meighen 1993; Daunert et al. 2000; Hakkila et al. 2002) to more precisely determine the nature of the induced effects (Ben-Israel et al. 1998).

Bioavailable/Bioaccessible Fraction Assessment Metals are ubiquitous in environmental matrices; however, their presence does not always induce toxic effects on the biosphere, in particular, due to their availability (adsorption on particulate matter, chemical complexation or precipitation, etc.) to an organism, that, bioavailability/bioaccessible (Semple et al. 2004) (Fig. 3). The use of bioindicators has become a relevant strategy to determine the bioavailable or bioaccessible fraction in a matrix, in complement with the overall fraction (determined by physicochemical methods). The first strategy was described in 1991 by Guzzo et al. (1991) and applied to aluminum detection. The strain used was obtained by a random insertion of promoterless luxAB reporter genes (Vibrio harveyi) in the E. coli chromosome, which induced a bioluminescence emission when in contact with aluminum. To obtain this result, the authors developed a library of approximately 3000 E. coli clones (only one was identified as able to detect aluminum). This strategy has proven to be particularly time and energy consuming. In 1993, Selifonova et al. (1993) and Corbisier et al. (1993) proposed similar strategies applied to bioavailable metal detection. In the publication of Selifonova et al., the mercury resistance operon was cloned upstream from the lux reporter genes (A. fischeri) in the E. coli host

33

Detection and Effects of Metal and Organometallic Compounds with Microbial. . .

833

Fig. 3 Bioavailability and bioaccessibility of metals in an environmental matrix (Semple et al. 2004)

strain, thus inducing bioluminescence in contact with mercury. The detection model has since been widely taken up by other teams to detect others metals in different environmental compartments (Stocker et al. 2003; Ivask et al. 2009; Magrisso et al. 2008; Gu et al. 2004; Verma and Singh 2005; Köhler et al. 2000; Bereza-Malcolm et al. 2015). In the case of organometallic compounds, the recognition strategy was less obvious (due to the lack of a known genetic resistance mechanism). The first microbial strain that was able to specifically detect organometallic compounds, that is, tributyltin, was described by Briscoe et al. in 1996 (Briscoe et al. 1996), and it was obtained by the random insertion of lux genes (luxAB of V. harveyi) into the E. coli chromosome (library of 3000 clones). The understanding of the genetic recognition mechanism has involved supplementary works, led by Gueuné et al. (2008) In some cases, genetic constructs designed to detect metals are also activated by organometallic compounds, such as methyl mercury (Ivask et al. 2009) (the promoter is described as being “specific” for mercury). Nevertheless, the example number dedicated to these chemicals remains relatively limited (Ivask et al. 2009; Gueune et al. 2009; Kabiersch et al. 2013).

834

S. Jouanneau et al.

Field Application: Biosensor Development To deploy these measurement methods outside of the laboratory (in the field or in industrial environments), some authors have focused on the integration of the biological detection method in dedicated measurement platforms named “biosensors.” A biosensor is defined by IUPAC as “a device that uses specific biochemical reactions mediated by isolated enzymes, immunosystems, tissues, organelles or whole cells to detect chemical compounds usually by electrical, thermal or optical signals.” (Nagel et al. 1992) Contrary to bioassays (laboratory methods), biosensors incorporate direct spatial contact between the biological recognition element and the transducer (Thevenot et al. 2001) and often integrate an automatic dimension of the measurement protocol. Here, the biosensors consist of a biological recognition element (bioluminescent strain) and a physical device (photosensitive transducer) that is able to convert the analogic signal of bioluminescence into a numerical signal. The first described biosensor using a bioluminescence microbial was proposed by Serat et al. in 1967 (Serat et al. 1967) based on Photobacterium phosphoreum (natural bioluminescent bacterial strain). In this device, bioluminescent cells are immobilized on a solid hydrogel (agar) and exposed to air pollutants. The toxicity of air contaminants is assessed from light inhibition. Despite this publication (which went relatively unnoticed), it was only in the 1990s that biosensors underwent real development (Jouanneau et al. 2016; Gu et al. 1996, 1999; Simpson et al. 1998). The proposed devices in the 1990s were mainly based on the bioreactor architecture (cells in a controlled liquid phase) and were dedicated to toxicity monitoring and not to the detection of metal or organometallic compounds. In 1999, Corbisier et al. took up the immobilization idea in a solid hydrogel (alginate) for the sake of facilitating the implementation of the measurement method in the field (Corbisier et al. 1999) for the detection of bioavailable metals. Five years later, Hakkila et al. (2004) proposed a multi-sensorial biosensor based on nine whole-cell bacterial sensors immobilized (alginate hydrogel) on optical fiber tips. Additionally, in 2004, Horry et al. (Horry et al. 2004) published research on another biosensor based on the bioreactor architecture (Fig. 4a) to assess the tributyltin (organometallic compound) concentration in water, which was followed in 2007 (Horry et al. 2007) by a device based on cells immobilized in a disposable card (Fig. 4b). In the following years, other devices were proposed mainly based on immobilized cells a priori that were more suitable to the reality of the field, contrary to the bioreactor architectures (Fig. 4c). The main limitation of using living cells concerns the stability of the cells over time. Indeed, cells tend to grow, thus inducing changes in their number in biological sensors (the signal intensity is directly associated with the cellular density) and in their physiological state. Moreover, exposure to environmental samples can produce harmful effects on biological sensors (Jouanneau et al. 2012; Affi et al. 2009) (contamination, competition for nutrients, bioavailable O2). To remove this strong limitation, Jouanneau et al. proposed a biosensor based on freeze-dried cells (Jouanneau et al. 2012) (Fig. 4d).

33

Detection and Effects of Metal and Organometallic Compounds with Microbial. . .

835

Biosensor proposed by Horry et al. 2004 - One strain - Detection of tributyltin - Cells in a controlled bioreactor - Transduction by a photomultiplier

a

Disposable card holder

O-ring

Optical fiber tip

Biosensor proposed by Horry et al. 2004 - One strain - Detection of tributyltin - Cells immobilized in a hydrogel matrix in a disposable card - Transduction by a photomultiplier

Heating element

Agarose matrix containing the bioluminescent bacteria Disposable card

b

Analysed sample

Bioluminescence measure by CCD camera

Removal card with immobilized bacteria

Biosensor proposed by Charrier et al. 2011 - Multi-strain - Detection of metals - Cells immobilized in a hydrogel matrix in a disposable card - Transduction by a CCD (Charge Coupled Device) sensor

Wells containing entrapped living cells

c

Peltier element (thermal regulation) Biosensor proposed by Jouanneau et al. 2012 - Multi-strain - Detection of metals - Cells freeze-dried in 96-well microplates - Transduction by a CCD sensor

d

Fig. 4 Example of biosensors developed to detect metals or organometallic compounds in waterCharrier (Charrier et al. 2011)

836

S. Jouanneau et al.

Advantages, Limits, and Perspectives As described by Bereza-Malcolm et al. (2015), the measurement methods using the microbial approach have some advantages. In the framework of metal and organometallic compound measurement, these approaches appear suitable to determine their bioavailable fraction and to assess their effects in different environmental compartments (at lower cost); nevertheless, some technical limits persist. The determination of the bioavailable fraction is based on the intensity of the light signal emitted by cells. The lack of a signal can be caused either by a concentration of metal lower than the detection limit of the biological sensor or a toxic concentration. The identification of the “true” hypothesis is primordial. For that purpose, the majority of authors use a supplementary strain as a toxicity control, that is, a non-inducible bioluminescent strain (constitutive strain). Roda et al. (2011) proposed an alternative strategy based on only one strain that has two distinct mechanisms of assessment (Fig. 5). The first mechanism is dedicated to the specific highlighting of copper, and the second mechanism allows the overall toxicity level of the tested sample to be determined. The second bottleneck concerns the specificity of inducible cells used to measure the bioavailable fraction of metals and organometallic compounds. Under favorable conditions (no nutrient or oxygen deficiencies), the bioluminescence level depends on two parameters, that is, the nature and concentration of chemicals, but, as shown though several publications (Ivask et al. 2009; Durand et al. 2003; Jouanneau et al. 2011), these biological sensors are very often induced by several chemicals (Fig. 6). To increase the specificity of these biological approaches, some authors have proposed cross-referencing the data provided by several biological descriptors using statistical tools. In 2008, Elad et al. proposed a first approach (Elad et al. 2008) based on the biological fingerprint (five biological descriptors) of the effects induced by a toxic compound. By comparing a new fingerprint (obtained from an unknown sample) to a reference bank (previously established), the authors were able to

Fig. 5 Schematic representation of the designed microbial bioreporter to allow the simultaneous assessment of the copper concentration and overall toxicity (Roda et al. 2011)

33

Detection and Effects of Metal and Organometallic Compounds with Microbial. . .

837

Fig. 6 Detected metals by several bioluminescent strains (Jouanneau et al. 2011)

classify samples with a data processing algorithm to deduce the nature of the chemical and its concentration in the analyzed samples. Although this proof of concept was not applied to metals or organometallic compounds, the use of strains specific to specific toxic effects (not specific to some compounds) are not restricted to only the tested compounds. This proof of concept was a first step toward an improvement of data processing, but the problem was limited to simple cases, that is, no mixtures of chemicals, no real samples, and a limited number of chemicals. As expressed previously in the case of specific bioluminescent cells, one strain can detect several compounds, but the reciprocal is also true, that is, one compound can be detected by several strains. It is on this second point that the strategy proposed by Jouanneau et al. (2011) was based, which notably builds on the crossings between the detection capabilities of several specific strains. For that purpose, the authors ran algorithms (decision tree, one model by compound) by supervised learning from a reference bank to regroup the bioluminescence data of five strains obtained from artificial samples until they contained four metals at four concentration levels. The approach was validated on these metals in a mixture and from complex samples, such as wastewater, industrial water, and river water, also integrating the notion of an overall toxicity assessment.

838

S. Jouanneau et al.

Evaluation of Toxicity by Raman Spectroscopy As mentioned in this chapter, the measurement of toxicity is often performed by mono-parametric methods. These highly selective methods are used mainly to obtain an accurate measurement of the concentrations of toxicants. Nevertheless, these methods focus only on a single parameter at a time or on a given toxicant without identifying its molecular target. New analytical methods have been proposed in the literature to fill this gap. Among the techniques proposed in the literature, Raman spectroscopy has been proposed for the evaluation of toxicity in different application contexts. In this part of the chapter, the main publications that propose this technique to evaluate the toxicity of heavy metals on living cells are discussed.

Principle of Raman Spectroscopy Raman spectroscopy is a vibrational technique based on the interaction between light and matter. Discovered in 1928, the Raman signal results from the inelastic scattering of light due to the interaction between the incident photons emitted by a monochromatic light source (laser) with the molecules constituting the sample. In terms of the optical efficiency, approximately 1 photon in 1000 is elastically scattered (without energy changes: Rayleigh scattering), whereas it is estimated that only 1 of 109 photons is elastically scattered (Raman signal) (Fig. 7) (Bumbrah and Sharma 2016). This low scattering efficiency has long been a handicap for the use of Raman spectroscopy as a reliable analytical technique for biological purposes (Kuhar et al. 2018). Currently, the use of confocal systems and highly sensitive detectors improve the specificity of this technique and allow the operator to analyze most biological cells (Smith et al. 2016; Gomes da Costa et al. 2019). Raman spectra can be considered to be the molecular fingerprints of the observed sample, as shown in the typical Raman spectrum of E. coli (Fig. 8). In fact, the Raman spectrum gives the

Raman scattering

Notch filter

detector Laser excitation

Mirror

laser

Rayleigh

microscope

sample

Chemical bond

Fig. 7 Schematic of a Raman spectrometer. The laser is focused on the sample by a confocal system; the Raman signal is filtered out from the laser light and dispersed on the CCD detector

33

Detection and Effects of Metal and Organometallic Compounds with Microbial. . .

839

Fig. 8 Raman spectrum of an Escherichia coli deposited on a gold surface. The Raman parameters were as follows: λ ¼ 785 nm, p ¼ 25 mW, t ¼ 15 s. (From Marine Bittel’s thesis)

vibrational states of the chemical bonds present in molecules, which reflect the molecular composition of the bacterial cell (Table 3). We report the presence of carbohydrates (glycosidic and polysaccharidic chains) at 407, 480, and 540 cm
1. Amino acid bands have many Raman bands at 665 cm
1 (guanine), 720 cm
1 (adenine), 830 cm
1 (tyrosine), 1000 cm
1 (phenylalanine), and 1570 cm
1 (adenine). Some regions of the spectrum give the compositions of lipids and proteins, as follows: 1245 and 1650–1680 cm
1 (amide II and I bands of protein), 1300 cm
1 (lipid), 1450 cm
1 (protein and fatty acid), and 1660 cm
1 (protein). The most important biological information is the signal of nucleic acid bands, that is, 1100 cm
1 (DNA), 780 and 810 cm
1 (phosphoesters in the DNA/RNA backbone), which are an important indicator of the physiological state of the cells (Maquelin et al. 2002; Assaf et al. 2014). Regardless of its advantages, Raman spectroscopy has been widely used as an alternative optical method in several application domains, such as medical (Neugebauer et al. 2015), food (Tao and Ngadi 2018), and microbial detection and classification (Ho et al. 2019). The main objective is to simplify the analysis process and reduce the investigation time (Harz et al. 2009; Lieutaud et al. 2019). Other examples of applications of Raman spectroscopy can be found elsewhere (Yang and Ying 2011; Galli et al. 2018; Araujo et al. 2018). Currently, Raman spectroscopy is opening new research perspectives for biological purposes and allows a fast and nondestructive method of observing living cells in near-physiological conditions (Puppels et al. 1991). In this part of the chapter, we describe the most important scientific publications concerning the use of Raman spectroscopy in the evaluation of toxicity in the environmental context. Only classical Raman spectroscopy is taken into consideration in our bibliographic research. Other variants of this technique exist, in particular, surface enhanced

840

S. Jouanneau et al.

Table 3 Main Raman bands observed for biological cells Raman shift (cm
1) 407 481 540 615–622 640–644 665–668 720–730 780 810–820 1000 1034 1100 1155 1175 1230–1295 1300 1430–1480 1573 1650–1680 1770

Molecule and chemical bond assignment Carbohydrates (glycosidic chains) Carbohydrates (polysaccharidic chains) C–O–C Phenylalanine (C–C twist) Tyrosine Guanine Adenine ADN ARN Phenylalanine Carbohydrates (C–C chains / C–O and C–O–H bonds) PO2
Carotenoid Tyrosine/Phenylalanine Amide III CH2 C–H C¼C Amides I/Fatty acids C¼O

Raman scattering (SERS), but the published studies are all interested in directly searching for the signature of the toxic agent and not the effect of pollutants on the living.

Non-targeted Approach for Measuring the Toxicity of Heavy Metals by Raman Spectroscopy The main challenges regarding toxicity in bioassays are the huge number of pollutants found in the environment. The best sensor should be reliable, reproducible, and sensitive to a wide range of pollutants. In fact, the increasing number of chemicals rejected in the environment and the need to identify and quantify of these compounds at a trace level promote the improvement of the sensitivity of sensors for widespread applications. For example, more than 398 different pesticides were found in the rivers of France in 2014 (http://eaufrance.fr/chiffres-cles). Nevertheless, for most sensors, direct research on pollutant(s) and/or the evaluation of its whole biological effects are proposed without any precision regarding its molecular target. As a result, there is a continuous and increasing demand for sensors that are able to evaluate the impacts of toxic agents to perform reliable monitoring. Raman spectroscopy has been proposed for the evaluation of the effects of toxicants, including antibiotics and many others molecules on cells (Jung et al.

33

Detection and Effects of Metal and Organometallic Compounds with Microbial. . .

841

Table 4 Most studied heavy metal and cellular models used to evaluate their toxicity by Raman spectroscopy Heavy metal Copper Lead and cadmium Nickel Cadmium Cadmium Arsenic

Cellular model Yeasts: Candida tropicalis and Schwanniomyces occidentalis Plant: Calendula officinalis Loess

Reference Radić et al. 2017 Fan et al. 2016

Bacterial cells: Streptomyces

Walter et al. 2012 Liu et al. 2020 Shen et al. 2019 Bittel et al. 2015a

Human cells: testis Animal cells: mice kidney Bacterial cells: Escherichia coli

2014; López-Díez et al. 2005; Ayala et al. 2018). Regardless of its advantages in the analysis of living samples, few research teams in the world use Raman spectroscopy for noninvasive investigations of heavy metal toxicity on living cells. Only 10 scientific publications were found in the literature after selecting studies related directly to our topic (Scopus: keywords, Raman spectroscopy, and heavy metal and toxicity). Different cellular models are proposed in the papers that were found (bacteria, yeast, and human cells), as described in Table 4. All of these publications use almost the same analytical protocol (Fig. 9). In fact, the analysis consists of first culturing cells in a suitable culture medium to obtain the biomass necessary for the test. After this phase, the cell cultures are transferred out and supplemented with a toxic solution to reach the required concentrations. Different concentrations of metal can be chosen to assess the existence of a “doseresponse” effect, and each experiment includes a control without a toxicant for comparison (Bittel et al. 2015a). Exposure tests are carried out depending on the time of cell division and should match with acute toxicity. To remove the culture medium, the samples are centrifuged and washed with water or adequate buffer solutions. The collected biomass is then spread onto a suitable surface and dried before analysis. Raman spectroscopy can be performed with different spectrometers; most of the measurements are performed at near infrared excitation (mostly at 785 nm) to reduce parasitic fluorescence. The spectral range is recorded from 400 to 1800 cm
1 because this region contains all of the useful Raman bands necessary for the extraction of essential information from the spectra. The preprocessing and data analysis, such as baseline correction and normalization, are generally performed by different software (Pieters et al. 2013; Gautam et al. 2015). Furthermore, the observed differences in Raman spectra can be matched with physiological events, such as fragmentation of dying cells or the decrease of protein synthesis. Consequently, the use of statistical methods is primordial to classify the results and to highlight the presence of a potential toxicity. With appropriate experimental conditions, numerous successful applications have been reported in the literature. In fact, confocal Raman spectroscopy presents reliable data and provides a novel method that is expected to be a promising strategy

Raman spectrometer

Deposit on gold surface & drynig 15 min

Incubation T°=30°C 250 rpm

Centrifugation at 7380 rpm, 5 min & washing 3 times with MgSO4 (10-2M)

Adding toxicant at exponential growth phase & re-incubation for 2 h

Fig. 9 Schematic protocol for the evaluation of metal toxicity on bacterial cells by Raman spectroscopy. Icons were uploaded from the Servier Medical ART website

Day 3: data analysis & statistical exploration

Cryotube stored at -80°C

Overnight T°=30°C 250 rpm

842 S. Jouanneau et al.

33

Detection and Effects of Metal and Organometallic Compounds with Microbial. . .

843

for the reproduction toxicity identification. This achievement provides a new perspective for in situ monitoring of substance changes in tissues and exhibits a more comprehensive understanding of the pathogenic mechanisms of heavy metals in molecular toxicology. Liu D et al. (2020) used confocal Raman spectroscopy to detect the biomolecular composition and changes in the testis under acute and chronic cadmium treatment. The data analysis showed the presence of specific Raman shifts associated with the mitochondria, nucleic acids, proteins, lipids, and cholesterol identified, which were distinguishing among groups undergoing different Cd treatment times. In another example, Shen et al. (2019) applied confocal Raman spectroscopy to map the pathological changes in situ in normal and Cd-exposed mice kidneys. All the biological investigations in the reported study were consistent with the Raman analysis, which revealed the progression and degree of lesions induced by Cd. Raman spectroscopy was also used for the determination of carbon nanotube toxicity on human alveolar carcinoma epithelial cells. In fact, Knief et al. (2009) demonstrate the potential of Raman spectroscopy to probe cytotoxicity after nanoparticle exposure. Analysis of the Raman band ratio demonstrates a dosedependent response that correlates to toxicological studies and demonstrates that clonogenic endpoints, and therefore the toxic response, can be potentially predicted from the spectra of cells exposed to undetermined doses, removing the need for costly and time-consuming biochemical assays. In the environmental context, Fan et al. (2016) worked on evaluating the impact of lead and cadmium on the cell walls of some Calendula officinalis plants. They reported that the Raman intensity of the bands at 2960 cm
1 increase under lead/ cadmium stress, which shows the changes in the arranging directions of cellulose molecules in cell wall samples. Fan et al. (2016) conclude that the components (pectin, protein, cellulose, etc.) and functional groups (-OH, N-H, C-O) of cell walls play an important role in the resistance process of cell walls to the stress of lead/ cadmium. In a different reported study, Radić et al. (2017) show the ability of Raman spectroscopy to evaluate the presence of copper in contaminated soil by analyzing microorganisms isolated from polluted sites. The Raman spectra of yeast isolated from this soil showed a certain increase in metallothione in production, which represents a specific response of the yeast species to the stress conditions. In another example, Liu Y et al. (2015) use confocal microprobe Raman spectroscopy for the detection of heavy metal copper in vetiver grass roots. He concludes that this technique could be used as a fast and simple tool to diagnose the content of heavy metal in vetiver grass, but the accuracy needs to be improved. Nevertheless, the most two important publications for the evaluation of toxicity by Raman spectroscopy are reported by Walter et al. (2012) and Bittel et al. (2015b) In fact, Walter et al. used Streptomyces as bioindicators for heavy metal contamination investigated by Raman spectroscopy. Their original approach proposes the study of a single cell to avoid time-consuming culturing. The identification of Raman spectra according to different Ni2+ concentration ranges is accomplished with a prediction accuracy of approximately 88%. Therefore, they conclude that Streptomyces can be used as a bioindicator to predict Ni2+ concentrations in the micro-molar range, but the low prediction rates need to be improved. Bittel et al. (2015b) highlight the potential of

844

S. Jouanneau et al.

Raman spectroscopy to evaluate arsenic toxicity. They propose this technique as an interesting multiparametric approach that gives the level of toxicity caused by arsenic on E. coli cells and suggest a methodology able to identify the molecular target of the relevant pollutant. An original statistical analysis method, called “Independent Component Analysis” (ICA) (Monakhova and Rutledge 2019), is proposed to identify molecules of bacterial cells impacted by the used heavy metal. In fact, ICA aims to decompose the multivariate signal (spectrum of a biological sample) into independent signals (contribution of each of the molecules) and allows a more specific chemical interpretation of the calculated components (Fig. 10).

Advantages, Limits, and Perspectives Raman spectroscopy offers an interesting multiparametric approach that provides an overview of all the physiological changes caused by a toxic agent. In this context, Raman spectroscopy could provide a screening method that is able to complete and overcome the limitations of other biosensors. Regardless of the advantages offered by Raman spectroscopy, many challenges remain for its toxicological applications. In fact, the chemical composition of living samples is continuously changing and depends on the life cycle of cells. Additionally, parameters such as temperature, oxygenation, and nutrients largely impact the physiology of cells and certainly affect their Raman spectra (Mlynáriková et al. 2015; Mukherjee et al. 2020). These changes make the obtained data subject to considerable variability. To guarantee an optimal discrimination, only bacteria in the same physiological state should be analyzed. Some studies connect the variations in the intensities of some Raman bands, particularly nucleic acid bands, to the physiological state of cells and suggest a methodology for the selection of good spectra based on the DNA/RNA ratio to elaborate the classification models (Assaf et al. 2014). The analysis of field samples adds a new challenge for Raman spectroscopy. In fact, each environment has a specific signature and impacts the signals of biological elements differently. This fact leads to the implementation of databases related to the studied environment but remains extremely complicated due to the specificity of each environment and the high number of the natural landscapes. The most appropriate solution is to develop and incorporate statistical methods capable of scanning the Raman spectrum and selecting only the areas that were previously identified as providing relevant information about a toxicant. Nevertheless, the identification of a toxicant is subtler in regard to categorizing the spectra of the same microorganism according to the exposure to several toxic compounds. In fact, toxic elements lead to variations in the molecular composition of bacteria, which necessarily provokes changes on their spectral fingerprint. The identification of the macromolecules that are impacted by the toxicity would define a “spectral-signature” of a given toxic compound. In this context, the main barriers come from the difficulty of extracting this signature from the global signal. The post-processing of data is a central issue, and here, the use of complex statistical methods, such as KNN or Neural Networks, is required.

Fig. 10 Averages of the Raman spectra of Escherichia coli after 40 min of exposure to four arsenic concentrations (0, 5, 50, and 500 μM). The Independent Components (ICs) reveal the arsenic effect on bacterial cells and give the Raman bands of molecules and cellular components identified as impacted by arsenic. The ICs with Raman bands highlighted in green correspond to the macromolecules that are the most impacted by the toxicity (IC3, IC4, IC5, IC6, and IC7). The profile of each component is characteristic of the contribution of specific cellular components

33 Detection and Effects of Metal and Organometallic Compounds with Microbial. . . 845

846

S. Jouanneau et al.

Additional technical challenges remain concerning the portability of the system. In fact, most studies propose lab systems with high performance but are not fielduseable. The recent technological developments should propose portable devices with acceptable prices for users and intelligent software that is able to display results without having to develop a new statistical tool every time so that this technique can be widely used.

References Affi M, Solliec C, Legentillomme P, Comiti J, Legrand J, Thouand G (2009) Numerical design of a card and related physicochemical phenomena occurring inside agarose-immobilized bacteria: a valuable tool for increasing our knowledge of biosensors. Sens Actuators B Chem 138(1): 310–317 Aloisi A, Della Torre A, De Benedetto A, Rinaldi R (2019) Bio-recognition in spectroscopy-based biosensors for *heavy metals-water and waterborne contamination analysis. Biosensors 9(3). https://doi.org/10.3390/bios9030096 Antizar-Ladislao B (2008) Environmental levels, toxicity and human exposure to tributyltin (TBT)contaminated marine environment. A review. Environ Int 34(2):292–308 Araujo CF, Nolasco MM, Ribeiro AMP, Ribeiro-Claro PJA (2018) Identification of microplastics using Raman spectroscopy: latest developments and future prospects. Water Res 142:426–440. https://doi.org/10.1016/j.watres.2018.05.060 Assaf A, Cordella CBY, Thouand G (2014) Raman spectroscopy applied to the horizontal methods ISO 6579:2002 to identify Salmonella Spp. in the food industry. Anal Bioanal Chem 406(20):4899–4910. https://doi.org/10.1007/s00216-014-7909-2 Atkins PW (2010) Shriver & Atkins’ inorganic chemistry. W.H. Freeman, New York Ayala OD, Wakeman CA, Pence IJ, Gaddy JA, Slaughter JC, Skaar EP, Mahadevan-Jansen A (2018) Drug-resistant Staphylococcus aureus strains reveal distinct biochemical features with Raman microspectroscopy. ACS Infect Dis 4(8):1197–1210. https://doi.org/10.1021/ acsinfecdis.8b00029 Ben-Israel O, Ben-Israel H, Ulitzur S (1998) Identification and quantification of toxic chemicals by use of Escherichia coli carrying lux genes fused to stress promoters. Appl Environ Microbiol 64(11):4346–4352 Bereza-Malcolm LT, Mann G, Franks AE (2015) Environmental sensing of heavy metals through whole cell microbial biosensors: a synthetic biology approach. ACS Synth Biol 4(5):535–546. https://doi.org/10.1021/sb500286r Bittel M, Cordella CBY, Assaf A, Jouanneau S, Durand MJ, Thouand G (2015a) Potential of Raman spectroscopy to monitor arsenic toxicity on bacteria: insights toward multiparametric bioassays. Environ Sci Technol 49(20):12324–12332. https://doi.org/10.1021/acs.est.5b03013 Bittel M et al (2015b) Potential of Raman spectroscopy to monitor arsenic toxicity on bacteria: insights toward multiparametric bioassays. PubMed – NCBI. https://www.ncbi.nlm.nih.gov/ pubmed/26398864. Accessed 24 Mar 2020 Briscoe SF, Diorio C, DuBow MS (1996) Luminescent biosensors for the detection of tributyltin and dimethyl sulfoxide and the elucidation of their mechanisms of toxicity. Environ Biotechnol 645–655. https://doi.org/10.1007/978-94-017-1435-8 Bulich AA (1982) A practical and reliable method for monitoring the toxicity of aquatic samples. Process Biochem 17(2):45–47 Bumbrah GS, Sharma RM (2016) Raman spectroscopy – basic principle, instrumentation and selected applications for the characterization of drugs of abuse. Egypt J Forensic Sci 6(3):209–215. https://doi.org/10.1016/j.ejfs.2015.06.001 Charrier T, Chapeau C, Bendria L, Picart P, Daniel P, Thouand G (2011) A multi-channel bioluminescent bacterial biosensor for the on-line detection of metals and toxicity. Part II:

33

Detection and Effects of Metal and Organometallic Compounds with Microbial. . .

847

Technical development and proof of concept of the biosensor. Anal Bioanal Chem 400(4): 1061–1070 Christophoridis C, Dedepsidis D, Fytianos K (2009) Occurrence and distribution of selected heavy metals in the surface sediments of Thermaikos Gulf, N. Greece. Assessment using pollution indicators. J Hazard Mater 168(2–3):1082–1091 Corbisier P, Ji G, Nuyts G, Mergeay M, Silver S (1993) LuxAB gene fusions with the arsenic and cadmium resistance operons of Staphylococcus aureus plasmid PI258. FEMS Microbiol Lett 110(SV):231–238 Corbisier P, Van der Lelie D, Borremans B, Provoost A, de Lorenzo V, Brown NL, Lloyd JR, Hobman JL, Csoregi E, Johansson G, Mattiasson B (1999) Whole cell- and protein-based biosensors for the detection of bioavailable heavy metals in environmental samples. Anal Chim Acta 387(3):235–244 Cornils B, Herrmann WA (eds) (2002) Applied homogeneous catalysis with organometallic compounds, 1st edn. Wiley. https://doi.org/10.1002/9783527618231 Cox PA (2004) Inorganic chemistry. In: Instant notes, 2nd edn. BIOS Scientific, Oxford Dameron C, International Programme on Chemical Safety (eds) (1998) Copper. Environmental health criteria. World Health Organization, Geneva Daunert S, Barrett G, Feliciano JS, Shetty RS, Shrestha S, Smith-Spencer W (2000) Genetically engineered whole-cell sensing systems: coupling biological recognition with reporter genes. Chem Rev 100(7):2705–2738. https://doi.org/10.1021/cr990115p Duffus JH (2002) “Heavy metals” a meaningless term? (IUPAC technical report). Pure Appl Chem 74(5):793–807. https://doi.org/10.1351/pac200274050793 Durand MJ, Thouand G, Dancheva-Ivanova T, Vachon P, DuBow M (2003) Specific detection of organotin compounds with a recombinant luminescent bacteria. Chemosphere 52(1):103–111 Elad T, Benovich E, Magrisso S, Belkin S (2008) Toxicant identification by a luminescent bacterial bioreporter panel: application of pattern classification algorithms. Environ Sci Technol 42(22):8486–8491 Fan C, Gao Y, Du B (2016) [Response of FTIR and Raman spectra on cell wall of calendula officinalis seedlings roots to the co-contamination stress of lead and cadmium in loess]. Guang Pu Xue Yu Guang Pu Fen Xi Guang Pu 36(7):2076–2081 Fernández-Luqueño F, López-Valdez F, Gamero-Melo P, Luna-Suárez S, Aguilera-González EN, Martínez AI, García-Guillermo MDS, Hernández-Martínez G, Herrera-Mendoza R, ÁlvarezGarza MA, Pérez-Velázquez IR (2013) Heavy metal pollution in drinking water – a global risk for human health: a review. Afr J Environ Sci Technol 7(7):567–584 Galli R, Preusse G, Schnabel C, Bartels T, Cramer K, Krautwald-Junghanns M-E, Koch E, Steiner G (2018) Sexing of chicken eggs by fluorescence and Raman spectroscopy through the shell membrane. PLoS One 13(2):e0192554. https://doi.org/10.1371/journal. pone.0192554 Gautam R, Vanga S, Ariese F, Umapathy S (2015) Review of multidimensional data processing approaches for Raman and infrared spectroscopy. EPJ Tech Instrum 2(1):1–38. https://doi.org/ 10.1140/epjti/s40485-015-0018-6 Gomes da Costa S, Richter A, Schmidt U, Breuninger S, Hollricher O (2019) Confocal Raman microscopy in life sciences. Morphologie 103(341):11–16. https://doi.org/10.1016/j. morpho.2018.12.003 Gu MB, Dhurjati PS, Van Dyk TK, LaRossa RA (1996) A miniature bioreactor for sensing toxicity using recombinant bioluminescent Escherichia coli cells. Biotechnol Prog 12(3):393–397. https://doi.org/10.1021/bp9600142 Gu MB, Gil GC, Kim JH (1999) A two-stage minibioreactor system for continuous toxicity monitoring. Biosens Bioelectron 14(4):355–361 Gu MB, Mitchell RJ, Kim BC (2004) Whole-cell-based biosensors for environmental biomonitoring and application. Adv Biochem Eng Biotechnol 87:269–305 Gueune H, Durand MJ, Thouand G, DuBow MS (2008) The YgaVP genes of Escherichia coli form a tributyltin-inducible operon. Appl Environ Microbiol 74(6):1954–1958 Gueune H, Thouand G, Durand MJ (2009) A new bioassay for the inspection and identification of TBT-containing antifouling paint. Mar Pollut Bull 58(11):1734–1738

848

S. Jouanneau et al.

Guzzo A, Diorio C, DuBow MS (1991) Transcription of the Escherichia coli FliC gene is regulated by metal ions. Appl Environ Microbiol 57(8):2255–2259 Hakkila K, Maksimow M, Karp M, Virta M (2002) Reporter genes LucFF, LuxCDABE, Gfp, and Dsred have different characteristics in whole-cell bacterial sensors. Anal Biochem 301(2): 235–242 Hakkila K, Green T, Leskinen P, Ivask A, Marks R, Virta M (2004) Detection of bioavailable heavy metals in EILATox-Oregon samples using whole-cell luminescent bacterial sensors in suspension or immobilized onto fibre-optic tips. J Appl Toxicol 24(5):333–342 Harz M, Rösch P, Popp J (2009) Vibrational spectroscopy-A powerful tool for the rapid identification of microbial cells at the single-cell level. Cytometry A 75(2):104–113. https://doi.org/ 10.1002/cyto.a.20682 Herrmann WA (2008) Ferrocene as a gasoline and fuel additive. In: Applied homogeneous catalysis with organometallic compounds. Wiley, pp 586–590. https://doi.org/10.1002/9783527618231. ch2j Ho C-S, Jean N, Hogan CA, Blackmon L, Jeffrey SS, Holodniy M, Banaei N, Saleh AAE, Ermon S, Dionne J (2019) Rapid identification of pathogenic bacteria using Raman spectroscopy and deep learning. Nat Commun 10(1). https://doi.org/10.1038/s41467-019-12898-9 Horry H, Durand MJ, Picart P, Bendriaa L, Daniel P, Thouand G (2004) Development of a biosensor for the detection of tributyltin. Environ Toxicol 19(4):342–345 Horry H, Charrier T, Durand MJ, Vrignaud B, Picart P, Daniel P, Thouand G (2007) Technological conception of an optical biosensor with a disposable card for use with bioluminescent bacteria. Sens Actuators B Chem 122(2):527–534 http://eaufrance.fr/chiffres-cles http://www.assainissement.developpement-durable.gouv.fr/recueil/ 01_TF/Dir%20substances%202013-39-UE.pdf. Accessed 20 May 2020 Ivask A, Rolova T, Kahru A (2009) A suite of recombinant luminescent bacterial strains for the quantification of bioavailable heavy metals and toxicity testing. BMC Biotechnol 9:41 Jaiser SR, Winston GP (2010) Copper deficiency myelopathy. J Neurol 257(6):869–881. https://doi. org/10.1007/s00415-010-5511-x Jouanneau S, Thouand G (2019) Biocapteurs pour la surveillance des polluants dans l’environnement. No. ref. article: chv1620 Jouanneau S, Durand M-J, Courcoux P, Blusseau T, Thouand G (2011) Improvement of the identification of four heavy metals in environmental samples by using predictive decision tree models coupled with a set of five bioluminescent bacteria. Environ Sci Technol 45(7): 2925–2931. https://doi.org/10.1021/es1031757 Jouanneau S, Durand MJ, Thouand G (2012) Online detection of metals in environmental samples: comparing two concepts of bioluminescent bacterial biosensors. Environ Sci Technol 46(21): 11979–11987. https://doi.org/10.1021/es3024918 Jouanneau S, Durand MJ, Lahmar A, Thouand G (2016) Main technological advancements in bacterial bioluminescent biosensors over the last two decades. In: Thouand G, Marks R (eds) Bioluminescence: fundamentals and applications in biotechnology. Advances in biochemical engineering/biotechnology, vol 3. Springer International Publishing, Cham, pp 101–116. https:// doi.org/10.1007/10_2015_333 Jung GB, Nam SW, Choi S, Lee G-J, Park H-K (2014) Evaluation of antibiotic effects on Pseudomonas aeruginosa biofilm using Raman spectroscopy and multivariate analysis. Biomed Opt Express 5(9):3238–3251. https://doi.org/10.1364/BOE.5.003238 Kabiersch G, Rajasärkkä J, Tuomela M, Hatakka A, Virta M, Steffen K (2013) Bioluminescent yeast assay for detection of organotin compounds. Anal Chem 85(12):5740–5745. https://doi. org/10.1021/ac4003062 Knief P, Clarke C, Herzog E, Davoren M, Lyng FM, Meade AD, Byrne HJ (2009) Raman spectroscopy – a potential platform for the rapid measurement of carbon nanotube-induced cytotoxicity. Analyst 134(6):1182–1191. https://doi.org/10.1039/B821393C Köhler S, Belkin S, Schmid RD (2000) Reporter gene bioassays in environmental analysis. Fresenius J Anal Chem 366(6):769–779

33

Detection and Effects of Metal and Organometallic Compounds with Microbial. . .

