Liposomes, Part F [1 ed.] 0123749697, 9780123749697

Citation preview

VOLUME FOUR HUNDRED AND SIXTY-FOUR

METHODS

IN

ENZYMOLOGY Liposomes, Part F

METHODS IN ENZYMOLOGY Editors-in-Chief

JOHN N. ABELSON AND MELVIN I. SIMON Division of Biology California Institute of Technology Pasadena, California, USA Founding Editors

SIDNEY P. COLOWICK AND NATHAN O. KAPLAN

VOLUME FOUR HUNDRED AND SIXTY-FOUR

METHODS

IN

ENZYMOLOGY Liposomes, Part F EDITED BY

¨ ZGU ¨ NES¸ NEJAT DU Department of Microbiology University of the Pacific Arthur A. Dugoni School of Dentistry San Francisco, California, USA

AMSTERDAM • BOSTON • HEIDELBERG • LONDON NEW YORK • OXFORD • PARIS • SAN DIEGO SAN FRANCISCO • SINGAPORE • SYDNEY • TOKYO Academic Press is an imprint of Elsevier

Academic Press is an imprint of Elsevier 525 B Street, Suite 1900, San Diego, CA 92101-4495, USA 30 Corporate Drive, Suite 400, Burlington, MA 01803, USA 32 Jamestown Road, London NW1 7BY, UK First edition 2009 Copyright # 2009, Elsevier Inc. All Rights Reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means electronic, mechanical, photocopying, recording or otherwise without the prior written permission of the publisher Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford, UK: phone (+44) (0) 1865 843830; fax (+44) (0) 1865 853333; email: permissions@ elsevier.com. Alternatively you can submit your request online by visiting the Elsevier web site at http://elsevier.com/locate/permissions, and selecting Obtaining permission to use Elsevier material Notice No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made For information on all Academic Press publications visit our website at elsevierdirect.com ISBN: 978-0-12-374969-7 ISSN: 0076-6879 Printed and bound in United States of America 09 10 11 12 10 9 8 7 6 5 4 3 2 1

CONTENTS

Contributors Preface Volumes in Series

Section I. Bioactive Liposomes 1. Tubular Liposomes with Variable Permeability for Reconstitution of FtsZ Rings

xi xvii xix

1 3

Masaki Osawa and Harold P. Erickson 1. Introduction 2. Reagents 3. Bacterial Expression of Membrane Targeting FtsZ 4. Purification of FtsZ-mts and FtsZ-YFP-mts 5. Renatured Preparation of FtsZ-YFP-mts 6. Tubular Multilamellar Liposome Preparation 7. Permeability of the Multilamellar Liposomes 8. Z-ring Formation in Liposomes 9. A Crude Flow Chamber to Exchange Buffer Outside Liposomes 10. Factors Affecting Z-ring Formation in Liposomes 11. Utility of the Liposomes Beyond FtsZ References

2. Detection and Analysis of Protein Synthesis and RNA Replication in Giant Liposomes

4 4 5 5 7 7 9 11 11 14 16 16

19

Takeshi Sunami, Hiroshi Kita, Kazufumi Hosoda, Tomoaki Matsuura, Hiroaki Suzuki, and Tetsuya Yomo 1. Introduction 2. Methods 3. Analysis of the FACS Data 4. Conclusions Acknowledgments References

20 21 27 28 29 29

v

vi

Contents

3. Construction of Cell-Sized Liposomes Encapsulating Actin and Actin-Cross-Linking Proteins

31

Kingo Takiguchi, Ayako Yamada, Makiko Negishi, Makoto Honda, Yohko Tanaka-Takiguchi, and Kenichi Yoshikawa 1. Introduction 2. Experimental Section 3. Morphogenesis of Giant Liposomes Encapsulating Actin and Its Cross-linking Proteins 4. Concluding Remarks Acknowledgments References

4. Reconstitution of Membrane Budding with Unilamellar Vesicles

32 37 42 49 50 50

55

Anna V. Shnyrova and Joshua Zimmerberg 1. Introduction 2. M Protein Purification 3. Evaluation of the Membrane Activity of M Protein Through its Interaction with Intermediate-Sized Unilamellar Liposomes 4. Reconstitution of M-Protein-Driven Membrane Budding on GUVs 5. Concluding Remarks References

Section II. Liposomes and Nanotechnology 5. Detection of Antimycolic Acid Antibodies by Liposomal Biosensors

56 57 58 66 73 73

77 79

Y. Lemmer, S. T. Thanyani, P. J. Vrey, C. H. S. Driver, L. Venter, S. van Wyngaardt, A. M. C. ten Bokum, K. I. Ozoemena, L. A. Pilcher, D. G. Fernig, A. C. Stoltz, H. S. Swai, and J. A. Verschoor 1. Introduction 2. Experimental 3. Conclusion Acknowledgments References

6. Solid Lipid Nanoparticle Formulations: Pharmacokinetic and Biopharmaceutical Aspects in Drug Delivery

80 81 102 102 102

105

Eliana B. Souto and Slavomira Doktorovova´ 1. Introduction 2. Production of SLN

106 107

Contents

3. Pharmacokinetics and Pharmacodynamics 4. Modified Release Profile 5. Biopharmaceutical Aspects of Administration Routes 6. Clinical Pharmacology 7. Concluding Remarks References

vii

107 112 114 117 121 122

7. Preparation of Complexes of Liposomes with Gold Nanoparticles 131 Chie Kojima, Yusuke Hirano, and Kenji Kono Introduction Preparation of Complexes of EYPC Liposomes with Au NPs Time-Dependent SPR of the Complexes TEM Analysis of the Complexes DLS Analysis of the Complexes Calcein Release from the Complexes Estimation of Numbers of the Au NP and the Liposome in the Complexes 8. Optimization of Lipid Components of the Complexes 9. Concluding Remarks Acknowledgment References 1. 2. 3. 4. 5. 6. 7.

8. Bio-Nanocapsule–Liposome Conjugates for In Vivo Pinpoint Drug and Gene Delivery

132 134 134 136 137 137 139 140 142 143 144

147

Takeshi Kasuya, Joohee Jung, Rie Kinoshita, Yasumasa Goh, Takashi Matsuzaki, Masumi Iijima, Nobuo Yoshimoto, Katsuyuki Tanizawa, and Shun’ichi Kuroda 1. Introduction 2. First-Generation Bio-Nanocapsules 3. Second-Generation BNCs 4. Retargeting of BNC–LP Conjugates 5. Overexpression of BNCs in S. cerevisiae 6. Conjugation of BNCs with LPs 7. Preparation of Antibody-Displaying BNC–LP Conjugates 8. Preparation of Biotin-Displaying BNC–LP Conjugates 9. Concluding Remarks Acknowledgments References

148 149 150 152 153 155 161 163 163 164 164

viii

Contents

9. Nanoliposomal Dry Powder Formulations

167

Gaurang Patel, Mahavir Chougule, Mandip Singh, and Ambikanandan Misra 1. Introduction 2. Preparation of Nanoliposomal DPFs 3. Physicochemical Characterization of NLDPFs 4. Concluding Remarks References

10. Lanthanide-Loaded Paramagnetic Liposomes as Switchable Magnetically Oriented Nanovesicles

168 169 174 187 188

193

Silvio Aime, Daniela Delli Castelli, and Enzo Terreno Introduction Paramagnetic Ln(III)-Based Shift Reagents Preparation of Osmotically Shrunken Liposomes NMR Characterization of Magnetically Oriented Nonspherical Liposomes 5. Sample Experiments 6. Concluding Remarks Acknowledgments References 1. 2. 3. 4.

11. Reconstitution of Membrane Proteins in Phospholipid Bilayer Nanodiscs

194 195 197 198 200 208 208 208

211

T. K. Ritchie, Y. V. Grinkova, T. H. Bayburt, I. G. Denisov, J. K. Zolnerciks, W. M. Atkins, and S. G. Sligar 1. Introduction 2. Overview of Nanodisc Technology 3. Reconstitution Considerations 4. Optimizing the Reconstitution for P-glycoprotein Acknowledgments References

12. DNA-Controlled Assembly of Liposomes in Diagnostics

212 212 218 223 228 228

233

Ulla Jakobsen and Stefan Vogel 1. Introduction 2. Probe Design 3. General Description of Materials and Techniques 4. Concluding Remarks Acknowledgment References

234 235 244 247 248 248

Contents

13. Soft Hybrid Nanostructures Composed of Phospholipid Liposomes Decorated with Oligonucleotides

ix

249

Martina Banchelli, Francesca Baldelli Bombelli, Debora Berti, and Piero Baglioni 1. Introduction 2. Materials 3. Liposome Preparation and Determination of Lipid Content 4. Incorporation of Oligonucleotides 5. Characterization of the Soft Hybrid Nanostructure 6. Applications of Oligo-Decorated Liposomes 7. Challenges and Perspectives Acknowledgments References

250 251 252 253 256 262 275 276 276

14. Synthesis, Characterization, and Optical Response of Gold Nanoshells Used to Trigger Release from Liposomes

279

Guohui Wu, Alexander Mikhailovsky, Htet A. Khant, and Joseph A. Zasadzinski 1. Introduction 2. Synthesis of HGNs 3. Optimization of HGN Dimensions for Maximum Absorption in the NIR 4. HGN Response to Femtosecond NIR Laser Pulses 5. Coupling HGN to Liposomes 6. Liposome Disruption and CF Release Due to Pulsed Laser Irradiation 7. Mechanism of Triggered Liposome Release 8. Effect of Proximity of HGNs to Liposomes 9. Conclusions Acknowledgments References

15. Complex Nanotube-Liposome Networks

280 283 286 290 294 299 300 303 304 304 304

309

Aldo Jesorka and Owe Orwar 1. Introduction 2. Network Fabrication Protocols 3. Complexity and Topology 4. Internal and Membrane Functionalization 5. Transport Phenomena and Controlled Mixing Procedures 6. Enzymatic Reactions in NVN 7. Concluding Remarks Acknowledgments References

309 310 314 315 318 320 323 323 324

x

Contents

16. Bionanotubules Formed from Liposomes

327

Josemar A. Castillo and Mark A. Hayes 1. Introduction 2. Bionanotubule Formation by Applying Electric Fields to Surface-Attached Liposomes 3. Bionanotubular Formation from Liposomes in Solution Using Electric Fields 4. Other Methods of Bionanotubular Formation from Liposomes 5. Concluding Remarks References

17. Engineering Cationic Liposome: siRNA Complexes for In Vitro and In Vivo Delivery

328 329 334 337 339 339

343

Jennifer E. Podesta and Kostas Kostarelos 1. Introduction 2. Cationic Liposome Systems for siRNA Delivery 3. Experimental Methods 4. Troubleshooting 5. Concluding Remarks References Author Index Subject Index

344 345 347 352 353 354 355 363

CONTRIBUTORS

Silvio Aime Department of Chemistry IFM and Molecular Imaging Center, University of Torino, Torino, Italy W. M. Atkins Department of Medicinal Chemistry, University of Washington, Seattle, Washington, USA Piero Baglioni Department of Chemistry and CSGI, University of Florence, Sesto Fiorentino, Florence, Italy Martina Banchelli Department of Chemistry and CSGI, University of Florence, Sesto Fiorentino, Florence, Italy T. H. Bayburt Department of Biochemistry, University of Illinois at Urbana-Champaign, Urbana, Illinois, USA Debora Berti Department of Chemistry and CSGI, University of Florence, Sesto Fiorentino, Florence, Italy Francesca Baldelli Bombelli Department of Chemistry and CSGI, University of Florence, Sesto Fiorentino, Florence, Italy Daniela Delli Castelli Department of Chemistry IFM and Molecular Imaging Center, University of Torino, Torino, Italy Josemar A. Castillo Department of Chemistry and Biochemistry, Arizona State University, Tempe, Arizona, USA Mahavir Chougule Pharmaceutics Department, College of Pharmacy and Pharmaceutical Science, Florida A&M University, Tallahassee, Florida, USA

xi

xii

Contributors

I. G. Denisov Department of Biochemistry, University of Illinois at Urbana-Champaign, Urbana, Illinois, USA Slavomira Doktorovova´ Department of Pharmaceutical Technology, Faculty of Health Sciences, Fernando Pessoa University, Porto, Portugal C. H. S. Driver Department of Chemistry, University of Pretoria, Pretoria, South Africa Harold P. Erickson Department of Cell Biology, Duke University Medical Center, Durham, North Carolina, USA D. G. Fernig School of Biological Sciences, University of Liverpool, United Kingdom Yasumasa Goh Beacle Inc., ORIC, Haga, Okayama, Japan Y. V. Grinkova Department of Biochemistry, University of Illinois at Urbana-Champaign, Urbana, Illinois, USA Mark A. Hayes Department of Chemistry and Biochemistry, Arizona State University, Tempe, Arizona, USA Yusuke Hirano Graduate School of Engineering, Osaka Prefecture University, Osaka, Japan Makoto Honda Division of Biological Science, Graduate School of Science, Nagoya University, Furo-cho, Chikusa-ku, Nagoya, Japan Current address: Stem Cell and Drug Discovery Institute, Shimogyo-ku, Kyoto, Japan Kazufumi Hosoda Department of Bioinformatics Engineering, Graduate School of Information Science and Technology, Osaka University, Suita, Osaka, Japan Masumi Iijima Institute of Scientific and Industrial Research, Osaka University, Ibaraki, Osaka, Japan, and Graduate School of Bioagricultural Sciences, Nagoya University, Chikusa, Nagoya, Japan Ulla Jakobsen Nucleic Acid Center, University of Southern Denmark, Odense, Denmark

Contributors

xiii

Aldo Jesorka Department of Chemical and Biological Engineering, Chalmers University of Technology, Go¨teborg, Sweden Joohee Jung Institute of Scientific and Industrial Research, Osaka University, Ibaraki, Osaka, Japan Present address: Institute for Innovative Cancer Research, ASAN Medical Center, Pungnap-2, Songpa, Seoul, Korea Takeshi Kasuya Institute of Scientific and Industrial Research, Osaka University, Ibaraki, Osaka, Japan Htet A. Khant Department of Chemical Engineering, University of California, Santa Barbara, California, USA Rie Kinoshita Beacle Inc., ORIC, Haga, Okayama, Japan Hiroshi Kita Exploratory Research for Advanced Technology (ERATO), Japan Science and Technology Agency ( JST), Chiyoda-ku, Tokyo, Japan Chie Kojima Nanoscience and Nanotechnology Research Center, Research Institutes for the Twenty First Century, Osaka Prefecture University, Osaka, Japan Kenji Kono Graduate School of Engineering, Osaka Prefecture University, Osaka, Japan Kostas Kostarelos Nanomedicine Laboratory, Centre for Drug Delivery Research, The School of Pharmacy, University of London, London, United Kingdom Shun’ichi Kuroda Institute of Scientific and Industrial Research, Osaka University, Ibaraki, Osaka, Japan, and Beacle Inc., ORIC, Haga, Okayama, Japan; Graduate School of Bioagricultural Sciences, Nagoya University, Chikusa, Nagoya, Japan Y. Lemmer Department of Biochemistry, University of Pretoria, Pretoria, South Africa, and Materials Science and Manufacturing, CSIR, Pretoria, South Africa Tomoaki Matsuura Department of Bioinformatics Engineering, Graduate School of Information Science and Technology, Osaka University, Suita, Osaka, Japan

xiv

Contributors

Takashi Matsuzaki Institute of Scientific and Industrial Research, Osaka University, Ibaraki, Osaka, Japan Alexander Mikhailovsky Department of Chemistry, University of California, Santa Barbara, California, USA Ambikanandan Misra TIFAC-CORE in NDDS, Pharmacy Department, Faculty of Technology and Engineering, The Maharaja Sayajirao University of Baroda, Kalabhavan, Vadodara, Gujarat, India Makiko Negishi Department of Physics, Graduate School of Science, Kyoto University, Sakyo-ku, Kyoto, Japan Owe Orwar Department of Chemical and Biological Engineering, Chalmers University of Technology, Go¨teborg, Sweden Masaki Osawa Department of Cell Biology, Duke University Medical Center, Durham, North Carolina, USA K. I. Ozoemena Department of Chemistry, University of Pretoria, Pretoria, South Africa Gaurang Patel TIFAC-CORE in NDDS, Pharmacy Department, Faculty of Technology and Engineering, The Maharaja Sayajirao University of Baroda, Kalabhavan, Vadodara, Gujarat, India L. A. Pilcher Department of Chemistry, University of Pretoria, Pretoria, South Africa Jennifer E. Podesta Nanomedicine Laboratory, Centre for Drug Delivery Research, The School of Pharmacy, University of London, London, United Kingdom T. K. Ritchie Department of Medicinal Chemistry, University of Washington, Seattle, Washington, USA Anna V. Shnyrova Laboratory of Cellular and Molecular Biology, Program in Physical Biology, Eunice Kennedy Shriver National Institute of Child Health and Human Development, Bethesda, Maryland, USA Mandip Singh Pharmaceutics Department, College of Pharmacy and Pharmaceutical Science, Florida A&M University, Tallahassee, Florida, USA

Contributors

xv

S. G. Sligar Department of Biochemistry, University of Illinois at Urbana-Champaign, Urbana, Illinois, USA Eliana B. Souto Department of Pharmaceutical Technology, Faculty of Health Sciences, Fernando Pessoa University, Porto, Portugal, and Institute of Biotechnology and Bioengineering, Centre of Genetics and Biotechnology, University of Tra´s-os-Montes and Alto Douro (CGB-UTAD/IBB), Vila Real, Portugal A. C. Stoltz Department of Biochemistry, and Department of Infectious Diseases, University of Pretoria, Pretoria, South Africa Takeshi Sunami Exploratory Research for Advanced Technology (ERATO), Japan Science and Technology Agency ( JST), Chiyoda-ku, Tokyo, Japan Hiroaki Suzuki Department of Bioinformatics Engineering, Graduate School of Information Science and Technology, Osaka University, Suita, Osaka, Japan H. S. Swai Materials Science and Manufacturing, CSIR, Pretoria, South Africa Kingo Takiguchi Division of Biological Science, Graduate School of Science, Nagoya University, Furo-cho, Chikusa-ku, Nagoya, Japan Yohko Tanaka-Takiguchi Division of Biological Science, Graduate School of Science, Nagoya University, Furo-cho, Chikusa-ku, Nagoya, Japan Katsuyuki Tanizawa Institute of Scientific and Industrial Research, Osaka University, Ibaraki, Osaka, Japan A. M. C. ten Bokum Department of Biochemistry, University of Pretoria, Pretoria, South Africa Enzo Terreno Department of Chemistry IFM and Molecular Imaging Center, University of Torino, Torino, Italy S. T. Thanyani Department of Biochemistry, University of Pretoria, Pretoria, South Africa S. van Wyngaardt Department of Biochemistry, University of Pretoria, Pretoria, South Africa

xvi

Contributors

L. Venter Department of Biochemistry, University of Pretoria, Pretoria, South Africa J. A. Verschoor Department of Biochemistry, University of Pretoria, Pretoria, South Africa Stefan Vogel Nucleic Acid Center, University of Southern Denmark, Odense, Denmark P. J. Vrey Department of Biochemistry, University of Pretoria, Pretoria, South Africa Guohui Wu Department of Chemical Engineering, University of California, Santa Barbara, California, USA Ayako Yamada Department of Physics, Graduate School of Science, Kyoto University, Sakyo-ku, Kyoto, Japan Current address: Department of Chemistry, Ecole Normale Superieure, Paris, France Tetsuya Yomo Department of Bioinformatics Engineering, Graduate School of Information Science and Technology, and Graduate School of Frontier Biosciences, Osaka University, Suita, Osaka, Japan, and Exploratory Research for Advanced Technology (ERATO), Japan Science and Technology Agency ( JST), Chiyoda-ku, Tokyo, Japan Kenichi Yoshikawa Department of Physics, Graduate School of Science, Kyoto University, Sakyo-ku, Kyoto, Japan Nobuo Yoshimoto Institute of Scientific and Industrial Research, Osaka University, Ibaraki, Osaka, Japan, and Graduate School of Bioagricultural Sciences, Nagoya University, Chikusa, Nagoya, Japan Joseph A. Zasadzinski Department of Chemical Engineering, University of California, Santa Barbara, California, USA Joshua Zimmerberg Laboratory of Cellular and Molecular Biology, Program in Physical Biology, Eunice Kennedy Shriver National Institute of Child Health and Human Development, Bethesda, Maryland, USA J. K. Zolnerciks Department of Medicinal Chemistry, University of Washington, Seattle, Washington, USA

PREFACE

Previous Methods in Enzymology volumes on ‘‘Liposomes’’ have described methods of liposome preparation and the physicochemical characterization of liposomes (Volume 367), and the use of liposomes in biochemistry, molecular cell biology (Volume 372), immunology, diagnostics, gene delivery, and gene therapy (Volume 373). Methods involved in the production and application of antibody- or ligand-targeted liposomes, environmentsensitive liposomes, and liposomal oligonucleotides were provided in Volume 387, as were methods for studying the in vivo fate of liposomes. Finally, Volume 391 presented methods in liposomal anticancer, antibacterial, antifungal, and antiviral agents, miscellaneous liposomal therapies and electron microscopy of liposomes. The latter volume also included a short introductory chapter on ‘‘The Origin of Liposomes: Alec Bangham at Babraham.’’ This new volume includes sections focusing on bioactive liposomes and the interface of liposomes and nanotechnology. I hope that these chapters will be helpful to graduate students, postdoctoral fellows, research associates, and established scientists initiating projects on liposomes, or shifting the focus of their research. Although the chapters are not written in ‘‘protocol’’ format, they describe the experimental methods in sufficient detail that can be adopted readily for the reader’s project. In addition, the chapters provide the perspective of the authors on the field, as well as examples of results obtained with the described methods. I would like to thank all the colleagues who graciously contributed to this volume with relatively short notice, Associate Editor Tara Hoey, Editorial Services Manager Delsy Retchagar, and Production Manager Radhakrishnan Lakshmanan of Elsevier for their help in preparing this volume, and Shirley Light, formerly of Academic Press, for her initiation of the ‘‘Liposomes’’ volumes about a decade ago. I would also like to express my gratitude to my supportive and loving family. I dedicate this volume to my wife, Diana, and my curious, creative, playful, and loving children, Avery and Maxine. NEJAT DU¨ZGU¨NES¸

xvii

This page intentionally left blank

METHODS IN ENZYMOLOGY

VOLUME I. Preparation and Assay of Enzymes Edited by SIDNEY P. COLOWICK AND NATHAN O. KAPLAN VOLUME II. Preparation and Assay of Enzymes Edited by SIDNEY P. COLOWICK AND NATHAN O. KAPLAN VOLUME III. Preparation and Assay of Substrates Edited by SIDNEY P. COLOWICK AND NATHAN O. KAPLAN VOLUME IV. Special Techniques for the Enzymologist Edited by SIDNEY P. COLOWICK AND NATHAN O. KAPLAN VOLUME V. Preparation and Assay of Enzymes Edited by SIDNEY P. COLOWICK AND NATHAN O. KAPLAN VOLUME VI. Preparation and Assay of Enzymes (Continued) Preparation and Assay of Substrates Special Techniques Edited by SIDNEY P. COLOWICK AND NATHAN O. KAPLAN VOLUME VII. Cumulative Subject Index Edited by SIDNEY P. COLOWICK AND NATHAN O. KAPLAN VOLUME VIII. Complex Carbohydrates Edited by ELIZABETH F. NEUFELD AND VICTOR GINSBURG VOLUME IX. Carbohydrate Metabolism Edited by WILLIS A. WOOD VOLUME X. Oxidation and Phosphorylation Edited by RONALD W. ESTABROOK AND MAYNARD E. PULLMAN VOLUME XI. Enzyme Structure Edited by C. H. W. HIRS VOLUME XII. Nucleic Acids (Parts A and B) Edited by LAWRENCE GROSSMAN AND KIVIE MOLDAVE VOLUME XIII. Citric Acid Cycle Edited by J. M. LOWENSTEIN VOLUME XIV. Lipids Edited by J. M. LOWENSTEIN VOLUME XV. Steroids and Terpenoids Edited by RAYMOND B. CLAYTON

xix

xx

Methods in Enzymology

VOLUME XVI. Fast Reactions Edited by KENNETH KUSTIN VOLUME XVII. Metabolism of Amino Acids and Amines (Parts A and B) Edited by HERBERT TABOR AND CELIA WHITE TABOR VOLUME XVIII. Vitamins and Coenzymes (Parts A, B, and C) Edited by DONALD B. MCCORMICK AND LEMUEL D. WRIGHT VOLUME XIX. Proteolytic Enzymes Edited by GERTRUDE E. PERLMANN AND LASZLO LORAND VOLUME XX. Nucleic Acids and Protein Synthesis (Part C) Edited by KIVIE MOLDAVE AND LAWRENCE GROSSMAN VOLUME XXI. Nucleic Acids (Part D) Edited by LAWRENCE GROSSMAN AND KIVIE MOLDAVE VOLUME XXII. Enzyme Purification and Related Techniques Edited by WILLIAM B. JAKOBY VOLUME XXIII. Photosynthesis (Part A) Edited by ANTHONY SAN PIETRO VOLUME XXIV. Photosynthesis and Nitrogen Fixation (Part B) Edited by ANTHONY SAN PIETRO VOLUME XXV. Enzyme Structure (Part B) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME XXVI. Enzyme Structure (Part C) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME XXVII. Enzyme Structure (Part D) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME XXVIII. Complex Carbohydrates (Part B) Edited by VICTOR GINSBURG VOLUME XXIX. Nucleic Acids and Protein Synthesis (Part E) Edited by LAWRENCE GROSSMAN AND KIVIE MOLDAVE VOLUME XXX. Nucleic Acids and Protein Synthesis (Part F) Edited by KIVIE MOLDAVE AND LAWRENCE GROSSMAN VOLUME XXXI. Biomembranes (Part A) Edited by SIDNEY FLEISCHER AND LESTER PACKER VOLUME XXXII. Biomembranes (Part B) Edited by SIDNEY FLEISCHER AND LESTER PACKER VOLUME XXXIII. Cumulative Subject Index Volumes I-XXX Edited by MARTHA G. DENNIS AND EDWARD A. DENNIS VOLUME XXXIV. Affinity Techniques (Enzyme Purification: Part B) Edited by WILLIAM B. JAKOBY AND MEIR WILCHEK

Methods in Enzymology

VOLUME XXXV. Lipids (Part B) Edited by JOHN M. LOWENSTEIN VOLUME XXXVI. Hormone Action (Part A: Steroid Hormones) Edited by BERT W. O’MALLEY AND JOEL G. HARDMAN VOLUME XXXVII. Hormone Action (Part B: Peptide Hormones) Edited by BERT W. O’MALLEY AND JOEL G. HARDMAN VOLUME XXXVIII. Hormone Action (Part C: Cyclic Nucleotides) Edited by JOEL G. HARDMAN AND BERT W. O’MALLEY VOLUME XXXIX. Hormone Action (Part D: Isolated Cells, Tissues, and Organ Systems) Edited by JOEL G. HARDMAN AND BERT W. O’MALLEY VOLUME XL. Hormone Action (Part E: Nuclear Structure and Function) Edited by BERT W. O’MALLEY AND JOEL G. HARDMAN VOLUME XLI. Carbohydrate Metabolism (Part B) Edited by W. A. WOOD VOLUME XLII. Carbohydrate Metabolism (Part C) Edited by W. A. WOOD VOLUME XLIII. Antibiotics Edited by JOHN H. HASH VOLUME XLIV. Immobilized Enzymes Edited by KLAUS MOSBACH VOLUME XLV. Proteolytic Enzymes (Part B) Edited by LASZLO LORAND VOLUME XLVI. Affinity Labeling Edited by WILLIAM B. JAKOBY AND MEIR WILCHEK VOLUME XLVII. Enzyme Structure (Part E) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME XLVIII. Enzyme Structure (Part F) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME XLIX. Enzyme Structure (Part G) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME L. Complex Carbohydrates (Part C) Edited by VICTOR GINSBURG VOLUME LI. Purine and Pyrimidine Nucleotide Metabolism Edited by PATRICIA A. HOFFEE AND MARY ELLEN JONES VOLUME LII. Biomembranes (Part C: Biological Oxidations) Edited by SIDNEY FLEISCHER AND LESTER PACKER

xxi

xxii

Methods in Enzymology

VOLUME LIII. Biomembranes (Part D: Biological Oxidations) Edited by SIDNEY FLEISCHER AND LESTER PACKER VOLUME LIV. Biomembranes (Part E: Biological Oxidations) Edited by SIDNEY FLEISCHER AND LESTER PACKER VOLUME LV. Biomembranes (Part F: Bioenergetics) Edited by SIDNEY FLEISCHER AND LESTER PACKER VOLUME LVI. Biomembranes (Part G: Bioenergetics) Edited by SIDNEY FLEISCHER AND LESTER PACKER VOLUME LVII. Bioluminescence and Chemiluminescence Edited by MARLENE A. DELUCA VOLUME LVIII. Cell Culture Edited by WILLIAM B. JAKOBY AND IRA PASTAN VOLUME LIX. Nucleic Acids and Protein Synthesis (Part G) Edited by KIVIE MOLDAVE AND LAWRENCE GROSSMAN VOLUME LX. Nucleic Acids and Protein Synthesis (Part H) Edited by KIVIE MOLDAVE AND LAWRENCE GROSSMAN VOLUME 61. Enzyme Structure (Part H) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME 62. Vitamins and Coenzymes (Part D) Edited by DONALD B. MCCORMICK AND LEMUEL D. WRIGHT VOLUME 63. Enzyme Kinetics and Mechanism (Part A: Initial Rate and Inhibitor Methods) Edited by DANIEL L. PURICH VOLUME 64. Enzyme Kinetics and Mechanism (Part B: Isotopic Probes and Complex Enzyme Systems) Edited by DANIEL L. PURICH VOLUME 65. Nucleic Acids (Part I) Edited by LAWRENCE GROSSMAN AND KIVIE MOLDAVE VOLUME 66. Vitamins and Coenzymes (Part E) Edited by DONALD B. MCCORMICK AND LEMUEL D. WRIGHT VOLUME 67. Vitamins and Coenzymes (Part F) Edited by DONALD B. MCCORMICK AND LEMUEL D. WRIGHT VOLUME 68. Recombinant DNA Edited by RAY WU VOLUME 69. Photosynthesis and Nitrogen Fixation (Part C) Edited by ANTHONY SAN PIETRO VOLUME 70. Immunochemical Techniques (Part A) Edited by HELEN VAN VUNAKIS AND JOHN J. LANGONE

Methods in Enzymology

xxiii

VOLUME 71. Lipids (Part C) Edited by JOHN M. LOWENSTEIN VOLUME 72. Lipids (Part D) Edited by JOHN M. LOWENSTEIN VOLUME 73. Immunochemical Techniques (Part B) Edited by JOHN J. LANGONE AND HELEN VAN VUNAKIS VOLUME 74. Immunochemical Techniques (Part C) Edited by JOHN J. LANGONE AND HELEN VAN VUNAKIS VOLUME 75. Cumulative Subject Index Volumes XXXI, XXXII, XXXIV–LX Edited by EDWARD A. DENNIS AND MARTHA G. DENNIS VOLUME 76. Hemoglobins Edited by ERALDO ANTONINI, LUIGI ROSSI-BERNARDI, AND EMILIA CHIANCONE VOLUME 77. Detoxication and Drug Metabolism Edited by WILLIAM B. JAKOBY VOLUME 78. Interferons (Part A) Edited by SIDNEY PESTKA VOLUME 79. Interferons (Part B) Edited by SIDNEY PESTKA VOLUME 80. Proteolytic Enzymes (Part C) Edited by LASZLO LORAND VOLUME 81. Biomembranes (Part H: Visual Pigments and Purple Membranes, I) Edited by LESTER PACKER VOLUME 82. Structural and Contractile Proteins (Part A: Extracellular Matrix) Edited by LEON W. CUNNINGHAM AND DIXIE W. FREDERIKSEN VOLUME 83. Complex Carbohydrates (Part D) Edited by VICTOR GINSBURG VOLUME 84. Immunochemical Techniques (Part D: Selected Immunoassays) Edited by JOHN J. LANGONE AND HELEN VAN VUNAKIS VOLUME 85. Structural and Contractile Proteins (Part B: The Contractile Apparatus and the Cytoskeleton) Edited by DIXIE W. FREDERIKSEN AND LEON W. CUNNINGHAM VOLUME 86. Prostaglandins and Arachidonate Metabolites Edited by WILLIAM E. M. LANDS AND WILLIAM L. SMITH VOLUME 87. Enzyme Kinetics and Mechanism (Part C: Intermediates, Stereo-chemistry, and Rate Studies) Edited by DANIEL L. PURICH VOLUME 88. Biomembranes (Part I: Visual Pigments and Purple Membranes, II) Edited by LESTER PACKER

xxiv

Methods in Enzymology

VOLUME 89. Carbohydrate Metabolism (Part D) Edited by WILLIS A. WOOD VOLUME 90. Carbohydrate Metabolism (Part E) Edited by WILLIS A. WOOD VOLUME 91. Enzyme Structure (Part I) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME 92. Immunochemical Techniques (Part E: Monoclonal Antibodies and General Immunoassay Methods) Edited by JOHN J. LANGONE AND HELEN VAN VUNAKIS VOLUME 93. Immunochemical Techniques (Part F: Conventional Antibodies, Fc Receptors, and Cytotoxicity) Edited by JOHN J. LANGONE AND HELEN VAN VUNAKIS VOLUME 94. Polyamines Edited by HERBERT TABOR AND CELIA WHITE TABOR VOLUME 95. Cumulative Subject Index Volumes 61–74, 76–80 Edited by EDWARD A. DENNIS AND MARTHA G. DENNIS VOLUME 96. Biomembranes [Part J: Membrane Biogenesis: Assembly and Targeting (General Methods; Eukaryotes)] Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 97. Biomembranes [Part K: Membrane Biogenesis: Assembly and Targeting (Prokaryotes, Mitochondria, and Chloroplasts)] Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 98. Biomembranes (Part L: Membrane Biogenesis: Processing and Recycling) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 99. Hormone Action (Part F: Protein Kinases) Edited by JACKIE D. CORBIN AND JOEL G. HARDMAN VOLUME 100. Recombinant DNA (Part B) Edited by RAY WU, LAWRENCE GROSSMAN, AND KIVIE MOLDAVE VOLUME 101. Recombinant DNA (Part C) Edited by RAY WU, LAWRENCE GROSSMAN, AND KIVIE MOLDAVE VOLUME 102. Hormone Action (Part G: Calmodulin and Calcium-Binding Proteins) Edited by ANTHONY R. MEANS AND BERT W. O’MALLEY VOLUME 103. Hormone Action (Part H: Neuroendocrine Peptides) Edited by P. MICHAEL CONN VOLUME 104. Enzyme Purification and Related Techniques (Part C) Edited by WILLIAM B. JAKOBY

Methods in Enzymology

xxv

VOLUME 105. Oxygen Radicals in Biological Systems Edited by LESTER PACKER VOLUME 106. Posttranslational Modifications (Part A) Edited by FINN WOLD AND KIVIE MOLDAVE VOLUME 107. Posttranslational Modifications (Part B) Edited by FINN WOLD AND KIVIE MOLDAVE VOLUME 108. Immunochemical Techniques (Part G: Separation and Characterization of Lymphoid Cells) Edited by GIOVANNI DI SABATO, JOHN J. LANGONE, AND HELEN VAN VUNAKIS VOLUME 109. Hormone Action (Part I: Peptide Hormones) Edited by LUTZ BIRNBAUMER AND BERT W. O’MALLEY VOLUME 110. Steroids and Isoprenoids (Part A) Edited by JOHN H. LAW AND HANS C. RILLING VOLUME 111. Steroids and Isoprenoids (Part B) Edited by JOHN H. LAW AND HANS C. RILLING VOLUME 112. Drug and Enzyme Targeting (Part A) Edited by KENNETH J. WIDDER AND RALPH GREEN VOLUME 113. Glutamate, Glutamine, Glutathione, and Related Compounds Edited by ALTON MEISTER VOLUME 114. Diffraction Methods for Biological Macromolecules (Part A) Edited by HAROLD W. WYCKOFF, C. H. W. HIRS, AND SERGE N. TIMASHEFF VOLUME 115. Diffraction Methods for Biological Macromolecules (Part B) Edited by HAROLD W. WYCKOFF, C. H. W. HIRS, AND SERGE N. TIMASHEFF VOLUME 116. Immunochemical Techniques (Part H: Effectors and Mediators of Lymphoid Cell Functions) Edited by GIOVANNI DI SABATO, JOHN J. LANGONE, AND HELEN VAN VUNAKIS VOLUME 117. Enzyme Structure (Part J) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME 118. Plant Molecular Biology Edited by ARTHUR WEISSBACH AND HERBERT WEISSBACH VOLUME 119. Interferons (Part C) Edited by SIDNEY PESTKA VOLUME 120. Cumulative Subject Index Volumes 81–94, 96–101 VOLUME 121. Immunochemical Techniques (Part I: Hybridoma Technology and Monoclonal Antibodies) Edited by JOHN J. LANGONE AND HELEN VAN VUNAKIS VOLUME 122. Vitamins and Coenzymes (Part G) Edited by FRANK CHYTIL AND DONALD B. MCCORMICK

xxvi

Methods in Enzymology

VOLUME 123. Vitamins and Coenzymes (Part H) Edited by FRANK CHYTIL AND DONALD B. MCCORMICK VOLUME 124. Hormone Action (Part J: Neuroendocrine Peptides) Edited by P. MICHAEL CONN VOLUME 125. Biomembranes (Part M: Transport in Bacteria, Mitochondria, and Chloroplasts: General Approaches and Transport Systems) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 126. Biomembranes (Part N: Transport in Bacteria, Mitochondria, and Chloroplasts: Protonmotive Force) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 127. Biomembranes (Part O: Protons and Water: Structure and Translocation) Edited by LESTER PACKER VOLUME 128. Plasma Lipoproteins (Part A: Preparation, Structure, and Molecular Biology) Edited by JERE P. SEGREST AND JOHN J. ALBERS VOLUME 129. Plasma Lipoproteins (Part B: Characterization, Cell Biology, and Metabolism) Edited by JOHN J. ALBERS AND JERE P. SEGREST VOLUME 130. Enzyme Structure (Part K) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME 131. Enzyme Structure (Part L) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME 132. Immunochemical Techniques (Part J: Phagocytosis and Cell-Mediated Cytotoxicity) Edited by GIOVANNI DI SABATO AND JOHANNES EVERSE VOLUME 133. Bioluminescence and Chemiluminescence (Part B) Edited by MARLENE DELUCA AND WILLIAM D. MCELROY VOLUME 134. Structural and Contractile Proteins (Part C: The Contractile Apparatus and the Cytoskeleton) Edited by RICHARD B. VALLEE VOLUME 135. Immobilized Enzymes and Cells (Part B) Edited by KLAUS MOSBACH VOLUME 136. Immobilized Enzymes and Cells (Part C) Edited by KLAUS MOSBACH VOLUME 137. Immobilized Enzymes and Cells (Part D) Edited by KLAUS MOSBACH VOLUME 138. Complex Carbohydrates (Part E) Edited by VICTOR GINSBURG

Methods in Enzymology

xxvii

VOLUME 139. Cellular Regulators (Part A: Calcium- and Calmodulin-Binding Proteins) Edited by ANTHONY R. MEANS AND P. MICHAEL CONN VOLUME 140. Cumulative Subject Index Volumes 102–119, 121–134 VOLUME 141. Cellular Regulators (Part B: Calcium and Lipids) Edited by P. MICHAEL CONN AND ANTHONY R. MEANS VOLUME 142. Metabolism of Aromatic Amino Acids and Amines Edited by SEYMOUR KAUFMAN VOLUME 143. Sulfur and Sulfur Amino Acids Edited by WILLIAM B. JAKOBY AND OWEN GRIFFITH VOLUME 144. Structural and Contractile Proteins (Part D: Extracellular Matrix) Edited by LEON W. CUNNINGHAM VOLUME 145. Structural and Contractile Proteins (Part E: Extracellular Matrix) Edited by LEON W. CUNNINGHAM VOLUME 146. Peptide Growth Factors (Part A) Edited by DAVID BARNES AND DAVID A. SIRBASKU VOLUME 147. Peptide Growth Factors (Part B) Edited by DAVID BARNES AND DAVID A. SIRBASKU VOLUME 148. Plant Cell Membranes Edited by LESTER PACKER AND ROLAND DOUCE VOLUME 149. Drug and Enzyme Targeting (Part B) Edited by RALPH GREEN AND KENNETH J. WIDDER VOLUME 150. Immunochemical Techniques (Part K: In Vitro Models of B and T Cell Functions and Lymphoid Cell Receptors) Edited by GIOVANNI DI SABATO VOLUME 151. Molecular Genetics of Mammalian Cells Edited by MICHAEL M. GOTTESMAN VOLUME 152. Guide to Molecular Cloning Techniques Edited by SHELBY L. BERGER AND ALAN R. KIMMEL VOLUME 153. Recombinant DNA (Part D) Edited by RAY WU AND LAWRENCE GROSSMAN VOLUME 154. Recombinant DNA (Part E) Edited by RAY WU AND LAWRENCE GROSSMAN VOLUME 155. Recombinant DNA (Part F) Edited by RAY WU VOLUME 156. Biomembranes (Part P: ATP-Driven Pumps and Related Transport: The Na, K-Pump) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER

xxviii

Methods in Enzymology

VOLUME 157. Biomembranes (Part Q: ATP-Driven Pumps and Related Transport: Calcium, Proton, and Potassium Pumps) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 158. Metalloproteins (Part A) Edited by JAMES F. RIORDAN AND BERT L. VALLEE VOLUME 159. Initiation and Termination of Cyclic Nucleotide Action Edited by JACKIE D. CORBIN AND ROGER A. JOHNSON VOLUME 160. Biomass (Part A: Cellulose and Hemicellulose) Edited by WILLIS A. WOOD AND SCOTT T. KELLOGG VOLUME 161. Biomass (Part B: Lignin, Pectin, and Chitin) Edited by WILLIS A. WOOD AND SCOTT T. KELLOGG VOLUME 162. Immunochemical Techniques (Part L: Chemotaxis and Inflammation) Edited by GIOVANNI DI SABATO VOLUME 163. Immunochemical Techniques (Part M: Chemotaxis and Inflammation) Edited by GIOVANNI DI SABATO VOLUME 164. Ribosomes Edited by HARRY F. NOLLER, JR., AND KIVIE MOLDAVE VOLUME 165. Microbial Toxins: Tools for Enzymology Edited by SIDNEY HARSHMAN VOLUME 166. Branched-Chain Amino Acids Edited by ROBERT HARRIS AND JOHN R. SOKATCH VOLUME 167. Cyanobacteria Edited by LESTER PACKER AND ALEXANDER N. GLAZER VOLUME 168. Hormone Action (Part K: Neuroendocrine Peptides) Edited by P. MICHAEL CONN VOLUME 169. Platelets: Receptors, Adhesion, Secretion (Part A) Edited by JACEK HAWIGER VOLUME 170. Nucleosomes Edited by PAUL M. WASSARMAN AND ROGER D. KORNBERG VOLUME 171. Biomembranes (Part R: Transport Theory: Cells and Model Membranes) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 172. Biomembranes (Part S: Transport: Membrane Isolation and Characterization) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER

Methods in Enzymology

xxix

VOLUME 173. Biomembranes [Part T: Cellular and Subcellular Transport: Eukaryotic (Nonepithelial) Cells] Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 174. Biomembranes [Part U: Cellular and Subcellular Transport: Eukaryotic (Nonepithelial) Cells] Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 175. Cumulative Subject Index Volumes 135–139, 141–167 VOLUME 176. Nuclear Magnetic Resonance (Part A: Spectral Techniques and Dynamics) Edited by NORMAN J. OPPENHEIMER AND THOMAS L. JAMES VOLUME 177. Nuclear Magnetic Resonance (Part B: Structure and Mechanism) Edited by NORMAN J. OPPENHEIMER AND THOMAS L. JAMES VOLUME 178. Antibodies, Antigens, and Molecular Mimicry Edited by JOHN J. LANGONE VOLUME 179. Complex Carbohydrates (Part F) Edited by VICTOR GINSBURG VOLUME 180. RNA Processing (Part A: General Methods) Edited by JAMES E. DAHLBERG AND JOHN N. ABELSON VOLUME 181. RNA Processing (Part B: Specific Methods) Edited by JAMES E. DAHLBERG AND JOHN N. ABELSON VOLUME 182. Guide to Protein Purification Edited by MURRAY P. DEUTSCHER VOLUME 183. Molecular Evolution: Computer Analysis of Protein and Nucleic Acid Sequences Edited by RUSSELL F. DOOLITTLE VOLUME 184. Avidin-Biotin Technology Edited by MEIR WILCHEK AND EDWARD A. BAYER VOLUME 185. Gene Expression Technology Edited by DAVID V. GOEDDEL VOLUME 186. Oxygen Radicals in Biological Systems (Part B: Oxygen Radicals and Antioxidants) Edited by LESTER PACKER AND ALEXANDER N. GLAZER VOLUME 187. Arachidonate Related Lipid Mediators Edited by ROBERT C. MURPHY AND FRANK A. FITZPATRICK VOLUME 188. Hydrocarbons and Methylotrophy Edited by MARY E. LIDSTROM VOLUME 189. Retinoids (Part A: Molecular and Metabolic Aspects) Edited by LESTER PACKER

xxx

Methods in Enzymology

VOLUME 190. Retinoids (Part B: Cell Differentiation and Clinical Applications) Edited by LESTER PACKER VOLUME 191. Biomembranes (Part V: Cellular and Subcellular Transport: Epithelial Cells) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 192. Biomembranes (Part W: Cellular and Subcellular Transport: Epithelial Cells) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 193. Mass Spectrometry Edited by JAMES A. MCCLOSKEY VOLUME 194. Guide to Yeast Genetics and Molecular Biology Edited by CHRISTINE GUTHRIE AND GERALD R. FINK VOLUME 195. Adenylyl Cyclase, G Proteins, and Guanylyl Cyclase Edited by ROGER A. JOHNSON AND JACKIE D. CORBIN VOLUME 196. Molecular Motors and the Cytoskeleton Edited by RICHARD B. VALLEE VOLUME 197. Phospholipases Edited by EDWARD A. DENNIS VOLUME 198. Peptide Growth Factors (Part C) Edited by DAVID BARNES, J. P. MATHER, AND GORDON H. SATO VOLUME 199. Cumulative Subject Index Volumes 168–174, 176–194 VOLUME 200. Protein Phosphorylation (Part A: Protein Kinases: Assays, Purification, Antibodies, Functional Analysis, Cloning, and Expression) Edited by TONY HUNTER AND BARTHOLOMEW M. SEFTON VOLUME 201. Protein Phosphorylation (Part B: Analysis of Protein Phosphorylation, Protein Kinase Inhibitors, and Protein Phosphatases) Edited by TONY HUNTER AND BARTHOLOMEW M. SEFTON VOLUME 202. Molecular Design and Modeling: Concepts and Applications (Part A: Proteins, Peptides, and Enzymes) Edited by JOHN J. LANGONE VOLUME 203. Molecular Design and Modeling: Concepts and Applications (Part B: Antibodies and Antigens, Nucleic Acids, Polysaccharides, and Drugs) Edited by JOHN J. LANGONE VOLUME 204. Bacterial Genetic Systems Edited by JEFFREY H. MILLER VOLUME 205. Metallobiochemistry (Part B: Metallothionein and Related Molecules) Edited by JAMES F. RIORDAN AND BERT L. VALLEE

Methods in Enzymology

xxxi

VOLUME 206. Cytochrome P450 Edited by MICHAEL R. WATERMAN AND ERIC F. JOHNSON VOLUME 207. Ion Channels Edited by BERNARDO RUDY AND LINDA E. IVERSON VOLUME 208. Protein–DNA Interactions Edited by ROBERT T. SAUER VOLUME 209. Phospholipid Biosynthesis Edited by EDWARD A. DENNIS AND DENNIS E. VANCE VOLUME 210. Numerical Computer Methods Edited by LUDWIG BRAND AND MICHAEL L. JOHNSON VOLUME 211. DNA Structures (Part A: Synthesis and Physical Analysis of DNA) Edited by DAVID M. J. LILLEY AND JAMES E. DAHLBERG VOLUME 212. DNA Structures (Part B: Chemical and Electrophoretic Analysis of DNA) Edited by DAVID M. J. LILLEY AND JAMES E. DAHLBERG VOLUME 213. Carotenoids (Part A: Chemistry, Separation, Quantitation, and Antioxidation) Edited by LESTER PACKER VOLUME 214. Carotenoids (Part B: Metabolism, Genetics, and Biosynthesis) Edited by LESTER PACKER VOLUME 215. Platelets: Receptors, Adhesion, Secretion (Part B) Edited by JACEK J. HAWIGER VOLUME 216. Recombinant DNA (Part G) Edited by RAY WU VOLUME 217. Recombinant DNA (Part H) Edited by RAY WU VOLUME 218. Recombinant DNA (Part I) Edited by RAY WU VOLUME 219. Reconstitution of Intracellular Transport Edited by JAMES E. ROTHMAN VOLUME 220. Membrane Fusion Techniques (Part A) Edited by NEJAT DU¨ZGU¨NES, VOLUME 221. Membrane Fusion Techniques (Part B) Edited by NEJAT DU¨ZGU¨NES, VOLUME 222. Proteolytic Enzymes in Coagulation, Fibrinolysis, and Complement Activation (Part A: Mammalian Blood Coagulation Factors and Inhibitors) Edited by LASZLO LORAND AND KENNETH G. MANN

xxxii

Methods in Enzymology

VOLUME 223. Proteolytic Enzymes in Coagulation, Fibrinolysis, and Complement Activation (Part B: Complement Activation, Fibrinolysis, and Nonmammalian Blood Coagulation Factors) Edited by LASZLO LORAND AND KENNETH G. MANN VOLUME 224. Molecular Evolution: Producing the Biochemical Data Edited by ELIZABETH ANNE ZIMMER, THOMAS J. WHITE, REBECCA L. CANN, AND ALLAN C. WILSON VOLUME 225. Guide to Techniques in Mouse Development Edited by PAUL M. WASSARMAN AND MELVIN L. DEPAMPHILIS VOLUME 226. Metallobiochemistry (Part C: Spectroscopic and Physical Methods for Probing Metal Ion Environments in Metalloenzymes and Metalloproteins) Edited by JAMES F. RIORDAN AND BERT L. VALLEE VOLUME 227. Metallobiochemistry (Part D: Physical and Spectroscopic Methods for Probing Metal Ion Environments in Metalloproteins) Edited by JAMES F. RIORDAN AND BERT L. VALLEE VOLUME 228. Aqueous Two-Phase Systems Edited by HARRY WALTER AND GO¨TE JOHANSSON VOLUME 229. Cumulative Subject Index Volumes 195–198, 200–227 VOLUME 230. Guide to Techniques in Glycobiology Edited by WILLIAM J. LENNARZ AND GERALD W. HART VOLUME 231. Hemoglobins (Part B: Biochemical and Analytical Methods) Edited by JOHANNES EVERSE, KIM D. VANDEGRIFF, AND ROBERT M. WINSLOW VOLUME 232. Hemoglobins (Part C: Biophysical Methods) Edited by JOHANNES EVERSE, KIM D. VANDEGRIFF, AND ROBERT M. WINSLOW VOLUME 233. Oxygen Radicals in Biological Systems (Part C) Edited by LESTER PACKER VOLUME 234. Oxygen Radicals in Biological Systems (Part D) Edited by LESTER PACKER VOLUME 235. Bacterial Pathogenesis (Part A: Identification and Regulation of Virulence Factors) Edited by VIRGINIA L. CLARK AND PATRIK M. BAVOIL VOLUME 236. Bacterial Pathogenesis (Part B: Integration of Pathogenic Bacteria with Host Cells) Edited by VIRGINIA L. CLARK AND PATRIK M. BAVOIL VOLUME 237. Heterotrimeric G Proteins Edited by RAVI IYENGAR VOLUME 238. Heterotrimeric G-Protein Effectors Edited by RAVI IYENGAR

Methods in Enzymology

xxxiii

VOLUME 239. Nuclear Magnetic Resonance (Part C) Edited by THOMAS L. JAMES AND NORMAN J. OPPENHEIMER VOLUME 240. Numerical Computer Methods (Part B) Edited by MICHAEL L. JOHNSON AND LUDWIG BRAND VOLUME 241. Retroviral Proteases Edited by LAWRENCE C. KUO AND JULES A. SHAFER VOLUME 242. Neoglycoconjugates (Part A) Edited by Y. C. LEE AND REIKO T. LEE VOLUME 243. Inorganic Microbial Sulfur Metabolism Edited by HARRY D. PECK, JR., AND JEAN LEGALL VOLUME 244. Proteolytic Enzymes: Serine and Cysteine Peptidases Edited by ALAN J. BARRETT VOLUME 245. Extracellular Matrix Components Edited by E. RUOSLAHTI AND E. ENGVALL VOLUME 246. Biochemical Spectroscopy Edited by KENNETH SAUER VOLUME 247. Neoglycoconjugates (Part B: Biomedical Applications) Edited by Y. C. LEE AND REIKO T. LEE VOLUME 248. Proteolytic Enzymes: Aspartic and Metallo Peptidases Edited by ALAN J. BARRETT VOLUME 249. Enzyme Kinetics and Mechanism (Part D: Developments in Enzyme Dynamics) Edited by DANIEL L. PURICH VOLUME 250. Lipid Modifications of Proteins Edited by PATRICK J. CASEY AND JANICE E. BUSS VOLUME 251. Biothiols (Part A: Monothiols and Dithiols, Protein Thiols, and Thiyl Radicals) Edited by LESTER PACKER VOLUME 252. Biothiols (Part B: Glutathione and Thioredoxin; Thiols in Signal Transduction and Gene Regulation) Edited by LESTER PACKER VOLUME 253. Adhesion of Microbial Pathogens Edited by RON J. DOYLE AND ITZHAK OFEK VOLUME 254. Oncogene Techniques Edited by PETER K. VOGT AND INDER M. VERMA VOLUME 255. Small GTPases and Their Regulators (Part A: Ras Family) Edited by W. E. BALCH, CHANNING J. DER, AND ALAN HALL

xxxiv

Methods in Enzymology

VOLUME 256. Small GTPases and Their Regulators (Part B: Rho Family) Edited by W. E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 257. Small GTPases and Their Regulators (Part C: Proteins Involved in Transport) Edited by W. E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 258. Redox-Active Amino Acids in Biology Edited by JUDITH P. KLINMAN VOLUME 259. Energetics of Biological Macromolecules Edited by MICHAEL L. JOHNSON AND GARY K. ACKERS VOLUME 260. Mitochondrial Biogenesis and Genetics (Part A) Edited by GIUSEPPE M. ATTARDI AND ANNE CHOMYN VOLUME 261. Nuclear Magnetic Resonance and Nucleic Acids Edited by THOMAS L. JAMES VOLUME 262. DNA Replication Edited by JUDITH L. CAMPBELL VOLUME 263. Plasma Lipoproteins (Part C: Quantitation) Edited by WILLIAM A. BRADLEY, SANDRA H. GIANTURCO, AND JERE P. SEGREST VOLUME 264. Mitochondrial Biogenesis and Genetics (Part B) Edited by GIUSEPPE M. ATTARDI AND ANNE CHOMYN VOLUME 265. Cumulative Subject Index Volumes 228, 230–262 VOLUME 266. Computer Methods for Macromolecular Sequence Analysis Edited by RUSSELL F. DOOLITTLE VOLUME 267. Combinatorial Chemistry Edited by JOHN N. ABELSON VOLUME 268. Nitric Oxide (Part A: Sources and Detection of NO; NO Synthase) Edited by LESTER PACKER VOLUME 269. Nitric Oxide (Part B: Physiological and Pathological Processes) Edited by LESTER PACKER VOLUME 270. High Resolution Separation and Analysis of Biological Macromolecules (Part A: Fundamentals) Edited by BARRY L. KARGER AND WILLIAM S. HANCOCK VOLUME 271. High Resolution Separation and Analysis of Biological Macromolecules (Part B: Applications) Edited by BARRY L. KARGER AND WILLIAM S. HANCOCK VOLUME 272. Cytochrome P450 (Part B) Edited by ERIC F. JOHNSON AND MICHAEL R. WATERMAN VOLUME 273. RNA Polymerase and Associated Factors (Part A) Edited by SANKAR ADHYA

Methods in Enzymology

xxxv

VOLUME 274. RNA Polymerase and Associated Factors (Part B) Edited by SANKAR ADHYA VOLUME 275. Viral Polymerases and Related Proteins Edited by LAWRENCE C. KUO, DAVID B. OLSEN, AND STEVEN S. CARROLL VOLUME 276. Macromolecular Crystallography (Part A) Edited by CHARLES W. CARTER, JR., AND ROBERT M. SWEET VOLUME 277. Macromolecular Crystallography (Part B) Edited by CHARLES W. CARTER, JR., AND ROBERT M. SWEET VOLUME 278. Fluorescence Spectroscopy Edited by LUDWIG BRAND AND MICHAEL L. JOHNSON VOLUME 279. Vitamins and Coenzymes (Part I) Edited by DONALD B. MCCORMICK, JOHN W. SUTTIE, AND CONRAD WAGNER VOLUME 280. Vitamins and Coenzymes (Part J) Edited by DONALD B. MCCORMICK, JOHN W. SUTTIE, AND CONRAD WAGNER VOLUME 281. Vitamins and Coenzymes (Part K) Edited by DONALD B. MCCORMICK, JOHN W. SUTTIE, AND CONRAD WAGNER VOLUME 282. Vitamins and Coenzymes (Part L) Edited by DONALD B. MCCORMICK, JOHN W. SUTTIE, AND CONRAD WAGNER VOLUME 283. Cell Cycle Control Edited by WILLIAM G. DUNPHY VOLUME 284. Lipases (Part A: Biotechnology) Edited by BYRON RUBIN AND EDWARD A. DENNIS VOLUME 285. Cumulative Subject Index Volumes 263, 264, 266–284, 286–289 VOLUME 286. Lipases (Part B: Enzyme Characterization and Utilization) Edited by BYRON RUBIN AND EDWARD A. DENNIS VOLUME 287. Chemokines Edited by RICHARD HORUK VOLUME 288. Chemokine Receptors Edited by RICHARD HORUK VOLUME 289. Solid Phase Peptide Synthesis Edited by GREGG B. FIELDS VOLUME 290. Molecular Chaperones Edited by GEORGE H. LORIMER AND THOMAS BALDWIN VOLUME 291. Caged Compounds Edited by GERARD MARRIOTT VOLUME 292. ABC Transporters: Biochemical, Cellular, and Molecular Aspects Edited by SURESH V. AMBUDKAR AND MICHAEL M. GOTTESMAN

xxxvi

Methods in Enzymology

VOLUME 293. Ion Channels (Part B) Edited by P. MICHAEL CONN VOLUME 294. Ion Channels (Part C) Edited by P. MICHAEL CONN VOLUME 295. Energetics of Biological Macromolecules (Part B) Edited by GARY K. ACKERS AND MICHAEL L. JOHNSON VOLUME 296. Neurotransmitter Transporters Edited by SUSAN G. AMARA VOLUME 297. Photosynthesis: Molecular Biology of Energy Capture Edited by LEE MCINTOSH VOLUME 298. Molecular Motors and the Cytoskeleton (Part B) Edited by RICHARD B. VALLEE VOLUME 299. Oxidants and Antioxidants (Part A) Edited by LESTER PACKER VOLUME 300. Oxidants and Antioxidants (Part B) Edited by LESTER PACKER VOLUME 301. Nitric Oxide: Biological and Antioxidant Activities (Part C) Edited by LESTER PACKER VOLUME 302. Green Fluorescent Protein Edited by P. MICHAEL CONN VOLUME 303. cDNA Preparation and Display Edited by SHERMAN M. WEISSMAN VOLUME 304. Chromatin Edited by PAUL M. WASSARMAN AND ALAN P. WOLFFE VOLUME 305. Bioluminescence and Chemiluminescence (Part C) Edited by THOMAS O. BALDWIN AND MIRIAM M. ZIEGLER VOLUME 306. Expression of Recombinant Genes in Eukaryotic Systems Edited by JOSEPH C. GLORIOSO AND MARTIN C. SCHMIDT VOLUME 307. Confocal Microscopy Edited by P. MICHAEL CONN VOLUME 308. Enzyme Kinetics and Mechanism (Part E: Energetics of Enzyme Catalysis) Edited by DANIEL L. PURICH AND VERN L. SCHRAMM VOLUME 309. Amyloid, Prions, and Other Protein Aggregates Edited by RONALD WETZEL VOLUME 310. Biofilms Edited by RON J. DOYLE

Methods in Enzymology

xxxvii

VOLUME 311. Sphingolipid Metabolism and Cell Signaling (Part A) Edited by ALFRED H. MERRILL, JR., AND YUSUF A. HANNUN VOLUME 312. Sphingolipid Metabolism and Cell Signaling (Part B) Edited by ALFRED H. MERRILL, JR., AND YUSUF A. HANNUN VOLUME 313. Antisense Technology (Part A: General Methods, Methods of Delivery, and RNA Studies) Edited by M. IAN PHILLIPS VOLUME 314. Antisense Technology (Part B: Applications) Edited by M. IAN PHILLIPS VOLUME 315. Vertebrate Phototransduction and the Visual Cycle (Part A) Edited by KRZYSZTOF PALCZEWSKI VOLUME 316. Vertebrate Phototransduction and the Visual Cycle (Part B) Edited by KRZYSZTOF PALCZEWSKI VOLUME 317. RNA–Ligand Interactions (Part A: Structural Biology Methods) Edited by DANIEL W. CELANDER AND JOHN N. ABELSON VOLUME 318. RNA–Ligand Interactions (Part B: Molecular Biology Methods) Edited by DANIEL W. CELANDER AND JOHN N. ABELSON VOLUME 319. Singlet Oxygen, UV-A, and Ozone Edited by LESTER PACKER AND HELMUT SIES VOLUME 320. Cumulative Subject Index Volumes 290–319 VOLUME 321. Numerical Computer Methods (Part C) Edited by MICHAEL L. JOHNSON AND LUDWIG BRAND VOLUME 322. Apoptosis Edited by JOHN C. REED VOLUME 323. Energetics of Biological Macromolecules (Part C) Edited by MICHAEL L. JOHNSON AND GARY K. ACKERS VOLUME 324. Branched-Chain Amino Acids (Part B) Edited by ROBERT A. HARRIS AND JOHN R. SOKATCH VOLUME 325. Regulators and Effectors of Small GTPases (Part D: Rho Family) Edited by W. E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 326. Applications of Chimeric Genes and Hybrid Proteins (Part A: Gene Expression and Protein Purification) Edited by JEREMY THORNER, SCOTT D. EMR, AND JOHN N. ABELSON VOLUME 327. Applications of Chimeric Genes and Hybrid Proteins (Part B: Cell Biology and Physiology) Edited by JEREMY THORNER, SCOTT D. EMR, AND JOHN N. ABELSON

xxxviii

Methods in Enzymology

VOLUME 328. Applications of Chimeric Genes and Hybrid Proteins (Part C: Protein–Protein Interactions and Genomics) Edited by JEREMY THORNER, SCOTT D. EMR, AND JOHN N. ABELSON VOLUME 329. Regulators and Effectors of Small GTPases (Part E: GTPases Involved in Vesicular Traffic) Edited by W. E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 330. Hyperthermophilic Enzymes (Part A) Edited by MICHAEL W. W. ADAMS AND ROBERT M. KELLY VOLUME 331. Hyperthermophilic Enzymes (Part B) Edited by MICHAEL W. W. ADAMS AND ROBERT M. KELLY VOLUME 332. Regulators and Effectors of Small GTPases (Part F: Ras Family I) Edited by W. E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 333. Regulators and Effectors of Small GTPases (Part G: Ras Family II) Edited by W. E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 334. Hyperthermophilic Enzymes (Part C) Edited by MICHAEL W. W. ADAMS AND ROBERT M. KELLY VOLUME 335. Flavonoids and Other Polyphenols Edited by LESTER PACKER VOLUME 336. Microbial Growth in Biofilms (Part A: Developmental and Molecular Biological Aspects) Edited by RON J. DOYLE VOLUME 337. Microbial Growth in Biofilms (Part B: Special Environments and Physicochemical Aspects) Edited by RON J. DOYLE VOLUME 338. Nuclear Magnetic Resonance of Biological Macromolecules (Part A) Edited by THOMAS L. JAMES, VOLKER DO¨TSCH, AND ULI SCHMITZ VOLUME 339. Nuclear Magnetic Resonance of Biological Macromolecules (Part B) Edited by THOMAS L. JAMES, VOLKER DO¨TSCH, AND ULI SCHMITZ VOLUME 340. Drug–Nucleic Acid Interactions Edited by JONATHAN B. CHAIRES AND MICHAEL J. WARING VOLUME 341. Ribonucleases (Part A) Edited by ALLEN W. NICHOLSON VOLUME 342. Ribonucleases (Part B) Edited by ALLEN W. NICHOLSON VOLUME 343. G Protein Pathways (Part A: Receptors) Edited by RAVI IYENGAR AND JOHN D. HILDEBRANDT VOLUME 344. G Protein Pathways (Part B: G Proteins and Their Regulators) Edited by RAVI IYENGAR AND JOHN D. HILDEBRANDT

Methods in Enzymology

xxxix

VOLUME 345. G Protein Pathways (Part C: Effector Mechanisms) Edited by RAVI IYENGAR AND JOHN D. HILDEBRANDT VOLUME 346. Gene Therapy Methods Edited by M. IAN PHILLIPS VOLUME 347. Protein Sensors and Reactive Oxygen Species (Part A: Selenoproteins and Thioredoxin) Edited by HELMUT SIES AND LESTER PACKER VOLUME 348. Protein Sensors and Reactive Oxygen Species (Part B: Thiol Enzymes and Proteins) Edited by HELMUT SIES AND LESTER PACKER VOLUME 349. Superoxide Dismutase Edited by LESTER PACKER VOLUME 350. Guide to Yeast Genetics and Molecular and Cell Biology (Part B) Edited by CHRISTINE GUTHRIE AND GERALD R. FINK VOLUME 351. Guide to Yeast Genetics and Molecular and Cell Biology (Part C) Edited by CHRISTINE GUTHRIE AND GERALD R. FINK VOLUME 352. Redox Cell Biology and Genetics (Part A) Edited by CHANDAN K. SEN AND LESTER PACKER VOLUME 353. Redox Cell Biology and Genetics (Part B) Edited by CHANDAN K. SEN AND LESTER PACKER VOLUME 354. Enzyme Kinetics and Mechanisms (Part F: Detection and Characterization of Enzyme Reaction Intermediates) Edited by DANIEL L. PURICH VOLUME 355. Cumulative Subject Index Volumes 321–354 VOLUME 356. Laser Capture Microscopy and Microdissection Edited by P. MICHAEL CONN VOLUME 357. Cytochrome P450, Part C Edited by ERIC F. JOHNSON AND MICHAEL R. WATERMAN VOLUME 358. Bacterial Pathogenesis (Part C: Identification, Regulation, and Function of Virulence Factors) Edited by VIRGINIA L. CLARK AND PATRIK M. BAVOIL VOLUME 359. Nitric Oxide (Part D) Edited by ENRIQUE CADENAS AND LESTER PACKER VOLUME 360. Biophotonics (Part A) Edited by GERARD MARRIOTT AND IAN PARKER VOLUME 361. Biophotonics (Part B) Edited by GERARD MARRIOTT AND IAN PARKER

xl

Methods in Enzymology

VOLUME 362. Recognition of Carbohydrates in Biological Systems (Part A) Edited by YUAN C. LEE AND REIKO T. LEE VOLUME 363. Recognition of Carbohydrates in Biological Systems (Part B) Edited by YUAN C. LEE AND REIKO T. LEE VOLUME 364. Nuclear Receptors Edited by DAVID W. RUSSELL AND DAVID J. MANGELSDORF VOLUME 365. Differentiation of Embryonic Stem Cells Edited by PAUL M. WASSAUMAN AND GORDON M. KELLER VOLUME 366. Protein Phosphatases Edited by SUSANNE KLUMPP AND JOSEF KRIEGLSTEIN VOLUME 367. Liposomes (Part A) Edited by NEJAT DU¨ZGU¨NES, VOLUME 368. Macromolecular Crystallography (Part C) Edited by CHARLES W. CARTER, JR., AND ROBERT M. SWEET VOLUME 369. Combinational Chemistry (Part B) Edited by GUILLERMO A. MORALES AND BARRY A. BUNIN VOLUME 370. RNA Polymerases and Associated Factors (Part C) Edited by SANKAR L. ADHYA AND SUSAN GARGES VOLUME 371. RNA Polymerases and Associated Factors (Part D) Edited by SANKAR L. ADHYA AND SUSAN GARGES VOLUME 372. Liposomes (Part B) Edited by NEJAT DU¨ZGU¨NES, VOLUME 373. Liposomes (Part C) Edited by NEJAT DU¨ZGU¨NES, VOLUME 374. Macromolecular Crystallography (Part D) Edited by CHARLES W. CARTER, JR., AND ROBERT W. SWEET VOLUME 375. Chromatin and Chromatin Remodeling Enzymes (Part A) Edited by C. DAVID ALLIS AND CARL WU VOLUME 376. Chromatin and Chromatin Remodeling Enzymes (Part B) Edited by C. DAVID ALLIS AND CARL WU VOLUME 377. Chromatin and Chromatin Remodeling Enzymes (Part C) Edited by C. DAVID ALLIS AND CARL WU VOLUME 378. Quinones and Quinone Enzymes (Part A) Edited by HELMUT SIES AND LESTER PACKER VOLUME 379. Energetics of Biological Macromolecules (Part D) Edited by JO M. HOLT, MICHAEL L. JOHNSON, AND GARY K. ACKERS VOLUME 380. Energetics of Biological Macromolecules (Part E) Edited by JO M. HOLT, MICHAEL L. JOHNSON, AND GARY K. ACKERS

Methods in Enzymology

xli

VOLUME 381. Oxygen Sensing Edited by CHANDAN K. SEN AND GREGG L. SEMENZA VOLUME 382. Quinones and Quinone Enzymes (Part B) Edited by HELMUT SIES AND LESTER PACKER VOLUME 383. Numerical Computer Methods (Part D) Edited by LUDWIG BRAND AND MICHAEL L. JOHNSON VOLUME 384. Numerical Computer Methods (Part E) Edited by LUDWIG BRAND AND MICHAEL L. JOHNSON VOLUME 385. Imaging in Biological Research (Part A) Edited by P. MICHAEL CONN VOLUME 386. Imaging in Biological Research (Part B) Edited by P. MICHAEL CONN VOLUME 387. Liposomes (Part D) Edited by NEJAT DU¨ZGU¨NES, VOLUME 388. Protein Engineering Edited by DAN E. ROBERTSON AND JOSEPH P. NOEL VOLUME 389. Regulators of G-Protein Signaling (Part A) Edited by DAVID P. SIDEROVSKI VOLUME 390. Regulators of G-Protein Signaling (Part B) Edited by DAVID P. SIDEROVSKI VOLUME 391. Liposomes (Part E) Edited by NEJAT DU¨ZGU¨NES, VOLUME 392. RNA Interference Edited by ENGELKE ROSSI VOLUME 393. Circadian Rhythms Edited by MICHAEL W. YOUNG VOLUME 394. Nuclear Magnetic Resonance of Biological Macromolecules (Part C) Edited by THOMAS L. JAMES VOLUME 395. Producing the Biochemical Data (Part B) Edited by ELIZABETH A. ZIMMER AND ERIC H. ROALSON VOLUME 396. Nitric Oxide (Part E) Edited by LESTER PACKER AND ENRIQUE CADENAS VOLUME 397. Environmental Microbiology Edited by JARED R. LEADBETTER VOLUME 398. Ubiquitin and Protein Degradation (Part A) Edited by RAYMOND J. DESHAIES VOLUME 399. Ubiquitin and Protein Degradation (Part B) Edited by RAYMOND J. DESHAIES

xlii

Methods in Enzymology

VOLUME 400. Phase II Conjugation Enzymes and Transport Systems Edited by HELMUT SIES AND LESTER PACKER VOLUME 401. Glutathione Transferases and Gamma Glutamyl Transpeptidases Edited by HELMUT SIES AND LESTER PACKER VOLUME 402. Biological Mass Spectrometry Edited by A. L. BURLINGAME VOLUME 403. GTPases Regulating Membrane Targeting and Fusion Edited by WILLIAM E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 404. GTPases Regulating Membrane Dynamics Edited by WILLIAM E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 405. Mass Spectrometry: Modified Proteins and Glycoconjugates Edited by A. L. BURLINGAME VOLUME 406. Regulators and Effectors of Small GTPases: Rho Family Edited by WILLIAM E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 407. Regulators and Effectors of Small GTPases: Ras Family Edited by WILLIAM E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 408. DNA Repair (Part A) Edited by JUDITH L. CAMPBELL AND PAUL MODRICH VOLUME 409. DNA Repair (Part B) Edited by JUDITH L. CAMPBELL AND PAUL MODRICH VOLUME 410. DNA Microarrays (Part A: Array Platforms and Web-Bench Protocols) Edited by ALAN KIMMEL AND BRIAN OLIVER VOLUME 411. DNA Microarrays (Part B: Databases and Statistics) Edited by ALAN KIMMEL AND BRIAN OLIVER VOLUME 412. Amyloid, Prions, and Other Protein Aggregates (Part B) Edited by INDU KHETERPAL AND RONALD WETZEL VOLUME 413. Amyloid, Prions, and Other Protein Aggregates (Part C) Edited by INDU KHETERPAL AND RONALD WETZEL VOLUME 414. Measuring Biological Responses with Automated Microscopy Edited by JAMES INGLESE VOLUME 415. Glycobiology Edited by MINORU FUKUDA VOLUME 416. Glycomics Edited by MINORU FUKUDA VOLUME 417. Functional Glycomics Edited by MINORU FUKUDA

Methods in Enzymology

xliii

VOLUME 418. Embryonic Stem Cells Edited by IRINA KLIMANSKAYA AND ROBERT LANZA VOLUME 419. Adult Stem Cells Edited by IRINA KLIMANSKAYA AND ROBERT LANZA VOLUME 420. Stem Cell Tools and Other Experimental Protocols Edited by IRINA KLIMANSKAYA AND ROBERT LANZA VOLUME 421. Advanced Bacterial Genetics: Use of Transposons and Phage for Genomic Engineering Edited by KELLY T. HUGHES VOLUME 422. Two-Component Signaling Systems, Part A Edited by MELVIN I. SIMON, BRIAN R. CRANE, AND ALEXANDRINE CRANE VOLUME 423. Two-Component Signaling Systems, Part B Edited by MELVIN I. SIMON, BRIAN R. CRANE, AND ALEXANDRINE CRANE VOLUME 424. RNA Editing Edited by JONATHA M. GOTT VOLUME 425. RNA Modification Edited by JONATHA M. GOTT VOLUME 426. Integrins Edited by DAVID CHERESH VOLUME 427. MicroRNA Methods Edited by JOHN J. ROSSI VOLUME 428. Osmosensing and Osmosignaling Edited by HELMUT SIES AND DIETER HAUSSINGER VOLUME 429. Translation Initiation: Extract Systems and Molecular Genetics Edited by JON LORSCH VOLUME 430. Translation Initiation: Reconstituted Systems and Biophysical Methods Edited by JON LORSCH VOLUME 431. Translation Initiation: Cell Biology, High-Throughput and Chemical-Based Approaches Edited by JON LORSCH VOLUME 432. Lipidomics and Bioactive Lipids: Mass-Spectrometry–Based Lipid Analysis Edited by H. ALEX BROWN VOLUME 433. Lipidomics and Bioactive Lipids: Specialized Analytical Methods and Lipids in Disease Edited by H. ALEX BROWN

xliv

Methods in Enzymology

VOLUME 434. Lipidomics and Bioactive Lipids: Lipids and Cell Signaling Edited by H. ALEX BROWN VOLUME 435. Oxygen Biology and Hypoxia Edited by HELMUT SIES AND BERNHARD BRU¨NE VOLUME 436. Globins and Other Nitric Oxide-Reactive Protiens (Part A) Edited by ROBERT K. POOLE VOLUME 437. Globins and Other Nitric Oxide-Reactive Protiens (Part B) Edited by ROBERT K. POOLE VOLUME 438. Small GTPases in Disease (Part A) Edited by WILLIAM E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 439. Small GTPases in Disease (Part B) Edited by WILLIAM E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 440. Nitric Oxide, Part F Oxidative and Nitrosative Stress in Redox Regulation of Cell Signaling Edited by ENRIQUE CADENAS AND LESTER PACKER VOLUME 441. Nitric Oxide, Part G Oxidative and Nitrosative Stress in Redox Regulation of Cell Signaling Edited by ENRIQUE CADENAS AND LESTER PACKER VOLUME 442. Programmed Cell Death, General Principles for Studying Cell Death (Part A) Edited by ROYA KHOSRAVI-FAR, ZAHRA ZAKERI, RICHARD A. LOCKSHIN, AND MAURO PIACENTINI VOLUME 443. Angiogenesis: In Vitro Systems Edited by DAVID A. CHERESH VOLUME 444. Angiogenesis: In Vivo Systems (Part A) Edited by DAVID A. CHERESH VOLUME 445. Angiogenesis: In Vivo Systems (Part B) Edited by DAVID A. CHERESH VOLUME 446. Programmed Cell Death, The Biology and Therapeutic Implications of Cell Death (Part B) Edited by ROYA KHOSRAVI-FAR, ZAHRA ZAKERI, RICHARD A. LOCKSHIN, AND MAURO PIACENTINI VOLUME 447. RNA Turnover in Bacteria, Archaea and Organelles Edited by LYNNE E. MAQUAT AND CECILIA M. ARRAIANO VOLUME 448. RNA Turnover in Eukaryotes: Nucleases, Pathways and Analysis of mRNA Decay Edited by LYNNE E. MAQUAT AND MEGERDITCH KILEDJIAN

Methods in Enzymology

xlv

VOLUME 449. RNA Turnover in Eukaryotes: Analysis of Specialized and Quality Control RNA Decay Pathways Edited by LYNNE E. MAQUAT AND MEGERDITCH KILEDJIAN VOLUME 450. Fluorescence Spectroscopy Edited by LUDWIG BRAND AND MICHAEL L. JOHNSON VOLUME 451. Autophagy: Lower Eukaryotes and Non-Mammalian Systems (Part A) Edited by DANIEL J. KLIONSKY VOLUME 452. Autophagy in Mammalian Systems (Part B) Edited by DANIEL J. KLIONSKY VOLUME 453. Autophagy in Disease and Clinical Applications (Part C) Edited by DANIEL J. KLIONSKY VOLUME 454. Computer Methods (Part A) Edited by MICHAEL L. JOHNSON AND LUDWIG BRAND VOLUME 455. Biothermodynamics (Part A) Edited by MICHAEL L. JOHNSON, JO M. HOLT, AND GARY K. ACKERS (RETIRED) VOLUME 456. Mitochondrial Function, Part A: Mitochondrial Electron Transport Complexes and Reactive Oxygen Species Edited by WILLIAM S. ALLISON AND IMMO E. SCHEFFLER VOLUME 457. Mitochondrial Function, Part B: Mitochondrial Protein Kinases, Protein Phosphatases and Mitochondrial Diseases Edited by WILLIAM S. ALLISON AND ANNE N. MURPHY VOLUME 458. Complex Enzymes in Microbial Natural Product Biosynthesis, Part A: Overview Articles and Peptides Edited by DAVID A. HOPWOOD VOLUME 459. Complex Enzymes in Microbial Natural Product Biosynthesis, Part B: Polyketides, Aminocoumarins and Carbohydrates Edited by DAVID A. HOPWOOD VOLUME 460. Chemokines, Part A Edited by TRACY M. HANDEL AND DAMON J. HAMEL VOLUME 461. Chemokines, Part B Edited by TRACY M. HANDEL AND DAMON J. HAMEL VOLUME 462. Non-Natural Amino Acids Edited by TOM W. MUIR AND JOHN N. ABELSON VOLUME 463. Guide to Protein Purification, 2nd Edition Edited by RICHARD R. BURGESS AND MURRAY P. DEUTSCHER VOLUME 464. Liposomes, Part F Edited by NEJAT DU¨ZGU¨NES,

This page intentionally left blank

S E C T I O N

O N E

BIOACTIVE LIPOSOMES

This page intentionally left blank

C H A P T E R

O N E

Tubular Liposomes with Variable Permeability for Reconstitution of FtsZ Rings Masaki Osawa and Harold P. Erickson

Contents Introduction Reagents Bacterial Expression of Membrane Targeting FtsZ Purification of FtsZ-mts and FtsZ-YFP-mts 4.1. For FtsZ-mts 4.2. For FtsZ-YFP-mts 5. Renatured Preparation of FtsZ-YFP-mts 6. Tubular Multilamellar Liposome Preparation 7. Permeability of the Multilamellar Liposomes 8. Z-ring Formation in Liposomes 9. A Crude Flow Chamber to Exchange Buffer Outside Liposomes 10. Factors Affecting Z-ring Formation in Liposomes 11. Utility of the Liposomes Beyond FtsZ References 1. 2. 3. 4.

4 4 5 5 6 6 7 7 9 11 11 14 16 16

Abstract We have developed a system for producing tubular multilamellar liposomes that incorporate the protein FtsZ on the inside. We start with a mixture of spherical multilamellar liposomes with FtsZ initially on the outside. Shearing forces generated by applying a coverslip most likely distort some of the spherical liposomes into a tubular shape, and causes some to leak and incorporate FtsZ inside. We describe protocols for liposome preparation, and for preparing membrane-targeted FtsZ that can assemble contractile Z rings inside the tubular liposomes. We also describe the characterization of the multilamellar liposomes in terms of the permeability or leakiness for a small fluorescent dye and larger protein molecules. These liposomes may be useful for reconstitution of other biological systems. Department of Cell Biology, Duke University Medical Center, Durham, North Carolina, USA Methods in Enzymology, Volume 464 ISSN 0076-6879, DOI: 10.1016/S0076-6879(09)64001-5

#

2009 Elsevier Inc. All rights reserved.

3

4

Masaki Osawa and Harold P. Erickson

1. Introduction FtsZ is a bacterial tubulin homologue that forms a ring structure called the ‘‘Z ring’’ at the division plane in bacteria. The Z ring is anchored to the membrane and constricts to divide the bacteria. FtsZ recruits a dozen other essential division proteins, which are mostly involved in remodeling the peptidoglycan cell wall. We recently succeeded in reconstituting Z rings inside tubular liposomes, and found that they generated a constriction force on the liposome wall (Osawa et al., 2008). The assembly of Z rings and the generation of the constriction force were achieved with FtsZ alone, and did not require any other division protein. This was an important discovery itself for understanding the mechanism of bacterial cell division. Now the liposome system we developed should provide a simple in vitro system for studying molecular details of how FtsZ works. To achieve these results we had to overcome two technical problems. The first problem was to tether FtsZ to the membrane. Pichoff and Lutkenhaus (2005) discovered that the carboxy terminus of FtsZ binds to FtsA, and FtsA has an amphipathic helix at its carboxy terminal that inserts into the membrane. We made an FtsZ that could tether itself to the membrane by fusing an amphipathic helix (membrane targeting sequence: mts) to the carboxy terminus of FtsZ. To visualize the protein, we inserted a yellow fluorescent protein (YFP) before the mts, giving FtsZ-YFP-mts. Here we provide detailed protocols for the purification of FtsZ-YFP-mts. The second problem was how to get the protein inside liposomes. We have succeeded in getting FtsZ-YFP-mts inside spherical unilamellar liposomes using the emulsion method (Noireaux and Libchaber, 2004; Pautot et al., 2003). However, we have never found Z rings assembled in such spherical unilamellar liposomes. Eventually, we discovered a procedure that produced tubular multilamellar liposomes, and incorporated FtsZ-YFP-mts inside, where it formed Z rings. Initially this was a fortunate accident, since the cylindrical geometry was not designed, and the FtsZ was initially on the outside. We have since refined the procedures for producing tubular multilamellar liposomes, and we now understand some aspects of the permeability or leakiness that lets FtsZ inside. We describe here our protocols for producing the tubular liposomes and the tests of permeability.

2. Reagents The following reagents are used in our experiments: 

Column buffer: 50 mM Tris/HCl, pH 7.9, 50 mM KCl, 1 mM EDTA, 10% (v/v) glycerol

FtsZ in Tubular Liposomes

     

5

HMKCG buffer: 50 mM HEPES/KOH, pH 7.7, 5 mM MgAc, 300 mM KAc, 50 mM KCl 10% (v/v) glycerol HMK50-350 buffer: 50 mM HEPES/KOH, pH 7.7, 5 mM MgAc, 50–350 mM KAc DOPG:1,2-dioleoyl-sn-glycero-3-[phospho-rac-(1-glycerol)] (Avanti) Egg PC: phosphatidylcholine (Avanti) HccA: 7-Hydroxycoumarin-3-carboxylic acid (Invitrogen) Teflon disc, 37 mm diameter

3. Bacterial Expression of Membrane Targeting FtsZ The FtsZ-YFP-mts is expressed from a pET-11b expression vector, with FtsZ366-YFP-mts or FtsZ366-mts genes inserted at NdeI/BamHI sites (366 indicates that the FtsZ was truncated there, removing the FtsA-binding C-terminal peptide). The YFP we use is the variety Venus (Nagai et al., 2002), which gave superior results in FtsZ fusions in E. coli (Osawa and Erickson, 2005). The mts used here is the amphipathic helix from E. coli MinD (Szeto et al., 2003). We have not yet tested the amphipathic helix from FtsA, which has three to five additional extra amino acids that extend the amphipathic helix (Pichoff and Lutkenhaus, 2005). The expression vector is transformed into E. coli strain C41 (Miroux and Walker, 1996), which gives better yields of soluble proteins than BL21. After transforming, colonies are selected on an LB (Luria broth) agar plate containing 100 mg/ml ampicillin. A colony is picked and cultured overnight in 50 ml LB media with 100 mg/ml ampicillin at 37  C. Five milliliters of the overnight culture is diluted in 500 ml LB and cultured at 37  C until the optical density at 600 nm reaches 0.8–1.0. Protein expression is induced by addition of 0.5 mM IPTG and at the same time the temperature of the shaker is set to 20  C (our shaker takes 1–2 h to reach 20  C). The cells are cultured overnight and spun down at 3750 rpm for 45 min in a Beckman GPR rotor.

4. Purification of FtsZ-mts and FtsZ-YFP-mts Since FtsZ-mts and FtsZ-YFP-mts are expressed as soluble proteins, we purify them using the same protocol as for wild-type FtsZ. The packed cells are resuspended in a final volume of 20 ml column buffer, and 1 mM phenylmethanesulphonylfluoride (PMSF) and 0.1–0.2 mg/ml

6

Masaki Osawa and Harold P. Erickson

lysozyme are added. The mixture is then incubated on a rotator at 4  C for 15 min. They are frozen at 80  C overnight or longer. Two cycles of freeze–thaw (fresh 1 mM PMSF is added after each thawing) method are performed. The resultant mixture is sonicated on ice until the viscosity is reduced. We usually sonicate it for three cycles of 20 s, with 1 min cooling intervals. It is then centrifuged at 32,000 rpm for 20 min at 4  C (Beckman 42.1 Ti rotor). The supernatant is collected and ammonium sulfate is added to 30% saturation (3.52 g dry ammonium sulfate to the 20 ml volume). This mixture is incubated for 20 min on ice and again centrifuged at 32,000 rpm for 20 min at 4  C (Beckman 42.1 Ti rotor). The supernatant is discarded and the pellet is resuspended in 10 ml column buffer and passed through a 0.22 mm filter. The protein is purified on an anion exchange column. A 1  10 cm Source Q column (Source 15Q, GE Healthcare) is used. The column is eluted with a 100 ml gradient from 50–500 mM KCl in column buffer.

4.1. For FtsZ-mts FtsZ has very low UV absorbance, so the peak is located by running each fraction on SDS–PAGE. The peak fractions are pooled and dialyzed into HMK350. The protein concentration is determined by the BCA method (Pierce). FtsZ produces 75% as much color as BSA (Lu et al., 1998), so it is necessary to correct for this. Aliquots are frozen and stored at 80  C.

4.2. For FtsZ-YFP-mts After elution from the Source 15Q column, the peak fractions are pooled. There are typically two peaks: a large main peak and a following small peak, and both peaks have an indistinguishable activity. These peaks can be identified by yellow fluorescence and confirmed by SDS–PAGE. They are concentrated using an Amicon Ultra-15 with centrifugation at 5000g. We have noted that incomplete boiling of FtsZ-YFP-MTS with SDS sample buffer generates two bands on the gel. The upper band (68 Kd) results from completely denatured protein and the lower band (60 Kd), which still has yellow fluorescence in the gel, is due to FtsZ-YFP-mts where the YFP is not denatured. The concentration of FtsZ-YFP-mts can be determined from its absorption at 515 nm, using the extinction coefficient for YFP-Venus 92,200 M 1 cm 1.

FtsZ in Tubular Liposomes

7

Our preferred buffer for FtsZ-YFP-mts is now HMKCG because FtsZ-YFP-mts seems to be more stable, as described below; we now use HMKCG for dialysis, dilution, reaction, and storage buffer.

5. Renatured Preparation of FtsZ-YFP-mts In our previous study (Osawa et al., 2008), we used a renaturing technique to prepare FtsZ366-YFP-mts. We developed this protocol because an early preparation of soluble protein as described above had no activity. We recently found that the soluble protein has full activity, equal to the best fractions of the renatured preparation. Because of the simplicity and much higher yield, we now use the soluble preparation for all work. One curious observation with the renatured FtsZ-YFP-mts was that it lost activity when dialyzed into HMK350. We found that addition of 10% glycerol and 50 mM chloride ion to HMK350 would preserve the activity of the renatured FtsZ-YFP-mts during dialysis. Although this precaution may only be necessary for the renatured FtsZ-YFP-mts, our preferred buffer is now HMKCG, which contains this glycerol and chloride ion.

6. Tubular Multilamellar Liposome Preparation PC and DOPG are dissolved separately in methanol at 100 mg/ml and mixed at a 4:1 ratio (20 ml/5 ml) in a 1.5 ml Eppendorf tube. The mixture is dried with an air current. 250 ml of milliQ water is added to the Eppendorf tube, and the dried lipid is suspended with vigorous vortexing. Many drops (5 ml each, total 250 ml) of the suspension are placed on a 37 mm diameter Teflon disc (Fig. 1.1A) and dried using an air current (Fig. 1.1B). The Teflon disc is placed in a beaker slightly larger than its diameter, and covered with 5 ml reaction buffer. It is then incubated at 37  C overnight to hydrate lipid (Fig. 1.1C). High salt buffers such as HMKCG or HMK350 will give aggregated multilamellar liposomes, which are desired to make tubular liposomes for Z-ring formation. The beaker is gently agitated to generate liposomes (Fig. 1.1D). Too vigorous agitation generates many small liposomes which interfere with observation of Z rings. About 1 ml of the suspension is pipetted out closest to the Teflon, which contains the most concentrated liposomes. This is placed in a 1.5 ml Eppendorf tube and left on a bench for 1 h (Fig. 1.1E). The multilamellar liposomes in HMKCG float to the surface. To obtain concentrated

8

Masaki Osawa and Harold P. Erickson

A

B

C

Liposome suspension (less than 5 ml)

Teflon disc D

Reaction buffer (HMKCG) E

Multilamellar liposome

F

HMKCG

Figure 1.1 Schematic illustration of the production of multilamellar liposomes. (A) A 37 mm Teflon disc is placed in a small beaker and 250 ml total aqueous suspension of lipid is deposited as many small drops (each drop is less than 5 ml). (B) The drops are dried with an air current and (C) the Teflon disc is covered with 5 ml of HMKCG and incubated overnight at 37  C. (D) The Teflon disc is gently agitated and multilamellar liposomes floated off. (E) One microliter of the most concentrated suspension, nearest the Teflon, is transferred to an Eppendorf tube and left on the bench for 1 h. The liposomes rise to the surface. (F) HMKCG is carefully removed from the bottom, leaving a concentrated suspension of liposomes.

liposomes, buffer from the bottom of the tube is carefully pipetted out, leaving the concentrated top layer (Fig. 1.1F). Liposomes that have a thinner wall (including unilamellar liposomes) can be easily obtained by using HMK100 with the same methods. If the KAc is reduced to 50 mM or less, many unilamellar liposomes will be obtained. Interestingly, liposomes prepared in HMK100 settle to the bottom of the Eppendorf tube over several hours. This is the opposite of multilamellar liposomes in HMKCG, which float to the top (probably because the 10% glycerol increases the density of the solution). To produce tubular liposomes, 5 ml of the aggregated multilamellar liposomes in HMKCG or HMK350 is placed on a glass slide and a coverslip is applied. To make liposomes with Z rings, liposomes are mixed first with 1 mM GTP and 4 mM FtsZ-YFP-mts. In this procedure, tubular liposomes are formed, probably generated by shear force when the suspension is spreading quickly through the narrow space between glasses. The tubular liposomes are always a minority, but with practice they can be found reproducibly. The tubular liposomes are

FtsZ in Tubular Liposomes

9

more abundant near the edges of the coverslip, perhaps because that is where shear is maximized. We have noticed that the fluorescence background is higher in the center, where the drop is initially deposited. This may reflect FtsZ-YFP-mts binding to the glass. We have not tested whether the concentration of soluble FtsZ-YFP-mts is substantially reduced by binding to the glass.

7. Permeability of the Multilamellar Liposomes To check the permeability of the multilamellar tubular liposomes that were produced by the above procedures, a polar fluorescence dye HccA is used as a small molecule indicator (MW 206.15 kDa) and FtsZ fused with YFP-Venus (FtsZ-YFP) as a large molecule indicator (MW 67 kDa). Neither of these probes can cross the lipid bilayer, so they should report on smaller or larger pores in the liposomes. Also, the FtsZ-YFP is missing the mts, and was given no GTP, so it would not make protofilaments or Z rings. It should just report the existence of pores large enough to admit the globular protein. When we mix the multilamellar liposomes with 1 mM HccA and 32 mM FtsZ-YFP and apply a coverslip, we find tubular liposomes with both HccA and YFP fluorescence. More than 80% of tubular liposomes have FtsZ-YFP inside and even more liposomes have HccA inside, indicating that multilamellar liposomes are leaky at least for some time after addition of the fluorescent probes and application of the coverslip. A simple FRAP (fluorescent recovery after photobleach) assay is used to determine the leakiness of liposomes 30 min after the coverslip is applied. The entire field of view is exposed to UV from the mercury lamp for 20 s, without any filter. This results in complete bleaching of both fluorescent molecules inside and outside the liposomes. HccA fluorescence on the outside recovers over 20–60 s as unbleached probe diffuses into the area from outside. At 20 s HccA has entered about half of the liposomes (Fig. 1.2F), and by 5 min almost all the liposomes contain fluorescent HccA. Diffusion of FtsZ-YFP is slower, requiring 5 min to recover 70% of original fluorescence on the outside. After 45 min about half of the liposomes have YFP fluorescence inside. We conclude that the multilamellar liposomes are almost all permeable or leaky to the small molecule probe, and about half have pores that will pass the larger protein probe. A curious phenomenon is the appearance of a bright ring on the inner layer of the liposome when the inside of the liposome is dark and the outside is bright (Fig. 1.2D and F). When we repeat the FRAP experiment with a confocal microscope, the bright ring is seen only rarely and is also very dim. We suggest that the bright ring may be an optical artifact seen in the wide field fluorescence microscopy.

10

Masaki Osawa and Harold P. Erickson

A pre

B 1 min

C

D 45 min

5 min

FtsZ-YFP

1 2 3

E pre

H 45 min

F 20 sec 1

HccA

2

G 5 min

3

I pre

J 45 min

DIC

20 mm

Figure 1.2 FRAP assay for checking permeability of multilamellar liposomes. The top row shows FtsZ-YFP (lacking the mts and GTP), the middle row shows HccA fluorescence, and the bottom row shows DIC. The first column shows prebleach images, and the numbers on the other panels show time after bleach. In (D), vesicles 1–3 show little or no recovery of FtsZ-YFP inside, but they do show a thin bright layer of internal fluorescence. The vesicle between 1 and 2 shows substantial recovery of FtsZ-YFP. In (F), (G) and (H) vesicles 1–3 show no recovery of HccA at 20 s, but full recovery at 5 min. The exposure of fluorescence images is kept at the same level from prebleach to postbleach. Notice also the change in shape of several vesicles over 45 min. (B) 50% magnification showing that fluorescence is completely bleached.

FtsZ in Tubular Liposomes

11

8. Z-ring Formation in Liposomes We have shown previously that FtsZ-YFP-mts can form Z rings inside tubular liposomes when FtsZ-YFP-mts plus GTP was mixed with multilamellar liposomes just before applying the coverslip (Osawa et al., 2008). To examine the leakiness of these liposomes with Z rings inside, we repeated the FRAP assay using 8 mM FtsZ-YFP-mts plus 2 mM GTP instead of 32 mM FtsZ-YFP. Most liposomes that had Z rings before the bleach showed no recovery of fluorescent Z rings, even though HccA fluorescence recovered rapidly (Fig. 1.3). Fluorescent FtsZ-YFP-mts could be seen coating the outside of the liposome but apparently did not enter the liposome, since no fluorescent Z rings were seen. We did, however, find some liposomes that recovered fluorescent Z rings after the bleach (Fig. 1.4). Figure 1.4C shows an intriguing example where HccA fluorescence invades through a hole at the top side of the tubular liposome (arrow in Fig. 1.4F). FtsZ-YFP-mts followed and fluorescent Z rings assembled as the nonbleached FtsZ-YFP-mts exchanged with the bleached (Fig. 1.4C and D), showing that occasional open liposomes can recover FtsZ from the outside and assemble Z rings. Recovery of Z rings with FtsZ-YFP-mts appeared to be less frequent than recovery of fluorescence of FtsZ-YFP. A potentially important difference is that FtsZ-YFPmts had GTP, so it should be mostly assembled into protofilaments 30 subunits long (Chen and Erickson, 2005). FtsZ-YFP was used without GTP so it should be much smaller protein monomers.

9. A Crude Flow Chamber to Exchange Buffer Outside Liposomes For various purposes one might want to change the buffer outside the liposomes, especially knowing that most of the liposomes are permeable to small molecules. The best production of tubular liposomes occurs with a small sample volume, which generates an optimal shear when the coverslip is applied. A 5 ml sample spread over a 4.8 cm2 coverslip gives a liquid layer 10 mm thick. If we try to perfuse buffer into this thin layer the liposomes are subject to high shear that results in severe elongation, loss of Z rings, and loss of liposomes. To minimize these effects, we first apply a 50 ml drop of perfusion buffer to one edge of the coverslip (Fig. 1.5B). We can then drain the liquid by applying a Kimwipe at the opposite edge, while applying a pipette tip or thin rod to prevent the coverslip from slipping. The arrangement shown in Fig. 1.5D, with a thin strip of Kimwipe applied to the coverslip and a large piece farther away, provides an optimal slow flow.

12

Masaki Osawa and Harold P. Erickson

B 1 min

C 12 min

D 45 min

E pre

F 20 sec

G 12 min

F 45 min

HccA

FtsZ-YFP

A pre

20 mm

Figure 1.3 A liposome containing multiple Z rings (arrowheads) before bleach shows recovery of HccA fluorescence both outside and inside. FtsZ-YFP-mts fluorescence is recovered on the outside surface of the liposome, but no fluorescent Z rings recover inside. The exposure of fluorescence images is kept at the same level from prebleach to postbleach. (D) We did not detect any fluorescent Z rings even when we optimized the contrast in this specific panel.

We tested the system by adding FtsZ-YFP-mts without GTP to multilamellar liposomes. Cylindrical liposomes were found with the protein inside but Z rings could not form without GTP. We then perfused HMKCG containing 1 mM GTP. As the GTP entered the liposomes, Z rings assembled (Fig. 1.5E). An additional advantage of this system is that the fluorescent protein outside the liposome is washed away during the perfusion. This removes the fluorescent protein in solution and binding to the outside surface of liposomes and enhances the contrast of the Z rings inside. Because there is no FtsZ outside to hydrolyze the GTP, the Z rings can last a long time.

13

FtsZ in Tubular Liposomes

A

B

C 1 min

D 12 min

45 min

FtsZ-YFP

pre

E

F

G 20 sec

H 12 min

45 min

HccA

pre

20 mm

Figure 1.4 A liposome with Z rings inside shows permeability to both FtsZ-YFP-mts and to HccA. (A) Arrowheads indicate Z rings prebleach. (B) YFP fluorescence is completely eliminated on the outside and inside of the liposome at 1 min after bleach. (C) About 12 min later YFP fluorescence has partially recovered on the outside, and fluorescent Z rings have formed on the inside, showing that unbleached FtsZ-YFP-mts has entered the leaky liposome. (D) At 45 min fluorescent Z rings have spread more toward the bottom. (E, F) About 20 s after bleaching, HccA has begun to enter the liposome, apparently through a hole at the top (note the zone of brighter fluorescence in the top 1/5 of the liposome). (G, H) At 12 min HccA fluorescence has fully recovered inside compared to the prebleached image in (E). Note that the FtsZ rings recover only at the top, near the hole, at 12 min. At 45 min they have spread through the top half of the liposome.

14

Masaki Osawa and Harold P. Erickson

A

E

Pipette

B

Inserted solution lifts cover glass up C

D

Thin rod

Kimwipe

5 mm

Figure 1.5 Changing buffer by flow/perfusion. (A) The normal preparation has a very thin layer of liquid between the slide and coverslip. (B) When a drop of the new buffer is placed on the side, the coverslip rises and is able to move back and forth. (C) To adsorb the liquid, the slide is tilted and a small piece of Kimwipe is placed on the edge opposite the drop. To prevent the coverslip from moving we used a thin rod or a pipette. (D) To minimize shear we slow the draining by placing a narrow piece of Kimwipe touching the coverslip and a larger piece farther away. (E) As a demonstration of buffer exchange, we prepared liposomes with FtsZ-YFP-mts but no GTP. FtsZ was trapped inside some tubular liposomes but did not form Z rings. We then flowed through a buffer containing GTP but no FtsZ. The GTP entered the liposome and initiated Z-ring assembly. This liposome was apparently leaky to GTP but not to protein, since the FtsZ did not leak out.

10. Factors Affecting Z-ring Formation in Liposomes During months of experimenting with the liposome system, we have made a number of observations that seem to be important for Z-ring formation. 

The shape of the liposome is very important, the cylindrical geometry being optimal for Z-ring assembly. We have rarely seen Z rings inside spherical liposomes. When these have internalized FtsZ-YFP-mts, the protein localizes to small patches or forms arcs that move unstably

15

FtsZ in Tubular Liposomes

(Fig. 1.6B). One possible interpretation is that when a Z ring forms inside a spherical liposome, the contractile force causes it to collapse to a small patch. Another interpretation is that FtsZ filaments cannot determine the direction to form stable Z rings. We have sometimes observed this situation in which a large round liposome had radially formed FtsZ filaments inside (Fig. 1.6B, liposome on right). With a cylindrical geometry, a contractile force will cause it to form a circle in a plane perpendicular to the axis.  The diameter of the liposome is important. Z-ring assembly is optimal in tubular liposomes less than 2 mm in diameter (bacteria are typically 1 mm in diameter). In larger diameter tubular liposomes, some Z rings are

A

B

*2

*2

*2 *2 *3 *1

*2

10 mm

*1

Figure 1.6 Behavior of FtsZ-YFP-mts in large multilamellar liposomes. (A) In a large tubular liposome (3 mm inside diameter), FtsZ-YFP-mts formed some normal Z rings indicated by two spots of equal intensity directly across from each other (labeled *1). However, many spirals were also formed, indicated by dots spaced irregularly (*2). Note the bright ring at (*3), which appears to have peeled away the inner layer of the multilamellar wall (DIC image on right). Figure 1.6A is reprinted from Osawa et al. (2008) with permission of the publisher. (B) Several incomplete Z rings, spirals or arc structures, (arrowheads) are formed in a large spherical liposome.

16

Masaki Osawa and Harold P. Erickson

formed, but many structures appear to be helices or spirals rather than closed rings (Fig. 1.6A).  A mixture of neutral (PC) and negatively charged (DOPG) lipids is important. We have obtained Z rings with 2–40% DOPG, but typically use 1:4 DOPG:PC. 100% PC or DOPG, or a 2:3 ratio of DOPG:PC failed to support Z-ring assembly. This is probably because the amphipathic helix has positively charged amino acids on the hydrophilic side, and these need some negatively charged phospholipids in the bilayer to bind to. Interestingly, tubular liposomes produced by E. coli polar lipid (Avanti) did not support Z rings. This particular batch of E. coli lipid that we used may have contained more than 40% of charged lipid.  We have made several attempts to assemble Z rings in unilamellar liposomes, without success. As mentioned above, the spherical shape of unilamellar liposomes may not support Z-ring assembly. Another factor may be the rigidity of the wall. The rigidity of the multilamellar wall may slow constriction and stabilize the Z rings.

11. Utility of the Liposomes Beyond FtsZ The liposome system that we have developed may have uses beyond the study of FtsZ. The tubular liposomes are similar in size to a bacterium, and can range up to the size of a yeast cell. Similar to our reconstitution of Z rings, they may eventually be useful in reconstituting the cytokinetic apparatus of yeast or larger animal cells. More generally, we have demonstrated that most of the liposomes are permeable or leaky to small molecules, while about half are leaky to larger proteins. They therefore constitute small femtoliter chambers in which one can enclose proteins and manipulate the small molecule environment.

REFERENCES Chen, Y., and Erickson, H. P. (2005). Rapid in vitro assembly dynamics and subunit turnover of FtsZ demonstrated by fluorescence resonance energy transfer. J. Biol. Chem. 280, 22549–22554. Lu, C., Stricker, J., and Erickson, H. P. (1998). FtsZ from Escherichia coli, Azotobacter vinelandii, and Thermotoga maritima—Quantitation, GTP hydrolysis, and assembly. Cell Motil. Cytoskel. 40, 71–86. Miroux, B., and Walker, J. E. (1996). Over-production of proteins in Escherichia coli: Mutant hosts that allow synthesis of some membrane proteins and globular proteins at high levels. J. Mol. Biol. 260, 289–298. Nagai, T., Ibata, K., Park, E. S., Kubota, M., Mikoshiba, K., and Miyawaki, A. (2002). A variant of yellow fluorescent protein with fast and efficient maturation for cellbiological applications. Nat. Biotechnol. 20, 87–90.

FtsZ in Tubular Liposomes

17

Noireaux, V., and Libchaber, A. (2004). A vesicle bioreactor as a step toward an artificial cell assembly. Proc. Natl. Acad. Sci. USA 101, 17669–17674. Osawa, M., and Erickson, H. P. (2005). Probing the domain structure of FtsZ by random truncation and insertion of GFP. Microbiology 151, 4033–4043. Osawa, M., Anderson, D. E., and Erickson, H. P. (2008). Reconstitution of contractile FtsZ rings in liposomes. Science 320, 792–794. Pautot, S., Frisken, B. J., and Weitz, D. A. (2003). Engineering asymmetric vesicles. Proc. Natl. Acad. Sci. USA 100, 10718–10721. Pichoff, S., and Lutkenhaus, J. (2005). Tethering the Z ring to the membrane through a conserved membrane targeting sequence in FtsA. Mol. Microbiol. 55, 1722–1734. Szeto, T. H., Rowland, S. L., Habrukowich, C. L., and King, G. F. (2003). The MinD membrane targeting sequence is a transplantable lipid-binding helix. J. Biol. Chem. 278, 40050–40056.

This page intentionally left blank

C H A P T E R

T W O

Detection and Analysis of Protein Synthesis and RNA Replication in Giant Liposomes Takeshi Sunami,† Hiroshi Kita,† Kazufumi Hosoda,* Tomoaki Matsuura,* Hiroaki Suzuki,* and Tetsuya Yomo*,†,‡

Contents 1. Introduction 2. Methods 2.1. Liposome preparation 2.2. Internal protein synthesis reactions followed by the reaction catalyzed by the synthesized proteins 2.3. Detecting reactions in liposomes by fluorescence-activated cell sorting 3. Analysis of the FACS Data 4. Conclusions Acknowledgments References

20 21 21 22 25 27 28 29 29

Abstract In living cells, biochemical reaction systems are enclosed in small lipidic compartments. To experimentally simulate various biochemical reactions occurring in extant cells, giant liposomes are used to reconstruct an artificial model cell. We present methods for conducting a protein synthesis reaction, followed by the reaction catalyzed by the synthesized proteins inside liposomes, and for measurement of the in liposome reaction using a fluorescence-activated cell sorter (FACS). These techniques enable us to perform detailed analysis of the biochemical reactions occurring in the microcompartments, and have the potential to reveal the role of compartmentalization in cellular systems.