849

Kuhar N, Sil S, Verma T, Umapathy S (2018) Challenges in application of Raman spectroscopy to biology and materials. RSC Adv 8(46):25888–25908. https://doi.org/10.1039/c8ra04491k Lewis JC, Feltus A, Ensor CM, Ramanathan S, Daunert S (1998) Peer reviewed: applications of reporter genes. Anal Chem 70(17):579A–585A. https://doi.org/10.1021/ac9819638 Lieutaud C, Assaf A, Gonçalves O, Wielgosz-Collin G, Thouand G (2019) Fast non-invasive monitoring of microalgal physiological stage in photobioreactors through Raman spectroscopy. Algal Res 42:101595. https://doi.org/10.1016/j.algal.2019.101595 Liu Y, Zhang Y, Jiang L (2015) Detection of heavy metal copper in vetiver grass roots based on Raman spectroscopy and resin adsorption technology. Nongye Gongcheng XuebaoTransactions Chin Soc Agric Eng 31:174–178. https://doi.org/10.11975/j.issn.1002-6819.2015.24.026 Liu D, Wan J, Liu Z, Zhao Z, Zhang G, Leng Y (2020) Determination of cadmium induced acute and chronic reproductive toxicity with Raman spectroscopy | SpringerLink. https://link.springer. com/article/10.1007/s10103-020-02976-6. Accessed 24 Mar 2020 López-Díez EC, Winder CL, Ashton L, Currie F, Goodacre R (2005) Monitoring the mode of action of antibiotics using Raman spectroscopy: investigating subinhibitory effects of amikacin on Pseudomonas aeruginosa. Anal Chem 77(9):2901–2906. https://doi.org/10.1021/ac048147m Magrisso S, Erel Y, Belkin S (2008) Microbial reporters of metal bioavailability. Microb Biotechnol 1(4):320–330 Maquelin K, Kirschner C, Choo-Smith L-P, van den Braak N, Endtz HP, Naumann D, Puppels GJ (2002) Identification of medically relevant microorganisms by vibrational spectroscopy. J Microbiol Methods 51(3):255–271. https://doi.org/10.1016/s0167-7012(02)00127-6 Meighen EA (1993) Bacterial bioluminescence: organization, regulation, and application of the lux genes. FASEB J 7(11):1016–1022 Mlynáriková K, Samek O, Bernatová S, Růžička F, Ježek J, Hároniková A, Šiler M, Zemánek P, Holá V (2015) Influence of culture media on microbial fingerprints using Raman spectroscopy. Sensors 15(11):29635–29647. https://doi.org/10.3390/s151129635 Monakhova Y, Rutledge D (2019) Independent components analysis (ICA) at the “cocktail-party” in analytical chemistry. Talanta 208:120451. https://doi.org/10.1016/j.talanta.2019.120451 Mukherjee R, Verma T, Nandi D, Umapathy S (2020) Understanding the effects of culture conditions in bacterial growth: a biochemical perspective using Raman microscopy. J Biophotonics 13(1):e201900233. https://doi.org/10.1002/jbio.201900233 Nagel B, Dellweg H, Gierasch LM (1992) Glossary for chemists of terms used in biotechnology (IUPAC recommendations 1992). Pure Appl Chem 64(1):143–168. https://doi.org/10.1351/ pac199264010143 Neugebauer U, Rösch P, Popp J (2015) Raman spectroscopy towards clinical application: drug monitoring and pathogen identification. Int J Antimicrob Agents 46:S35–S39. https://doi.org/ 10.1016/j.ijantimicag.2015.10.014 Pieters S, Vander Heyden Y, Roger J-M, D’Hondt M, Hansen L, Palagos B, De Spiegeleer B, Remon J-P, Vervaet C, De Beer T (2013) Raman spectroscopy and multivariate analysis for the rapid discrimination between native-like and non-native states in freeze-dried protein formulations. Eur J Pharm Biopharm 85(2):263–271. https://doi.org/10.1016/j.ejpb.2013.03.035 Preston S, Coad N, Townend J, Killham K, Paton GI (2000) Biosensing the acute toxicity of metal interactions: are they additive, synergistic, or antagonistic? Environ Toxicol Chem 19(3): 775–780. https://doi.org/10.1002/etc.5620190332 Puppels GJ, Olminkhof JHF, Segers-Nolten GMJ, Otto C, de Mul FFM, Greve J (1991) Laser irradiation and Raman spectroscopy of single living cells and chromosomes: sample degradation occurs with 514.5 nm but not with 660 nm laser light. Exp Cell Res 195(2):361–367. https://doi. org/10.1016/0014-4827(91)90385-8 Qiao Y, Yang Y, Zhao J, Tao R, Xu R (2013) Influence of urbanization and industrialization on metal enrichment of sediment cores from Shantou Bay, South China. Environ Pollut 182:28–36 Radić DS, Pavlović VP, Lazović MM, Jovičić-Petrović JP, Karličić VM, Lalević BT, Raičević VB (2017) Copper-tolerant yeasts: Raman spectroscopy in determination of bioaccumulation

850

S. Jouanneau et al.

mechanism. Environ Sci Pollut Res 24:21885–21893. PubMed – NCBI. https://www.ncbi.nlm. nih.gov/pubmed/28779342. Accessed 24 Mar 2020 Roda A, Roda B, Cevenini L, Michelini E, Mezzanotte L, Reschiglian P, Hakkila K, Virta M (2011) Analytical strategies for improving the robustness and reproducibility of bioluminescent microbial bioreporters. Anal Bioanal Chem 401(1):201–211 Selifonova O, Burlage R, Barkay T (1993) Bioluminescent sensors for detection of bioavailable Hg (II) in the environment. Appl Environ Microbiol 59(9):3083–3090 Semple KT, Doick KJ, Jones KC, Burauel P, Craven A, Harms H (2004) Defining bioavailability and bioaccessibility of contaminated soil and sediment is complicated. Environ Sci Technol 38(12):228A–231A Serat WF, Budinger FE, Mueller PK (1965) Evaluation of biological effects of air pollutants by use of luminescent bacteria. J Bacteriol 90(3):832–833 Serat WF, Budinger FE, Mueller PK (1967) Toxicity evaluation of air pollutants by use of luminescent bacteria. Atmos Environ 1(1):21–32. https://doi.org/10.1016/0004-6981(67) 90105-9 Shen R et al (2019) Ex vivo detection of cadmium-induced renal damage by using confocal Raman spectroscopy. J Biophotonics. Wiley Online Library. https://onlinelibrary.wiley.com/doi/abs/10. 1002/jbio.201900157. Accessed 24 Mar 2020 Simpson ML, Sayler GS, Applegate BM, Ripp S, Nivens DE, Paulus MJ, Jellison GE (1998) Bioluminescent-bioreporter integrated circuits form novel whole-cell biosensors. Trends Biotechnol 16(8):332–338 Smith R, Wright KL, Ashton L (2016) Raman spectroscopy: an evolving technique for live cell studies. Analyst 141(12):3590–3600. https://doi.org/10.1039/C6AN00152A Stocker J, Balluch D, Gsell M, Harms H, Feliciano J, Daunert S, Malik KA, van der Meer JR (2003) Development of a set of simple bacterial biosensors for quantitative and rapid measurements of arsenite and arsenate in potable water. Environ Sci Technol 37(20):4743–4750 Tao F, Ngadi M (2018) Recent advances in rapid and nondestructive determination of fat content and fatty acids composition of muscle foods. Crit Rev Food Sci Nutr 58(9):1565–1593. https:// doi.org/10.1080/10408398.2016.1261332 Thevenot DR, Toth K, Durst RA, Wilson GS (2001) Electrochemical biosensors: recommended definitions and classification. Biosens Bioelectron 16(1–2):121–131 Ulitzur S, Barak M (1988) Detection of genotoxicity of metallic compounds by the bacterial bioluminescence test. J Biolumin Chemilumin 2(2):95–99 US EPA (2019) Analytical methods approved for drinking water compliance monitoring of inorganic contaminants and other inorganic constituents. https://nepis.epa.gov/Exe/ZyPDF. cgi?Dockey¼P100WD2D.txt. Accessed 6 Jan 2020 Verma N, Singh M (2005) Biosensors for heavy metals. Biometals 18(2):121–129 Walter A et al (2012) Raman spectroscopic detection of Nickel impact on single Streptomyces cells – possible bioindicators for heavy metal contamination. J Raman Spectrosc. Wiley Online Library. https://onlinelibrary.wiley.com/doi/abs/10.1002/jrs.3126. Accessed 24 Mar 2020 Yang D, Ying Y (2011) Applications of Raman spectroscopy in agricultural products and food analysis: a review. Appl Spectrosc Rev 46(7):539–560. https://doi.org/10.1080/ 05704928.2011.593216 Zoroddu MA, Aaseth J, Crisponi G, Medici S, Peana M, Nurchi VM (2019) The essential metals for humans: a brief overview. J Inorg Biochem 195:120–129. https://doi.org/10.1016/j. jinorgbio.2019.03.013

Microbial Biosensors for the Detection of Organic Pollutants

34

Benjamin Shemer and Shimshon Belkin

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Naphthalene . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Benzene, Toluene, Ethylbenzene and Xylene (BTEX) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Aliphatic Hydrocarbons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Nitroaromatic Explosives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Pharmaceuticals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Hormones and Endocrine-Disrupting Compounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Antibiotics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Pesticides and Other Agrochemicals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Halogenated Organic Pollutants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions and Future Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

852 852 856 857 859 861 861 862 864 866 868 868

Abstract

Over the last 30 years, numerous scientific publications have described the design, construction, testing, and characterization of diverse whole-cell bioreporter strains for the detection and quantification of organic pollutants. In this chapter we attempt to review these reports, providing the relevant information regarding the sensor strains’ construction principles and performance characteristics, with a special emphasis on the detection thresholds of either specific target compounds or classes of such chemicals. Keywords

Bioreporter · Biosensor · Organic pollutants · Endocrine-disrupting compounds · Explosives · Pesticides · BTEX · Halogenated organics B. Shemer · S. Belkin (*) Institute of Life Sciences, The Hebrew University of Jerusalem, Jerusalem, Israel e-mail: [email protected] © Springer Nature Switzerland AG 2022 G. Thouand (ed.), Handbook of Cell Biosensors, https://doi.org/10.1007/978-3-030-23217-7_92

851

852

B. Shemer and S. Belkin

Introduction The rapid expansion in industrial, pharmaceutical, and agricultural technologies throughout the past century, often augmented either by the absence of adequate environmental regulations or by the lack of their enforcement, has led to the discharge of high quantities of diverse organic pollutants to the environment. The environmental persistence of some of these pollutants, and their significant deleterious effects on environmental well-being and on human health, prompted a worldwide effort toward their mitigation and remediation. An initial step in any remediation effort involves the careful analysis of the nature of the pollutants, the degree of contamination, and an assessment of the risks it harbors. Chemical analysis, clearly a crucial element in this process, is nevertheless limited by its inability to provide information on the toxicity of a chemical or its bioavailability. As a partial response to this need, different bioassays have been developed to provide the missing complementary evidence regarding availability and biological effects of either specific pollutants or of uncharacterized environmental samples. Such information is essential for assessing the actual risks involved in any pollution event or contaminated site. Aided by the accelerated development of synthetic biology tools over the past decades, cell-based assays have emerged as an attractive option for pollution monitoring. Characterized by ease of growth and maintenance, large and homogeneous populations, low costs, rapid responses, and – maybe most important – facility of genetic manipulations, numerous whole-cell bioreporters have been described in the scientific literature. In this chapter we attempt to briefly review those sensor strains that detect and quantitatively report the presence of organic molecules. We have limited the review to those reports describing sensors of either specific organic compounds or specific classes of such chemicals, thus excluding sensors of general biological effects such as toxicity, genotoxicity, or oxidative stress. The selected reports are divided in the review below according to the chemical class of the target organic molecules; the same classification is also implemented in Table 1, in which the diverse microbial reporter strains mentioned in the present review and their target chemicals are listed.

Naphthalene Aromatic compounds, most notably naphthalene, were among the first to be used as target molecules for microbial bioreporters. Naphthalene, a polycyclic aromatic hydrocarbon (PAH) with possible carcinogenic effects (Boffetta et al. 1997), has been in commercial use for almost two centuries, mostly as a precursor to phthalic anhydride in the plastic industry. Its intensive use and distribution in the environment, along with its adverse health effects to humans (Agency for Toxic Substances and Disease Registry (ATSDR) 2005), promoted the efforts to develop an effective biosensing device for its detection. The first report of a naphthalene bioreporter was published by the Sayler group (Burlage et al. 1990; King et al. 1990). The promoter region of the naphthalene

34

Microbial Biosensors for the Detection of Organic Pollutants

853

Table 1 Microbial sensors for organic pollutants mentioned in this chapter

Host microorganism

Reporting element or monitored paramter

References

Pseudomonas Acinetobacter E. coli

luxCDABE luxCDABE ß-galactosidase

Burlage et al. (1990) Sun et al. (2017) Cho et al. (2014)

E. coli

Firefly luciferase Oxygenase assay

m-Xylene

P. putida, Burkholderia cepacia, P. mendocina, E. coli P. putida

Willardson et al. (1998) Tizzard and LloydJones (2007)

Toluene

E. coli

Aequorin

BTEX Toluene, xylene

P. putida E. coli

luxCDABE Alkalinephosphatase, β-galactosidase

Aliphatic hydrocarbons Short- and middlechained alkanes

E. coli

luxAB

Acinetobacter baylyi Yarrowia lipolytica

luxCDABE

Target analyte Naphthalene Naphthalene Naphthalene Salicylate, naphthalene BTEX 3-Methylbenzylalcohol, 3-xylene Toluene, other BTEX compounds

Alkanes Short- and middlechained alkanes Long-chain alkanes Trace explosives TNT DNT DNT DNT TNT TNT

Alcanivorax borkumensis, E. coli

ß-galactosidase, luxCDABE

Oxygen consumption luxAB, eGFP

De Las Heras et al. (2008) and de las Heras and de Lorenzo (2011a) Zeinoddini et al. (2010) Applegate et al. (1998) Paitan et al. (2004)

Sticher et al. (1997) and Minak-Bernero et al. (2004) Zhang et al. (2012) Alkasrawi et al. (1999) Kumari et al. (2011)

Unspecified E. coli

GFP GFP, luxCDABE

E. coli Saccharomyces cerevisiae Dictyosphaerium chlorelloides E. coli

GFP GFP

Burlage et al. (1999) Yagur-Kroll et al. (2014, 2015) Davidson et al. (2012) Radhika et al. (2007)

Chlorophyll fluorescence GFP

Altamirano et al. (2004) Looger et al. (2003)a (continued)

854

B. Shemer and S. Belkin

Table 1 (continued)

Host microorganism

Reporting element or monitored paramter

E. coli E. coli E. coli E. coli

β-galactosidase luxCDABE GFP luxCDABE

Genotoxic agents (nalidixic acid)

E. coli

Various antibiotic groups

E. coli panel

phoA (alkaline phosphatase), luxCDABE luxCDABE

Hormones Thyroid hormone

S. cerevisiae

β-galactosidase

Androgen

S. cerevisiae

Androgen Glucocorticoid

S. cerevisiae S. cerevisiae

Photinus pyralis luciferase luxCDABE β-galactosidase

Several hormone groups Progesterone

S. cerevisiae S. cerevisiae

β-galactosidase β-galactosidase

Estrogen

S. cerevisiae

β-galactosidase

Target analyte Antibiotics Tetracyclines Tetracyclines Tetracyclines Β-Lactam

References Chopra et al. (1990) Korpela et al. (1998) Bahl et al. (2005) Valtonen et al. (2002) and Smolander et al. (2009) Biran et al. (2011)

Melamed et al. (2012, 2014) and Kao et al. (2018) Li et al. (2008) and Shiizaki et al. (2010) Michelini et al. (2008)

Estrogen Aspergillus nidulans Estrogen, progesterone, Arxula androgen adeninivorans Androgen, estrogen, S. cerevisiae progesterone Estradiol, bisphenol A E. coli Pesticides and other agrochemicals Organophosphates E. coli Organophosphates P. putida Organophosphates E. coli

β-galactosidase CFP, GFP, DsRed2 β-galactosidase

Eldridge et al. (2007) Wright and Gustafsson (1992) Miller et al. (2010) Klotz et al. (1997) and Berg et al. (2000) Routledge and Sumpter (1996) and Arnold et al. (1996) Zutz et al. (2017) Chamas et al. (2017a, b) Gaido et al. (1997)

Impedance

Furst et al. (2017)

Amperometric Amperometric RFP

Organophosphates

E. coli

Amperometric

Lindane

E. coli

Amperometric

Rainina et al. (1996) Lei et al. (2005) Chong and Ching (2016) Mulchandani et al. (1998) Anu Prathap et al. (2012) (continued)

34

Microbial Biosensors for the Detection of Organic Pollutants

855

Table 1 (continued)

Target analyte Lindane Atrazine, endrin Atrazine, simazine, diuron DNOC, atrazine, simazine, isoproturon, diuron Simazine, atrazine, propazine, terbuthylazine, linuron Atrazine, cyanuric acid Organophosphates Organophosphates, carbamates Halogenated organics 4CBA 2-CPA Various chlorinated and brominated hydrocarbons TCE TCE PCB PCB a

Host microorganism Streptomyces strain M7 Scenedesmus subspicatus Chlorella vulgaris C. vulgaris

Reporting element or monitored paramter Impedance Chlorophyll fluorescence Chlorophyll fluorescence Chlorophyll fluorescence

References López Rodriguez et al. (2015) Frense et al. (1998) Naessens et al. (2000) Védrine et al. (2003)

D. chlorelloides, Scenedesmus intermedius, Scenedesmus sp. E. coli E. coli S. cerevisiae

Chlorophyll fluorescence

Peña-Vázquez et al. (2009)

lux CDABE Amperometric Colorimetric

Hua et al. (2015) Rainina et al. (1996) Li et al. (2013)

E. coli

luxCDABE

E. coli Rhodococcus sp.

luxCDABE Potentiometric

Rozen et al. (1999) and Köhler et al. (2000) Tauber et al. (2001) Hutter et al. (1995) and Peter et al. (1996)

P. aeruginosa P. putida E. coli Pseudomonas sp.

Potentiometric Potentiometric luxCDABE Colorimetric

Han et al. (2001) Hnaien et al. (2011) Layton et al. (1998) Gavlasova et al. (2008)

Later contested by Reimer et al. (2014)

dioxygenase-encoding nahA gene, from the NAH7 plasmid of Pseudomonas putida, was fused to the bioluminescence luxCDABE gene cassette from Aliivibrio fischeri (formerly Vibrio fischeri); the plasmid-borne fusion was then transformed into a P. putida strain, to yield a bioreporter that emitted light in a dose-dependent manner in the presence naphthalene or salicylate. By measuring light intensity, Burlage et al. (1990) were also able to track the rate of gene expression of the naphthaleneassociated genes in real time and showed that the rate of naphthalene metabolism was dependent on the culture’s growth phase. This construct was successfully used to monitor the long-term (ca. 2 years) bioremediation process of soil contaminated by naphthalene, anthracene, and phenanthrene (Ripp et al. 2000). Several recent studies applied a similar methodology to construct other naphthalene bioreporters. Sun et al. (2017) used an Acinetobacter ADPWH_lux, a strain in

856

B. Shemer and S. Belkin

which the luxCDABE cassette was inserted between the salA and salR genes on the chromosome, as host for a plasmid containing the P. putida nahAD operon. This construct was shown to respond to soil and water samples containing naphthalene, but not other PAHs. Construction of an Escherichia coli-based biosensor for naphthalene was reported by Cho et al. (2014). This strain carries a lacZ fusion of the nahR gene promoter, encoding a regulatory element of the naphthalene degradation pathway. By assaying β-galactosidase activity, this strain successfully detected naphthalene in both aqueous and gaseous samples.

Benzene, Toluene, Ethylbenzene and Xylene (BTEX) Benzene, toluene, ethylbenzene, and xylene (BTEX) are major environmental pollutants. These volatile aromatic compounds are found in crude oil, petroleum, and natural gas deposits, as well as in motor vehicles and aircraft emissions and even cigarette smoke. Among these compounds, the water-soluble benzene is considered the most hazardous to humans due to its carcinogenic potential (Wilbur et al. 2008); other members of this family also exert significant deleterious health effects, including damage to neuronal membrane as well as reproductive, respiratory, and blood disorders (Agency for Toxic Substances and Disease Registry (ATSDR) 2004). Some of the early attempts to construct a microbial biosensor for BTEX detection made use of the P. putida TOL plasmid pWWO. This toluene-degrading soil bacterium is capable of degrading toluene and toluene-related compounds to Krebs cycle intermediates (Worsey and Williams 1975; Burlage et al. 1989; Assinder and Williams 1990). The first stage in this process includes the binding of toluene (and related compounds) to the XylR transcriptional regulator, which then allows the interaction of XylR and the Pu promoter region. This activates the transcription of several genes involved in the biotransformation of toluene to benzoate. Another set of genes is then transcribed through the Pm promoter, transforming the benzoate to Krebs cycle intermediates. The Pu promoter, activated immediately upon exposure to toluene, was employed as the sensing element of bioreporters aimed at its detection. One such system was reported by Willardson et al. (1998). The XylR and Pu elements were fused to firefly luciferase and the resulting plasmid transformed into an E. coli host. The resulting bioreporter emitted a luminescent signal upon exposure to micromolar levels of XylR-binding compounds such as toluene, benzene, and xylene. This strain was shown to detect BTEX compounds in aquifer and soil samples. P. putida’s innate ability to degrade BTEX was also used by Applegate et al. (1998). The promoter of the tod operon, encoding several enzymes capable of transforming benzene, toluene, and p-xylene, was fused to the luxCDABE gene cassette in the bacterial chromosome. The resulting strain detected 30 μg/L of toluene and responded to jet fuel residues in aqueous solution. In another report, with E. coli as host and the previously described XylR-Pu as the sensing mechanism, aequorin served as the reporting element (Zeinoddini et al.

34

Microbial Biosensors for the Detection of Organic Pollutants

857

2010). This protein uses coelenterazine hcp (a synthetic derivative of coelenterazine) and O2 as substrates. Upon exposure to Ca2+, the coelenterazine is converted to coelentramide, a process accompanied by the emission of an instant flash of blue light. This process is advantageous in that while the accumulation of the protein is a long and steady process, the bioluminescent reaction is very rapid. The detection limit for toluene obtained with this design was lower than 1 μM. A more elaborate design for BTEX sensing, also based on XylR and the Pu promoter, was reported by de las Heras and de Lorenzo (2011a). The Pu promoter and the luxCDABE cassette of Photorhabdus luminescens were fused and randomly inserted into the P. putida chromosome using a mini-transposon delivery vector. A similar method was used to insert a fusion of XylR, and its constitutive promoter Pr, but in an opposite orientation. The resulting microbial bioreporter was then lyophilized and immobilized in gelatin capsules and spread on a soil microcosm composed of a filter paper soaked with m-xylene and covered with soil. By applying water or increasing the humidity in the chamber, the gelatin dissolved, allowing the bacteria to be released to the surface. A bioluminescent signal was observed using a CCD camera after 32–58 h, depending on the cryoprotectants employed during lyophilization. A P. putida strain with a similar design and lacZ as reporter gene, previously constructed by the same authors (De Las Heras et al. 2008), was used as control. Another approach for constructing a BTEX bioreporter was reported by Tizzard and Lloyd-Jones (2007), based on the fact that bacterial degradation of highly reduced hydrocarbons is initiated by oxygenase-mediated substrate oxidation. The high specificity, stability, and activity render such oxygenases ideal in vivo biosensing components. The signal output of this system was based on measuring the rate of catabolism, as manifested by oxygen consumption measured using BD Oxygen Biosensor™ plates. The assay measured oxygen uptake by a test bacterium (different strains were used for each analyte) exposed to benzene, toluene, ethylbenzene, and the three isomers of xylene, in comparison to the same bacterium lacking the appropriate oxygenase or in which respiratory activity was inactivated. Using this methodology, 125 μM of toluene were detected in water samples. In a different report by Paitan et al. (2004), the xylS promoter of the TOL operon was fused to either lacZ (β-galactosidase, with p-aminophenyl-β-Dgalactopyranoside as substrate) or phoA (alkaline phosphatase, with p-aminophenyl phosphate). In both cases the product of the enzymatic reaction is p-aminophenol, which is oxidized by an electrode and converted to a measurable current signal by use of the chronoamperometric technique. The resulting bioreporter could detect 40–50 μM of toluene and m-xylene within 20–30 min from exposure.

Aliphatic Hydrocarbons Aliphatic hydrocarbons are among the most prevalent environmental pollutants, primarily due to their use as fuels but also through other industrial processes. Significant contamination of soil, air, aquifers, and marine environments by these

858

B. Shemer and S. Belkin

compounds draws great public concern, due to the adverse health effects reported to occur after exposure to many members of this group (Laffon et al. 2016; Azuma et al. 2018). A microbial biosensor for the detection of aliphatic hydrocarbons was described by Sticher et al. (1997). It consisted of an E. coli host, transformed with two plasmids: one harbored a fusion between the promoter region of alkB, an alkane 1-monooxygenase originating from Pseudomonas oleovorans, to the luxAB genes of Vibrio harveyi, and the other containing alkS, encoding the transcriptional activator of alkB. The resulting strain detected octane concentrations as low as 24.5 nM and exhibited good specificity toward short- and middle-chained alkanes. Another E. coli-based bioreporter, harboring a plasmid containing the P. oleovorans alcohol dehydrogenase alkJ and constitutively expressed V. harveyi luxAB genes, was reported by Minak-Bernero et al. (2004). The alkJ gene recognizes C5-C12 linear primary alcohols and converts them to aldehydes. These, in turn, act as substrates in the luciferase-catalyzed luminescent reaction. To allow detection of octane, the alkane hydroxylase alkBFG genes, also originating from P. oleovorans, were introduced via plasmid, allowing the oxidation of octane to a primary alcohol, which could then be detected by the mechanism previously described. This strain was shown to detect 10–200 μM of octane. A bioreporter capable of detecting longer-chain alkanes and alkenes, reported by Zhang et al. (2012), was based on Acinetobacter baylyi as host. This bacterium, ubiquitous in natural soil and marine environments, is capable of degrading alkanes with 12- to 36-long carbon chains (Ratajczak et al. 1998). The chromosomally integrated luxCDABE genes were controlled by the alkR regulation system. Introduction of several point mutations to the promoter region of alkR shortened the response time. This construct was demonstrated to determine the presence of mineral and crude oils in water, seawater, and soil. The marine oil-degrading bacterium Alcanivorax borkumensis, ubiquitous in oilpolluted environments (Hara et al. 2003), was also employed to detect long-chained alkanes (Kumari et al. 2011). The promoter region of its n-tetradecane- and hexadecane-inducible alkSB1GHJ operon, along with its transcriptional regulator AlkS, was used as the sensing element. This plasmid-based construct was used to transform both E. coli and A. borkumensis hosts, which were shown to detect octane, tetradecane, and crude oil, with the latter strain being more sensitive to low concentrations of pure alkanes and oil (Sevilla et al. 2015). A yeast biosensor for middle-chain alkanes, capable of performing in temperatures as low as 5
C, was reported by Alkasrawi et al. (1999). Yarrowia lipolytica, a psychrotrophic yeast strain isolated from diesel oil-contaminated alpine soil and capable of alkane degradation, was used as host, and the concentration of alkanes was assayed as oxygen uptake during consumption of the tested chemicals. While less sensitive than the E. coli-based system reported by Sticher et al. (1997), the low temperature capabilities of this sensor may be advantageous under environmental conditions.

34

Microbial Biosensors for the Detection of Organic Pollutants

859

Nitroaromatic Explosives The detection of nitroaromatic explosives is of high importance due to a combination of humanitarian, security, and environmental concerns. The extensive production of explosives in the past two centuries resulted in vast contamination of soils and aquifers. As some of these compounds, namely, 2,4,6-trinitrotoluene (TNT) and 2,4-dinitrotoluene (2,4-DNT), are toxic and possibly carcinogenic (Agency for Toxic Substances and Disease Registry (ATSDR) 1995, 2016), development of rapid, reliable, and cheap techniques for their detection is highly desirable. Bioreporters have also been suggested as means for the remote detection of buried landmines, thus mitigating the risk for involved personnel. Most landmines produced throughout the past century contain a TNT-based explosive charge enveloped in a metallic or plastic casing. While TNT is the most abundant compound in the explosive charge, it also contains the more volatile impurities 2,4-DNT and 2,4-dinitrobenzene (DNB). These compounds are constantly released from the landmine and migrate to the surrounding soil. Due to its high volatility and stability, 2,4-DNT is considered the tracer of choice for TNT-based explosives and is therefore the target in most reports of explosives’ whole-cell sensors. The concept of a microbial-based system for landmine detection was first described in a patent by Burlage et al. (1999). This system was based on a genetically modified bacterium (the author did not specify the strain, P. putida and Bacillus subtilis were both mentioned in the patent document) sprayed directly on the area containing the landmines. Upon exposure to TNT/DNT vapors transpiring from the buried landmines, the bacterium was programmed to produce a green fluorescent protein (GFP) which could then be detected using UV lamps. While a preliminary success with this construct was reported (MacDonald et al. 2003), no further publications regarding its development were reported to date. Several additional microbial bioreporters for explosives detection were reported throughout the past two decades, employing diverse hosts, reporter genes, and sensing elements (Shemer et al. 2015, 2017). A notable example is Radhika et al. (2007), who harnessed elements of the rat olfactory system, shown to respond to DNT, and inserted them to a yeast strain while coupling their activation to the transcription of GFP. Another interesting approach was reported by Davidson et al. (2012), who used a riboswitch, an aptamer element in the 50 untranslated region of some RNAs that can bind specific target molecules. The aptamer binds to the DNT molecule and induces a change in the RNA’s secondary structure, which enables transcription of an otherwise repressed protease. This enzyme then cleaves a bond between GFP and a nonfluorescent mutant yellow fluorescent protein, and thus unquenches the GFP fluorescence. A different concept for TNT detection was reported by Altamirano et al. (2004), who used two types of microalgae in a single assay. The toxic effect of TNT causes inhibition of the fluorescent signal emitted from PSII of the host organism. By measuring the difference in PSII fluorescence in the presence of TNT in a strain sensitive toward TNT and in a TNT-resistant mutant, the authors were able to detect as low as 0.5 mg/L of TNT after 3 min of exposure.