* {

{

Department of Bioinformatics Engineering, Graduate School of Information Science and Technology, Osaka University, Suita, Osaka, Japan Exploratory Research for Advanced Technology (ERATO), Japan Science and Technology Agency ( JST), Chiyoda-ku, Tokyo, Japan Graduate School of Frontier Biosciences, Osaka University, Suita, Osaka, Japan

Methods in Enzymology, Volume 464 ISSN 0076-6879, DOI: 10.1016/S0076-6879(09)64002-7

#

2009 Elsevier Inc. All rights reserved.

19

20

Takeshi Sunami et al.

1. Introduction Liposomes are small vesicles composed of a lipid bilayer membrane, which are widely used in both basic and applied research in life sciences (Luisi and Walde, 2000; Torchilin and Weissig, 2003). Especially, giant liposomes (diameter > 1 mm) are of particular interest as a cell model and a microreactor containing biochemical reaction systems in a cell-sized volume (Hanczyc et al., 2003; Kaneko et al., 1998; Noireaux and Libchaber, 2004; Nomura et al., 2003; Tsumoto et al., 2001). Reconstruction of such a cell model, which we call here as an artificial cell, from defined biochemical components, with which the characteristics of the extant cells are expected to be elicited (Luisi et al., 2006; Szostak et al., 2001), is a challenging task for biochemists and chemical engineers. Researchers have been attempting to synthesize artificial cells aiming to identify various aspects of living systems. Attempts to synthesize artificial cells can contribute to a better understanding of the origin of life, as this understanding will provide a physically possible path that could have lead to primitive living cells. Furthermore, experimentally increasing the complexity of the artificial cell, starting from the simplest one, will provide an opportunity to simulate evolutionary processes to the development of more complex organisms, and eventually current organisms (Deamer, 2005; Forster and Church, 2006; Luisi et al., 2006; Szostak et al., 2001). While the experimental construction of artificial cells has been proposed many times over the last several years (Deamer, 2005; Forster and Church, 2006; Luisi et al., 2006; Szostak et al., 2001), progress toward this goal has been proceeding in discrete steps, with researchers assembling elements partially fulfilling the properties of a living system. For example, it was shown to be possible to generate artificial lipid vesicles (liposomes) of the same size as small bacteria from amphiphilic molecules (Bangham and Horne, 1964). Artificial vesicles were also shown to be capable of autocatalytic growth, and to even be able to undergo repeated cycles of growth and division (Hanczyc et al., 2003; Oberholzer et al., 1995; Takakura and Sugawara, 2004). Various types of biological reactions have been successfully performed within the environment provided by liposomes (Fischer et al., 2002; Murtas et al., 2007; Noireaux and Libchaber, 2004; Nomura et al., 2003; Tsumoto et al., 2001; Walde and Ichikawa, 2001; Yu et al., 2001). These studies represent significant steps toward the assembly of an artificial cell. What is common among most of these studies is the usage of giant liposomes (diameter > 1 mm) that are similar in size to extant living cells, which provides an environment that is biologically relevant. Researchers including us have been attempting to encapsulate various biochemical reactions, including the protein synthesis reaction, in giant

Biochemical Reactions in Giant Liposomes

21

liposomes (Fischer et al., 2002; Murtas et al., 2007; Noireaux and Libchaber, 2004; Nomura et al., 2003; Tsumoto et al., 2001; Walde and Ichikawa, 2001; Yu et al., 2001). For the protein synthesis reaction, the cell-free protein synthesis system consisting of large number of components are used as the material for encapsulation. When encapsulating this system in liposomes, it is necessary that the all components required for the reaction are being encapsulated. If the concentrations of the components are below nM, however, and the size of liposomes is below mm, it is possible that some of the components will be missing. It is often the case that components involved in the protein synthesis reaction are at sub-nM concentrations (Shimizu et al., 2001), and therefore, for the protein synthesis reaction to occur in liposomes, giant liposomes have to be used. Here, we describe the strategy for preparing giant liposomes used for encapsulating the b-glucuronidase synthesis reaction and the RNA replication by self-encoded replicase, both of which utilize the cell-free protein synthesis system. We also describe the methods to detect the reaction in individual liposomes using a fluorescence-activated cell sorter (FACS), and to analyze the data. FACS allows the evaluation of large quantities of individual liposomes, at a rate of more than 20,000 liposomes/s. By the quantitative evaluation of the internal reaction within liposomes, properties of the reactions conducted in a small compartment (Kita et al., 2008), and an interesting property of the internal structure of liposomes can be identified (Hosoda et al., 2008).

2. Methods 2.1. Liposome preparation For encapsulation of the cell-free protein synthesis system, the choice of vesicle formation method is critical, since the concentration and encapsulation efficiency of the number of individual components in such a dense suspension may vary over a wide range (Gregoriadis et al., 1999; Monnard et al., 1997; Pupo et al., 2005; Walde and Ichikawa, 2001). Among the various vesicle formation methods proposed to date, the freeze-dried empty liposomes (FDEL) method (Kikuchi et al., 1999; Kirby and Gregoriadis, 1984; Murtas et al., 2007; Torchilin and Weissig, 2003) is one of the most promising candidates. The FDEL method has long been used as a means of vesicle formation with high entrapment efficiency even for biomolecules (Kirby and Gregoriadis, 1984; Torchilin and Weissig, 2003). In this method, liposomes in suspension are lyophilized (freeze-dried) to form empty lamellae of dry lipid film. Upon rehydration, a suspension containing biochemical molecules permeates into the lamellae, which swell to form material-containing vesicles. This method provides facile, stable, and reproducible formation of lipid

22

Takeshi Sunami et al.

vesicles containing complex and dense reaction mixtures. It is also important to note that this method is applicable for almost any combination of buffer and components. For the preparation of FDEL membranes, 1-palmitoyl-2-oleoyl-snglycero-3-phosphocholine (POPC; Avanti Polar Lipids, Alabaster, AL), cholesterol (Nacalai Tesque, Kyoto, Japan), and 1,2-distearoyl-sn-glycero3-phosphoethanolamine-n-[methoxy(polyethylene glycol)-5000] (DSPEPEG5000; NOF Corporation, Tokyo, Japan) dissolved in dichloromethane/diethyl ether (1:1, v/v) are mixed at a molar ratio of 58:39:3 and dried using a rotary evaporator, followed by complete removal of solvent in the vacuum chamber. Note that the lipid composition can be altered as desired, depending on the objective of the experiments, while some lipid compositions may strongly inhibit the internal reactions. The dried lipid film is hydrated with Milli-Q water (12 mM lipid), and this suspension is subjected to vortex mixing for 20 s and sonication for 5 s. After passing through a polycarbonate filter with a pore size of 0.4 mm (Nuclepore TrackEtch Membranes; Whatman, Maidstone, Kent, UK), the suspension is dispensed into small aliquots (40 mL each), and freeze-dried overnight (Labconco Corp., Kansas, MO). After purging the tubes with Argon gas, the freeze-dried membranes are stored in a freezer. Figure 2.1A shows the scanning electron microscope image of the freeze-dried liposome membrane. This dried membrane has a highly porous structure in the micrometer scale. To encapsulate the reaction mixture in liposomes, the solution is simply injected over this dried membrane (Fig. 2.1B).

2.2. Internal protein synthesis reactions followed by the reaction catalyzed by the synthesized proteins In practice, cell-free protein synthesis systems have been used to produce functional proteins, such as green fluorescent protein (GFP) (Ishikawa et al., 2004; Sunami et al., 2006; Yu et al., 2001), T7 RNA polymerase (Ishikawa et al., 2004), Qb RNA replicase, b-galactosidase (Kita et al., 2008), and b-glucuronidase (Hosoda et al., 2008) in liposomes. Among these, we describe the synthesis of b-glucuronidase (Fig. 2.2A) (Hosoda et al., 2008) and Qb replicase (Fig. 2.2B) (Kita et al., 2008). Qb replicase has been used to construct a ‘‘self-encoding system,’’ which is described in detail below. 2.2.1. b-Glucuronidase synthesis A plasmid encoding b-glucuronidase (pET-uidA) (Hosoda et al., 2008) is mixed with the cell-free protein synthesis system (PURESYSTEM classic II (Post Genome, Tokyo, Japan); Shimizu et al., 2001), 50 mM 5-(pentafluorobenzoylamino) fluorescein di-b-D-glucuronide (PFB-FDGlcU; Invitrogen, Carlsbad, CA), and a red fluorescent protein (allophycocyanin, hereafter APC, Molecular Probes, USA) at a high concentration (500 nM final

A

B

Scale bar: 5 mm

Figure 2.1 Microscopic pictures of liposomes prepared by the FDEL method. (A) Scanning electron microscope (SEM) image of freezedried liposomes. Freeze-dried liposomes are observed in a SEM (Keyence, VE-9800, Japan) at 950 magnification with 1 kV acceleration voltage after evaporation of gold (a few hundred nanometers). (B) Microscopic images of liposomes encapsulating R-PE. Upper images: bright-field observation; lower images: fluorescence observation.

24

Takeshi Sunami et al.

A

B Rep(+)Gal(-) RNA b-sub PFB-FDGlcU

Fluorophore

b-glucuronidase

Replicase lacZ

mRNA DNA

b-galactosidase CMFDG Fluorophore

Figure 2.2 Schematic of the encapsulated reactions. (A) b-Glucuronidase synthesis. The mRNA is transcribed from the DNA template, and b-glucuronidase is translated from the mRNA using the cell-free protein synthesis system within the liposomes. Synthesized b-glucuronidase hydrolyzes the fluorogenic substrate (PFB-FDGlcU) to yield the green fluorophore PFB-fluorescein. Red fluorescent protein (APC) is added as a marker of the internal aqueous volume. (B) RNA replication by self-encoded replicase. Rep(þ)Gal( ) RNA encodes the Qb replicase b-subunit and an antisense sequence of the b-galactosidase gene (lacZ). Synthesized b-subunit assembles three host proteins of Escherichia coli: ribosomal protein S1, elongation factor Tu (EF-Tu), and Ts (EF-Ts) that are present in the cell-free system to generate active Qb replicase. Upon the replication of the sense strand RNA by self-encoded replicase, the sense sequence of lacZ appears, which then yield the b-galactosidase. Synthesized b-galactosidase then hydrolyzes the fluorogenic substrate (CMFDG) to yield the green fluorophore CM-fluorescein.

concentration). PFB-FDGlcU is a fluorogenic substrate of b-glucuronidase that turns into a fluorescent molecule (PFB-fluorescein) as a product of the enzyme reaction and emits green fluorescence. APC is used as a marker for the internal aqueous volume. The liposomes containing the reaction system are prepared by adding 10 mL of reaction mixture into an aliquot of freezedried liposomes (final lipid concentration 48 mM ). The liposome suspension is diluted 20-fold with the dilution buffer (reaction mixture without DNA, substrate, and APC). Protease (final concentration 1 mg/mL) is also included in the dilution buffer to suppress completely the enzyme reaction that may occur outside of the vesicles. All preparation procedures mentioned above are performed on ice. The reaction is then initiated by incubating at 37  C, and the suspension is time-sampled and subjected to FACS measurements. 2.2.2. RNA replication by self-encoded replicase In all living systems, the genome is replicated by proteins encoded within the genome itself. This universal reaction is essential for evolvability of the system. We previously constructed a simplified system termed a self-encoding system, where the genetic information is replicated by self-encoded replicase

Biochemical Reactions in Giant Liposomes

25

in liposomes (Kita et al., 2008). The self-encoding system is assembled using RNA, as shown in Fig. 2.2B. The Rep(þ)Gal( ) RNA encodes the sense strand of the b-subunit of Qb replicase, an RNA-dependent RNA polymerase responsible for replicating the RNA genome of coliphage Qb (Blumenthal and Carmichael, 1979), and an antisense sequence of the bgalactosidase gene (lacZ ). In addition, the RNA is designed to serve as a template for the replication by Qb replicase. When Rep(þ)Gal( ) RNA is added to the cell-free protein synthesis system, Qb replicase is first synthesized, which then replicates its own template to produce the antistrand RNA. The antistrand RNA also serves as a template of Qb replicase. As the RNA is replicated by its own encoding gene, we termed this as a self-encoding system. The occurrence of the replication reaction can be probed by the production of b-galactosidase, as it is encoded on the antistrand RNA. Synthesized b-galactosidase then hydrolyzes the fluorogenic substrate. In practice, Rep(þ)Gal( ) RNA is mixed with the cell-free protein synthesis system (PURESYSTEM, customized as described in Kita et al. (2008)), 100 mM fluorogenic substrate 5-chloromethylfluorescein di-b-Dgalactopyranoside (CMFDG; Invitrogen), and R-phycoerythrin (R-PE; Invitrogen) at high concentration (400 nM at a final concentration). CMFDG is a fluorogenic substrate of b-galactosidase that turns into a fluorescent molecule (CM-fluorescein) as a product of the enzyme reaction and emits green fluorescence. R-PE is used as a marker for internal aqueous volume. Note that both R-PE and APC can be used as an internal volume marker. Liposomes containing the reaction system are prepared and initiated essentially the same as the b-glucuronidase synthesis.

2.3. Detecting reactions in liposomes by fluorescence-activated cell sorting Fluorescence-activated cell sorting is used to measure various reactions in individual cells by detecting the emission intensities of multiple fluorescent markers upon laser irradiation. FACS can also be used to measure the internal reactions in liposomes (Kageyama et al., 2007; Taly et al., 2007). For example, green fluorescent products in liposomes can be quantified by measuring the green fluorescence intensity of individual liposomes, whereas the inner aqueous volume of each liposome can be quantified from the intensity of the red fluorescent marker protein encapsulated. As the number of green product molecules and the internal aqueous volume can be simultaneously measured, the concentration of the product molecule in individual liposomes can be estimated. The rate of measurement by FACS is more than 20,000 liposomes/s, significantly faster than that of microscopic observations. Typical results of FACS measurements are shown in Fig. 2.3. Note that FACS can be used not only to measure the internal reaction of individual liposomes but also to sort the liposomes with desired fluorescence

Internal aqueous volume (fL)

A

103

0 min

103

30 min

103

60 min

103

102

102

102

102

10

10

10

10

1 103

104

105

1 106 103

104

1 105 106 103 104 PFB-fluorescein (molecules)

105

1 106 103

90 min

104

105

106

B Internal aqueous volume (fL)

103

0 min

103

120 min

103

200 min

103

102

102

102

102

10

10

10

10

1 103

104

105

1 106 103

104

1 105 106 103 104 CM-fluorescein (molecules)

105

1 106 103

275 min

104

105

106

Figure 2.3 Time course of the reaction analyzed by FACS. Typical results of in liposome (A) b-glucuronidase synthesis and (B) RNA replication by self-encoded replicase. The product (horizontal) and internal aqueous volume (vertical) of each liposome are shown. Dots represent the data of individual liposomes. In all plots, 20,000 data points from a total of 100,000 obtained are shown for clarity. The reactions proceed only in a fraction of the liposomes. This is because not all liposomes carry the plasmid DNA or the reaction efficiency is very low (Hosoda et al., 2008; Kita et al., 2008).

Biochemical Reactions in Giant Liposomes

27

signals, and the sorted samples can be used for further analysis such as microscopic observations (e.g., Hosoda et al., 2008). For FACS measurements, the sampled liposome suspension is diluted by at least 10-fold with sheath solution (FACSFlow; Becton Dickinson, San Jose, CA) to reduce the measurement frequency to smaller than 20,000 events/s. Then, red and green fluorescence signals, reflecting the whole internal aqueous volume and the amount of product, respectively, are measured with a FACS (FACSAria; Becton Dickinson). Typically, 100,000 data are sufficient for each measurement. The time course of the internal reaction can also be measured by analyzing the liposomes at different time points. APC or R-PE, used as a measure for the internal aqueous volume, is excited with a HeNe laser (633 nm) (or semiconductor laser (488 nm) in the case of R-PE), and the emission is detected through a 660  10 nm band-pass filter (or 576  13 nm band-pass filter in the case of R-PE). The number of R-PE molecules is converted from the red fluorescence signal detected (FIR) using the linear relation between the intensity from fluorescent beads carrying a known amount of R-PE (QuantiBRITE PE Quantitation Kit; BD Biosciences Clontech, Palo Alto, CA). The number of APC molecules is converted from the red fluorescence signal detected (FIR) using the linear relation between the intensity from fluorescent beads carrying a known amount of RPE, and the linear relation between fluorescence intensity of R-PE and APC. The quantity of the whole volume (Vw) is calculated by dividing the number of APC or R-PE molecules by the concentration of encapsulated APC (500 nM) or R-PE (400 nM). The final equation used for the conversion is Vw ¼ 0.054 FIR, or Vw ¼ 0.094 FIR, with APC or R-PE, respectively. The number of product molecules is derived from the intensity of the green fluorescence signal (FIG). PFB-fluorescein and CM-fluorescein, the reaction product of b-glucuronidase and b-galactosidase, respectively, are excited with a 488-nm semiconductor laser and the emission is detected through a 530  15 nm band-pass filter. For conversion, the linear relation between the green fluorescence intensities obtained by FACS and the number of product molecules (PFB-fluorescein or CM-fluorescein) is obtained. This is obtained by determining the intensity of a known amount of fluorescent molecules, and the linear relation between the intensity of the fluorescent molecule and the product molecules. The final equation used for conversion of PFB-fluorescein and CM-fluorescein is NG ¼ 330 FIG and NG ¼ 920 FIG, respectively, where NG is the number of product molecules.

3. Analysis of the FACS Data The kinetics of the internal reaction can be obtained from the FACS data. Figure 2.4A shows the histogram (frequency distribution) of the product concentration of RNA replication by self-encoded replicase in

28

Takeshi Sunami et al.

0.1 Frequency

B

0.12 0.08 0.06

0 min 120 min 200 min 275 min

0.04 0.02 0 0.1

1 10 100 CM-fluorescein (mM )

CM-fluorescein (mM )

A

20 15 10

Internal volume (fL) 1.2-4.0: 4.0-13: 13-40: 40-130:

5 0

0 100

200 300 400 Time (min)

500

Figure 2.4 Analysis of the FACS data. (A) Histogram (frequency distribution) of the CM-fluorescein concentration in liposomes with an internal volume of 4–13 fL at different time points. This is obtained from the data shown in Fig. 2.3B. (B) Time course of the reaction in liposomes with different internal volumes.

liposomes with different internal aqueous volumes and at different time points. In the measurement data, liposomes in which reaction proceeded (reacted liposomes) can be distinguished from those remained non-reacted based on the intensity of green fluorescence. The median of the histogram of the reacted liposomes are employed as a representative product concentration in liposomes with a given aqueous volume. The median values are used to trace the representative behavior of the internal reaction. As an example, we describe the strategy to obtain the time course data from the data shown in Fig. 2.4A. First, median values of the product concentration at the end of reaction, where the population of reacted liposomes can be readily distinguished, are obtained for each liposome internal volume (1.2–4.0, 4.0–13, 13–40, 40–130 fL). The rank in the product concentration (rank 1 ¼ highest product concentration) of the liposomes exhibiting the median value is then obtained. To obtain the representative product concentration at intermediate time points, where the distributions of reacted and non-reacted liposomes overlap, the product concentrations of the liposomes of identical rank to those obtained above are subsequently obtained for each liposome internal volume at different reaction times. These give the time courses of the reaction within liposomes shown in Fig. 2.4B.

4. Conclusions The presented method for in liposome protein synthesis followed by the measurement by FACS allows us to assess the kinetics of reactions in liposomes. This method can thus be used to elucidate the reaction dynamics in a microcompartment. In all living cells, cytosolic biochemical reactions are enclosed in a lipidic compartment. The effect of the microcompartment

Biochemical Reactions in Giant Liposomes

29

properties on the internal reaction has not been fully investigated. It may be possible to elucidate this effect with the presented method, for example, by carrying out the reaction in liposomes with different lipid composition and/ or different size. Thus, by providing an environment mimicing that of living cells, and by investigating the reactions inside, it may be possible to reveal the role of compartmentalization in cellular systems.

ACKNOWLEDGMENTS This research was supported in part by ‘‘Special Coordination Funds for Promoting Science and Technology: Yuragi Project’’ and ‘‘Global COE (Centers of Excellence) Program’’ of the Ministry of Education, Culture, Sports, Science, and Technology, Japan.

REFERENCES Bangham, A. D., and Horne, R. W. (1964). Negative staining of phospholipids and their structural modification by surface-active agents as observed in the electron microscope. J. Mol. Biol. 8, 660–668. Blumenthal, T., and Carmichael, G. G. (1979). RNA replication: Function and structure of Qbeta-replicase. Annu. Rev. Biochem. 48, 525–548. Deamer, D. (2005). A giant step towards artificial life? Trends Biotechnol. 23, 336–338. Fischer, A., Franco, A., and Oberholzer, T. (2002). Giant vesicles as microreactors for enzymatic mRNA synthesis. ChemBioChem 3, 409–417. Forster, A. C., and Church, G. M. (2006). Towards synthesis of a minimal cell. Mol. Syst. Biol. 2, 45. Gregoriadis, G., McCormack, B., Obrenovic, M., Saffie, R., Zadi, B., and Perrie, Y. (1999). Vaccine entrapment in liposomes. Methods 19, 156–162. Hanczyc, M. M., Fujikawa, S. M., and Szostak, J. W. (2003). Experimental models of primitive cellular compartments: Encapsulation, growth, and division. Science 302, 618–622. Hosoda, K., Sunami, T., Kazuta, Y., Matsuura, T., Suzuki, H., and Yomo, T. (2008). Quantitative study of the structure of multilamellar giant liposomes as a container of protein synthesis reaction. Langmuir 24, 13540–13548. Ishikawa, K., Sato, K., Shima, Y., Urabe, I., and Yomo, T. (2004). Expression of a cascading genetic network within liposomes. FEBS Lett. 576, 387–390. Kageyama, Y., Toyota, T., Murata, S., and Sugawara, T. (2007). Study on structural changes in supramolecular assemblies composed of amphiphilic nicotinamide and its dihydronicotinamide derivative by flow cytometry. Soft Matter 3, 699–702. Kaneko, T., Itoh, T. J., and Hotani, H. (1998). Morphological transformation of liposomes caused by assembly of encapsulated tubulin and determination of shape by microtubuleassociated proteins (MAPs). J. Mol. Biol. 284, 1671–1681. Kikuchi, H., Suzuki, N., Ebihara, K., Morita, H., Ishii, Y., Kikuchi, A., Sugaya, S., Serikawa, T., and Tanaka, K. (1999). Gene delivery using liposome technology. J. Control. Release 62, 269–277. Kirby, C., and Gregoriadis, G. (1984). Dehydration-rehydration vesicles—A simple method for high-yield drug entrapment in liposomes. Biotechnology 2, 979–984.

30

Takeshi Sunami et al.

Kita, H., Matsuura, T., Sunami, T., Hosoda, K., Ichihashi, N., Tsukada, K., Urabe, I., and Yomo, T. (2008). Replication of genetic information with self-encoded replicase in liposomes. ChemBioChem 9, 2403–2410. Luisi, P. L., and Walde, P. (2000). Giant Vesicles. John Wiley & Sons Inc., New York. Luisi, P. L., Ferri, F., and Stano, P. (2006). Approaches to semi-synthetic minimal cells: A review. Naturwissenschaften 93, 1–13. Monnard, P. A., Oberholzer, T., and Luisi, P. (1997). Entrapment of nucleic acids in liposomes. Biochim. Biophys. Acta 1329, 39–50. Murtas, G., Kuruma, Y., Bianchini, P., Diaspro, A., and Luisi, P. L. (2007). Protein synthesis in liposomes with a minimal set of enzymes. Biochem. Biophys. Res. Commun. 363, 12–17. Noireaux, V., and Libchaber, A. (2004). A vesicle bioreactor as a step toward an artificial cell assembly. Proc. Natl. Acad. Sci. USA 101, 17669–17674. Nomura, S. M., Tsumoto, K., Hamada, T., Akiyoshi, K., Nakatani, Y., and Yoshikawa, K. (2003). Gene expression within cell-sized lipid vesicles. ChemBioChem 4, 1172–1175. Oberholzer, T., Wick, R., Luisi, P. L., and Biebricher, C. K. (1995). Enzymatic RNA replication in self-reproducing vesicles: An approach to a minimal cell. Biochem. Biophys. Res. Commun. 207, 250–257. Pupo, E., Padron, A., Santana, E., Sotolongo, J., Quintana, D., Duenas, S., Duarte, C., de la Rosa, M. C., and Hardy, E. (2005). Preparation of plasmid DNA-containing liposomes using a high-pressure homogenization-extrusion technique. J. Control. Release 104, 379–396. Shimizu, Y., Inoue, A., Tomari, Y., Suzuki, T., Yokogawa, T., Nishikawa, K., and Ueda, T. (2001). Cell-free translation reconstituted with purified components. Nat. Biotechnol. 19, 751–755. Sunami, T., Sato, K., Matsuura, T., Tsukada, K., Urabe, I., and Yomo, T. (2006). Femtoliter compartment in liposomes for in vitro selection of proteins. Anal. Biochem. 357, 128–136. Szostak, J. W., Bartel, D. P., and Luisi, P. L. (2001). Synthesizing life. Nature 409, 387–390. Takakura, K., and Sugawara, T. (2004). Membrane dynamics of a myelin-like giant multilamellar vesicle applicable to a self-reproducing system. Langmuir 20, 3832–3834. Taly, V., Kelly, B. T., and Griffiths, A. D. (2007). Droplets as microreactors for highthroughput biology. ChemBioChem 8, 263–272. Torchilin, V., and Weissig, V. (2003). Liposomes: A Practical Approach. 2nd ed. Oxford University Press, Oxford. Tsumoto, K., Nomura, S. M., Nakatani, Y., and Yoshikawa, K. (2001). Giant liposome as a biochemical reactor: Transcription of DNA and transportation by laser tweezers. Langmuir 17, 7225–7228. Walde, P., and Ichikawa, S. (2001). Enzymes inside lipid vesicles: Preparation, reactivity and applications. Biomol. Eng. 18, 143–177. Yu, W., Sato, K., Wakabayashi, M., Nakaishi, T., Ko-Mitamura, E. P., Shima, Y., Urabe, I., and Yomo, T. (2001). Synthesis of functional protein in liposome. J. Biosci. Bioeng. 92, 590–593.

C H A P T E R

T H R E E

Construction of Cell-Sized Liposomes Encapsulating Actin and Actin-Cross-linking Proteins Kingo Takiguchi,* Ayako Yamada,†,1 Makiko Negishi,† Makoto Honda,*,2 Yohko Tanaka-Takiguchi,* and Kenichi Yoshikawa†

Contents 1. Introduction 1.1. Actin-cross-linking proteins coencapsulated in cell-sized liposomes 1.2. Methodologies to encapsulate actin and actin-cross-linking proteins into giant liposomes 2. Experimental Section 2.1. Preparation of actin and actin-cross-linking proteins for encapsulation 2.2. Natural swelling method for coencapsulation of actin and fascin, a-actinin, filamin, or BBMI into liposomes 2.3. Spontaneous transfer method for coencapsulation of actin and HMM (ActoHMM) into liposomes 3. Morphogenesis of Giant Liposomes Encapsulating Actin and Its Cross-linking Proteins 3.1. G-Actin and its polymerization in liposomes (natural swelling method) 3.2. Actin and fascin (natural swelling method) 3.3. Actin and a-actinin (natural swelling method) 3.4. Actin and filamin (natural swelling method) 3.5. Actin and BBMI (natural swelling method) 3.6. F-Actin (spontaneous transfer method)

* { 1 2

32 33 34 37 37 37 38 42 42 43 43 43 45 45

Division of Biological Science, Graduate School of Science, Nagoya University, Furo-cho, Chikusa-ku, Nagoya, Japan Department of Physics, Graduate School of Science, Kyoto University, Sakyo-ku, Kyoto, Japan Current address: Department of Chemistry, Ecole Normale Superieure, Paris, France Current address: Stem Cell and Drug Discovery Institute, Shimogyo-ku, Kyoto, Japan

Methods in Enzymology, Volume 464 ISSN 0076-6879, DOI: 10.1016/S0076-6879(09)64003-9

#

2009 Elsevier Inc. All rights reserved.

31

32

Kingo Takiguchi et al.

3.7. ActoHMM (spontaneous transfer method) 3.8. Actin and S-1 (spontaneous transfer method) 3.9. Mechanism determining liposome morphology 4. Concluding Remarks Acknowledgments References

45 46 46 49 50 50

Abstract To shed light on the mechanism underlying the active morphogenesis of living cells in relation to the organization of internal cytoskeletal networks, the development of new methodologies to construct artificial cell models is crucial. Here, we describe the successful construction of cell-sized liposomes entrapping cytoskeletal proteins. We discuss experimental protocols to prepare giant liposomes encapsulating desired amounts of actin and cross-linking proteins including molecular motor proteins, such as fascin, a-actinin, filamin, myosin-I isolated from brush border (BBMI), and heavy meromyosin (HMM). Subfragment 1 (S-1) is also studied in comparison to HMM, where S-1 and HMM are singleheaded and double-headed derivatives of conventional myosin (myosin-II), respectively. In the absence of cross-linking proteins, actin filaments (F-actin) are distributed homogeneously without any order within the liposomes. In contrast, when actin is encapsulated together with an actin-cross-linking protein, mesh structures emerge that are similar to those in living motile cells. Optical microscopic observations on the active morphological changes of the liposomes are reported.

1. Introduction Living cells and their organelles are compartmentalized by biomembranes in a self-organized manner, and each has a specific shape depending on its function. Cellular morphologies are thought to be determined and maintained by cytoskeletal networks (Hotani et al., 2003; Rodriguez et al., 2003). F-actin is a major component of the cytoskeleton, and is involved in a variety of cellular functions. Such functions include the extension of microspikes from cells (Albrecht-Buehler and Lancaster, 1976), the movement of filopodia in neural growth cones (Mitchison and Cramer, 1996; Mitchison and Kirschner, 1988), the extension or retraction of pseudopods during amoeboid movement (Taylor and Condeelis, 1979), and the contraction of contractile rings during cell division (Schroeder, 1973). F-actins also provide mechanical support, including stress fibers (Byers and Fujikawa, 1982). It is therefore important to study the role of the actin cytoskeleton in relation to membrane morphogenesis. Along this line, artificial model systems using giant liposomes containing actin and its associating proteins have been developed (Ba¨rmann et al., 1992; Cortese et al., 1989; Fygenson

33

Construction of Actin-Encapsulating Liposomes

et al., 1997; Ha¨ckl et al., 1998; Limozin et al., 2003, 2005; Limozin and Sackmann, 2002; Miyata and Hotani, 1992; Miyata et al., 1999; Pontani et al., 2009). In this chapter, we use the term ‘‘cell-sized liposome’’ to indicate giant liposomes with sizes of several to several tens of micrometers, which constitute closed thin membranes of phospholipids in aqueous solution. Cell-sized liposomes are expected to serve as a useful model in the real world, allowing us to make real-time observations using optical microscopy (Bangham, 1995; Cortese et al., 1989; Hotani et al., 2003; Lasic, 1995; Lipowsky, 1991; Miyata et al., 1999).

1.1. Actin-cross-linking proteins coencapsulated in cell-sized liposomes F-actins usually function in vivo in bundles or networks, and they are organized by various actin-binding, especially actin-cross-linking, proteins. Accordingly, we studied the effects of five different actin-cross-linking proteins, fascin, a-actinin, filamin, BBMI, and HMM on the morphogenesis of liposomes caused by actin assembly, by coencapsulating actin and one of those proteins together in liposomes. Actin-cross-linking proteins are characterized by the manner of cross-linking, that is, by the distance and angular flexibility between adjacent cross-linked F-actin molecules, as illustrated in Fig. 3.1. Fascin is a 55 kDa globular protein and is responsible for the tight crosslinking of F-actin to form bundles (Cant et al., 1994; Edwards and Bryan, 1995; Edwards et al., 1995; Otto et al., 1980; Yamashiro-Matsumura and Matsumura, 1985). a-Actinin is a larger protein (about 100 kDa), and Proteins

Fascin

a-actinin

Filamin

BBMI

HMM

S-1

Domain structure

Crosslinking pattern

Figure 3.1 Model of F-actin bundles or networks formed by actin-cross-linking proteins. From left to right, cross-linked F-actins mediated with fascin, a-actinin, filamin, BBMI, or HMM are shown. The upper row shows the molecular structure of each cross-linking protein. The lower row illustrates the arrangement of F-actins crosslinked by each cross-linking protein. S-1 and S-1-associating F-actins are also shown (right). F-actin is depicted as a line, while fascin, a-actinin, filamin, BBMI, HMM, and S-1 are drawn as globular monomers, dumbbell-shaped dimers, V-shaped dimers, single-headed musical note-like structures, double-headed structures, and single-headed structures, respectively.

34

Kingo Takiguchi et al.

antiparallel side-by-side associated homodimers of a-actinin form F-actin bundles with much looser packing (Imamura et al., 1988; Meyer and Aebi, 1990; Mimura and Asano, 1987). Filamin is a protein of approximately 250 kDa with a flexible elongated structure, and end-to-end associated homodimers of filamin are able to mediate flexible and high-angle linking between two adjacent F-actins (Bretscher, 1991; Hartwig and Kwiatkowski, 1991; Hock et al., 1990). The dynamics and mechanical properties of actin networks caused by these actin-cross-linking proteins have been well studied (Collins et al., 1991; Grazi et al., 1994; Hou et al., 1990; Wachsstock et al., 1993, 1994; Xu et al., 1998; Yamashiro et al., 1998). However, the relationship between the actin network organization and the morphogenesis of biological membranes has not been clarified yet. In living cells, the cytoskeleton is thought to be anchored to the membrane through membrane proteins, which probably regulate their own specific membranous morphology, position, and duration. BBMI exhibits a unique property in comparison with the other actin-cross-linking proteins mentioned above (fascin, a-actinin, filamin, or HMM). BBMI consists of a single motor domain (head) and a tail domain, and since the tail domain can bind both F-actin and a lipid membrane, BBMI can play a role as a cross-linker between F-actin and the membrane, as well as between F-actins (Coluccio, 1997). HMM has been frequently studied as a representative of double-headed myosin motors, and it can form and transform bundles and gels of F-actin (Takiguchi, 1991; Tanaka-Takiguchi et al., 2004). On the other hand, S-1 has often been studied as a representative of simple single-headed myosin motors. S-1 has only one actin-binding motor domain and is unable to cross-link F-actins, thus S-1 often has been studied in comparison to HMM.

1.2. Methodologies to encapsulate actin and actin-cross-linking proteins into giant liposomes Through the methodology of natural swelling, which is a method to obtain liposomes by the hydration of dried lipid films, we have successfully prepared liposomes containing monomeric actin (G-actin) and an actincross-linking protein (such as fascin, a-actinin, filamin, or BBMI) in the presence of 30 mM KCl (Honda et al., 1999). As the encapsulated G-actin polymerizes into F-actins, they form bundles or gels depending on the type of coencapsulated actin-cross-linking protein, causing various morphological changes of liposomes. The differences in morphology among the transformed liposomes indicate that the actin-cross-linking proteins determine the liposome shape by organizing their specific actin networks. On the other hand, in living cells, and depending on the cell type and the intracellular location, actin is expressed up to about 300 mM and undertakes its functions with the cooperation of various myosin motor proteins under physiological salt conditions (several mM Mg2þ and several tens of

Construction of Actin-Encapsulating Liposomes

35

mM Kþ) ( Janson et al., 1991; Pollard et al., 2000). However, using the natural swelling method, it is difficult or almost impossible to entrap the desired amounts of F-actin and myosin simultaneously in giant liposomes at physiological salt conditions. This is mainly due to the difficulty in preparing giant liposomes in the presence of salt, especially divalent cations such as Mg2þ, and the difficulty in controlling the amount of macromolecules to be incorporated within the liposomes, because of the passive nature of the encapsulation process. To overcome those problems, we have developed the spontaneous transfer method through an oil/water interface, by adapting water-in-oil phospholipid-coated cell-sized-droplets (W/O droplets) as precursors of giant liposomes (Yamada et al., 2006). Recently, there have been several attempts to employ W/O droplets coated by phospholipids as a model for living cells instead of liposomes (Hase and Yoshikawa, 2006), or as a precursor of liposomes (Hamada et al., 2008; Noireaux and Libchaber, 2004; Pautot et al., 2003a,b; Pontani et al., 2009; Takiguchi et al., 2008; Yamada et al., 2006). The W/O droplets are prepared easily by emulsifying an aqueous solution, together with oil, containing phospholipids. The process enables us to encapsulate biomolecules at a controlled concentration under any salt strength, into cell-sized compartments covered with a monolayer of phospholipids. By taking advantage of this, the behavior of F-actins entrapped in a closed space (Claessens et al., 2006) or conformational changes of actin molecules interacting with a phospholipid membrane in the presence of Mg2þ (Hase and Yoshikawa, 2006) have been studied. In addition, we have succeeded in developing a system for the controlled fusion of two droplets containing a substrate and an enzyme, respectively (Hase et al., 2007). Consequently, methodologies to obtain liposomes by transferring phospholipid-coated W/O droplets from an oil phase to an aqueous phase through their interface using centrifugation (Noireaux and Libchaber, 2004; Pautot et al., 2003a,b; Pontani et al., 2009) or without an external force (the spontaneous transfer method; Hamada et al., 2008; Takiguchi et al., 2008; Yamada et al., 2006) have been developed. With those methodologies, one can obtain liposomes with sizes of 10–100 mm containing desired amounts of molecules. Furthermore, by taking advantage of microfluidic techniques, new methodologies to obtain monodisperse liposomes from monodisperse droplets have been actively studied (Funakoshi et al., 2007; Shum et al., 2008; Stachowiak et al., 2008). Table 3.1 compares the various methodologies used to prepare giant liposomes. Using one of the simplest methods, that is, the spontaneous transfer method, we succeeded in constructing giant liposomes encapsulating 200 mM F-actin in the presence of 5 mM MgCl2 and 50 mM KCl. Moreover, this method enabled us to succeed in encapsulating desired amounts of HMM or S-1 with F-actin into giant liposomes (Takiguchi et al., 2008).

Table 3.1 Methods for giant liposome preparation Natural swellinga

Preparation Required time for encapsulation Yield par observation field Required amount of the sample Control of inner/ outer condition Control of the size Asymmetric membrane Physiological salt concentration Problem of oil contamination a b c d e f g h

Easy Long Low

Electroformationb

Centrifugationc

Spontaneous transferd

Jettinge

Double emulsion f

Easy Comparably long High

Easy Short

Easy Short

Small

Difficult Comparably short Comparably low Large

Difficult Comparably short Comparably low Large

High

Comparably large Difficult

Large

Comparably low Small

Difficult

Easy

Easy

Easy

Easy

Difficult

Difficult

Difficult

Easy

Easy

Difficult

Difficult

Easyg

In principle easy Easyh

Difficult

Difficult

Comparably difficult No

Difficult

Easy

Easy

Easy

Easy

No

Yes

Yes

Yes

Yes

Robust method to prepare liposomes of almost any content, but yield is sometimes low. Multilamellar liposomes tend to be obtained (Bangham et al., 1965). High yield of unilamellar liposomes. A special technique is required for salt concentrations higher than 10 mM (Angelova and Dimitrov, 1986). The principle is similar to that of the spontaneous transfer method. Liposomes are free in an aqueous phase (Noireaux and Libchaber, 2004; Pautot et al., 2003b). Easy encapsulation. Liposomes are anchored to an oil/water interface (Yamada et al., 2006). Microfluidic technique is required. Highly monodisperse liposomes can be obtained (Funakoshi et al., 2007; Stachowiak et al., 2008). Microfluidic technique and use of volatile oil are required. Highly monodisperse liposomes can be obtained (Shum et al., 2008). Pautot et al. (2003a). Hamada et al. (2008).

Construction of Actin-Encapsulating Liposomes

37

2. Experimental Section 2.1. Preparation of actin and actin-cross-linking proteins for encapsulation G-Actin is obtained by incubating dried muscle (acetone powder) prepared from rabbit skeletal muscle with cold water followed by a polymerization– depolymerization cycle (Ebashi and Ebashi, 1965; Higashi-Fujime, 1983). Fascin is isolated from porcine brain extracts using the actin gel method, following DE-52 and hydroxylapatite (Bio-Rad) column chromatographies (Honda et al., 1999; Yamashiro-Matsumura and Matsumura, 1985). a-Actinin is purified from an extract of minced rabbit skeletal muscle that had been incubated overnight at room temperature using a repeating fractionation with ammonium sulfate (Ebashi and Ebashi, 1965). Filamin is isolated from chicken gizzard by a modification (Muguruma et al., 1990) of the method of Molony et al. (1987). BBMI is purified from an extract of homogenized brush borders isolated from chicken intestines using a series of column chromatographies, Sepharose CL-4B, CM sepharose, and monoQ HR5/5 (Collins et al., 1991; Jontes and Milligan, 1997). Skeletal muscle myosin (myosin-II) was obtained from rabbit skeletal muscles (white muscles of the back and hind legs). HMM and S-1 are obtained by digestion of skeletal muscle myosin with chymotrypsin in the presence of 1 mM MgCl2/ 0.5 M KCl and of 1 mM EDTA (ethylenediamine tetraacetic acid)/120 mM KCl, respectively (Tanaka-Takiguchi et al., 2004).

2.2. Natural swelling method for coencapsulation of actin and fascin, a-actinin, filamin, or BBMI into liposomes To obtain giant liposomes encapsulating actin and its cross-linking proteins, dried phospholipid films are hydrated with an aqueous solution containing the proteins (Honda et al., 1999; Hotani and Miyamoto, 1990; Kaneko et al., 1998; Miyata and Hotani, 1992). The phospholipids used are L-aphosphatidylcholine (PC), L-a-phosphatidylethanolamine (PE) and L-aphosphatidylglycerol (PG). Each phospholipid has been isolated from egg yolk and is purchased from Sigma (St. Louis, MO) or Avanti Polar Lipid (Alabaster, AL), and is dissolved in a chloroform/methanol solution (98:2, v/v) at a concentration of 10 mM. Methanol is added to protect lipid molecules from damage resulting from the moisturization of chloroform. The phospholipid solutions are mixed in a glass test tube (total of 200 mg lipid). Typically, 20 ml of the solution is poured into a glass tube (10 mm in diameter and 30 mm in height) which has been cleaned with the chloroform/methanol solution prior to use. The organic solvent of the phospholipid solution is evaporated under a flow of nitrogen gas, and the lipids are

38

Kingo Takiguchi et al.

further dried under vacuum for at least 90 min with a vacuum pump (G-20D, ULVAC, Chigasaki, Japan). Note that the phospholipid solutions are stocked at 4  C and are protected from light, and the dried lipid films are stocked in an evacuated dessiccator at room temperature and are also protected from light. To generate liposomes from the film, 40 ml buffer A (5 mM Tris–HCl, pH 8.0, 0.2 mM ATP, 5 mM dithiothreitol, 30 mM KCl) containing various concentrations of G-actin and an actin-cross-linking protein are added to the dried lipid films on ice. The lipid films immediately started swelling to form liposomes containing the proteins, and further swelling is facilitated by occasionally agitating the test tubes by hand. After 30 min, the liposome suspensions are diluted 30-fold with buffer A to prevent the polymerization of the G-actin outside the liposomes. The polymerization of G-actin to form F-actin inside the liposomes is then initiated by raising the temperature to 25  C under dark-field microscopy (BHF, Olympus, Tokyo, Japan) (Hotani, 1984). 2.2.1. Visualization of actin filament (F-actin) To visualize F-actin localization within liposomes, rhodamine-phalloidin (R-415, Molecular Probes, Eugene, OR) in methanol is added to the lipid solution prior to the film preparation. The final concentration of rhodamine-phalloidin in the specimens is about 20 nM. The rhodamine-phalloidin specifically binds F-actin, so that actin filaments become visible by fluorescence microscopy (Miyata and Hotani, 1992). 2.2.2. Visualization of liposome membranes Giant liposomes can be visualized without any labeling by several types of optical microscopes, such as phase contrast, differential interference contrast (DIC), and dark-field. To observe liposome membranes by fluorescence microscopy, a fluorescence-labeled phospholipid is mixed with other phospholipids (about 1:200, w/w) (Honda et al., 1999). NBD- or BODIPYconjugated phospholipids are frequently used as the fluorescence-labeled phospholipid. The fluorescence-labeled liposomes are then observed with fluorescence optics usually assembled into another microscope, that is, a phase contrast, DIC, or dark-field microscope.

2.3. Spontaneous transfer method for coencapsulation of actin and HMM (ActoHMM) into liposomes A schematic representation of the experimental setup is illustrated in Fig. 3.2 (Takiguchi et al., 2008; Yamada et al., 2006).

39

Construction of Actin-Encapsulating Liposomes

Egg PC in mineral oil

A

5 mm

PDMS Buffer

Cover glass

4 mm

Objective lens B

ActoHMM

Oil

Z X

Buffer

Figure 3.2 (A) Experimental setup for construction and observation of actoHMMencapsulating giant liposomes using the spontaneous transfer method. (B) Schematic representation of the transformation from a W/O droplet in the oil phase (left) to a liposome in the aqueous phase (right). F-actin is illustrated with a line, and HMM is illustrated as a double-headed structure. The interface of the droplet is depicted as a lipid monolayer, while a multilayered interface may be generated to some extent in these experiments. The figure is adopted from Takiguchi et al. (2008).

2.3.1. Observation chamber An observation chamber is prepared consisting of a cylindrical hole (ca. 4 mm in diameter) in a poly(dimethylsiloxane) (PDMS) sheet (ca. 5 mm thick) on a glass microscope slide (0.12–0.17 mm thick, Matsunami Glass, Osaka, Japan). To obtain the PDMS sheet, the base solution and a curing agent of Silpot 184 W/C (Dow Corning Toray, Tokyo, Japan) are mixed well at a ratio of 10:1 (w/w) and are poured into a plastic Petri dish to harden. Note that bubbles generated during the mixing disappear spontaneously in about 1 h at room temperature. One can speed the hardening process by baking but the temperature must not be higher than 85  C to avoid melting the Petri dish. It takes about 24 h at room temperature or 2 h at 85  C

40

Kingo Takiguchi et al.

to harden the PDMS. A piece of the PDMS sheet is cut off and a hole with ca. 4 mm diameter is made in it with a leather-craft punch. One can make several holes in each PDMS sheet to compare different conditions. Note that if the diameter is larger than 4 mm, it is difficult to maintain a proper oil/water interface in the chamber because the water phase tends to be a spherical shape due to surface tension, since the PDMS surface is rather hydrophobic. These PDMS sheets are repeatedly used after washing with ethanol or isopropanol. The PDMS surface is cleaned with scotch tape before being set on a glass cover slide, which is cleaned by baking at 500  C for 1 h. The PDMS and the glass surface must be tightly attached. 2.3.2. Oil containing phospholipids Chloroform or a chloroform/methanol solution of PC is prepared at a concentration of 10 mM and poured into a glass test tube. Typically, 25–50 ml of the solution is poured into a glass tube (7 mm in diameter and 30 mm in height) (Maruemu, Osaka, Japan), which has been cleaned with acetone prior to use. The organic solvent of the phospholipid solution is then evaporated under a nitrogen flow and is dried under vacuum for more than 15 min to produce a dry film at the bottom of the glass tube. Typically, 500 ml of mineral oil (Nacalai Tesque, Kyoto, Japan) is then added to the tube and is sealed tightly to avoid moisture prior to ultrasonication for 60 min at 50  C and vortex mixing. For sealing, Parafilm (Pechiney Plastic Packaging, Chicago, IL) alone cannot be used because it melts under these conditions. A combination of a cap from the tube and Parafilm, or Dura Seal (Diversified Biotech, Boston, MA) and Parafilm can be used. Note that the vortex mixing must be performed immediately after ultrasonication to avoid aggregation of the phospholipid. The final concentration of PC is 0.5 or 1.0 mM, and these prepared oils must be used in 1 day. 2.3.3. Oil/water interface in the chamber Firstly, 10 ml of an aqueous phase, that is, the outer solution of liposomes, is placed at the bottom of the observation chamber. Here, we use a solution consisting of 25 mM imidazol-HCl, 5 mM MgCl2, 50 mM KCl, and 10 mM DTT at pH 7.5 (buffer B) and up to 100 mM sucrose, or the concentration of buffer B is increased up to twice instead of adding sucrose to adjust the osmotic pressure to that of the inner solution. It is recommended to touch the glass surface with the pipette tip so that one can fill the chamber from the bottom. The glass surface should be totally wet without air bubbles and the air/water interface must be rather flat and horizontal. Ten microliters of the PC-containing oil is then gently added along the PDMS wall to obtain an oil/water interface covered with

Construction of Actin-Encapsulating Liposomes

41

phospholipid between the upper oil phase and the lower aqueous phase. The observation chamber is then set on the optical microscope. An objective lens with a magnification not higher than 20 times can be employed to focus on the formed interface. 2.3.4. Preparation of W/O droplets and transfer into liposomes One hundred microliters of the PC-containing oil is placed in an Eppendorf tube and 5 ml of the inner solution, that is, the target solution to be encapsulated in liposomes, is then added to the tube. As the inner solution, buffer B containing F-actin or actoHMM is prepared. Immediately after addition of the aqueous solution, the solution is emulsified by pipetting it up and down for about 30 s. Ten microliters of the obtained W/O droplet solution is then placed on the oil phase in the chamber. One can observe that, in the oil phase, the W/O droplets gradually fall down onto the oil/water interface because of gravity. Interestingly, the droplets then spontaneously move through the interface into the aqueous solution keeping their spherical shape. In our experimental conditions, the transferred droplets, or liposomes, are anchored onto the interface (Fig. 3.2). Although it is possible to transfer the liposomes further into the bulk aqueous phase by encapsulating higher density solution than the outer phase (Hamada et al., 2008), or using centrifugation (Pontani et al., 2009), we performed the observations on liposomes anchored to the interface since we could then monitor the full process of the transfer and the subsequent transformation of each specific liposome. As for the transformation of a droplet in oil into a liposome in water, we have already discussed the full details of the process (Yamada et al., 2006). Observations are performed using a Zeiss Axiovert 100 inverted microscope equipped with an LSM 510 module for confocal microscopy (Carl Zeiss, Jena, Germany). To observe liposome membranes by fluorescence microscopy, fluorescence-labeled phospholipids are mixed with other phospholipids (1:100–1:1000, mol/mol). 2.3.5. Visualization of F-actin Actin prepared as described above is polymerized in buffer C (2 mM Tris– HCl, pH 8.0, 30 mM KCl, and 0.2 mM ATP) and then used for the experiments. To visualize F-actin entrapped within liposomes, a mixture of rhodamine-phalloidin in methanol and buffer C, or alternatively rhodamine-phalloidin redissolved in buffer C after volatilizing the methanol, is added to the actin solution before emulsification in the lipid-containing oil or mixing with HMM. The molecular ratio of rhodamine-phalloidin to the actin monomer was approximately 1:40, which depends on the amount and thickness of the actin bundles formed.

42

Kingo Takiguchi et al.

3. Morphogenesis of Giant Liposomes Encapsulating Actin and Its Cross-linking Proteins 3.1. G-Actin and its polymerization in liposomes (natural swelling method) Liposomes made of neutral and acidic phospholipids assumed spherical or tubular shapes in buffer A, and their shapes always fluctuated in solution (Fig. 3.3A) until the encapsulated actin polymerized. Liposomes containing only actin-cross-linking protein or G-actin also assumed spherical or tubular shapes in buffer A and always fluctuated in shape. The line-like images of liposomes obtained by dark-field microscopy have a uniform brightness which is almost identical to that of single-layered membranes from A

B

C

Figure 3.3 Morphological changes of actin-containing liposomes caused by actin assembly without an actin-cross-linking protein. (A) Dark-field microscopic images of liposomes made of neutral (PC) and acidic (PG) phospholipids (1:1, mol/mol). (B) Transformed liposomes obtained by the polymerization of encapsulated G-actin into F-actin. Four disk-shaped and one flat spoon-shaped liposomes are shown (viewed from the top). The lipid composition was PE and PG (1:1, mol/mol), and the concentration of encapsulated actin was 100 mM. (C) A fluorescent micrograph that shows the F-actin localization within a disk-shaped liposome obtained by actin assembly (top view). The assembled F-actin was labeled with rhodamine-phalloidin. The lipid composition was PC and PG (1:1, mol/mol), and the concentration of the encapsulated G-actin was 50 mM. All scale bars represent 5 mm. Note that here we can distinguish a disk-shaped liposome from a spherical one by shifting the focus plane of the microscope, or by continuous observation of each free-tumbling liposome in a microscope specimen. The figure is adopted from Honda et al. (1999).

Construction of Actin-Encapsulating Liposomes

43

erythrocyte ghosts whose associated proteins have been removed by protease treatment. Polymerization of the encapsulated G-actin into F-actin was achieved by raising the temperature, and the subsequent morphological changes of liposomes were monitored. When the actin polymerized, the liposomes transformed into flat disk or spoon shapes (Fig. 3.3B) (Miyata and Hotani, 1992). The F-actin that had polymerized in liposomes spontaneously aligned along the periphery of the flattened liposomes to form bundles (Fig. 3.3C). These liposomes are fairly rigid, as judged by the low fluctuation in their shapes. Their membranes were quiescent, and bending motions were restrained.

3.2. Actin and fascin (natural swelling method) When G-actin was encapsulated into liposomes together with fascin and then polymerized, the F-actin bundled tightly. Nearly 60% of liposomes transformed into lemon shapes and straight rigid projections subsequently developed from the tip(s), and eventually they became elongated straight tube(s) (Fig. 3.4A). Fluorescence labeled F-actin was always detected in the long projecting membrane regions (Fig. 3.4A, left), indicating that the fascin-mediated actin bundles are responsible for the formation of the membrane projections. The frequency of transformed liposomes and the lengths of their projections increased at higher concentrations of encapsulated actin or fascin.

3.3. Actin and a-actinin (natural swelling method) Unlike fascin, a-actinin had only a weak effect on morphological changes of actin-containing liposomes. When actin polymerized, most liposomes coencapsulating a-actinin transformed into disk shapes similar to those generated by encapsulating and polymerizing G-actin in the absence of a linking protein. The polymerized actin bundled and then aligned along the periphery. Liposomes transformed into disk shapes possessing long projections only in rare cases (1 M) solution during

58

Anna V. Shnyrova and Joshua Zimmerberg

purification is critical for the disassembly of the interior of NDV from its envelope. However, experimental observations indicate that M protein interaction with NP protein is not totally disrupted during the purification procedure based only on salt concentration changes, and that the aggregation of M protein may be in part due to the formation of this M–NP complexes. Thus, addition of calcium in millimolar concentration during M protein purification allows for a better separation of M protein from the NP complex, presumably through disruption of some electrostatic interactions, therefore leading to highly purified M protein. The protein purity is estimated from the reducing SDS–PAGE by analysis of the areas under the intensity peaks correspondent to protein bands, as was described previously (Coorssen et al., 2002).

3. Evaluation of the Membrane Activity of M Protein Through its Interaction with Intermediate-Sized Unilamellar Liposomes GUVs provide an excellent tool for studying the membrane activity of a particular protein, as they allows for real time observation of shape transformations by bright-field or fluorescence microscopy. However, experiments on GUVs may end up frustrating if the binding of the protein to a particular lipid composition is not previously analyzed by quantitative methods. In this aspect, large (200–500 nm in diameter) and intermediatesized (100–200 nm in diameter) unilamellar liposomes (LUVs and IUVs, respectively) are suitable for the initial checkup of the protein–membrane interactions.

3.1. Remarks about IUV formation and characterization There are various methods for IUV preparation described elsewhere (a great reference source is New, 1990b). The choice of a particular method mainly depends on the desired lipid composition of the vesicles and the buffer used. For the compositions and conditions described below, a combination of gentle hydration method followed by extrusion through appropriate polycarbonate membrane gives reproducible results. However, we encourage the reader to explore a variety of methods for IUV preparation in order to find an appropriate one or a combination of several that suits the particular experimental case. The success in the IUV preparation relies on achieving both the desirable size and the unilamellarity of the vesicles. The first parameter is crucial in terms of the curvature as well as the lateral membrane tension of the vesicles. For example, the membrane binding of many proteins depends on

Modeling Viral Budding with Unilamellar Vesicles

59

membrane curvature (e.g., Peter et al., 2004; Vaccaro et al., 1993). Thus, if the vesicle population exhibits high size variability, results of the binding assays may be truly confusing. Hence, the size distribution of the IUV sample is to be measured routinely, for example, by dynamic light-scattering. The lamellarity of the vesicles is extremely important when quantitative results are obtained from a binding assay. Even if the phospholipid concentration is determined for a particular IUV sample, the amount of the lipid accessible to the proteins, that is, lipids in the outer monolayer of a vesicle, may be up to orders of magnitude lower than for a unilamellar vesicle. Multilamellarity also affects the amount of encapsulated material into the vesicle and the effective membrane rigidity. This may lead to potential artifacts in experiments, for example, in measurements of the proteininduced leakage of encapsulated material by fluorescence spectroscopy. Thus, the routine of checking the unilamellarity of the IUVs should be adopted as well, for example, by TNBS assay (New, 1990a). A direct measurement of the IUVs lamellarity is offered by the electron microscopy imaging of ammonium molibdate negatively stained samples, as described previously (Bugelski et al., 1990). The use of uranyl acetate (UA) stain with this kind of samples is highly discouraged, as UA interferes with lipid membranes, especially with the charged ones (Caffrey et al., 1987; Parsegian et al., 1981).

3.2. IUV preparation For the M protein experiments, IUVs are prepared by a combination of the gentle hydration method (Reeves and Dowben, 1969) with the extrusion procedure (Mayer et al., 1986). IUV are formed from mixtures of the lipid stocks in chloroform of dioleoyl phosphatidylcholine (DOPC), palmitoyl oleoyl phosphatidylcholine (POPC), dioleoyl phosphatidylethanolamine (DOPE), dioleoyl phophatidylglycerol (DOPG) and cholesterol (chol), doped with 0.2 molar percent of lipid fluorescence marker Lissamine Rhodamine B DOPE (Rh-DOPE). All lipids are from Avanti Polar Lipids (AL). IUVs used for the binding assay with M protein are prepared from the following lipid mixtures (in molar ratios): DOPC:DOPE:Chol:Rh-PE 60:29.8:10:0.2 (PE-chol neutral mixture); POPC:Chol:Rh-DOPE 69.8:30:0.2 (chol neutral mixture); POPC:Rh-PE 99.8:0.2 (neutral mixture); POPC:DOPG:Rh-PE 84.8:15:0.2 (charged mixture). The total lipid concentration in the final IUV sample for these mixtures is 1 g/L. For the assay of the M-protein-induced IUV leakage, the PE–chol neutral mixture is employed, with total lipid concentration of 0.5 g/L in the IUV sample. Each lipid mixture (initially in chloroform) is added to a round bottom of 5 or 10 mL flask, suitable for a rotary–evaporator system (such as VV Micro Evaporator (Heidolph Brinkmann LLC, IL) or equivalent). Chloroform and methanol (9:1, v/v) are added to the flask until a final volume of

60

Anna V. Shnyrova and Joshua Zimmerberg

0.5 mL is achieved. The flask if rapidly transferred to the rotary– evaporator, where the evaporation of the organic solvents is carried out with gentle warming of the flask (30 ºC or above the phase transition temperature(s) of the lipids in the mixture) and an intermediate flask rotation speed, in order to obtain a thin and uniform lipid film. Once the visible traces of the organic solvent have disappeared, the evaporation continues for another 15–30 min. Then, the flask is transferred to a highpressure lyophilizer at room temperature for 1 h to ensure the removal of residual solvents. The desired aqueous buffer is added to the flask and the lipid film is left to hydrate overnight at 37 ºC. Special care is taken to avoid shaking the flask during the hydration process. In the case of the IUV preparation for the binding assay, the aqueous buffer is 100 mM KCl, 20 mM HEPES, and 0.2 mM EDTA, pH 7.4 (buffer C). For leakage measurements, an aqueous fluorescent dye should be entrapped inside the IUVs. Thus, buffer D (12.5 mM ANTS, 45 mM DPX, 20 mM KCl, 0.2 mM EDTA, 20 mM HEPES, pH 7.4) contains 8-aminonaphthalene-1,3,6-trisulfonate (ANTS) quenched with p-xylenebis(piridinium bromide) (DPX) (Molecular Probes, Invitrogen Corp., CA) (Du¨zgu¨nes¸, 2003a). The quenching efficiency is proportional to the concentration of ANTS/DPX. The dyes are loaded in to IUV at high concentration ensuring effective quenching of ANTS fluorescence. The leakage measurements are based on the increase of the ANTS fluorescence upon dilution. A small (10-nm fluorescence-labeled polystyrene beads for bioimaging) and adopting the electroporation method for the good manufacturing practice (GMP)-based production of BNC-based nanomedicines would be very difficult. We recently found that the N-terminal half of the L protein possesses membrane fusogenic activity (Matsuzaki et al., unpublished data). BNCs spontaneously form an 150-nm rigid complex with LPs (BNC–LP conjugate) in which multiple BNCs are embedded on the surface of LPs (Fig. 8.2) ( Jung et al., 2008). This property of BNC allows incorporation of various therapeutic materials into BNC–LP conjugates as follows. First, various materials (even 40-kbp plasmid, 100-nm fluorescent polystyrene beads) are constantly incorporated into LPs by conventional methods. Second, LPs are covered with BNCs harboring tissue specificity and high infectivity by the fusogenic activity of BNC. Third, BNC–LP conjugates can deliver various materials incorporated specifically and efficiently into human liver-derived tumors in a mouse xenograft model

Liposome

Incorporation

DNA

siRNA BNC

Conjugation Protein

Chemical compounds

BNC-liposome conjugate

Figure 8.2

Preparation of BNC–LP conjugates (bars: 50 nm).

152

Takeshi Kasuya et al.

through intravenous injection ( Jung et al., 2008). Additionally, this method can be expanded to GMP-based mass production more readily than electroporation. We therefore designated the BNC–LP conjugate as a ‘‘secondgeneration BNC’’ (Kasuya and Kuroda, 2009). With respect to the attachment of BNC–LP conjugates to human liver cells, the N-terminal half of the pre-S region has as a specific receptor for the human liver, which was demonstrated in HBV and BNCs. The intracellular drug release from BNC–LP conjugates may be mediated by the cellular internalization activity (another fusogenic activity) of the C-terminal half of the pre-S region and S region (see Fig. 8.1), as demonstrated in HBV (Glebe and Urban, 2007). In fact, an engineered BNC lacking the N-terminal half of the pre-S region can form a BNC–LP conjugate and deliver incorporated materials to target cells (Kasuya et al., 2008c). BNC–LP conjugates therefore possess the advantages of LPs and viral vectors, namely the use of versatile materials (as with LP) and the HBV-derived infection and active targeting machinery. These results indicate that BNC–LP conjugates are more promising than BNC per se as in vivo pinpoint DDS carriers.

4. Retargeting of BNC–LP Conjugates Due to the narrow tropism of HBV (because of the function of the N-terminal half of the pre-S region), systemically injected BNCs accumulate specifically in human liver-derived tissues in vivo. For expanding the indications of BNC-based nanomedicines, it is important to establish the methodology for retargeting BNC from human liver to nonliver tissues by substitution of the pre-S region by other targeting molecules (e.g., cytokines, growth factors, receptors, antibodies, glycans, lectins, aptamers). First, the pre-S (3–77) region is replaced with EGF by genetic engineering. The EGF-displaying BNC lost the specificity to human liver cells and obtained new specificity to EGFR-overexpressing A431 cells in vitro (Yamada et al., 2003). This approach needs a time-consuming step for constructing the expression system and sometimes fails to achieve the high productivity of BNCs in yeast cells. Next, a large part (50–159) of the pre-S region is replaced with the Staphylococcus aureus protein A-derived IgG Fc-binding ZZ domain. Beyond expectation, the ZZ domain-displaying BNC (ZZ–BNC) is efficiently synthesized in yeast cells. After ZZ–BNC is mixed with anti-EGFR IgG, the mixture is injected intracranially to a glioma-transplanted mouse orthograft model. The anti-EGFR IgG-displaying BNC accumulated in the EGFR-overexpressing transplanted glioma in the mouse corpus striatum (Kurata et al., 2008; Tsutsui et al., 2007), indicating that the antibody-displaying ZZ–BNC is applicable for

Bio-Nanocapsule–Liposome Conjugate

153

in vivo use. In 2008, because the ZZ domain contains many Lys residues, ZZ–BNC was modified with N-hydroxysuccinimide (NHS)-biotin to display biotin molecules onto the BNC surface. The biotinylated ZZ– BNC can be used for displaying various biotinylated targeting molecules (see above) by using an avidin (e.g., streptavidin, neutravidin) as an adaptor. For instance, the Phaseolus vulgaris agglutinin-L4 isolectin (L4-PHA)-displaying BNC (PHA–BNC) has been shown to accumulate in vivo in highly metastatic malignant tumors which overexpress b1-6-branching N-acetylglucosamine (GlcNAc), a specific ligand of L4-PHA (Kasuya et al., 2008c). PHA–BNC–LP conjugate can deliver luciferase-expressing plasmid to the b1-6GlcNAc-overexpressing cells, showing that the lack of a large part of the pre-S region does not affect the formation of BNC–LP conjugates. We recently revealed that the fusogenic activity of BNC required for LP conjugation is delineated in the short sequence at the N-terminal of pre-S region, which remains in ZZ–BNC (Matsuzaki et al., unpublished data). These data strongly suggested that the engineered BNCs (ZZ–BNC, biotinylated ZZ–BNC) and their LP conjugates can be used for lesion-specific pinpoint DDS carriers. In this chapter, we describe methods to prepare BNC–LP conjugates containing DNA, and BNC–LP conjugates containing an anticancer drug (doxorubicin (DOX)), and the effects of these BNC–LP conjugates in in vitro and in vivo systems.

5. Overexpression of BNCs in S. cerevisiae BNC is produced in S. cerevisiae AH22R (a leu2 his4 can1 cirþ pho80) strain (Kobayashi et al., 1988). The yeast cells are transformed by the spheroplast method (Hinnen et al., 1978) with the YEp plasmid pGLDLIIP39-RcT (Kuroda et al., 1992), which encodes the N-terminal chicken lysozyme signal sequence-fused HBV envelope L protein (subtype adr) under the glyceraldehyde-3-phosphate dehydrogenase (GLD; also called TDH3) gene promoter. For production of ZZ–BNC, the DNA segment encoding the pre-S region (50–159) in the plasmid pGLDLIIP39-RcT is replaced with the DNA segment encoding the IgG Fc-binding ZZ domain derived from S. aureus protein A (Tsutsui et al., 2007). The LEU2þ yeast transformants are cultured in a synthetic selection medium 8S5N-P400 (Yamada et al., 2001) at 30  C for 7 days, harvested by centrifugation, and stored at 80  C. The amount of BNC in yeast cells is estimated to be 40% of total soluble proteins (Kuroda et al., 1992).

154

Takeshi Kasuya et al.

5.1. Purification of BNCs by ultracentrifugation (10 mg protein per lot) BNC and ZZ–BNC can be readily purified by ultracentrifugation owing to the overexpression of BNCs in yeast cells. The yeast cells (about 20 g wet weight) are disrupted with glass beads (0.5 mm in diameter, 175 ml) using a BEAD-BEATER (BioSpec Products, Bartlesville, OK, USA) in 160 ml of 0.1 M sodium phosphate buffer (pH 7.2) containing 7.5 M urea, 15 mM ethylenediaminetetraacetic acid (EDTA), 4 mM phenylmethylsulfonyl fluoride (PMSF), 0.1 mM 4-amidinophenyl-methanesulfonyl fluoride (APMSF), and 0.1% (v/v) Tween 80. The crude extract obtained by centrifugation at 34,780g at 4  C for 30 min (about 6 g of protein) is mixed with PEG 6000 solution (15%, w/v at final concentration) to precipitate BNCs. The precipitants (about 3 g of protein) are subjected to CsCl isopycnic ultracentrifugation (10–40%, w/v) using an SW28 rotor (Beckman Coulter, Inc., Fullerton, CA, USA) at 24,000 rpm at room temperature for 15 h. Fractions containing BNCs are determined by sandwich enzyme-linked immunosorbent assay (ELISA) for BNC using an IMx HBsAg assay system (Abbott Laboratories, Abbott Park, IL, USA). Positive fractions are subjected twice to sucrose density gradient ultracentrifugation (10–50%, w/v) using an SW28 rotor at 24,000 rpm at room temperature for 15 h. Fractions containing BNCs are concentrated by ultrafiltration using an Amicon Ultra 100,000 NMWL (Millipore, Billerica, MA, USA) and subjected to a Sephacryl S-500 HR (GE Healthcare, Waukesha, WI, USA) gel-filtration column equilibrated with phosphate-buffered saline (PBS) containing 1 mM EDTA. The protein concentration of BNCs is measured by a bicinchoninic acid (BCA) assay kit (Sigma-Aldrich, St Louis, MO, USA) using bovine serum albumin (BSA) as a calibration standard. Approximately 10 mg (as a protein) of BNC is obtained from recombinant yeast cells grown in 2 l of culture medium (Yamada et al., 2001).

5.2. Purification of BNCs using column chromatography The ultracentrifugation step is rate-limiting, time-consuming, and produces a low yield, so obtaining >10 mg of BNC per lot in 1 week is difficult. Based on the heat stability of BNCs (Yamada et al., 2001), a large amount of yeast crude extract can be processed immediately by heat treatment ( Jung et al., unpublished data). The crude extract of yeast cells is obtained by the method described earlier, and then dialyzed three times against PBS containing 1 mM EDTA at 4  C for 2 h to remove urea. The crude extract (about 6 g of protein) is dispensed to 40-ml plastic tubes, incubated at 70  C for 20 min, and then centrifuged at 34,780g at 4  C for 30 min to remove yeast-derived proteins. The supernatant (about 800 mg of protein) is subjected to a sulfate-cellulofine column (1.6  20 cm; Chisso Corp., Tokyo,

Bio-Nanocapsule–Liposome Conjugate

155

Japan) equilibrated with PBS containing 150 mM NaCl. After the stepwise elution of PBS containing 1 M NaCl, the fraction containing a BNC is concentrated by ultrafiltration using an Amicon Ultra 100,000 NMWL (about 120 mg of protein) and then subjected to a Sephacryl S-500 HR gel-filtration column (1.6  60 cm) equilibrated with PBS containing 1 mM EDTA. About 60 mg of the highly purified BNC is obtained from the yeast cells grown in 2 l of culture medium.

5.3. Purification of ZZ–BNC using column chromatography Porcine IgG is precipitated from porcine serum (Sigma-Aldrich) with (NH4)2SO4 (40% saturation), dissolved in PBS, and dialyzed against PBS at 4  C for 48 h. The dialyzed solution is mixed with two-times volume of 60 mM acetate buffer (pH 4.8) and then mixed with n-caprylic acid to achieve a final concentration of 6.8% (v/v). After incubation at room temperature for 30 min with gentle stirring, the supernatant containing purified IgG is obtained by brief centrifugation. Purified porcine IgG is conjugated to an NHS-activated Sepharose 4B Fast Flow (GE Healthcare) column according to manufacturer’s instructions. The IgG-conjugated Sepharose column is equilibrated with PBS. Preparation of the crude extract of yeast and heat treatment are the same as the BNC protocol. The supernatant is subjected to the porcine IgGconjugated Sepharose column (affinity chromatography) to capture ZZ– BNC specifically using the IgG–ZZ domain interaction. The column is washed extensively with 75 mM Tris–HCl (pH 7.2) containing 10 mM NaCl, and then ZZ–BNC is obtained by the stepwise elution of 10 mM Tris–HCl (pH 7.2) containing 3.5 M NaSCN, 500 mM NaCl, and 10 mM EDTA. Fractions containing ZZ–BNC are subjected to a Sephacryl S-500 HR gel-filtration column (1.6  60 cm) equilibrated with PBS containing 1 mM EDTA. About 30 mg of highly purified ZZ–BNC can be obtained from yeast cells grown in 2 l of culture medium.

6. Conjugation of BNCs with LPs The LPs used for BNC conjugation (Fig. 8.2) can be prepared by conventional methods such as solvent evaporation (Bangham and Horne, 1964), ethanol injection (Batzri and Korn, 1973), and reverse-phase evaporation (Szoka and Papahadjopoulos, 1978). The lipid composition of LPs should be optimized for the materials to be incorporated, for example, cationic lipids for the preparation of DNA- or siRNA-containing LPs (i.e., lipoplexes; Chapter 14 of volume 465, Du¨zgu¨nes¸ et al., 2002; Li and Huang, 2006), and anionic/neutral lipids for chemical compounds

156

Takeshi Kasuya et al.

containing LPs. To our knowledge, all LPs that have been employed have resulted in the formation of BNC–LP conjugates.

6.1. Example 1.1: Preparation of BNC–LP conjugates containing DNA (BNC–lipoplex conjugates) For preparation of a DNA–LP complex (lipoplex), we routinely use commercially available lyophilized cationic LPs containing O,O0 -ditetradecanoyl-N-(a-trimethylammonioacetyl)diethanolamine chloride (DC-6-14; Kikuchi et al., 1999) as the major cationic lipid, an essential component for the cellular uptake of DNA, for example, Coatsoame-EL-01-D (NOF, Tokyo, Japan) and LipoTrust series (Hokkaido System Science, Sapporo, Japan). In the case of Coatsome-EL-01-D, 1.5 mg of lyophilized LP (as lipids) is dissolved in 1 ml of 250 mg/ml luciferase (LUC) expression vector pGL3 (Promega, Madison, WI, USA) and incubated for 15 min to allow lipoplex formation. Aliquots of lipoplex (100 mg LP, 16.7 mg DNA) are mixed with freeze-dried BNC (100 mg as protein) and incubated at room temperature for 15 min. By changing the amount of LP used for 16.7 mg DNA, a series of lipoplex and BNC–lipoplex conjugates are prepared with an N/P ratio (molar ratio of nitrogen-atom content in cationic lipids to phosphorous-atom content in plasmid DNA) from 0.3 to 2.4. The z-averaged sizes and z-potentials of lipoplex and BNC–lipoplex conjugate are measured at 25  C using a Zetasizer Nano-ZS (Malvern Instruments Ltd., Worcestershire, UK). When the content of cationic lipids in lipoplex is increased, z-averaged sizes and z-potentials are apt to increase to >1 mm and >0 mV, respectively, in accordance with the increase in N/P ratio (Fig. 8.3A). BNCs are found to keep the sizes of BNC–lipoplex conjugates to 0 the vesicles flip by 90 . The sign of the chemical shift of the intraliposomal water

Department of Chemistry IFM and Molecular Imaging Center, University of Torino, Torino, Italy Methods in Enzymology, Volume 464 ISSN 0076-6879, DOI: 10.1016/S0076-6879(09)64010-6

#

2009 Published by Elsevier Inc.

193

194

Silvio Aime et al.

resonance is positive in the former case and negative in the latter, respectively. The liposome orientation can be switched by incorporating in the liposome bilayer suitable amphiphilic paramagnetic lanthanide(III) complexes. The sign of DwLIPO, and consequently the magnetic alignment, will correspond to the sign of the magnetic susceptibility anisotropy of the metal complex. The magnetic susceptibility anisotropy is dependent on both the electronic configuration of the lanthanide ion and the structural characteristics of the amphiphilic complex incorporated in the liposome membrane. The magnetic orientation of such vesicles is maintained in vivo, thus opening promising perspectives for the application of nonspherical liposomes in medical imaging.

1. Introduction Liposomes are bilayered nanovesicles extensively used in the biomedical field, primarily as delivery systems. In particular, paramagnetic liposomes have recently attracted much attention as diagnostic probes for magnetic resonance imaging (MRI), thanks to their ability to carry a high number of imaging reporter to the site of interest, thus improving the sensitivity of the imaging technique for molecular imaging applications (Delli Castelli et al., 2008a). Paramagnetic liposomes are very versatile MRI probes, because they can generate contrast by exploiting different mechanisms depending on either the characteristics of the loaded paramagnetic species or the orientation of the liposome in the magnetic field. The two more important MRI contrast mechanisms rely on the shortening of the longitudinal (T1) or the transverse (T2) relaxation times of water protons. Furthermore, a new class of contrast agents has been recently introduced whose contrast mechanism deals with a saturation transfer process mediated by chemical exchange (CEST, chemical exchange saturation transfer) (Woods et al., 2006; Zhou and van Zijl, 2006). The most intriguing property of CEST agents relies on the possibility of visualizing different probes simultaneously present in the same region of the MR image (Aime et al., 2005a; Mc Mahon et al., 2008; Terreno et al., 2008a). Conventional liposomes loaded with Gd(III)- or Mn(II)-complexes can act as both T1- or T2-agents, whereas those loaded with paramagnetic Ln (III) ions different from Gd(III) lack the ability to generate T1-contrast, but maintain the possibility to act as T2-agents (Terreno et al., 2008b) and, moreover, can exhibit a very efficient CEST-contrast (lipoCEST agents) (Aime et al., 2005b). A noticeable improvement in the possibility to visualize different lipoCEST agents in the same image can be achieved upon orienting the vesicles in the magnetic field (Delli Castelli et al., 2008a,b).

Magnetically Oriented Paramagnetic Liposomes

195

Lipid-based nanosystems, such as bicelles and phospholipidic nanotubes, have peculiar magnetic properties that enable them to be oriented within a magnetic field (Karp et al., 2006; Prosser et al., 2006). The driving-force of this phenomenon is the interaction between the anisotropy of the magnetic susceptibility tensor (Dw) of the particle and the applied magnetic field (Prosser and Shyianovskaya, 2001); orientation can only occur when the particle shape is not isotropic. Systems made of phospholipids orient within the field according to their negative Dw value, but this ‘‘natural’’ alignment can be modified upon varying the Dw sign. This task has been accomplished by incorporating in the bilayer diamagnetic amphiphiles with positive magnetic anisotropy (Park et al., 2008) or paramagnetic systems, mostly lanthanide-based metal complexes, having positive Dw values (Crowell and Macdonald, 2001; Marcotte et al., 2006; Prosser et al., 1999). Although it has been reported that spherical multilamellar liposomes can orient within high magnetic fields (Brumm et al., 1992), conventional unilamellar spherical liposomes do not display a preferential orientation with respect to a magnetic field. Nonspherical unilamellar liposomes can be prepared by shrinking spherical ones through osmotic stress (Boroske et al., 1981; Menager and Cabuil, 2002; Terreno et al., 2006). In fact, when challenged with hyperosmotic stress, liposomes readily lose water, and rearranged into nonspherical systems, for example, assuming discoidal, oblate, cigar-like shapes (Hirota, 2003). As observed for other phospholipid-based membranes, the spontaneous orientation of diamagnetic, nonspherical, liposomes is to lie with the long axis parallel to the magnetic field due to the negative Dw value of the phospholipids. This orientation can be switched by 90 upon incorporating in the liposomal membrane paramagnetic species that change the Dw sign. In this chapter, we address the experimental workup that leads to the osmotically shrunken paramagnetic liposomes and describe the NMR method for determining their orientation properties.

2. Paramagnetic Ln(III)-Based Shift Reagents The paramagnetic Ln(III)-based shift reagents used in this work are illustrated in Chart 10.1. The metal complexes can be classified in two groups, depending on their hydrophilic/hydrophobic ratio. Ln–HPDO3A complexes are hydrophilic species and are the paramagnetic payload that may be hosted in the inner aqueous cavity of liposomes. Such complexes can be prepared according to the procedure reported in the literature for the Gd(III) ion (Dischino et al., 1991).

-OOC

COON

N

Ln-HPDO3A

Ln N

-OOC

-OOC N

N N

OH

COO-

COO-

N

Ln-DOTA-1

Ln -OOC

N

-OOC

N CON

Ln-DTPA-2a

CONH

Ln

N

CONH N COOCOON NOC

N

Ln-DTPA-1

COO-

Ln N

COO-

COON

COO-OOC

O CONH

N N N Ln N

-OOC -OOC

COO-

N

Ln-DTPA-2b

O

Ln

N

CONH -OOC -OOC

O

O

O O

N N

N N

Ln-TRIAZINO-1

Chart 10.1 Lanthanide complexes used for the preparation of magnetically oriented unilamellar liposomes.

197

Magnetically Oriented Paramagnetic Liposomes

The latter group deals with the shift reagents that have an amphiphilic character as they contain two long (typically C18) aliphatic chains covalently attached to the metal chelate. These species are poorly soluble or insoluble in water, but can be readily incorporated in the liposome bilayer (see Section 4). These compounds differ either in the structure of the coordination cage (DOTA-like, DTPA-like, and a triazino-based ligand) or in the modality through which the aliphatic chains are attached to the chelating unit (the chains are bound to a single or to different nitrogen atoms of the ligand). Ln–DOTA-1 and Ln–DTPA-2a complexes are synthesized according to the literature (Anelli et al., 2001; Prosser et al., 1999, respectively). The ligand TRIAZINO-1 is synthesized according to the procedure illustrated in Scheme 10.1, whereas the Tm(III) complex is synthesized following the procedure reported by Anelli et al. (2001). Ln–DTPA-1 complexes can be obtained from Bracco Imaging S.p.A.

3. Preparation of Osmotically Shrunken Liposomes Nonspherical large unilamellar paramagnetic liposomes suspended in isotonic buffer are prepared by using the conventional thin lipid film method. A given amount (typically 20 mg) of the lipid material (DPPC: dipalmitoylphosphatidylcholine, DSPE-PEG2000: distearoylphosphoethanolamine-polyethyleneglycol-2000, and amphiphilic Ln(III) complexes, if Cl N N

Cl N

Cl

Cl N

(C18H37)2NH

N

(C18H37)2N

Acetone H2O Na2CO3

HN NH2 NH2NH2.H2O

N

N N

(C18H37)2N

N HN NH2

Cl BrCH2COOtBu K2CO3 CH3CN COOH

COOtBu

HN N N (C18H37)2N

HN N COOH

N COOH

N HN N

CF3COOH

N

COOtBu N

(C18H37)2N N

COOtBu HN N

COOH

COOtBu

Scheme 10.1 Synthetic procedure followed for the synthesis of the ligand TRIAZINO-1.

198

Silvio Aime et al.

required) is dissolved in chloroform and the organic solution is slowly evaporated to remove the solvent until a thin film was formed on the wall of a round-bottom flask. The solvent removal is completed by keeping the film containing flask at the vacuum line for a couple of hours. The dry film is then hydrated by adding a hypotonic aqueous solution (usually 1 ml, 40 mOsm) containing the neutral Ln–HPDO3A (Ln ¼ Gd, Dy, Er, or Tm) complex. The film hydration is carried out at a temperature above the gel-to-liquid phase transition of the main phospholipid component (ca. 55  C for DPPC-based formulations) to promote the formation of a heterogeneous suspension of multilamellar vesicles entrapping the solution of the hydrophilic shift reagent. The unilamellarity is achieved by extruding the suspension several times (at least six) through polycarbonate filters with a pore diameter of 200 nm (Lipex extruder, Northern Lipids Inc., Canada) always keeping the temperature at 55  C via circulating water through the water jacket around the extruder. This workup leads to spherical liposomes, entrapping the shift reagent in their inner aqueous cavity, suspended in the hydration solution that is iso-osmolar with respect to the liposome cavity. The osmotic shrinkage of the vesicles occurs during the purification of the liposomes from the nonencapsulated shift reagent that is carried out by dialyzing the liposomes at 4  C against an isotonic (300 mOsm) HEPES/NaCl buffer (pH 7.4). Two dialysis cycles of 4 h each (volume ratio 500:1) are sufficient to remove the shift reagent. The vesicles are then subjected to a dynamic light-scattering investigation (Zetasizer NanoZS, Malvern, UK) in order to assess the mean hydrodynamic size and the polydispersity of the system.

4. NMR Characterization of Magnetically Oriented Nonspherical Liposomes The paramagnetic contribution to the chemical shift of the water protons (DINTRALIPO) that share the inner core of the liposomes with a Ln(III)-based metal complex is the sum of two contributions: DINTRALIPO ¼ DHYP þ DBMS

ð10:1Þ

The DHYP term refers to the hyperfine contribution that is mediated by the chemical coordination of the water molecules to the Ln(III) ion. Quantitatively, when the exchange rate of the metal bound water is fast, this term is directly proportional to the chemical shift of the water protons at the metal site (DLn) weighted on the molar fraction of the intraliposomal water protons that are coordinated to the metal center:

199

Magnetically Oriented Paramagnetic Liposomes

q½SRŠ ð10:2Þ 111:2 where q indicates the number of water protons coordinated to the Ln(III) ion, [SR] is the intraliposomal concentration of the shift reagent, and 111.2 is the molar concentration of water protons in the aqueous core. In principle, DLn is the sum of a dipolar (also termed pseudocontact) and a contact term, although it has been demonstrated that the paramagnetic effect induced by Ln(III) ions on the coordinated water protons predominantly arises from the dipolar contribution. On this basis, DLn is proportional to the magnetic anisotropy of the Ln (III) complex DwLnL and to the geometric factor G that comprises the metal– proton distance r and the orientation of the metal–proton vector with respect to the magnetic axis of the metal complex (defined by the polar angles y and ’). Finally, DwLnL is proportional to the value of the Bleaney’s constant CJ (that characterizes each paramagnetic Ln(III) ion, see Table 10.1) and to the ligand field coefficient (A02 hr 2 i) of the metal complex (Babailov, 2008): DHYP ¼ DLn

DHYP  Dw LnL  G

where Dw LnL / CJ  A02 hr 2 i

and Gðr;y;’Þ ð10:3Þ

The second contribution to DINTRALIPO in Eq. (10.1) is represented by the so-called bulk magnetic susceptibility (BMS) shift term. The peculiarity of this contribution is its strong dependence on the shape and orientation within the static magnetic field of the compartment containing the paramagnetic species. This shift contribution is null when the paramagnetic complex is entrapped in a spherical region and this means that in

Table 10.1 Effective magnetic moments and Bleaney’s constants for paramagnetic Ln(III) ions Ln

eff

CJ

Pr Nd Eu Gd Tb Dy Ho Er Tm Yb

3.62 3.68 3.4–3.6 7.9 9.7 10.6 10.6 9.6 7.6 4.5

11 4.2 4.0 0 86 100 39 33 53 22

200

Silvio Aime et al.

conventional spherical liposomes encapsulating a paramagnetic shift reagent, DHYP is the only operative shift contribution for the intraliposomal water protons. An accurate quantitative description of the BMS contribution is only possible for highly symmetric compartment shapes and welldefined orientations (Chu et al., 1990). For instance, in the case of a paramagnetic SR contained in a cylinder parallel to the magnetic field, DBMS is given by the following equation (Corsi et al., 2001): 1557:23  ½SRŠ  m2eff  s ð10:4Þ T where meff is the effective magnetic moment of the Ln(III) ion (see Table 10.1), T is the absolute temperature, and s is a ‘‘shape/orientation’’ constant that in this case is equal to 1/3. If the same cylinder is flipped by 90 and it is now perpendicular to the static field, s changes value and sign ( 1/6). On this basis, it is possible to predict that an SR entrapped in a nonspherical compartments oriented in the static field will display a positive BMS shift when the long axis of the compartment is parallel to the field, and a negative BMS shift when it is oriented perpendicularly. DBMS ¼

5. Sample Experiments A nice experiment for describing the effect on the osmotic shrinkage on the chemical shift of the intraliposomal water protons in liposomes encapsulating a paramagnetic shift reagent is illustrated in Fig. 10.1. The experiment was carried out on a stealth liposome (DPPC/DSPE-PEG2000 75/5 mole ratio) entrapping an hypotonic solution (40 mM) of the neutral shift reagent Tm-HPDO3A and incorporating in the bilayer the amphiphilic complex Tm-DOTA-1 (20% in moles). Upon their formation, the liposomes are suspended in an iso-osmolar buffer (40 mOsm), where the vesicles are expected to be spherical. The NMR spectrum of the suspension displays a resonance at about 0.4 ppm downfield to the bulk water. This resonance can be assigned to the intraliposomal water protons that are poorly shifted owing to the low paramagnetic payload of the vesicle. In fact, for a spherical liposome the only contribution to DINTRALIPO is the dipolar one, which is proportional to the intraliposomal concentration of the SR. The intrinsic shifting efficiency (DLn) of Tm-HPDO3A is about 20 ppm/M (Aime et al., 2006), so a 40 mM solution is expected to induce a chemical shift of ca. 0.8 ppm. The lower observed shift is likely due to the presence of the amphiphilic SR that, at least for the portion pointing inward the cavity, can affect the DINTRALIPO value. The observation that the overall observed shift is lower than that expected for the hydrophilic SR

201

Magnetically Oriented Paramagnetic Liposomes

300 mOsm (isotonic)

220 mOsm

22.3

0.6

12.1

160 mOsm

8.2 110 mOsm 4 72 mOsm

0.4

40 mOsm (iso-osmolar) 40

30

20

10

0

−10

−20

Figure 10.1 1H NMR spectra (14 T, 25  C) of liposomes made of DPPC/Tm-DOTA1/DSPE-PEG2000 (65/30/5 mol%) encapsulating 40 mM Tm-HPDO3A and suspended in a buffered medium (pH 7.4) with increasing osmolarity. The numbers indicate the DINTRALIPO values.

is an indication that the sign of the hyperfine shift contribution for the amphiphilic SR is opposite with respect to the hydrophilic one. Upon adding sodium chloride to the liposome suspension the vesicle are exposed to an osmotic stress. As a result of this stress, the resonance of the intraliposomal water protons moves away from the bulk water signal reaching a DINTRALIPO value of 22.4 ppm when the suspension is isotonic (ca. 300 mOsm). The increase of the shift is accompanied by a signal broadening and a decrease in its integral as reported in Fig. 10.2. Both observations can be accounted for in terms of the osmotic shrinkage of the nanovesicles as the water leakage (decrease in the signal integral) increases the intravesicular concentration of paramagnets that, in turn, results in an increase of the shift and of the line broadening of the intraliposomal water resonance. The data reported in Fig. 10.2 nicely fit a simple monoexponential decay that allows the estimation of the water fraction released from the liposomes undergone to the osmotic stress. On this basis, the vesicles appear to lose ca. the 75% of their entrapped water during the shrinkage and this means that the intraliposomal concentration of the Tm

202

Silvio Aime et al.

0.8

Signal integral

0.6

0.4

0.2

0.0 0

100

200 300 400 Osmolarity (mOsm)

500

600

Figure 10.2 Integral values for the intraliposomal water protons NMR signal (calculated using as reference the residual HOD signal of the deuterated solvent) as a function of the osmolarity of the liposomes suspension reported in Fig. 10.1.

(III)-based SRs in the nonspherical vesicle is almost fourfold larger than the original one. Focusing the attention on Tm-HPDO3A, its final concentration in the vesicles should be around 160 mM that corresponds to a DHYP shift of 3.2 ppm. The corresponding BMS contribution is much higher, about 16 ppm, as calculated from Eq. (4), assuming that the liposomes are oriented parallel to the static magnetic field and behave as a cylinder. Concerning the amphiphilic SR, its contribution to DHYP should have a negative sign, but the BMS contribution will be positive and, as discussed for Tm-HPDO3A, it significantly exceeds the hyperfine one. Hence, in spite of the approximations of the calculations (especially regarding the s parameter in Eq. (4)), the magnitude of the observed shift of 22.4 ppm appears reasonable. Figure 10.3 displays a cryo-TEM image of a specimen of osmotically shrunken liposomes (Terreno et al., 2009). The suspension is characterized by the presence of ellipsoidal-like vesicles whose shape in the TEM image is dependent upon their orientation with respect to the electronic beam (i.e., perpendicular to the image). The average long axis size ranged from 120 to 160 nm, in good agreement with the hydrodynamic size determined by dynamic light-scattering measurements. As already anticipated in the introduction, nonspherical liposomes align in the static magnetic field with their long axis parallel to B0 in virtue of the negative magnetic anisotropy of the self-assembled diamagnetic phospholipids (Scheme 10.2). Thus, the BMS shift contribution from such vesicles should be always positive. Figure 10.4 reports about the effect of encapsulating paramagnetic complexes in the aqueous cavity. Liposomes loaded

203

Magnetically Oriented Paramagnetic Liposomes

Electronic beam B

A

B B

A

Carbon Ice layer Lipo-CEST 200 nm

Figure 10.3 Cryo-TEM image showing a 2D projection of osmotically shrunken liposomes encapsulating 40 mM of Tm-HPDO3A and incorporating 20 mol% of Tm-DOTA-1. (A) Lens-shaped vesicle with the short axis positioned in plane with the electron beam, and (B) lens-shaped vesicle positioned with the short axis perpendicular to the electron beam. The dashed circle is displayed to guide the eye. Shrunken liposomes with all different orientations are observed.

Conventional liposome

ΔINTRALIPO = DHYP

(ΔBMS= 0)

B0 Non spherical liposome (magnetically orientable) ΔINTRALIPO = Δ HYP + ΔBMS

ΔχLIPO < 0

ΔχLIPO > 0

ΔBMS > 0

ΔBMS < 0

Paramagnetic amphiphilic metal complex (ΔχLnL < 0) Paramagnetic hydrophilic metal complex Paramagnetic amphiphilic metal complex (ΔχLnL > 0)

Scheme 10.2 Basic correlations among the physicochemical variables involved in the magnetic orientation and chemical shift of intraliposomal water protons for paramagnetic liposomes.

204

Silvio Aime et al.

12

DINTRALIPO (ppm)

10 8 6 4 2 0 -2 -4 0.0

0.3 0.4 0.1 0.2 Osmolarity of the hydration solution

Figure 10.4 DINTRALIPO values, measured at 25  C, as a function of the osmolarity of the solution used for hydrating the thin lipidic film (DPPC/DSPE-PEG 95:5 molar ratio) and containing an hydrophilic shift reagent (squares: Dy-HPDO3A, circles: Tm-HPDO3A). Filled symbols refer to spherical or slightly deformed vesicles. Open symbols refer to nonspherical liposomes. The straight lines are reported only for guiding eyes.

with Tm-HPDO3A or Dy-HPDO3A at different concentrations were purified against isotonic buffer and their DINTRALIPO values were determined and plotted as a function of the osmolarity of the hydration solution containing the metal complexes (Terreno et al., 2007). At high osmolarity values the liposomes were spherical and the DINTRALIPO values were rather small (only DHYP contribution present) and of opposite sign, as expected for complexes with the same structure (i.e., similar ligand field coefficients), but containing Ln(III) ions with CJ constants of opposite sign (Tm > 0, Dy < 0, see Table 10.1). As the osmolarity was decreased, the liposomes were subjected to the osmotic shrinkage, and the paramagnetic effect on the shift became less pronounced (mainly due to the decreased [SR] in the vesicle), but below 0.1 Osm the shrinkage was more evident and the DINTRALIPO values became positive for both the liposome types. Note also the increased absolute shift values due to the higher efficiency of the BMS contribution. To change the orientation of the nonspherical liposomes within the static magnetic field it is necessary to change the sign of the magnetic anisotropy susceptibility of the liposomal membrane (DwLIPO) from negative to positive. This task can be pursued by incorporating amphiphilic paramagnetic complexes with positive magnetic anisotropy (DwLnL > 0). Due to the much higher magnetic susceptibility of a paramagnetic species with respect to a diamagnetic system, small amounts of paramagnetic compounds are sufficient to address this task. As recalled above, the magnetic anisotropy of a paramagnetic metal complex is dependent on the metal ion, through

Magnetically Oriented Paramagnetic Liposomes

205

the CJ constant, and on the crystal field coefficient, which is mainly correlated to the structure of the coordination cage. Complexes with positive CJ values like Eu(III) or Tm(III) are excellent systems for changing the magnetic orientation of phospholipid-based assemblies as it was also reported for bicelles. Our work has shown that the incorporation of Tm– DOTA-1 complex leads to positive DINTRALIPO values, that is, the shift is along the same direction as observed for nonspherical liposomes without any paramagnetic species incorporated. Further insights into this matter has been gained by preparing a liposome incorporating the Tm–DTPA-2 complex, that is, the same complex successfully used for changing the orientation of bicelles (Prosser et al., 1999). Interestingly, the DINTRALIPO value for this system was negative (Fig. 10.5A), thus confirming the ability of this metal chelate to switch the magnetic anisotropy susceptibility of the liposomal membrane from a negative to a positive value. This result prompted us to investigate the properties of other paramagnetic amphiphiles differing in the Ln(III) ion and in the conjugation scheme of the hydrophobic tails to the ligand skeleton. The results are summarized in Fig. 10.5. By considering metal complexes with the same Ln(III) ion, for example, Tm(III) (Fig. 10.5A), the DINTRALIPO values were negative (liposomes oriented with the long axis perpendicular to B0, Scheme 10.2) for the two DTPA-2 ligands in which the two hydrocarbon chains are linked to different nitrogen atoms of the ligand, and positive (liposomes oriented with the long axis parallel to B0, Scheme 10.2) for the three ligands with different coordination cage (DOTA-1, DTPA-1, and TRIAZINO-1), but with the two hydrophobic tails bound to same nitrogen atom. This finding confirms that the orientation of nonspherical liposomes is strongly affected by the structure of the incorporated paramagnetic amphiphile. On the other hand, for a given structure, the replacement of the Ln(III) ion with a metal with a CJ value of opposite sign (e.g., Tm(III) and Dy(III)) induces the change in the orientation of the vesicles as reported in Fig. 10.5B. A peculiar behavior is expected in the presence of an amphiphilic Gd(III) complex. Gd(III) ion has a meff value (7.9) similar to Tm(III), but the symmetric distribution of its unpaired electrons makes him unable to induce a dipolar shift (CJ ¼ 0). For this reason, the incorporation of a Gd(III) complex in the membrane does not affect the DwLIPO value and the nonspherical liposomes maintain the same orientation displayed by diamagnetic vesicles and the resulting DINTRALIPO shifts are invariantly positive (Fig. 10.5C). However, the relatively high meff value allows the exploitation of the BMS shift contribution when an hydrophilic Gd(III) complex is encapsulated in an osmotically shrunken liposomes (Aime et al., 2007). In other words, the differences in the DINTRALIPO values between nonspherical vesicles oriented in the same way and loaded with Tm(III) and Gd(III)-chelates of the same ligand are primarily due to the presence of the dipolar shift contribution for the former system.

A

B Tm-DTPA-2b Dy-DTPA-2a Tm-DTPA-2a Tm-DTPA-2a Tm-DTPA-1 Dy-DOTA-1

Tm-TRIAZINO-1

Tm-DOTA-1

Tm-DOTA-1

20

10 0 -10 -20 DINTRALIPO (ppm)

-30

30

20

10 0 -10 -20 -30 -40 -50 DINTRALIPO (ppm)

C Gd-DTPA-1

Gd-DTPA-2b

Gd-DTPA-2a

Gd-DOTA-1

0

5

10 15 DINTRALIPO (ppm)

20

Figure 10.5 (A) (Top): DINTRALIPO values, measured at 25  C, for Tm(III)-loaded liposomes differing in the amphiphilic paramagnetic complex incorporated in the membrane (DPPC/TmL/DSPE-PEG 75:20:5 molar ratio). All systems contain Tm-HPDO3A (40 mM in the hydration solution) in the inner aqueous cavity. (B) (Middle): DINTRALIPO values, measured at 25  C, for Tm(III)- and Dy(III)-loaded liposomes for two different ligands (membrane composition DPPC/LnL/DSPE-PEG 75:20:5 molar ratio). Tm-HPDO3A or Dy-HPDO3A

207

Magnetically Oriented Paramagnetic Liposomes

A relevant issue related to the potential MRI applications of magnetically oriented liposomes deals with the maintenance of their orientation in biological systems. Figure 10.6 illustrates the results obtained from two experiments in which the lipoCEST probes were injected subcutaneously (on the left) (Terreno et al., 2008b) or intramuscularly (on the right), respectively. In both cases the CEST contrast was detected upon irradiation at the same frequency offset previously determined in aqueous solution for lipoCEST probes used in this in vivo experiment. This is a convincing demonstration that the nanovesicles maintain their magnetic orientation in the biological environment and highlights the possibility of using lipoCEST-based probes for the visualization of different biological targets in the same anatomical region.

∆LIPO 7 ppm

7 ppm

0.55

−17 ppm

3.5 ppm

−17 ppm

0.6

∆LIPO −17 ppm

∆LIPO 3.5 ppm

∆LIPO −17 ppm

0.7 0.6

0.3 0.28 0.26

0.5 0.5

0.24

0.4

0.22 0.2

0.4 0.3

0.18

0.35

0.25

0.28 0.26 0.24 0.22

0.45

0.3

0.3

0.18 0.16

0.16

0.14

0.14

0.12

0.12

0.1

0.2 0.1

0.2

Figure 10.6 In vivo MR-CEST experiments in mice. Two different lipoCEST probes are subcutaneously (left) or intramuscularly (right) injected. At the top an MRI T2w anatomical image is acquired just after the injection (the syringes indicate the injection site). At the bottom, MR-CEST parametric maps, obtained after the saturation at the frequency offset characteristics of the administered probes, are superimposed to the morphological images.

(40 mM in the hydration solution) are present in the inner aqueous cavity. (C) (Bottom): DINTRALIPO values, measured at 25  C, for Gd(III)-loaded liposomes differing in the amphiphilic paramagnetic complex incorporated in the membrane (DPPC/GdL/DSPE-PEG 75:20:5 molar ratio). Gd-HPDO3A (40 mM in the hydration solution) is encapsulated in the aqueous cavity.

208

Silvio Aime et al.

6. Concluding Remarks In this chapter, we have shown how to achieve the transformation of spherical unilamellar liposomes into nonspherical ones. Upon their loading with paramagnetic Ln(III) complexes, such systems exhibit the ability to align themselves in a static magnetic field. The parallel or perpendicular orientation is determined by the attainment of the minimum of energy of the interaction between the magnetic field and the magnetic susceptibility anisotropy of the self-assembled amphiphilic components of the bilayer membrane. The peculiar magnetic properties of paramagnetic Ln(III)based complexes can be exploited for switching from one orientation to the other and the magnetic alignment adopted by a given nanoparticles can be easily determined by measuring the 1H NMR chemical shift of the intraliposomal water protons. Currently, the most important field of application for these systems deals with the development of highly sensitive MR imaging probes in the CEST modality. Moreover, it is expected that, in analogy with other phospholipid-based magnetically oriented self-assemblies, like bicelles and nanotubes, nonspherical liposomes with their long axis perpendicular to the magnetic field could be useful systems for the determination of protein structures. Finally, they may be useful for the NMR characterization of large biomolecules associated with the membrane bilayer.

ACKNOWLEDGMENTS Economic and scientific support from local (Regione Piemonte, Nano-IGT and C 130 projects) and national government (FIRB RBNE03PX83_006, FIRB RBIP06293N, and PRIN 2005039914 projects), EC-FP6 (DiMI: LSHB-CT-2005-512146, EMIL: LSHCCT-2004-503569, and MEDITRANS: NMP4-CT-2006-026668), and EC-FP7 projects (ENCITE: 201842), EC-COST D38 action is gratefully acknowledged. Bracco Imaging S.p.A. (CRM Colleretto Giacosa (TO), Italy) is acknowledged for the longstanding and very fruitful collaboration. Discussions with S. Langereis, H. Gruell, and coworkers at Philips (Eindhoven, NL) have been very useful for the design and development of the systems reported in this contribution.

REFERENCES Aime, S., Carrera, C., Delli Castelli, D., Geninatti Crich, S., and Terreno, E. (2005a). Tunable imaging of cells labeled with MRI PARACEST agents. Angew. Chemie Int. Ed. 44, 1813–1815. Aime, S., Delli Castelli, D., and Terreno, E. (2005b). Highly sensitive MRI chemical exchange saturation transfer agents using liposomes. Angew. Chemie Int. Ed. 44, 5513–5515.

Magnetically Oriented Paramagnetic Liposomes

209

Aime, S., Geninatti Crich, S., Gianolio, E., Giovenzana, G. B., Tei, L., and Terreno, E. (2006). High sensitivity lanthanide(III) based probes for MR-medical imaging. Coord. Chem. Rev. 250, 1562–1579. Aime, S., Delli Castelli, D., Lawson, D., and Terreno, E. (2007). Gd-loaded liposomes as T1, susceptibility, and CEST agents, all in one. J. Am. Chem. Soc. 129, 2430–2431. Anelli, P. L., Lattuada, L., Lorusso, V., Schneider, M., Tournier, H., and Uggeri, F. (2001). Mixed micelles containing lipophilic gadolinium complexes as MRA contrast agents. MAGMA 12, 114–120. Babailov, S. P. (2008). Lanthanide paramagnetic probes for NMR spectroscopic studies of molecular conformational dynamics in solution: Applications to macrocyclic molecules. Prog. Nucl. Mag. Res. Sp. 52, 1–21. Boroske, E., Elwenspoek, M., and Helfrich, W. (1981). Osmotic shrinkage of giant egglecithin vesicles. Biophys. J. 34, 95–109. Brumm, T., Mops, A., Dolainsky, C., Bruckner, S., and Bayerl, T. M. (1992). Macroscopic orientation effects in broadline NMR-spectra of model membranes at high magnetic field strength. Biophys. J. 61, 1018–1024. Chu, S. C.-K., Xu, Y., Balschi, J., and Springer, C. S. Jr (1990). Bulk magnetic susceptibility shifts in NMR studies of compartmentalized samples: Use of paramagnetic reagents. Magn. Res. Med. 13, 239–262. Corsi, D. M., Platas-Iglesias, C., van Bekkum, H., and Peters, J. A. (2001). Determination of paramagnetic lanthanide(III) concentrations from bulk magnetic susceptibility shifts in NMR spectra. Magn. Res. Chem. 39, 723–726. Crowell, K. J., and Macdonald, P. M. (2001). Europium III binding and the reorientation of magnetically aligned bicelles: Insights from deuterium NMR spectroscopy. Biophys. J. 81, 255–265. Delli Castelli, D., Gianolio, E., Geninatti Crich, S., Terreno, E., and Aime, S. (2008b). Metal containing nanosized systems for MR-molecular imaging applications. Coord. Chem. Rev. 252, 2424–2443. Delli Castelli, D., Terreno, E., Carrera, C., Giovenzana, G. B., Mazzon, R., Rollet, S., Visigalli, M., and Aime, S. (2008a). Lanthanide-loaded paramagnetic liposomes as switchable magnetically oriented nanovesicles. Inorg. Chem. 47, 2928–2930. Dischino, D. D., Delaney, E. J., Emswiler, J. E., Gaughan, G. T., Prasad, J. S., Srivastava, S. K., and Tweedle, M. F. (1991). Synthesis of nonionic gadolinium chelates useful as contrast agents for magnetic resonance imaging. 1,4,7-Tris(carboxymethyl)-l0substituted- 1,4,7,10-tetraazacyclododecanes and their corresponding gadolinium chelates. Inorg. Chem. 30, 1265–1269. Hirota, S. (2003). Viscometric determination of axial ratio of ellipsoidal DNA-lipid complex. Methods Enzymol. 367, 177–199. Karp, E. S., Inbaraj, J. J., Lryukhin, M., and Lorigan, G. A. (2006). Electron paramagnetic resonance studies of an integral membrane peptide inserted into aligned phospholipid bilayer nanotube arrays. J. Am. Chem. Soc. 128, 12070–12071. Marcotte, I., Belanger, A., and Auger, M. (2006). The orientation effect of gramicidin A on bicelles and Eu3þ-doped bicelles as studied by solid-state NMR and FT-IR spectroscopy. Chem. Phys. Lipids 139, 137–149. Mc Mahon, M. T., Gilad, A. A., De Liso, M. A., Cromer Berman, S. M., Bulte, J. W. M., and van Zijl, P. C. M. (2008). New ‘‘multicolor’’ polypeptide diamagnetic chemical exchange saturation transfer (DIACEST) contrast agents for MRI. Magn. Res. Med. 60, 803–812. Menager, C., and Cabuil, V. (2002). Reversible shrinkage of giant magnetoliposomes under an osmotic stress. J. Phys. Chem. B 106, 7913–7918.

210

Silvio Aime et al.