860

B. Shemer and S. Belkin

An innovative approach was reported by Looger et al. (2003), who attempted to redesign the ribose-binding site of an E. coli ribose-binding protein to recognize TNT. This was apparently achieved by analysis of the protein’s high-resolution three-dimensional structure, and prediction of an amino acid sequence that will complement the structure of TNT. This design was integrated into a synthetic twocomponent signal transduction pathway in E. coli, with a very specific and sensitive response to TNT. This report, however, was later contested by Reimer et al. (2014), who demonstrated that no binding of the protein to TNT occurred, nor was any signal detected in response to TNT. Several other studies also reported the construction of microbes capable of detecting DNT and TNT (Garmendia et al. 2008; De Las Heras et al. 2008; de las Heras and de Lorenzo 2011b; Lönneborg et al. 2012; Kim et al. 2008; Tan et al. 2015). Harnessing such genetically engineered microbes for actual landmine detection was reported by Belkin et al. (2017). This study made use of E. coli-based bioreporters harboring a plasmid-borne fusion of a mutated yqjF gene promoter to gfpmut2 (Yagur-Kroll et al. 2014, 2015). The constructed bioreporter produced the fluorescent protein in the presence of TNT, DNT, and DNB. When encapsulated in alginate beads and spread over the test area, the fluorescent signal produced by the bacteria was measured by a laser-equipped telescope. This system successfully detected buried antipersonnel landmines buried in soil from a distance of 20 m. A schematic depiction of the detection concept is presented in Fig. 1.

Fig. 1 Remote detection of buried landmines using a bacterial bioreporter. The E. coli bioreporters are encapsulated in Ca-alginate beads and spread across the target area. Upon exposure to the explosives’ vapors emitted from buried landmines, a regulatory protein binds to the explosive tracer and then triggers the yqjF gene promoter, driving GFP synthesis. A telescope equipped with a laser source scans the target area, excites the GFP molecules, and collects the emitted fluorescence, thus monitoring the spatial distribution of GFP accumulation. Comparison of the fluorescent signal across the scanned area at t = 0 and after 22 h allowed the detection of most targets. (Belkin et al. 2017)

34

Microbial Biosensors for the Detection of Organic Pollutants

861

Pharmaceuticals Hormones and Endocrine-Disrupting Compounds The increased consumption, and subsequent discharge, of either hormones or hormone-like substances to the environment has been a global concern for several decades, as have been that of diverse endocrine-disrupting compounds. The ineffectiveness of traditional wastewater treatment facilities in removing such substances, along with the significant biological effects even upon exposure to very small concentrations of estrogens, progesterones, androgens, and dioxinlike compounds (Luo et al. 2014), called for the development of new techniques for rapid detection of these substances. One of the techniques introduced over the past decade for this purpose is based on the use of genetically modified yeast strains. Yeast cells, which on the one hand are eukaryotic in structure but on the other hand lack endogenous hormone receptor elements, provide a robust sensor model which is superior to prokaryotic platforms when attempting to assess a potential threat to humans. A common design of a yeast-based biosensor for hormone detection (Fig. 2) includes the binding of the target to a constitutively expressed human hormone receptor (HR), thus allowing its binding to a hormone response

Fig. 2 A common design for a yeast-based biosensor for detection of hormones or hormonelike activity. A human hormone receptor, located in the chromosome, is under the control of a constitutive promoter (A). The target analyte enters the cell (B), and binds to the hormone receptor (C). The binding causes a conformational change in the hormone receptor, which allows the hormone-receptor complex to bind to the hormone response element (HRE, D), located on a plasmid and fused to a reporter gene. The binding of the complex to the HRE promotes transcription of the reporter protein (E) and the generation of a measurable signal (F)

862

B. Shemer and S. Belkin

element (HRE), which then drives expression of reporter genes (Tsai and O’Malley 1994). This concept, with mild variations, was successfully employed for diverse hormone receptors, including the human thyroid receptor (Li et al. 2008; Shiizaki et al. 2010), androgen receptor (Gaido et al. 1997; Eldridge et al. 2007; Michelini et al. 2008), glucocorticoid receptor (Wright and Gustafsson 1992), progesterone (Gaido et al. 1997; Klotz et al. 1997; Berg et al. 2000), and estrogen (Routledge and Sumpter 1996; Arnold et al. 1996; Gaido et al. 1997). Recently, a fungal-based estrogen bioreporter was also reported (Zutz et al. 2017), exhibiting improved performance with raw samples. In most of these cases, the inducible HRE is plasmid-borne, whereas the constitutive HR is either chromosomally integrated or is carried on a second plasmid. To address the issue of a potential imbalance in transcript copy number between the HR and the HRE, Miller et al. (2010) constructed a single-plasmid containing both the receptor element and the reporting element. This design was successfully applied for several hormone groups. An interesting approach for screening several hormone groups in a single assay was recently reported by Chamas et al. (2017a). Following sample separation via thin-layer chromatography, the plate containing the fractionated sample was immersed in a solution containing a modified Arxula adeninivorans bioreporter strain (Chamas et al. 2017b). This strain produces recombinant human progesterone, androgen, and estrogen receptors, with each response element coupled to a differently colored fluorescent reporter protein: cyan (CFP), green (GFP), and red (DsRed2). As a result, bands which contain compounds with progesterone, androgen, or estrogen activity will induce fluorescence of a different color in the A. adeninivorans overlay. However, this potentially multi-parallel detection test yielded poor reproducibility when analyzing environmental samples. A different approach, employing E. coli cells that express the human estrogen receptor (ERα) on the cell’s surface (Furst et al. 2017), was recently suggested. This method uses an electrochemical electrode, modified with a protein that binds to ERα only when a target ligand is present. The binding of a target estrogen to the HR causes attachment of the bacterial cells to the electrode, which in turn causes a quantifiable change in impedance. This methodology allowed the detection of subppb levels of estradiol and ppm levels of bisphenol A in a complex matrix, using very low sample volumes (10 μL).

Antibiotics The increasing use of antibiotics in the food industry for both preventive and therapeutic purposes has led to a rise in public concern regarding the presence of residues of these compounds in consumer products. In addition to the undesirable and uncontrolled exposure to pharmaceuticals, this may increase the emergence and spreading of antibiotic resistance. One of the earliest attempts at constructing a microbial biosensor for detecting antibiotic residues was reported by Chopra et al. (1990). E. coli cells were transformed with a plasmid in which the tetA gene, encoding a tetracycline efflux pump

34

Microbial Biosensors for the Detection of Organic Pollutants

863

which promotes the cell’s resistance to this antibiotic, was fused to an enzymatically active β-galactosidase segment. TetR, a repressor of tetA, is released upon exposure to tetracycline, which allows the transcription of tetA and thereby of lacZ. Korpela et al. (1998) employed a similar concept, by using the tetA-tetR complex, on a TN10 transposon, to push the transcription of the luxCDABE cassette of P. luminescens. The resulting strain specifically detected picomole amounts of several tetracyclines and was successfully employed to detect four tetracycline derivatives in poultry muscle tissue (Virolainen et al. 2008). A potential disadvantage of the TN10-controlled bioreporter reported by Korpela et al. (1998) lies in the fact that tetA transcription reduces intracellular tetracycline concentrations due to its efflux pump activity, thus mitigating the activity of the bioreporter. This was addressed by Bahl et al. (2005), who used the tetM gene of Enterococcus faecalis as a sensing element. This gene encodes a ribosomal protection protein, which allows the release of tetracycline from tetracycline-ribosome complexes and permits protein synthesis to occur. This mechanism ensures that tetracycline levels in the cell will not decrease and possibly even increase due to the release of tetracycline from tetracycline-ribosome complexes. By fusing tetM to gfp, the bioreporter successfully responded to concentrations ranging between 5 ng/mL and 16 μg/mL, a considerable improvement over previous antibiotic bioreporters. A bacterial biosensor for the detection of β-lactams was reported by Valtonen et al. (2002): an E. coli strain harboring a plasmid containing a fusion of the ampR regulatory element of Citrobacter freundii induced in the presence of β-lactams, to the luxCDABE cassette. The response toward various β-lactams was studied, and detection thresholds ranging from several ng to μg per mL were reported. The same strain was later used by Smolander et al. (2009) to detect novel substances with β-lactam activity in a high-throughput assay coupled to a support vector machine classifying algorithm. As antibiotic residues concentrations in food products or water samples are expected to be extremely low, lowering the detection threshold of microbial biosensors is not only desirable, but sometimes a requisite in the construction of a commercial biosensing platform. An interesting example was reported by Biran et al. (2011), who managed to increase the sensitivity of an E. coli based biosensor for the detection of DNA-damaging agents such nalidixic acid. The sensing-reporting elements of this reporter were the sulA gene promoter, involved in the cell’s SOS response, and the alkaline phosphatase-coding phoA gene. To increase its sensitivity toward the target analytes, several gene deletions were introduced in the host genome: rfaE (enhancing membrane permeability), umuD, and uvrA (inhibiting DNA damage repair mechanisms). Certain combinations of these deletion mutations caused a significant increase in the reporter’s sensitivity toward nalidixic acid. Melamed et al. (2012) analyzed the response pattern of a panel of microbial bioreporters, each responding to a different group of antibiotics, and defined a typical response pattern for each of the antibiotics groups tested. This allowed to successfully detect and categorize samples containing “unknown” antibiotics at varying concentrations. Further enhancement of this panel’s capabilities was obtained by altering the cell’s membrane permeability and efflux capacity, which

864

B. Shemer and S. Belkin

resulted in a lower detection threshold for several antibiotics (Melamed et al. 2014). One member of this panel, carrying a plasmid-borne recA::luxCDABE fusion, was later incorporated into a biosensor device for the detection of ciprofloxacin residues in food (Kao et al. 2018).

Pesticides and Other Agrochemicals Due to their extensive use in agriculture, pesticide residues, many of them highly persistent, are commonly found in soil, in surface and groundwaters, in plant and wildlife tissues, and sometimes in food products (Alavanja 2009). Coupled with the well documented acute and chronic health effects of some pesticides (Rosenstock et al. 1991; Alavanja et al. 2004; Sanborn et al. 2012), the need for cost-effective, easy, and rapid detection systems is obvious. A simple design for an organophosphate microbial biosensor was suggested two decades ago by Rainina et al. (1996). This system is based on an E. coli host cell, harboring the organophosphate hydrolase (OPH) gene of Pseudomonas diminuta. This gene encodes an enzyme which hydrolyzes p-nitrophenyl derivatives to p-nitrophenol, which is then degraded by the cell’s native metabolism. Electrode-based enzymatic biosensors based on this enzyme were studied extensively (Mulchandani et al. 1999, 2001a, b, c). While simple to use, this reporter exhibited relatively high detection thresholds for the target analytes (250 ppb for paraoxon), probably due to mass transport resistance across the periplasmic and cytoplasmic membranes. A similar E. coli-based design was applied by Mulchandani et al. (1998) and immobilized on a pH electrode. The resulting device detected 2 μM of paraoxon, methyl parathion, and diazinon and allowed multiple uses of the same electrode. An attempt to improve the performance of OPH-based biosensors was reported by Lei et al. (2005), who used P. putida, a natural p-nitrophenol degrader, as host and immobilized the cells on a carbon-paste electrode. By measuring the electrooxidization current created by the intermediates, detection threshold of ca. 0.3 ppb for all three tested organophosphates (paraoxon, methyl parathion, and parathion) was obtained, with good selectivity. Recently, a similar concept for the detection of p-nitrophenyl-substituted organophosphates was reported by Tang et al. (2014), making use of an E. coli strain genetically engineered to express OPH on its outer membrane. The cells were immobilized on a glass carbon electrode modified with ordered mesoporous carbons. The assay’s sensitivity was improved by using an OPH mutant with improved stability and enzymatic activity. This design allowed detection of 9 nM paraoxon, 10 nM parathion, and 15 nM methyl parathion. In a recent report, the construction of a microbial biosensor capable of directly responding to organophosphates by using the transcription regulator DmpR (Chong and Ching 2016) was described. This molecule is activated by various methyl phenols, and a mutated version of it also responds to 4-nitrophenol, a common hydrolysis product of organophosphates. The transcription of the dmpR promoter was coupled to mRFP1, resulting in red fluorescence in the presence of parathion,

34

Microbial Biosensors for the Detection of Organic Pollutants

865

methyl-parathion, and fenitrothion. Random mutagenesis was employed to enhance the sensitivity of the dmpR promoter, yielding two dmpR mutants with improved sensitivity. Lindane (hexachlorocyclohexane, HCH) is an organochlorine insecticide which is a major environmental contaminant (Phillips et al. 2005). It acts as a neurotoxin by blocking the GABA-gated chloride channel and reducing neuronal inhibition, which results in hyperexcitation of the central nervous system (Bloomquist 1993). The high toxicity and carcinogenicity in humans of lindane (Loomis et al. 2015), renders it an important target for development of rapid and sensitive detection systems. A microbial biosensor for the detection of lindane was reported by Anu Prathap et al. (2012). This bioreporter is based on an E. coli host with the Sphingomonas paucimobilis linA2 HCH dehydrochlorinase overexpressed using the T7 strong inducible promoter. This gene is involved in the initial steps of lindane biotransformation through a three-step dehydrochlorination, which leads to the release of hydrochloric acid (HCl). The authors used polyaniline as a pH transducer; upon protonation by the excreted HCl, its conductivity is increased by up to 10 orders of magnitude. By using pulsed amperometry, lindane concentrations as low as 2 ppt were detected. López Rodriguez et al. (2015) also used a lindane-degrading microbe as a platform for developing a microbial biosensor for this pesticide. Streptomyces strain M7 (SM7), isolated from a pesticide-contaminated soil, is capable of growing with lindane as the sole carbon source (Benimeli et al. 2007). By use of impedimetric analysis to measure the variation in metabolic activity in the presence of lindane, the authors were able to detect changes as low as 120 μg/L in lindane concentration in a liquid matrix. An attractive advantage of this development is the use of bacterial spores as the recognition element (López Rodriguez et al. 2014), adding to the robustness and durability of the assay. Hua et al. (2015) reported a successful E. coli-based assay for the detection of atrazine and its degradation product cyanuric acid. Two bioluminescent reporter strains were developed, one for both compounds and the other for cyanuric acid only, employing different Pseudomonas sp. pADP-1 atz gene promoters as the sensing elements. Cyanuric acid detection limits were reported to comply with WHO recommendations for this compound. Photosynthetic organisms, such as microalgae, have also been studied as biosensors of agrochemicals. A simple design consists of measuring the chlorophyll fluorescence at 682 nm, which increases upon exposure to toxicants that inhibit photosystem II (Merz et al. 1996). This design was applied with various hosts and herbicides, including atrazine, endrine, simazine, diuron, dinitro-ortho-cresol (DNOC), and isoproturon (Frense et al. 1998; Naessens et al. 2000; Védrine et al. 2003). As such sensors detect all PSII-targeting pesticides, their specificity is limited. This issue was addressed by Peña-Vázquez et al. (2009), who obtained simazine specificity by combining, in a solgel matrix, sensor strains with different sensitivities to this herbicide. This design follows the concept reported by Altamirano et al. (2004) for improving a reporter’s specificity toward TNT, reviewed earlier in this chapter.

866

B. Shemer and S. Belkin

Acetylcholinesterase (AChE) inhibition assays for the detection and quantification of organophosphates and carbamates, both inhibitors of this enzyme, are well documented (Villatte et al. 1998). It has been shown that insect AChE is more sensitive than the mammalian one (Villatte et al. 1998). However, purification of this enzyme is costly and time-consuming. To overcome this difficulty, Li et al. (2013) used S. cerevisiae to express the Drosophila melanogaster AChE on the cell’s surface. This strain was used as an immobilized platform, and AChE inhibition was measured in a commercially available colorimetric kit (Ellman et al. 1961), successfully and sensitively detecting organophosphates and carbamates.

Halogenated Organic Pollutants Halogenated organic compounds (HOCs) are members of a large group of organic substances used extensively in agriculture (herbicides, insecticides, and fungicides), in diverse industries (solvents, process components, and byproducts), and in water treatment (chlorination byproducts). Chlorinated and brominated polycyclic compounds have favorable attributes in terms of stability at high temperatures and hydraulic and electrical properties. Unfortunately, these attributes are also related to accumulation in fatty tissues, low environmental mobility, and low biodegradability for many of the members of this group (Pearson 1982). This has led to the large number of HOCs in the US Environmental Protection Agency’s “Priority Pollutant List” (U.S. Environmental Protection Agency 1972, 2014). While many HOCs are highly recalcitrant, some organisms are capable either partial or full mineralization of certain members of this group. For the most part, biodegradation processes of HOCs are initiated by the cleavage of the carbonhalogen bond by dehalogenases (Fetzner and Lingens 1994). One such enzyme, the Arthrobacter SU 4-chlorobenzoic acid (4CBA) dehalogenase, was employed by Rozen et al. (1999) to construct a bioreporter for the detection of 4CBA. The gene encoding this enzyme, fcbA, was fused to the A. fischeri luxCDABE genes and transformed via plasmid to an E. coli host. The resulting strain detected 60 mg/L of 4CBA with rather good specificity. The detection limits of this bioreporter were significantly improved (28 μM of 4CBA, equivalent to 4.3 mg/L) when a membrane-leaky mutant was used as host and immobilized in alginate beads (Köhler et al. 2000). A similar approach was used by Tauber et al. (2001) to design a bioreporter for detecting 2-chloropropionic acid (2-CPA). The promoter region of the DL-DEX dehalogenase of a Pseudomonas sp., encoding for an enzyme responsible for the dehalogenation of DL-2-haloalkanoic acids, was fused to the luxCDABE genes of P. luminescens and transformed via plasmid to an E. coli host. While exhibiting good dose-dependency, the sensitivity of the constructed bioreporter was low, with a detection threshold of 100 mg/L. Hutter et al. (1995) employed a different approach to detect halogenated hydrocarbons. This design used the innate ability of a Rhodococcus strain containing an

34

Microbial Biosensors for the Detection of Organic Pollutants

867

alkyl-halidohydrolase, an enzyme which removes halogen ions from halogenated hydrocarbons. Once free, these ions are quantified using ion sensitive potentiometric chloride or bromide electrodes. Using this method, the authors were able to detect 50 μg/L of chlorinated hydrocarbons and 10 μg/L of brominated hydrocarbons. The methodology was later improved by immobilizing the bacteria directly on the electrode’s surface (Peter et al. 1996). Another attempt to use microbial biosensors to detect halogenated hydrocarbons was reported by Han et al. (2001), who used Pseudomonas aeruginosa JI 104 to detect trichloroethylene (TCE). This strain, isolated from soil near a gasworks, is capable of full mineralization of TCE, during which the chloride ion is released. Quantification of TCE was performed by coupling the activity of this bacterium to a chloride ion electrode. A dose-dependent response was observed, with a detection threshold of 0.1 mg/L TCE. However, the device exhibited a very limited shelf life; activity was reduced by 50% after 3 days. More recently, an interesting approach was applied to detect TCE by immobilizing a P. putida strain to a surface of gold interdigitated microelectrode (Hnaien et al. 2011). This strain degrades TCE to glyoxylate and formate by a toluene dioxygenase (TOD) enzyme. The design consisted of two electrodes; the working electrode, onto which the wild-type strain was immobilized, and the reference electrode, on which was immobilized a TOD-defective strain. Upon exposure to the target analyte, formate and glyoxylate are formed in the working electrode but not in the reference electrode. Quantification is achieved by measuring the conductance differences between the electrodes. Using this design, a rapid, dosedependent response to TCE was observed, with a detection threshold of 0.07 μM. In addition, shelf life was shown to be superior to previous designs, maintaining 92% activity after 7 weeks of storage. Polychlorinated biphenyls (PCBs), banned from use in many countries several decades ago, are still prevalent environmental pollutants. Many studies reported the high toxicity of PCBs and their transformation products, leading to a great interest in the development of new detection techniques. A bioreporter for the detection of PCBs was first reported by Layton et al. (1998), who fused the bphA1 gene promoter region, encoding the subunit alpha of the biphenyl dioxygenase of the PCBdegrading strain Cupriavidus necator (formerly known as Ralstonia eutropha), to the A. fischeri luxCDABE genes. This fusion was transformed via plasmid to an E. coli host, and the resulting strain was able to detect 0.15 mg/L of 4-chlorobiphenyl and 1.5 mg/L of Aroclor. Another approach for PCBs detection was reported by Gavlasova et al. (2008), who immobilized a Pseudomonas sp. in a silica gel matrix. The catabolic process by which this bacterium degrades PCBs includes the formation of a yellow metabolite which absorbs light at 398 nm. By monitoring the formation of this metabolite, the authors were able to detect 2,3,40 -trichlorobiphenyl, 2,4,40 -trichlorobiphenyl, and 2,5,40 -trichlorobiphenyl with a detection threshold of 0.2–0.5 mg/L. The ability of this bioreporter to detect PCBs in mineral oil and soil originating from a landfill was also tested with satisfactory results.

868

B. Shemer and S. Belkin

Conclusions and Future Outlook As clearly emerges from the information provided in this chapter, a large variety of microorganisms have been molecularly engineered as live bioreporters of diverse organic compounds of environmental and health-related significance. Impressive ingenuity is displayed regarding the microbiological and molecular approaches employed to achieve this end, employing laboratory “workhorses” such as E. coli or S. cerevisiae, but also isolates from niche ecosystems, such as the fungus Y. lipolytica. With the ever-increasing availability of yet more sophisticated synthetic biology and computational tools, it is easy to predict that this trend will continue into the near future. Another facet of the topic that may be gleaned from what is portrayed in this review, is actually hinted at by what it does not include: in spite of the remarkable array of sensor systems described above, and of the spectrum of deleterious chemicals they can detect, almost none of them has been adopted by regulatory authorities as recommended monitoring tools. A part of this phenomenon may be explained by the necessary conservativism of environmental regulations, and the long timelines required for the introduction, acceptance, and commercialization of new methodologies. Just as important, however, is that many of the reports are satisfied with a description of a potentially successful sensor and its preliminary characterization, reporting a validation of a concept which is not yet suitable as a basis for a real application. One can only hope that in the future we will see more of the rigorous characterization efforts involved in the development of such an application, along with a parallel development of the hardware required for its implementation as a robust and commercially viable tool.

References Agency for Toxic Substances and Disease Registry (ATSDR) (1995) Toxicological profile for 2,4,6trinitrotoluene (TNT). U.S. Department of Health and Human Services, Public Health Service, Atlanta Agency for Toxic Substances and Disease Registry (ATSDR) (2004) Interaction profile for: benzene, toluene, ethylbenzene, and xylenes (BTEX). U.S. Department of Health and Human Services, Public Health Service, Atlanta Agency for Toxic Substances and Disease Registry (ATSDR) (2005) Toxicological profile for naphthalene, 1-methylnaphthalene, and 2-methylnaphthalene. U.S. Department of Health and Human Services, Public Health Service, Atlanta Agency for Toxic Substances and Disease Registry (ATSDR) (2016) Toxicological profile for dinitrotoluenes. U.S. Department of Health and Human Services, Public Health Service, Atlanta Alavanja MC (2009) Introduction: pesticides use and exposure, extensive worldwide. Rev Environ Health 24:303–310 Alavanja MC, Hoppin JA, Kamel F (2004) Health effects of chronic pesticide exposure: cancer and neurotoxicity. Annu Rev Public Health 25:155–197 Alkasrawi M, Nandakumar R, Margesin R et al (1999) A microbial biosensor based on Yarrowia lipolytica for the off-line determination of middle-chain alkanes. Biosens Bioelectron 14:723–727. https://doi.org/10.1016/S0956-5663(99)00046-9

34

Microbial Biosensors for the Detection of Organic Pollutants

869

Altamirano M, Garcıa-Villada L, Agrelo M et al (2004) A novel approach to improve specificity of algal biosensors using wild-type and resistant mutants: an application to detect TNT. Biosens Bioelectron 19:1319–1323 Anu Prathap MU, Chaurasia AK, Sawant SN, Apte SK (2012) Polyaniline-based highly sensitive microbial biosensor for selective detection of lindane. Anal Chem 84:6672–6678. https://doi. org/10.1021/ac301077d Applegate B, Kehrmeyer S, Sayler G (1998) A chromosomally based tod-luxCDABE wholecell reporter for benzene, toluene, ethybenzene, and xylene (BTEX) sensing. Appl Environ Microbiol 64:2730–2735 Arnold SF, Robinson MK, Notides AC et al (1996) A yeast estrogen screen for examining the relative exposure of cells to natural and xenoestrogens. Environ Health Perspect 104:544–548. https://doi.org/10.1289/ehp.96104544 Assinder SJ, Williams PA (1990) The TOL plasmids: determinants of the catabolism of toluene and the xylenes. Adv Microb Physiol 31:1–69. Elsevier Azuma K, Tanaka-Kagawa T, Jinno H (2018) Health risk assessment of inhalation exposure to longchain aliphatic hydrocarbons and aldehydes, TMB, MCH, and MIBK in indoor environments. Environ Health Perspect. Abstract presented at the International Society of Exposure Science and the International Society for Environmental Epidemiology (ISES-ISEE) 2018 Joint Annual Meeting Bahl MI, Hansen LH, Sørensen SJ (2005) Construction of an extended range whole-cell tetracycline biosensor by use of the tet(M) resistance gene. FEMS Microbiol Lett 253:201–205. https://doi. org/10.1016/j.femsle.2005.09.034 Belkin S, Yagur-Kroll S, Kabessa Y et al (2017) Remote detection of buried landmines using a bacterial sensor. Nat Biotechnol 35:308–310 Benimeli CS, Castro GR, Chaile AP, Amoroso MJ (2007) Lindane uptake and degradation by aquatic Streptomyces sp. strain M7. Int Biodeterior Biodegrad 59:148–155. https://doi.org/ 10.1016/j.ibiod.2006.07.014 Berg M, Undisz K, Thiericke R et al (2000) Miniaturization of a functional transcription assay in yeast (human progesterone receptor) in the 384- and 1536-well plate format. J Biomol Screen 5:71–76. https://doi.org/10.1177/108705710000500203 Biran A, Ben-Yoav H, Yagur-Kroll S et al (2011) Microbial genotoxicity bioreporters based on sulA activation. Anal Bioanal Chem 400:3013–3024 Bloomquist JR (1993) Toxicology, mode of action and target site-mediated resistance to insecticides acting on chloride channels. Comp Biochem Physiol C Pharmacol Toxicol Endocrinol 106:301–314. https://doi.org/10.1016/0742-8413(93)90138-B Boffetta P, Jourenkova N, Gustavsson P (1997) Cancer risk from occupational and environmental exposure to polycyclic aromatic hydrocarbons. Cancer Causes Control 8(3):444–472 Burlage R, Hooper S, Sayler G (1989) The TOL (pWW0) catabolic plasmid. Appl Environ Microbiol 55:1323 Burlage RS, Sayler GS, Larimer F (1990) Monitoring of naphthalene catabolism by bioluminescence with nah-lux transcriptional fusions. J Bacteriol 172:4749–4757 Burlage RS, Everman KR, Patek DR (1999) Method for detection of buried explosives using a biosensor. US Patent US5972638A Chamas A, Pham HTM, Jähne M et al (2017a) Separation and identification of hormone-active compounds using a combination of chromatographic separation and yeast-based reporter assay. Sci Total Environ 605–606:507–513. https://doi.org/10.1016/j.scitotenv.2017.06.077 Chamas A, Pham HTM, Jähne M et al (2017b) Simultaneous detection of three sex steroid hormone classes using a novel yeast-based biosensor. Biotechnol Bioeng 114:1539–1549. https://doi.org/ 10.1002/bit.26249 Cho JH, Lee DY, Lim WK, Shin HJ (2014) A recombinant Escherichia coli biosensor for detecting polycyclic aromatic hydrocarbons in gas and aqueous phases. Prep Biochem Biotechnol 44:849–860. https://doi.org/10.1080/10826068.2014.887577 Chong H, Ching CB (2016) Development of colorimetric-based whole-cell biosensor for organophosphorus compounds by engineering transcription regulator DmpR. ACS Synth Biol 5:1290–1298. https://doi.org/10.1021/acssynbio.6b00061

870

B. Shemer and S. Belkin

Chopra I, Hacker K, Misulovin Z, Rothstein DM (1990) Sensitive biological detection method for tetracyclines using a tetA-lacZ fusion system. Antimicrob Agents Chemother 34:111–116. https://doi.org/10.1128/AAC.34.1.111 Davidson ME, Harbaugh SV, Chushak YG et al (2012) Development of a 2, 4-dinitrotolueneresponsive synthetic riboswitch in E. coli cells. ACS Chem Biol 8:234–241 de las Heras A, de Lorenzo V (2011a) In situ detection of aromatic compounds with biosensor Pseudomonas putida cells preserved and delivered to soil in water-soluble gelatin capsules. Anal Bioanal Chem 400:1093–1104 de las Heras A, de Lorenzo V (2011b) Cooperative amino acid changes shift the response of the σ54-dependent regulator XylR from natural m-xylene towards xenobiotic 2, 4-dinitrotoluene. Mol Microbiol 79:1248–1259 De Las Heras A, Carreño CA, De Lorenzo V (2008) Stable implantation of orthogonal sensor circuits in Gram-negative bacteria for environmental release. Environ Microbiol 10:3305–3316 Eldridge ML, Sanseverino J, Layton AC et al (2007) Saccharomyces cerevisiae BLYAS, a new bioluminescent bioreporter for detection of androgenic compounds. Appl Environ Microbiol 73:6012–6018 Ellman GL, Courtney KD, Andres V Jr, Featherstone RM (1961) A new and rapid colorimetric determination of acetylcholinesterase activity. Biochem Pharmacol 7:88–95 Fetzner S, Lingens F (1994) Bacterial dehalogenases: biochemistry, genetics, and biotechnological applications. Microbiol Rev 58:641–685 Frense D, Müller A, Beckmann D (1998) Detection of environmental pollutants using optical biosensor with immobilized algae cells. Sens Actuators B Chem 51:256–260. https://doi.org/ 10.1016/S0925-4005(98)00203-2 Furst AL, Hoepker AC, Francis MB (2017) Quantifying hormone disruptors with an engineered bacterial biosensor. ACS Cent Sci 3:110–116. https://doi.org/10.1021/acscentsci.6b00322 Gaido KW, Leonard LS, Lovell S et al (1997) Evaluation of chemicals with endocrine modulating activity in a yeast-based steroid hormone receptor gene transcription assay. Toxicol Appl Pharmacol 143:205–212. https://doi.org/10.1006/taap.1996.8069 Garmendia J, De Las Heras A, Galvão TC, De Lorenzo V (2008) Tracing explosives in soil with transcriptional regulators of Pseudomonas putida evolved for responding to nitrotoluenes. Microb Biotechnol 1:236–246 Gavlasova P, Kuncova G, Kochankova L, Mackova M (2008) Whole cell biosensor for polychlorinated biphenyl analysis based on optical detection. Int Biodeterior Biodegrad 62:304–312. https://doi.org/10.1016/j.ibiod.2008.01.015 Han T-S, Kim Y-C, Sasaki S et al (2001) Microbial sensor for trichloroethylene determination. Anal Chim Acta 431:225–230. https://doi.org/10.1016/S0003-2670(00)01329-5 Hara A, Syutsubo K, Harayama S (2003) Alcanivorax which prevails in oil-contaminated seawater exhibits broad substrate specificity for alkane degradation. Environ Microbiol 5:746–753 Hnaien M, Lagarde F, Bausells J et al (2011) A new bacterial biosensor for trichloroethylene detection based on a three-dimensional carbon nanotubes bioarchitecture. Anal Bioanal Chem 400:1083–1092. https://doi.org/10.1007/s00216-010-4336-x Hua A, Gueuné H, Cregut M et al (2015) Development of a bacterial bioassay for atrazine and cyanuric acid detection. Front Microbiol 6:Article 211. https://doi.org/10.3389/fmicb.20 15.00211 Hutter W, Peter J, Swoboda H et al (1995) Development of a microbial bioassay for chlorinated and brominated hydrocarbons. Anal Chim Acta 306:237–241. https://doi.org/10.1016/0003-2670 (94)00679-G Kao W-C, Belkin S, Cheng J-Y (2018) Microbial biosensing of ciprofloxacin residues in food by a portable lens-free CCD-based analyzer. Anal Bioanal Chem 410:1257–1263 Kim J-W, Kim J-H, Tung S (2008) Nanoscale flagellar-motor based MEMS biosensor for explosive detection. IEEE, pp 630–632 King JMH, DiGrazia PM, Applegate B et al (1990) Rapid, sensitive bioluminescent reporter technology for naphthalene exposure and biodegradation. Science 249:778–781. https://doi. org/10.1126/science.249.4970.778