Park, S. H., Loudet, C., Marassi, F. M., Dufourc, E. J., and Opella, S. J. (2008). Solid-state NMR spectroscopy of a membrane protein in biphenyl phospholipid bicelles with the bilayer normal parallel to the magnetic field. J. Magn. Res. 193, 133–138. Prosser, R. S., and Shyianovskaya, I. V. (2001). Lanthanide ion assisted magnetic alignment of model membranes and macromolecules. Concepts Magnetic Res. 13, 19–31. Prosser, R. S., Bryant, H., Bryant, R. G., and Vold, R. R. (1999). Lanthanide chelates as bilayer alignment tools in NMR studies of membrane-associated peptides. J. Magn. Res. 141, 256–260. Prosser, R. S., Evanics, F., Kitevski, L., and Al-Abdoul-Wahid, M. S. (2006). Current applications of bicelles in NMR studies of membrane-associated amphiphiles and proteins. Biochemistry 45, 8453–8465. Terreno, E., Cabella, C., Carrera, C., Delli Castelli, D., Mazzon, R., Rollet, S., Stancanello, J., Visigalli, M., and Ame, S. (2007). From spherical to osmotically shrunken paramagnetic liposomes: An improved generation of LIPOCEST MRI agents with highly shifted water protons. Angew. Chemie Int. Ed. 46, 966–968. Terreno, E., Delli Castelli, D., Milone, L., Rollet, S., Stancanello, J., Violante, E., and Aime, S. (2008a). First ex-vivo MRI co-localization of two LIPOCEST agents. Contrast Media Mol. Imaging 3, 38–43. Terreno, E., Delli Castelli, D., Cabella, C., Dastru`, W., Sanino, A., Stancanello, J., Tei, L., and Aime, S. (2008b). Paramagnetic liposomes as innovative contrast agents for magnetic resonance (MR) molecular imaging applications. Chem. Biodivers. 5, 1901–1912. Terreno, E., Delli Castelli, D., Violante, E., Sanders, H. M. H. F., Sommerdijk, N. A. J., and Aime, S. (2009). Osmotically shrunken LIPOCEST agents: An innovative class of magnetic resonance imaging contrast media based on chemical exchange saturation transfer. Chem. Eur. J. 15, 1440–1448. Woods, M., Woessner, D. E., and Sherry, A. D. (2006). Paramagnetic lanthanide complexes as PARACEST agents for medical imaging. Chem. Soc. Rev. 35, 500–511. Zhou, J., and van Zijl, P. C. M. (2006). Chemical exchange saturation transfer imaging and spectroscopy. Prog. Nucl. Mag. Res. Sp. 48, 109–136.

C H A P T E R

E L E V E N

Reconstitution of Membrane Proteins in Phospholipid Bilayer Nanodiscs T. K. Ritchie,* Y. V. Grinkova,† T. H. Bayburt,† I. G. Denisov,† J. K. Zolnerciks,* W. M. Atkins,* and S. G. Sligar†

Contents 1. Introduction 2. Overview of Nanodisc Technology 2.1. Structure and properties of Nanodiscs 2.2. MSP expression 2.3. MSP purification 3. Reconstitution Considerations 3.1. Preparing the reconstitution mixture 3.2. Reconstitution of bR trimer 3.3. Assembly of monomeric rhodopsin Nanodiscs 4. Optimizing the Reconstitution for P-glycoprotein 4.1. P-gp as a target for incorporation 4.2. Reconstitution of P-gp 4.3. Functional activity of P-gp in liposomes versus Nanodiscs Acknowledgments References

212 212 213 216 217 218 220 221 223 223 225 225 226 228 228

Abstract Self-assembled phospholipid bilayer Nanodiscs have become an important and versatile tool among model membrane systems to functionally reconstitute membrane proteins. Nanodiscs consist of lipid domains encased within an engineered derivative of apolipoprotein A-1 scaffold proteins, which can be tailored to yield homogeneous preparations of disks with different diameters, and with epitope tags for exploitation in various purification strategies. A critical aspect of the self-assembly of target membranes into Nanodiscs lies in the optimization of the lipid:protein ratio. Here we describe strategies for performing this optimization and provide examples for reconstituting * {

Department of Medicinal Chemistry, University of Washington, Seattle, Washington, USA Department of Biochemistry, University of Illinois at Urbana-Champaign, Urbana, Illinois, USA

Methods in Enzymology, Volume 464 ISSN 0076-6879, DOI: 10.1016/S0076-6879(09)64011-8

#

2009 Elsevier Inc. All rights reserved.

211

212

T. K. Ritchie et al.

bacteriorhodopsin as a trimer, rhodopsin, and functionally active P-glycoprotein. Together, these demonstrate the versatility of Nanodisc technology for preparing monodisperse samples of membrane proteins of wide-ranging structure.

1. Introduction As this volume highlights, model membrane systems are essential for ongoing research aimed at understanding lipid dynamics in complex biological membranes, membrane protein function, and molecular recognition between lipids and proteins or small molecules. In addition, several lipid membrane-based systems have been developed for drug delivery or other applications. Over the course of the past several decades the study of membrane proteins has been accelerated by membrane models including detergent micelles, mixed detergent/lipid micelles, bicelles, and liposomes, facilitating structural determination and functional studies. Although each of these established systems has distinct advantages, none are perfect for all applications and, in fact, each has significant limitations. Therefore, when considering methods for reconstituting membrane proteins, or designing lipid-based nanodevices, a recently established tool based on self-assembling lipid bilayer Nanodiscs is an important development (Bayburt and Sligar, 2002, 2003; Bayburt et al., 2002, 2006, 2007; Chougnet et al., 2007; Denisov et al., 2004; Marin et al., 2007; Morrissey et al., 2008; Nath et al., 2007a; Sligar, 2003). Nanodisc technology provides many advantages for controlling the physical parameters of protein–lipid particles, and they are likely to have utility as components to be incorporated into more complex nanodevices (Das et al., 2009; Goluch et al., 2008; Nath et al., 2008; Zhao et al., 2008). Here we describe the methods used for self-assembly of Nanodiscs and their application for reconstituting various membrane proteins into soluble nanoscale lipid bilayers with controlled composition and stoichiometry.

2. Overview of Nanodisc Technology Phospholipid bilayer Nanodiscs are similar in structure to nascent discoidal high-density lipoprotein particles. They consist of a circular fragment of the phospholipid bilayer encapsulated by two copies of a membrane scaffold protein (MSP) derived from apolipoprotein A-1 (Bayburt et al., 2002; Denisov et al., 2004), as illustrated in Fig. 11.1. A detailed review of the structural and biological aspects of apolipoprotein A-1 and its modification to yield MSPs has been presented (Nath et al., 2007a). Currently available MSP constructs are represented in Table 11.1. They consist of an N-terminal

Reconstitution of Membrane Proteins in Phospholipid Bilayer Nanodiscs

213

Figure 11.1 Structure of Nanodiscs, modeled with POPC as lipid. Lipid bilayer fragment (white space filling) is encircled by two amphipathic helices of MSP (gray ribbon). The graphic was generated using the PyMOL Molecular Graphics system.

hexahistidine tag, a linker containing a protease site enabling the tag to be removed, and the main MSP sequences. Incorporation of membrane proteins into Nanodiscs with the histidine tag removed after purification of MSP enables the separation of empty disks from those containing histidine-tagged target proteins. The main MSP sequence can be varied by changing the number of amphipathic helices punctuated by prolines and glycines, to allow for disks of varying sizes. As summarized in Table 11.1, these scaffold proteins provide a collective set of tools to generate Nanodiscs ranging in outer diameter from 9.8 to 17 nm, which can accommodate a range of membrane proteins.

2.1. Structure and properties of Nanodiscs Optimization of the lipid:protein stoichiometry during the self-assembly process allows production of Nanodiscs of uniform size. The effect of scaffold protein length has been examined by determining the concentration

Table 11.1 Membrane scaffold protein constructs

Protein a

MSP1 MSP1TEV MSP1D1a MSP1D1 D73C MSP1D1(–) MSP1E1a MSP1E1D1 MSP1E2a MSP1E2D1 MSP1E3a MSP1E3D1a MSP1E3D1 D73C MSP1D1– 22 MSP1D1– 33 MSP1D1– 44 MSP2 MSP2N2 MSP2N3 MSP1FC MSP1FN

N-terminus

Disk size (nm) b

c

MW (Da)

e280 (M-1cm-1)

Features

FX TEV TEV TEV

9.7 /9.8 9.7b/10c 9.5b/9.7c 9.6b

24,608 25,947 24,662 24,650

23,950 26,930 21,430 21,430

TEV FX TEV FX TEV FX TEV TEV

9.6b/9.6c 10.4b/10.6c 10.5b 11.1b/11.9c 11.1b 12.1b/12.9c 12.1b 12.0b

22,044 27,494 27,547 30,049 30,103 32,546 32,600 32,588

18,450 32,430 29,910 32,430 29,910 32,430 29,910 29,910

TEV

9.4b

23,404

21,430

Original MSP1 (deletion 1–43 mutant of human Apo A-1) MSP1 with removable 7-his tag Deletion 1–11 mutant of MSP1TEV Cysteine in helix 2, Apo A-1 numbering, mutant of MSP1D1 MSP1D1 with removed 7-His tag Extended MSP1, helix 4 repeated Extended MSP1D1, helix 4 repeated Extended MSP1, helices 4 and 5 repeated Extended MSP1D1, helices 4 and 5 repeated Extended MSP1, helices 4, 5, and 6 repeated Extended MSP1D1, helices 4, 5, and 6 repeated Cysteine in helix 2, Apo A-1 numbering, mutant of MSP1E3D1 Deletion 1–22 mutant of MSP1TEV

TEV

9.0b

22,055

15,930

Deletion 1–33 mutant of MSP1TEV

TEV

8.6b

20,765

15,930

Deletion 1–44 mutant of MSP1TEV

FX TEV TEV TEV TEVF

9.5b 15.0b/16.5c 15.2b/17c 9.7b 9.6b

48,020 45,541 46,125 25,714 25,714

47,900 39,430 39,430 22,400 22,400

Fusion of two MSP1 with GT-linker Fusion of MSP1D1–11 and MSP1D1–22 with GT-linker Fusion of MSP1D1–11 and MSP1D1–17 with GT-linker MSP1D1 with C-terminal FLAG-tag MSP1D1 with N-terminal FLAG-tag

FX ¼ GHHHHHHIEGR; TEV ¼ GHHHHHHHDYDIPTTENLYFQG; TEVF ¼ GHHHHHHHDYDIPTTENLYFQGSDYKDDDDKG. a The plasmid is available through Addgene (http://www.addgene.org). b Stokes hydrodynamic diameter, determined by size-exclusion chromatography (Denisov et al., 2004). c Nanodisc diameter determined by SAXS (Denisov et al., 2004).

215

Reconstitution of Membrane Proteins in Phospholipid Bilayer Nanodiscs

of radiolabeled lipid and scaffold protein in the Nanodisc-containing size exclusion peak (Denisov et al., 2004). These results, summarized in Fig. 11.2, illustrate an interesting trend. Insertion of extra helices in the central portion of the scaffold protein (MSP1E1, MSP1E2, and MSP1E3) results in Nanodiscs of increasing size, while deletions of the affinity tag and the first 22 amino acids of the N-terminus do not significantly decrease the size of the disk formed, implying that the first 22 amino acids are marginally, if at all, involved in the self-assembly process and resultant stabilization of the discoidal nanoparticle. Truncation past the first 22 amino acids leads to a gradual decrease in lipid:protein ratio accompanied by a decrease in the major monodisperse Nanodisc component and an increase in aggregated fractions. Systematic studies of the lipid:protein ratio in Nanodiscs made from different MSP constructs have shown that the number of lipids per Nanodisc, NL, and the number of amino acids in the scaffold protein, M, can be described by the following simple relationship (Eq. (11.1), modified Eq. (11.2) from Denisov et al., 2004): NL S ¼ ð0:423M

9:75Þ2

ð11:1Þ

where S represents the mean surface area per lipid used to form the ˚ 2. The quadratic relationship between the number Nanodisc, measured in A of lipid molecules per Nanodisc and the length of the scaffold protein confirms the flat two-dimensional morphology of Nanodisc particles,

40

1 SP TE 1T V EV M ( SP −) 1Δ 1M SP 11 1Δ 1M SP 22 1Δ 1M 33 SP 1Δ 144

20

SP 1

2 2N

3

SP

1E

M

M

SP M

1E

1 1E

SP M

SP M

M

2

100

60

M

200

80

SP

300

M

Number of DPPC molecules per leaflet

B

SP 1

Number of DPPC molecules per leaflet

A

Figure 11.2 Number of DPPC molecules per Nanodisc determined experimentally using tritiated lipids. Panel A: number of lipids in Nanodiscs formed with extended MSP proteins. Panel B: number of lipids in Nanodiscs formed with truncated MSP proteins. For the description of MSP constructs, see Table 11.1.

216

T. K. Ritchie et al.

illustrated in Fig. 11.1. The size similarity of Nanodiscs formed using the same scaffold protein but different lipids clearly indicates that the length of the protein’s amphipathic helix is the sole determinant of Nanodisc diameter, while different lipid:protein stoichiometries are due to the different surface area per lipid. For example, 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC) is in gel state below 314 K, with the area per lipid in the ˚ 2, while 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphochorange of 52–57 A line (POPC) is in liquid crystalline state above 268 K, with the area per lipid approximately 70 A˚2.

2.2. MSP expression MSPs are expressed using the pET expression system (Novagen) with the BL21-Gold (DE3) strain (Stratagene) as a host. The expression is very efficient, and a large amount of protein is produced in just a few hours after induction with IPTG. However, MSPs are noticeably susceptible to proteolysis, and prolonged postinduction growth results in significant decrease of the MSP yield. Different modifications of the N- and C-termini of the MSP can affect stability in vivo, and for some MSPs (e.g., fusion constructs or epitope tagged MSPs), shortening of the postinduction time and/or lowering the temperature during the growth in comparison with the standard protocol improves yield. The highest yield is achieved with a rich medium such as terrific broth (TB); however, minimal medium was used successfully for production of the isotope labeled MSP (Li et al., 2006). Relatively high oxygenation level, which is essential for good yields, can be easily maintained in a fermenter, such as Bio-Flow III. However, satisfactory yields can also be achieved in flasks by using relatively small culture volume (e.g., 500 mL in a 2-L Fernbach flask). The detailed method is outlined below: (1) A starting culture is prepared as follows: 30 mL of Luria Broth (LB) medium containing kanamycin (30 mg/L) is inoculated with a single colony from a freshly streaked plate. The suspension is incubated at 37  C with shaking at 250 rpm until the OD600 is approximately 0.4–0.6 (usually 5–6 h). At this point the culture can be used immediately or stored overnight at 4  C. (2) 2.5 L TB medium is prepared and sterilized and the fermenter parameters (37  C, 500 rpm, and air—3 L/min) are set. When the temperature reaches 37  C, 25 mg kanamycin and a few drops of antifoam are added and the fermenter is inoculated with the starting culture. (3) OD600 is checked every hour. When the OD reaches 2.5–3.0 (usually in 3–4 h), the culture is induced with 1 mM IPTG. The fermentation is stopped 3 h after induction. Typically, OD600 reaches 10–15 by the end of fermentation.

Reconstitution of Membrane Proteins in Phospholipid Bilayer Nanodiscs

217

(4) The cells are harvested by centrifugation at 8000g for 10 min. The weight of the wet pellet collected from 2.5 L of culture grown on TB medium is usually between 50 and 60 g. The cell pellet is stored at 80  C.

2.3. MSP purification MSPs are purified using Chelating Sepharose FF (GE Healthcare), charged with Ni2þ, following the general protocol for purification of polyhistidinetagged proteins with additional washing steps using detergent-containing buffers to disrupt interaction of MSP with other proteins: (1) The metal-chelating column (3.4  6 cm) is charged by passing through 1 bed volume (50 mL) of 0.1 M NiSO4, followed by 100 mL of water. The column is equilibrated with 250 mL of 40 mM phosphate buffer, pH 7.4. (2) The cell pellet collected from 2.5 L fermentation (40–60 g) is resuspended in 200 mL of 20 mM phosphate buffer, pH 7.4. Phenylmethylsulfonyl fluoride (PMSF) is added from a stock solution in ethanol to make 1 mM. After the cells are completely resuspended, Triton X-100 is added to a final concentration of 1%. Approximately 5 mg of deoxyribonuclease I (Sigma, DN-25) is added. The cells are lysed by sonication (three 1-min rounds). The lysate is clarified by centrifugation at 30,000g for 30 min. (3) The lysate is loaded on the column. Care should be taken to make sure the flow rate does not exceed 10 mL/min (about 1 mL/min cm2). The column is washed with 250 mL of each of the following:  40 mM Tris/HCl, 0.3 M NaCl, 1% Triton X-100, pH 8.0  40 mM Tris/HCl, 0.3 M NaCl, 50 mM Na-cholate, 20 mM imidazole, pH 8.0  40 mM Tris/HCl, 0.3 M NaCl, 50 mM imidazole, pH 8.0 (4) MSP is eluted with 40 mM Tris/HCl, 0.3 M NaCl, 0.4 M imidazole. 10–14 mL fractions are collected, and protein is checked with Coomassie G-250 reagent (Pierce). The fractions containing MSP are pooled and the sample is dialyzed against buffer 1 (20 mM Tris/HCl, 0.1 M NaCl, 0.5 mM EDTA, pH 7.4) at 4  C. The protein sample is filtered using 0.22 mm syringe filter, and 0.01% NaN3 is added for storage. (5) Analyze the sample: protein purity is checked by running SDS–PAGE and performing electrospray mass spectrometry (see Table 11.1 for molecular masses). Absorbance is measured at 280 nm using 1 mm path length quartz cuvette against standard buffer, and protein concentration is calculated. If necessary, it is concentrated to 4–10 mg/mL. MSP can be stored for several days at 4  C. For long-term storage, the sample is frozen or lyophilized, and is stored at 20  C or below.

218

T. K. Ritchie et al.

(6) After purification, the column is regenerated with 50 mM EDTA, is washed with water, and is equilibrated with 20% ethanol. The column is regenerated after every purification round.

3. Reconstitution Considerations As of 2009, the list of membrane proteins reconstituted into Nanodiscs for functional studies include the cytochromes P450 (Baas et al., 2004; Bayburt and Sligar, 2002; Civjan et al., 2003; Das et al., 2007, 2009; Denisov et al., 2006, 2007; Duan et al., 2004; Grinkova et al., 2008; Kijac et al., 2007; Nath et al., 2007b) bacteriorhodopsin as a monomer and trimer (Bayburt and Sligar, 2003; Bayburt et al., 2006), G-protein coupled receptors as monomers and dimers (Bayburt et al., 2007; Leitz et al., 2006; Marin et al., 2007), other receptors (Boldog et al., 2006, 2007; Mi et al., 2008), toxins (Borch et al., 2008), blood coagulation protein tissue factor (Morrissey et al., 2008; Shaw et al., 2007), protein complexes of the translocon (Alami et al., 2007; Dalal et al., 2009), and monoamine oxidase (Cruz and Edmondson, 2007). The potential of Nanodiscs is exemplified by their utility in diverse biochemical and biophysical methodologies, including solid state NMR (Kijac et al., 2007; Li et al., 2006), single molecule fluorescence experiments (Nath et al., 2008), and solubilizing functional receptors (Bayburt et al., 2007; Boldog et al., 2007; Leitz et al., 2006; Mi et al., 2008). Importantly, these methods may be modified to accommodate other membrane proteins. As an example, we describe the methods of reconstitution of bacteriorhodopsin (bR) trimer and rhodopsin monomer. Assembly of membrane proteins into Nanodiscs follows the rules for empty Nanodiscs. Cholatesolubilized phospholipids (see Section 3.1) are mixed with MSP and detergent-solubilized membrane protein. Following detergent removal with adsorbent beads (Bio-beads SM-2, Biorad or Amberlite XAD-2; SigmaAldrich), the assembly is analyzed and purified by size-exclusion chromatography. Additional parameters to consider are the choice of detergent to initially solubilize the protein from its membrane, choice of Nanodisc size, and the lipid to MSP to membrane protein ratios. Incorporation of a membrane protein into Nanodiscs requires the protein to be initially solubilized by treatment with a detergent. For a practical guide to membrane protein solubilization, see Hjelmeland and Chrambach (1984). The crude solubilized protein can be put directly into Nanodiscs or purified beforehand. A distinct advantage of using the crude-solubilized membrane is that membrane proteins tend to be labile in detergent, and affinity purification can be done after the target is in the Nanodiscs. The use of protein purified in detergent has the advantage that the native lipid is mostly removed, thus simplifying determination of the correct MSP to phospholipid ratio.

Reconstitution of Membrane Proteins in Phospholipid Bilayer Nanodiscs

219

When using purified protein, however, the presence of relatively high glycerol concentrations can interfere with the assembly process, so the final concentration in the reconstitution mixture should be kept below 4%. As with empty disks, the phospholipid:MSP ratio must be satisfied (see Table 11.2). Stated more precisely, the surface area of the target plus phospholipid bilayer needs to be matched to the size of Nanodisc being assembled. It should be recognized that target protein, along with any associated native lipid, will displace exogenously added phospholipid from the Nanodisc structure. ˚ 2 for DPPC, 57 A ˚ 2 for 1,2The mean surface area per lipid in Nanodiscs is 52 A 2 ˚ dimyristoyl-sn-glycero-3-phosphocholine (DMPC), and 69 A for POPC (Bayburt et al., 2002, 2006; Denisov et al., 2004). These numbers can be used as a starting point for determining the necessary amount of phospholipid, shown in Table 11.2 for empty Nanodiscs. If the structure of the target is known, an estimate of displaced lipid can be made based on cross-sectional area of the membrane domain. If the structure is not known, an estimate can be made using an area of 140 A˚2 per transmembrane helix. The Swiss-Prot database (http://www.expasy.org) annotates potential transmembrane helices for proteins in its database and ExPASy provides links to topology prediction tools for unknown proteins. One then simply subtracts the number of phospholipids displaced by the target protein, and any native lipid present, from the amount of lipid that would be used to form empty Nanodiscs of the same size and phospholipid type. Bacteriorhodopsin was found to displace 37 DMPC molecules and rhodopsin displaced 50 POPC molecules based on chemical and spectral analysis of purified Nanodiscs (Bayburt et al., 2006, 2007). The experimentally determined numbers are consistent with the cross-sectional areas of bR trimer corresponding to 40 DMPC and rhodopsin corresponding to 43 POPC estimated from the crystal structures. These results indicate that a simple subtraction of phospholipid to account for the surface area of protein is a valid approximation. Endogenous lipid must also be accounted for when reconstituting from whole solubilized membrane. A crude approximation is that the weight of lipid is equal to the weight of total protein in a membrane. We estimate the concentration of lipid using the molecular weight of POPC (MW 760). It is often convenient to use a large excess of MSP and synthetic phospholipid

Table 11.2 Reconstitution ratios for empty disks POPC

MSP1D1 65 MSP1E1D1 85 MSP1E2D1 105 MSP1E3D1 130

DPPC

DMPC

˚ 2) Bilayer area per Nanodisc (A

90 115 145 180

80 100 130 160

4400 5700 7200 8900

220

T. K. Ritchie et al.

compared to native membrane lipid so that the contribution of native lipid and membrane protein can be neglected. Once a reconstitution has been performed and analyzed by size-exclusion chromatography, the lipid:MSP ratio can be adjusted to optimize the formation of Nanodiscs. Another obvious consideration is the choice of the Nanodisc size. The bilayer area for several disk sizes is given in Table 11.2. Importantly, a critical number of phospholipids associated with the Nanodisc–protein complex may be necessary for nativelike structure. Theoretically, three bR can fit into MSP1 Nanodiscs, but the trimer only forms in the larger Nanodiscs, which suggests that sufficient phospholipid must be present to allow unperturbed oligomer formation. A final consideration is the target protein to disk ratio in the assembly mixture. Single monomeric membrane proteins will assemble into Nanodiscs as long as the ratio of Nanodisc to target is high (i.e., the number per Nanodisc follows the Poisson distribution for noninteracting target). If an oligomeric membrane protein is desired then one must consider the strength of interaction, as increasing the phospholipid component can dissociate oligomers by a surface dilution effect. For weakly interacting proteins, such as the bR homotrimer, the choice of Nanodisc to target ratio is critical (Bayburt et al., 2006). Experimentally, the ratio of bR to Nanodisc was varied to find the optimal ratio. A similar approach was used for the Tar receptor (Boldog et al., 2006). Bacteriorhodopsin trimer exhibits exciton formation that was used as a convenient assay for trimer formation. In the case of Tar, a functional assay suggested that a trimer of dimers formed at a specific reconstitution ratio. A few simple tests for assembly of a target protein with Nanodiscs can be performed to ensure efficient reconstitution. Separation of the reconstituted sample using a calibrated Superdex 200 column will allow determination of size and homogeneity of the Nanodiscs. If excess empty disks are present, column fractions can be analyzed for the presence of target by techniques such as SDS– PAGE or activity assays. Upon reinjection, the peak target fraction should elute at the same position without degradation or aggregation; size changes in the peak fraction indicate improper Nanodisc formation. The amount of phospholipid can be measured and should correspond to the expected value, as described above. For the measurement to be meaningful, however, the target-containing Nanodiscs must be isolated first from any empty Nanodiscs.

3.1. Preparing the reconstitution mixture Lipid stocks are prepared in chloroform at 25–100 mM and stored at 20  C in glass vials with Teflon-lined screw caps. The concentration of the stock solution is determined by phosphate analysis (Chen et al., 1956; Du¨zgu¨nes¸, 2003). The desired amount of chloroform lipid stock is dispensed into a disposable glass culture tube and dried using a gentle stream of nitrogen gas in a fume hood; a thin film on the lower walls of the tube can be obtained by rotating the tube while holding it at an angle. To remove

Reconstitution of Membrane Proteins in Phospholipid Bilayer Nanodiscs

221

residual solvent, the tube is placed in a vacuum dessicator under high vacuum overnight. Buffer containing sodium cholate is added to the dried lipid film. Typically, cholate is added to twice the desired concentration of lipid, for example, if 200 mL of 100 mM lipid stock was used, 200 mL of 200 mM cholate or 400 mL of 100 mM cholate is added. The tube is vortexed, heated under hot tap water (about 60  C), and sonicated in an ultrasonic bath until the solution is completely clear, and no lipid remains on the walls of the tube. Scaffold protein is added to cholate-solubilized phospholipid to yield desired lipid:protein ratio, ensuring the final cholate concentration in the reconstitution mixture is between 12 and 40 mM, supplementing with standard buffer or cholate stock solution if necessary. The mixture is incubated at the appropriate incubation temperature, which is dependent on the lipid used, for 15 min or longer. The temperature of the self-assembly should be near the Tm of the lipid being used. Assembly with POPC is done on ice or at 4  C, DMPC at room temperature, and DPPC at 37  C. Prepared disk reconstitution mixtures can be used immediately to make Nanodiscs or incorporate membrane proteins, or lyophilized for prolonged storage. Specific examples in the following subsections demonstrate these steps with different proteins.

3.2. Reconstitution of bR trimer Purple membrane is isolated from Halobacterium salinarum JW-3 cultures and solubilized with 4% (w/v) Triton X-100 as described (Dencher and Heyn, 1978; Oesterhelt and Stoeckenius, 1974). MSP1E3 stock solutions ( 200 mM) and a DMPC/cholate mixture (200 mM/400 mM in buffer 1, prepared as described above) are added to bR (200 mM) in a microfuge tube to give MSP1E3:bR:DMPC ratio of 2:3:160. Protease inhibitors can be included in the assembly. The final concentration of DMPC should be above 7 mM, below which poor disk formation occurs (Bayburt et al., 2006). If low phospholipid concentrations are necessary, Nanodisc formation can be aided by using sodium cholate at a final concentration of 14 mM. After 1 h incubation at room temperature, detergent is removed by treatment for 3–4 h at room temperature with 500 mg wet Bio-beads SM2 per mL of solution, with gentle agitation to keep the beads suspended. Bio-beads SM-2 or Amberlite XAD-2 are prepared by suspending in methanol, washing with several volumes of methanol in a sintered glass funnel, and rinsing with large amounts of Milli-Q treated water (Millipore) to remove traces of methanol. Amberlite XAD-2 additionally requires removal of fine particles by decantation. Prepared beads are stored in water containing 0.01% (w/v) NaN3 as preservative. Incubation temperature and amount of beads are factors in the rate and completeness of detergent removal (Rigaud et al., 1998). We generally use an equal volume of beads to sample and an overnight incubation to remove detergents at

222

T. K. Ritchie et al.

4  C. Room temperature or 37  C assemblies require several hours. A table of adsorption capacities for various detergents has been compiled (Rigaud et al., 1998). If it is critical that the amount of residual detergent is known, the assembly should be tested using radiolabeled detergent. Bio-beads are removed by punching a hole in the bottom of the microfuge tube with a needle, placing the tube snugly through a hole made in the cap of a 15-mL Falcon tube (Corning), and punching a vent hole in the cap of the microfuge tube. The assembly is centrifuged briefly using the Falcon tube to collect the sample. The sample is filtered using a 0.22-mm filter and injected onto the gel filtration column run at 0.5 mL/min while monitoring A280 and A560. A typical elution profile after assembly of trimer is shown in Fig. 11.3, panel A. The reconstitution was made using optimal amount of phospholipid, yet the Nanodisc peak is still accompanied by larger aggregates that also contain bR. One possible explanation for the presence of aggregates is that multiple bR interactions promote an aggregation pathway as opposed to formation of Nanodiscs of fixed size. Fractions containing the bR Nanodiscs are pooled and the presence of trimer is assessed by measuring the visible circular dichroism spectrum which shows a positive and negative peak, due to exciton splitting (Bayburt et al., 2006).

A

B Rhodopsin nanodisc

bR nanodisc

0.5 OD

5 mOD 500 nm

Aggregates

280 nm 560 nm

0

5

10 15 20 Elution time (min)

Aggregates

25

15

20

25 30 Elution time (min)

35

Figure 11.3 Elution profile from Nanodisc reconstitutions. Panel A: elution profile of MSP1E3 bR trimer Nanodiscs after assembly. After detergent removal the sample was injected onto a Superdex 200 prep grade column at a flow rate of 0.5 mL/min. The main peak corresponds to Nanodiscs containing three bR. Panel B: elution profile of MSP1E3 rhodopsin Nanodisc assembly mixture produced from solubilized rod outer segments. The sample was injected onto a Superdex 200 HR 10/30 column run at a flow rate of 0.5 mL/min.

Reconstitution of Membrane Proteins in Phospholipid Bilayer Nanodiscs

223

3.3. Assembly of monomeric rhodopsin Nanodiscs The assembly described herein uses whole membrane and added synthetic phospholipid to generate rhodopsin monomer Nanodiscs. Rhodopsin is handled in a darkroom under dim red light (Kodak #1 filter, 7.5 W bulb). Rod outer segments (Papermaster, 1982) are solubilized in 135 mM nonyl glucoside to give 143 mM solubilized rhodopsin. Rod outer segments contain on the order of 100 native phospholipids per rhodopsin. MSP1E3D1 (183 mM) and DMPC (0.1 M in buffer 1 containing 0.2 M cholate) are mixed with solubilized membranes at ratios of 1:168:0.05 (MSP:DMPC:rho) on ice followed by overnight removal of detergent with Bio-beads at 4  C with gentle agitation. The sample is filtered and run on a Superdex 200 HR 10/30 column run at 0.5 mL/min. The elution profile monitored at 500 nm is given in Fig. 11.3, panel B. The elution profile shows a sharp Gaussian peak, though there are small amounts of larger aggregates. The aggregates indicate that the amount of DMPC in the reconstitution could be lowered somewhat to optimize assembly of Nanodiscs.

4. Optimizing the Reconstitution for P-glycoprotein When embarking on the incorporation of a new target into Nanodiscs, one must not only consider the requirements of the Nanodisc system but also any unique requirements of the target of interest. Herein we describe the tailoring of the reconstitution to an important mammalian protein, P-glycoprotein (P-gp). P-gp is a member of the ATP-binding cassette (ABC) transporter family which has been implicated in the phenomenon of multidrug resistance in tumor cells (Higgins, 2007), as well as the absorption and disposition of many pharmaceutical compounds (Zhou, 2008), yet there is still a great deal about the mechanism and interaction with substrates that is unknown. In fact, structure–function studies of P-gp have been seriously hampered by the difficulty of obtaining large quantities of stable P-gp. Presumably, this difficulty results from the structural complexity of P-gp which comprises a 1280 amino acid protein with 12 transmembrane helices punctuated by two cytoplasmic nucleotide-binding domains (NBDs) (Higgins et al., 1997). A recent crystal structure of mouse P-gp (Abcb1a, 87% homology with human P-gp) is shown in Fig. 11.4, to illustrate the domain architecture (Aller et al., 2009). P-gp is known to be sensitive to both the lipid environment (Orlowski et al., 2006) and the detergent used during the purification process (Bucher et al., 2007). Disruption of the lipid–protein interface has been shown to

224

T. K. Ritchie et al.

TMDs

NBDs

Figure 11.4 Crystal structure of mouse P-gp (PDB: 3G5U) in the nucleotide-free state, as seen from the plane of the membrane (Aller et al., 2009). The TMDs are embedded in the membrane, while the NBDs protrude into the interior of the cell. The graphic was generated using the PyMOL Molecular Graphics system.

result in almost complete inactivation of the protein (Callaghan et al., 1997); in fact, a common practice in the purification of P-gp is to add external lipid to maintain this crucial interface (Ambudkar et al., 1998; Taylor et al., 2001). Many detergents commonly used to solubilize membrane proteins disrupt the protein–lipid interaction, and are thus detrimental for use with P-gp (Naito and Tsuruo, 1995). N-Dodecyl-b-D-maltoside (DDM) is a mild, nonionic detergent that is commonly used in the solubilization and reconstitution of P-gp (Kimura et al., 2007; McDevitt et al., 2008), and which has also previously been used in the formation of Nanodiscs (Alami et al., 2007; Boldog et al., 2006; Dalal et al., 2009). It was, therefore, chosen to use in the incorporation of P-gp into Nanodiscs. The standard lipid used during the purification and liposomal reconstitution of P-gp is an Escherichia coli total lipid extract (Kim et al., 2006; Taylor et al., 2001), which is a mixture of phosphatidylethanolamine (57.5%), phosphatidylglycerol (15.1%), cardiolipin (9.8%), and ‘‘other’’ lipids (17.6%). This mixture seems to satisfy the requirement P-gp has for the lipid content, as exemplified by high levels of

Reconstitution of Membrane Proteins in Phospholipid Bilayer Nanodiscs

225

drug-stimulated ATPase activity in reconstituted proteoliposomes (Ambudkar et al., 1998; Taylor et al., 2001) and has also been used, with DDM, in the formation of Nanodiscs (Alami et al., 2007; Dalal et al., 2009).

4.1. P-gp as a target for incorporation There are currently four in vitro systems routinely utilized to study P-gp: whole cells overexpressing P-gp (Adachi et al., 2001; Polli et al., 2001; Takano et al., 1998; Wang et al., 2002), membrane fractions from those cells (Loo and Clarke, 2005; Loo et al., 2003; Zolnerciks et al., 2007), purified protein that has been solubilized in detergent (Liu et al., 2000; Qu et al., 2003; Rosenberg et al., 2005), and purified protein that has been reconstituted into proteoliposomes (Kim et al., 2006; Lu et al., 2001; Taylor et al., 2001). Each system has strengths and weaknesses; in the whole cell and membrane fraction systems the protein is in the most native form but there is the obvious concern about the complexity of the system. Human P-gp that has been detergent-solubilized shows no ATPase activity, whereas protein that has been reconstituted into proteoliposomes has ATPase activity (Ambudkar et al., 1998), but is not particularly stable. In fact, at room temperature P-gp-proteoliposomes have a half-life of less than 1 day (Heikal et al., 2009). Nanodiscs afford an attractive system to study P-gp because they allow for a relatively simple, controlled system in which P-gp is solubilized, yet in an active form.

4.2. Reconstitution of P-gp Baculovirus-encoding dodeca-histidine-tagged-P-gp was a generous gift from Dr. Kenneth Linton (Imperial College, London). Production of P-gp containing insect cell membranes and protein purification is performed as previously described (Taylor et al., 2001), with modifications. Briefly, insect cell membrane fractions are solubilized in solubilization buffer (20 mM Tris, 150 mM NaCl, 1.5 mM MgCl2, 20% glycerol, 0.4% lipid (80:20 E. coli total lipid:cholesterol), and 2% DDM, pH 6.8) with repeated extrusion through a 25-gauze needle. Insoluble protein is separated by centrifugation at 100,000g for 40 min. The resulting solubilized protein is incubated with ProBond Nickel-Chelating Resin (Invitrogen) for 1 h at 4  C with constant agitation, with the addition of 20 mM imidazole to reduce nonspecific binding. The resin is washed with 20 bed volumes of wash buffer (20 mM Tris, 150 mM NaCl, 1.5 mM MgCl2, 20% glycerol, 0.1% DDM, pH 8) with increasing concentrations of imidazole (80–150 mM). P-gp containing fractions are eluted with 500 mM imidazole in elution buffer (same as wash buffer, pH 6.8) and stored at 80  C until used.

226

T. K. Ritchie et al.

(1) A lipid film of 12 mmol E. coli total lipid (molar concentration determined as described above) is prepared and vacuum desiccated overnight. (2) The lipid film is resuspended in 17 mmol DDM and 1 mL buffer 1 (20 mM Tris, 100 mM NaCl, pH 7.4), then sonicated and vortexed until the solution is clear and free of lumps of lipid. (3) 500 mL of purified P-gp in elution buffer is added, as welll as, protease inhibitors (20 mM leupeptin, 1 mM benzamidine, and 1 mM pepstatin), 100 nmol MSP1E3D1, and enough buffer 1 to make a total volume of 2.5 mL, ensuring the final glycerol concentration is less than 4%. The mixture is then incubated at room temperature with constant agitation for 1 h. (4) To initiate self-assembly, 0.6 g/mL washed Bio-beads SM-2 are added and incubated at room temperature for 2 h with constant agitation. (5) Reconstituted Nanodiscs are removed from Bio-beads with a 25-gauge needle and is stored at 4  C until used. (6) Empty Nanodiscs can be made in parallel, adding 500 mL of elution buffer in place of purified P-gp.

4.3. Functional activity of P-gp in liposomes versus Nanodiscs Functional characterization of a transporter protein in Nanodiscs has unique challenges. A disadvantage of using Nanodiscs to study transporters, such as P-gp, is the inability to study true vectorial transport, per se, because there is no internal or external compartment. Fortunately, a majority of the substrates transported by P-gp stimulate ATPase activity, which can be used as a surrogate for many of the conformational and chemical processes functionally coupled to transport (Polli et al., 2001). As mentioned previously, human P-gp has no detectable ATPase activity when solubilized in DDM, but regains activity when reconstituted. The amount of lipid is stringently controlled during the reconstitution process to prevent the concurrent formation of liposomes. Thus, the activity that is determined after reconstitution can be attributed to P-gp in Nanodiscs. For an initial characterization, the activity of P-gp reconstituted in Nanodiscs was determined by measuring the basal and drug-stimulated ATPase activity in MSP1E3D1 disks and in proteoliposomes, the standard reconstitution system for P-gp. Proteoliposomes are formed as previously described, with modifications (Taylor et al., 2001). Briefly, a mixture of E. coli lipid and cholesterol (80:20, w/w) is dried to a lipid film, before rehydration in elution buffer without DDM. The solution is sonicated and vortexed to make unilamelar liposomes. DDM is added to completely solubilize the lipid, and the solution is incubated at room temperature for 1 h to equilibrate. Equal volumes of the solubilized lipid and purified P-gp are incubated with protease inhibitors for 30 min at room temperature with constant agitation. Detergent is selectively

227

Reconstitution of Membrane Proteins in Phospholipid Bilayer Nanodiscs

removed by addition of 0.3 g/mL of Bio-beads SM-2 for 2 h at room temperature with constant agitation. Proteoliposomes are recovered with a 25-gauge needle and stored on ice until used. Basal and drug-stimulated ATPase activity was determined by phosphate release using a colorimetric assay, as previously described (Chifflet et al., 1988), at 50 mM nicardipine, with varying concentrations of ATP (Taylor et al., 2001). Empty disks or liposomes made in parallel were used as a control. Figure 11.5 shows the comparison of basal and nicardipine-stimulated activity of P-gp in MSP1E3D1 Nanodiscs and liposomes. A twofold increase in the maximum drug-stimulated ATPase activity in Nanodiscs, compared to liposomes, is seen, while the Km values are comparable. This could be due to the uniform orientation of P-gp in Nanodiscs, whereas in liposomes there are two possible orientations: right-side-out (NBDs on the interior of the liposomes, and therefore inaccessible to ATP) and inside-out (NBDs on the exterior of the liposomes, and therefore accessible to ATP). This scrambled orientation in liposomes is consistent with incorporation of the protein using completely solubilized lipid (Rigaud, 2002). An increase in basal activity is also seen in disks as compared to liposomes, where the basal activity is almost undetectable. These data not only show that P-gp is functionally active when reconstituted into Nanodiscs, but that it exhibits higher specific activity than the current standard reconstitution system as well. P-gp is a complex, integral membrane protein containing 12 transmembrane helices that was incorporated into Nanodiscs in a fairly straightforward manner, after small modifications to the standard procedure. This will facilitate a more

nmol Pi/min/mg P-gp

1600

1200

800

400

0 0.0

0.5

1.0 1.5 mM ATP

2.0

2.5

Figure 11.5 ATPase activity of P-gp in MSP1E3D1 Nanodiscs as compared to proteoliposomes. Squares represent the activity of P-gp in MSP1E3D1 Nanodiscs and circles represent activity in liposomes. Open symbols show basal activity in the absence of drug and filled symbols show activity in the presence of 50 mM nicardipine.

228

T. K. Ritchie et al.

detailed study into the mechanism of P-gp and its interaction with substrates and serves to exemplify the utility of Nanodiscs in the study of membrane proteins.

ACKNOWLEDGMENTS The work described here was supported by Grants GM 33775 and GM 31756 to S. G. S. and GM 32165 to W. M. A.

REFERENCES Adachi, Y., Suzuki, H., and Sugiyama, Y. (2001). Comparative studies on in vitro methods for evaluating in vivo function of MDR1 P-glycoprotein. Pharm. Res. 18, 1660–1668. Alami, M., Dalal, K., Lelj-Garolla, B., Sligar, S. G., and Duong, F. (2007). Nanodiscs unravel the interaction between the SecYEG channel and its cytosolic partner SecA. EMBO J. 26, 1995–2004. Aller, S. G., Yu, J., Ward, A., Weng, Y., Chittaboina, S., Zhuo, R., Harrell, P. M., Trinh, Y. T., Zhang, Q., Urbatsch, I. L., and Chang, G. (2009). Structure of Pglycoprotein reveals a molecular basis for poly-specific drug binding. Science 323, 1718–1722. Ambudkar, S. V., Lelong, I. H., Zhang, J., and Cardarelli, C. (1998). Purification and reconstitution of human P-glycoprotein. Methods Enzymol. 292, 492–504. Baas, B. J., Denisov, I. G., and Sligar, S. G. (2004). Homotropic cooperativity of monomeric cytochrome P450 3A4 in a nanoscale native bilayer environment. Arch. Biochem. Biophys. 430, 218–228. Bayburt, T. H., and Sligar, S. G. (2002). Single-molecule height measurements on microsomal cytochrome P450 in nanometer-scale phospholipid bilayer disks. Proc. Natl. Acad. Sci. USA 99, 6725–6730. Bayburt, T. H., and Sligar, S. G. (2003). Self-assembly of single integral membrane proteins into soluble nanoscale phospholipid bilayers. Protein Sci. 12, 2476–2481. Bayburt, T. H., Grinkova, Y. V., and Sligar, S. G. (2002). Self-assembly of discoidal phospholipid bilayer nanoparticles with membrane scaffold proteins. Nano Lett. 2, 853–856. Bayburt, T. H., Grinkova, Y. V., and Sligar, S. G. (2006). Assembly of single bacteriorhodopsin trimers in bilayer nanodiscs. Arch. Biochem. Biophys. 450, 215–222. Bayburt, T. H., Leitz, A. J., Xie, G., Oprian, D. D., and Sligar, S. G. (2007). Transducin activation by nanoscale lipid bilayers containing one and two rhodopsins. J. Biol. Chem. 282, 14875–14881. Boldog, T., Grimme, S., Li, M., Sligar, S. G., and Hazelbauer, G. L. (2006). Nanodiscs separate chemoreceptor oligomeric states and reveal their signaling properties. Proc. Natl. Acad. Sci. USA 103, 11509–11514. Boldog, T., Li, M., and Hazelbauer, G. L. (2007). Using Nanodiscs to create water-soluble transmembrane chemoreceptors inserted in lipid bilayers. Methods Enzymol. 423, 317–335. Borch, J., Torta, F., Sligar, S. G., and Roepstorff, P. (2008). Nanodiscs for immobilization of lipid bilayers and membrane receptors: Kinetic analysis of cholera toxin binding to a glycolipid receptor. Anal. Chem. 80, 6245–6252.

Reconstitution of Membrane Proteins in Phospholipid Bilayer Nanodiscs

229

Bucher, K., Belli, S., Wunderli-Allenspach, H., and Kramer, S. D. (2007). P-glycoprotein in proteoliposomes with low residual detergent: The effects of cholesterol. Pharm. Res. 24, 1993–2004. Callaghan, R., Berridge, G., Ferry, D. R., and Higgins, C. F. (1997). The functional purification of P-glycoprotein is dependent on maintenance of a lipid-protein interface. Biochim. Biophys. Acta 1328, 109–124. Chen, P. S., Toribara, T. Y., and Warner, H. (1956). Microdetermination of phosphorus. Anal. Chem. 28, 1756–1759. Chifflet, S., Torriglia, A., Chiesa, R., and Tolosa, S. (1988). A method for the determination of inorganic phosphate in the presence of labile organic phosphate and high concentrations of protein: Application to lens ATPases. Anal. Biochem. 168, 1–4. Chougnet, A., Grinkova, Y., Ricard, D., Sligar, S., and Woggon, W.-D. (2007). Fluorescent probes for rapid screening of potential drug-drug interactions at the CYP3A4 level. ChemMedChem 2, 717–724. Civjan, N. R., Bayburt, T. H., Schuler, M. A., and Sligar, S. G. (2003). Direct solubilization of heterologously expressed membrane proteins by incorporation into nanoscale lipid bilayers. Biotechniques 35, 556–558, 560, 562–563. Cruz, F., and Edmondson, D. E. (2007). Kinetic properties of recombinant MAO-A on incorporation into phospholipid nanodisks. J. Neural Transm. 114, 699–702. Dalal, K., Nguyen, N., Alami, M., Tan, J., Moraes, T. F., Lee, W. C., Maurus, R., Sligar, S. G., Brayer, G. D., and Duong, F. (2009). Structure, binding, and activity of Syd, a SecY-interacting protein. J. Biol. Chem. 284, 7897–7902. Das, A., Grinkova, Y. V., and Sligar, S. G. (2007). Redox potential control by drug binding to cytochrome P 450 3A4. J. Am. Chem. Soc. 129, 13778–13779. Das, A., Zhao, J., Schatz, G. C., Sligar, S. G., and Van Dyne, R. P. (2009). Screening of type I and II drug binding to human cytochrome P450–3A4 in nanodiscs by localized surface plasmon resonance spectroscopy. Anal. Chem. 81, 3754–3759. Dencher, N. A., and Heyn, M. P. (1978). Formation and properties of bacteriorhodopsin monomers in the non-ionic detergents octyl-beta-D-glucoside and Triton X-100. FEBS Lett. 96, 322–326. Denisov, I. G., Grinkova, Y. V., Lazarides, A. A., and Sligar, S. G. (2004). Directed selfassembly of monodisperse phospholipid bilayer nanodiscs with controlled Size. J. Am. Chem. Soc. 126, 3477–3487. Denisov, I. G., Grinkova, Y. V., Baas, B. J., and Sligar, S. G. (2006). The ferrous-dioxygen intermediate in human cytochrome P450 3A4: Substrate dependence of formation of decay kinetics. J. Biol. Chem. 281, 23313–23318. Denisov, I. G., Grinkova, Y. V., McLean, M. A., and Sligar, S. G. (2007). The one-electron autoxidation of human cytochrome P450 3A4. J. Biol. Chem. 282, 26865–26873. Duan, H., Civjan, N. R., Sligar, S. G., and Schuler, M. A. (2004). Co-incorporation of heterologously expressed Arabidopsis cytochrome P450 and P450 reductase into soluble nanoscale lipid bilayers. Arch. Biochem. Biophys. 424, 141–153. Du¨zgu¨nes¸, N. (2003). Preparation and quantitation of small unilamellar liposomes and large unilamellar reverse-phase evaporation liposomes. Methods Enzymol. 367, 23–27. Goluch, E. D., Shaw, A. W., Sligar, S. G., and Liu, C. (2008). Microfluidic patterning of nanodisc lipid bilayers and multiplexed analysis of protein interaction. Lab Chip 8, 1723–1728. Grinkova, Y. V., Denisov, I. G., Waterman, M. R., Arase, M., Kagawa, N., and Sligar, S. G. (2008). The ferrous-oxy complex of human aromatase. Biochem. Biophys. Res. Commun. 372, 379–382. Heikal, A., Box, K., Rothnie, A., Storm, J., Callaghan, R., and Allen, M. (2009). The stabilisation of purified, reconstituted P-glycoprotein by freeze drying with disaccharides. Cryobiology 58, 37–44.

230

T. K. Ritchie et al.

Higgins, C. F. (2007). Multiple molecular mechanisms for multidrug resistance transporters. Nature 446, 749–757. Higgins, C. F., Callaghan, R., Linton, K. J., Rosenberg, M. F., and Ford, R. C. (1997). Structure of the multidrug resistance P-glycoprotein. Semin. Cancer Biol. 8, 135–142. Hjelmeland, L. M., and Chrambach, A. (1984). Solubilization of functional membrane proteins. Methods Enzymol. 104, 305–318. Kijac, A. Z., Li, Y., Sligar, S. G., and Rienstra, C. M. (2007). Magic-angle spinning solid-state NMR spectroscopy of nanodisc-embedded human CYP3A4. Biochemistry 46, 13696–13703. Kim, I. W., Peng, X. H., Sauna, Z. E., FitzGerald, P. C., Xia, D., Muller, M., Nandigama, K., and Ambudkar, S. V. (2006). The conserved tyrosine residues 401 and 1044 in ATP sites of human P-glycoprotein are critical for ATP binding and hydrolysis: Evidence for a conserved subdomain, the A-loop in the ATP-binding cassette. Biochemistry 45, 7605–7616. Kimura, Y., Kioka, N., Kato, H., Matsuo, M., and Ueda, K. (2007). Modulation of drug-stimulated ATPase activity of human MDR1/P-glycoprotein by cholesterol. Biochem. J. 401, 597–605. Leitz, A. J., Bayburt, T. H., Barnakov, A. N., Springer, B. A., and Sligar, S. G. (2006). Functional reconstitution of beta 2-adrenergic receptors utilizing self-assembling nanodisc technology. Biotechniques 40, 601–602, 604, 606, 608, 610, 612. Li, Y., Kijac, A. Z., Sligar, S. G., and Rienstra, C. M. (2006). Structural analysis of nanoscale self-assembled discoidal lipid bilayers by solid-state NMR spectroscopy. Biophys. J. 91, 3819–3828. Liu, R., Siemiarczuk, A., and Sharom, F. J. (2000). Intrinsic fluorescence of the P-glycoprotein multidrug transporter: Sensitivity of tryptophan residues to binding of drugs and nucleotides. Biochemistry 39, 14927–14938. Loo, T. W., and Clarke, D. M. (2005). Do drug substrates enter the common drug-binding pocket of P-glycoprotein through ‘‘gates’’? Biochem. Biophys. Res. Commun. 329, 419–422. Loo, T. W., Bartlett, M. C., and Clarke, D. M. (2003). Drug binding in human P-glycoprotein causes conformational changes in both nucleotide-binding domains. J. Biol. Chem. 278, 1575–1578. Lu, P., Liu, R., and Sharom, F. J. (2001). Drug transport by reconstituted P-glycoprotein in proteoliposomes. Effect of substrates and modulators, and dependence on bilayer phase state. Eur. J. Biochem. 268, 1687–1697. Marin, V. L., Bayburt, T. H., Sligar, S. G., and Mrksich, M. (2007). Functional assays of membrane-bound proteins with SAMDI-TOF mass spectrometry. Angew. Chem. Int. Ed. 46, 8796–8798. McDevitt, C. A., Shintre, C. A., Grossmann, J. G., Pollock, N. L., Prince, S. M., Callaghan, R., and Ford, R. C. (2008). Structural insights into P-glycoprotein (ABCB1) by small angle X-ray scattering and electron crystallography. FEBS Lett. 582, 2950–2956. Mi, L. Z., Grey, M. J., Nishida, N., Walz, T., Lu, C., and Springer, T. A. (2008). Functional and structural stability of the epidermal growth factor receptor in detergent micelles and phospholipid nanodiscs. Biochemistry 47, 10314–10323. Morrissey, J. H., Pureza, V., Davis-Harrison, R. L., Sligar, S. G., Ohkubo, Y. Z., and Tajkhorshid, E. (2008). Blood clotting reactions on nanoscale phospholipid bilayers. Thromb. Res. 122, S23–S26. Naito, M., and Tsuruo, T. (1995). Reconstitution of purified P-glycoprotein into liposomes. J. Cancer Res. Clin. Oncol. 121, 582–586.

Reconstitution of Membrane Proteins in Phospholipid Bilayer Nanodiscs

231

Nath, A., Atkins, W. M., and Sligar, S. G. (2007a). Applications of phospholipid bilayer nanodiscs in the study of membranes and membrane proteins. Biochemistry 46, 2059–2069. Nath, A., Grinkova, Y. V., Sligar, S. G., and Atkins, W. M. (2007b). Ligand binding to cytochrome P450 3A4 in phospholipid bilayer nanodiscs: The effect of model membranes. J. Biol. Chem. 282, 28309–28320. Nath, A., Koo, P. K., Rhoades, E., and Atkins, W. M. (2008). Allosteric effects on substrate dissociation from cytochrome P450 3A4 in nanodiscs observed by ensemble and singlemolecule fluorescence spectroscopy. J. Am. Chem. Soc. 130, 15746–15747. Oesterhelt, D., and Stoeckenius, W. (1974). Isolation of the cell membrane of Halobacterium halobium and its fractionation into red and purple membrane. Methods Enzymol. 31, 667–678. Orlowski, S., Martin, S., and Escargueil, A. (2006). P-glycoprotein and ‘lipid rafts’: Some ambiguous mutual relationships (floating on them, building them or meeting them by chance?). Cell. Mol. Life Sci. 63, 1038–1059. Papermaster, D. S. (1982). Preparation of retinal rod outer segments. Methods Enzymol. 81, 48–52. Polli, J. W., Wring, S. A., Humphreys, J. E., Huang, L., Morgan, J. B., Webster, L. O., and Serabjit-Singh, C. S. (2001). Rational use of in vitro P-glycoprotein assays in drug discovery. J. Pharmacol. Exp. Ther. 299, 620–628. Qu, Q., Russell, P. L., and Sharom, F. J. (2003). Stoichiometry and affinity of nucleotide binding to P-glycoprotein during the catalytic cycle. Biochemistry 42, 1170–1177. Rigaud, J. L. (2002). Membrane proteins: Functional and structural studies using reconstituted proteoliposomes and 2-D crystals. Braz. J. Med. Biol. Res. 35, 753–766. Rigaud, J. L., Levy, D., Mosser, G., and Lambert, O. (1998). Detergent removal by nonpolar polystyrene beads. Applications to membrane protein reconstitution and twodimensional crystallization. Eur. Biophys. J. 27, 305–319. Rosenberg, M. F., Callaghan, R., Modok, S., Higgins, C. F., and Ford, R. C. (2005). Three-dimensional structure of P-glycoprotein: The transmembrane regions adopt an asymmetric configuration in the nucleotide-bound state. J. Biol. Chem. 280, 2857–2862. Shaw, A. W., Pureza, V. S., Sligar, S. G., and Morrissey, J. H. (2007). The local phospholipid environment modulates the activation of blood clotting. J. Biol. Chem. 282, 6556–6563. Sligar, S. G. (2003). Finding a single-molecule solution for membrane proteins. Biochem. Biophys. Res. Commun. 312, 115–119. Takano, M., Hasegawa, R., Fukuda, T., Yumoto, R., Nagai, J., and Murakami, T. (1998). Interaction with P-glycoprotein and transport of erythromycin, midazolam and ketoconazole in Caco-2 cells. Eur. J. Pharmacol. 358, 289–294. Taylor, A. M., Storm, J., Soceneantu, L., Linton, K. J., Gabriel, M., Martin, C., Woodhouse, J., Blott, E., Higgins, C. F., and Callaghan, R. (2001). Detailed characterization of cysteine-less P-glycoprotein reveals subtle pharmacological differences in function from wild-type protein. Br. J. Pharmacol. 134, 1609–1618. Wang, E. J., Lew, K., Casciano, C. N., Clement, R. P., and Johnson, W. W. (2002). Interaction of common azole antifungals with P glycoprotein. Antimicrob. Agents Chemother. 46, 160–165. Zhao, J., Das, A., Schatz, G. C., Sligar, S. G., and Van Duyne, R. P. (2008). Resonance localized surface plasmon spectroscopy: Sensing substrate and inhibitor binding to cytochrome P450. J. Phys. Chem. C 112, 13084–13088. Zhou, S. F. (2008). Structure, function and regulation of P-glycoprotein and its clinical relevance in drug disposition. Xenobiotica 38, 802–832. Zolnerciks, J. K., Wooding, C., and Linton, K. J. (2007). Evidence for a Sav1866-like architecture for the human multidrug transporter P-glycoprotein. FASEB J. 21, 3937–3948.

This page intentionally left blank

C H A P T E R

T W E LV E

DNA-Controlled Assembly of Liposomes in Diagnostics Ulla Jakobsen and Stefan Vogel

Contents 1. Introduction 2. Probe Design 2.1. Double membrane anchor single DNA-probe design 2.2. Single membrane anchor dual-probe design 2.3. Chain length dependence 2.4. Thermal denaturation experiments 2.5. Light scattering 3. General Description of Materials and Techniques 3.1. Measurement of transition temperatures (Tm) 3.2. Experimental procedure for liposome assembly 3.3. Preparation of POPC liposomes 3.4. DNA synthesis of lipid-modified DNA conjugates 3.5. HPLC purification 4. Concluding Remarks Acknowledgment References

234 235 236 237 237 240 242 244 244 244 245 245 246 247 247 248

Abstract DNA-encoding of solid nanoparticles requires surface chemistry, which is often tedious and not generally applicable. In the presented method, noncovalent attachment of DNA is used to assemble soft nanoparticles (liposomes) in solution. This process displays remarkably sharp thermal transitions from the assembled to disassembled state, thus enabling easy and fast detection of polynucleotides (e.g., DNA or RNA), including single nucleotide polymorphisms (SNPs).

Nucleic Acid Center, University of Southern Denmark, Odense, Denmark Methods in Enzymology, Volume 464 ISSN 0076-6879, DOI: 10.1016/S0076-6879(09)64012-X

#

2009 Elsevier Inc. All rights reserved.

233

234

Ulla Jakobsen and Stefan Vogel

1. Introduction DNA-controlled assembly of solid nanoparticles requires surface chemistry to stabilize colloidal nanoparticles chemically and to avoid nonspecific non-DNA-controlled aggregation. For most solid nanoparticles, surface chemistry needs to be developed specifically for each material based on the respective chemical composition of the particles. A very broad range of different chemical compositions (e.g., metals, metal oxides, metal sulfides) and possible combinations in hybrid materials exist, but only very few general procedures are available for attachment of DNA to the respective surfaces. Many of the chemical procedures used for surface stabilization and attachment of recognition sites rely on weak bonding, such as sulfur chemistry for gold surfaces, and some of the chemical procedures have to be adjusted according to the desired particle size during the particle growing process. In light of the many difficulties and tedious practical procedures, noncovalent attachment of DNA to soft nanoparticles such as liposomes becomes an attractive technology. A number of applications such as DNA-controlled tethering to surfaces toward liposome arrays and DNA-controlled fusion of liposomes have been reported (Chandra et al., 2006; Pfeiffer and Ho¨o¨k, 2004; Stengel et al., 2007; Chan et al., 2009; Yoshina-Ishii and Boxer, 2003, Yoshina-Ishii et al., 2005; Zhang et al., 1996). Our approach describes DNA-controlled assembly of liposomes in solution and on solid supported membranes ( Jakobsen et al., 2008a). Remarkably sharp thermal transitions between assembled and disassembled state and discrimination of single mismatches, deletions, and insertions in the resulting DNA-target duplex allow applications in the diagnostics of single nucleotide polymorphisms (SNPs). The basic design is based on the noncovalent attachment of DNAprobes (single-stranded DNA with terminal lipid membrane anchors, for membrane anchor structure see Fig. 12.5) to a liposome surface (Fig. 12.1). The complementary polynucleotide target can be unmodified DNA or RNA that allows the method to be used in the detection of biological polynucleotide targets ( Jakobsen et al., 2008). The conformationally flexible single-stranded DNA-probe is presumably anchored reversibly on the liposomes since DNA strands permanently anchored at both ends of the DNA strand would not allow efficient hybridization to a complementary target sequence. After hybridization of the DNA-probe to a complementary target DNA, both ends of the corresponding duplex are not able to be simultaneously anchored into the same liposome, as this would require bending of the rigid double-stranded DNA (for mechanical properties of duplex DNA, see Cloutier and Widom, 2005). Therefore, one of the membrane anchors is released into solution and subsequent interliposomal membrane anchoring occurs and is highly favored. The interliposomal anchoring of the DNA duplex leads to

DNA-Controlled Assembly of Liposomes in Diagnostics

235

Liposome(s)

= Lipid membrane anchor

= DNA target

= DNA probe

Figure 12.1 Double membrane anchor single DNA-probe design. Schematic representation of liposome aggregation upon duplex formation between a lipophilic probe DNA and an unmodified target DNA. Liposomes and DNA strands are not drawn to scale.

liposome cross-linking and liposome assembly (Fig. 12.1). The process continues until target or probe DNA strands have been consumed since liposomes are present in excess. The required duplex rigidity is not compromised by single mismatches but multiple mismatches are not tolerated. Cooperative effects such as DNA melting and the entropically favored disassembly of liposome aggregates are presumably responsible for the sharp thermal transitions observed by UV spectroscopy measurements. The narrow temperature range for the disassembly process (thermal denaturation is compared to unmodified DNA), and nonoverlapping thermal transitions for complementary and single-mismatched sequences can be utilized to turn the liposome assembly process into a powerful detection system for SNPs. The DNA-controlled assembly of liposomes is general and efficient for liposomes of sizes between 50 and 200 nm in solution and on supported membranes. In contrast to solid nanoparticles, the lipophilic DNA-probe strand is inserted noncovalently, which avoids development of tedious conjugation chemistry and enables a rapid assembly process (assembly process requires only minutes to occur at nM and mM DNAprobe concentrations). This method is, to the best of our knowledge, the first application of liposome assembly to the detection of polynucleotides with single mismatch discrimination power in solution ( Jakobsen et al., 2008).

2. Probe Design The key concept of the method is schematically illustrated in Fig. 12.1 for a double membrane anchor single DNA-probe design and in Fig. 12.2 for a single membrane anchor dual DNA-probe design. The membrane

236

Ulla Jakobsen and Stefan Vogel

Probe A Liposomes Target Probe B

= DNA probes

= DNA target

Figure 12.2 Single membrane anchor dual-probe design. Schematic representation of liposome aggregation upon duplex formation between a lipophilic probe DNA and an unmodified target DNA. Liposomes and DNA strands are not drawn to scale.

anchoring is spontaneous upon addition of liposomes to the mixture of DNA-probe and target. The membrane anchor of the DNA-probe consists of a macrocyclic building block with at least two lipid substituents (e.g., cholesteryl or saturated alkyl chains, for chemical structures see Fig. 12.5) to ensure permanent anchoring of the DNA-probe into the liposome surface. The macrocyclic membrane anchor scaffold (Vogel et al., 2003, 2006) can be substituted with lipids at two positions and subsequently incorporated into DNA using standard automated DNA-synthesis.

2.1. Double membrane anchor single DNA-probe design The method is based on lipid-modified DNA-probes and liposomes (POPC). Introduction of lipid membrane anchors in DNA-probes has been shown earlier to increase the thermal stability of the corresponding DNA duplexes ( Jakobsen et al., 2007a,b, 2008a; Rohr and Vogel, 2006). The lipid membrane anchors are incorporated at both ends of the DNAprobe strand to enable dual anchoring of the DNA-probe into a lipid bilayer (e.g., liposomes). In order to avoid self-aggregation and surfactant properties of the corresponding DNA-probes in aqueous solution, we added up to three T-nucleosides on both ends of the DNA-probe which sufficiently suppressed self-aggregation at concentrations below 2 mM. The singlestranded DNA-probe is partitioned noncovalent into the liposome bilayer and dynamically (reversible) anchored in contrast to covalently bound DNA on solid nanoparticle surfaces (e.g., Au-surfaces, Li et al., 2002; Mirkin et al., 1996). The dynamic anchoring is a requirement to enable hybridization to a

DNA-Controlled Assembly of Liposomes in Diagnostics

237

complementary target DNA since a DNA permanently anchored on both ends of the DNA strand would not be able to hybridize with a complementary target strand. Presumably, one of the two anchors is released from the membrane bilayer while hybridizing to a complementary target strand and does not anchor into the same liposome after hybridization. The corresponding double-stranded DNA duplex is mechanically very stiff (see Cloutier and Widom, 2005) and forces one of the membrane anchors to be released into the solution since anchoring of the double-stranded DNA into the same liposome would require bending of the duplex which is energetically very unfavorable for short double-stranded DNA. The released membrane anchor is subsequently anchored into another liposome and will initiate assembly of liposomes (Fig. 12.1). The exclusive interliposomal anchoring leads to cross-linking and formation of liposome aggregates with aggregate sizes depending on DNA-probe and target concentrations.

2.2. Single membrane anchor dual-probe design Another design consists of two probes (A and B) in which both DNA sequences are complementary to half of the target sequence and feature one terminal membrane anchor for each DNA-probe strand. The dualprobe design allows encoding of two batches of liposomes with separate addition of the respective DNA-probe. The liposome assembly is subsequently initiated by addition of a target with a sequence complementary to both probes A and B (Fig. 12.2). The target bridges the DNA probes A and B, and forms a mechanically stiff duplex. The process of liposome assembly, based on a design with two DNA probes, is equally efficient compared to the single DNA-probe design, but requires two separate DNA probes and introduces a DNA duplex with two double-stranded subunits of different thermal stability. The probe design becomes more complicated since a mismatch in the thermally more stable region of the duplex may not be destabilizing enough to see a difference in the overall thermal stability of the DNA-probe:target duplex. However, using appropriate sequence design, the mismatch discrimination and remarkable sharpness of the corresponding thermal transitions equals that of the double membrane anchor single DNA-probe design.

2.3. Chain length dependence Liposome assembly depends highly on the nature of the membrane anchor (chain length and chemical structure). For saturated alkyl chains as membrane anchors, lipids of a minimal chain length of 12-carbons are required to initiate assembly of liposomes; however, the process becomes significantly more efficient when 14-carbon or 16-carbon chains are used. Membrane anchors with increased chain length of up to 20-carbon chains lead to

238

Ulla Jakobsen and Stefan Vogel

0.34 5⬘-TGTGGAAGAAGTTGGT 3⬘-TTTX2C16ACACCTTCTTCAACCACX2C16TTT

Apparent absorbance at 260 nm

0.32 0.30 0.28 0.26

5⬘-TGTGGAAGAAGTTGGT

0.24

3⬘-TTTX2C12ACACCTTCTTCAACCACX2C12TTT

0.22 5⬘-TGTGGAAGAAGTTGGT

0.20

3⬘-TTTX2C10ACACCTTCTTCAACCACX2C10TTT

0.18 0.16 20

30

40

50

60

70

80

90

Temperature (⬚C)

Figure 12.3 UV-monitored thermal denaturation data for DNA-probe target duplexes in the presence of liposomes and saturated lipid membrane anchors of different chain length.

DNA-probes that are anchored so strongly into the liposome surface that hybridization does not occur for the measured DNA-probes with a duplex length up to 17 base pairs (bp). Structurally different lipids such as conformationally less flexible cholesteryl-derived membrane anchors also result in efficient membrane anchoring but display, in contrast to 16-carbon saturated lipid chains (e.g., palmityl), sequence dependency. The efficiency of DNAcontrolled assembly with different lipid membrane anchor length is shown in Fig. 12.3 (Jakobsen and Vogel, 2008b). For practical purposes, a chain length of 16-carbons has been sufficient for the permanent anchoring of DNA-probes ranging from 9-mer to 27-mer DNA probes. The lipid anchor scaffold is based on an aza crown ether and the macrocyclic core structure has two secondary amine functions available for substitution by various lipids (Fig. 12.5). Conformational restriction by a pyridino group in the macrocycle separates the corresponding lipid chains from each other (Fig. 12.5) which may enable independent anchoring of each lipid substituent into the liposome surface. Our experiments with a single 16-carbon chain lipid membrane anchor correspond to previously published data, that two lipid anchors are required to permanently anchor DNA (20-mer duplex) into lipid membranes (Pfeiffer and Ho¨o¨k, 2004). DNA-probes with a single 16-carbon chain lipid membrane anchor and an otherwise identical

239

DNA-Controlled Assembly of Liposomes in Diagnostics

5´ O

O

O N H

O

3´

Y

Figure 12.4 Single palmitoyl-chain membrane anchor building block Y. Experiments are conducted with identical sequences except for the membrane anchor (62 nM probe DNA concentration in 110 mM Naþ, 50 -TTTYTGTGGAAGAAGTTGGTGYTTT and 30 -TTTYACACCTTCTTCAACCACYTTT with the respective complementary DNA, HEPES buffer, pH 7.0).

3´ O

P

O−

H

O X2Chol :

R=

X2C10 :

R=

N

X2C12 :

R=

O

X2C16 :

R=

X2C20 :

R=

H

H

N O

O

N

N

R

R

O 5´

Figure 12.5 General structure of the lipid-modified macrocyclic monomer and lipid substitution pattern.

sequence design have not been able to initiate liposome assembly in solution, which was attributed to insufficient membrane anchoring of the corresponding duplexes. However, thermal denaturation studies in the absence of liposomes have shown that hybridization occurs and the corresponding duplex is not destabilized by modification Y (Fig. 12.4). The lipid-substituted macrocycles (Fig. 12.5) can be incorporated into DNA by automated DNA-synthesis using the phosphoramidite approach at any desired position in a given sequence including the possibility for multiple incorporations.

240

Ulla Jakobsen and Stefan Vogel

The chemical synthesis of aza macrocycles is straightforward and can easily be upscaled to multigram amounts of the corresponding phosphoramidite building blocks (see Fig. 12.5) for automated DNA-synthesis ( Jakobsen et al., 2007a,b; Rohr and Vogel, 2006).

2.4. Thermal denaturation experiments The process of liposome assembly is reversible and can be monitored by ultraviolet (UV) spectroscopy and dynamic light scattering (DLS). Monitoring of the process by UV spectroscopy can be done at 260 nm or higher wavelengths (measured between 260 and 420 nm), although with decreasing signal intensity toward longer wavelengths. The assembly of liposomes is fully reversible and can be repeated by cycling the temperature around the thermal denaturation temperature (Tm) of the corresponding duplex. The thermal denaturation curves, as monitored by UV spectroscopy at 260 nm, are inversed compared to measurements in the absence of liposomes and of much higher intensity (2 orders of magnitude). This behavior is caused by the presence of liposomes which assemble at a temperature below the Tm of the corresponding DNA-probe:target duplex and disassemble at a temperature above the Tm. The process is monitored at 260 nm which represents the average absorption maximum for natural DNA bases but at 62 nM concentration, a concentration commonly used for the experiments with liposomes, the DNA absorption changes upon thermal denaturation will not be visible but only the changes in scattering intensity of the liposomes. The optical properties of the solution after addition of liposomes are no longer dominated by the UV absorption of the DNA but the apparent absorption caused by light scattering of the liposomes. At temperatures below Tm of the corresponding duplex, liposome assembly occurs and leads to high apparent absorbance through increased light scattering which is caused by increased average particle sizes of the liposome aggregates in solution. Heating of the liposome aggregates above the Tm of the corresponding duplex cause disassembly of the aggregates and results in strongly decreased apparent absorption due to decreased light scattering of individual liposomes compared to the liposome aggregates (Fig. 12.6). For diagnostic applications, the most important property of the DNAcontrolled liposome assembly processes are the remarkably sharp thermal transitions observed during denaturation of the corresponding duplexes. The narrow temperature range for the disassembly process of only 2–4  C compared to 15–20  C for thermal denaturation of duplexes in the absence of liposomes turns this process into a powerful method for the detection of SNPs. Thermal melting profiles with similar thermal transitions have only been reported for a nanoparticle system based on DNA-controlled assembly of gold colloids (Mirkin et al., 1996; Rosi and Mirkin, 2005; Taton et al., 2000).

241

DNA-Controlled Assembly of Liposomes in Diagnostics

Apparent absorbance at 260 nm

0.00

3⬘-TTTXACACCTTCTTCAACCACXTTT 5⬘-TGTGGAAGAAGTTGGTG

Mismatch position

−0.04

ΔT > 10 ⬚C −0.08

C:G - matched sequence

C:T C:A C:C

−0.12

Single mismatches

−0.16 20

30

40

50 60 Temperature (⬚C)

70

80

Figure 12.6 UV spectroscopy Tm data of a 17-mer DNA-probe hybridized to the corresponding single mismatches (17-mer DNA-target strands) and a fully matched target sequence.

The nonoverlapping thermal denaturation curves simplify data analysis significantly which is an important advantage over fluorophore-based detection systems and allow application of this method for the analysis of weakly discriminated mismatches in DNA-targets. The current system can detect low nM concentrations of target DNA using a label-free setup. Single mismatch discrimination data in the presence and absence of liposomes resemble closely data for unmodified DNA, which allows for traditional probe sequence design (Table 12.1). Future improvements of the currently label-free method will take advantage of the structure of liposomes which allows encapsulation of labels (e.g., fluorophores) or smaller nanoparticles (e.g., quantum dots) inside the liposome to shift the detection wavelength to the spectral range of visual light or magnetic particles to separate the target detection components from a complex sample matrix (e.g., serum). Discrimination of SNPs based on DTm-values from thermal denaturation profiles as measured by UV spectroscopy is comparable to DNA in the absence of liposomes but with much sharper transitions. However, the thermal stability of the corresponding duplexes is considerably increased in the presence of liposomes as shown in Table 12.1, since the recorded Tm-values are very similar in the absence and presence of liposomes despite a much lower DNA-probe concentration of only 62 nM compared to 1 mM in the absence of liposomes. A lower DNA-probe concentration should otherwise destabilize the duplex and result in lower

242

Ulla Jakobsen and Stefan Vogel

Table 12.1 Mismatch discrimination dataa,b Tm ( C)

DTm ( C)

Tm ( C)

Duplex

No.

1 2 3 4 a b

c d

30 TTTX2C16ACACCTTCTT CAACCACX2C16TTT

With liposomesc

Without liposomesd

50 -TGTGGAAGAAGTT GGTG 50 -TGTGTAAGAAGTT GGTG 50 -TGTGAAAGAAGTT GGTG 50 -TGTGCAAGAA GTTGGTG

48

47

–

DTm ( C)

–

36

12

37

10

38

10

38

11

36

12

37

11

X denotes the polyaza crown ether base surrogate (see Fig. 12.5). Conditions: Tm values ( C) (DTm ¼ change in Tm value) calculated relative to the DNA:DNA reference duplex measured as the maximum of the first derivative of the melting curve (A260 vs. temperature) recorded in medium salt buffer (10 mM HEPES, 110 mM Naþ, pH 7.0). 62 nM concentrations of the two complementary strands in the presence of liposomes. 1 mM concentrations of the two complementary strands in the absence of liposomes. Exp. error: 1  C.

Tm-values but this is presumably counterbalanced by the effect of the liposomes and the increased thermal stability by introduction of the lipid membrane anchor building blocks (Rohr and Vogel, 2006).

2.5. Light scattering The readout of thermal denaturation experiments based on UV spectroscopic measurements is mainly attributed to scattered light from liposomes and liposome aggregates. The confirmation of liposome assembly upon addition of a complimentary target strand has been achieved by DLS titration studies. Addition of 0.25, 0.5, 0.75, and 1 equiv. of complementary DNA below the Tm of the corresponding DNA-probe:target duplex results in a significant increase of the average particle size from 72 nm (DNA without a complementary target) to 145 nm (1 equiv. of target, Figs. 12.7 and 12.8). The increase in average particle size corresponds linearly to the increase in target DNA concentration (Fig. 12.8) and confirms that aggregation is initiated by addition of a complementary target sequence. The DLS titration is the method of choice to follow the DNA-controlled aggregation over time and shows the fast kinetics of the process (liposome aggregation within minutes). Addition of a complementary target causes liposome aggregation within less than 15 min for a 17-mer DNA-probe (50 -TTT-X-TGTGGAAGAAGTTGGTG-X-TTT:30 -ACACCTTCTTCAACCAC, X ¼ X2C16).

243

DNA-Controlled Assembly of Liposomes in Diagnostics

0.5 eq. cDNA 0.75 eq. cDNA

0.25 eq. cDNA

100

ssDNA + liposomes

Intensity

1 eq. cDNA

50

0

50

100

150 Size (nm)

200

250

300

Figure 12.7 Dynamic light scattering measurements of liposome assembly. Singlestranded DNA (ssDNA, TTT-X-TGTGGAAGAAGTTGGTG-X-TTT, X ¼ X2C16, for the chemical structure see Fig. 12.5) is titrated with complementary DNA (cDNA, 30 -ACACCTTCTTCAACCAC, 0.25, 0.5, 0.75, 1.0 equiv.). 150 140 130

Size (nm)

120 110 100 90 80 70 0.0

0.2

0.4 0.6 Complementary DNA (equivalents)

0.8

1.0

Figure 12.8 Dynamic light scattering measurements of liposome assembly. Linear fit of average particle size versus the number of equivalents complementary target.

244

Ulla Jakobsen and Stefan Vogel

3. General Description of Materials and Techniques Materials: 1-Palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC; lyophilized powder) was purchased from Avanti Polar Lipids, and solvents (HPLC grade) were purchased commercially and used without further purification. Chemicals for DNA synthesis were purchased from Glen Research or Proligo. All other chemicals were obtained from Sigma in the best quality available. Ultrapure Milli-Q water is used in all experiments. 4-(2-Hydroxyethyl)piperazine-1-ethanesulfonic acid (HEPES) buffer (10 mM HEPES, 110 mM Naþ) is prepared by mixing appropriate amounts of HEPES, HEPES sodium salt, and NaCl followed by correction of pH to 7.0.

3.1. Measurement of transition temperatures (Tm) Thermal denaturation experiments are carried out on a Perkin-Elmer UV/VIS spectrometer Lambda 30 with a PTP-6 (Peltier Temperature Programmer) device by using PE TempLab 2.0 software or a Varian Cary 3E/300 equipped with a Peltier controlled 6  6 sample changer and Cary WinUV software. Melting temperatures (Tm, in  C) are determined as a first derivative of thermal denaturation curves, which are obtained by recording absorbance at 260 nm as a function of temperature at a rate of 0.5  C/min for measurements with liposomes and of 1  C/min for measurements without liposomes. The solutions are heated to 90  C, maintained for 5 min at this temperature, and then gradually cooled before performing thermal denaturation experiments. All melting temperatures are reported with an uncertainty of 1  C, as determined from multiple experiments.

3.2. Experimental procedure for liposome assembly A particular order for the mixing of components (e.g., liposomes, DNAprobes, target strand) is not required for the experiments mentioned. In a typical experiment we use POPC liposomes with an average diameter of 65  15 nm and a lipid concentration of 10 mM, prepared by extrusion through 50 nm filters. A final concentration of 0.5 mM is used for all experiments. Addition of lipophilic probe DNA in HEPES buffer solution (pH 7.0, 110 mM Naþ, 10 mM HEPES, 62 nM DNA-probe strand) followed by addition of the target sequence (62 nM DNA target) initiates assembly. Studies at higher DNA concentrations (2 mM ) results in particle aggregation as observed visually by a rapidly increasing turbidity and finally precipitation of the liposome aggregates with fully intact liposomes as

245

DNA-Controlled Assembly of Liposomes in Diagnostics

shown by an arsenazo III dye assay ( Jakobsen et al., 2008). Experiments with lower target concentrations (target concentration < 100 nM) produce stable aggregates as observed by light scattering, but without macroscopic precipitation, even after prolonged time (72 h).

3.3. Preparation of POPC liposomes POPC is suspended in aqueous HEPES buffer at 10 mM lipid concentration with application of sonication or stirring to help the formation of a uniform suspension. Liposomes are prepared by repeated extrusion (10 times) through double polycarbonate filters with a 50 or 100 nm pore size using compressed N2 (20–40 bar) and a LIPEXTM Extruder from Northern Lipids (Olson et al., 1979).

3.4. DNA synthesis of lipid-modified DNA conjugates Synthesis of lipid-modified DNA conjugates is performed in 0.2 or 1 mmol scale on an automated DNA synthesizer using the phosphoramidite approach (Caruthers, 1991). The facile chemical synthesis results in lipophilic phosphoramidite building blocks which can be used during DNA-synthesis with a slightly modified standard protocol. The solubility of the lipid-substituted macrocycles in acetonitrile is decreased with increasing chain length which requires use of solvents such as 1,2-dichloroethane instead of acetonitrile for the lipid membrane anchor phosphoramidite building blocks (Fig. 12.9).

O NC

P N O

X2C10, X2C12, X2C16, X2C20 : R = n = 1, 3, 6, 10

N O

R

O

N

N

R

N

H ODMTr X2Chol

:R=

H

H

O

Figure 12.9 General structure and substitution pattern of lipophilic phosphoramidite building blocks for automated DNA-synthesis.

246

Ulla Jakobsen and Stefan Vogel

Prior to and after coupling of the phosphoramidite, a washing step with 1,2-dichloroethane is suggested to avoid precipitation of remaining phosphoramidite solution in the tubing of the DNA synthesizer. All of the phosphoramidite building blocks have a considerably higher molecular weight than standard nucleoside phosphoramidites (A,T,G,C) and sterically demanding substituents (e.g., cholesteryl, palmityl); therefore, prolonged coupling times (30 min) and more concentrated solutions (0.1 M instead of 0.05 M) are used to improve coupling yields and to shorten coupling times. Standard activators (tetrazole, dicyanoimidazol) lead to low coupling yields ( 80, those with open nanostructures keep growing, indicating that the more flexible and less bulky open nanostructures can be more densely packed on the lipid surface. In the inset of Fig. 13.5B, we also report the hydrodynamic radius of lipid/DNA hybrids for N ¼ 30 at different

269

Liposomes Decorated with Oligonucleotides

A

44

RH (nm)

42

40

1 day

38

36

3

4

5

6 7 8 9

2

3

4

5

6 7 8 9

10

100 Time (min)

B

60

Pseudo-hexagon Open

RH (nm)

55

50 48

45

46 44

40

42 0.4

0.8

1.2

1.6

cPOPC [M]

35 0

20

40

60

80 N

100

120

140

Figure 13.5 (A) Time dependence of the hydrodynamic radius of the lipid/DNA hexagon hybrid for two grafting densities: (●) N ¼ 9, [POPC] ¼ 0.785 mM, [ODN] ¼ 0.18 mM; (□) N ¼ 30, [POPC] ¼ 0.235 mM, [ODN] ¼ 0.18 mM. The empty circle and the filled square represent the final equilibrium values reached after 1 day. (B) Hydrodynamic radii of lipid/DNA hybrids as a function of the occupancy number, [POPC] ¼ 1.33 mM. The inset shows the behavior of the hydrodynamic radius of the hybrid as a function of liposome density for N ¼ 30. (□) Pseudohexagons, (▲) open nanostructures.

liposome densities, the results show a slight increase in the absolute value, while PDI is of the same order of magnitude. This means that liposome density is not critical for the successful formation of the complexes.

270

Martina Banchelli et al.

The samples are monitored with time, and they are stable for weeks at room temperature, demonstrating the effectiveness of this method. It is important to underline that the achievement of a thermodynamically stable system requires variable equilibration times depending on the surface grafting density. 6.3.2. Stepwise strategy Ring and open nanostructures are also built on the liposome surface by a stepwise strategy, which consists of the sequential addition of equimolar amounts of each strand to ON-Chol-decorated liposomes, in stoichiometric fashion with respect to the anchoring sites, and then recruitment by the partially built hybrid. Two stepwise procedures differing by the mixing order of the oligonucleotides, illustrated in Fig. 13.7, are adopted for the construction of the more complex close nano-object to distinguish a possible kinetic control of each step due to the different shapes of the partially built nanoconstructs. Each strategy consists of eight steps where step 1 corresponds to POPC liposomes and step 2 to the insertion of ON-Chol into the lipid membrane. The spontaneous insertion of ON-Chol and the following coupling events between complementary oligonucleotides are revealed as an increase of the radius of the liposome, due to the added hydrodynamic thickness, by DLS (Fig. 13.7). The study is performed on eight different samples for each strategy representing the composing different steps (in the asymmetric strategy for close and open nanostructures the first seven steps coincide) prepared under the same conditions: the same ODN strand is added to different step samples at the same time, and the interval between two sequential additions is set at 1 h, and the final volume is adjusted with PBS buffer. ON-Chol is added previously to the liposomes, and eventually the samples are equilibrated overnight before the sequential addition of the other strands. First, samples are prepared at different grafting densities keeping constant the lipid concentration (1.33 mM) as a function of the added oligonucleotide concentration. For the open construct, the stepwise strategy gives comparable results, in terms of the hydrodynamic radius, to those obtained with the single-step procedure in all the grafting density ranges investigated. For the construction of closed nanostructures at this lipid concentration, instead, we need to distinguish between low (N < 15) and high grafting densities: pure isolated hybrid structures have only been obtained in the former condition. In fact, for low grafting densities, both symmetric and asymmetric strategies have given comparable RH values in the final step, in agreement with those obtained with the single-step, procedure, as well as a RH trend in the intermediate construction steps consistent with the expected partially built nanoconstructs (see Fig. 13.6A). However, the symmetric strategy ends up being more effective than the asymmetric one

271

Liposomes Decorated with Oligonucleotides

A = 9 (step-by-step) = 9 (single step) = 30 (step-by-step) = 30 (single step)

44

RH (nm)

42

40

38

36

34 1

2

3

4

5

6

7

8

Step B 44

42 RH (nm)

1 day

40

38 3

4

5 6 7 8 9

2

10

3

4

5

6 7 8 9

100 Time (min)

Figure 13.6 (A) Effect of the liposome number density on the formation of lipid/DNA nanohybrids for two different POPC/DNA ratios obtained with the symmetric stepwise procedure. The radii of the nanoconstructs obtained by addition of preformed hexagons are reposted with filled symbols for comparison. Readapted from Baldelli Bombelli et al. (2009). (B) Time dependence of the hydrodynamic radius of the lipid/ DNA hexagon hybrid for two grafting densities: (●) N ¼ 9, [POPC] ¼ 0.785 mM, [ODN] ¼ 0.18 mM; (□) N ¼ 30, [POPC] ¼ 0.235 mM, [ODN] ¼ 0.18 mM.

in terms of polydispersity and grafting density range. For occupancy numbers >15, although the intermediate steps are characterized by hydrodynamic radii in agreement with the expected assemblies, the addition of the

272

Martina Banchelli et al.

38

37

RH (nm)

36

35

Asymmetric FA−2 F⬘E⬘ DE D⬘C⬘ BC B⬘A⬘ *B⬘X(open)

34

Symmetric FA−2 B⬘A⬘ BC F⬘E⬘ D⬘C⬘ DE

33

32 2

4

6

8

Step

Figure 13.7 Trend of the hydrodynamic radii for the step-by-step construction of DNA pseudohexagons and open nanostructures on liposomes for N  10. Symmetric (filled circles) and asymmetric strategies (filled squares) used for close nanostructures are showed. The open nanostructures are formed by asymmetric strategy and the first seven steps coincide with those of the asymmetric strategy for the close nanostructure (empty square). The tables on the left indicate the addition order of the oligonucleotides in the two strategies (see Scheme 13.1). Adapted from Baldelli Bombelli et al. (2009).

last oligonucleotide strand induces the formation of large aggregates with a consequent significant increase in PDI. Size distribution analysis of the autocorrelation functions for these samples does not give well-defined populations, but we rather obtain a single broad population probably composed of isolated hybrid structures, together with dimers and trimers. The fact that we do not observe the same aggregation process for the addition of the last strand in the formation of the open nanostructure, suggests that the closure of the ring is the driving mechanism for the formation of aggregates. On the basis of this hypothesis, a determining factor in the aggregation process could be a too short average distance between liposomes in solution, causing the association of neighboring hybrid structures during the ring closure. This parameter can be optimized to make the symmetric strategy effective for higher occupancy numbers also. Hence, a different series of samples is prepared according to the symmetric strategy keeping constant

Liposomes Decorated with Oligonucleotides

273

DNA concentration (0.18 mM) and varying lipid concentration between 0.235 and 0.785 mM that corresponds to occupancy numbers of 30 and 9, respectively. Aggregation is not detected upon addition of the last ODN strand for these samples and the resulting hybrid nanostructures hydrodynamic sizes and PDI are comparable to those obtained for corresponding samples prepared with the single-step procedure (Fig. 13.6A). Moreover, the hybridization process is monitored with time, as reported in Fig. 13.6B. We do not observe a significant difference, as for the singlestep procedure, in the achievement of the ‘‘equilibrium’’ state between two grafting densities and after 10 min the hydrodynamic radius is invariant with time for both samples. It is important to stress that the formation of DNA nanoassemblies in solution requires a lengthy annealing process (heating at 90  C and then cooling down to 20  C with a constant temperature gradient within 6 h), while the step-by-step procedure on liposomes is performed at room temperature. The success of the stepwise strategy highlights the important role of the immobilization of ODN on the lipid surface, working as a sort of catalyzer for the hybridization reaction. We conclude that the stepwise strategy is a sucessful method for ‘‘in situ’’ construction of more complex DNA nanostructures on the liposome surface at room temperature, but, to avoid aggregation, special care is needed regarding the liposome density in solution.

6.4. Kinetics aspects DLS is shown to be an extremely powerful tool for studying the formation of soft hybrid nanostructures composed of phospholipid liposomes decorated with oligonucleotides nanostructures. DLS analysis allows the comparison of different preparation methods to infer a final optimized protocol for the preparation of monodisperse, isolated DNA/lipid nanohybrid structures. Several parameters have to be considered in the different preparation procedures and the optimal conditions may not be the same for diverse strategies. In fact, one of the main differences between single-step and stepwise procedures seems to be the kinetics of the attainment of a stable complex. Moreover, in the stepwise strategy the determining parameter is liposome crowding, that can promote aggregation during the closure step of the ring-like structure, while in the single-step protocol a longer time scale is needed to reach the equilibrium state of the system at higher grafting density. To study the kinetics of hybridization of the determining step of the formation of these hybrids, the time course of the 260 nm absorbance after mixing the solutions containing the coupling sides is followed via stoppedflow (see Section 4.1). The time-course of the absorbance is recorded between 0.04 and 17 min (longest time measurable with this equipment).

274

Martina Banchelli et al.

A 0.400

A

0.398

0.396

0.394

0 B

5

10 t (min)

15

0.400 0.398 0.396

A

0.394 0.392 0.390 0.388 0.386 0.384 0

200

400 600 t (min)

800

1000

Figure 13.8 (A) Kinetics of ON-Chol/pseudohexagon-2 hybridization onto POPC vesicles (single-step process) at two different occupancy numbers: 9 (empty circles) and 30 (filled squares). The absorbance has been normalized to 0.4 for a better comparison. (B) Kinetics of hybridization process that occurs at the step 8 of the symmetric stepwise procedure presented in Fig. 13.5 (DE addition).

In particular, for the stepwise procedure we monitor the absorbance decrease upon addition of the DE strand to the step 7 solution obtained according to symmetric strategy (Fig. 13.8A), while for the single-step procedure a preformed DNA pseudohexagon solution is added to oligoloaded vesicles (Fig. 13.8B). The final ODN concentration is kept constant in all experiments to 0.18 mM, while the lipid concentration is set to 0.235 and 0.785 mM which corresponds to N ¼ 30 and 9, respectively. The data are reported in Fig. 13.8A and B and, since the total absorption is strongly affected by vesicle scattering, the curves have been normalized to

Liposomes Decorated with Oligonucleotides

275

achieve a better comparison between different samples. Nonetheless, a quantitative estimation of the hybridization cannot be precisely determined. In the single-step procedure, hybridization kinetics curves cannot be analyzed assuming a single process with an uniform association rate constant, but are rather composed of an initial faster dynamics followed by a slower process. Presumably, the first process is driven by the diffusion of the nanostructures approaching the lipid surface, as suggested by the superimposition of the kinetics curves in Fig. 13.8A for the first minutes of the process. The higher steric and repulsive barrier of the more crowded interface comes into play later, and determines the association rate of the second process. Of course, there are several processes involved in the formation of the DNA/lipid hybrid, and a more accurate analysis is needed to be able to model it. Unfortunately, the hybridization process is not completed by 17 min, which is the longest experimental time achievable with this experimental setup. Nevertheless, 17 min seems to be a good time scale to study the kinetics of the ring closure mechanism in the symmetric stepwise preparation upon addition of the DE strand. In this case, the hybridization is mainly composed of a single association process characterized by a faster rate constant and completed by the investigated time scale. These data are in good agreement with that observed by DLS, confirming that different kinetics of binding are involved in the determining step of the two preparation procedures. This enhances the necessity to choose the right conditions depending on the chosen methodology to succeed in the preparation of DNA/lipid hybrid structures.

7. Challenges and Perspectives DNA coupling to hard nanoparticles (e.g., AuNp) to direct their hierarchical self-assembly is a relatively mature research field (Alivisatos et al., 1996; Mirkin et al., 1996; Nykypanchuk et al., 2008). Conversely, the literature on DNA coupling to soft nanoparticles (i.e., liposomes) is still at an early stage, even if interesting and innovative applications of this procedure are now being reported in the literature (Beales and Kyle Vanderlick, 2007; Jakobsen et al., 2008; Pfeiffer and Ho¨o¨k, 2004; Stengel et al., 2007). Coupling DNA to soft nanoparticles requires a noncovalent approach, that is, a lipid modification of the oligonucleotide, and its insertion into the membrane is ruled by thermodynamics, and can therefore be altered in response to experimental conditions (temperature, concentration, salinity, and so on). Hence, the preparation and characterization of DNAdecorated liposomes is probably more challenging, but should nevertheless be pursued in view of the many envisaged uses.

276

Martina Banchelli et al.

Several key factors should be taken into account: the choice of the lipidanchoring unit, the presence and nature of a spacer toward the membrane proximal end, and, of course, the length and the composition of the oligonucleotide portion, which is more directly dictated by the purpose of the application. It is difficult to rationalize these different contributions, in view of the still limited reports in the literature. The role of the lipid composition of the membrane has not investigated in detail, but we can easily predict that the affinity of the lipid-ON conjugate will be highly dependent on this parameter. It should be mentioned, however, that whatever the long-term envisaged application, the reliable knowledge of structural parameters, such as the size of the hybrid aggregates, oligonucleotide conformation, and grafting density, are vital for the use of these objects as functional building blocks for nanostructure arrays.

ACKNOWLEDGMENTS Financial support from CSGI, MIUR-PRIN, CNR-FUSINT, and the European Commission’s Sixth Framework Program (Project Reference AMNA, Contract No. 013575) are acknowledged. Dr. Alessio Innocenti is acknowledged for help in Stopped Flow Experiment. Dr. Gabriella Caminati is acknowledged for fruitful discussions.

REFERENCES Alivisatos, A. P., Johnsson, K. P., Peng, X. G., Wilson, T. E., Loweth, C. J., Bruchez, M. P., and Schultz, P. G. (1996). Organization of ‘nanocrystal molecules’ using DNA. Nature 382, 609–611. Baldelli Bombelli, F., Gambinossi, F., Lagi, M., Berti, D., Caminati, G., Brown, T., Sciortino, F., Norde´n, B., and Baglioni, P. (2008). DNA closed nanostructures: A structural and Monte Carlo simulation study. J. Phys. Chem. B 112, 15283–15294. Baldelli Bombelli, F., Betti, F., Gambinossi, F., Caminati, G., Brown, T., Baglioni, P., and Berti, D. (2009). Closed nanostructures assembled by step-by-step ss-DNA coupling assisted by phospholipid membranes. Soft Matter 5, 1639–1645. Banchelli, M., Betti, F., Berti, D., Caminati, G., Baldelli Bombelli, F., Brown, T., Wilhelmsson, L. M., Norde´n, B., and Baglioni, P. (2008). Phospholipid membranes decorated by cholesterol-based oligonucleotides as soft hybrid nanostructures P. J. Phys. Chem. B 112, 10942–10952. Banchelli, M., Gambinossi, F., Berti, D., Caminati, G., Brown, T., and Baglioni, P. (2009a). Anchoring amphiphilic DNA to phospholipid membranes. Part II: Modulation of grafting density and conformation in vesicles. J. Phys. Chem. B submitted. Banchelli, M., Gambinossi, F., Berti, D., Caminati, G., Brown, T., and Baglioni, P. (2009b). Anchoring amphiphilic DNA to phospholipid membranes. Part I: Modulation of grafting density and orientation in supported lipid bilayers. J. Phys. Chem. B manuscript in preparation. Barrow, D. A., and Lentz, B. R. (1980). Large vesicle contamination in small, unilamellar vesicles. Biochim. Biophys. Acta 597, 92–99.

Liposomes Decorated with Oligonucleotides

277

Beales, P. A., and Kyle Vanderlick, T. J. (2007). Specific binding of different vesicle populations by the hybridization of membrane-anchored DNA. J. Phys. Chem. A 111, 12372–12380. Cevc, G. (1993). Phospholipids Handbook. Marcel Dekker Inc., New York. Chan, Y.-H. M., Van Lengerich, B., and Boxer, S. G. (2008). Lipid-anchored DNA mediates vesicle fusion as observed by lipid and content mixing. Biointerphases 3, FA17–FA21. Chan, Y.-H. M., Van Lengerich, B., and Boxer, S. G. (2009). Effects of linker sequences on vesicle fusion mediated by lipid-anchored DNA oligonucleotides. Proc. Natl. Acad. Sci. 106, 979–984. Fry, D. W., White, C., and Goldman, D. J. (1978). Rapid separation of low molecular weight solutes from liposomes without dilution. Anal. Biochem. 90, 809–815. Gao, Y., Wolf, L. K., and Georgiadis, R. M. (2006). Secondary structure effects on DNA hybridization kinetics: A solution versus surface comparison. Nucleic Acids Res. 34, 3370–3377. Gosse, C., Boutorine, A., Aujard, I., Chami, M., Kononov, A., Cogne-Laage, E., Allemand, J.-F., Li, J., and Jullien, L. (2004). Micelles of lipid-oligonucleotide conjugates: Implication for membrane anchoring and base pairing. J. Phys. Chem. B 108, 6485–6497. Henry, M. R., Stevens, P. W., Sun, J., and Kelso, D. M. (1999). Real-time measurements of DNA hybridization on microparticles with fluorescence resonance energy transfer. Anal. Biochem. 276, 204–214. Jakobsen, U., Simonsen, A. C., and Vogel, S. (2008). DNA-controlled assembly of soft nanoparticles. J. Am. Chem. Soc. 130, 10462–10463. Kalyanasundaram, K., and Thomas, J. K. (1977). Environmental effects on vibronic band intensities in pyrene monomer fluorescence and their application in studies of micelar systems. J. Am. Chem. Soc. 99, 2039–2044. Koppel, D. E. (1972). Analysis of macromolecular polydispersity in intensity correlation spectroscopy: The method of cumulants. J. Chem. Phys. 57, 4814–4820. Mirkin, C. A., Letsinger, R. L., Mucic, R. C., and Storhoff, J. J. (1996). A DNA-based method for rationally assembling nanoparticles into macroscopic materials. Nature 382, 607–609. Nykypanchuk, D., Maye, M. M., Van der Lelie, D., and Gang, O. (2008). DNA-guided crystallization of colloidal nanoparticles. Nature 451, 549–552. Pfeiffer, I., and Ho¨o¨k, F. (2004). Bivalent cholesterol-based coupling of oligonucleotides to lipid membrane assemblies. J. Am. Chem. Soc. 126, 10224–10225. Stengel, G., Zahn, R., and Hook, F. (2007). DNA-induced programmable fusion of phospholipid vesicles. J. Am. Chem. Soc. 129, 9584–9585. Stewart, J. C. M. (1980). Colorimetric determination of phospholipids with ammonium ferrothiocyanate. Anal. Biochem. 104, 10–14. Tumpane, J., Sandin, P., Kumar, R., Powers, V. E. C., Lundberg, E. P., Gale, N., Baglioni, P., Lehn, J.-M., Albinsson, B., Lincoln, P., Wilhelmsson, L. M., Brown, T., et al. (2007). Addressable high-information-density DNA nanostructures. Chem. Phys. Lett. 440, 125–129. Wetmur, J. G., and Davidson, N. (1968). Kinetics of renaturation of DNA. J. Mol. Biol. 31, 349–370. Zana, R. (1987). Surfactant Solutions: New Methods of Investigation. Marcel Dekker, New York.

This page intentionally left blank

C H A P T E R

F O U R T E E N

Synthesis, Characterization, and Optical Response of Gold Nanoshells Used to Trigger Release from Liposomes Guohui Wu,* Alexander Mikhailovsky,† Htet A. Khant,* and Joseph A. Zasadzinski*

Contents 1. Introduction 2. Synthesis of HGNs 3. Optimization of HGN Dimensions for Maximum Absorption in the NIR 4. HGN Response to Femtosecond NIR Laser Pulses 5. Coupling HGN to Liposomes 5.1. Pulsed laser optics 5.2. Continuous-wave laser irradiation 6. Liposome Disruption and CF Release Due to Pulsed Laser Irradiation 7. Mechanism of Triggered Liposome Release 8. Effect of Proximity of HGNs to Liposomes 9. Conclusions Acknowledgments References

280 283 286 290 294 298 299 299 300 303 304 304 304

Abstract Liposomes show great promise as intravenous drug delivery vehicles, but it is often difficult to combine stability in the circulation with rapid, targeted release at the site of interest. Targeting to specific tissues requires developing highly specific ligands with strong affinities to receptors overexpressed on diseased cells; a new cellular target requires developing new ligands and identifying new receptors. Novel photoactivated, hollow, gold nanoshell (HGN)/liposome composites provide a new approach to both controlled * {

Department of Chemical Engineering, University of California, Santa Barbara, California, USA Department of Chemistry, University of California, Santa Barbara, California, USA

Methods in Enzymology, Volume 464 ISSN 0076-6879, DOI: 10.1016/S0076-6879(09)64014-3

#

2009 Elsevier Inc. All rights reserved.

279

280

Guohui Wu et al.

release and specific targeting. HGN are extremely efficient near infrared (NIR) light absorbers, and are not susceptible to photobleaching like conventional dyes. Near-complete liposome contents release can be initiated within seconds by irradiating HGNs with an NIR pulsed laser. Targeting the drug is limited only by the dimensions of the laser beam; no specific ligands or antibodies are required, so different tissues and cells can be targeted with the same HGN/liposomes. HGNs can be encapsulated within liposomes or tethered to the outer surface of liposomes for the most efficient drug release. HGNs in liposome solutions can also trigger release, but with lower efficiency. Drug release is induced by adsorbing femto- to nanosecond NIR light pulses that cause the HGNs to rapidly increase in temperature. The resulting large temperature gradients lead to the formation of vapor microbubbles in aqueous solutions, similar to the cavitation bubbles induced by sonication. The collapse of the unstable vapor bubbles causes liposome-membrane rupture and contents release, with minimal damage to the surroundings, and little overall heating of the solution.

1. Introduction The therapeutic efficacy of many drugs can be improved by maximizing their concentration at the disease site; toxicity can be reduced simultaneously by lowering the concentration elsewhere in the body. Liposomes and other lipid-based drug carriers sequester toxic drugs within a lipid membrane to provide significant advantages over systemic therapy by altering drug biodistribution, maximizing efficacy, while minimizing damage to healthy organs and tissues (Allen and Cullis, 2004; Sengupta et al., 2005). Submicron liposomes and other lipid-based nanocarriers can remain in the bloodstream for extended periods allowing for accumulation in regions of tumor growth or inflammation due to the poorly formed and leaky vasculature, a mechanism known as the enhanced permeation and retention effect (EPR) (Allen and Cullis, 2004). However, it is difficult for a given liposome to combine the necessary physical integrity and drug retention in circulation (to maximize drug accumulation at the disease site) with rapid contents release at the disease site (to affect therapy and minimize drug resistance). For example, liposomal doxorubicin (a chemotherapy drug) reduces drug-related toxicity; however, its therapeutic activity is reduced despite its efficient delivery to tumors because of slow release from the liposome carriers (Abraham et al., 2005). Therefore, one of the current challenges for liposomes and other carriers is how to initiate the release of encapsulated drugs with both spatial and temporal controls. External signals such as ultrasound (Huang and MacDonald, 2004) and visible light (Mueller et al., 2000; Shum et al., 2001) have been used to induce contents release from liposomes, but these methods are limited to surface-accessible areas

Synthesis, Characterization, and Optical Response

281

such as the eye and skin. For delivery to deeper tissues, other strategies have emerged, ranging from liposomes sensitized to general hyperthermia (Ponce et al., 2006), and receptor-targeted (Noble et al., 2004), and pH- or enzymatically triggered liposomes (Davidsen et al., 2003; Simo˜es et al., 2006). It is difficult, however, to incorporate a destabilizing agent into the liposome membrane to promote release without compromising long-term stability and drug retention in the circulation; by contrast, liposomes optimized to be robust and resistant to leakage in the circulation are hampered by suboptimal drug release. Active targeting requires specific ligands with high affinities to receptors overexpressed on diseased cells that can lead to ‘‘bindingsite barriers’’ where the tightly bound nanocarriers prevent drug penetration into the tissue (Peer et al., 2007). In addition, targeting a different site requires identifying an appropriate receptor as well as the synthesis and characterization of new ligands. To address the joint challenges of controlled release and specific targeting, we coupled hollow gold nanoshells (HGNs) that strongly absorb near infrared (NIR) light, to remotely trigger content release from conventional liposomes and ‘‘vesosomes’’ (multicompartment lipid-based carriers, or larger liposomes encapsulating multiple smaller liposomes, Kisak et al., 2004) within seconds, using an external, pulsed laser source. The HGNs can be tethered chemically to the liposome surface, encapsulated within the liposomes, or even just be in solution with the liposomes. Absorbing pulsed laser light causes the HGNs to rapidly increase in temperature, leading to the formation of microscopic vapor bubbles (Huang et al., 2006; Tong et al., 2007) in the vicinity of the liposome bilayer, the collapse of these bubbles causes transient liposome membrane rupture and contents release (Wu et al., 2008). The effects on the liposome membrane are similar to those induced by ultrasound-induced cavitation; sonication is a commonly used method to create small, unilamellar vesicles from a lamellar dispersion and is well known to disrupt bilayer membranes. In addition to membrane disruption and fast contents release, only HGN/liposome complexes directly irradiated by the laser are ruptured, providing the necessary spatial control of contents release. Alternatively, continued irradiation of the HGNs can induce localized hyperthermia or permeabilized cell bilayers, both of which can promote drug uptake by cells, or even lead directly to cell death (Chen et al., 2007; Hirsch et al., 2003; Norman et al., 2008; Tong et al., 2007). The great advantage of NIR to activate liposome drug-release is that tissue, blood, etc. are relatively transparent to 700–1100 nm wavelength light, allowing penetration depths of several centimeters (Weissleder, 2001). Gold nanostructures designed to have a plasmon resonance at NIR wavelengths, that is, silica core/gold nanoshells (Hirsch et al., 2003; Prasad et al., 2005), gold nanorods (Huang et al., 2006; Norman et al., 2008), and HGNs or nanocages (Chen et al., 2007; Prevo et al., 2008; Sun et al., 2003) are especially effective at absorbing NIR light and converting this energy into

282

Guohui Wu et al.

heat. HGNs are similar to silica core/gold nanoshells that have been used both in vitro and in vivo to accumulate NIR light (Hirsch et al., 2003), except that HGNs have a hollow core, which allows easier synthesis, (Prevo et al., 2008) smaller overall dimensions (Prevo et al., 2008; Sun et al., 2003), and no silica to interact with tissues. Gold nanoshells and nanorods illuminated with NIR light have been used successfully to noninvasively heat and eradicate diseased cells and tissues both in vivo and in vitro (Chen et al., 2007; Hirsch et al., 2003; Huang et al., 2006; Norman et al., 2008; Tong et al., 2007). Heat-transfer analysis confirms experimental observations that absorption of nano- to femtosecond pulses of NIR light causes the temperature of the HGNs to reach the melting temperature of gold, causing the collapse of the nanoshells into solid nanospheres (Prevo et al., 2008; Wu et al., 2008). The conversion of the optical energy into heat is so fast (nanoseconds) that thermal energy dissipation to the surrounding fluids occurs after the HGN reaches its maximum temperature (Link et al., 1999a). As the high-temperature HGNs equilibrate with the surrounding fluid, large-temperature gradients induce the boiling of microscopic amounts of water within microseconds (Lapotko et al., 2006; Lin and Kelly, 1998; Wu et al., 2008). These microbubbles are unstable, and the large volume of cold water surrounding the bubbles causes them to collapse in the same way as sonication-induced cavitation bubbles. The bubble collapse induces mechanical stresses in the surrounding fluid, that can tear lipid membranes apart, thereby releasing the contents of liposomes or other lipid-based drug carriers (Wu et al., 2008). The solution returns to equilibrium less than a millisecond after the initial laser pulse (Wu et al., 2008). Depending on the laser power and the pulse repetition rate, the average solution temperature is increased by no more than a few degrees (Prevo et al., 2008). There is a minimum energy threshold for drug release and a characteristic acoustic response of the solution on irradiation (Wu et al., 2008), similar to sonication-induced cavitation (Lin and Kelly, 1998). Neither the liposomes nor their contents are degraded chemically by the irradiation and release. The potential advantages of this new photoactivated release include (Sengupta et al., 2005) (i) synergistic disease-cell targeting by combining drug-carrying particles (liposomes) and energy-absorbing particles (HGNs) (Allen and Cullis, 2004), (ii) localizing release without harmful effects on surrounding healthy tissues, with no cytotoxicity or cutaneous photosensitivity as in photodynamic therapy, since the gold nanoparticles are inert (Abraham et al., 2005), (iii) triggering up to several centimeters inside the body as most tissues are transparent to NIR light (Huang and MacDonald, 2004), and (iv) creating high-localized concentrations of drug with both spatial and temporal controls. A variety of liposome or polymeric (Discher et al., 1999) carriers could be modified by tethering or encapsulating HGNs to produce a system for rapid, targeted release on demand via NIR irradiation. In

283

Synthesis, Characterization, and Optical Response

addition, HGN-induced liposome disruption could be used to induce rapid diffusional mixing to permit the study of fast chemical kinetics in nanoenvironments mimicking cell membranes (Chiu et al., 1999).