34

Microbial Biosensors for the Detection of Organic Pollutants

871

Klotz DM, Ladlie BL, Vonier PM et al (1997) o,p0 -DDT and its metabolites inhibit progesteronedependent responses in yeast and human cells. Mol Cell Endocrinol 129:63–71. https://doi.org/ 10.1016/S0303-7207(96)04041-5 Köhler S, Bachmann TT, Schmitt J et al (2000) Detection of 4-chlorobenzoate using immobilized recombinant Escherichia coli reporter strains. Sens Actuators B Chem 70:139–144. https://doi. org/10.1016/S0925-4005(00)00583-9 Korpela MT, Kurittu JS, Karvinen JT, Karp MT (1998) A recombinant Escherichia coli sensor strain for the detection of tetracyclines. Anal Chem 70:4457–4462. https://doi.org/10.1021/ ac980740e Kumari R, Tecon R, Beggah S et al (2011) Development of bioreporter assays for the detection of bioavailability of long-chain alkanes based on the marine bacterium Alcanivorax borkumensis strain SK2. Environ Microbiol 13:2808–2819. https://doi.org/10.1111/j.14622920.2011.02552.x Laffon B, Pásaro E, Valdiglesias V (2016) Effects of exposure to oil spills on human health: updated review. J Toxicol Environ Health B 19:105–128. https://doi.org/10.1080/10937404.2016. 1168730 Layton AC, Muccini M, Ghosh MM, Sayler GS (1998) Construction of a bioluminescent reporter strain to detect polychlorinated biphenyls. Appl Environ Microbiol 64:5023–5026 Lei Y, Mulchandani P, Wang J et al (2005) Highly sensitive and selective amperometric microbial biosensor for direct determination of p-nitrophenyl-substituted organophosphate nerve agents. Environ Sci Technol 39:8853–8857. https://doi.org/10.1021/es050720b Li J, Ma M, Wang Z (2008) A two-hybrid yeast assay to quantify the effects of xenobiotics on thyroid hormone-mediated gene expression. Environ Toxicol Chem 27:159–167 Li J, Ba Q, Yin J et al (2013) Surface display of recombinant Drosophila melanogaster acetylcholinesterase for detection of organic phosphorus and carbamate pesticides. PLoS One 8:e72986 Lönneborg R, Varga E, Brzezinski P (2012) Directed evolution of the transcriptional regulator DntR: isolation of mutants with improved DNT-response. PLoS One 7:e29994 Looger LL, Dwyer MA, Smith JJ, Hellinga HW (2003) Computational design of receptor and sensor proteins with novel functions. Nature 423:185–190 Loomis D, Guyton K, Grosse Y et al (2015) Carcinogenicity of lindane, DDT, and 2, 4dichlorophenoxyacetic acid. Lancet Oncol 16:891–892 López Rodriguez ML, Madrid RE, Giacomelli CE (2014) The optimization of the culture medium to design Streptomyces sp. M7 based impedimetric biosensors. Sens Actuators B Chem 193:230–237. https://doi.org/10.1016/j.snb.2013.11.066 López Rodriguez ML, Benimeli C, Madrid RE, Giacomelli CE (2015) A simple Streptomyces spore-based impedimetric biosensor to detect lindane pesticide. Sens Actuators B Chem 207:447–454. https://doi.org/10.1016/j.snb.2014.10.030 Luo Y, Guo W, Ngo HH et al (2014) A review on the occurrence of micropollutants in the aquatic environment and their fate and removal during wastewater treatment. Sci Total Environ 473–474:619–641. https://doi.org/10.1016/j.scitotenv.2013.12.065 MacDonald J, Lockwood J, McFee J, et al (2003) Alternatives for landmine detection. DTIC Document Melamed S, Lalush C, Elad T et al (2012) A bacterial reporter panel for the detection and classification of antibiotic substances. Microb Biotechnol 5:536–548. https://doi.org/10.1111/ j.1751-7915.2012.00333.x Melamed S, Naftaly S, Belkin S (2014) Improved detection of antibiotic compounds by bacterial reporter strains achieved by manipulations of membrane permeability and efflux capacity. Appl Microbiol Biotechnol 98:2267–2277 Merz D, Geyer M, Moss DA, Ache H-J (1996) Chlorophyll fluorescence biosensor for the detection of herbicides. Fresenius J Anal Chem 354:299–305. https://doi.org/10.1007/s0021663540299 Michelini E, Cevenini L, Mezzanotte L et al (2008) A sensitive recombinant cell-based bioluminescent assay for detection of androgen-like compounds. Nat Protoc 3:1895–1902 Miller CA, Tan X, Wilson M et al (2010) Single plasmids expressing human steroid hormone receptors and a reporter gene for use in yeast signaling assays. Plasmid 63:73–78. https://doi. org/10.1016/j.plasmid.2009.11.003

872

B. Shemer and S. Belkin

Minak-Bernero V, Bare RE, Haith CE, Grossman MJ (2004) Detection of alkanes, alcohols, and aldehydes using bioluminescence. Biotechnol Bioeng 87:170–177 Mulchandani A, Mulchandani P, Kaneva I, Chen W (1998) Biosensor for direct determination of organophosphate nerve agents using recombinant Escherichia coli with surface-expressed organophosphorus hydrolase. 1. Potentiometric microbial electrode. Anal Chem 70: 4140–4145. https://doi.org/10.1021/ac9805201 Mulchandani A, Mulchandani P, Chen W et al (1999) Amperometric thick-film strip electrodes for monitoring organophosphate nerve agents based on immobilized organophosphorus hydrolase. Anal Chem 71:2246–2249 Mulchandani A, Chen W, Mulchandani P et al (2001a) Biosensors for direct determination of organophosphate pesticides. Biosens Bioelectron 16:225–230 Mulchandani P, Chen W, Mulchandani A (2001b) Flow injection amperometric enzyme biosensor for direct determination of organophosphate nerve agents. Environ Sci Technol 35:2562–2565 Mulchandani P, Chen W, Mulchandani A et al (2001c) Amperometric microbial biosensor for direct determination of organophosphate pesticides using recombinant microorganism with surface expressed organophosphorus hydrolase. Biosens Bioelectron 16:433–437 Naessens M, Leclerc JC, Tran-Minh C (2000) Fiber optic biosensor using Chlorella vulgaris for determination of toxic compounds. Ecotoxicol Environ Saf 46:181–185. https://doi.org/ 10.1006/eesa.1999.1904 Paitan Y, Biran I, Shechter N et al (2004) Monitoring aromatic hydrocarbons by whole cell electrochemical biosensors. Anal Biochem 335:175–183. https://doi.org/10.1016/j.ab.2004. 08.032 Pearson CR (1982) Halogenated aromatics. In: Bock KJ, Daum KA, Merian E et al (eds) Anthropogenic compounds. Springer Berlin Heidelberg, Berlin/Heidelberg, pp 89–116 Peña-Vázquez E, Maneiro E, Pérez-Conde C et al (2009) Microalgae fiber optic biosensors for herbicide monitoring using sol–gel technology. Biosens Bioelectron 24:3538–3543. https://doi. org/10.1016/j.bios.2009.05.013 Peter J, Hutter W, Stöllnberger W, Hampel W (1996) Detection of chlorinated and brominated hydrocarbons by an ion sensitive whole cell biosensor. Biosens Bioelectron 11:1215–1219. https://doi.org/10.1016/0956-5663(96)88086-9 Phillips TM, Seech AG, Lee H, Trevors JT (2005) Biodegradation of hexachlorocyclohexane (HCH) by microorganisms. Biodegradation 16:363–392. https://doi.org/10.1007/s10532-0042413-6 Radhika V, Proikas-Cezanne T, Jayaraman M et al (2007) Chemical sensing of DNT by engineered olfactory yeast strain. Nat Chem Biol 3:325–330 Rainina EI, Efremenco EN, Varfolomeyev SD et al (1996) The development of a new biosensor based on recombinant E. coli for the direct detection of organophosphorus neurotoxins. Biosens Bioelectron 11:991–1000. https://doi.org/10.1016/0956-5663(96)87658-5 Ratajczak A, Geißdörfer W, Hillen W (1998) Expression of alkane hydroxylase from Acinetobacter sp. strain ADP1 is induced by a broad range of n-alkanes and requires the transcriptional activator AlkR. J Bacteriol 180:5822–5827 Reimer A, Yagur-Kroll S, Belkin S et al (2014) Escherchia coli ribose binding protein based bioreporters revisited. Sci Rep 4:5626 Ripp S, Nivens DE, Ahn Y et al (2000) Controlled field release of a bioluminescent genetically engineered microorganism for bioremediation process monitoring and control. Environ Sci Technol 34:846–853 Rosenstock L, Keifer M, Daniell WE et al (1991) Chronic central nervous system effects of acute organophosphate pesticide intoxication. Lancet 338:223–227 Routledge EJ, Sumpter JP (1996) Estrogenic activity of surfactants and some of their degradation products assessed using a recombinant yeast screen. Environ Toxicol Chem 15:241–248 Rozen Y, Nejidatl A, Gartemann K-H, Belkin S (1999) Specific detection of p-chlorobenzoic acid by Escherichia coli bearing a plasmid-borne fcbA0 ::lux fusion. Chemosphere 38:633–641. https://doi.org/10.1016/S0045-6535(98)00210-0

34

Microbial Biosensors for the Detection of Organic Pollutants

873

Sanborn M, Cole D, Sanin LH, Bassil K (2012) Systematic review of pesticide human health effects. Ontario College of Family Physicians. https://ocfp.on.ca/docs/pesticides-paper/2012systematic-review-of-pesticide.pdf Sevilla E, Yuste L, Rojo F (2015) Marine hydrocarbonoclastic bacteria as whole-cell biosensors for n-alkanes. Microb Biotechnol 8:693–706. https://doi.org/10.1111/1751-7915.12286 Shemer B, Palevsky N, Yagur-Kroll S, Belkin S (2015) Genetically engineered microorganisms for the detection of explosives’ residues. Front Microbiol 6:1175 Shemer B, Koshet O, Yagur-Kroll S, Belkin S (2017) Microbial bioreporters of trace explosives. Curr Opin Biotechnol 45:113–119 Shiizaki K, Asai S, Ebata S et al (2010) Establishment of yeast reporter assay systems to detect ligands of thyroid hormone receptors α and β. Toxicol In Vitro 24:638–644. https://doi.org/ 10.1016/j.tiv.2009.10.001 Smolander O-P, Ribeiro AS, Yli-Harja O, Karp M (2009) Identification of β-lactam antibiotics using bioluminescent Escherichia coli and a support vector machine classifier algorithm. Sens Actuators B Chem 141:604–609. https://doi.org/10.1016/j.snb.2009.06.019 Sticher P, Jaspers MC, Stemmler K et al (1997) Development and characterization of a whole-cell bioluminescent sensor for bioavailable middle-chain alkanes in contaminated groundwater samples. Appl Environ Microbiol 63:4053–4060 Sun Y, Zhao X, Zhang D et al (2017) New naphthalene whole-cell bioreporter for measuring and assessing naphthalene in polycyclic aromatic hydrocarbons contaminated site. Chemosphere 186:510–518. https://doi.org/10.1016/j.chemosphere.2017.08.027 Tan J, Kan N, Wang W et al (2015) Construction of 2, 4, 6-trinitrotoluene biosensors with novel sensing elements from Escherichia coli K-12 MG1655. Cell Biochem Biophys 72:417–428 Tang X, Zhang T, Liang B et al (2014) Sensitive electrochemical microbial biosensor for p-nitrophenylorganophosphates based on electrode modified with cell surface-displayed organophosphorus hydrolase and ordered mesopore carbons. Biosens Bioelectron 60:137–142. https://doi.org/10.1016/j.bios.2014.04.001 Tauber M, Rosen R, Belkin S (2001) Whole-cell biodetection of halogenated organic acids. Talanta 55:959–964 Tizzard AC, Lloyd-Jones G (2007) Bacterial oxygenases: in vivo enzyme biosensors for organic pollutants. Biosens Bioelectron 22:2400–2407. https://doi.org/10.1016/j.bios.2006.08.027 Tsai M, O’Malley BW (1994) Molecular mechanisms of action of steroid/thyroid receptor superfamily members. Annu Rev Biochem 63:451–486 U.S. Environmental Protection Agency (1972) Clean Water Act, 33 U.S.C §§ 1251et seq. https:// www.epa.gov/sites/production/files/2015-09/documents/priority-pollutant-list-epa.pdf. Accessed 21 Jan 2019 United State Environmental Protection Agency (2014) https://www.epa.gov/sites/production/files/ 2015-09/documents/priority-pollutant-list-epa.pdf Valtonen SJ, Kurittu JS, Karp MT (2002) A luminescent Escherichia coli biosensor for the high throughput detection of β-lactams. J Biomol Screen 7:127–134 Védrine C, Leclerc J-C, Durrieu C, Tran-Minh C (2003) Optical whole-cell biosensor using Chlorella vulgaris designed for monitoring herbicides. Biosens Bioelectron 18:457–463. https://doi.org/10.1016/S0956-5663(02)00157-4 Villatte F, Marcel V, Estrada-Mondaca S, Fournier D (1998) Engineering sensitive acetylcholinesterase for detection of organophosphate and carbamate insecticides. Biosens Bioelectron 13:157–164 Virolainen NE, Pikkemaat MG, Elferink JWA, Karp MT (2008) Rapid detection of tetracyclines and their 4-epimer derivatives from poultry meat with bioluminescent biosensor bacteria. J Agric Food Chem 56:11065–11070. https://doi.org/10.1021/jf801797z Wilbur S, Wohlers D, Paikoff S et al (2008) ATSDR evaluation of health effects of benzene and relevance to public health. Toxicol Ind Health 24:263–398. https://doi.org/10.1177/ 0748233708090910

874

B. Shemer and S. Belkin

Willardson BM, Wilkins JF, Rand TA et al (1998) Development and testing of a bacterial biosensor for toluene-based environmental contaminants. Appl Environ Microbiol 64:1006–1012 Worsey MJ, Williams PA (1975) Metabolism of toluene and xylenes by Pseudomonas putida (arvilla) mt-2: evidence for a new function of the TOL plasmid. J Bacteriol 124:7–13 Wright A, Gustafsson J-A (1992) Glucocorticoid-specific gene activation by the intact human glucocorticoid receptor expressed in yeast. Glucocorticoid specificity depends on low level receptor expression. J Biol Chem 267:11191–11195 Yagur-Kroll S, Lalush C, Rosen R et al (2014) Escherichia coli bioreporters for the detection of 2, 4-dinitrotoluene and 2, 4, 6-trinitrotoluene. Appl Microbiol Biotechnol 98:885–895 Yagur-Kroll S, Amiel E, Rosen R, Belkin S (2015) Detection of 2,4-dinitrotoluene and 2,4,6trinitrotoluene by an Escherichia coli bioreporter: performance enhancement by directed evolution. Appl Microbiol Biotechnol 99:7177–7188. https://doi.org/10.1007/s00253-015-6607-0 Zeinoddini M, Khajeh K, Behzadian F et al (2010) Design and characterization of an aequorinbased bacterial biosensor for detection of toluene and related compounds. Photochem Photobiol 86:1071–1075 Zhang D, He Y, Wang Y et al (2012) Whole-cell bacterial bioreporter for actively searching and sensing of alkanes and oil spills. Microb Biotechnol 5:87–97. https://doi.org/10.1111/j.17517915.2011.00301.x Zutz C, Wagener K, Yankova D et al (2017) A robust high-throughput fungal biosensor assay for the detection of estrogen activity. Steroids 126:57–65

Microbial Fuel Cells, Concept, and Applications

35

Carlo Santoro, Mike Brown, Iwona Gajda, John Greenman, Oluwatosin Obata, Maria José Salar García, Pavlina Theodosiou, Alexis Walter, Jonathan Winfield, Jiseon You, and Ioannis Ieropoulos

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Bioelectrochemical Systems and Microbial Fuel Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A Range of Organics to be Degraded . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Microbial Fuel Cell Main Components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Anode Materials and Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cathode Materials and Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Membrane and Separator Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MFC Designs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Scaling-Up for Energy Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Microbial Fuel Cell for Wastewater Treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Microbial Fuel Cell and Practical Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Microbial Fuel Cell as Biosensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sensors for BOD Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sensors for Toxicity Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sensors for Microbial Activity Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Microbial Fuel Cells (MFCs) for Monitoring Bio-Corrosion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Other Sensor Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

876 878 878 879 881 881 882 883 885 886 888 889 892 893 896 898 899 900 902

Abstract

The first published report of microbial fuel cells (MFCs) was over 100 years ago, yet it is only recently that interest in the technology has grown exponentially with the discovery that bacteria can transfer electrons to the anode without the need for external mediators. Diverse bioelectrochemical technologies have since been C. Santoro · M. Brown · I. Gajda · J. Greenman · O. Obata · M. J. S. García · P. Theodosiou · A. Walter · J. Winfield · J. You · I. Ieropoulos (*) Bristol BioEnergy Centre, Bristol Robotics Laboratory, University of the West of England, Bristol, UK e-mail: [email protected] © Springer Nature Switzerland AG 2022 G. Thouand (ed.), Handbook of Cell Biosensors, https://doi.org/10.1007/978-3-030-23217-7_93

875

876

C. Santoro et al.

developed. Microbial fuel cells have captured the attention of scientists due to the simultaneous removal of organics and pollutants and generation of electricity. Therefore, the MFC technology can become an integrated part of wastewater treatment as a renewable power system for low power consuming devices or even for real-time biosensing. In this work, a brief story of microbial fuel cells is presented followed by the description of existing bioelectrochemical systems. The diverse range of organic compounds treated in MFCs is presented followed by the description of the main MFC components (anode, cathode, and separator) their development and optimisation. Finally, the implementation of the technology for wastewater treatment and practical implementations are discussed. A final detailed part is dedicated to the utilisation of bioelectrochemical systems for biosensing.

Introduction Microbial fuel cell (MFC) is a fascinating bioelectrochemical technology that embraces the most diverse disciplines ranging from physics, chemistry, electrochemistry, biology/microbiology, and engineering. The addition of biology/microbiology to the already-complex electrochemistry makes the microbial fuel cell even more complicated to fully understand, especially when microorganisms are involved (Bennetto 1990; Santoro et al. 2017). In contrast to “classic” fuel cells, MFCs have at least one of the electrodes (usually the anode) in which the reaction (oxidation or reduction) is carried out by “electroactive” microorganisms. At the anode, electroactive bacteria are capable of using organic compounds and, through their respiration, are able to release electrons onto the anode electrode. The extracellular electron transfer (EET) mechanism from the bacteria to the electrode can occur following direct or indirect pathways (Kumar et al. 2017). The direct electron transfer can occur through the outer-membrane ctype cytochromes or through conductive pili or pilon-like structures known as microbial nanowires (Kumar et al. 2017). Indirect EET can occur due to the presence of catabolites or mediators that are (i) redox molecules self-excreted by the microorganisms, (ii) added externally to favour electron transfer, or (iii) naturally occurring in the solution. These molecules are also known as electron shuttles (Kumar et al. 2017). Another important difference between biological and chemical fuel cells is the operating temperature. In “classic” fuel cells, the temperature is increased in order to enhance the reaction kinetics on both anode and cathode. MFCs can operate at lower temperatures, required by the mesophilic microorganisms for survival. Despite some cases in which extremophilic bacteria thrive at higher temperature, MFCs tend to operate in temperatures varying between 15
C and 45
C with lower temperatures affecting negatively the bacterial metabolism (Larrosa-Guerrero et al. 2010). Generally, the majority of the MFC experiments is carried out at room temperature (20
C) or held stable at 30
C.

35

Microbial Fuel Cells, Concept, and Applications

877

A further difference between chemical and biological fuel cells exists in the operating pH levels. In fact, extreme pH conditions (acidic or alkaline) are preferred, because the cathodic oxygen reduction reaction (ORR) utilises H+ and OH as reactants for completing the ORR (Rojas-Carbonell et al. 2018). Once again, due to the presence of live microorganisms, with the exclusion of extremophilic bacteria, MFCs operate at (circum)neutral pH to preserve the electroactive bacteria at the anode. Typical pH ranges between 5 and 9, lower or higher pH levels are generally not advised, and usually if the pH is extreme, buffers are used to maintain the pH close to neutral levels (Logan et al. 2006). Finally, the fuel utilised in both the anode and cathode half-cells is another differentiator between chemical and biological fuel cells. “Classic” fuel cells usually operate with well-refined, processed, and purified anodic fuels (e.g., hydrogen, methanol, etc.). Moreover, the fuel is composed of low molecular weight substances. Electroactive biofilms on the other hand have been shown to be able to degrade more complex organic molecules, single substrates (Pandey et al. 2016), or multiple substrates (Pandey et al. 2016) like in the case of municipal or industrial wastewater (Pant et al. 2010a). Concerning the reaction occurring at the cathode, in chemical fuel cells, oxygen is the main oxidant used due to its abundance, high redox potential, and negligible weight. Consequently, the majority of MFC technologies are built considering the cathode ORR (Wang et al. 2014). However, if the main goal is not to generate useful electricity, other oxidants can be reduced at the cathode. This is of extreme interest if nitrate or nitrite needs to be reduced to nitrogen gas at the cathode and so enhancing the denitrification process (Virdis et al. 2010). Also sulfate has been used successfully as an oxidant at the cathode (Ucar et al. 2017). Several heavy metals (e.g., Hg, Cu, Cr, V, U, etc.) have also been reduced at the cathode (Ucar et al. 2017) with the goal of minimising the presence of these compounds into water bodies and reducing their toxicity. Finally, as the biological kinetic reactions are much slower compared to the chemical kinetics, the electricity production by MFCs is lower (4–6 orders of magnitude) compared to chemical fuel cells. Therefore, the method of harvesting electricity from MFCs and its use for practical applications has become a fascinating challenge to overcome. Particularly, specific power management systems (PMSs) are developed for harvesting low electricity output from MFCs (Wang et al. 2015). Practically PMSs rely on the combination of the MFCs with external supercapacitors or batteries capable of storing the energy and delivering it when needed, also known as intermittent behaviour (Melhuish et al. 2006; Ieropoulos et al. 2005a). Practical applications reported in the literature will be discussed further in the following sections. This book chapter intends to consider several important topics related with the microbial fuel cell technology. The discussion will touch briefly upon the history of bioelectrochemistry and microbial fuel cells moving then to the description of different bioelectrochemical systems involving the production of hydrogen, the reduction of carbon dioxide, and desalination. A brief overview of the organic

878

C. Santoro et al.

compounds used as anodic fuel is also presented. The main components of the MFCs, anode, cathode, and separators are briefly presented, and the improvements are also discussed. Different MFC designs are presented, and the advantages and disadvantages are examined. Microbial fuel cells as a technology for wastewater treatment and operation of low power devices is presented. Finally, emphasis is given to the potential of MFCs for biosensing of BOD and heavy metals, microbial activity, microbial corrosion, and other applications.

History The earliest discovery linking biology and electrical energy dates back to the eighteenth century, when Luigi Galvani observed for the first time the bioelectrical properties of animal tissue. This inspired further developments of electrolytic research by Alessandro Volta and the invention of the electrochemical pile as the first battery. Historically, the earliest report of bacteria-generating electricity dates to 1911, when Michael C. Potter discovered that “the disintegration of organic compounds by microorganisms is accompanied by the liberation of electrical energy” (Potter 1911), observing a current flow between two electrodes emerged in Escherichia coli and Saccharomyces cultures. In 1931 Barnett Cohen demonstrated the first stack of multiple MFC units and obtained a voltage of 35 V (Cohen 1931). In the 1960s, the electrochemical abilities of microbes were further investigated using E. coli (Davis and Yarbrough 1962) and photosynthetic bacteria (Berk and Canfield 1964) when NASA investigated bioelectricity as a means to recycling human waste in space missions. In the 1980s, Peter Bennetto developed mediator-based MFCs (Thurston et al. 1985) and designed an analytical type of bioreactor (Bennetto 1990), which was commercially available. However, only after year 2000, there have been both a significant progress and increased interest in the field, linking the MFCs with wastewater treatment, energy production, and biosensing by exploring both fundamental knowledge and technology applicability.

Bioelectrochemical Systems and Microbial Fuel Cells Bioelectrochemical systems (BESs) are a large technological platform having in common the electroactive biofilm performing the oxidation of organics at the anode (Logan et al. 2006). Microbial fuel cells (MFCs) are BESs that degrade organics and simultaneously generate electricity (Fig. 1a). This type of BES is the most investigated due to the possibility of being integrated into wastewater treatment and/or producing valuable electricity (Santoro et al. 2017). Another BES is microbial electrolysis cell (MEC), which unlike the MFC, consumes electricity rather than producing it. The advantage of an MEC, other than simultaneously oxidising organics, is the production of hydrogen on the cathode compartment (Logan et al. 2008). This occurs when utilising the potential from the anode and an additional power source, which enables the hydrogen evolution (Fig. 1b). The extra potential needed in MEC for generating hydrogen is lower compared to the one needed in a “classic” electrolysis cell (Logan et al. 2008).

35

a

Microbial Fuel Cells, Concept, and Applications

e-

Rext

e-

b

e-

e-

Organic molecules

H2O OH-

2eH 2 O2 HO2-

OXIDATION

H+

4e-

2eREDUCTION

CO2

O2

Smaller molecules

e-

POWER SOURCE

C A T H O D E

e-

e-

e-

A N O D E

Organic molecules

OXIDATION

H+ CO2

M E M B R A N E

Smaller molecules

CO CH4 CH3OH CH2CH2 CH3CH2OH CH2O HCOOH H2C2O4 etc.

REDUCTION

POWER SOURCE

e-

A N O D E

H2 OH-

Organic molecules

M E M B R A N REDUCTION E

OXIDATION

H+ CO2

H2O

Smaller molecules

d

C A T H O D E

CO2

e-

e-

e-

A N O D E

c

879

e-

Rext AEM

eOrganic molecules

A N O D E

Cl-

ClOXIDATION

e-

CEM Na+

Na+

H+

e-

2eH 2O2 HO2-

H2O OH-

4e-

2eCl-

CO2 Smaller molecules

C A T H O D E

Na+

REDUCTION

O2

C A T H O D E

SALT WATER

Fig. 1 Schematic of (a) microbial fuel cell, (b) microbial electrolysis cell, (c) microbial electrochemical cell, and (d) microbial desalination cell

In general, other compounds can be synthesised at the cathode of BES in the so-called microbial electrochemical cells (MECs) (Rabaey and Rozendal 2010). In these MECs, external power is needed, and the cathode containing inorganic catalysts or specific bacteria is polarised to certain negative values in which the reaction takes place (Harnisch and Schröder 2010). Acetate, formate, and alcohols are of extreme interest especially when considering that they can be produced by the reduction of carbon dioxide (Roy et al. 2016; Fig. 1c). It was shown that also carbon monoxide and methane could be formed as the product of CO2 electroreduction (Bajracharya et al. 2017; Fig. 1c). Microbial desalination cell (MDC) is also an interesting BES, firstly introduced in 2009 (Cao et al. 2009), in which organics are degraded, electricity is produced, and the salinity content decreased into the desalination chamber as shown in Fig. 1d.

A Range of Organics to be Degraded Like all other living organisms, electroactive microorganisms in microbial fuel cells require energy for reproduction, growth, development, and maintenance. Substrates (feedstock) provide MFC bacteria with carbon energy, electrons, and nutrients

880

C. Santoro et al.

including various salts and amino acids. In the MFC context, a substrate can be considered in the following aspects: (i) target waste to treat, (ii) fuel for electrical energy generation, (iii) material for resource recovery, and (iv) biosensing. MFC technology is capable of using organic waste as a substrate; consequently various types of waste both solid and liquid forms have been tested. These include a wide range of waste from common wastewater types such as domestic (Corbella and Puigagut 2018), dairy (Luo et al. 2017), or food processing wastewater (Cecconet et al. 2018) to “not-so-easy to clean” types of wastewater such as landfill leachate (Greenman et al. 2009), mining wastewater (Ni et al. 2016) and pharmaceutical wastewater (Amari and Boshrouyeh Ghandashtani 2019). Solid wastes such as municipal solid waste (Chiu et al. 2016) and fruit and vegetable wastes have also been investigated (Venkata Mohan et al. 2010). MFCs are often compared to anaerobic digestion technology since both technologies focus on energy production from waste/wastewater using anaerobic microbial metabolisms (Pham et al. 2006). Table 1 summarises relevant studies investigating organic waste for MFC substrate in terms of waste treatment system performance and specifications.

Table 1 Waste or wastewater treatment performance of MFCs Working volume

Treatment HRT efficiency

Wastewater Domestic wastewater

(L) 1

(h) 24

Dairy industrial wastewater Food processing wastewater Landfill leachate Mining wastewater Pharmaceutical wastewater

0.43

Municipal solid waste Composite vegetable waste Potato pulp

% COD 83

Treatment capacity mgCOD Ld1 Ld1 1 517

10

83

1

2,490

0.028

240

97

0.003

35

1.043 Wm2 Luo et al. (2017)

0.343

400

–

0.021

–

1.7 Wm3

0.033

768

–

0.001

–

0.32 Wm3

30

93

0.08

558

20.5 Wm3

480

–

0.2

–

1.818 Wm2

0.43

96

63

0.11

441

0.934 Wm3

0.025

720

55

0.001

217

3.21 Wm3

0.1

4

PMAX (Wm3; References Wm2) 0.346 Wm3 Corbella and Puigagut (2018) Cecconet et al. 27 Wm3 (2018)

Sonawane et al. (2017) Ni et al. (2016) Amari and Boshrouyeh Ghandashtani (2019) Chiu et al. (2016) Venkata Mohan et al. (2010) Tian et al. (2017)

35

Microbial Fuel Cells, Concept, and Applications

881

Comprehensive reviews on waste and wastewater used in MFCs can be found in the literature (Pandey et al. 2016). On the other hand, the substrate can be seen as a fuel source for the MFCs. Intensive studies on the relationship between substrate properties and MFC power output have been carried out. For this purpose of studies, two types of substrates are used: simple/defined feedstock and undefined feedstock. Pant et al. reported that simple compounds such as acetate and butyrate are easier to degrade in MFCs, whereas complex substrates are favoured by diverse electroactive bacterial communities (Pant et al. 2010a, 2016). In addition to substrate types, physical or chemical properties of substrates also have a great effect on MFC power production. The relationship between substrate concentration and MFC current generation usually follows saturation kinetics if there are no other limitations for the anodic biofilm to function (You et al. 2018). Therefore, increasing substrate concentration leads to higher power output up to a certain (saturation) level.

Microbial Fuel Cell Main Components Anode Materials and Development The anode electrode is of great importance for the bacterial attachment and the biofilm development. As mentioned above, the electroactive bacteria are core to the MFC technology, and their activity needs to be supported (Logan et al. 2019). An effective selection of the anode electrode material is necessary for supporting the oxidation reaction and therefore the entire MFC performance (Guo et al. 2015). Anode electrodes need to have specific characteristics for being suitable for application in MFCs (Guo et al. 2015). The anode materials need to (i) have electrical conductivity, (ii) possess adequate mechanical and chemical properties, (iii) possess high surface area in order to enhance the biotic/abiotic interface, (iv) have anode geometry and structure that ensures perfusible conditions, and (v) have to be environmentally friendly and biocompatible for a better interaction between bacteria-electrode. The cost is an important parameter to take into account especially considering scaling up and large-scale applications. Carbonaceous materials are suitable to work as anode electrode and have been by far the most reported in the literature (Mustakeem 2015). Carbon felt, carbon cloth, carbon paper, carbon veil, carbon brush, carbon mesh, graphite plate, and granulated carbon are the most adopted anode materials in MFCs that are also widely available (Wei et al. 2011). Carbon rod, glassy carbon, granular activated carbon, reticulated vitreous carbon, and carbonised cardboard have also been successfully employed in MFCs as shown in Fig. 2. The surface of these carbon-based materials can be modified with functional groups in order to enhance the hydrophilicity of the surface and therefore improve the bacterial attachment and decrease start-up time (Guo et al. 2013). Recently, metallic materials have also been identified as suitable for MFC anode. Much effort was spent toward stainless steel (SS) because of its lower cost compared to carbonaceous materials and therefore more suitable for large-scale applications

882

C. Santoro et al.

Fig. 2 Anode electrode materials used for MFCs: (a) granular graphite, (b) granular activated carbon, (c) reticulated vitrified carbon, (d) carbon felt, (e) carbon brush, (f) carbon paper, (g) graphite plate, (h) carbon cloth, (i) carbon mesh, and (j) carbon veil. (Figure 2 a–i reprinted and rearranged from Wei J, Liang P, Huang X (2011). Recent progress in electrodes for microbial fuel cells. Bioresour Technol 102:9335–9344, with permission of Elsevier)

(Pocaznoi et al. 2012; Fig. 2e). SS shapes used were flat plate, mesh, and scrubber. The main drawback is the lower surface area compared to carbon-based materials and the possibility of corrosion if the SS grade is not adequate (Pocaznoi et al. 2012). Other metals such as copper, silver, nickel, titanium, and molybdenum have shown interesting initial results when employed as anode materials (Baudler et al. 2015).

Cathode Materials and Development As mentioned above, due to the unfavourable pH environment, the ORR is sluggish, and therefore a catalyst is needed for improving the reaction. Two types of cathode are used in MFCs: (i) fully coated hydrophilic and (ii) air-breathing open to air (Wang et al. 2017). The first configuration is used for double-chamber MFCs in which the cathode is fully submerged into the cathode chamber. The second configuration instead is used in membraneless or membrane-based, open to air MFCs. In the second configuration, the cathode structure is designed for enhancing the threephase interface (TPI) in which gas, liquid, and solid are present. Oxygen in the gas phase is preferred to the dissolved oxygen since it is present in higher concentrations. Three main catalysts are used for enhancing ORR kinetics and can be classified in (i) platinum group metal (PGM) materials, (ii) carbonaceous-based materials, and (iii) platinum group metal-free (PGM-free) materials containing transition metals such as Fe, Ni, Mn, or Co (Yuan et al. 2016). Platinum was the most used catalyst especially at the beginning, and it was inherited by the more mature technology of hydrogen/air fuel cells (Wang et al. 2014). As MFCs is a low-cost technology, the extensive use of platinum is

35

Microbial Fuel Cells, Concept, and Applications

883

infeasible due to its high cost. Moreover, platinum is oxophilic and therefore prone to pollution in the presence of anions (Santoro et al. 2016a). High surface area carbonaceous-based materials have been found to be a good alternative catalyst for ORR. Several materials such as carbon black, graphene, activated carbon, carbon nanotubes, and carbon nanofibers were successfully tested as cathode catalysts (Antolini 2015). Carbonaceous materials enhance the electrical conductivity of the cathode structure, thereby improving the output performance. Moreover, they possess high surface area that is beneficial for enhancing the sluggish ORR in neutral media. Activated carbon mixed with polytetrafluorethylene (PTFE) and pressed over a current collector (carbon-based or metal-based) is nowadays the most used cathode in MFC operations (Zhang et al. 2009). The best power output was achieved with the use of PGM-free catalysts. These materials can be prepared through (i) impregnating heterocyclic macrocycle organic compounds (e.g., porphyrins and phthalocyanines) containing transitional metals on a carbonaceous material (Costa de Oliveira et al. 2017) or (ii) pyrolysing a mixture containing a metal salt and a nitrogen-rich organic precursor (Kodali et al. 2017a; Santoro et al. 2018). It was also shown that Fe containing catalysts outperformed systems working with Co (second best performing), Ni, and Mn (Rojas-Carbonell et al. 2017), indicating that the utilisation of Fe is the most appropriate for higher power output levels (Kodali et al. 2017b).