2. Synthesis of HGNs

H )n 2 CH 2O

S-(CH2CH2O)n-H

S-( CH

2 CH 2 O) n -H

C S-(

-H O) n

2

NIR

CH

S-(

-H O) n

2

CH

2

CH

S-(

CH

Au NP

2 CH 2 O) n -H

-H O) n

800 nm

2

CH H2

-H O) n

2

CH

S-(

S-( 2 CH 2 O) n -H

S-(CH 2CH2O)n-H

C S-(

2 CH 2 O) n -H

CH

CH

2

S-( S-( CH

CH

2

CH

2

-H O) n

Au NS

CH

2

S-(CH2CH2O)n-H

H-(OCH2CH2)n-S

Ag

CH

S-(

S-(

PEG750-SH S-(CH2CH2O)n-H

+Au+3

CH H2

2

-H O) n

The templated galvanic replacement reaction of silver for gold (Chen et al., 2007; Hao et al., 2004; Liang et al., 2005; Prevo et al., 2008; Schwartzberg et al., 2006; Sun and Xia, 2004; Sun et al., 2002; Wiley et al., 2004, 2005) provides a simple and reproducible, nontoxic route to HGNs 20–50 nm in diameter for use in biomedical applications (Scheme 14.1). Silver nanoparticles with diameters of the desired HGN core size are synthesized first, then are sacrificed by adding a gold salt to the solution; the gold is reduced to metal because it has a greater standard reduction potential than the silver template, which is oxidized to a molecular solution (Sun et al., 2002). The gold plates onto the outside of the dissolving silver nanoparticle, resulting in an HGN of controlled diameter; the shell thickness is determined by the relative amount of gold salt to the silver template. The ratio of shell diameter to shell thickness governs the wavelength of the HGN absorbance (Oldenburg et al., 1998); the surface plasmon resonance (SPR) absorbance of nanoshells made in this fashion can be tuned very simply by applying Turkevich’s basic colloidal growth chemistry to the sacrificial silver nanoparticles (Link et al., 1999a; Turkevich, 1985; Turkevich et al., 1951, 1954). The emphasis here is not on making shape-specific or extremely monodisperse nanoshells, but rather on a simple and scalable route to nanoshells for practical applications, with tunable sizes and absorbance profiles that require minimal experimental footprints (reduced heating, minimal separations, etc.), and minimal exposure to toxic solvents, reagents, or intermediates that would be detrimental to biomedical applications. The major benefit of this synthesis is that it is

Scheme 14.1 Schematic illustration of the synthesis and stabilization of gold nanoshells as well as their structural change after NIR laser irradiation.

284

Guohui Wu et al.

rapid, stable, highly scalable, and in many cases is a true ‘‘one-pot’’ synthesis (Hao et al., 2004; Liang et al., 2005; Sun and Xia, 2004). Silver templates are prepared at 60  C in a well-stirred 600 ml solution of 0.2 mM silver nitrate (AgNO3; Fisher Scientific, Atlanta, GA) with 0.6 ml 1.0 M sodium borohydride (NaBH4; Fisher Scientific) in the presence of 0.5 mM sodium citrate ( J.T. Baker Chemical Co., Phillipsburg, NJ). The solution is stirred for at least 2 h to allow the NaBH4 to fully hydrolyze. The addition of sodium borohydride accelerates the chemical reactions, and the resulting nanoparticles are 15–25 nm in diameter (Prevo et al., 2008). After cooling to room temperature, larger silver nanoparticles could be grown from these stock sols, if desired. Silver particle growth is initiated by adding 0.5 ml of 2.0 M hydroxylamine hydrochloride solution (NH2OH
HCl, Aldrich, Milwaukee, WI) to the silver sol, followed by stirring for 5 min (Turkevich, 1985; Turkevich et al., 1951, 1954), addition of 1.25 ml 0.1 M AgNO3 (0–1 ml), and stirring overnight. The growing silver nanoparticles turn the sol a darker yellow or orange, depending on the amount of additional AgNO3. Gold nanoshells could then be made via galvanic replacement chemistry from the template silver sols without the need to isolate the silver nanoparticles. First, a given silver sol (50 ml) is heated to 60  C and the necessary amount (3.2 ml for the nanoshells in Fig. 14.1) of 25 mM tetrachloroauric acid (HAuCl4, Aldrich) is added dropwise (depending on the initial silver template size). Silver (Agþ/Ag 0.8V, vs. SHE) has a lower redox potential than gold (AuCl4 /Au 0.99V, vs. SHE) and the replacement reaction is (Sun et al., 2002): 3AgðsÞ þ AuCl4ðaqÞ ! AuðsÞ þ 3Agþ ðaqÞ þ 4ClðaqÞ Upon the addition of the concentrated HAuCl4, the solution turns from yellow/orange to gray/yellow to blue/gray to blue/turquoise within seconds as the silver and gold are oxidized and reduced, respectively. The reactions are monitored using UV/vis/NIR spectroscopy, and stopped when the silver peak located near 400 nm vanishes (usually within a few minutes, although the reaction mixture is generally stirred for at least 1 h after gold addition). This which occurs when the gold/silver ratio in the reaction vessel approaches the stoichiometric ratio of 1:3 (to err on the side of completion, the ratio is usually 37:1). Once the reaction is complete, the samples are cooled, silver chloride is allowed to precipitate, and the supernatant containing the gold nanoshells are transferred to another vessel and stored at 4  C until further use. The size distribution and particle morphology are analyzed by transmission electron microscopy using an FEI Tecnai T20 microscope (Fig. 14.1). A typical HGN is spherical and hollow, although some have irregular shapes or incomplete shells. This particular set of concentrations gave HGNs of diameter 33  13 nm and shell

Synthesis, Characterization, and Optical Response

285

Figure 14.1 TEM images of HGNs with a variety of morphologies. All of the nanoshells have a hollow core, that is, the light gray area near the center of each nanoshell, surrounded by the gold shell, which is dark gray to black. Some nanoshells are not complete (arrow). Bar is 50 nm.

thicknesses of 2  0.9 nm, and had a maximum adsorption around 820 nm (see Fig. 14.7) using a Jasco V-530 UV/vis spectrometer ( JASCO Corp., Tokyo). The synthesized HGNs are quite stable in the dilute synthesis solutions because of electrostatic repulsion caused by the adsorption of citrate ions, resulting in an overall negative charge on the HGN. However, to improve their stability against aggregation in physiological buffers and other high-ionic strength solutions, the HGNs are further stabilized sterically by tethering poly(ethylene glycol) (PEG) of molecular weight 750 Da to the HGN via a thiol linker. Methoxypolyethylene glycol amine (750PEG-NH2, PEG molecular weight of 750 Da, Aldrich) is converted to methoxypolyethylene glycol thiol (750PEG-SH), using a twofold molar excess of 2-iminothiolane
HCI (also known as Traut’s reagent; Sigma-Aldrich, St. Louis, MO) in buffer (2.28 mM Na2HPO4, pH 8.8). One-half milliliter of the prepared 0.0379 M 750PEG-SH is added to 600 ml of the HGN to achieve a 1000:1 ratio of thiol:gold. The PEG-stabilized HGN solution is centrifuged at 21,000g for 30 min and redispersed in Milli-Q water twice to remove any unattached and/or soluble chemicals. The pellet readily redisperses in

286

Guohui Wu et al.

buffer. Complete gold nanoshells and nanorods with various coatings can also be purchased directly from Nanopartz Co. (Salt Lake City, Utah). In our case, the PEG-modified HGNs are stable for at least 1 year in physiologic buffers (longer term stability is still being tested). No flocculation is observed after mixing the HGN solution with concentrated carboxyfluorescein (CF) in buffer at basic pH; severe precipitation occurs instantly for HGNs without 750PEG-SH. For in vivo applications, PEG coatings can minimize recognition by the immune system and minimize nonspecific binding to blood proteins and cells. Similar PEG-modified gold nanoparticles show little cytotoxicity in vitro (Niidome et al., 2006). After intravenous injection, PEG-modified gold nanoparticles circulate in mice with a halflife of approximately 1 h, and, for at least 72 h, there is no accumulation in major organs except for the liver (Niidome et al., 2006). The free methoxy end of the 750PEG-SH can also be modified to display various ligands or antibodies to allow the HGNs to be tethered to various lipids or proteins. Unlike gold nanorod syntheses, this approach does not require the shape-determining detergent hexadecyltrimethylammonium bromide (CTAB), which is cytotoxic. CTAB adsorbs to gold very strongly and it is difficult to replace CTAB with other ligands; removal of CTAB can result in nanoparticle aggregation (Niidome et al., 2006). An additional advantage of this galvanic replacement synthesis is the small size of the HGNs, which is especially useful when enclosing HGNs within liposomes for controlled release. In comparison, the silica core/gold shell nanoparticles pioneered by Hirsch et al. (2003) are at least 100 nm in diameter, and the dielectric core material (either silica or polystyrene) brings the additional concern of potential biological effects of silica or the products of silica degradation. An added benefit is that, compared with the one-pot synthesis of HGNs, the silica core/gold shell nanoparticle synthesis is time-consuming and laborintensive (Shi et al., 2005).

3. Optimization of HGN Dimensions for Maximum Absorption in the NIR An idealized HGN structure allows for analytical predictions of light absorption and scattering by means of classical field theories, such as the Mie scattering formalism (Kreibig and Genzel, 1985). Analytical solutions exist for the interaction of light with spherical metal nanoparticles, nanoshells, and nanorods, that is, structures possessing a high degree of symmetry. The optical properties of less symmetric particles or aggregates require numerical solutions of Maxwell’s equations. For spherical shell nanoparticles such as HGNs, analytical solutions for the far-field extinction, absorption, and scattering cross sections have been known for almost a century and can be found, for

287

Synthesis, Characterization, and Optical Response

example, in the classic book by Bohren and Huffman (1983). These cross sections are related to the attenuation of photons, N, according to N =N0 ¼ expð nsext xÞ, in which sext is the HGN extinction cross section, n is the number/volume of HGN, and x is the path length of light with initial number of photons, N0. The extinction cross section is the sum of the scattering cross section, ssca, and absorption cross section, sabs, of the incident light. To model the absorption and scattering of HGNs, analytical solutions of Mie scattering for core/shell nanoparticles can be used as described by Bohren and Huffman (1983). The dielectric function (or complex index of refraction) for gold is taken from various sources ( Johnson and Christy, 1972), but similar results are obtained for all sets of material properties. The extinction efficiency, Qext, is the ratio of the extinction cross section to the geometric cross section: Qext ¼ sext/pa2, a is the HGN radius. The absorption efficiency, Qabs, is the difference between Qext and Qsca, the scattering efficiency: Qabs ¼ Qext Qsca ¼ sabs/pa2. Ideally, all of the radiation would be adsorbed by the HGN (Qabs >> Qsca) to provide the maximum HGN heating. The light scattered by the HGNs is lost for useful purposes. Figure 14.2 shows the absorption and scattering efficiencies as a function of wavelength for an idealized HGN with an overall diameter of 33 nm and a shell thickness of 1.7 nm. For wavelengths around 800 nm, absorption dominates scattering for these HGNs, suggesting that almost all of the laser energy is being converted to heat. The peak value of the absorption cross section for an HGN with these dimensions is 9.12  10 11 cm2. Figure 14.3 shows the calculated optimal shell thickness d as a function of the HGN diameter D. HGNs with these dimensions exhibit the maximum

12

Efficiency

10 Scattering Absorption

8 6 4 2 0 400

600 800 Wavelength (nm)

1000

Figure 14.2 Theoretical absorption and scattering efficiency for HGN with a diameter of 33 nm and shell thickness of 1.75 nm in water.

288

Guohui Wu et al.

7

Shell thickness (nm)

6

5

4

3

2

20

40

60 80 Particle diameter (nm)

100

Figure 14.3 Optimal shell thickness providing the maximum absorption efficiency at 800 nm as a function of HGN diameter.

absorption at 800 nm (optimal wavelength for a Ti:Sapphire laser). This result shows that the smaller the HGN, the thinner the gold shell must be to insure a maximum adsorption in the NIR range of the optical spectrum. For example, a 40-nm diameter HGN should have a shell thickness of 2 nm, while an 80-nm diameter HGN should have a shell thickness of  5 nm to maximize the absorption at 800 nm. Figure 14.4 shows the calculated size-dependent absorption efficiency, Qabs, and absorption cross section, sabs at 800 nm. The absorption efficiency peaks for HGNs with D  50 nm, but the absorption cross section peaks for D  85 nm. For a given mass of gold, the HGN with the highest efficiency should be chosen as the mass/HGN increases faster with HNG diameter than does the optical cross section. The peak absorption cross section is 4  10 10 cm2. For an optical energy density of 2.2 mJ/cm2 corresponding to the cavitation threshold, HGN of any size will be heated well above the melting point of bulk gold (Prevo et al., 2008). Smaller HGN will be heated to higher temperatures due to their smaller mass, but particles with D  85 nm will absorb the maximum energy/particle (Prevo et al., 2008). Thus, it is difficult to predict a priori which HGN size is optimal for a given application. While the direct implementation of Mie theory does a good job of predicting the wavelength at which the maximum absorbance occurs for the experimentally determined HGN size distribution (Fig. 14.5), it fails to

289

Synthesis, Characterization, and Optical Response

14

12 300 10 200 8 100

Absorption cross-section (cm2)

Absorption efficiency

400 ⫻ 10−12

6 20

40

60 80 HGN diameter (nm)

100

Figure 14.4 Absorption efficiency (red) and cross section (blue) at 800 nm of HGNs with optimized geometry as a function of diameter.

describe the inhomogenously broadened extinction spectrum of the HGN ensemble (Fig. 14.6). The statistical distribution of shell thickness and HGN diameter are determined from the TEM micrographs and fit by a lognormal distribution function (Fig. 14.5). The extinction cross section of the HGN ensemble has been modeled using these distribution functions. Figure 14.6 shows that the simulation predicts less broadening of the absorption spectra than what is observed experimentally. The experimental broadening may be due to the shape polydispersity, as well as possible chemical variations due to any residual silver alloy in the HGN. An additional complication is that the shell thickness in the HGNs is significantly less than the electron mean free path in gold (50 nm), and the dielectric function can be strongly affected by the scattering of electrons at the nanoparticle boundaries. To address this effect a size-dependent term is introduced into the dielectric function of gold according to Alvarez et al. (1997) and the simulation optimized numerically to achieve the best match between the experimental extinction spectra and the theoretical model. The use of the size-dependent dielectric function does not resolve completely the discrepancy between the experimental and theoretical extinction spectra. This may be explained by the fact that thin layers on the metal surface may be depleted/enriched in electron density (also the silver alloying), depending on the properties of the environment, which can change the effective dimension of the HGN (Kreibig, 1995).

290

Guohui Wu et al.

200

Occurence

150

100

50

0 0

20

40 60 Particle diameter (nm)

80

Occurence

15

10

5

0 0

1

2 3 4 Shell thickness (nm)

5

Figure 14.5 Measured diameter (left) and shell thickness (right) from TEM images for HGNs synthesized as described in text.

4. HGN Response to Femtosecond NIR Laser Pulses The HGNs prepared as discussed above have a broad absorption peak at 780–820 nm (Figs. 14.6 and 14.7), which overlaps well with the Ti: Sapphire pulsed laser (Spectraphysics Spitfire), which has an FWHM of 12 nm centered around 800 nm. Compared with conventional dyes, HGNs are much stronger than NIR absorbers and are less susceptible to photobleaching or other forms of chemical degradation. However, sufficient irradiation does lead to collapse of the HGN into solid gold

291

Synthesis, Characterization, and Optical Response

1.4

Absorbance (a.u.)

1.2 1.0 0.8

Experiment Simulation

0.6 0.4 0.2 0.0 400

500

600 700 800 Wavelength (nm)

900

1000

Figure 14.6 Measured (red) and calculated (blue) extinction spectra of the experimental distribution of HGN diameters and shell thicknesses from Fig. 14.5.

nanoparticles; this collapse leads to a loss of absorption in the NIR (Fig. 14.3). The original color of the HGN suspensions is dark blue (Fig. 14.7); after pulsed NIR laser irradiation at 800 nm at a power density of 16.1 mJ/cm2 for 8 min (120fs pulses at 1 kHz repetition rate), the HGN suspension turns a dark red color. The corresponding UV–VIS absorption spectra exhibit a shift of the absorption peak from 780 to 470 nm, indicating the conversion of the HGN to solid nanoparticles (Prasad et al., 2005; Prevo et al., 2008). TEM images of irradiated particles confirm the change in the HGN morphology; the hollow center of the nanoshell collapses and the particles anneal into the more stable solid spheres, consistent with the change in color and the shift of the adsorption maxima (Fig. 14.7). These observations confirm that HGNs reach sufficiently high temperatures after femtosecond pulses of NIR light to melt and anneal into more stable shapes. For a solid-core, spherical Au particle with size from 2 to 100 nm, the SPR is from 520 to 570 nm (Kreibig, 1977). For other gold morphologies, such as rods, cages, thin plates, and aggregates, the absorption peak due to the SPR shifts to lower energy (higher wavelength) in the NIR window. The peak at 470 nm can be explained by the formation of Au–Ag alloy nanoparticles during the laser-induced heating. Ag has an SPR at 400 nm; the plasmon absorption of Au–Ag alloy falls between the SPRs of Ag and Au nanoparticles and varies linearly with the Au mole fraction (Link et al., 1999b). X-ray photoelectron spectroscopy confirms the presence of residual silver in the HGN (Prevo et al., 2008). A second explanation for the 470-nm peak is the presence of Au clusters less than 2 nm in size

292

Guohui Wu et al.

2 NIR

Extinction (a.u.)

1.5

1

0.5

0

400

600 800 Wavelength (nm)

1000

Figure 14.7 Extinction spectra (normalized at 350 nm) of: (blue) HGN suspension; (red) HGN suspension laser-irradiated at a power density of 16.1 mJ/cm2 for 8 min (120 fs pulses at 1-kHz repetition rate). Below the peaks of extinction curves are photographs of the corresponding dispersions. The spectra are consistent with the collapse and annealing of the HGN into solid gold/silver alloy nanoparticles.

due to laser-induced particle heating and fragmentation (Link et al., 1999b). The existence of such fine Au particles has also been postulated to explain the similar absorption peak at 460 nm of Au/SiO2 after ultrasound treatments (Link et al., 1999b). Rapid nonradiative relaxation processes convert the absorbed light energy mainly to heat (Grua et al., 2003). Petrova et al. (2006) found that at temperatures higher than 250  C, nanorods anneal into spherical particles in less than 1 h; we have found that HGN also anneals into solid particles within hours at 250  C. This is significantly below the melting point of bulk gold (1064  C). Time-resolved spectroscopy studies, however, showed that nanorods (in aqueous solution) maintained their shape in spite of a rapid lattice temperature increase to 1000 K achieved by ultrafast pulsed laser excitation. Petrova et al. (2006) explained that ‘‘the difference in the temperature stability of the nanorods under continuous thermal heating compared to laser-induced heating is attributed to thermal diffusion: the rods do not stay hot for long enough after ultrafast excitation for significant structural transformation to occur.’’ On the other hand, the melting of HGN suggests that the structural changes in HGN after femtosecond laser-induced heating requires that the temperature in the HGN gold lattice must be significantly higher than 1000 K. The amount of rearrangement required to collapse an HGN into a solid sphere is also much less than that required to change a solid gold nanorod to a sphere.

Synthesis, Characterization, and Optical Response

293

The maximum temperature reached by the HGNs is also limited by the rate of energy dissipation into the surrounding liquid. However, the thermal diffusivity of water limits the conduction of heat from the HGNs, and heat dissipation from the HGN to the surrounding water is much slower (microseconds) than the electron dynamics involved in plasmon-mediated heating (nanoseconds) (Link et al., 1999a; Prasad et al., 2005; Prevo et al., 2008; Roper et al., 2007). Essentially, all of the optical energy input to the HGN goes into heating the HGN (Prevo et al., 2008). The high temperature HGN then dissipates its thermal energy into the surrounding water by conduction; large temperature gradients are generated in the vicinity of the HGN, which cause the surrounding water to boil and form rapidly growing microbubbles. These microbubbles cannot grow indefinitely as there is not sufficient energy within the HGN to raise the bulk of the solvent to the boiling temperature (Prevo et al., 2008). The bubbles become unstable and can undergo a violent collapse which produces shock waves (Pecha and Gompf, 2000) or microjets (Popinet and Zaleski, 2002). Recent time-resolved X-ray scattering experiments show that femtosecond laser excitation of gold nanoparticles leads to the compression of the solvent, which is consistent with bubble formation (Kotaidis et al., 2006). The growth and collapse of unstable vapor bubbles also produces a detectable pressure change in the bulk solution, which gives a photoacoustic signal that can be ‘‘heard’’ by a hydrophone (Model #TC4013, Reson, Goleta, CA). The bandwidth of the hydrophone is 1 Hz–170kHz. The hydrophone, which is 0.5 cm in diameter and  2 cm long, is immersed into the HGN solution 5 mm above the laser beam in a quartz cuvette with 10-mm light path. The solution volume is 2.5 ml. The sample is not stirred during the laser irradiation to minimize the acoustic noise. Data collection is synchronized with the laser cavity dumping event by using triggering signal from the laser control electronics, and the output of the hydrophone is collected by a digital oscilloscope (Tektronix TDS5032B). The acoustic transients are averaged over several hundred individual laser pulses to improve the signal/noise ratio. Figure 14.9A shows a typical acoustic signal of pressure fluctuations in a 0.142 mM HGN solution as recorded by a hydrophone between two 130 fs laser pulses (1-KHz repetition rate). No pressure fluctuations above background occurred in control solutions of phosphate buffer solution, or 4.71 mM CF solution dissolved in phosphate buffer, irradiated with the highest pulsed laser power density of 16.2 W/cm2 (Fig. 14.9B). Laser-induced heating, however, does not always produce an acoustic signal. For spherical gold particles in aqueous solution, explosive boiling happens only above a threshold temperature of 85% of the critical temperature (Tc ¼ 374  C for water) (Kotaidis et al., 2006). We have observed that a similar threshold of the laser energy density is required to generate a photoacoustic signal. The amplitude of the pressure fluctuations remains at

294

Guohui Wu et al.

Figure 14.8 TEM images showing the collapse of the HGN into solid nanoparticles after 16.1 mJ/cm2 NIR-laser irradiation.

background up to a laser power density of 2.3 mJ/cm2, above which the amplitude of the pressure fluctuations increases with increasing laser power density (Lin and Kelly, 1998) (Fig. 14.9A). Calculations show that the increased power density leads to higher HGN temperatures (Prasad et al., 2005; Prevo et al., 2008), which are then translated into larger pressure fluctuations (more water boils and larger bubbles are formed) as this energy is dissipated. The pressure fluctuations die out within a few hundred microseconds after the light pulse; the HGN and the surrounding solution equilibrate and return to ambient temperature prior to the next pulse from the laser. Although the gold nanoshells melt (Fig. 14.8), the temperature increase of the bulk solution is small; the sample reaches a steady temperature only a few degrees above ambient, which depends on the laser power density and the HGN concentration. This is because the absolute amount of energy deposited into the solution is quite small (the laser power here is 0.67 W). Therefore, any significant pulsed laser-induced heating is limited to the immediate vicinity (microns) of the HGNs. This is important for the biomedical application of HGNs as only liposomes, vesosomes (liposomes encapsulating multiple smaller liposomes), or tissues directly adjacent to the irradiated HGNs will be affected by the light pulse. However, it is well established that transient cavitation bubbles due to sonication are capable of disrupting cell and liposome membranes (Huang, 2008). The irradiated HGNs might be thought of as optically triggered, nanosonicators, which can cause the transient rupture of liposomes resulting in rapid contents release.

5. Coupling HGN to Liposomes As the extent of the heated zone is quite small, the method of coupling the HGN to liposomes or cell membranes is important. We have examined three different methods of rupturing liposomes with HGNs. First, HGNs

295

Synthesis, Characterization, and Optical Response

B

cm J/

2

) Accoustic signal (mV

0.4 0.2 0.0 −0.2

) Accoustic signal (mV

16 .1 m

9. 48 m

J/ 7. cm 44 2 m J/ 3. cm 59 2 m 2. J/ 31 cm 2 m J/ 1. cm 34 2 m J/ 0. cm 72 2 m J/ 0m cm 2 J/ cm

2

A

0.4 0.2 0.0 −0.2

Buffer,

2

J/cm

6.1 m on, 1

luti CF so

−0.4 −0.6

2

1000

−0.4

HGN

800

−0.6

2

cm

16.1 mJ/

/cm

J , 0m

0

Ti me

200

(ms

)

600

400

0

200

800 600 (ms) ime

400

1000

T

Figure 14.9 Acoustic signal amplitude of (A) 0.142 mM HGN solution as a function of pulsed laser energy is recorded by a hydrophone after a single laser pulse of various laser energy densities. The magnitude of the photoacoustic signal of pressure fluctuations associated with cavitation (Paliwal and Mitragotri, 2006) increases with increasing laser power density, consistent with larger energy absorption by the HGN and subsequent larger or more numerous cavitation bubbles. However, a threshold value of laserenergy density of 2.3 mJ/cm2 is required to induce the photoacoustic signals above background in the HGN solutions. (B) Acoustic signal amplitude in control solutions. No signal was observed in buffer or CF solution without HGN, and no signal was observed in HGN solution without laser irradiation.

can be encapsulated within dipalmitoylphosphatidylcholine (DPPC) liposomes via the interdigitated phase transition, which causes lipid membranes to form flat open sheets at low temperatures that close to form unilamellar vesicles at higher temperatures (Boyer and Zasadzinski, 2007; Kisak et al., 2002) (Scheme 14.2A). Second, the nanoshells can be tethered to the outside of preformed liposomes using a thiol/PEG–lipid linkage (Scheme 14.2B). Third, the HGN solutions can be mixed with preformed liposomes so that the HGNs are exclusively outside the liposomes (Scheme 14.2C). Scheme 14.2A: Encapsulation of HGN and CF inside liposomes. 6-Carboxyfluorescein (50 mM) (CF; Invitrogen; Eugene, OR) is dissolved in water together with 6 equiv. of concentrated NaOH, which converts the CF from its acid form to the water-soluble salt form. The CF solution is used to disperse the PEG-stabilized HGNs described previously. Liposomes are prepared as described by Boyer and Zasadzinski (2007) and Kisak et al. (2002). DPPC dried from chloroform is hydrated with Milli-Q water by vortexing at 55  C to form a lamellar dispersion, which is then cooled to room temperature. Vesicles are prepared by sonication at room temperature using a 60 Sonic Dismembrator (Fisher Scientific) for 4 min at a power of 4 W.

296

Guohui Wu et al.

A

Inside

Interdigitated lipid sheets B

Tethered NIR Liposome Lipid-PEG-SH Freely outside

C Liposome

Scheme 14.2 HGNs can be located: (A) inside the liposomes; (B) tethered to the liposomes; and (C) free outside the liposomes.

At room temperature, DPPC bilayers are in the gel or Lb0 phase; the interdigitated bilayer phase (LI) is induced by the addition of 0.106 ml of ethanol (3 M net ethanol concentration) to 0.5 ml of a 50 mg/ml DPPC vesicle suspension. The initially bluish vesicle suspension turns milky white, and its viscosity increases significantly. The vesicles fuse and burst open, forming stacks of open bilayer sheets many micrometers in size (Boyer and Zasadzinski, 2007; Kisak et al., 2002). After annealing at 4  C overnight, the interdigitated sheets are centrifuged at 3000g and dispersed in pure water three times to remove any ethanol. The pellet of interdigitated DPPC sheets is mixed with the solution of 32 mM CF and 12 mM HGN and heated at 50  C for 2 h under vortex mixing. Raising the temperature causes a phase transition from LI to the liquid crystalline La phase and the bilayer sheets become much more flexible, allowing them to close around the HGN in suspension to form interdigitation–fusion vesicles (Boyer and Zasadzinski, 2007; Kisak et al., 2002). The internal concentrations in the liposomes are 32 mM CF (110 mOsm), 12 mM HGN, and the overall lipid concentration is 22 mg/ml DPPC. Based on earlier work (Boyer and Zasadzinski, 2007), about 50–60% of the HGN is encapsulated in liposomes. Cryo-EM confirms that the HGN are encapsulated within the liposomes by this procedure (Wu et al., 2008). After encapsulation, phosphate-buffered saline (PBS; 20 mM Na2HPO4/ NaH2PO4, 34.5 mM NaCl, pH ¼ 7.4) is used to disperse the liposomes to minimize osmotic stress across the membrane. The unencapsulated CF is

297

Synthesis, Characterization, and Optical Response

removed by size-exclusion chromatography using a Sephadex G-75 column (Amersham Biosciences Corp., Piscataway, NJ) eluted with PBS buffer. The eluted suspension is centrifuged at 100g for 20 min and redispersed in PBS buffer twice to remove any unencapsulated HGN prior to irradiation. Scheme 14.2B: Tethering HGN to liposomes containing CF. A pellet of interdigitated DPPC sheets is prepared as described above. The interdigitated sheets are mixed with a 32 mM CF in water solution containing 2 mol% (relative to the amount of DPPC used) DSPE-2000PEG-NH2 powder (Avanti Polar Lipids, Alabaster, AL). The mixture is then heated at 50  C for 1 h under vortex mixing. The DSPE-2000PEG-NH2 partitions into the DPPC bilayers in the La phase as the temperature increases and the interdigitation fusion vesicles are formed (Sou et al., 2000; Zalipsky et al., 1996). Next, the amine groups at the liposome surfaces are converted to thiol by mixing with 100% excess 2-iminothiolane solution (0.29 M). The thiolated liposomes encapsulating CF are incubated with a solution of HGN and CF for 48 h to allow HGN to tether via the thiol linkages to the outer surfaces of liposomes. The final concentrations in the solution are 18 mg/ml phospholipid (98 mol% DPPC and 2 mol% DSPE-2000PEG-SH), 18 mM HGN and 32 mM CF. The liposomes with tethered HGNs are eluted through a Sephadex G-75 size-exclusion column to remove any unencapsulated CF, and centrifuged at 200g to remove untethered HGNs. Figure 14.10 shows a cryo-EM tilt series that confirms that the HGN are tethered to the liposomes (Wu et al., 2008). Thin films of the liposome/ HGN solution are spread on holey carbon TEM grids (Structure Probe, West Chester, PA) under controlled temperature and humidity conditions using a VitRobot (FEI Company, Hillsboro, OR) (Frederik and Hubert, 2005), then vitrified by rapid plunging into liquid ethane (Chiruvolu et al., 1994a,b; Frederik and Hubert, 2005; Jung et al., 2002). Cryo-EM imaging is performed on an FEI Tecnai T20 microscope, operating at 200 kV with a Gatan liquid nitrogen specimen cryo-holder. Single-axis tomographic imaging is performed with a JEOL 2010A microscope from 60 to þ60 tilt

−45⬚

0⬚

+45⬚

200 nm

200 nm

200 nm

Figure 14.10 Cryo-TEM tilt series showing that the HGNs are tethered to the liposome surfaces. Arrows point out the same HGNs as they rotate with the surface of the liposome.

298

Guohui Wu et al.

angles in 2 increments with a total dose of less than 100 electrons/A˚2. Three representative images at goniometer tilt angles of 45 , 0 , and 45 are shown in Fig. 14.10. The arrows mark specific HGNs on the surface of the liposomes followed during the 90 rotation to confirm that HGNs are tethered to the surface. Scheme 14.2C: Liposomes containing CF with external HGNs. DPPC liposomes containing 32 mM CF are prepared by the same interdigitation– fusion method described earlier, except no HGNs are added to the solution prior to vesicle formation. The preformed vesicles are eluted through a Sephadex G-75 column to remove external CF, and then dispersed in different concentrations of HGN solutions as needed.

5.1. Pulsed laser optics Liposome disruption is triggered by irradiating the liposome/HGNs with the output of the femtosecond Ti:Sapphire regenerative amplifier (Spectraphysics Spitfire) running at a repetition rate of 1 kHz. The setup is illustrated in Fig. 14.11. The laser beam is collimated by a Galilean telescope to achieve a Gaussian diameter of 2.3 mm. The pulse duration is monitored by a home-built single-shot optical autocorrelator and is kept at about 120 fs. The spectral FWHM of the laser radiation is 12 nm centered around 800 nm. The laser beam is directed onto the sample by a system of mirrors; no focusing optics are used. The energy of the optical pulse is controlled by Schott neutral density glass filters. A thermopile power meter (Newport Inc., Irvine, CA) is used to measure the incident optical power. The maximum power available is 670 mW, which corresponds to

Tunable fs (ps) laser light (700–1000 nm) Beam shaping optics

Sample

Emission pickup optics and filters Two-photon absorption

Emission Spectrometer

CCD

Figure 14.11 Schematic illustration of Laser irradiation/CF release detection setup.

Synthesis, Characterization, and Optical Response

299

670 mJ/pulse and an energy density of 16.1 mJ/cm2 (or mean power density of 16.1 W/cm2). The temperature of the HGN suspensions is measured using an Omegaette HH306 digital thermometer (Omega) with a K-type thermocouple probe (Omega Engineering Inc., Stamford, CT), which is immersed into the solution 5 mm above the laser beam. The solution is stirred to ensure good mixing during irradiation. Luminescence is excited in the sample via a two-photon absorption process. The emission is collected at a 90 angle by a system of lenses and focused on the entrance slit of a monochromator (Acton Research SpectraPro 300). The laser radiation is blocked by a Schott colored glass filter (BG38). The light dispersed by the monochromator is detected by a spectroscopic CCD camera (PI Acton PIXIS-400) and transferred into a personal computer for analysis. The evolution of the photoluminescence is recorded by collecting consecutive spectra over a 600 nm bandwidth with a constant interval. To quantify the fractional release of CF, fluorescence is measured using a PTI QuantaMaster spectrofluorimeter (Photon Technology International, Lawrenceville, NJ). Any release from the liposomes is detected by an increase in fluorescence intensity (from the background) as the external concentration of CF increased. The fractional release can be quantified as fractional release ¼ (Ilaser I0)/(Imax I0), where Ilaser is the fluorescence intensity of the solution after laser treatment, Imax is the maximum fluorescence intensity after lysing the liposomes with reduced Triton X-100 (a nonionic surfactant which has a hydrophilic polyethylene oxide group and a hydrophobic 4-(1,1,3,3-tetramethylbutyl)-phenyl group) (Boyer and Zasadzinski, 2007), and I0 is the background fluorescence intensity before either treatment.

5.2. Continuous-wave laser irradiation To illustrate the differences between pulsed and continuous-wave irradiation, the liposome/HGN samples are also irradiated with continuous NIR light, using a Spectraphysics 3900S Ti:Sapphire CW laser. The laser wavelength is tuned to 820 nm and the output power is controlled by changing the power of the pump laser source (Spectraphysics Beamlok-2060) to be 0.7 W. The Gaussian beam diameter of the CW laser is 1.0 mm.

6. Liposome Disruption and CF Release Due to Pulsed Laser Irradiation Irradiation with femtosecond NIR light pulses with an energy density >2.2 W/cm2 triggers a near instantaneous increase in the measured fluorescence in the solution of DPPC liposomes encapsulating HGNs and CF

300

Guohui Wu et al.

Table 14.1 Comparison of triggered content release

Laser

Solution

Pulsed fs

CF CF þ Au NS Liposomes containing CF, but no Au NS Au NS suspended freely outside of liposomes containing CF Au NS and CF encapsulated inside liposomes Au NS tethered to the outer surface of liposomes containing CF Au NS and CF encapsulated inside liposomes

CW

Release (%)

02 12 12 28  2 71  1 93  2 12

(Table 14.1), but has no effect on control solutions of unencapsulated CF, a mixture of HGNs and CF, or DPPC liposomes with CF, but no HGNs. In all samples, the CF concentrations are matched to give similar concentrations averaged over the sample volume. Continuous-wave (unpulsed) laser irradiation at 800 nm with a higher average power density of 89 W/cm2 leads to no increase in the fluorescence intensity, and hence no CF release, even after 4 h of irradiation. When continuous-wave laser irradiation is used, the nanoshell is always close to being in thermal equilibrium with its surroundings, and there are insufficient temperature gradients to give rise to microbubble formation (Prasad et al., 2005; Prevo et al., 2008). From these observations, it is clear that minor changes in the solution and liposome temperature are not responsible for the rapid release of CF from the liposomes.

7. Mechanism of Triggered Liposome Release To identify the mechanism of release, the laser power density is varied as in Fig. 14.9, while comparing the total fluorescence intensity after 9 min of irradiation with 120 fs long pulses at a 1 kHz repetition rate. Figure 14.12 shows a distinct power density threshold of NIR light necessary to trigger CF release from the liposomes: no fluorescence increase is detected for a power density lower than 1.5 W/cm2, while the total release is roughly constant at about 74% for power densities greater than 4.3 W/cm2. The rate of fluorescence increase during NIR irradiation for liposomes encapsulating HGNs also increases with the laser power density (Fig. 14.13). The in situ fluorescence intensity is constant for an irradiation power density of 1.3 mJ/cm2, which is below the threshold. Above the threshold, the curves could be fit with a single exponential, with a time constant increasing with decreased

301

Synthesis, Characterization, and Optical Response

100

50

80

40

60

30

40

20

20

10

0

0 Au NS freely outside liposomes

80

40

60

30

40

20

20

10

0

Derivative

Dye release (%)

Au NS inside liposomes

0 0

2 4 6 8 10 12 14 Laser power density (mJ/cm2)

16

Figure 14.12 Effect of energy density on CR release from liposomes with HGNs encapsulated inside and suspended outside (black data points). The solid red curves are sigmoidal fits to the data; the blue curves are the derivative of the fit. The peak in the derivative gives the threshold energy for CF release, which is the same for both samples. The maximum release is quite different due to the different average distances between HGNs and the liposome bilayers.

laser power density (time constant t ¼ 5, 52, and 112 at the power density of 14.9, 13.0, and 7.1 W/cm2, respectively). At the higher power densities, release is complete within seconds. The derivative of the release versus laser-energy density curves gives an estimate of the threshold energy density for contents release (Fig. 14.12). For both HGNs inside and free outside the liposomes, the threshold value is the same, 2.2 mJ/cm2, which coincides with the threshold for the photoacoustic signal of cavitation in the solution (Fig. 14.9A). The power threshold suggests that the mechanism of triggered release is through perforation of lipid bilayers mediated by transient cavitation, that is, microbubble formation and collapse (Paliwal and Mitragotri, 2006; Pecha and Gompf, 2000; Tong et al., 2007). Several recent studies have also suggested that the effects of plasmon-resonant nanoparticles on cell membrane rupture are

302

Guohui Wu et al.

Intensity (a.u.)

3 ⫻ 104

14.9

m2

mJ/c

2 ⫻ 104

2 J/cm

m 13.0

1 ⫻ 104 7.1

2 cm

mJ/

0 2 m

J/c

m 1.3

500 400

300 ) 200 e (s m 100 Ti 0

Figure 14.13 Kinetics of in situ fluorescence shows the rate of liposomal content-release induced by encapsulated HGNs at various laser-energy densities. Time zero is the beginning of laser irradiation. Below a threshold energy density (2.2 mJ/cm2), there was no fluorescence increase. Above the threshold, there was a near instantaneous increase in the fluorescence intensity, followed by a more gradual increase. The increase of fluorescence increase can be fit by single exponential: F ¼ Fo þ A e x=t , with t ¼ 5, 52, and 112 s at laser-energy densities of 14.9, 13.0, and 7.1 mJ/cm2, respectively.

linked with cavitation dynamics and transient bubble formation (Lapotko et al., 2006; Pitsillides et al., 2003; Yao et al., 2005. While the permeability of DPPC liposomes increases near the gel to liquid crystalline transition temperature of 41  C (Ponce et al., 2006), this is not the source of the rapid increase in CF fluorescence observed here. The overall temperature increase of the solution due to the HGN absorption of NIR light is only a few degrees above ambient (Prevo et al., 2008). Even at 41  C, the permeability increase is only such that CF release would take minutes, not seconds as found here. The only way to get this rate of release is by the generation of large defects in the bilayer, as observed in liposomes after sonication (Zasadzinski, 1986). Only minor differences in liposome morphology are visible by cryo-EM after irradiation (Wu et al., 2008); the bilayers are less spherical, and hence under less tension after irradiation, consistent with a decrease in the positive osmotic pressure difference following CF release. These minor changes in the liposome shapes or sizes after irradiation suggest that cavitation induces transient defects in the bilayer, enabling drug release, after which membrane integrity is restored.

303

Synthesis, Characterization, and Optical Response

8. Effect of Proximity of HGNs to Liposomes Permeabilizing lipid membranes with microbubble cavitation should be induced by any HGNs in the solution, as long as there is an HGN within some maximum distance from a liposome bilayer. To test this hypothesis, DPPC liposomes encapsulating CF dye are mixed with increasing concentrations of HGNs, according to Scheme 14.2C. Upon pulsed laser irradiation, CF release is triggered. Figure 14.14 shows that the fractional CF release increases with HGN concentration up to an HGN concentration of 0.0315 mM (at higher HGN concentrations, the CF fluorescence is quenched by the HGNs (Dulkeith et al., 2002), data not shown). Hence, increasing the HGN concentration allows more liposomes to be within a critical distance of an irradiated HGN, which causes more liposomes to be ruptured, leading to greater CF release and an increase in the fluorescence. Minimizing and maintaining the distance between the HGN and the liposome bilayer lead us to tether HGNs to the outer surface of liposomes (Scheme 14.2B). Tethering the HGN directly to the outer surface of the liposomes increases the maximum release fraction to 96% (Table 14.1). The efficiency of phototriggered contents release is strongly affected by the proximity of HGN to the bilayer, consistent with the hypothesis that mechanical disruption by microbubbles is responsible for release (Tong et al., 2007). 60 50

Dye release (%)

40 30 20 10 0 −10 0.00

0.01 0.02 External HGN (mM)

0.03

Figure 14.14 Effect of unencapsulated HGNs on CF release from DPPC liposomes after NIR irradiation at 16.1 mJ/cm2. The solid line is a linear fit to the data.

304

Guohui Wu et al.

9. Conclusions Femtosecond pulses of NIR light absorbed by HGNs tethered to, encapsulated within, or in solution with liposomes trigger the near instantaneous release of liposome contents. The high temperatures reached by the HGN (Prasad et al., 2005; Prevo et al., 2008) induce production of unstable microbubbles, similar to the cavitation bubbles produced by ultrasound (Paliwal and Mitragotri, 2006). The mechanical and thermal effects of the microbubble collapse (Pecha and Gompf, 2000; Pitsillides et al., 2003) causes disruption of the liposome carriers, similar to the disruption caused by sonication. Neither the liposomes nor the CF appears to be altered chemically during this process, and the overall temperature rise of the bulk solution is only a few degrees. NIR light can penetrate up to 10 cm into tissue, which should allow these liposome/HGN complexes to be addressed noninvasively within a reasonable fraction of the human body. Any liposome carrier could be modified by tethering or encapsulating HGN to produce a system for rapid release on demand via NIR irradiation. This should eventually allow for better control of drug delivery to selected disease sites while minimizing systemic toxicity.

ACKNOWLEDGMENTS We thank Dr. Samir Mitragotri, Dr. Sumit Paliwal and Dr. Makoto Ogura for helpful discussions on cavitation and generously lending the hydrophone. This work was supported by the NIH Program of Excellence in Nanotechnology Grant HL080718: Nanotherapy for Vulnerable Plaques.

REFERENCES Abraham, S. A., Waterhouse, D. N., Mayer, L. D., Cullis, P. R., Madden, T. D., and Bally, M. B. (2005). The liposomal formulation of doxorubicin. Methods Enzymol. 391, 71–97. Allen, T. M., and Cullis, P. R. (2004). Drug delivery systems: Entering the mainstream. Science 303, 1818–1822. Alvarez, M. M., Khoury, J. T., Schaaff, T. G., Shafigullin, M. N., Vezmar, I., and Whetten, R. L. (1997). Optical absorption spectra of nanocrystal gold molecules. J. Phys. Chem. B 101, 3706–3712. Bohren, C. F., and Huffman, D. R. (1983). Absorption and Scattering of Light by Small Particles. Wiley, New York, 530 pp. Boyer, C., and Zasadzinski, J. A. (2007). Multiple lipid compartments slow vesicle contents release in lipases and serum. ACS Nano 1, 176–182. Chen, J., Wang, D., Xi, J., Au, L., Siekkinen, A., Warsen, A., Li, Z.-Y., Zhang, H., Xia, Y., and Li, X. (2007). Immuno gold nanocages with tailored optical properties for targeted photothermal destruction of cancer cells. Nano Lett. 7, 1318–1322.

Synthesis, Characterization, and Optical Response

305

Chiruvolu, S., Warriner, H. E., Naranjo, E., Idziak, S., Radler, J. O., Plano, R. J., Zasadzinski, J. A., and Safinya, C. R. (1994a). A phase of liposomes with entangled tubular vesicles. Science 266, 1222–1225. Chiruvolu, S., Walker, S., Leckband, D., Israelachvili, J., and Zasadzinski, J. (1994b). Higher order self-assembly of vesicles by ligand-receptor interactions. Science 264, 1753–1756. Chiu, D. T., Wilson, C. F., Ryttsen, F., Stromberg, A., Farre, C., Karlsson, A., Nordholm, S., Gaggar, A., Modi, B. P., Moscho, A., Garza-Lopez, R. A., Orwar, O., et al. (1999). Chemical transformations in individual ultrasmall biomimetic containers. Science 283, 1892–1895. Davidsen, J., Jorgensen, K., Andresen, T. L., and Mouritsen, O. G. (2003). Secreted phospholipase A(2) as a new enzymatic trigger mechanism for localised liposomal drug release and absorption in diseased tissue. Biochim. Biophy. Acta 1609, 95–101. Discher, B. M., Won, Y. Y., Ege, D. S., Lee, J.-M., Bates, F. S., and Hammer, D. (1999). Polymersomes: Tough vesicles made from diblock copolymers. Science 284, 1143–1146. Dulkeith, E., Morteani, A. C., Niedereichholz, T., Klar, T. A., Feldmann, J., Levi, S. A., van Veggel, F. C. J. M., Reinhoudt, D. N., Moller, M., and Gittins, D. I. (2002). Fluorescence quenching of dye molecules near gold nanoparticles: Radiative and nonradiative effects. Phys. Rev. Lett. 89, 203002/203001–203002/203004. Frederick, P. M., and Hubert, D. H. (2005). Cryoelectron microscopy of liposomes. Methods Enzymol. 391, 431–448. Grua, P., Morreeuw, J. P., Bercegol, H., Jonusauskas, G., and Vallee, F. (2003). Electron kinetics and emission for metal nanoparticles exposed to intense laser pulses. Phys. Rev. B Condens. Matter Mater. 68, 035424/03542.1–035424/0354.12. Hao, E., Li, S. Y., Bailey, R. C., Zou, S. L., Schatz, G. C., and Hupp, J. T. (2004). Optical properties of metal nanoshells. J. Phys. Chem. B 108, 1224–1229. Harris, N., Ford, M. J., and Cortie, M. B. (2006). Optimization of plasmonic heating by gold nanospheres and nanoshells. J. Phys. Chem. B 110, 10701–10707. Hirsch, L. R., Stafford, R. J., Bankson, J. A., Sershen, S. R., Rivera, B., Price, R. E., Hazle, J. D., Halas, N. J., and West, J. L. (2003). Nanoshell-mediated near-infrared thermal therapy of tumors under magnetic resonance guidance. Proc. Natl. Acad. Sci. USA 100, 13549–13554. Huang, S.-L. (2008). Liposomes in ultrasonic drug and gene delivery. Adv. Drug Deliv. Rev. 60, 1167–1176. Huang, S.-L., and MacDonald, R. C. (2004). Acoustically active liposomes for drug encapsulation and ultrasound-triggered release. Biochim. Biophy. Acta 1665, 134–141. Huang, X., El-Sayed, I. H., Qian, W., and El-Sayed, M. A. (2006). Cancer cell imaging and photothermal therapy in the near-infrared region by using gold nanorods. J. Am. Chem. Soc. 128, 2115–2120. Johnson, P. B., and Christy, R. W. (1972). Optical constants of the noble metals. Phys. Rev. B 6, 4370–4379. Jung, H. T., Lee, S. Y., Coldren, B., Zasadzinski, J. A., and Kaler, E. W. (2002). Gaussian curvature and the equilibrium between cylinders, discs and spheres. Proc. Natl. Acad. Sci. USA 99, 15318–15322. Kisak, E. T., Coldren, B., and Zasadzinski, J. A. (2002). Nanocompartments enclosing vesicles, colloids, and macromolecules via interdigitated lipid bilayers. Langmuir 18, 284–288. Kisak, E. T., Coldren, B., Evans, C. A., Boyer, C., and Zasadzinski, J. A. (2004). The vesosome—A multicompartment drug delivery vehicle. Curr. Med. Chem. 11, 199–219. Kotaidis, V., Dahmen, C., von Plessen, G., Springer, F., and Plech, A. (2006). Excitation of nanoscale vapor bubbles at the surface of gold nanoparticles in water. J. Chem. Phys. 124, 184702/184701–184702/184707.

306

Guohui Wu et al.

Kreibig, U. (1977). Anomalous frequency and temperature dependence of the optical absorption of small gold particles. J. Phys. (Paris) Colloq. 2, 97–103. Kreibig, U., and Genzel, L. (1985). Optical absorption of small metallic particles. Surf. Sci. 156, 678–700. Kreibig, U., and Vollmer, M. (1995). In ‘‘Optical Properties of Metal Clusters,’’ Springer Series in Materials Science, 532 pp. Lapotko, D. O., Lukianova, O. E., and Oraevsky, A. A. (2006). Selective laser nanothermolysis of human leukemia cells with microbubbles generated around clusters of gold nanoparticles. Lasers Surg. Med. 38, 631–642. Liang, H. P., Wan, L. J., Bai, C. L., and Jiang, L. (2005). Gold hollow nanospheres: Tunable surface plasmon resonance controlled by interior-cavity sizes. J. Phys. Chem. B 109, 7795–7800. Lin, C. P., and Kelly, M. W. (1998). Cavitation and acoustic emission around laser-heated microparticles. Appl. Phys. Lett. 72, 2800–2802. Link, S., Burda, C., Nikoobakht, B., and El-Sayed, M. A. (1999a). How long does it take to melt a gold nanorod? A femtosecond pump-probe absorption spectroscopic study. Chem. Phys. Lett. 315, 12–18. Link, S., Wang, Z. L., and El-Sayed, M. A. (1999b). Alloy formation of gold-silver nanoparticles and the dependence of the plasmon absorption on their composition. J. Phys. Chem. B 103, 3529–3533. Mueller, A., Bondurant, B., and O’Brien, D. F. (2000). Visible-light-stimulated destabilization of PEG-liposomes. Macromolecules 33, 4799–4804. Niidome, T., Yamagata, M., Okamoto, Y., Akiyama, Y., Takahashi, H., Kawano, T., Katayama, Y., and Niidome, Y. (2006). PEG-modified gold nanorods with a stealth character for in vivo applications. J. Control. Release 114, 343–347. Noble, C. O., Kirpotin, D. B., Hayes, M. E., Mamot, C., Hong, K., Park, J. W., Benz, C. C., Marks, J. D., and Drummond, D. C. (2004). Development of ligandtargeted liposomes for cancer therapy. Expert Opin. Ther. Targets 8, 335–353. Norman, R. S., Stone, J. W., Gole, A., Murphy, C. J., and Sabo-Attwood, T. L. (2008). Targeted photothermal lysis of the pathogenic bacteria, pseudomonas aeruginosa, with gold nanorods. Nano Lett. 8, 302–306. Oldenburg, S. J., Averitt, R. D., Westcott, S. L., and Halas, N. J. (1998). Nanoengineering of optical resonances. Chem. Phys. Lett. 288, 243–247. Paliwal, S., and Mitragotri, S. (2006). Ultrasound-induced cavitation: Applications in drug and gene delivery. Expert Opin. Drug Deliv. 3, 713–726. Pecha, R., and Gompf, B. (2000). Microimplosions: Cavitation collapse and shock wave emission on a nanosecond time scale. Phys. Rev. Lett. 84, 1328–1330. Peer, D., Karp, J. M., Hong, S., Farokhzad, O. C., Margalit, R., and Langer, R. (2007). Nanocarriers as an emerging platform for cancer therapy. Nat. Nanotechnol. 2, 751–760. Petrova, H., Perez Juste, J., Pastoriza-Santos, I., Hartland, G. V., Liz-Marzan, L. M., and Mulvaney, P. (2006). On the temperature stability of gold nanorods: Comparison between thermal and ultrafast laser-induced heating. Phys. Chem. Chem. Phys. 8, 814–821. Pitsillides, C. M., Joe, E. K., Wei, X., Anderson, R. R., and Lin, C. P. (2003). Selective cell targeting with light-absorbing microparticles and nanoparticles. Biophys. J. 84, 4023–4032. Ponce, A. M., Vujaskovic, Z., Yuan, F., Needham, D., and Dewhirst, M. W. (2006). Hyperthermia mediated liposomal drug delivery. Int. J. Hyperthermia 22, 205–213. Popinet, S., and Zaleski, S. (2002). Bubble collapse near a solid boundary: A numerical study of the influence of viscosity. J. Fluid Mech. 464, 137–163. Prasad, V., Mikhailovsky, A., and Zasadzinski, J. A. (2005). Inside-out disruption of silica/ gold core-shell nanoparticles by pulsed laser irradiation. Langmuir 21, 7528–7532.