Membrane and Separator Materials Along with anode and cathode materials, the separator is also a critical factor in MFC design, since both the material and the size of the separator strongly affect the power output. Its main functions are (i) to prevent physical contact between electrodes resulting in short circuit, (ii) to minimise the substrate losses caused by crossover from the anode to the cathode, and (iii) to reduce oxygen transfer from the cathode to the anode. The separator plays an important role in MFCs since it has a big influence on the internal resistance of these bioelectrochemical devices and therefore on the power output of the system. A good MFC separator should meet the following criteria: (i) high ionic conductivity and low oxygen permeability, (ii) low biofouling and good long-term stability, (iii) low intrinsic resistance, and (iv) low cost (Hernandez-Fernandez et al. 2015). One of the critical factors for the implementation of this technology in real applications is the cost of the separator. Commercial polymer-based membranes such as Nafion are commonly used as separators in MFCs due to their high proton conductivity; however, their high cost, oxygen permeability, and tendency to biofouling, limit technological scale-up. Alternative MFC configurations involve a membraneless architecture design, with the main drawback being oxygen transfer from the cathode to the anode (Liu and Logan 2004). So far the materials used as separator in MFCs can be grouped into (i) ion exchange membranes, (ii) bridge salts, and (iii) porous membranes as shown in Fig. 3.

884

C. Santoro et al. CLASSIFICATION

Ion Exchange Membranes (IEMs)

CEMs Cation Exchange Membrane

AEMs Anion Exchange Membrane

Nafion SPEEK PSEBS Ultrex CMI 7000 Hyflon Zirfon etc

AMI 7001 QUAPPO Etc.

BPMs Bipolar Exchange Membrane

CEM and AEM mounted together

Salt Bridge

Porous Membranes

Microporous Filtration Membranes

Course-pore filter materials

Microfiltration membranes (MFMs) Ultrafiltration membranes (UFMs) etc.

J-cloth Non-woven Earthen nylon cloth (NWC) pot glass fiber filter

Bio-degradable membranes: Natural rubber

Fig. 3 Classification of separators for MFC applications

Ion exchange membranes are the most common separators used in MFCs and can be grouped according to their selectivity to transfer positively or negatively charged ions into (i) cation exchange membranes (CEMs), (ii) anion exchange membranes (AEMs), and (iii) bipolar membranes (BPMs) (Daud et al. 2015). Among the CEMs, Nafion is the most widespread commercial membrane due to its high cation conductivity (Oh and Logan 2006). A less expensive alternative is Ultrex CMI 7000 that exhibits higher ohmic resistance but has similar cation conductivity (Harnisch et al. 2008). Other commercial membranes are Hyflon or Zirfon; the former exhibits higher ohmic resistance than Nafion (Ieropoulos et al. 2010a). Regarding Zirfon, compared with Nafion, this material shows lower ohmic resistance but higher oxygen permeability (Pant et al. 2010b). The main limitation of these commercial CEMs is the high selectivity to other cations (e.g., Na+, K+, or Ca+) under neutral conditions. This leads to unwanted accumulation of protons in the anodic chamber, leading to the cathode alkalinisation that can cause salt precipitation and lower the cathode activity. CEMs also are prone to biofouling, which limits their large-scale implementation (Harnisch and Schröder 2009). The use of low-cost polymer matrix such as sulfonated polyether ether ketones (SPEEK) or sulfonated polystyrene-ethylene-butylene-polystyrene (SPSEBS) has also been reported as alternative materials to commercial CEMs, but their low ionic conductivity significantly reduces the utilisation in MFCs (Ghasemi et al. 2013). Ionic liquid-based membranes have been successfully used as MFC separator as an alternative to Nafion (Salar-García et al. 2015, 2016). Alternative ion exchange-type membranes are AEMs, which allow the anion transfer (OH, Cl, etc.) reducing the pH splitting between the chambers allowing MFCs to reach higher values of both power output and coulombic efficiency levels (Varcoe et al. 2014). Regarding bipolar membranes (BPMs), they are composed of two monopolar membranes mounted together allowing the flux of protons and hydroxide ions through their structure. Despite the acidification of the anodic chamber being limited, the internal resistance is increased compared with both CEMs and AEMs (Harnisch et al. 2008).

35

Microbial Fuel Cells, Concept, and Applications

885

Salt bridges is a low-cost alternative to IEMs in which the ions are transferred between the chambers through an electrolyte solution containing potassium chloride and phosphate buffer and/or agar usually in a glass tube. Salt bridges allow MFCs to reach high values of coulombic efficiency caused by the low oxygen diffusion from the cathodic to anodic chamber; however, the ohmic resistance is very high, and the power output is negatively affected (Daud et al. 2015). Porous size-selective membranes use a separation mechanism that is based on the particle size instead of its charge, and they can be grouped into microporous microfiltration (MF) membranes and ultrafiltration (UF) membranes or pore-filter materials. MF membranes have similar ohmic resistance to Nafion despite being able to reduce both the pH splitting and biofouling. The power output levels of these systems are similar to those working with Nafion as a separator but reaching higher values of COD removal (Huang et al. 2017). The term pore-filter separator covers different kinds of permeable materials such as fabric (J-cloths, glass fiber, nylon mesh, etc.), ceramic materials (earthen, terracotta, earthenware, and clayware), natural rubber, or cellulose, among others. The cost of IEMs and microporous membranes is significantly lower and their proton permeability higher, which favour the pH balance. However, the oxygen transfer as well as the substrate crossover is higher due to the porous size, which adversally affects the power output by the system (Daud et al. 2015). Ceramic-based materials have also been used as MFC separators in recent years due to their low cost and natural availability (Behera et al. 2010). Earthen, terracotta, earthenware, and clayware are some of the most commonly used since they exhibit good long-term stability and low maintenance requirements, which facilitate their use in scaled-up MFCs. These results might be related to the low ohmic resistance of these natural materials (Winfield et al. 2016). Another biodegradable alternative is natural rubber, which also exhibited promising results as MFC separator reaching higher power densities than commercial IEMs (Winfield et al. 2013).

MFC Designs Although MFC designs have much evolved over the last 20 years, in general, they can be categorised in terms of their scale and presence or absence of a membrane. New cathode materials have also been developed that can act as both a cathode and a membrane. Nonetheless, the number of chambers can determine the difference between membraneless and membrane-based MFCs. A membraneless MFC has a single reaction chamber, and both anode and cathode share the same electrolyte. The liquid solution contains organics and therefore serves as fuel as well as the liquid electrolyte of the MFC. A membrane-based MFC will always have a solid physical separator between the anode and the cathode. The presence of the membrane can sometimes be associated with the presence of the cathodic chamber. The reason for choosing multichamber designs is to combine the generation of electrical energy and wastewater treatment with other processes, occurring in the cathodic

886

C. Santoro et al.

compartment, such as the production of value added products e.g. hydrogen (Cusick et al. 2011) or a disinfectant solution (Gajda et al. 2016). One of the most researched areas of MFC design is the what absolute size each anodic chamber should take. The first large-scale (660-gallon; 2500 L) microbial fuel cell system was aimed at producing 2 kW but in the end produced significantly less. In contrast, at about the same time, Ieropoulos et al. used small-scale MFC to energise EcoBot-I and continued with small-scale systems for building EcoBots II and III (Ieropoulos et al. 2008). This approach is based on miniaturisation and multiplication of small-scale MFCs, as a viable way to scale up for power density (of a stack) and also for creating small-scale cascades (Winfield et al. 2012). Other marked differences between large- and small-scale MFC units include the surface area (of electrodes and separator)-to-anodic volume ratio which increases with smaller volumes. At large scale the fraction of total living cells within the anodic chamber in planktonic suspension is similar in number to the population of active cells attached to the electrode. The smaller the MFC, the less the proportion of planktonic cells there are in the anodic chamber, compared to the attached biofilm cells, at any time. The planktonic contribution toward possible “side reactions” (e.g., methanogenesis) becomes less important as the anode chamber decreases in size. In very small MFC, the biofilm population outnumbers the planktonic numbers by 10–100-fold, depending on the degree of miniaturisation. The dilution rate of largescale MFCs is either zero (i.e., run in batch mode) or low, generally below 0.1 h1, and this results in low growth rates, slow metabolic rates and sluggish performance of the biofilm, which suboptimum. The small degree of hydrodynamic shear is insufficient to remove daughter cells, thus resulting in thick (diffusion-limited) biofilms. In the presence of excess carbon energy, until accumulating biofilm conditions, large amounts of EPS also contribute to increasing diffusion limitation, slowing metabolic activity and power production. Thick biofilm formation is more likely in the case of large electrode blocks or plates that are impermeable to hydraulic perfusion flow (Guo et al. 2015). If highly perfusible electrodes are used in combination with small-scale volumes, moderate hydraulic rates ensure that the biofilm remains free of diffusion limitations throughout its lifetime (Fig. 4).

Scaling-Up for Energy Production In the last two decades, there has been substantial progress toward scale-up and practical implementation of microbial fuel cell (MFC) technology. Studies investigating MFCs reported reactor volumes from micro-L to 1000-L scale (Ge et al. 2013) ranging from laboratory-based experiments to long-term pilot-scale investigations. In the case of power production, having a MFC system suitable for realworld applications implies achieving high power densities at a large scale while keeping the system operational for long periods. A MFC is by nature a relatively low power-generating device. The maximum theoretical open-circuit voltage (OCV) of 1.14 V is determined by the redox potential difference between NADH (0.32 V vs. SHE) and oxygen (+0.82 V vs.

35

Microbial Fuel Cells, Concept, and Applications

Power Density (W/m3)

a

887

104 103 102 101 100 10–1 10–2 10–3 10–4

Energy Recovery (kWh/m3)

b

Energy Recovery (kWh/kg COD)

c

100 10–1 10–2 10–3 10–4 10–5 10–6 101 100 10–1 10–2 10–3 10–4 10–5 10–6 –3 10 10–2 10–1 100 101 102 103 104 MFC anode liquid volume (mL)

105

Fig. 4 Power generation (a), energy normalised as a function of the anode solution (b), and energy normalised as a function of COD consumption (c) for MFCs having different volume sizes. (Figure 4 reprinted and adapted with permission from Ge Z, Li J, Xiao L, et al. (2013), Environ Sci Technol Lett 1:137–141. Copyright (2014) American Chemical Society, Ref. (Ge et al. 2013))

SHE) at pH 7. However, with the exception of few reports in which specific conditions allowed reaching an open-circuit voltage level very close to the max (Ieropoulos et al. 2007; Santoro et al. 2016b), in practice, OCV reaches ~0.8 V and a working voltage of ~0.35–0.40 V. Hence, to reach high power outputs, two strategies

888

C. Santoro et al.

are pursued; either increasing the size of the MFC or assembling a plurality of MFCs into stacks. To increase reactor size however, it is important to keep the areal density of the electrode reactions constant. Moreover, the only way not to lose performance while increasing the size is for the density of the reaction sites/surfaces to be independent from the size and the diffusion distances to be kept minimal and constant (Walter et al. 2016). This challenging approach proved to be efficient for practical implementation in two cases: benthic MFCs (Ewing et al. 2017; Tender et al. 2008) and water column MFCs (Ieropoulos et al. 2016a; Walter et al. 2018). In these systems appropriate electrical connections (series, parallel) must be adhered to since the units share the same electrolyte. This condition often requires the use of DC/DC converters to boost the voltage to desired levels. A second strategy involves the assembly of a plurality of smaller units into a stack. Besides maintaining high power densities (Ieropoulos et al. 2008), such approach allows for series or parallel electrical connections between the stacked MFCs and, thus, enables increasing the voltage and current outputs. These strategies or combination of both methods is pursued to increase the power output toward usable levels for practical implementation. The challenges in the stacking strategy are voltage reversal (Aelterman et al. 2006) as well as the system design, its longevity and operation both in terms of substrate supply and distribution, power management and energy harvesting. Voltage reversal generally occurs due to the non-uniformity between stacked units, and it is inherent in the biological component of the system (Greenman et al. (2011); Ledezma et al. 2013a). It can be eliminated by an addition of a supporting anode in parallel (Kim et al. 2015), controlling the inner current via ohmic resistance tuning (Logan et al. 2018), or by the presence of parallel elements (Papaharalabos et al. 2017) forming groups (modules) which then are connected in series. Multiple reactors can be fluidically isolated by an air gap (Ieropoulos et al. 2008; Ledezma et al. 2013a), overflow (Feng et al. 2014), or baffles (Zhuang et al. 2012) between modules to avoid direct hydraulic connections and limit losses.

Microbial Fuel Cell for Wastewater Treatment Microbial fuel cells were first described in 1911 (Potter 1911), but at that time, the focus was not on wastewater treatment but on the concept that microorganisms could generate electricity from synthetic media (cane sugar and glucose). It is perhaps surprising that it was not until the 1990s that MFCs were first linked to wastewater treatment (Habermann and Pommer 1991). Since then there have been hundreds of lab-scale studies employing MFCs to treat a wide range of wastewater types. These findings have highlighted that MFCs are versatile and providing there is organic matter in the liquid and conditions are favourable (i.e., temperature, pH), then MFCs will adapt to the environment. There have been arrays of wastewater types used (Pandey et al. 2016). This has demonstrated MFCs’ unique ability to deal with wastewaters that have vastly different characteristics (Pandey et al. 2016). MFCs offer many advantages and benefits as a technology for wastewater treatment. They

35

Microbial Fuel Cells, Concept, and Applications

889

are less energy intensive and in addition they can actually contribute positively by producing electricity concomitant to treatment. It is not just the removal of organic matter and electricity production that makes MFCs such an attractive technology, but specific pollutants can also be targeted either for removal (generally at the anode) or for recovery (cathode). This is particularly relevant when considering that different wastewaters will have vastly varying characteristics. For example, animal and agricultural wastewater will be high in potassium, nitrogen, and phosphorus, which are the three main components of fertiliser that can be removed anodically (Zhang et al. 2019) or via precipitation at the cathode (Almatouq and Babatunde 2018). MFC researchers should appreciate that oxygen is not the only sustainable cathodic electron acceptor because other elements with high redox potential such as some heavy metals can also be utilised (Ucar et al. 2017). By running wastewater contaminated with heavy metal ions through the cathode, the high redox ions react with the incoming electrons coming from the anode, resulting in the precipitation of that metal on the cathode surface. This is a good example of how MFCs can both remove problem pollutants and recover valuable elements; examples include chromium, vanadium, arsenic, gold, silver, copper, cobalt, selenium, platinum, and even uranium (Mathuriya and Yakhmi 2014). To date there are just a small number of examples of pilot or field trials at wastewater treatment companies. As already mentioned, the MFC pilot installed at Foster’s brewery in Australia (2007) (Keller and Rabaey 2008) has been an invaluable experience that revealed some of the challenges that accompany in situ operation particularly the issue with the accumulation of salts at the cathode and also the unexpected impact simply from the exposure to the elements. On a similar note, a smaller-scale trial was carried out positioning MFCs in biological trickling filters in the UK, with swarms of flies being one of the major challenges (Ieropoulos et al. 2016b). Other pilot systems at treatment plants have been installed at smaller scales but still valuable in the information they provide. An example utilizing brewery wastewater is the 20 L pilot scale that was run for almost 1 year with high COD removal up to 95% (Lu et al. 2017). Furthermore, a different group operated a 45 L system over a 9-month period and evaluated that the MFC system was more energy- and cost-efficient than anaerobic digestion technology (Hiegemann et al. 2016). A larger pilot was conducted using a 90 L for brewery wastewater (Dong et al. 2015), and the output was more than sufficient to power pumps over the period. A 100 L volume pilot scale was also used for treating municipal wastewater and charging an ultracapacitor (Ge et al. 2015). Finally, Stoll et al. demonstrated that MFCs could contribute toward an energy-neutral environment for wastewater treatment companies (when compared to aeration processes) but that true neutrality could only occur if MFCs were fed high-strength wastewater (Stoll et al. 2018).

Microbial Fuel Cell and Practical Applications MFC technology has continued to attract increased attention as the power generation processes are being improved. However, despite significant improvements in power generation recorded over the last three decades, the technology is yet to reach its

890

C. Santoro et al.

full commercial potential. A recent review documents the performance of liter-scale MFCs treating real wastewater at continuous flow mode, thus illustrating the potential for implementation and technology readiness (Lu et al. 2017). Up until 2010, only three pilot-scale trials had been tested (Logan 2010). Since then, several other field trials have been conducted: a microbial electrochemical system for hydrogen production from wastewater (Cusick et al. 2011), benthic MFCs (Tender et al. 2008), MFCs in constructed wetland for wastewater treatment (Corbella et al. 2015), a prototype to be integrated in wastewater treatment plant (Jiang et al. 2011; Martinucci et al. 2015; Hiegemann et al. 2016; Ge et al. 2015), floating MFCs combined with plants that act as autonomous sensors able to transmit a signal in natural water bodies (Schievano et al. 2017), and MFC-based urinal system (Ieropoulos et al. 2016a; Walter et al. 2018). The first practical demonstrations using MFCs, which have not yet reached field trial stage, are to be found in the world of robots. “Gastronome” (Fig. 5a) was a three-wagon train that employed an artificial stomach (front wagon) in which sugars (dextrose) were metabolised by using artificial mediators to facilitate the electron transfer. Metabolised sugars along with the chemically reduced artificial mediator (HNQ which was part of the “stomach” content) were then pumped into six abiotic

e

c

a Gastronome

EcoBot II

2000

2004

2013

2010

2002

b

EcoBot IV

f

d EcoBot I

2015

EcoBot III

Row-Bot

Fig. 5 MFC-powered robots in chronological order until 2015. Gastronome (a), EcoBot-I (b), EcoBot-II (c), EcoBot-III (d), EcoBot-IV (e), and Row-Bot (f). (Figure 5a Reprinted from Wilkinson (2000) with permission of Springer. Figure 5b, c, and d source Wikipedia (https://en. wikipedia.org/wiki/EcoBot))

35

Microbial Fuel Cells, Concept, and Applications

891

fuel cells, and the energy produced was used to charge a battery pack (Ni-Cd), which powered the train’s motors and pumps (rear wagon) (Wilkinson 2000). Wilkinson’s invention was the first example of microbial metabolism incorporated into a robotic system. EcoBots made their appearance in 2002. EcoBots integrated robotics with the MFC technology providing autonomy by using wastewater and organic matter as their fuel. EcoBot-I (Fig. 5b) was the first of the series directly powered by MFCs. It employed eight E. coli-inoculated MFCs which with the aid of artificial mediators (methylene blue) oxidised the sugars and charged the onboard electrolytic capacitors. This robot was able to perform phototaxis every time the capacitor bank reached a certain threshold (Ieropoulos et al. 2003, 2004). Two years later, its successor, EcoBot-II (Fig. 5c), was reported (Ieropoulos et al. 2005b). The differences with its predecessor were the avoidance of using synthetic mediators, the use of air-breathing cathodes and mixed bacterial sludge culture. These advancements resolved the aforementioned problems and resulted in higher power outputs and wider substrate utilisation possibilities. Thus, EcoBot-II was more advanced in functionality than EcoBot-I as on top of phototaxis it performed sensing (environmental temperature monitoring) information processing, and wireless communication of the temperature information through an onboard wireless transmitter (Melhuish et al. 2006). In 2010, EcoBot-III was reported (Fig. 5d) which was the first to exhibit autonomous behaviour as it was able to collect food and water from its environment, metabolise the collected food, and egest the waste (Ieropoulos et al. 2010b). EcoBot-III was able to perform temperature logging, communicate telemetry, move towards food and water sources, and actuate its numerous pumps, motors and micro-controller. Three years later the team developed EcoBot-IV (Fig. 5e) that was the worlds’ first fully self-sustainable robot employing 24 MFC units that could also receive commands by a human operator to change its behavioural repertoire, if the need arose. It had bioinspired digestion and ingestion systems and was able to power its electronics and transmit data about its performance to a computer wirelessly. The most recent bioinspired MFC-powered autonomous robot is Row-Bot (Rossiter et al. 2015; Fig. 5f). This followed from the EcoBot example, but changed the environment from terrestrial to aquatic, since it was inspired by the water boatman beetle, an aquatic insect that floats on the water surface, feeds on algae and dead plants. The stomach of the robot consisted of two MFCs, which digested the organic fluid and turned it to electricity that powered the sof-bodied robot to open and close its mouths and move (row itself) forward to continue its operation (Philamore et al. 2015, 2016). The first successful field trial of benthic MFCs was also able to power sensors but in marine environments (Shantaram et al. 2005; Tender et al. 2008). In these studies, the power management circuitry comprised DC-DC converters, power harvester, and capacitors/batteries. Recently other advancements in practical applications powered by benthic MFCs have been reported with the utilisation of larger MFCs and more efficient energy harvesters (Ewing et al. 2017; Arias-Thode et al. 2017).

892

C. Santoro et al.

Fig. 6 Examples of practical applications reported at the University of the West of England. Pee Power ® urinal tested on the University of the West of England, Frenchay campus (a), MFC stack capable of powering a smartphone (b), wristwatch powered by MFCs (c), windmill powered by a single MFC. (Figure 6b adapted from Walter et al. 2017 published by Elsevier under CC-BY-4.0. Figure 6c adapted from Papaharalabos et al. (2013) with permission of Elsevier. Figure 6d adapted from Gajda et al. (2015) with permission of Elsevier)

Another important step in materialising practical applications of MFCs was the use of power generated from MFCs for switching on LEDs in toilets (Ieropoulos et al. 2016a; Fig. 6a), powering a mobile phone (Ieropoulos et al. 2013) and more recently a smartphone (Walter et al. 2017; Fig. 6b), operating a wristwatch (Papaharalabos et al. 2013; Fig. 6c), or running a windmill normally powered by solar panels (Gajda et al. 2015; Fig. 6d). Moreover, MFCs were used for powering DC pumps that were feeding and hydrating the MFC stack, thus demonstrating selfsustainability (Ledezma et al. 2013b) and also a motorised air freshener, in the context of autonomous odour-neutralisation (You et al. 2016). More recently, pilot tests have been carried out involving the use of power generated from MFC stacks to light LED modules in festivals and rural communities. Particularly, at the Glastonbury Festival in 2015, stacks of MFCs consisting of 432 individual units generated a mean power of 300 mW and powered internal lightings of the toilet units for the duration of the festival (Ieropoulos et al. 2016a). An improved design of the MFC system was deployed the next year at the festival (2016) resulting in the generation of 30% higher power despite the reduction in size, thus providing more lighting for the toilet units (Walter et al. 2018). This demonstrated the feasibility of MFC systems’ implementation as self-sustaining lit urinal systems, which can be deployed to remote locations (Ieropoulos et al. 2016a).

Microbial Fuel Cell as Biosensor One promising area proposed for the implementation of microbial fuel cell (MFC) technology in recent years is as a biosensor. A biosensor is defined as any analytical device, which utilises a biological based feedback mechanism and is transduced into

35

Microbial Fuel Cells, Concept, and Applications

893

an electrical output (Mehrotra 2016). The constant monitoring of water quality by the wastewater industry is paramount in providing a good service. Currently the main methods for measuring biological oxygen demand (BOD), chemical oxygen demand (COD), and other specific molecules or ions are time-consuming and relatively expensive. Particularly, the traditional method for measuring BOD takes 5 days, and COD can be available within 3 hours using colorimetric methods. Faster but not yet immediate response can be achieved in measuring nitrogen, phosphorus, and specific metals, also in this case through colorimetric methods or ion chromatography. However, both the practicality and costs of repeating reagent tests and the length of testing procedures are not always ideal. This leaves a void which a system such as an MFC can be used as an “online” in situ biosensor for measuring the target analyte(s) of interest. In fact, an MFC can give feedback in real-time on the characteristics of a monitored analyte liquid. In the anode of an MFC, the biofilm is composed of electroactive microorganisms whose response is directly linked to their metabolic activity rates, and therefore power output can be correlated to concentration of a substance/substrate within given media under controlled conditions (Fig. 7a). In principle, any metabolite that is broken down via the biofilms metabolism is reflected in the power/signal output (depending on what is being measured) and as a result has the potential to be used as the MFC-based biosensor measuring the concentration of organics into the electrolyte (Fig. 7b). Moreover, MFCs can also be used as a “shock” sensor for large differences in physicochemical conditions such as the presence of toxic compounds in general. In fact the excessive concentration of specific pollutants such as disinfectants or heavy metals can decrease the power output by negatively affecting the activity of the electroactive biofilm (Fig. 7c).

Sensors for BOD Monitoring Biochemical oxygen demand (BOD) is an environmental parameter widely used to indicate the amount of biodegradable organic material in water bodies. It is used extensively as the measurement of organic pollution in wastewater, detecting only the portion of the total carbon that can be oxidised by microorganisms. The BOD test relies on a measurable depletion of dissolved oxygen (DO) over a specified period of time, as it represents the amount of dissolved oxygen needed for microbial oxidation of the organic matter. BOD5 determination is time-consuming as the test requires 5 days to obtain the results, and consequently it is not suitable for online process monitoring. Therefore, rapid determination of BOD is particularly attractive as special attention is paid to the application of biosensing in real environments - this is where the MFC technology can be particularly useful. The biofilm attached to the anode electrode is considered as the main sensing element for BOD concentration as well as for biotoxicity monitoring. It measures the respiration rate of the immobilised microorganisms on the electrode surface as a result of the concentration of the biodegradable compounds. The first use of the BOD sensing microbial electrode was reported by Karube et al. (1977) that used

894

C. Santoro et al. e-

a

SIGNAL

Signal

e-

A N O D E

Time

Organic compounds e-

b

Generic ion

SIGNAL

e-

Signal

BOD ADDITION

A N O D E

Time

Organic compounds e-

c

Generic ion

SIGNAL

e-

A N O D E

Signal

ADDITION OF TOXIC SUBSTANCE

Time

Toxic substance Organic compounds

Generic ion

Fig. 7 Principles of an anodic electroactive biofilm operating as biosensor: steady state (a), addition of organics as BOD (b), and addition of toxic substances (c)

immobilised microorganisms with an oxygen probe. Later in 1999, Kim and Kwon presented a microbial BOD sensor that was free from the inhibitive effect of toxic heavy metal ions as well as the pH, demonstrating practically a useful BOD

35

Microbial Fuel Cells, Concept, and Applications

895

monitoring device (Kim and Kwon 1999). This led to the first instance of MFC technology being utilised as biosensor, and it can be attributed to Kim et al. (2003) who used a mediator-less MFC for detecting BOD for almost 5 years. In their work, the authors were able to successfully show that a MFC’s output responded to the addition of BOD resulting in voltage fluctuations (Kim et al. 2003). It presented that the amount of electrical charge (expressed in coulombs) generated from a mediatorless MFC is proportional to the concentration of fuel used; therefore, MFCs can be used as a BOD sensor. Since then there have been numerous other examples reported for the development of MFCs as biosensors for a range of topics from water security to healthcare (Su et al. 2011). Particularly, MFCs have been used in the past decade as biosensors for several substrates involving COD and BOD (Yang et al. 2015), heavy metals (Ivars-Barceló et al. 2018), glucose (Grattieri and Minteer 2018), and others (Su et al. 2011) for a range of different media/solutions and consequently for the application in different sectors. There are numerous factors that affect the sensitivity and stability of the MFC-based BOD sensors, such as the concentration of the dissolved oxygen, external resistance, and substrate flow rate as well as the impact of other parameters such as pH, temperature, and other environmental factors. Through the past 20 years, significant improvements have been made by eliminating respiratory inhibitors on the current generation (Chang et al. 2005), achieving faster response time through miniaturisation (Di Lorenzo et al. 2009), and improving practical aspects of the design (Peixoto et al. 2011) that led to the development of more robust bioelectrochemical sensors (Modin and Wilén 2012). MFC-based biosensors have an outstanding potential for the real-time, rapid, and in situ monitoring of water quality (ElMekawy et al. 2013). A particular case for MFC biosensors is that through an increased power generation, there is promise that they can function as an online, “real-time” monitoring device (Chang et al. 2004) that could be operating as an autonomous sensor. To demonstrate this potential, MFCs and an energy management device were combined, upon reaching a level of charge that set off visual and sound alarms to alert the user that the BOD concentration of pollutants present in water reached levels beyond pre-set thresholds (Pasternak et al. 2017). Such a floating biosensor was developed for online water quality monitoring where the energy required by the sound and light alarm connected to the biosensor was produced by four MFCs connected in parallel charging capacitors via the energy management system. The important innovation in this demonstration was the fact that the frequency of the sound and light alarm switching ON was directly proportional to the concentration of the subject organic pollutant. The schematic representation of the biosensor operation shown in Fig. 8 are (1) biosensor operates in uncontaminated freshwater under open-circuit conditions; (2) in the presence of urine, the sensor open-circuit voltage increases; and (3) the energy management system switches ON, resulting in charging the capacitor; the sound and visual alarm is activated by the capacitor, when full, causing the latter to discharge; the frequency of activation is directly proportional to the concentration of pollutant in water. Recent developments in specially designed energy harvesting for low-power and cost-effective MFC-based sensors (Yamashita et al. 2019) can enable

896

C. Santoro et al.

Fig. 8 Schematic of the first autonomous, self-powered MFC biosensor utilised for online monitoring BOD. (Adapted from Pasternak et al. (2017) published by Elsevier under CC-BY-4.0)

off-the-grid applications of MFC-driven autonomous environmental sensing in the future.

Sensors for Toxicity Assessment Gas chromatography, gas chromatography coupled to mass spectrometry, highperformance liquid chromatography, and liquid chromatography coupled to mass spectrometry are some of the techniques most commonly used for detecting different kinds of compounds. However, their use is not feasible for on-site measuring in realtime monitoring processes because of their high cost (Choi and Gu 2003). MFCbased sensors overcome these limitations and may be a suitable alternative for the in situ detecting of toxic compounds. MFC power performance is directly related to an adequate biofilm growth, and the presence of toxic substances inhibits its

35

Microbial Fuel Cells, Concept, and Applications

897

developing. This fact may be used as the signal to detect the presence of toxic compounds. The inhibitory effect named as “I” can be determined as follows (Kim et al. 2007): I ð% Þ ¼

jCY nor  CY tox j  100 CY nor

In the above equation, CYnor and CYtox represent the coulombic yield of the MFC fed with nontoxic and toxic influent, respectively. These parameters can be calculated by integrating the current output by the system over time. The presence of a toxin in the MFC inlet feedstock will affect the biofilm growth in the anode, limiting their capacity to oxidise organic matter and, therefore, reduce the voltage/power output of the system. For this reason, MFC-based biosensors can be used to quantitatively determine the toxicity of different types of effluents since the inhibition rate is directly related to the amount of toxins. Kim et al. 2007 reported the use of MFCs for both on-site and online biomonitoring of the presence of toxic substances in water. The authors observed up to 76% of inhibition ratio when 1 mg L1 of cadmium and lead (II) mixture was added to real wastewater. The biosensor was able to detect up to 0.1 mg L1 of the toxic mixture. Among the benefits of using MFC-based sensors for detecting pollutants are the real-time monitoring and the long-term stability of the system, which does no need any maintenance requirements or chemical supply. By contrast, it is needed to avoid the downstream microbial adaptation to toxins in the influent, which will provide wrong measurements. A few years later, Davila et al. (2011) designed a microsensor based on a doublechamber MFC inoculated with Geobacter sulfurreducens. The authors reported that the system was sensitive to the presence of formaldehyde, with 0.1% being the lowest amount of this compound detected by the biosensor. They observed an irreversible inactivation of the biofilm activity for all amounts of formaldehyde tested. In addition to cadmium/lead mixture, and formaldehyde, MFC sensors have also being used to detect sodium dodecyl sulfate and bentazon in the concentration range of 10–50 mg.L1 and 1–3 mg.L1 (Stein et al. 2012), respectively, copper (II) (Shen et al. 2013), or hydrochloric acid (Shen et al. 2012), amongst other compounds. Linear relationship between the current output by the system and the substrate concentration is widely used to quantify the amount of toxin present in the influent. It is crucial to keep the anodic potential of the MFC in a suitable range as well as control the pH to avoid a false-positive measurement. Paper-based MFC was also used as portable, low-cost “shock sensor” for detecting chromium (Xu et al. 2016). A miniaturised electrochemical toxicity biosensor based on graphene oxide quantum dots/carboxylated carbon nanotubes was also used as a “shock sensor” for evaluating the toxicity of Cd, Hg, Pb, 2,4-dinitrophenol, 2,4,6-trichlorophenol, and pentachlorophenol (Zhu et al. 2017). Biosensors using electroactive biofilm were also used for assessing the toxicity of different volatile organic compounds (VOCs) (Santoro et al. 2016c; Kannan et al. 2019).

898

C. Santoro et al.

However, fluctuations and interferences are very common when real wastewater is analysed due to its complex content and the wide variety of compounds contained. In order to overcome this limitation, modeling tools such as nonlinear modeling methods or artificial neural networks (ANN) have being employed to analyze the signals from biosensors based on MFCs (Feng et al. 2013; Feng and Harper 2013). Despite the multiple benefits of using MFC-based sensors for detecting toxins, there are still some limitations that need to be overcome to promote their practical application. Some of the most important are the need for microorganisms to be highly active, the low sensitivity, as well as the optimisation of the control modes or the flow configurations (Jiang et al. 2015).

Sensors for Microbial Activity Monitoring Applications of microbial fuel cell technology for the monitoring of microbial activities have attracted the attention of the scientific community in recent years, due to the laborious and time-consuming nature of the traditional microbiological approaches currently being employed (Cui et al. 2019). The approach is based on current generation by MFCs as a response to different substrates, and it is hought to be a fast and easy way to collect original data of microbial activities from diverse environments and particularly for the monitoring of microbial activities in situ (Cui et al. 2019; Mathuriya and Yakhmi 2016). One of the earlier reports on the use of MFC systems in this regard was proposed by Tront et al. (2008), who developed an approach to monitor substrate concentration and rate of microbial respiration in groundwater in situ. In their experiments, the authors inoculated column systems with Geobacter sulfurreducens for the utilisation of acetate as the substrate. The results showed that the level of electrical signals generated provided a real-time data on the electron donor availability and the metabolic activity of the bacteria. The authors concluded that the metabolic responses of the organism were a function of substrate utilisation as well as current generation. The system also showed the potential to provide valuable information about the presence of toxic substances or poor microbial growth highlighted by the response and recovery to shock oxygen loading. The ability of MFCs to monitor microbial activity was also tested by Williams et al. (2010) who used graphite anodes installed downgradient from a region containing acetate with graphite cathode embedded at the ground surface. The authors showed a strong correlation between current generation and uranium removal/acetate utilisation. Analysis of the microbial community in that experiment showed the dominance of Geobacter species (up to 80%). The authors concluded that the electrode was capable of generating detectable current despite the long separation between the cathode and the anode (6 m) and that the Geobacter species were the primary source of current generation. In another study, a submersible MFC system was developed to monitor microbial activity in groundwater in the presence or absence of a developed biofilm. The results showed that anode electrode without developed biofilms was suitable for monitoring microbial activities, while those with

35

Microbial Fuel Cells, Concept, and Applications

899

developed anode were only suitable for BOD measurements. The results of that study showed that current density generated was proportional to the active microorganism concentration, which was expressed as microbial adenosine-triphosphate concentration (Zhang and Angelidaki 2011). The authors, however, highlighted the impact of factors such as temperature, pH, and conductivity on the sensitivity of the MFC sensor. To detect the presence or activity of microbial life in outer space, Abrevaya et al. (2010) reported the use of a simple and compact cylindrical MFC reactor with no anode chamber. The authors highlighted the possibility of using the technology as a reliable sensing tool for microbial detection in outer space, based on the microbial activity, which is a universal characteristic of life. The potential of MFC for the monitoring of microbial activity constitutes a new paradigm for MFC application and highlights the possibility of generating inexpensive and accurate real-time microbial activity data irrespective of location. It comes with additional safeguarding with the ability to differentiate the presence of toxic materials, which may impact the microbial activity resulting in possible false results.

Microbial Fuel Cells (MFCs) for Monitoring Bio-Corrosion Biosensor applications using MFCs can be used from monitoring environmental contaminants to detecting toxicants and detecting the presence of microbial biofilms that can cause bio-corrosion. Microorganisms inherently adhere on surfaces or substrata, where they grow and perform a variety of metabolic reactions, breaking down organic molecules. The products of this metabolism may influence corrosion by promoting the deterioration of the underlying substratum (Xu and Gu 2014). In the cases where the substratum consists of metal or metal alloys, this reaction is referred to as bio-corrosion. Microbially induced corrosion (MIC) or bio-corrosion changes the electrochemical conditions of the metal/solution interface due to the biofilm formation (Videla and Herrera 2014). Bio-corrosion posed significant problems to operators of industrial water, gas, and oil systems, everywhere in the world (Geesey 1991). The repair and replacement of corroded pipes cost billions of pounds to these utility industries; if left untreated, corroded pipes reduce performance and efficiencies as well as introduce environmental hazards (Geesey 1991). Hence, the early detection and treatment of biocorrosion is of great importance; however, first and foremost the underlying processes involved need to be understood. Primarily bio-corrosion is induced in anaerobic conditions and is caused by the two anaerobic metabolisms: respiration and fermentation. The former is often caused by sulfate-reducing bacteria (SRB) and the latter by acid-producing bacteria (APB). SRB usually use sulfate as the electron acceptor and organic carbon as the electron donor, to produce maintenance energy. If organic carbon is not present (carbon starvation), which is the case in biofilm-covered steel surfaces, SRB switch to elemental iron (Fe0) as the electron donor for energy production (Xu et al. 2013).

900

C. Santoro et al.

The electrons that are released from the iron oxidation are taken directly by microbes which grow on the iron surface (substratum) as biofilm, facilitating electron transfer and causing bio-corrosion. Conversely, APB secrete corrosive metabolites which are oxidants such as formic acid and acetic acid. Acetic acid attack is a frequent problem in the oil and gas industry. This type of bio-corrosion can be detected easily as it is associated with low pH; however, SRB-induced corrosion is more challenging to be detected (Xu et al. 2013). It is apparent that detecting bio-corrosion is important for the smooth operation of industrial pipelines, as it will impact the decision-making process for their appropriate treatment. In detecting bio-corrosion, existing sensors face a challenge of distinguishing a metal film from a biofilm; also these sensors require external voltage input to work. A sensor having a sacrificial anode based on zinc and a cathode based of stainless steel has been used to detect the biofouling within pipeline (Cristiani et al. 2008). The voltage output increased with the formation of biofilm on the cathode, and when a certain threshold was reached, chlorine was used to remove the biofilm. The negative aspect of this sensor was the utilisation of a sacrificial anode that was consumed over time in order to provide electrons to the cathode (Cristiani et al. 2008). Recently it was suggested that a microbial fuel cell (MFC) approach can be used to detect a corrosive biofilms since such a biofilm will have electroactive abilities (transfer electrons across a cell wall) (Gu 2012). The proposed MFC-based biofilm detector will be membraneless and will have a solid-state electrode for anode, in a sulfate- and nitrate-rich solution, along with a bio-cathode. The idea behind this approach is that a pipeline can have different inert elements such as graphite, allowing a biofilm to grow on them (Yang et al. 2015). This element can then be removed and added into the MFC-based sensor as a bio-cathode. The shift in potential between the solid-state electrode and bio-cathode (after calibration) will be an indication of the presence of bio-corrosion (Yang et al. 2015). It is envisaged that this nondestructive biofilm detection approach can be easily implemented in the near future, pending further experimentation and development.

Other Sensor Applications Besides the aforementioned MFC biosensors, other potential sensing applications using MFCs have been explored. The amperometric detection of gaseous compounds is one of them. The research on biosensing gas started in the early 1980s. Karube et al. reported on a sensor using immobilised microorganisms and oxygen cathode for the detection of methyl alcohol, ethyl alcohol, acetic acid (Karube and Suzuki 1988), and also ammonium (Karube et al. 1981). Since then, research on bioelectrosensing has more recently progressed towards the detection of VFAs (volatile fatty acids) such as acetate, propionate, or butyrate, which are intermediate compounds of the methane formation pathway (Merlin Christy et al. 2014). Researchers dedicated attention to the fact that VFAs can be a reliable indicator to monitor the anaerobic digestion (AD) process which is a well-known complex

35

Microbial Fuel Cells, Concept, and Applications

901

organic waste valorisation (Schievano et al. 2018). Therefore, monitoring VFAs in the AD process can prevent the process breakdown and enable efficient operation of the process. Choi et al. used various types of VFA as MFC substrate and showed that the utilisation rate of two preferred substrates, acetate and propionate, slows down when other VFAs are present (Choi et al. 2011). Kaur et al. investigated the correlation between current/voltage responses and the three major VFA species, i.e., acetate, propionate, and butyrate, individually, using 250 mL H-type MFCs (Kaur et al. 2014). A recent study demonstrated promising results of VFA biosensor, which showed a linear relationship with the VFA concentration ranging between 0 and 200 mM at the specific reaction time from 10 to 120 min (Jiang et al. 2019). This sensor also had good selectivity when other organic compounds such as cellulose, yeast extract, peptone, and glyceryl trioleate were present. Within environmental monitoring, water quality is an important topic. For example, studies have focused on the use of microbial fuel cells for monitoring of the dissolved oxygen concentrations (DOC) in water bodies (Song et al. 2019; Zhang and Angelidaki 2012). The work of Zhang and Angelidaki showed very good correlation between DOC and signal, for several types of water (seawater, wastewater, lake waters) and with relatively quick response times of around 4 min. A recent field trial (67 days) carried out by Song et al. also showed a good correlation between the in situ DOC of a lake and the recorded signal, thus demonstrating the potential use of bioelectrosensors for DO measurements (Song et al. 2019). Following a similar line of thought, Wang and Jiang proposed a system for monitoring the bulking of sediments which is caused by the discharge of industrial wastewater or municipal sewage (Wang and Jiang 2019). In the case of the DO measurement developed by Song et al., the setup consisted of a single anode, buried in the sediments, and multiple cathodes spaced regularly along a water column. Conversely, the setup developed by Wang and Jiang to detect sediment bulking employed as single cathode and an array of anodes placed along the sediment/ water column interface. In both cases, these types of environmental sensors aimed at the early detection of eutrophication events through real-time monitoring (Song et al. 2019; Wang and Jiang 2019). Kim and Han explored another possible application of MFC biosensor that can detect and quantify E. coli (Kim and Han 2013). They tested environmental as well as laboratory samples, which showed good results although the response time was about 250 min longer than ones with laboratory samples. Other sensor applications using MFCs such as antibiotic such as levofloxacin (Zeng et al. 2017), oil (Dai et al. 2019), or light (Ieropoulos et al. 2016b) have also been reported. A more pragmatic study recently focused on the possibility of employing MFCs to detect cocaine metabolites in urine (Catal et al. 2019). In this work, Catal et al. have demonstrated the inverse correlation between the metabolite concentration (benzoylecgonine) and the voltage measured though a fixed load. The concentration of benzoylecgonine found in urine samples of users can be as high as 700 ng mL1. In the study, the lowest concentration tested was 100 ng.mL1 (concentration normally found in real samples) and gave a significant and detectable voltage drop, thus confirming the potential for future applications.

902

C. Santoro et al.

References Abrevaya XC, Mauas PJD, Cortón E (2010) Microbial fuel cells applied to the metabolically based detection of extraterrestrial life. Astrobiology 10:965–971 Aelterman P, Rabaey K, Pham HT et al (2006) Continuous electricity generation at high voltages and currents using stacked microbial fuel cells. Environ Sci Technol 40:3388–3394 Almatouq A, Babatunde AO (2018) Identifying optimized conditions for concurrent electricity production and phosphorus recovery in a mediator-less dual chamber microbial fuel cell. Appl Energ 230:122–134 Amari S, Boshrouyeh Ghandashtani M (2019) Non-steroidal anti-inflammatory pharmaceutical wastewater treatment using a two-chambered microbial fuel cell. Water Environ J. https://doi. org/10.1111/wej.12476 Antolini E (2015) Composite materials for polymer electrolyte membrane microbial fuel cells. Biosens Bioelectron 69:54–70 Arias-Thode YM, Hsu L, Anderson G et al (2017) Demonstration of the SeptiStrand benthic microbial fuel cell powering a magnetometer for ship detection. J Power Sources 356:419–429 Bajracharya S, Srikanth S, Mohanakrishna G et al (2017) Biotransformation of carbon dioxide in bioelectrochemical systems: state of the art and future prospects. J Power Sources 356:256–273 Baudler A, Schmidt I, Langner M et al (2015) Does it have to be carbon? Metal anodes in microbial fuel cells and related bioelectrochemical systems. Energy Environ Sci 8:2048–2055 Behera M, Jana PS, Ghangrekar MM (2010) Performance evaluation of low cost microbial fuel cell fabricated using earthen pot with biotic and abiotic cathode. Bioresour Technol 101:1183–1189 Bennetto HP (1990) Electricity generation by microorganisms. Biotechnol Educ I 4:163–168 Berk RS, Canfield JH (1964) Bioelectrochemical energy conversion. Appl Microbiol 12:10–12 Cao X, Huang X, Liang P et al (2009) A new method for water desalination using microbial desalination cells. Environ Sci Technol 43:7148–7152 Catal T, Kul A, Atalay VE et al (2019) Efficacy of microbial fuel cells for sensing of cocaine metabolites in urine-based wastewater. J Power Sources 414:1–7 Cecconet D, Molognoni D, Callegari A et al (2018) Agro-food industry wastewater treatment with microbial fuel cells: energetic recovery issues. Int J Hydrog Energy 43:500–511 Chang IS, Jang JK, Gil GC et al (2004) Continuous determination of biochemical oxygen demand using microbial fuel cell type biosensor. Biosens Bioelectron 19:607–613 Chang IS, Moon H, Jang JK, Kim BH (2005) Improvement of a microbial fuel cell performance as a BOD sensor using respiratory inhibitors. Biosens Bioelectron 20:1856–1859 Chiu H, Pai T, Liu M et al (2016) Electricity production from municipal solid waste using microbial fuel cells. Waste Manag Res 34:619–629 Choi SH, Gu MB (2003) Toxicity biomonitoring of degradation byproducts using freeze-dried recombinant bioluminescent bacteria. Anal Chim Acta 481:229–238 Choi JDR, Chang HN, Han JI (2011) Performance of microbial fuel cell with volatile fatty acids from food wastes. Biotechnol Lett 33:705–714. https://doi.org/10.1007/s10529-010-0507-2 Cohen B (1931) The bacterial culture as an electrical half-cell. J Bacteriol 21:18–19 Corbella C, Puigagut J (2018) Improving domestic wastewater treatment efficiency with constructed wetland microbial fuel cells: influence of anode material and external resistance. Sci Total Environ 631–632:1406–1414 Corbella C, Guivernau M, Vinas M et al (2015) Operational, design and microbial aspects related to power production with microbial fuel cells implemented in constructed wetlands. Water Res 84:232–242 Costa de Oliveira MA, Mecheri B, D’Epifanio A et al (2017) Graphene oxide nanoplatforms to enhance catalytic performance of iron phthalocyanine for oxygen reduction reaction in bioelectrochemical systems. J Power Sources 356:381–388 Cristiani P, Perboni G, Debenedetti A (2008) Effect of chlorination on the corrosion of cu/Ni 70/30 condenser tubing. Electrochim Acta 54:100–107 Cui Y, Lai B, Tang X (2019) Microbial fuel cell-based biosensors. Biosensors 9:92

35

Microbial Fuel Cells, Concept, and Applications

903

Cusick RD, Bryan B, Parker DS et al (2011) Performance of a pilot-scale continuous flow microbial electrolysis cell fed winery wastewater. Appl Microbiol Biotechnol 89(6):2053–2063 Dai Z, Xu Z, Wang T et al (2019) In-situ oil presence sensor using simple-structured upward openchannel microbial fuel cell (UOC-MFC). Biosens Bioelectron X 1:100014 Daud SM, Kim BH, Ghasemi M et al (2015) Separators used in microbial electrochemical technologies: current status and future prospects. Bioresour Technol 195:170–179 Davila D, Esquivel JP, Sabate N, Mas J (2011) Silicon-based microfabricated microbial fuel cell toxicity sensor. Biosens Bioelectron 26:2426–2430 Davis JB, Yarbrough HF (1962) Preliminary experiments on a microbial fuel cell. Science 137:615–616 Di Lorenzo M, Curtis TP, Head IM et al (2009) A single-chamber microbial fuel cell as a biosensor for wastewaters. Water Res 43:3145–3154 Dong Y, Qu Y, He W et al (2015) A 90-liter stackable baffled microbial fuel cell for brewery wastewater treatment based on energy self-sufficient mode. Bioresour Technol 195:66–72 ElMekawy A, Diels L, De Wever H, Pant D (2013) Valorization of cereal based biorefinery byproducts: reality and expectations. Environ Sci Technol 47:9014–9027 Ewing T, Ha PT, Beyenal H (2017) Evaluation of long-term performance of sediment microbial fuel cells and the role of natural resources. Appl Energy 192:490–497 Feng Y, Harper WF (2013) Biosensing with microbial fuel cells and artificial neural networks: laboratory and field investigations. J Environ Managem 130:369–374 Feng Y, Kayode O, Harper WF (2013) Using microbial fuel cell output metrics and nonlinear modeling techniques for smart biosensing. Sci Total Environ 449:223–228 Feng YJ, He WH, Liu J et al (2014) A horizontal plug flow and stackable pilot microbial fuel cell for municipal wastewater treatment. Bioresour Technol 156:132–138 Gajda I, Stinchcombe A, Greenman J et al (2015) Ceramic MFCs with internal cathode producing sufficient power for practical applications. Int J Hydrog Energy 40:14627–14631 Gajda I, Greenman J, Melhuish C et al (2016) Electricity and disinfectant production from wastewater: microbial fuel cell as a self-powered electrolyser. Sci Rep 6:25571 Ge Z, Li J, Xiao L et al (2013) Recovery of electrical energy in microbial fuel cells: brief review. Environ Sci Technol Lett 1:137–141 Ge Z, Wu L, Zhang F et al (2015) Energy extraction from a large-scale microbial fuel cell system treating municipal wastewater. J Power Sources 297:260–264 Geesey G (1991) What is biocorrosion? In: Flemming HC, Geesey GG (eds) Biofouling and biocorrosion in industrial water systems. Springer, Berlin/Heidelberg Ghasemi M, Daud WRW, Ismail AF et al (2013) Simultaneous wastewater treatment and electricity generation by microbial fuel cell: performance comparison and cost investigation of using Nafion 117 and SPEEK as separators. Desalination 325:1–6 Grattieri M, Minteer SD (2018) Self-powered biosensors. ACS Sensors 3:44–53 Greenman J, Gálvez A, Giusti L et al (2009) Electricity from landfill leachate using microbial fuel cells: comparison with a biological aerated filter. Enzym Microb Technol 44:112–119 Greenman J, Ieropoulos IA, Melhuish C (2011) Microbial Fuel Cells – Scalability and their Use in Robotics. In: Eliaz N (eds) Applications of Electrochemistry and Nanotechnology in Biology and Medicine I. Modern Aspects of Electrochemistry, vol 52. Springer, New York, NY Gu T (2012) Methods and devices for the detection of biofilms. World Intellectual Property Organization: Patent WO2012/149487 Guo K, Freguia S, Dennis PG et al (2013) Effects of surface charge and hydrophobicity on anodic biofilm formation, community composition, and current generation in bioelectrochemical systems. Environ Sci Technol 47:7563–7570 Guo K, Prévoteau A, Patil SA et al (2015) Engineering electrodes for microbial electrocatalysis. Curr Opin Biotechnol 33:149–156 Habermann W, Pommer EH (1991) Biological fuel-cells with sulfide storage capacity. Appl Microbiol Biotechnol 35(1):128–133 Harnisch F, Schröder U (2009) Selectivity versus mobility: separation of anode and cathode in microbial bioelectrochemical systems. Chem Sus Chem 2:921–926

904

C. Santoro et al.

Harnisch F, Schröder U (2010) From MFC to MXC: chemical and biological cathodes and their potential for microbial bioelectrochemical systems. Chem Soc Rev 39:4433–4448 Harnisch F, Schröder U, Scholz F (2008) The suitability of monopolar and bipolar ion exchange membranes as separators for biological fuel cells. Environ Sci Technol 42:1740–1746 Hernandez-Fernandez FJ, de los Rios AP, Salar-García MJ et al (2015) Recent progress and perspectives in microbial fuel cells for bioenergy generation and wastewater treatment. Fuel Processing Technol 138:284–297 Hiegemann H, Herzer D, Nettmann E et al (2016) An integrated 45 L pilot microbial fuel cell system at a full-scale wastewater treatment plant. Bioresour Technol 218:115–122 Huang L, Li X, Ren Y et al (2017) Preparation of conductive microfiltration membrane and its performance in a coupled configuration of membrane bioreactor with microbial fuel cell. RCS Adv 7:20824–20832 Ieropoulos I, Melhuish C, Greenman J (2003) Artificial metabolism: towards true energetic autonomy in artificial life. Adv Artif Life Proc 2801:792–799 Ieropoulos I, Melhuish C, Greenman J (2004) Energetically autonomous robots. 8th Intell Auton Syst Conf 128–135 Ieropoulos I, Melhuish C, Greenman J, Horsfield I (2005a) EcoBot-II: an artificial agent with a natural metabolism. Int J Adv Robot Syst 2:295–300 Ieropoulos I, Greenman J, Melhuish C et al (2005b) Energy accumulation and improved performance in microbial fuel cells. J Power Sources 145:253–256 Ieropoulos I, Melhuish C, Greenman J (2007) Artificial gills for robots: MFC behaviour in water. Bioinspir Biomim 2(3):S83 Ieropoulos I, Greenman J, Melhuish C (2008) Microbial fuel cells based on carbon veil electrodes: stack configuration and scalability. Int J Energy Res 32(13):1228–1240 Ieropoulos I, Greenman J, Melhuish C (2010a) Improved energy output levels from small-scale microbial fuel cells. Bioelectrochemistry 78:44–50 Ieropoulos I, Greenman J, Melhuish C, Horsfield I (2010b) EcoBot-III: a robot with guts. 12th Int Conf synth Simul living Syst 733–740 Ieropoulos IA, Ledezma P, Stinchcombe A et al (2013) Waste to real energy: the first MFC powered mobile phone. Phys Chem Chem Phys 15:15312–15316 Ieropoulos I, Stinchcombe A, Gajda I et al (2016a) Pee power urinal – microbial fuel cell technology field trials in the context of sanitation. Environ Sci Water Res Technol 2:336–343 Ieropoulos I, Winfield J, Gajda I et al (2016b) Chapter 12 – the practical implementation of microbial fuel cell technology. In: Scott K (ed) Microbial electrochemical and fuel cells. Woodhead Publishing, Cambridge, pp 357–380 Ivars-Barceló F, Zuliani A, Fallah M et al (2018) Novel applications of microbial fuel cells in sensors and biosensors. Appl Sci 8:1184 Jiang D, Curtis M, Troop E et al (2011) A pilot-scale study on utilizing multi-anode/cathode microbial fuel cells (MAC MFCs) to enhance the power production in wastewater treatment. Int J Hydrog Energy 36:876–884 Jiang Y, Liang P, Zhang CY et al (2015) Enhancing the response of microbial fuel cell based toxicity sensors to Cu (II) with the applying of flow-through electrodes and controlled anode potentials. Bioresour Technol 190:367–372 Jiang Y, Chu N, Zeng RJ (2019) Submersible probe type microbial electrochemical sensor for volatile fatty acids monitoring in the anaerobic digestion process. J Clean Prod 232:1371–1378 Kannan P, Jogdeo P, Mohidin AF et al (2019) A novel microbial-bioelectrochemical sensor for the detection of n-cyclohexyl-2-pyrrolidone in wastewater. Electrochim Acta 317:604–611 Karube I, Suzuki S (1988) Biochemical energy-conversion by immobilized photosynthetic Bacteria. Methods Enzymol 137:668–674 Karube I, Matsunaga T, Mitsuda S, Suzuki S (1977) Microbial electrode BOD sensors. Biotechnol Bioeng 19:1535–1547 Karube I, Okada T, Suzuki S (1981) Amperometric determination of Ammonia gas with immobilized nitrifying Bacteria. Anal Chem 53:1852–1854

35

Microbial Fuel Cells, Concept, and Applications

905

Kaur A, Ibrahim S, Pickett CJ et al (2014) Anode modification to improve the performance of a microbial fuel cell volatile fatty acid biosensor. Sensors Actuators B Chem 201:266–273 Keller J, Rabaey K (2008) Experiences from MFC Pilot Plant Operation. Available from: https://www.yumpu.com/en/document/view/47122823/experiences-from-mfc-pilot-plant-opera tion-microbial-fuel-cells Kim T, Han JI (2013) Fast detection and quantification of Escherichia coli using the base principle of the microbial fuel cell. J Environ Manag 130:267–275 Kim M-N, Kwon H-S (1999) Biochemical oxygen demand sensor using Serratia marcescens LSY 4. Biosens Bioelectron 14:1–7 Kim BH, Chang IS, Gil GC et al (2003) Novel BOD (biological oxygen demand) sensor using mediator-less microbial fuel cell. Biotechnol Lett 25:541–545 Kim M, Hyun SM, Gadd GM, Kim HJ (2007) A novel biomonitoring system using microbial fuel cells. J Environ Monitor 9:1323–1328 Kim B, Lee BGP, Kim BH et al (2015) Assistance current effect for prevention of voltage reversal in stacked microbial fuel cell systems. ChemElectroChem 2(5):755–760 Kodali M, Gokhale R, Santoro C et al (2017a) High performance platinum group metal-free cathode catalysts for microbial fuel cell (MFC). J Electrochem Soc 164:H3041–H3046 Kodali M, Santoro C, Serov A et al (2017b) Air breathing cathodes for microbial fuel cell using Mn, Fe-, Co- and Ni-containing platinum group metal-free catalysts. Electrochim Acta 231:115–124 Kumar A, Huan-Hsuan Hsu L, Kavanagh P et al (2017) The ins and outs of microorganism–electrode electron transfer reactions. Nat Rev Chem 1:24 Larrosa-Guerrero A, Scott K, Head IM et al (2010) Effect of temperature on the performance of microbial fuel cells. Fuel 89:3985–3994 Ledezma P, Greenman J, Ieropoulos I (2013a) MFC-cascade stacks maximise COD reduction and avoid voltage reversal under adverse conditions. Bioresour Technol 134:158–165 Ledezma P, Stinchcombe A, Greenman J et al (2013b) The first self-sustainable microbial fuel cell stack. Phys Chem Chem Phys 15:2278–2281 Liu H, Logan BE (2004) Electricity generation using an air-cathode single chamber microbial fuel cell in the presence and absence of a proton exchange membrane. Environ Sci Technol 38:4040–4046 Logan BE (2010) Scaling up microbial fuel cells and other bioelectrochemical systems. Appl Microbiol Biotechnol 85:1665–1671 Logan BE, Hamelers B, Rozendal R et al (2006) Microbial fuel cells: methodology and technology. Environ Sci Technol 40:5181–5192 Logan BE, Call D, Cheng S et al (2008) Microbial electrolysis cells for high yield hydrogen gas production from organic matter. Environ Sci Technol 42:8630–8640 Logan BE, Zikmund E, Yang W et al (2018) The impact of Ohmic resistance on measured electrode potentials and maximum power production in microbial fuel cells. Environ Sci Technol 52:8977–8985 Logan BE, Rossi R, Ragab A et al (2019) Electroactive microorganisms in bioelectrochemical systems. Nat Rev Microbiol 17:307–319 Lu M, Chen S, Babanova S et al (2017) Long-term performance of a 20-L continuous flow microbial fuel cell for treatment of brewery wastewater. J Power Sources 356:274–287 Luo H, Xu G, Lu Y et al (2017) Electricity generation in a microbial fuel cell using yogurt wastewater under alkaline conditions. RSC Adv 7:32826–32832 Martinucci E, Pizza F, Perrino D et al (2015) Energy balance and microbial fuel cells experimentation at wastewater treatment plant Milano-Nosedo. Int J Hydrog Energy 40:14683–14689 Mathuriya AS, Yakhmi JV (2014) Microbial fuel cells to recover heavy metals. Environ Chem Lett 12(4):483–494 Mathuriya AS, Yakhmi JV (2016) Microbial fuel cells – applications for generation of electrical power and beyond. Crit Rev Microbiol 42:127–143

906

C. Santoro et al.

Mehrotra P (2016) Biosensors and their applications – a review. J Oral Biol Craniofac Res 6(2):153–159 Melhuish C, Ieropoulos I, Greenman J et al (2006) Energetically autonomous robots: food for thought. Auton Robot 21:187–198 Merlin Christy P, Gopinath LR, Divya D (2014) A review on anaerobic decomposition and enhancement of biogas production through enzymes and microorganisms. Renew Sust Energ Rev 34:167–173 Modin O, Wilén BM (2012) A novel bioelectrochemical BOD sensor operating with voltage input. Water Res 46:6113–6120 Mustakeem (2015) Electrode materials for microbial fuel cells: nanomaterial approach. Mater Renew Sustain Energy 4:22 Ni G, Christel S, Roman P et al (2016) Electricity generation from an inorganic sulfur compound containing mining wastewater by acidophilic microorganisms. Res Microbiol 167:568–575 Oh SE, Logan BE (2006) Proton exchange membrane and electrode surface areas as factors that affect power generation in microbial fuel cells. App Microbiol Biotechnol 70:162–169 Pandey P, Shinde VN, Deopurkar RL et al (2016) Recent advances in the use of different substrates in microbial fuel cells toward wastewater treatment and simultaneous energy recovery. Appl Energy 168:706–723 Pant D, Van Bogaert G, Diels L et al (2010a) A review of the substrates used in microbial fuel cells (MFCs) for sustainable energy production. Bioresour Technol 101:1533–1543 Pant D, Van Bogaert G, De Smet M et al (2010b) Use of novel permeable membrane and air cathodes in acetate microbial fuel cells. Electrochim Acta 55:7710–7716 Pant D, Van Bogaert G, Alvarez-Gallego Y et al (2016) Evaluation of bioelectrogenic potential of four industrial effluents as substrate for low cost microbial fuel cells operation. Environ Eng Manag J 51(8):1897–1904 Papaharalabos G, Greenman J, Melhuish C et al (2013) Increased power output from micro porous layer (MPL) cathode microbial fuel cells (MFC). Int J Hydrog Energy 38:11552–11558 Papaharalabos G, Stinchcombe A, Horsfield I et al (2017) Autonomous energy harvesting and prevention of cell reversal in MFC stacks. J Electrochem Soc 164(3):H3047–H3051 Pasternak G, Greenman J, Ieropoulos (2017) Self-powered, autonomous Biological Oxygen Demand biosensor for online water quality monitoring. Sens Actuators B Chem 244:815–822 Peixoto L, Min B, Martins G et al (2011) In situ microbial fuel cell-based biosensor for organic carbon. Bioelectrochemistry 81:99–103 Pham TH, Rabaey K, Aelterman P et al (2006) Microbial fuel cells in relation to conventional anaerobic digestion technology. Eng Life Sci 6:285–292 Philamore H, Rossiter J, Stinchcombe A et al (2015) Row-bot: an energetically autonomous artificial water boatman. 2015 IEEE/RSJ international conference on Intelligent Robots and Systems (IROS). Sept 28 – Oct 2, 2015. Hamburg Philamore H, Ieropoulos I, Stinchcombe A et al (2016) Toward energetically autonomous foraging soft robots. Soft Robot 3:186–197 Pocaznoi D, Calmet A, Etcheverry L et al (2012) Stainless steel is a promising electrode material for anodes of microbial fuel cells. Energy Environ Sci 5:9645–9652 Potter MC (1911) Electrical effects accompanying the decomposition of organic compounds. Proc R Soc B Biol Sci 84:260–276 Rabaey K, Rozendal RA (2010) Microbial electrosynthesis – revisiting the electrical route for microbial production. Nat Rev Microbiol 8:706 Rojas-Carbonell S, Santoro C, Serov A et al (2017) Transition metal-nitrogen-carbon catalysts for oxygen reduction reaction in neutral electrolyte. Electrochem Commun 75:38–42 Rojas-Carbonell S, Artyushkova K, Serov A et al (2018) Effect of pH on the activity of platinum group metal-free catalysts in oxygen reduction reaction. ACS Catal 8:3041–3053 Rossiter J, Philamore H, Stinchcombe A et al (2015) Row-bot: an energetically autonomous artificial water boatman. In: Proceedings of 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). pp 3888–3893

35

Microbial Fuel Cells, Concept, and Applications

907

Roy S, Schievano A, Pant D (2016) Electro-stimulated microbial factory for value added product synthesis. Bioresour Technol 213:129–139 Salar-García MJ, Ortiz-Martínez VM, de los Rios AP et al (2015) A method based on impedance spectroscopy for predicting the behavior of novel ionic liquid-polymer inclusion membranes in microbial fuel cells. Energy 89:648–654 Salar-García MJ, Ortiz-Martínez VM, Baicha Z et al (2016) Scaled-up continuous up-flow microbial fuel cell based on novel embedded ionic liquid-type membrane-cathode assembly. Energy 101:113–120 Santoro C, Serov A, Stariha L et al (2016a) Iron based catalysts from novel low-cost organic precursors for enhanced oxygen reduction reaction in neutral media microbial fuel cells. Energy Environ Sci 9:2346–2353 Santoro C, Babanova S, Erable B, Schuler A, Atanassov P (2016b) Bilirubin oxidase based enzymatic air-breathing cathode: operation under pristine and contaminated conditions. Bioelectrochemistry 108:1–7 Santoro C, Mohidin AF, Grasso LL et al (2016c) Sub-toxic concentrations of volatile organic compounds inhibit extracellular respiration of Escherichia coli cells grown in anodic bioelectrochemical systems. Bioelectrochem 112:173–177 Santoro C, Arbizzani C, Erable B et al (2017) Microbial fuel cells: from fundamentals to applications. A Rev J Power Sources 356:225–244 Santoro C, Rojas-Carbonell S, Awais R et al (2018) Influence of platinum group metal-free catalyst synthesis on microbial fuel cell performance. J Power Sources 375:11–20 Schievano A, Colombo A, Grattieri M et al (2017) Floating microbial fuel cells as energy harvesters for signal transmission from natural water bodies. J Power Sources 340:80–88 Schievano A, Colombo A, Cossettini A et al (2018) Single-chamber microbial fuel cells as on-line shock-sensors for volatile fatty acids in anaerobic digesters. Waste Manag 71:785–791 Shantaram A, Beyenal H, Veluchamy RRA et al (2005) Wireless sensors powered by microbial fuel cells. Environ Sci Technol 39:5037–5042 Shen YJ, Lefebvre O, Tan Z et al (2012) Microbial fuel-cell-based toxicity sensor for fast monitoring of acidic toxicity. Water Sci Technol 65(7):1223 Shen YJ, Wang M, Chang IS et al (2013) Effect of shear rate on the response of microbial fuel cell toxicity sensor to cu(II). Bioresour Technol 136:707–710 Sonawane JM, Adeloju SB, Ghosh PC (2017) Landfill leachate: A promising substrate for microbial fuel cells. Int J Hydrogen Energy 42:23794–23798 Song N, Yan Z, Xu H et al (2019) Development of a sediment microbial fuel cell-based biosensor for simultaneous online monitoring of dissolved oxygen concentrations along various depths in lake water. Sci Tot Environ 673:272–280 Stein NE, Hamelers HV, Straten GV et al (2012) Effect of toxic components on microbial fuel cell-polarization curves and estimation of the type of toxic inhibition. Biosensors 2:255–268 Stoll ZA, Dolfing J, Xu P (2018) Minimum performance requirements for microbial fuel cells to achieve energy-neutral wastewater treatment. Water 10:243 Su L, Jia W, Hou C et al (2011) Microbial biosensors: a review. Biosens Bioelectron 26:1788–1799 Tender LM, Gray SA, Groveman E et al (2008) The first demonstration of a microbial fuel cell as a viable power supply: powering a meteorological buoy. J Power Sources 179 (2):571–575 Tian Y, Mei X, Liang Q, Wu D, Ren N, Xing D (2017) Biological degradation of potato pulp waste and microbial community structure in microbial fuel cells. RSC Adv 7(14):8376–8380 Thurston CF, Bennetto HP, Delaney GM et al (1985) Glucose metabolism in a microbial fuel cell. Stoichiometry of product formation in a Thionine-mediated Proteus vulgaris fuel cell and its relation to coulombic yields. Microbiol 131:1393–1401 Tront JM, Fortner JD, Plötze M et al (2008) Microbial fuel cell biosensor for in situ assessment of microbial activity. Biosens Bioelectron 24:586–590 Ucar D, Zhang Y, Angelidaki I (2017) An overview of electron acceptors in microbial fuel cells. Front Microbiol 8:643

908

C. Santoro et al.

Varcoe JR, Atanassov P, Dekel DR (2014) Anion-exchange membranes in electrochemical energy systems. Energy Environ Sci 7:3135–3191 Venkata Mohan S, Mohanakrishna G, Sarma PN (2010) Composite vegetable waste as renewable resource for bioelectricity generation through non-catalyzed open-air cathode microbial fuel cell. Bioresour Technol 101:970–976 Videla HA, Herrera LK (2014) Studies in surface science and catalysis. Chapter 7 Biocorrosion. Elsevier Book series volume 151:193–218 Virdis B, Rabaey K, Rozendal RA et al (2010) Simultaneous nitrification, denitrification and carbon removal in microbial fuel cells. Water Res 44:2970–2980 Walter XA, Gajda I, Forbes S et al (2016) Scaling-up of a novel, simplified MFC stack based on a self-stratifying urine column. Biotechnol Biofuels 9(1):93 Walter XA, Stinchcombe A, Greenman J et al (2017) Urine transduction to usable energy: a modular MFC approach for smartphone and remote system charging. Appl Energy 192:575–581 Walter XA, Merino-Jiménez I, Greenman J et al (2018) PEE POWER ® urinal II – urinal scale-up with microbial fuel cell scale-down for improved lighting. J Power Sources 392:150–158 Wang C, Jiang H (2019) Real-time monitoring of sediment bulking through a multi-anode sediment microbial fuel cell as reliable biosensor. Sci Tot Environ 697:134009 Wang Z, Cao C, Zheng Y, Chen S et al (2014) Abiotic oxygen reduction reaction catalysts used in microbial fuel cells. Chem Electro Chem 1:1813–1821 Wang H, Park JD, Ren ZJ (2015) Practical energy harvesting for microbial fuel cells: a review. Environ Sci Technol 49:3267–3277 Wang Z, Mahadevan GD, Wu Y et al (2017) Progress of air-breathing cathode in microbial fuel cells. J Power Sources 356:245–255 Wei J, Liang P, Huang X (2011) Recent progress in electrodes for microbial fuel cells. Bioresour Technol 102:9335–9344 Wilkinson S (2000) “Gastrobots” – benefits and challenges of microbial fuel cells in food powered robot applications. Auton Robot 9(2):99–111 Williams KH, Nevin KP, Franks A et al (2010) Electrode-based approach for monitoring in situ microbial activity during subsurface bioremediation. Environ Sci Technol 44:47–54 Winfield J, Ieropoulos I, Greenman J (2012) Investigating a cascade of seven hydraulically connected microbial fuel cells. Bioresour Technol 110:245–250 Winfield J, Ieropoulos I, Rossiter J et al (2013) Biodegradation and proton exchange using natural rubber in microbial fuel cells. Biodegradation 24:733–739 Winfield J, Gajda I, Greenman J et al (2016) A review into the use of ceramics in microbial fuel cells. Bioresour Technol 215:296–303 Xu D, Gu T (2014) Carbon source starvation triggered more aggressive corrosion against carbon steel by the Desulfovibrio vulgaris biofilm. Int Biodeterior Biodegrad 91:74–81 Xu D, Li Y, Song F, Gu T (2013) Laboratory investigation of microbiologically influenced corrosion of C1018 carbon steel by nitrate reducing bacterium bacillus licheniformis. Corros Sci 77:385–390 Xu Z, Liu Y, Williams et al (2016) Disposable self-support paper-based multi-anode microbial fuel cell (PMMFC) integrated with power management system (PMS) as the real time “shock” biosensor for wastewater. Biosens Bioelectron 85:232–239 Yamashita T, Hayashi T, Iwasaki H et al (2019) Ultra-low-power energy harvester for microbial fuel cells and its application to environmental sensing and long-range wireless data transmission. J Power Sources 430:1–11 Yang H, Zhou M, Liu M et al (2015) Microbial fuel cells for biosensor applications. Biotechnol Lett 37:2357–2364 You J, Greenman J, Melhuish C et al (2016) Electricity generation and struvite recovery from human urine using microbial fuel cells. J Chem Technol Biotechnol 91:647–654 You J, Greenman J, Ieropoulos I (2018) Novel analytical microbial fuel cell design for rapid in situ optimisation of dilution rate and substrate supply rate, by flow, volume control and anode placement. Energies 11:2377

35

Microbial Fuel Cells, Concept, and Applications

909

Yuan H, Hou Y, Abu-Reesh IM et al (2016) Oxygen reduction reaction catalysts used in microbial fuel cells for energy-efficient wastewater treatment: a review. Mater Horiz 3:382–401 Zeng L, Li X, Shi Y et al (2017) FePO4 based single chamber air-cathode microbial fuel cell for online monitoring levofloxacin. Biosens Bioelectron 91:367–373 Zhang Y, Angelidaki I (2011) Submersible microbial fuel cell sensor for monitoring microbial activity and BOD in groundwater: focusing on impact of anodic biofilm on sensor applicability. Biotechnol Bioeng 108:2339–2347 Zhang Y, Angelidaki I (2012) A simple and rapid method for monitoring dissolved oxygen in water with a submersible microbial fuel cell (SBMFC). Biosens Bioelectron 38:189–194 Zhang F, Cheng S, Pant D et al (2009) Power generation using an activated carbon and metal mesh cathode in a microbial fuel cell. Electrochem Commun 11:2177–2179 Zhang L, Fu G, Zhang Z (2019) Simultaneous nutrient and carbon removal and electricity generation in self-buffered biocathode microbial fuel cell for high-salinity mustard tuber wastewater treatment. Bioresour Technol 272:105–113 Zhu X, Wu G, Lu N et al (2017) A miniaturized electrochemical toxicity biosensor based on graphene oxide quantum dots/carboxylated carbon nanotubes for assessment of priority pollutants. J Hazardous Mater 324:272–280 Zhuang L, Yuan Y, Wang Y et al (2012) Long-term evaluation of a 10-liter serpentine-type microbial fuel cell stack treating brewery wastewater. Bioresour Technol 123:406–412

Organic Matter BOD Biosensor Monitoring

36

Akihito Nakanishi, Wataru Yoshida, and Isao Karube

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Appearance of Microbial BOD Biosensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Progress in Microbial BOD Biosensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Microbial Fuel Cell (MFC)-Based BOD Biosensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Selection of Microorganisms for BOD Biosensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Immobilization Technique for Microorganisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Autonomous Microbial BOD Biosensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions and Future Directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

912 912 913 915 919 919 921 922 923

Abstract

From a proposal of a microbial Biochemical Oxygen Demand (BOD) biosensor by Karube et al. in 1977, components (recognition elements, transducer, etc.) have so far been developed from various perspectives such as microbiology and material engineering. As a result, recent progresses in biosensors have resulted in high sensitivity, compact body, and a wide spectrum of targets. Furthermore, self-powered microbial BOD biosensors are currently developed as independent devices even when electric power is not supplied. In this section, recent progress and development in microbial BOD biosensors are summarized based on actual reports on microbial BOD biosensors. Keywords

Dissolved oxygen · Biochemical oxygen demand · Assimilable organic compounds · BOD biosensor · Microorganisms A. Nakanishi · W. Yoshida · I. Karube (*) School of Bioscience and Biotechnology, Tokyo University of Technology, Tokyo, Japan e-mail: [email protected]; [email protected]; [email protected]; [email protected] © Springer Nature Switzerland AG 2022 G. Thouand (ed.), Handbook of Cell Biosensors, https://doi.org/10.1007/978-3-030-23217-7_95

911

912

A. Nakanishi et al.

Introduction Anthropogenic activity has led to a significant increase in organic carbon concentration. As a result, the quality of rivers and lakes has changed over the past 40 years (Pasternak et al. 2017). A considerable amount of wastewater is produced as a byproduct of the paper-producing process (60 m3 of wastewater per ton of produced paper) in the paper manufacturing industry. If wastewater is discharged to the environment without appropriate treatment processes, the environment becomes significantly polluted (Raud et al. 2011; Thompson et al. 2001). The accumulation of hazardous matters in rivers and lakes negatively affects biodiversity and the ecosystem (Gosset et al. 2016) and thus, the problem of organic contamination in these water environments is very serious in some countries (Oturan 2014). Biochemical Oxygen Demand (BOD) is widely used as a parameter to detect the content of biodegradable organic pollutants in wastewater (Liu et al. 2016; Zhao et al. 2017). The European Union has proposed the evaluation of water quality using BOD as a criterion, defining the BOD concentration range from class I to V (3–7 mg/L) (Environmental standard of the European Union 2008). China advocates that the standard water quality should be defined as lower than 10 mg/L BOD (National standard of the People’s Republic of China 2014). In evaluating such a low BOD, the construction of a system that precisely detects BOD is very important for monitoring the water quality (Hu et al. 2017). So far, organic contaminants in wastewater have been detected based on a method of evaluating the BOD developed in 1936 (APHA 1985; Cheng et al. 2014; Karube et al. 1977; Liang et al. 2018). The classical and typical evaluation of BOD, which is closely related to the organic matter in wastewater, is carried out by measuring the amount of dissolved oxygen in an offline-incubated culture for 5–7 days as BOD5–7. This method, however, is inconvenient and inaccurate because of long-term culture steps and complicated methods, amount of time required, poor reproducibility, and requirement of proficient operation skills. Consequently, the method cannot be used as an on-time and on-site monitoring tool (APHA 1985, 2012; Cheng et al. 2014; Liang et al. 2018; Pasternak et al. 2017; Raud et al. 2011; Zhao et al. 2017). Evaluation methods in wastewater treatment systems require rapidity, accuracy, reproducibility, and convenience, meaning that BOD5–7 evaluation is not suitable as an environmental monitoring tool (Raud et al. 2011). In order to solve these problems, novel BOD sensors using microorganisms have been developed to evaluate organic pollutants in wastewater (Karube et al. 1977; Zhao et al. 2017). Since the invention of microbial BOD biosensors in 1977 (Karube et al. 1977), BOD biosensors using microorganisms have been vigorously researched and developed as tools that can quickly and easily monitor the dissolved oxygen (DO) in wastewater (Chee et al. 1999; Liu et al. 2000; Riedel et al. 1998; Sangeetha et al. 1996).

Appearance of Microbial BOD Biosensor Clark-type microbial BOD biosensors basically remain as the main structure consisting of a DO sensor as an electrochemical transducer and membraneimmobilized microorganisms as a biological element, although various types of

36

Organic Matter BOD Biosensor Monitoring

913

BOD biosensors have been developed to date (Thévenot et al. 2001). Under diffusion of DO and substrates via the microbial membrane, the microbial BOD biosensor measures the DO that results from respiration by immobilized microorganisms (Raud et al. 2011). Adding substrates into the system, the respiration rate of the microorganism in the microbial membrane increases, so that the oxygen sensor equipped on the microbial BOD biosensor monitors the decrease in DO (Raud et al. 2011). The BOD could be calculated from the decrease in DO depending on the correlation between the output signal of the DO sensor, the concentration of the target substrate, and the BOD (Karube et al. 1977; Raud et al. 2011). A number of microbial BOD biosensors have been designed using different microorganisms such as yeast (Hikuma et al. 1979; Yang et al. 1997), Bacillus subtilis (Riedel et al. 1988), and Serratia marcescens (Kim and Kwon 1999) to evaluate BOD by monitoring the amount of DO, and these biosensors have been used in various occasions (e.g., diagnosis of seawater and soil) (Nakamura et al. 2007). However, in terms of costs, sensitivity at low DO, and maintenance of sensitivity, these detection systems are limited by the need for highly sensitive DO electrodes and the low solubility of oxygen in water (8.84 mg O2/L at 1 atm, 20  C) (Cheng et al. 2014; Nakamura et al. 2007).

Progress in Microbial BOD Biosensor Various microbial BOD biosensors have been developed to solve the problems of costs, sensitivity at low DO, and maintenance of sensitivity (Fig. 1). As a nextgeneration type of technology, microbial BOD sensors have been developed by the use of mediators (artificial electron mediators) as electron acceptors (Jordan et al. 2010, 2013; Nakamura et al. 2007; Pasco et al. 2000) (Fig. 2a). The mediators in microbial BOD sensors also accelerate the oxidation reaction of biochemical organic matters, which is why BOD biosensors with mediators have attracted great attention (Zaitseva et al. 2017). For instance, Yoshida et al. developed a BOD biosensor that (c)

(b)

(a) Substrate

Toxin + Substrate

O2

O2

Immobilized microorganisms

Immobilized microorganisms

O2

O2

Oxygen electrode

Oxygen electrode

Monitoring for shifting O2 concentration

Monitoring for shifting O2 concentration

(d) Substrate Mediator (oxidized)

Substrate

Immobilized microorganisms Mediator (reduced)

Immobilized microorganisms CO2, NH3, H+, electron etc

Electrode Monitoring for shifting reduced-mediator concentration

Detectors Monitoring for shifting target concentration

Fig. 1 Principle of microbial biosensor. (a) Respiration activity measurement type for assailable compounds; (b) Respiration activity measurement type for toxic compounds; (c) Mediator measurement type; (d) Metabolite measurement type

914

A. Nakanishi et al.

Improving microbial usability on microbial BOD-biosensor (a) By using mediators as electron acceptor Single mediator system Popular scheme of mediator system (for prokaryotic cell) Mediator In a cell

(b) By using electric current generated by microbial activities in situ MFC-based BOD biosensors

Reduced enzyme

Ferricyanide e.g. P. fluorescens (Yoshida et al. 2000)

(Jordan et al. 2010; Jordan et al. 2013; Nakamura et al. 2007; Pasco et al. 2000) Double-mediator system (for eukaryotic cell) Mediators

Organic substrate H+ CO2 e-

MFC-based BOD biosensor

H2O Cathode

Oxidized enzyme

Biofilm containing microorganisms

Anode

Mediator

Electrode

e-

Transceiver unit

Processing unit

Sensing unit External load

H+ O2 e-

+

+ For analyzing output current response

For detecting substrate

For sharing data with the end user

(Chouler and Di Lorenzo 2015; Di Lorenzo et al. 2014) (Chouler and Di Lorenzo 2015) (c) By using microorganisms in industrial wastewater to meet BOD values between BOD5-7 and microbial BOD-biosensor Detector Porous membrane

… + Ferricyanide

Microorganisms assimilating present substrates in waste water

+ Menadione

Ferrocene

e.g. S. cerevisiae (Nakamura et al. 2007)

Ring

Methylene blue

Porous membrane

e.g. D. hansenii (Zaitseva et al. 2017)

e.g. Paenibacillus sp.-based biosensors (Raud et al. 2011)

(d) By using immobilization technique for microorganisms Popular use of immobilized organisms as sensing element

PVA system

Nylon mesh + Immobilized organisms

Nylon mesh (Zhao et al. 2017)

N-vinylpyrrolidone

PVA

PVA

Ring

䞉 Chemical and biological stability 䞉 Non-toxicity 䞉 Biological compatibility (Marks et al. 2007)

Additionally, 䞉 Improved mechanical strength (Arlyapov et al. 2013)

Collagen fiber system + Collagen 䞉 Porous structure 䞉 Excellent water-binding capacity 䞉 Biocompatibility (Liao et al. 2004)

Collagen Additionally, 䞉 Excellent physical and chemical stability 䞉 Anti-biodegradability 䞉 Absorbability of various biomaterials

Zr4+ Zirconium ion

(Zhao et al. 2017)

Biofilm system

B. subtilis Biofilm 䞉 Negative charge in neutral pH region on anode 䞉 Self-maintenans for biofilm (Fein et al. 1997; Li et al. 2013)

B. subtilis Biofilm

+

+ n PPy

Additionally, 䞉Easy delivery of electrons to the electrode using ferrocyanide as electro mediator

Graphene

(Hu et al. 2017)

Fig. 2 Application and progress of microbial BOD-biosensor. (a) Mediator systems; (b) MFC-based BOD biosensor system; (c) Microorganism in situ-using system; (d) Improved immobilizing system

applied ferricyanide to increase the responsiveness of Pseudomonas fluorescens bacteria as a single-mediator system (Yoshida et al. 2000). Ferricyanide was proven to be a highly effective mediator for shuttling electrons from reduced bacterial enzymes to the electrode in the presence of organic matters. On the other hand, ferricyanide as a mediator did not show the same performance in eukaryotes and thus, the mediator system needed to be improved for use in eukaryotic cells (Kaláb and SkládalKalab 1994; Ramsay and Turner 1988). As a result, the mediator system in microbial BOD biosensors was developed into a double-mediator system by using second mediators facilitating electron mediation to eukaryotes as well as prokaryotes such as P. fluorescens (Nakamura et al. 2007). Nakamura et al. developed a doublemediator system that used ferricyanide and menadione as electron acceptors using yeast Saccharomyces cerevisiae and demonstrated a working range of 6.6–220 mg

36

Organic Matter BOD Biosensor Monitoring

915

O2/L with a low sample volume of 560 μL of 15 min (Nakamura et al. 2007). The double-mediator system in the microbial BOD biosensor was additionally developed by Zaitseva et al., showing long-term stability of 43 days, improved sensitivity (2.5 mg O2/L as a lower limit), and improved detection time (minimum of 10 min) using a mediator couple of ferrocene-methylene blue as electron acceptors with eukaryotic yeast Debaryomyces hansenii as a recognition element (Zaitseva et al. 2017). Furthermore, evaluation performed by the microbial BOD biosensor showed excellent correlation (R2 = 0.99) with the value derived from the existing BOD5, resulting in high accuracy as a BOD biosensor (Zaitseva et al. 2017). Other microbial BOD biosensors using transducers aside from DO sensor were also researched and developed. For instance, luminescence-sensing BOD biosensors using Photobacterium phosphoreum (Hyun et al. 1993) and surface photovoltage-sensing BOD biosensors using Trichosporon cutameum (Murakami et al. 1998) were constructed. In particular, the luminous bacterium P. phosphoreum consumes assimilable organic compounds (AOCs) in the intracellular metabolic process and emits light accurately according to its amount so that the microbial BOD biosensor consisting of P. phosphoreum and photodiodes was highly accurate in the evaluation of AOC concentration in wastewater (Hyun et al. 1993; Reynolds and Ahmad 1997). As a result, a linear correlation was shown between the fluorescence intensity detected by the BOD biosensor and the real BOD in wastewater because of the high accuracy (Reynolds and Ahmad 1997). However, microbial BOD sensors using a single organism such as P. phosphoreum have difficulties in evaluating the BOD stably and maintaining the broad detection range because the single organism depends on the metabolic reaction of a specific substrate (Kim et al. 2003a). Membrane-type BOD sensors have other problems of membrane fouling, non-portability, and inadequate design for on-line use (Cheng et al. 2014; Kim et al. 2003a; Nakamura et al. 2007). Therefore, novel ideas are required in the design of detection systems that allow rapid, user-friendly, and accurate on-line measurements of organic matter in wastewater.

Microbial Fuel Cell (MFC)-Based BOD Biosensors Microbial BOD biosensors based on MFCs show the possibility of responding to these demands. MFC-based BOD biosensors are simple structures that enable sustainable monitoring of target analytes in water as an in situ and on-line system (Table 1) (Kim et al. 2007b). Since the MFC-based BOD biosensor is similar to the bioreactor-based biosensor and provides the advantages of on-line monitoring of biological processes and corresponding BOD, it is practical and can be an alternative to BOD5 for the evaluation of water quality (Chang et al. 2004; Jouanneau et al. 2014). In MFCs, microorganisms with electrochemical activities oxidize organic matter into carbon dioxide and protons to donate electrons to the anode, whereas appropriate molecules receive the electrons from the cathode to enable the flow of electric current in the circuit (Logan et al. 2006; Pant et al. 2010; Stein et al. 2012c).

916

A. Nakanishi et al.

Table 1 List of MFC-based BOD (COD) biosensors

Microorganism Activated sludge

Enriched microbial consortium Freshly prepared sludge suspension

Divice configuration for microorganisms Biofilm on anode

Anode compartment in double-chamber microbial fuel cells Anode compartment in double-chamber microbial fuel cells

Electrochemically active microorganisms Cells in waste water

Anode compartment in double-chamber microbial fuel cells Single-chamber microbial fuel cells

Cells in ground water

Submarsible MFC

Target Waste water

Waste water

Synthetic waste water Waste water from sewage treatment plant Artificial waste water Waste water

Ground water

Limit detection at lower 400 mg O2/L (COD evaluation) 17.2 ppm (BOD evaluation) 10 ppm (BOD evaluation) 92.8 ppm (BOD evaluation) 50 mg O2/L (BOD evaluation) 60 ppm (COD evaluation) 92 mg O2/L (BOD evaluation)

Reference Gil et al. (2003) Kim et al. (2003a) Kim et al. (2003b)

Moon et al. (2004) Di Lorenzo et al. (2009a) Zhang and Angelidaki (2011)

According to another perspective, MFCs are biological energy transducers resulting from electroactive microorganisms that oxidize organic matter and use electrodes as the final electron acceptor (Pasternak et al. 2017). MFC-based BOD biosensors are able to work based on the principle of biocatalytic response to AOCs in the sample and the generation of electric energy (Rasmussen et al. 2016). For instance, in MFC-based BOD biosensors using CH3COO as an AOC, the number of electrons resulting from the oxidation of acetate ions by the microorganisms immobilized on the electrode correlates with the intensity of the current, such that the current intensity can be evaluated by the acetate ion concentration (Logan et al. 2006; Pant et al. 2010). The amount of electricity in MFCs is directly correlated with the AOC concentration (Pasternak et al. 2017). Research on MFC-based BOD biosensors took the first step to progress with the development of a technique of immobilizing Shewanella putrefaciens, an electrochemically active and metalreducing bacterium, carried out by Kim et al. (1999a, 2002). In these reports, S. putrefaciens was immobilized on the anode in MFC-oxidized AOCs by its metabolism. The BOD could then be evaluated directly by measuring the electron quantity in the circuit (Kim et al. 1999a, 2002). In fact, the current flowing in the MFC-based biosensor is directly related to the metabolic activities of electroactive microorganisms in the anodic biofilm (Di Lorenzo et al. 2014; Kim et al. 1999a, 2002). As these reports have shown, MFC-based BOD biosensors generate electric

36

Organic Matter BOD Biosensor Monitoring

917

current by exchanging electrons via electroactive microorganisms (Geobacter spp., Shewanella spp., etc.) on the solid electrode so that they could be used as sensors (Tront et al. 2008). In another case, the MFC-based BOD biosensor showed the ability to evaluate trace-level AOCs in oxygen-saturated seawater (Cheng et al. 2014). Anodic microorganisms consisting of biofilms in MFC-based BOD biosensors are not only converters of chemical energy into electric energy but also act as a recognition component (bioreceptor) (Chouler and Di Lorenzo 2015) (Fig. 2b). In the MFC-based BOD biosensor, current is evaluated as the measurements of organic matter concentration in the MFCs and the metabolic activity of bacteria, since the decomposition rate of organic matter is related to the rate of metabolism of microorganisms (Stein et al. 2012c). The change in current intensity correlates with a single specific disturbance applied as long as operational parameters such as pH, temperature, and conductivity of feeding solution are kept constant. As a result, the MFC-based BOD biosensor could be used as practical sensors of specific disturbance, including changes in pH, temperature, bioactive compound, fuel concentration, and conductivity (Du et al. 2007; Su et al. 2011). This means that under non-saturated fuel conditions, MFC-based BOD biosensors show the ability to detect only the targeted AOCs because of the correlation between the changes in AOC concentration and the quantity of electrons transferred to the anode electrode, evaluated as output current (Di Lorenzo et al. 2014; Stein et al. 2010). Therefore, MFC-based BOD biosensors have been used for not only AOCs but also toxicants in water (Chouler and Di Lorenzo 2015). Various niche applications of MFCs have been designed but their use as a sensor for water quality evaluation is the most realistic (Das and Mangwani 2010; Su et al. 2011). Up till now, MFCs have been demonstrated to measure BOD in several types of water examples including wastewater (Chang et al. 2004; Di Lorenzo et al. 2009b; Gil et al. 2003; Kim et al. 2003a). In particular, the MFCs used in the biosensors can be an effective way of sensing water in developing countries because they are composed of simple structures and are economical as sensors (Chouler and Di Lorenzo 2015). The MFC-based BOD biosensors are simple in terms of structure and electronic control in design and implementation. Therefore, they do not require an additional transducer to convert a biological response into a signal (Chouler and Di Lorenzo 2015). In these simple MFC-based biosensors, the presence of pollutants in the feeding stream can be detected instantly as a precise change in current in the system (Chouler and Di Lorenzo 2015). In several cases, the use of the sensors in pure culture systems has been reported but their use in media containing a mixture of natural microorganisms has shown high stability and better performance (Kim et al. 1999b, 2003b). Additionally, an electroactive biofilm is spontaneously formed on the biocompatible surface of the anode without requiring a time-consuming immobilization step (Di Lorenzo et al. 2009a). MFC-based BOD biosensors can be used for the continuous on-site monitoring of water quality in real time. Furthermore, the sensors easily exhibited the concept of a self-powered system because of their properties as a cell, showing the possibility to be used stably in remote areas without energy access (Melhuish et al. 2006). On the other hand, there was a report whereby the MFC-based BOD biosensor, as a mediator-free sensor, showed high service life

918

A. Nakanishi et al.

for more than 5 years in 2003 (Kim et al. 2003a). Thus, based on these findings, the MFC-based BOD biosensors have started to be used as practical and conventional tools (Sun et al. 2015). According to many previous reports, MFCs can be used as a biosensor to detect toxicants in water and as a microorganism consortium in which electronic exchange is active (Chang et al. 2005; Feng et al. 2013; Franks and Nevin 2010; Kim et al. 2007a, b; Patil et al. 2010; Logan and Regan 2006). MFC-based biosensors have been studied in recent years for monitoring not only AOCs in wastewater but also toxicants and biologically active components (Chouler and Di Lorenzo 2015). Electrical power output correlates with the existence of toxicants regardless of organic/inorganic substrates (Patil et al. 2010). Although MFCs can generate stable current under constant condition, toxicant contamination in MFCs generally decreases the electric current by affecting the microorganisms (Kim et al. 2007b; Stein et al. 2010). The use of MFCs as a toxicity sensor was proposed from a report by Kim et al. in 2007, and the detection targets were Pb, Hg, diazinon (an organophosphorous pesticide), and polychlorinated biphenyl (Kim et al. 2007b). In order to detect these toxicants continuously, MFC biosensors are used as a shock sensor (Chouler and Di Lorenzo 2015). Since then, the sensitivities of MFC-based biosensors have already been demonstrated against several toxic metals such as Cd2+, which was assessed quantitatively within a concentration range of 0.1–100 μg/L in 12 min (Di Lorenzo et al. 2014). In order to measure BOD concentration in toxicants, MFC-based biosensors have been improved in various ways, and several reports have shown detection coefficients of 0.99 (Di Lorenzo et al. 2009a; Kim et al. 2003a). In the detection and evaluation of toxicants using MFCs, the evaluation stability is important. The enzyme inhibition kinetics of biochemical and electrochemical reactions at the anode were evaluated using the Butler Volmer Monod model developed to describe the polarization (Stein et al. 2012c). Theoretically, this model shows four types of inhibition kinetics of enzymatic reactions as below: (1) inhibition of whole bacteria as an irreversible inhibitor; (2) inhibition of the rate constant correlated between electrochemical and biochemical reactions; (3) inhibition of the ratio of forward to backward reaction rate constants due to oxidation of substrate; and (4) inhibition of substrates to bind redox complex by toxic components as antagonists (Stein et al. 2012c). In order to use MFCs as a toxicant biosensor efficiently, the selection of an optimum MFC-based biosensor is mostly necessary with consideration of such inhibitory effects on the target toxicants (Stein et al. 2012c). Based on this information, various sensors have been researched and developed towards functional improvement in topics such as membranes, external residence, shear rate, single-chamber devices, and miniaturization in order to use MFCs as BOD biosensors to detect toxic substances (Chang et al. 2004; Dávila et al. 2011; Di Lorenzo et al. 2009b, 2014; Karube et al. 1977; Kim et al. 2003a, 2007b; Liu et al. 2014; Min and Logan 2004; Moon et al. 2004; Stein et al. 2010, 2012a, b; Shen et al. 2013; Patil et al. 2010; Zhang and Angelidaki 2011). For instance, in 2010, a study reported the dramatic economical cost cut by using ceramics simultaneously in the construction of a separation membrane and as a structural material (Behera et al. 2010; Ieropoulos et al. 2015; Pasternak et al. 2016).

36

Organic Matter BOD Biosensor Monitoring

919

Selection of Microorganisms for BOD Biosensor Biosensors using microorganisms might be unsuitable for detection of macromolecules and refractory compounds, but the method of conventional BOD5–7 is possible and suitable for their detection (Raud et al. 2011). Such a method causes a discrepancy between the BOD detected by the sensor and the BOD5–7 (Zhao et al. 2017). In the evaluation of wastewater from paper mills and aspen pulp mills, B. subtilisand Paenibacillus sp.-based biosensors yielded over-estimated values of BOD7 (Raud et al. 2011). Therefore, in order to obtain similar results from the BOD biosensor and BOD7, the targeted sample should be decomposed of macromolecules by enzymatic treatment (Chee et al. 2001, 2007; Kwong et al. 1998; Reiss et al. 1998; Tag et al. 2000). To avoid discrepancies in the results in wastewater and complications in the measurement process, the selection of microorganisms is especially important in constructing a reliable BOD biosensor that intends to use microorganisms in industrial wastewater (Raud et al. 2011). The microorganisms in industrial wastewater can usually assimilate persistent substrates in the wastewater so that the values of BOD7 and measurements obtained by the biosensor will be similar (Raud et al. 2011). Semi-specific biosensors have better accuracy and reproducibility in the evaluation of cellulose-derived organic contamination. In particular, Paenibacillus sp.-based biosensors showed superior performance (Raud et al. 2011) (Fig. 2c).

Immobilization Technique for Microorganisms With conventional microbial BOD biosensors, microorganisms must be adjusted to withstand common measurements (Jordan et al. 2013). Even with the use of optimized microorganisms, repeated and complicated operations such as microorganism culture, centrifugation, and washing are normally required before each measurement (Fig. 2d). In turn, these processes are barriers to rapid BOD measurements and environmental monitoring as an online system (Hu et al. 2017). In recent years, economic and convenient microorganism-immobilizing techniques have drawn huge attention for use in BOD in response to these problems (Hu et al. 2017). The microbial film used to immobilize microorganisms is an important element of the microbial BOD sensor that has serious effects on the sensitivity, linearity, response time, and service life of the sensor (Zhao et al. 2017). The sensitivity and durability of microbial receptor elements have been improved with regard to the immobilization of microorganisms, and the development of new biomaterial immobilization methods has been strongly promoted (Arlyapov et al. 2013). Progress in microorganism-immobilizing techniques helps to protect these microorganisms from environmental toxins, resulting in the ease of use. As a result, the technique has advantages in terms of economy and convenience for practical use in microbial BOD biosensors (Dhall et al. 2012; Hu et al. 2017). For immobilization, there are several techniques such as cross-linking, sandwiching, adsorption, and embedding (Jouanneau et al. 2014; Ponomareva et al. 2011). Depending on the reported

920

A. Nakanishi et al.

practical examples to date, these techniques are designed to enable microorganisms to access and react with DO and organic compounds and to pass the electrons produced by their reactions to an electrode (Jouanneau et al. 2014; Ponomareva et al. 2011). Calcium alginate (Kumlanghan et al. 2008; Sergio and Bustos 2009), polycarbonate (Suriyawattanakul et al. 2002), agar (Elkahlout et al. 2017; Raud et al. 2012), poly (carbamoyl) sulfonate (Chan et al. 1999), Al2O3 sol-gel (Jiang et al. 2006), cellulose (Fijałkowski et al. 2015), and poly(vinyl alcohol) (PVA) (Wang et al. 2010) have been used as immobilizing materials for microorganisms (Hu et al. 2017). In particular, gel-based entrapment with PVA was significantly used to immobilize microorganisms because of the PVA’s properties of chemical and biological stability, nontoxicity, and biological compatibility (Marks et al. 2007). On the other hand, the gel-based PVA has no mechanical strength and is not the most ideal component as a biosensor receptor element (Marks et al. 2007). To improve the mechanical strength of PVA, many methods have been developed using UV-irradiation, boronic acid, and N-vinyl pyridine as copolymer components (Wang et al. 2010). These methods have adversely affected the immobilized microorganisms and their reactivities, resulting in decreased sensitivity of the biosensor (Arlyapov et al. 2013). A method of immobilizing microorganisms using N-vinylpyrrolidone as a component to modify PVA has been developed in recent years, and this method was close to solving the problems of immobilization (Arlyapov et al. 2013). The immobilization method with N-vinylpyrrolidone makes it possible to increase the service life by the mechanical strength of the receptor element, and sensitivity was also improved without exhibiting toxicity to the immobilized microorganisms (Alferov et al. 2011; Arlyapov et al. 2013). Arlyapov et al. constructed a microbial BOD biosensor by immobilizing Debaryomyces hansenii using the abovementioned method and the biosensor was evaluated using 34 kinds of substrates. The results shown by the constructed sensor correlated with the collected data from BOD5 (Arlyapov et al. 2013). Many immobilization carriers have shown insufficient performance in terms of porosity and substrate diffusion in biofilms, and because they consist of a thin layer, they result in high cost and low usability (Wang et al. 2010; Zhao et al. 2017). Because of these problems, immobilization of microorganisms using collagen fibers has also been considered (Liao et al. 2004). Collagen fibers, which are natural polymer materials extracted from animal skin (pig hide and cattle hide), have a porous structure, excellent water-binding capacity, and biocompatibility (Liao et al. 2004). In addition, Zr (IV)-loaded collagen fibers (ZrCF), developed to further enhance functionality, has many advantages including excellent physical and chemical stability, anti-biodegradability, and absorbability of various biomaterials (proteins, enzymes, and microorganisms) (Huang et al. 2008). E. coli and S. cerevisiae were immobilized to ZrCF and sandwiched in a nylon mesh to form a biofilm, which was equipped to an oxygen electrode as a component of the biosensor (Zhao et al. 2017). The ZrCF-based BOD biosensor was compared with BOD5 in terms of detection performance against various substrates such as sugar and organic acid as a basic assessment. The sensor was also used to evaluate BOD in river water and synthetic seawater, using synthetic wastewater from The Organisation for

36

Organic Matter BOD Biosensor Monitoring

921

Economic Co-operation and Development as an applied assessment (Zhao et al. 2017). ZrCF showed a long service life for 42 days because of its high stability in practical use in real samples (Zhao et al. 2017). Additionally, the storage time of the biofilm was half a year at 4  C, showing the excellent usability of the biofilm for practical use (Zhao et al. 2017). Recently, the three-dimensional (3D) porous structure of graphene has attracted considerable attention as an application in biosensor technology because of its own internal 3D porous structure and physical stability based on π-π stacking of graphene sheets (Qiu et al. 2017; Wang et al. 2015). Graphene obtained by steam reduction has advantages over the development of sensors based on several functionalities. Not only conventional enzymes (Zanon et al. 2013) but also microorganisms have gained interest as immobilized targets on the graphene structure (Hu et al. 2017). Research on microorganism immobilization with biofilms on electrodes has also advanced (Li et al. 2013). B. subtilis is an eco-friendly biofilm-forming bacterium used for a long time as a conventional strain. The bacterial membrane formed by B. subtilis has a negative charge in the neutral pH region (Fein et al. 1997; Li et al. 2013). On the other hand, polypyrrole (PPy) is a material that greatly contributed to the development of sensors, including biosensors, owing to its excellent chemical stability and biocompatibility (Liu et al. 2012). Depending on its characteristic structure, PPy could be easily used to control and immobilize various biomolecules such as enzymes and DNA by adsorption, covalent bonding, and cross-linking (Jiang and Lin 2005). Additionally, in recent years, PPy has been used as a base carrier not only for adsorption, covalent bonding, and cross-linking but also for Coulomb coupling to a negative charge because of the positive charge possessed by PPy during electro-polymerization (Le et al. 2015). Based on these properties of B. subtilis and PPy, a novel BOD was developed that electronically immobilized B. subtilis with its own biofilm on 3D porous graphene-polypyrrole that delivered electrons to the electrode using ferrocyanide as an electron mediator (Hu et al. 2017). The correlation between the amperometric responses and the standard BOD concentrations showed excellent linearity from 4 to 60 mg O2/L with a lowest detection limit of 1.8 mg at a signal-to-noise ratio of 3 or higher (Hu et al. 2017).

Autonomous Microbial BOD Biosensor BOD biosensors currently require online usability in powerless environments and durability against environmental changes (Pasternak et al. 2017). Thus, self-powered biosensors that enable long-term monitoring without a power supply have been developed. Pasternak et al. developed a BOD biosensor that detects the presence of urine in pure water and autonomously turns on a visual and sound cue (Pasternak et al. 2017). The sensor could supply electric power by electroactive microorganisms and function autonomously for at least 5 months. The system turns ON when the concentration exceeds the minimum threshold of urine concentration (57.7  4.8 mg O2/L). Once the alarm turns ON, the system will not turn OFF until the urine concentration has decreased to 15.3  1.9 mg O2/L.

922

A. Nakanishi et al.

Quick BOD Type α5000 (Central Kagaku Corporation) O2 electrode

O2 detector

Permeated O2

Aqueous phase targeted for BOD analysis

Porous membrane Microorganisms derived from target aqueous phase Spacer Ring

Dissolved O2

Porous membrane

Fig. 3 Typical commercial use of microbial BOD-sensor

A self-powered autonomous BOD biosensor for water quality monitoring has been researched and developed. Recent self-powered BOD biosensors demonstrated high responsiveness, showing enough signal in a short time of several minutes as compared to other detection methods (Chouler and Di Lorenzo 2015; Di Lorenzo et al. 2014) (Fig. 2b). A report has found that the time required to achieve a stable current of 95% was only 2.8 min (Di Lorenzo et al. 2014). The development of the family of self-powered BOD biosensors has been advanced to date. However, the electric current generated in situ was only for the function of the BOD sensor but not for other device functions (Li et al. 2011). Li et al. (2011) demonstrated real self-powered biosensors that generated an electric current in the presence of homoserine-lactones using Pseudomonas aeruginosa mutants and monitored the self-produced electric power in an external device (Li et al. 2011). At present, self-powered devices have been developed as flexible piezo-electric energy harvesters (nanogenerators), and these devices have been developed mainly for implantable electronics in health care (Cheng et al. 2015; Lee et al. 2014). Miyake et al. developed a self-powered biosensor that converted simple sugars such as glucose and fructose into usable energy (electric energy) and functioned as light-emitting diodes (Miyake et al. 2011). MFC-based BOD biosensors based on enzymes have never been used for long-term online monitoring, and their service life is normally several days because of the lifetime of the enzyme (Rubenwolf et al. 2011). On the other hand, the MFC-based BOD biosensors using microorganisms will be expected to be used for long-term monitoring as the microorganism can maintain the sensitivity of the sensor by itself (Pasternak et al. 2017).

Conclusions and Future Directions As described above, the development of microbial BOD biosensors depends on various aspects such as analytical chemistry, physical chemistry, and microbiology resulting that microbial BOD biosensor is already on the market (Fig. 3). In the future, microbial BOD biosensors will be developed and improved with the aim of assessing the environment as the center of research. Sensors will be optimized

36

Organic Matter BOD Biosensor Monitoring

923

to meet diversified needs such as high speed, convenience, on-line usability, and autonomy.

References Alferov VA, Filatova NM, Asulyan LD et al (2011) Formation of the stable receptor element of a biosensor by immobilization of Gluconobacter oxydans bacterial cells in a film of poly(vinyl alcohol) modified by N-vinylpyrrolidone. Izv TulGU Nat Sci 1:210–219 APHA (1985) Standard methods for examination of water and wastewater. American Public Health Association, Washington, DC APHA (2012) Standard methods for the examination of water and wastewater, 19th edn. American Public Health Association, Washington, DC Arlyapov VA, Yudina NY, Asulyan LD et al (2013) BOD biosensor based on the yeast Debaryomyces hansenii immobilized in poly(vinyl alcohol) modified by N-vinylpyrrolidone. Enzyme Microb Technol 53:257–262 Behera M, Jana PS, Ghangrekar MM (2010) Performance evaluation of low cost microbial fuel cell fabricated using earthen pot with biotic and abiotic cathode. Bioresour Technol 101:1183–1189 Chan C, Lehmann M, Tag K et al (1999) Measurement of biodegradable substances using the salttolerant yeast Arxula adeninivorans for a microbial sensor immobilized with poly(carbamoyl) sulfonate (PCS) part I: construction and characterization of the microbial sensor. Biosens Bioelectron 14:131–138 Chang IS, Jang JK, Gil GC et al (2004) Continuous determination of biochemical oxygen demand using microbial fuel cell type biosensor. Biosens Bioelectron 19:607–613 Chang IS, Moon H, Jang JK et al (2005) Improvement of a microbial fuel cell performance as a BOD sensor using respiratory inhibitors. Biosens Bioelectron 20:1856–1859 Chee GJ, Nomura Y, Karube I (1999) Biosensor for the estimation of low biochemical oxygen demand. Anal Chim Acta 379:185–191 Chee GJ, Nomura Y, Ikebukuro K et al (2001) Biosensor for the evaluation of biochemical oxygen demand using photocatalytic pretreatment. Sens Actuators B 80:15–20 Chee GJ, Nomura Y, Ikebukuro K et al (2007) Stopped-flow system with ozonizer for the estimation of low biochemical oxygen demand in environmental samples. Biosens Bioelectron 22: 3092–3098 Cheng L, Quek SB, Cord-Ruwisch R (2014) Hexacyanoferrate-adapted biofilm enables the development of a microbial fuel cell biosensor to detect trace levels of assimilable organic carbon (AOC) in oxygenated seawater. Biotechnol Bioeng 111:2412–2420 Cheng L, Yuan M, Gu L et al (2015) Wireless, power-free and implantable nanosystem for resistance-based biodetection. Nano Energy 15:598–606 Chouler J, Di Lorenzo M (2015) Water quality monitoring in developing countries; can microbial fuel cells be the answer? Biosensors 5:450–470 Das S, Mangwani N (2010) Recent developments in microbial fuel cells: a review. J Sci Ind Res 69:727–731 Dávila D, Esquivel JP, Sabaté N et al (2011) Silicon-based microfabricated microbial fuel cell toxicity sensor. Biosens Bioelectron 26:2426–2430 Dhall P, Kumar R, Kumar A (2012) BOD beads – a ready to use seeding material for estimation of organic load of wastewater. Anal Methods 4:4101–4106 Di Lorenzo M, Curtis TP, Head IM et al (2009a) A single-chamber microbial fuel cell as a biosensor for wastewaters. Water Res 43:3145–3154 Di Lorenzo M, Scott K, Curtis TP et al (2009b) Continuous feed microbial fuel cell using an air cathode and a disc anode stack for wastewater treatment. Energy Fuels 23:5707–5716 Di Lorenzo M, Thomson AR, Schneider K et al (2014) A small-scale air-cathode microbial fuel cell for on-line monitoring of water quality. Biosens Bioelectron 62:182–188

924

A. Nakanishi et al.

Du Z, Li H, Gu T (2007) A state of the art review on microbial fuel cells: a promising technology for wastewater treatment and bioenergy. Biotechnol Adv 25:464–482 Elkahlout K, Alipour S, Eroglu I et al (2017) Long-term biological hydrogen production by agar immobilized Rhodobacter capsulatus, in a sequential batch photobioreactor. Bioprocess Biosyst Eng 40:589–599 Environmental standard of the European Union (2008) Data sheets for surface water quality standards. https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=LEGISSUM%3Al2 8180. Accessed 15 May 2018 Fein JB, Daughney CJ, Yee N et al (1997) A chemical equilibrium model for metal adsorption onto bacterial surfaces. Geochim Cosmochim Acta 61:3319–3328 Feng Y, Barr W, Harper WF (2013) Neural network processing of microbial fuel cell signals for the identification of chemicals present in water. J Environ Manag 120:84–92 Fijałkowski K, Peitler D, Rakoczy R et al (2015) Survival of probiotic lactic acid bacteria immobilized in different forms of bacterial cellulose in simulated gastric juices and bile salt solution. LWT Food Sci Technol 68:322–328 Franks AE, Nevin KP (2010) Microbial fuel cells, a current review. Energies 3:899–919 Gil GC, Chang IS, Kim BH et al (2003) Operational parameters affecting the performance of a mediator-less microbial fuel cell. Biosens Bioelectron 18:327–324 Gosset A, Ferro Y, Durrieu C (2016) Methods for evaluating the pollution impact of urban wet weather discharges on biocenosis: a review. Water Res 89:330–354 Hikuma M, Suzuki H, Yasuda T et al (1979) Amperometric estimation of BOD by using living immobilized yeast. Eur J Appl Microbiol Biotechnol 8:289–297 Hu J, Li Y, Gao G et al (2017) A mediated BOD biosensor based on immobilized B. subtilis on three-dimensional porous graphene-polypyrrole composite. Sensors 17. pii: E2594 Huang X, Jiao LM, Liao XP et al (2008) Adsorptive removal of As(III) from aqueous solution by Zr(IV)-loaded collagen fiber. Ind Eng Chem Res 47:5623–5628 Hyun CK, Tamiya E, Takeuchi T et al (1993) Novel BOD sensor based on bacterial luminescence. Biotechnol Bioeng 41:1107–1111 Ieropoulos I, Stinchcombe A, Gajda I et al (2015) Pee power urinal – microbial fuel cell technology field trials in the context of sanitation. Environ Sci Water Res Technol 2:336–343 Jiang X, Lin X (2005) Overoxidized polypyrrole film directed DNA immobilization for construction of electrochemical micro-biosensors and simultaneous determination of serotonin and dopamine. Anal Chim Acta 537:145–151 Jiang Y, Xiao LL, Zhao L et al (2006) Optical biosensor for the determination of BOD in seawater. Talanta 70:97–103 Jordan MA, Welsh DT, Teasdale PR et al (2010) A ferricyanide-mediated activated sludge bioassay for fast determination of the biochemical oxygen demand of wastewaters. Water Res 44:5981–5988 Jordan MA, Welsh DT, John R et al (2013) A sensitive ferricyanide-mediated biochemical oxygen demand assay for analysis of wastewater treatment plant influents and treated effluents. Water Res 47:841–849 Jouanneau S, Recoules L, Durand MJ et al (2014) Methods for assessing biochemical oxygen demand (BOD): a review. Water Res 49:62–82 Kaláb T, SkládalKalab P (1994) Evaluation of mediators for development of amperometric microbial bioelectrodes. Electroanalysis 6:1004–1008 Karube I, Matsunaga T, Mitsuda S et al (1977) Microbial electrode BOD sensors. Biotechnol Bioeng 19:1535–1547 Kim MN, Kwon HS (1999) Biochemical oxygen demand sensor using Serratia marcescens LSY 4. Biosens Bioelectron 14:1–7 Kim BH, Kim HJ, Hyun MS et al (1999a) Direct electrode reaction of Fe(III) reducing bacterium, Shewanella putrefaciens. J Microbiol Biotechnol 9:127–131 Kim H, Hyun M, Chang I et al (1999b) A microbial fuel cell type lactate biosensor using a metalreducing bacterium, Shewanella putrefaciens. J Microbiol Biotechnol 9:365–367

36

Organic Matter BOD Biosensor Monitoring

925

Kim HJ, Park HS, Hyun MS et al (2002) A mediator-less microbial fuel cell using a metal reducing bacteirium, Shewanella putrefaciens. Enzyme Microb Technol 30:145–152 Kim BH, Chang IS, Gil GC et al (2003a) Novel BOD (biological oxygen demand) sensor using mediator-less microbial fuel cell. Biotechnol Lett 25:541–545 Kim M, Youn SM, Shin SH et al (2003b) Practical field application of a novel BOD monitoring system. J Environ Monit 5:640–643 Kim BH, Chang IS, Gadd GM (2007a) Challenges in microbial fuel cell development and operation. Appl Microbiol Biotechnol 76:485–494 Kim M, Hyun MS, Gaddb GM et al (2007b) A novel biomonitoring system using microbial fuel cells. J Environ Monit 9:1323–1328 Kumlanghan A, Kanatharana P, Asawatreratanakul P et al (2008) Microbial BOD sensor for monitoring treatment of wastewater from a rubber latex industry. Enzyme Microb Technol 42:483–491 Kwong AWK, Chan CY, Renneberg R (1998) Monitoring biodegradable substances with highmolecular content with a microbial sensor. Anal Lett 31:2309–2325 Le DQ, Takai M, Suekuni S (2015) Development of an observation platform for bacterial activity using polypyrrole films doped with bacteria. Anal Chem 87:4047–4052 Lee SH, Jeong CK, Hwang GT et al (2014) Self-powered flexible inorganic electronic system. Nano Energy 14:111–125 Li Z, Rosenbaum M, Venkataraman A et al (2011) Bacteria-based AND logic gate: a decisionmaking and self-powered biosensor. Chem Commun 47:3060–3062 Li J, Yang R, Chen K et al (2013) Mediated microbial biosensor using Bacillus subtilis for wastewater BOD measurement. Biochem Eng J 8:219–225 Liang Q, Yamashita T, Yamamoto-Ikemoto R et al (2018) Flame-oxidized stainless-steel anode as a probe in bioelectrochemical system-based biosensors to monitor the biochemical oxygen demand of wastewater. Sensors 18. pii: E607 Liao XP, Lu Z, Du X et al (2004) Collagen fiber immobilized Myrica rubra tannin and its adsorption to UO22+. Environ Sci Technol 38:324–328 Liu J, Björnsson L, Mattiasson B (2000) Immobilised activated sludge based biosensor for biochemical oxygen demand measurement. Biosens Bioelectron 14:883–893 Liu J, Wang Z, Zhao Y et al (2012) Three-dimensional graphene-polypyrrole hybrid electrochemical actuator. Nanoscale 4:7563–7568 Liu B, Lei Y, Li B (2014) A batch-mode cube microbial fuel cell based “shock” biosensor for wastewater quality monitoring. Biosens Bioelectron 62:308–314 Liu C, Li Z, Jiang D et al (2016) Demonstration study of biofilm reactor based rapid biochemical oxygen demand determination of surface water. Sens Bio-Sens Res 8:8–13 Logan BE, Regan JM (2006) Microbial fuel cells: challenges and applications. Environ Sci Technol 40:5172–5180 Logan BE, Hamelers B, Rozendal R et al (2006) Microbial fuel cells: methodology and technology. Environ Sci Technol 40:5181–5192 Marks RS, Cullen DC, Karube I et al (eds) (2007) Handbook of biosensors and biochips. Hoboken. 356 Melhuish C, Ieropoulos I, Greenman J et al (2006) Energetically autonomous robots: food for thought. Auton Robots 21:187–198 Min B, Logan BE (2004) Continuous electricity generation from domestic wastewater and organic substrates in a flat plate microbial fuel cell. Environ Sci Technol 38:5809–5814 Miyake T, Haneda K, Nagai N et al (2011) Enzymatic biofuel cells designed for direct power generation from biofluids in living organisms. Energy Environ Sci 4:5008–5012 Moon H, Chang IS, Kang KH et al (2004) Improving the dynamic response of a mediator-less microbial fuel cell as a biochemical oxygen demand (BOD) sensor. Biotechnol Lett 26: 1717–1721 Murakami Y, Kikuchi T, Yamaura A et al (1998) An organic pollution sensor based on surface photovoltage. Sens Actuators B 53:163–172

926

A. Nakanishi et al.

Nakamura H, Suzuki K, Ishikuro H et al (2007) A new BOD estimation method employing a double-mediator system by ferricyanide and menadione using the eukaryote Saccharomyces cerevisiae. Talanta 72:210–216 National standard of the People’s Republic of China, environmental quality standards for surface water (2014.) http://english.gov.cn/archive/laws_regulations/2014/08/23/content_2814749830 42375.htm. Accessed 15 May 2018 Oturan MA (2014) Electrochemical advanced oxidation technologies for removal of organic pollutants from water. Environ Sci Pollut Res 21:8333–8335 Pant D, Bogaert GV, Diels L et al (2010) A review of the substrates used in microbial fuel cells (MFCs) for sustainable energy production. Bioresour Technol 101:1533–1543 Pasco N, Baronian K, Jeffries C et al (2000) Biochemical mediator demand-a novel rapid alternative for measuring biochemical oxygen demand. Appl Microbiol Biotechnol 53:613–618 Pasternak G, Greenman J, Ieropoulos I (2016) Comprehensive study on ceramic membranes for low-cost microbial fuel cells. ChemSusChem 9:88–96 Pasternak G, Greenmana J, Ieropoulos I (2017) Self-powered, autonomous Biological Oxygen Demand biosensor foronline water quality monitoring. Sens Actuators B Chem 244:815–822 Patil S, Harnisch F, Schröder U (2010) Toxicity response of electroactive microbial biofilms – a decisive feature for potential biosensor and power source applications. ChemPhysChem 11:2834–2837 Ponomareva ON, Arlyapov VA, Alferov VA (2011) Microbial biosensors for detection of biological oxygen demand (a review). Appl Biochem Microbiol 47:1–11 Qiu H, Guan Y, Luo P et al (2017) Recent advance in fabricating monolithic 3D porous graphene and their applications in biosensing and biofuel cells. Biosens Bioelectron 89:85–95 Ramsay G, Turner APF (1988) Development of an electrochemical method for the rapid determination of microbial concentration and evidence for the reaction mechanism. Anal Chim Acta 215:61–69 Rasmussen M, Abdellaoui S, Minteer SD (2016) Enzymatic biofuel cells: 30 years of critical advancements. Biosens Bioelectron 76:91–102 Raud M, Tutt M, Jõgi E et al (2011) BOD biosensors for pulp and paper industry wastewater analysis. Environ Sci Pollut Res Int 19:3039–3045 Raud M, Toomas T, Eerik J et al (2012) Comparative study of semi-specific Aeromonas hydrophila and universal Pseudomonas fluorescens biosensors for BOD measurements in meat industry wastewaters. Enzyme Microb Technol 50:221–226 Reiss M, Heibges A, Metzger J et al (1998) Determination of BOD-values of starch-containing waste water by a BOD biosensor. Biosens Bioelectron 13:1083–1090 Reynolds DM, Ahmad SR (1997) Rapid and direct determination of wastewater BOD values using a fluorescence technique. Water Res 31:2012–2018 Riedel K, Renneberg R, Kühn M et al (1988) A fast estimation of biochemical oxygen demand using microbial sensors. Appl Microbiol Biotechnol 28:316–318 Riedel K, Lehmann M, Tag K et al (1998) Arxula adeninivorans based sensor for the estimation of BOD. Anal Lett 31:1–12 Rubenwolf S, Kerzenmacher S, Zengerle R et al (2011) Strategies to extend the lifetime of bioelectrochemical enzyme electrodes for biosensing and biofuel cell applications. Appl Microbiol Biotechnol 89:1315–1322 Sangeetha S, Sugandhi G, Murugesan M et al (1996) Torulopsis candida based sensor for the estimation of biochemical oxygen demand and its evaluation. Electroanalysis 8:698–701 Sergio AMD, Bustos TY (2009) Biodegradation of wastewater pollutants by activated sludge encapsulated inside calcium-alginate beads in a tubular packed bed reactor. Biodegradation 20:709–712 Shen Y, Wang M, Chang IS et al (2013) Effect of shear rate on the response of microbial fuel cell toxicity sensor to Cu(II). Bioresour Technol 136:707–710

36

Organic Matter BOD Biosensor Monitoring

927

Stein NE, Hamelers HVM, Buisman CNJ (2010) Stabilizing the baseline current of a microbial fuel cell-based biosensor through overpotential control under non-toxic conditions. Bioelectrochemistry 78:87–91 Stein NE, Hamelers HV, Buisman CN (2012a) Influence of membrane type, current and potential on the response to chemical toxicants of a microbial fuel cell based biosensor. Sens Actuators B Chem 163:1–7 Stein NE, Hamelers HV, Buisman CN (2012b) The effect of different control mechanisms on the sensitivity and recovery time of a microbial fuel cell based biosensor. Sens Actuators B Chem 171–172:816–821 Stein NE, Hamelers HVM, van Straten G et al (2012c) Effect of toxic components on microbial fuel cell-polarization curves and estimation of the type of toxic inhibition. Biosensors 2:255–268 Su L, Jia W, Hou C et al (2011) Microbial biosensors: a review. Biosens Bioelectron 26:1788–1799 Sun JZ, Kingori GP, Si RW et al (2015) Microbial fuel cell-based biosensors for environmental monitoring: a review. Water Sci Technol 71:801–809 Suriyawattanakul L, Surareungchai W, Sritongkam P et al (2002) The use of co-immobilization of Trichosporon cutaneum and Bacillus licheniformis for a BOD sensor. Appl Microbiol Biotechnol 59:40–44 Tag K, Kwong AWK, Lehmann M et al (2000) Fast detection of high molecular weight substances in wastewater based on an enzymatic hydrolysis combined with the Arxula BOD sensor system. Chem Technol Biotechnol 75:1080–1082 Thévenot DR, Toth K, Durst RA et al (2001) Electrochemical biosensors: recommended definitions and classification. Biosens Bioelectron 16:121–131 Thompson G, Swain J, Kay M et al (2001) The treatment of pulp and paper mill effluent: a review. Bioresour Technol 77:275–286 Tront JM, Fortner JD, Plötze M et al (2008) Microbial fuel cell biosensor for in situ assessment of microbial activity. Biosens Bioelectron 24:586–590 Wang J, Zhang Y, Wang Y et al (2010) An innovative reactor-type biosensor for BOD rapid measurement. Biosens Bioelectron 25:1705–1709 Wang B, Yan S, Shi Y (2015) Direct electrochemical analysis of glucose oxidase on a graphene aerogel/gold nanoparticle hybrid for glucose biosensing. J Solid State Electrochem 19:307–314 Yang Z, Suzuki H, Sasaki S et al (1997) Comparison of the dynamic transient- and steady-state measuring methods in a batch type BOD sensing system. Sensors Actuators 45:217–222 Yoshida N, Yano K, Morita T et al (2000) A mediator-type biosensor as a new approach to biochemical oxygen demand estimation. Analyst 125:2280–2284 Zaitseva AS, Arlyapov VA, Yudina NY et al (2017) Use of one- and two-mediator systems for developing a BOD biosensor based on the yeast Debaryomyces hansenii. Enzyme Microb Technol 98:43–51 Zanon E, Mancini A, Pavanello IG et al (2013) A review on thermal exfoliation of graphene oxide. J Mater Res 95:5343–5350 Zhang Y, Angelidaki I (2011) Submersible microbial fuel cell sensor for monitoring microbial activity and BOD in groundwater: focusing on impact of anodic biofilm on sensor applicability. Biotechnol Bioeng 108:2339–2347 Zhao L, He L, Chen S et al (2017) Microbial BOD sensors based on Zr (IV)-loaded collagen fiber. Enzyme Microb Technol 98:52–57

Cell-Based Biosensor for Rapid Screening of Pathogens and Toxins

37

Celina To, Pratik Banerjee, and Arun K. Bhunia

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cell-Based Biosensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Microbes as Biosensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Mammalian Cells as Biosensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B Cell . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Vero Cell . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Mast Cell . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Neuron . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

930 931 931 938 939 940 940 941 942 942

C. To Molecular Food Microbiology Laboratory, Department of Food Science, Purdue University, West Lafayette, IN, USA Hygiena, Camarillo, CA, USA P. Banerjee Division of Epidemiology, Biostatistics, and Environmental Health, The University of Memphis, Memphis, TN, USA e-mail: [email protected] A. K. Bhunia (*) Molecular Food Microbiology Laboratory, Department of Food Science, Purdue University, West Lafayette, IN, USA Department of Comparative Pathobiology, Purdue University, West Lafayette, IN, USA Purdue Institute of Inflammation, Immunology and Infectious Disease, Purdue University, West Lafayette, IN, USA e-mail: [email protected] © Springer Nature Switzerland AG 2022 G. Thouand (ed.), Handbook of Cell Biosensors, https://doi.org/10.1007/978-3-030-23217-7_102

929

930

C. To et al.

Abstract

Exposure to pathogens, toxins, and pollutants is a major public health concern. Often air, water, and food are the major sources. Efforts to develop and deploy highly sensitive detection and diagnostic platforms are of great importance to protect the health of consumers, military personnel, and civilians. Biosensors employing a proper transducer have the potential to digitally amplify signals for prompt analysis of small amounts of analyte and the consequent institution of remedial actions. Cell-based biosensor (CBB) using mammalian and microbial cells is referred to as a functional biosensor since it interrogates the interaction between the living cells and an analyte to provide physiologically relevant information regarding the functionality of an analyte. Thus, CBB is advantageous over immunosensor or nucleic acid-based biosensors, which may not assess functionality. However, the major drawback of CBB is the maintenance of viability of the cells during deployment in a rugged environment or in the point-of-care use. Keywords

Cell-based sensor · Bacteria · Bacteriophage · Mammalian cell · Detection · Pathogen · Toxin · Diagnosis

Introduction Fast and accurate detection and diagnostic methods are critical for preventing infectious disease outbreaks or diagnosing diseases for physicians to initiate a therapeutic intervention. Traditional culture-based methods are highly accurate and reliable, but prolonged assay time may make them inconvenient. Biosensor methods are sensitive and provide an amplified digital signal that relies on pathogen or toxin interaction with an appropriate ligand or receptors immobilized on a sensor (transducer) platform (Bhunia 2014). Specificity and the sensitivity of an assay also depend on the ligand used in the form of antibody, aptamer, DNA/RNA probe, or receptor molecules and the physiological state of the analyte to be interrogated. Based on the signal output, biosensors are divided into electrochemical, optical, mass-based, or thermometric (Bhunia 2014; Velusamy et al. 2010). Of these, optical sensors have been widely used because of their relative ease in operation and improved data interpretation capabilities (Cho et al. 2014; Singh and Bhunia 2018). A majority of these traditional biosensors are designed to interpret the presence or absence of targeted pathogens, toxins, or genes encoding virulence factors, but do not necessarily reveal if these analytes are active, i.e., viable or harmful at the time of analysis. Especially for food safety applications, food receives various antimicrobial and processing treatments which may make the target pathogens/toxins injured, inactive, or denatured. Therefore, it is highly desirable that the detection method can unequivocally differentiate viable cells or active toxins from a

37

Cell-Based Biosensor for Rapid Screening of Pathogens and Toxins

931

dead or inactive form of the same analyte and to demonstrate that the contaminant does not pose any threat to consumer health.

Cell-Based Biosensor Whole cell-based biosensor (CBB) offers a unique option of measuring the functionality and toxicity of an analyte (Banerjee and Bhunia 2009, 2010; Banerjee et al. 2013; Ye et al. 2019). This concept relies on the interaction between higher eukaryotic or microbial cells and the target analyte to generate a response or signal that can be measured by a detector (Fig. 1). Upon exposure to the analyte, living cells acting as transducers can activate signaling pathways that can be measured by different modalities including, but not limited to, colorimetric, fluorometric, electrochemical, or amperometric methods. In contrast, dead pathogenic microbes or denatured or inactive toxins or pollutants may not interact with the whole cells; therefore, they may not stimulate a signaling event, thus providing a negative response. For this reason, a major flaw behind immunosensor or DNA-based sensors is its inability to discriminate live from dead microbes or active from inactive or denatured toxins, thus providing misleading inaccurate information regarding public health consequences of such contaminant. In addition, immunosensors such as enzyme-linked immunosorbent assays (ELISA) are costly and difficult to adopt in a resource-limited setting (Fu et al. 2011).

Microbes as Biosensor Microbial cell-based biosensors employ bacteria, viruses, cyanobacteria, yeast, and algae as bioreporters, and often these cells are engineered to express bioluminescence or fluorescence (e.g., luxCDABE operon from Vibrio fischeri or gfp from jellyfish) with an inducible promoter (Table 1). Microbial biosensors have found greater utility in monitoring environmental contaminants and pollutants, for example, benzene, toluene, ethylbenzene, and xylenes (BTEX) (Xu et al. 2003, 2014), and heavy metals such as arsenic (As), chromium (Cr), cadmium (Cd), lead (Pb), copper (Cu), and others (Teo and Wong 2014). Furthermore, bacteria have been used for monitoring genotoxicity or mutagenic agents that cause DNA damage, avoiding the use of animals as a test subject. The mutagenic agent may include mycotoxin, antibiotics, acridine, ethidium bromide, ionizing radiation, UV, aromatic hydrocarbon, dimethylnitrosamine, and heavy metals (Ni, Cd, As) (Kaina and Fritz 2006). The most familiar bacterial genotoxicity testing assay is the “Ames test,” where it quantifies reversion rate from defined mutation to the wild type after exposure to the mutagenic agent (Ames et al. 1973). Salmonella enterica serovar Typhimurium has been commonly used for such testing. In recent years, genetically engineered bacteria carrying a promoter-reporter fusion gene that can quantify the gene repair event in real time after being exposed to the mutagenic agent are used (Biran et al. 2010). The most common inducible promoters

Fig. 1 Schematics of cell-based biosensor comprising of microbial and mammalian cells

932 C. To et al.

Soil

Soil and ground water NA

Water

Water

Water

Water

Water

BTEX

BTEX

Arsenic (As)

Cadmium (cd)

Copper (Cu)

Cu

Cu, Lead (Pb), Cd

BTEX

Sample type Simulated aquatic oil spill Soil

Target analyte Benzene, toluene, ethylbenzene xylene (BTEX) Benzene, toluene, xylene

Fluorescence

Amperometry

30 min