Synthesis, Characterization, and Optical Response

307

Prevo, B. G., Esakoff, S. A., Mikhailovsky, A., and Zasadzinski, J. A. (2008). Scalable routes to gold nanoshells with tunable sizes and properties: Monitoring their response to NIR pulsed laser irradiation. Small 4, 1183–1195. Roper, D. K., Ahn, W., and Hoepfner, M. (2007). Microscale heat transfer transduced by surface plasmon resonant gold nanoparticles. J. Phys. Chem. C 111, 3636–3641. Schwartzberg, A. M., Oshiro, T. Y., Zhang, J. Z., Huser, T., and Talley, C. E. (2006). Improving nanoprobes using surface-enhanced Raman scattering from 30-nm hollow gold particles. Anal. Chem. 78, 4732–4736. Sengupta, S., Eavarone, D., Capila, I., Zhao, G., Watson, N., Kiziltepe, T., and Sasisekharan, R. (2005). Temporal targeting of tumour cells and neovasculature with a nanoscale delivery system. Nature 436, 568–572. Shi, W., Sahoo, Y., Swihart, M. T., and Prasad, P. N. (2005). Gold nanoshells on polystyrene cores for control of surface plasmon resonance. Langmuir 21, 1610–1617. Shum, P., Kim, J.-M., and Thompson, D. H. (2001). Phototriggering of liposomal drug delivery systems. Adv. Drug Deliv. Rev. 53, 273–284. Simo˜es, S., Fonseca, C., Pedroso de Lima, M., and Du¨zgu¨nes¸, N. (2006). pH responsive liposomes. State of the art from biophysics to therapy. In ‘‘Smart Nanoparticles in Nanomedicine,’’ (R. Arshady and K. Kono, eds.), Vol. 8, pp. 217–250. Kentus Books, London. Sou, K., Endo, T., Takeoka, S., and Tsuchida, E. (2000). Poly(ethylene glycol)-modification of the phospholipid vesicles by using the spontaneous incorporation of PEG into the vesicles. Bioconjug. Chem. 11, 372–379. Sun, Y. G., and Xia, Y. N. (2004). Mechanistic study on the replacement reaction between silver nanostructures and chloroauric acid in aqueous medium. J. Am. Chem. Soc. 126, 3892–3901. Sun, Y. G., Mayers, B. T., and Xia, Y. N. (2002). Template-engaged replacement reaction: A one-step approach to the large-scale synthesis of nanostructures with hollow interiors. Nano Lett. 2, 481–485. Sun, Y., Mayers, B., and Xia, Y. (2003). Metal nanostructures with hollow interiors. Adv. Mater. 15, 641–646. Tong, L., Zhao, Y., Huff, T. B., Hansen, M. N., Wei, A., and Cheng, J.-X. (2007). Gold nanorods mediate tumor cell death by compromising membrane integrity. Adv. Mater. 19, 3136–3141. Turkevich, J. (1985). A history of colloidal gold. Gold Bull. 18, 86–91. Turkevich, J., Stevenson, P. C., and Hillier, J. (1951). A study of the nucleation and growth processes in the synthesis of colloidal gold. Dis. Faraday Soc. 11, 55–65. Turkevich, J., Garton, G., and Stevenson, P. C. (1954). The color of colloidal gold. J. Colloid Sci. 9, S26–S35. Weissleder, R. (2001). A clearer vision for in vivo imaging. Nat. Biotechnol. 19, 316–317. Wiley, B., Herricks, T., Sun, Y. G., and Xia, Y. N. (2004). Polyol synthesis of silver nanoparticles: Use of chloride and oxygen to promote the formation of single crystal, truncated cubes and tetrahedrons. Nano Lett. 4, 1733–1739. Wiley, B., Sun, Y. G., Chen, J. Y., Cang, H., Li, Z. Y., Li, X. D., and Xia, Y. N. (2005). Shape controlled synthesis of silver and gold nanostructures. MRS Bull. 30, 356–361. Wu, G., Mikhailovsky, A., Khant, H. A., Fu, C., Chiu, W., and Zasadzinski, J. A. (2008). Remotely triggered liposome release by near-infrared light absorption via hollow gold nanoshells. J. Am. Chem. Soc. 130, 8175–8177. Yao, C. P., Rahmanzadeh, R., Endl, E., Zhang, Z., Gerdes, J., and Huttmann, G. (2005). Elevation of plasma membrane permeability by laser irradiation of selectively bound nanoparticles. J. Biomed. Opt. 10, 064012/064011–064012/064018. Zalipsky, S., Brandeis, E., Newman, M. S., and Woodle, M. C. (1994). Long circulating, cationic liposomes containing amino-peg-phosphatidylethanolamine. FEBS Lett. 353, 71–74. Zasadzinski, J. A. (1986). Transmission electron microscopy observations of sonicationinduced changes in liposome structure. Biophys. J. 49, 1119–1130.

This page intentionally left blank

C H A P T E R

F I F T E E N

Complex Nanotube-Liposome Networks Aldo Jesorka and Owe Orwar

Contents 1. Introduction 2. Network Fabrication Protocols 3. Complexity and Topology 4. Internal and Membrane Functionalization 5. Transport Phenomena and Controlled Mixing Procedures 6. Enzymatic Reactions in NVN 7. Concluding Remarks Acknowledgments References

309 310 314 315 318 320 323 323 324

Abstract Surfactant nanotube-vesicle networks (NVN) belong to the smallest artificial devices known to date for performing controlled chemical operations with enzymes. Newly established means for transport of chemical reactants between containers, as well as advancements in initiation and control of chemical reactions in such systems have opened pathways to new devices with a resolution down to the single-molecule level. Here, we summarize the fabrication and functionalization of complex nanotube-liposome networks for such devices, and discuss related aspects of their application for studying chemical kinetics and materials transport phenomena in ultrasmall-scale biomimetic environments.

1. Introduction In biological cells, nature has achieved complex, truly nanoscale systems for computation and information processing, direct and catalyzed synthesis, sensing and other life sustaining biochemical and biophysical tasks. It is a challenging transdisciplinary endeavor to create biomimetic, Department of Chemical and Biological Engineering, Chalmers University of Technology, Go¨teborg, Sweden Methods in Enzymology, Volume 464 ISSN 0076-6879, DOI: 10.1016/S0076-6879(09)64015-5

#

2009 Published by Elsevier Inc.

309

310

Aldo Jesorka and Owe Orwar

fully or partly artificial devices that imitate some key features of biological systems, with the overall goal of providing new concepts for complex chemical operations on the nanoscale level. A key point of interest is, from a physicochemical perspective, to understand how chemical reactions proceed in a small-scale environment, and which kinetic and mechanistic principles govern these transformations. In particular, modeling and experimentally studying chemistry in ultrasmall devices containing only a few molecules is an emerging field of nanoscience and nanotechnology (Karlsson et al., 2003a; Levene et al., 2003; Xie et al., 2008), with growing importance in analytical chemistry and materials science. Methods for controlling the initiation of chemical reactions, mixing, and transport phenomena in biomimetic nanoscale compartments with designed functionalities are of considerable relevance for detailed understanding of fundamental (bio)chemical phenomena, such as enzymatic reactions, protein synthesis and transport, properties of signaling pathways, and others. An unconventional, yet fully biocompatible environment can be created by means of soft-matter fabrication methods that use phospholipid bilayer membranes. It comprises biomimetic containers as model reactors and lipotube-interconnected containers as complex network devices that have been demonstrated to be useful for investigations of chemical reactions, polymer dynamics, mass transport, and other phenomena at the micrometer and nanometer level. A set of well-defined procedures and methods has been generated from within this multidisciplinary approach, to be briefly introduced in the following five subsections.

2. Network Fabrication Protocols The concepts used for nanoscale network and device design are based on self-organized phospholipid membranes that drew inspiration from biological systems (Evans et al., 1996; Karlsson et al., 2001). A set of methods based on self-assembly of individual phospholipid molecules to membrane leaflets, self-organization, and forced shape transformations using precise micromanipulation tools can be utilized to form fully closed lipid bilayer shells and attached nanoconduits with certain structural and functional properties similar to biological micro- and nanocompartments (Karlsson et al., 2004; Lizana et al., 2009). These procedures are in fact unconventional bottom-up fabrication routes with extraordinary flexibility to yield three-dimensional soft-matter devices at a length scale that is difficult to reach with modern solid-state clean room technology. Figure 15.1 shows schematically the fabrication strategy. This fabrication concept involving forced shape transitions is interesting for creating specialized membrane-encapsulated volumes of

311

Complex Nanotube-Liposome Networks

A

B

+

Pi

C Pi

− F = 1–10 pN

GUV

VB VA

MLV ftube ~ 100 nm fGUV ~ 10-100 mm D

Pi

VC

E

F

VA

ΨB/A =

3rtube VB = VA 2rGUV(A)

VB

Figure 15.1 Schematic representation of a fabrication strategy for complex nanofluidic networks composed of buffer-filled giant unilamellar phospholipid vesicles (GUV) and nanotubes. (A) Penetration of a GUV with a glass injection pipette after electroporation with a DC pulse. The injection pipette is backfilled with a solution that can be the same as or differ from the liquid content of the vesicle, for example, a buffered enzyme solution, here indicated by a different color. Attached to the GUV is an MLV, which serves as a lipid reservoir for subsequent network building. (B) By retracting the needle from the vesicle, using a micromanipulation device, a nanotube is drawn from the GUV. (C) Injecting the content of the injection pipette with positive backpressure Pi leads to formation of a second, the so-called daughter vesicle, at the end of the nanotubes. (D) The procedure can be repeated with differently backfilled injection pipettes, leading to a content-differentiated complex nanofluidic network. (E) For the creation of a new daughter vesicle, the ratio C of the two volumes involved is dependent only on the dimensions of the nanotubes connecting them and the size of the mother vesicle.

particular volume, connectivity and functionality. Several advanced micromanipulation protocols based on micropipette injection technology and electroporation have been developed, including vesicle fission, pipette writing, and vesicle inflation (Karlsson et al., 2004). In particular the vesicle inflation method, as depicted in Fig. 15.1, allows for the fabrication of fluid-state lipid nanotube-vesicle networks (NVN) of high geometrical complexity, where each node within a network can feature uniquely differentiated chemistry. The networks are composed of giant surface-immobilized phospholipid bilayer vesicles (typically 20 mm in diameter, 10–12 l, with a membrane thickness of  5 nm) interconnected by  100 nm wide lipid nanotubes. The construction and design of nanofluidic channels only a few times larger than individual biomacromolecules are thus possible. A 100-nm-radius tube has a cross-sectional area of 3.14  10 14 m2. In comparison, the size of a

312

Aldo Jesorka and Owe Orwar

common enzyme, alkaline phosphatase, is approximately 3.5 nm in diameter (Lenghaus et al., 2003), giving a cross-sectional area of 9.6  10 18 m2. The volume element of a 100-nm-long tube segment is 3.14  10 22 m3, and if such a volume is hosting a single molecule, then the concentration will be 53 nM. The total system volume of a vesicle nanotubes network is at least 6 orders of magnitude smaller than that of traditional microfluidic channels. The experimental setup for the construction of NVN is schematically depicted in Fig. 15.2. It can easily be adapted to any inverted and confocal microscope setup. The main components are a pair of water hydraulic micromanipulators (high graduation: Narishige MWH-3, coarse graduation: Narishige MC-35A). The electro-assisted injections are controlled by a microinjection pumping system (Eppendorf Femtojet) and a pulse generator (Digitimer Stimulator DS9A, 0–40 V, ms/ms/s pulse lengths, single pulse and pulse series generation). Occasionally, a manual injection pump is required, in particular when lipid material has to be removed gently from a unilamellar vesicle. This is conveniently achieved with an Eppendorf CellTram Vario piston pump. Carbon fiber microelectrodes measuring 30 mm in length and 5–10 mm in diameter (ProCFE, Axon Instruments) are employed as

Femtojet injection pump Waterhydraulic micromanipulator 1

V Waterhydraulic micromanipulator 2

Pa

DC puls generator +

Injection needle and eletroporation electrode

Carbon fiber microelectrode

GUVs Microscope cover slip Inverted microscope objective 40x

Figure 15.2 Experimental setup for the fabrication of nanotube-vesicle networks. The individual components are described in the text.

Complex Nanotube-Liposome Networks

313

counterelectrode, while a thin silver wire serves as main electrode in the backfilled capillaries. Injection tips are prepared from borosilicate capillaries (GC100TF-10, Clark Electromedical Instruments) that were carefully flame-forged in the back ends (Narishige MF 900 Microforge) to avoid that the sharp glass ends wear out the capillary holder. The capillaries are flushed with a stream of nitrogen gas to remove dust particles and tapered on a CO2-laser-puller (P-2000, Sutter Instrument) immediately before use. The outer-tip diameter of the capillaries produced by this method is typically 0.25–1 mm. Giant vesicles for network building are most conveniently prepared starting from a 5-ml droplet of 1 mg/ml L-a-phosphatidylcholine dispersed in phosphate-buffered saline (PBS) buffer (Trizma base 5 mM, K3PO4/ KH2PO4 30 mM, MgSO4 1 mM, EDTA 0.5 mM, adjusted with H3PO4 to pH 7.8). The droplet is pipetted onto a borosilicate coverslip #1, dehydrated in a vacuum desiccator under membrane pump vacuum for 15 min. The system contains 1% (v/v) glycerol as additive to avoid complete dehydration. After deposition and drying, the lipid is carefully rehydrated by covering it with 0.5 ml PBS buffer. Within a few minutes, several hundreds of cellsized unilamellar vesicles are formed. An aliquot (300 ml) is placed on the top face of a second coverslip, which is finally used in the experiment. The liposomes are immobilized on the coverslip surface by spontaneous adsorption. It is often advantageous to spincoat the negative expoxy photoresist SU-8 (Microchem) onto the cover glass prior to use, which after baking and short UV–ozone exposure forms a solid, moderately hydrophobic film on the glass surface. It prevents strong adhesion and disintegration of the surface-immobilized vesicles and extends the lifetime of the network. It has to be noted that UV exposure and cross-linking without the final oxidizing treatment renders the surface very hydrophobic, leading to lipid monolayer formation (Czolkos et al., 2007). A tapered and rear-end forged borosilicate glass micropipette is backfilled with aqueous medium, mounted onto the electroinjection system and micromanipulated onto the membrane of a surface-immobilized GUV, which itself is attached to a multilamellar lipid reservoir. With one or several anodic rectangular DC-voltage pulses of field strengths between 10 and 40 V/cm and duration of 1–4 ms applied over the micropipette, the liposome membrane can be penetrated (Fig. 15.1A). The micropipette is then pulled out and away from the liposome (about 5 mm/s), forming a lipid nanotube connection between the liposome and the pipette tip. When the nanotube has reached the desired length, aqueous medium is slowly pressure-injected (5  10 14 l/s) into the nanotubes. Drawing lipid material from the reservoir, a small vesicle is formed at the outlet of the micropipet tip. The size of the vesicle can be set by controlling injection time and volume, and the newly created structure is finally deposited on the

314

Aldo Jesorka and Owe Orwar

surface (Fig. 15.1B). Several vesicles can be fabricated successively, forming a nanotube-interconnected network (Fig. 15.1C). Already in their simplest configuration, these liposome systems can be used to study and control chemical reactions in a well-defined nanoscale environment, and their potential as constituents in advanced devices for nanofluidics and nanochemistry applications has been already partly exploited. In particular, enzymatic reactions, transport studies, and incorporation of membrane proteins will be discussed in a later section.

3. Complexity and Topology To form network structures of higher order topologies, vesicles within a network must be connected using membrane fusion induced by, for example, a focused electric field (Karlsson et al., 2002a; Stromberg et al., 2000). Circular networks as well as fully connected networks with threedimensional nanotube layers are examples of such systems (Fig. 15.3). To create a closed system, a simple network is constructed (Fig. 15.3A and B); then, a small nanotube-conjugated satellite vesicle is formed and positioned in close contact to a surface-immobilized vesicle container within the network (Fig. 15.3C). A localized electrical field is applied by placing a carbon fiber microelectrode adjacent to the fusion partners. Fusion of the vesicle containers is stimulated by application of one or several short B

A

E MLV

D

C

+

−

10 mm

Figure 15.3 Construction of complex nanotube-vesicle networks using an electroporation/membrane fusion protocol. A three-node network is constructed using the vesicle injection procedure (A, B). By means of a micropipette, a small fusion vesicle is generated and brought into contact with a surface-immobilized target vesicle. Application of a rectangular DC pulse fuses the vesicles (C, D), leading to a closed network of genus 1. (E) Differential interference contrast enhanced bright-field micrograph of an 8-GUV nanotubes-interconnected network. The highest connectivity is 6 (center vesicle), MLV denotes the multilamellar vesicle that serves as a membrane reservoir. Reprinted by permission of the American Chemical Society.

Complex Nanotube-Liposome Networks

315

rectangular DC-voltage pulses of field strengths between 40 and 80 V/cm and durations of 1–4 ms between the micropipette and the counterelectrode (Fig. 15.3D). Withdrawal of the micropipette can subsequently be performed without vesicle deformation, leakage, or disruption. Residual lipid material occasionally adhering to the pipette is removed by applying one or several cathodic DC-voltage pulses over the micropipette. The result is a closed, circular NVN (Fig. 15.3D). An extension of the procedure to very complex networks is possible, provided there is a sufficiently large membrane reservoir (MLV, multilamellar vesicle) attached (Fig. 15.3D, upper right corner). Employing the above-described micromanipulation technologies to move vesicles and produce localized electric fields and mechanical point loads, configurational transformations can be performed within and between lipid NVN. These manipulations include building and breaking connections within networks or between networks. Specifically, separate networks can be joined together using a combination of vesicle translocation and electrofusion (Karlsson et al., 2002a). Self-organization can also be utilized to create network subdomains of bifurcating lipid nanotubes with the surface-immobilized vesicles arranged at the periphery. Networks are in this instance produced from initial geometries with doubly nanotube-conjugated vesicles. Self-organization is triggered by mechanical or electromechanical action to merge two nanotubes. This is a process driven by spontaneous minimization of surface free energy of the phospholipid membrane. The dimensions of nanotubes, the network connectivity as well as the size, location, and contents of individual containers can be designed and manipulated with high precision (Lobovkina et al., 2005, 2008).

4. Internal and Membrane Functionalization Strong arguments for using lipid membranes as a material for constructing nanoscale soft-matter devices are their compatibility with biological components and their capability to host certain protein functionalities, such as ion channels, receptors, and enzymes (Criado and Keller, 1987; Eytan, 1982). Proteins can reside in the phospholipid membrane, having the fundamental bilayer structure of biological membranes. Not only water-soluble proteins and biomacromolecular assemblies, such as enzymes, but also synthetic polymers of a simpler structure can be confined to the interiors of vesicles or nanotubes. Several proteins can be extracted from cellular systems and subsequently be reconstituted into an artificial lipid host environment. Alternatively, biomacromolecules and lipids can be extracted directly from cultured

316

Aldo Jesorka and Owe Orwar

biological cells (Bauer et al., 2006). A high degree of compositional complexity can be achieved with both strategies. Davidson et al. (2003) demonstrated successfully the formation of NVN with reconstituted membrane protein from red blood cells (Fig. 15.4A and B). The network itself is prepared as described earlier. For the reconstitution, erythrocyte hosts with an eosin-5-maleimide-labeled anion exchanger AE1 are employed. The labeled material containing 1 mg/ml protein is solubilized for 1 h with 5 mM Triton X-100 at pH 7.8 in a solution consisting of 0.12 M KCl, 0.5 mM EDTA, 1 mM MgSO4, and 1% glycerol. After centrifugation (27,000g, 1 h), the supernatant is diluted two times with 10 mM Triton X-100 in the same buffer and incubated for 1 h. All steps are performed at 0  C. Finally, removal of the detergent with Biobeads SM-2 at room temperature and filtering through 200 nm membrane filters, a proteoliposome solution is obtained, which can be transformed into giant unilamellar vesicles, using the KPi-buffer dehydration/rehydration technique described earlier. The activity of the AE1 after reconstitution is verified by single-channel ion conductance measurements in excised inside-out patches from the vesicle membranes. The distribution of protein in the network membrane is established by means of fluorescence microscopy to be homogeneous. The labeled protein could diffuse via nanotube interconnections from vesicle to vesicle. A

C

B

D

Figure 15.4 Examples of internal and membrane functionalization of NVNs. (A, B) Fluorescence micrograph of a four-vesicle network with reconstituted membrane proteins (fluorescently labeled AE1 ion channels). The center vesicle is also schematically represented. The lower two vesicles in this panel are additionally filled with small unilamellar vesicles, containing the same labeled protein. (C) Fluorescence micrograph of a three vesicle network with different fluorescent dyes in each container. (D) Compartmentalization of a GUV by means of a hydrogel derived from a thermoresponsive LCST polymer hydrogel. The diameter of a GUV in each panel is  25 mm. Reprinted by permission of the American Chemical Society.

Complex Nanotube-Liposome Networks

317

Another recent example of incorporation of membrane proteins directly from cultured biological cells was demonstrated by Bauer et al. (2006). Here, a combination of dithiothreitol (DTT) and formaldehyde is employed to produce plasma membrane vesicles that are subsequently shaped into networks using the micropipette-assisted method for network fabrication. A single cell is fully sufficient to derive all material needed to build a small network. This protocol allows, similar to direct reconstitution of protein in vesicles, full control over the solution environment. Most importantly, the proteins remain in their proper orientation. However, the DTT/formaldehyde treatment may lead to loss of some protein function due to cross-linking; alternative protocols can be utilized to minimize this effect. Generally, the protein and lipid content in the networks is determined by employing different cell types, for instance by prior overexpression or the use of specialized cell types as sources for specific proteins. Aside from membrane modification and functionalization, the networks offer the unique possibility to differentiate the contents of individual vesicles. It represents a major benefit of the bottom-up production methodology, giving the ability to assemble complex chemical reaction systems (Karlsson et al., 2001). A simple example of a three-vesicle network, having each node modified with a different internalized fluorescent dye, is depicted in Fig. 15.4C. This is the foundation for the utilization of NVNs as enzymatic model reactors, or artificial cells. As already described, the exchange of injection pipettes at the time of network generation allows for fabrication of vesicles with individually defined internal composition and excellent control over the content composition and concentrations. In contrast, direct incorporation using ordinary liposome techniques, that is, enclosing the entire content upon formation of a liposome, usually suffers from low encapsulation efficiency ( Jesorka and Orwar, 2008). Several examples of this concept have been presented in the past, ranging from simple ionic solutions and colloidal particles for visualization, via small vesicles (Bolinger et al., 2004) and water-soluble polymers for compartmentalization ( Jesorka et al., 2005; Long et al., 2005), to biomacromolecules like DNA and alkaline phosphatase in order to investigate transport phenomena and enzymatic reactions on the size scale of a biological cell (Sott et al., 2006; Tokarz et al., 2005). Material is typically injected in different nodes within a network to yield an initially heterogeneous distribution of these species. Diffusional relaxation then occurs rapidly for materials small enough to traverse the nanotubes. Furthermore, well-defined internal structures can be formed in vesiclenanotube networks. An example is the controlled generation of hydrogels within the internal compartments (Markstrom et al., 2007) (Fig. 15.4D). Polymer solutions or finely dispersed suspensions can be microinjected into vesicles that subsequently undergo controlled sol–gel transitions. The thermoresponsive poly(N-isopropyl acrylamide), which undergoes a reversible

318

Aldo Jesorka and Owe Orwar

phase transition at 32  C in pure water and 27  C in 100 mM phosphate buffer solution, allows injection, mixing, and manipulation of clear solutions while enabling reversible solidification and compartmentalization as well as entrapping and release of colloidal material and larger biomacromolecules on demand. Vesicles and networks with hydrogel interiors of high density allow, for example, the study of cellular enzyme kinetics under simulated macromolecular crowding (cell-like) conditions (Helfrich et al., 2002; Johansson et al., 2000).

5. Transport Phenomena and Controlled Mixing Procedures In NVN, injected or locally generated materials can move through the network structure. Different, passive diffusional or active, tension or field-driven means of transport (Dommersnes et al., 2005; Karlsson et al., 2002b, 2005; Lizana and Konkoli, 2005; Tokarz et al., 2005) govern their mobility. Due to their minute size and high structural flexibility, NVNs allow for controlled transport of ultrasmall numbers of molecules. Size and optical properties of the networks allow generally for direct monitoring of transport processes by optical means, provided fluorescent or fluorescently labeled species are employed. Understanding and characterization of mixing and diffusion processes within such confined spaces have become the foundation of our investigations of intranetwork enzyme reaction kinetics. To initiate chemical (for example, enzymatic) reactions in the nodes of nanofluidic NVNs, control of fluid delivery is of paramount importance. The simplest and most direct method of transport and mixing of the contents of interconnected vesicles in a network is their integration by nanotube-mediated fusion, thereby delivering discrete quantities of material in a controlled fashion. A mobile, pipette-suspended vesicle is merged with and emptied into a stationary vesicle to achieve mixing, the final concentration being well defined by the two initially spherical volumes (Karlsson et al., 2003b). Another useful technique involves mixing of two individual solutions in a growing GUV conjugated by a suspended nanotube on one side and contacted by an injection pipette on the other (Fig. 15.5A). The far end of the nanotube is connected to a larger mother vesicle and a lipid membrane reservoir. The inflation of the daughter vesicle forces membrane material from the mother to traverse via the lipid nanotube to the daughter vesicle, producing a liquid flow inside the nanotube due to nonslip conditions and viscous coupling. During inflation, the different liquid contents of both pipette and nanotube will mix. At the limit of diffusion, the mixing ratio is only

319

Complex Nanotube-Liposome Networks

+p

A

rGUV

Solvent B

Solvent A

2rtube

3 mm 5 %A 15 mm 1 %A 30 mm 0.5 %A

Membrane reservoir

B

C D1

D2

MV

2.56 1.92 1.28 0.64 0 (s)

D 10 mm

Figure 15.5 Mixing and diffusional transport in NVNs. (A) Schematic drawing of mixing driven by membrane flow. Solvent A (blue arrows) from the left-hand vesicle is transported by viscous drag through the interconnecting nanotube as the right-hand vesicle grows during injection of solvent B (green arrows). The red figures represent the increasing diameter of the inflated vesicle, the blue figures are the corresponding percentages of solvent A in that vesicle. Membrane material is supplied from a membrane reservoir (red arrows). (B) Fluorescence micrograph of a three vesicle system constructed by micropipette injection according to panel (A). Mixing in a chain of consecutively microinjected vesicles leads to a dilution series that is represented by the decreasing fluorescence of the fluorescein solution originating from the mother vesicle (MV), as it is increasingly diluted with the injected buffer solution in daughter vesicles D1 and D2. (C) Schematic representation of two vesicle network interconnected by a nanotube. The red enclosure is the area of the nanotube where diffusive transport of a 30 nm single latex particle has been followed over a period of 2.56 s. (D) Time-dependent positional information of the particle in the enclosed area from panel (C). The color represents the point in time the particle was detected in the particular position. Reprinted by permission of the American Chemical Society.

dependent on the diameters of the inflated vesicle and the nanotube, but not on inflation rate (see also Fig. 15.1E). For example, in a system of several sequentially generated nanotube-conjugated vesicles, dilution series can be obtained with mixing ratios that cover almost 3 orders of magnitude between the first and last container (Fig. 15.5B). Thus, vesicles can be prepared with a well-defined volume of the two solutions at a pre-calculated mixing ratio. Three fundamental mechanisms of molecular transport within lipid nanotubes have been investigated to date. The first, Marangoni transport, is based on membrane tension gradients and the dynamic and fluid character

320

Aldo Jesorka and Owe Orwar

of the bilayer membrane (Dommersnes et al., 2005). The second mechanism is electrophoresis (Tokarz et al., 2005), a well-established way of distributing fluid and solutes in micro- and nanofluidic devices. The third and in practical terms most important mode is diffusion, a very effective means of transport over short distances (Davidson et al., 2005; Lizana and Konkoli, 2005). NVNs have several unique properties with respect to diffusion as an effective means of transport. They are sufficiently small, the chemical potential in the networks can be controlled over time by injection of different concentrations of relevant species into individual containers and, finally, the geometry of the networks can be changed over time. Reactants such as enzymes and substrates can thus be transported in a controlled manner from one end to the other and sequentially catalyze reactions in different nodes, as has been shown for diffusive-directed transport of alkaline phosphatase (Sott et al., 2006). Not only small molecules but also nanoscale objects such as submicron particles can be efficiently distributed by diffusion within NVNs. Single nanoparticles exhibited stochastic motion inside a nanotube. This movement is mainly Brownian, but contributions from random membrane tension differences are conceivable. The stochastic motion of a 30-nm-diameter particle inside a 100-nm-radius nanotube is largely one dimensional (Fig. 15.5C and D). The figure shows five different time point observations of how a particle spontaneously occupies a 6-mm segment of a lipid nanotube in a random fashion during a 2.56-s observation period (Karlsson et al., 2002b). The theoretical diffusion coefficient (Brenner and Gaydos, 1977) for a 30-nm-diameter particle at room temperature in a tube of 100 nm radius is 9.05 mm2/s, and the corresponding root-mean-square displacement of 6.81 mm during a 2.56-s time interval is in good agreement with the experiment.

6. Enzymatic Reactions in NVN In chemical reaction systems with a characteristic length scale of micro- and nanometers, which rapidly gain importance in modern technological soft-matter applications, the understanding of kinetics of chemical reactions in highly confined spaces is of increasing interest. As discussed earlier, at this small-scale, diffusion is the predominant mode of transport and mixing, thus the interplay between reactions and diffusion-dominated transport is of special relevance. Figure 15.6A and B shows as an example the progression of fluorescein diffusion in a network constructed of three vesicles. Central to the discussion about the differences between small- and large-scale reactions is the question about which transport modes dominate

321

Complex Nanotube-Liposome Networks

C

A

D MLV

V1

V2

V3

t = 0 min B V1

V2

V3 V1

V2

V1 + 2 10 mm

t = 10 min 2

3

1

4

t = 10 min

F 2

3

4

Normalized intensity

G

E

−6 −4 −2 0

1

t = 90 min

Vesicle 1 Vesicle 2 Vesicle 3 Vesicle 4

−8

0

20 40 60 80 100 120 Time (min)

Figure 15.6 Transport and enzymatic reactions in NVN. In the fluorescence micrographs, the boundary of the vesicles and the connecting nanotubes are highlighted with dashed lines. V1–V4 denotes the volumes. Fluorescence images are digitally edited to improve image quality. (A, B) Diffusion of fluorescein through an NVN at injection time (A) and after 10 min (B), followed by laser-induced fluorescence microscopy (LIF—fluorescein emission upon excitation at 488 nm). (C, D) Contrast-enhanced bright-field micrograph of the nanotube-mediated vesicle fusion technique employed to start a chemical reaction in a network. Vesicle 1, suspended by a pipette is translated toward vesicle 2, where content mixing and initiation of the reaction occurs. A multilamellar vesicle is attached to the original mother vesicle in order to provide lipid material during network construction. (E, F) LIF-images (fluorescein emission) of product formation in a four-vesicle network with linear geometry. Initially, vesicle 1 is injected with enzyme, while vesicles 2–4 are filled with FDP. Scale bar represents 10 mm. (G) Plots of normalized fluorescence intensity versus time (solid lines), representing the rate of product formation in vesicles 1–4 in the NVN shown panels (E) and (F). Dash-dotted lines show the theoretically modeled time dependencies of product formation. Reprinted by permission of the American Chemical Society.

in, and which reactions benefit from, structured spaces. There are a number of reactions that run faster in a network-like geometry. The interplay between geometry and the nature of a reaction scheme becomes important only when the individual reaction steps start to influence each other. Such reactions contain antagonistic catalytic influences in the intermediate stages of a multistep reaction scheme. This approach is, for example, an alternative way to explain certain aspects of cytoarchitecture (Konkoli, 2005).

322

Aldo Jesorka and Owe Orwar

Due to their versatility, flexibility, and favorable materials properties, vesicle-nanotube networks present a potent system to investigate chemical reactions in confined and structured space and under macromolecular crowding conditions. Reaction conditions can include interactions with biomacromolecules and cell components in their native environment, while the fluid character of the phospholipid membrane offers practical advantages, such as facile interfacing to injection equipment and active surfaces. We have investigated enzymatic reactions in unstructured and complex networks and utilized the dynamic features of the fluid membrane for reaction control. A simple technique to initiate enzymatic reactions inside NVN is nanotube-mediated merging of two interconnected vesicles (Fig. 15.6C and D). Here, vesicles are filled consecutively with the reaction partners (substrate and enzyme, respectively) by microinjection. Initially, the containers are spatially separated; diffusive mixing is insignificant due to the long distance and the narrow conduit. To initiate a reaction, the vesicles are brought in contact with each other, shortening the nanotube, and at a critical distance, the vesicles merge to form one spherical reactor with mixed contents. The concept is demonstrated on the stepwise enzymatic dephosphorylation of fluorescein diphosphate (FDP), which yields fluorescein monophosphate and finally fluorescein as products. Typical starting conditions are 10 mM FDP in a buffered solution consisting of 10 mM Trizma base and 100 mM KCl adjusted to pH 8.9; and 11 units/ml of alkaline phosphatase in a buffer of 5 mM Trizma base, 1 mM MgCl2, 0.1 mM ZnCl2, and 100 mM KCl at pH 8.9. Product formation is followed over time by fluorescence recovery after photobleaching (FRAP) microscopy. The enzyme concentration inside the vesicle after merging, here approximately 15 enzyme molecules, can be determined through comparison with bulk measurements. To gain theoretical understanding of the system, a simple enzyme substrate reaction model is developed and solved using a survival probability approach with the assumption of infinite enzyme substrate reaction rate. To further explore the potential of complex NVNs in a biomimetic context, the alkaline phosphatase/FDP reaction model was extended. Transition from a compact geometry (a single spherical container) to a structured geometry (several spherical containers connected by nanotubes) in NVNs induces the rate of the dephosphorylation reaction to display wave-like properties. By tuning the geometry of the network, the reaction dynamics can be directly influenced (Lizana et al., 2008). If, in this system, an enzyme is introduced into a terminal node of the network, it diffuses into neighboring nodes through the interconnecting nanotubes (Fig. 15.6E and F). The directionality of enzyme diffusion can be controlled exactly through establishing a concentration gradient. All the nodes except the enzymecontaining node bear FDP, which becomes fluorescent when converted to fluorescein and can thus be monitored by fluorescence microscopy.

Complex Nanotube-Liposome Networks

323

The temporal pattern of front propagation as well as the rate of reaction is dependent on geometry of the network. Figure 15.6G shows as an example the product formation dynamics in an unbranched linear NVN as wave-like propagation phenomenon (Sott et al., 2006). More extensive theoretical studies on both diffusion and reaction phenomena in similar networks have been performed, showing fascinating and nonintuitive behavior on front propagation and reaction optimization (Konkoli, 2007; Lizana and Konkoli, 2005).

7. Concluding Remarks Networks composed of phospholipid nanotubes and giant unilamellar vesicles are versatile chemical microreactors with nanofluidic interconnections. Unconventional yet comparatively facile fabrication strategies have been developed, employing phospholipids and biomacromolecules to construct enclosed membrane devices. The fabrication methodology involves self-assembly, forced shape transformations, and micromanipulation protocols such as electroinjection. The devices combine complex structure, flexibility and biocompatibility at a length scale that is difficult to reach even with the most advanced solid-state microtechniques. The networks are stable, flexible, and suitable for studies of chemical reactions in ultrasmall volumes, especially in a biologically relevant microenvironment. Reactants can be directly introduced and reliably confined in individual nodes, and strategies for microcompartmentalization and membrane modification, for example with ion channel proteins, exist. Due to the small spatial dimensions and short path lengths for molecules to travel, diffusion is the predominant, but not the only possible, material transport mode, and the impermeability of the phospholipid membrane confines all ionic and many nonpolar reactants to the network interior. Active and passive transport of small molecules, submicron particles, and biopolymers through nanotubes enables defined initiation and controlled progression of chemical reactions, while the fluid membrane boundary offers support through its structural dynamics. The uniquely flexible supramolecular architecture with its biomimetic foundation has bearing for understanding enzymatic reactions in biological systems.

ACKNOWLEDGMENTS The work was supported by the Royal Swedish Academy of Sciences, the Swedish Research Council (VR), and the Swedish Foundation for Strategic Research (SSF).

324

Aldo Jesorka and Owe Orwar

REFERENCES Bauer, B., Davidson, M., and Orwar, O. (2006). Direct reconstitution of plasma membrane lipids and proteins in nanotube-vesicle networks. Langmuir 22, 9329–9332. Bolinger, P. Y., Stamou, D., and Vogel, H. (2004). Integrated nanoreactor systems: Triggering the release and mixing of compounds inside single vesicles. J. Am. Chem. Soc. 126, 8594–8595. Brenner, H., and Gaydos, L. J. (1977). Constrained Brownian-movement of spherical particles in cylindrical pores of comparable radius—Models of diffusive and convective transport of solute molecules in membranes and porous media. J. Colloid Interface Sci. 58, 312–356. Criado, M., and Keller, B. U. (1987). A membrane fusion strategy for single-channel recordings of membranes usually non-accessible to patch-clamp pipette electrodes. FEBS Lett. 224, 172–176. Czolkos, I., Erkan, Y., Dommersnes, P., Jesorka, A., and Orwar, O. (2007). Controlled formation and mixing of two-dimensional fluids. Nano Lett. 7, 1980–1984. Davidson, M., Karlsson, M., Sinclair, J., Sott, K., and Orwar, O. (2003). Nanotube-vesicle networks with functionalized membranes and interiors. J. Am. Chem. Soc. 125, 374–378. Davidson, M., Dommersnes, P., Markstrom, M., Joanny, J. F., Karlsson, M., and Orwar, O. (2005). Fluid mixing in growing microscale vesicles conjugated by surfactant nanotubes. J. Am. Chem. Soc. 127, 1251–1257. Dommersnes, P. G., Orwar, O., Brochard-Wyart, F., and Joanny, J. F. (2005). Marangoni transport in lipid nanotubes. Europhys. Lett. 70, 271–277. Evans, E., Bowman, H., Leung, A., Needham, D., and Tirrell, D. (1996). Biomembrane templates for nanoscale conduits and networks. Science 273, 933–935. Eytan, G. D. (1982). Use of liposomes for reconstitution of biological functions. Biochim. Biophys. Acta 694, 185–202. Helfrich, M. R., Mangeney-Slavin, L. K., Long, M. S., Djoko, Y., and Keating, C. D. (2002). Aqueous phase separation in giant vesicles. J. Am. Chem. Soc. 124, 13374–13375. Jesorka, A., and Orwar, O. (2008). Liposomes: Technologies and analytical applications. Ann. Rev. Anal. Chem. 1, 801–832. Jesorka, A., Markstrom, M., Karlsson, M., and Orwar, O. (2005). Controlled hydrogel formation in the internal compartment of giant unilamellar vesicles. J. Phys. Chem. B 109, 14759–14763. Johansson, H. O., Brooks, D. E., and Haynes, C. A. (2000). International Review of Cytology—A Survey of Cell Biology 192, pp. 155–170. Academic Press Inc., San Diego, CA. Karlsson, A., Karlsson, R., Karlsson, M., Cans, A. S., Stromberg, A., Ryttsen, F., and Orwar, O. (2001). Molecular engineering—Networks of nanotubes and containers. Nature 409, 150–152. Karlsson, M., Sott, K., Davidson, M., Cans, A. S., Linderholm, P., Chiu, D., and Orwar, O. (2002a). Formation of geometrically complex lipid nanotube-vesicle networks of higherorder topologies. Proc. Natl. Acad. Sci. USA 99, 11573–11578. Karlsson, R., Karlsson, M., Karlsson, A., Cans, A. S., Bergenholtz, J., Akerman, B., Ewing, A. G., Voinova, M., and Orwar, O. (2002b). Moving-wall-driven flows in nanofluidic systems. Langmuir 18, 4186–4190. Karlsson, A., Karlsson, M., Karlsson, R., Sott, K., Lundqvist, A., Tokarz, M., and Orwar, O. (2003a). Nanofluidic networks based on surfactant membrane technology. Anal. Chem. 75, 2529–2537. Karlsson, R., Karlsson, A., and Orwar, O. (2003b). A nanofluidic switching device. J. Am. Chem. Soc. 125, 8442–8443.

Complex Nanotube-Liposome Networks

325

Karlsson, M., Davidson, M., Karlsson, R., Karlsson, A., Bergenholtz, J., Konkoli, Z., Jesorka, A., Lobovkina, T., Hurtig, J., Voinova, M., and Orwar, O. (2004). Biomimetic nanoscale reactors and networks. Annu. Rev. Phys. Chem. 55, 613–649. Karlsson, A., Sott, K., Markstrom, M., Davidson, M., Konkoli, Z., and Orwar, O. (2005). Controlled initiation of enzymatic reactions in micrometer-sized biomimetic compartments. J. Phys. Chem. B 109, 1609–1617. Konkoli, Z. (2005). Interplay between chemical reactions and transport in structured spaces. Phys. Rev. E 72, 011917. Konkoli, Z. (2007). Diffusion-controlled reactions in small and structured spaces as a tool for describing living cell biochemistry. J. Phys. Condens. Matter 19, 065149. Lenghaus, K., Dale, J. W., Henderson, J. C., Henry, D. C., Loghin, E. R., and Hickman, J. J. (2003). Enzymes as ultrasensitive probes for protein adsorption in flow systems. Langmuir 19, 5971–5974. Levene, M. J., Korlach, J., Turner, S. W., Foquet, M., Craighead, H. G., and Webb, W. W. (2003). Zero-mode waveguides for single-molecule analysis at high concentrations. Science 299, 682–686. Lizana, L., and Konkoli, Z. (2005). Diffusive transport in networks built of containers and tubes. Phys. Rev. E 72, 026305. Lizana, L., Bauer, B., and Orwart, O. (2008). Controlling the rates of biochemical reactions and signaling networks by shape and volume changes. Proc. Natl. Acad. Sci. USA 105, 4099–4104. Lizana, L., Konkoli, Z., Bauer, B., Jesorka, A., and Orwar, O. (2009). Controlling chemistry by geometry in nanoscale systems. Annu. Rev. Phys. Chem. 60, 449–468. Lobovkina, T., Dommersnes, P. G., Joanny, J. F., Bassereau, P., Karlsson, M., and Orwar, O. (2005). Pattern formation of different geometries and topologies as well as knotted structures in lipid nanotube networks. Biophys. J. 88, 208a. Lobovkina, T., Dommersnes, P. G., Tiourine, S., Joanny, J. F., and Orwar, O. (2008). Shape optimization in lipid nanotube networks. Eur. Phys. J. E 26, 295–300. Long, M. S., Jones, C. D., Helfrich, M. R., Mangeney-Slavin, L. K., and Keating, C. D. (2005). Dynamic microcompartmentation in synthetic cells. Proc. Natl. Acad. Sci. USA 102, 5920–5925. Markstrom, M., Gunnarsson, A., Orwar, O., and Jesorka, A. (2007). Dynamic microcompartmentalization of giant unilamellar vesicles by sol gel transition and temperature induced shrinking/swelling of poly(N-isopropyl acrylamide). Soft Matter 3, 587–595. Sott, K., Lobovkina, T., Lizana, L., Tokarz, M., Bauer, B., Konkoli, Z., and Orwar, O. (2006). Controlling enzymatic reactions by geometry in a biomimetic nanoscale network. Nano Lett. 6, 209–214. Stromberg, A., Ryttsen, F., Chiu, D. T., Davidson, M., Eriksson, P. S., Wilson, C. F., Orwar, O., and Zare, R. N. (2000). Manipulating the genetic identity and biochemical surface properties of individual cells with electric-field-induced fusion. Proc. Natl. Acad. Sci. USA 97, 7–11. Tokarz, M., Akerman, B., Olofsson, J., Joanny, J. F., Dommersnes, P., and Orwar, O. (2005). Single-file electrophoretic transport and counting of individual DNA molecules in surfactant nanotubes. Proc. Natl. Acad. Sci. USA 102, 9127–9132. Xie, X. S., Choi, P. J., Li, G.-W., Lee, N. K., and Lia, G. (2008). Single-molecule approach to molecular biology in living bacterial cells. Annu. Rev. Biophys. 37, 417–444.

This page intentionally left blank

C H A P T E R

S I X T E E N

Bionanotubules Formed from Liposomes Josemar A. Castillo and Mark A. Hayes

Contents 1. Introduction 2. Bionanotubule Formation by Applying Electric Fields to Surface-Attached Liposomes 2.1. Materials and methods 3. Bionanotubule Formation from Liposomes in Solution Using Electric Fields 3.1. Materials and methods 4. Other Methods of Bionanotubule Formation from Liposomes 5. Concluding Remarks References

328 329 331 334 336 337 339 339

Abstract Bionanotubules are lipid-bound cylindrical structures with typical diameters in the tens of nanometers and length than can span up to hundreds of micrometers. Besides being observed in nature, bionanotubules can be prepared synthetically by various methods, some of which involve the extension of these structures from lipid vesicles. We describe the formation of lipid nanotubules from liposomes prepared with various lipid mixtures including phosphatidylcholine, phosphatidic acid, and various fluorescent phospholipids. We depict the methods used to extend bionanotubules from surface-attached vesicles, using electric fields as the driving force for bilayer extension and tubular growth. These methods include liposome preparation, surface attachment, and tubular extension by applying modest electric fields (10 mm) are preferred. Also, using fluorescently labeled phospholipids (charged or zwitterionic) as components of the liposomes is required to monitor the charge and the phospholipid distribution during nanotubular formation by fluorescence microscopy. 2.1.1.1. Buffer preparation Phosphate buffer is made from prepared 10 mM solutions of sodium phosphate dibasic and sodium phosphate monobasic. A specific volume of sodium phosphate monobasic is titrated with enough sodium phosphate dibasic until the desired pH of 7.4 is obtained. Sodium phosphate dibasic anhydrous (>95%) and sodium dihydrogen phosphate (99%) are obtained from Sigma-Aldrich (St. Louis, MO). 2.1.1.2. NBD-labeled liposomes Liposomes composed of PC, phosphatidic acid (PA), and N-4-nitrobenz-2-oxa-1,3-diazole phosphatidic acid (NBD-PA) (90:9:1 by mass) are also prepared by reverse-phase evaporation (Szoka and Papahadjopoulos, 1978). All phospholipids are obtained from Avanti Polar Lipids (Alabaster, AL). Lipids (10–20 mg) dissolved in chloroform (CHCl3) solutions are added to a round-bottom flask. While rotating the flask, the CHCl3 is rapidly evaporated off using a gentle stream of nitrogen gas, leaving a thin, uniform solid gel coating of lipids on the interior of the flask. Remaining CHCl3 is removed by vacuum. The dry lipids are then prehydrated with a few microliters of nanopure 18 MO water and the round-bottom flask is placed in a rotary evaporator with a water temperature of 39  C. Phosphate buffer (3.00 ml, pH ¼ 7.4) is added to the flask, which is then allowed to rotate at 39  C for 2 h. This method produces giant uni- and multilamellar vesicles. 2.1.1.3. Oregon Green-labeled liposomes Oregon GreenÒ 488, 1,2dihexadecanoyl-sn-glycerol-3-phosphoethanolamine (OG-DHPE), is obtained from Molecular Probes (Eugene, OR). Liposomes composed of various ratios of PC/PA/OG-DHPE are prepared by reverse-phase evaporation (Szoka and Papahadjopoulos, 1978). The ratios used are 89:10:1,

332

Josemar A. Castillo and Mark A. Hayes

89:30:1, and 89:60:1 PC:PA:OG-DHPE by mass, respectively. OG-DHPE is obtained as a solid and is dissolved in CHCl3 (1.00 ml) to obtain a final concentration of 1 mg/ml. In brief, 10 mg of total CHCl3-dissolved lipids are added to a round-bottom flask. While rotating the flask, the CHCl3 is rapidly evaporated off using nitrogen gas, leaving a thin, uniform solid gel coating of lipids on the interior of the flask. Remaining CHCl3 is removed by vacuum. The dry lipids are prehydrated with a few microliters of water with rotation at 39  C. Phosphate buffer (pH ¼ 7.4) is then added to the flask, which is allowed to rotate at 39  C for 2 h. 2.1.2. Surface attachment 2.1.2.1. Sample wells Rectangles of approximately 2  4 cm are defined with silicon sealant (PermatexÒ ) on a microscope glass slide (Fig. 16.2A). The height of the silicon wall is adjusted to ensure that wells are capable of holding approximately 2 ml of solution. Wells can be open to air, however, it also possible to used a second piece of glass to partially close the structure while leaving open areas to serve as access for the placement of electrodes.

A

B

− V +

Low voltage power supply

Glass slide

1.7 cm Platinum wires (1 mm diameter)

Microscope stage

Silicone well (2 × 4 cm)

Figure 16.2 (A) Schematic (top view) of the experimental apparatus used for data collection along with a fluorescence image of a tubule growing from an immobilized liposome. The image illustrates the direction of tubular growth when an electric field is applied to surface-attached vesicles. (B) View of the experimental apparatus on the microscope stage during data collection (Castillo et al., 2009).

Bionanotubules Formed from Liposomes

333

2.1.2.2. Substrate coating Wells are cleaned and treated with a cationic surfactant according to methods developed for the treatment of capillary electrophoresis surfaces (Katayama et al., 1998). After thoroughly rinsing the wells with water, they are rinsed with 10 mM NaOH for 5 min, then with water for 5 min, and gently patted dry with VWR light-duty tissue wipes (West Chester, PA). The well is then filled with 7.5% polybrene (PB) (w/v in water) for 15 min, followed by a water rinse for 5 min. Hexadimethrine bromide (PB) is obtained from Sigma-Aldrich. 2.1.2.3. Liposome attachment The liposome preparation is allowed to reside in the prepared glass slide well for 5 min. Phosphate buffer (pH ¼ 7.4) is then used to overflow the well with the purpose of displacing the nonattached liposomes from the remaining solution while preventing the drying of the attachment surface. This procedure is performed until the remaining solution is visibly clear, indicating that most of the nonattached liposomes are removed from the surface. The outside of the well and the rest of the glass slide are completely dried with tissue wipes prior to beginning the imaging with the microscope. Substrate coating procedures are performed at room temperature.

2.1.3. Apparatus for electric field application The apparatus shown in Fig. 16.2A and B is assembled in-house. It consists of a simple circuit powered by low-voltage power supply with a voltmeter attached in parallel. One centimeter sections of two platinum wires, 1-mm diameter and separated by a distance of 1.7 cm, are submerged in the sample well perpendicularly to the long axis of the microscope slide. Both electrodes should rest on the substrate to ensure stability and submersion in the solution. As shown in Fig. 16.2B, the alligator clips holding the electrodes are attached to an easily movable block to add rigidity and portability to the system as well as to minimize realignment of the electrodes prior to each experiment (Castillo et al., 2009). 2.1.4. Tubular extension After electrodes are placed in the well, tubules are extended by starting with no electric field and progressively increasing the magnitude of the voltage applied. Generally, voltage magnitude is increase by 2 V/cm every 15 min (or longer). The mother liposomes are giant uni- and/or multilamellar vesicles with typical diameters of 10–50 mm. Well-behaved and stable growth of nanotubules can be observed with modest electric fields ( 0

ΔBMS < 0

Paramagnetic amphiphilic metal complex (ΔχLnL < 0) Paramagnetic hydrophilic metal complex Paramagnetic amphiphilic metal complex (ΔχLnL > 0)

Silvio Aime et al., Scheme 10.2 Basic correlations among the physicochemical variables involved in the magnetic orientation and chemical shift of intraliposomal water protons for paramagnetic liposomes.

Liposome(s)

= Lipid membrane anchor

= DNA target

= DNA probe

Ulla Jakobsen and Stefan Vogel, Figure 12.1 Double membrane anchor single DNAprobe design. Schematic representation of liposome aggregation upon duplex formation between a lipophilic probe DNA and an unmodified target DNA. Liposomes and DNA strands are not drawn to scale.

Probe A Liposomes Target Probe B

= DNA probes

= DNA target

Ulla Jakobsen and Stefan Vogel, Figure 12.2 Single membrane anchor dual-probe design. Schematic representation of liposome aggregation upon duplex formation between a lipophilic probe DNA and an unmodified target DNA. Liposomes and DNA strands are not drawn to scale.

A 0.8

Abs

0.6

0.4

0.2

0.0 220

240

260 280 Wavelength (nm)

300

320

220

240

260 280 Wavelength (nm)

300

320

B 0.8

Abs

0.6

0.4

0.2

0.0

Martina Banchelli et al., Figure 13.2 UV absorption spectra of the different fractions from the gel separation experiment: system before the separation (black), first fraction (red), second fraction (green), third fraction (blue); (A) POPC vesicles (1.3 mM) and ON-Chol (4.1 mM) and (B) free ON-Chol as the control.

100% 90% 80% 70% 60%

Fraction 3 Fraction 2 Fraction 1

50% 40% 30% 20% 10% 0% Liposome + ON-Chol

ON-Chol

Martina Banchelli et al., Figure 13.3 Relative values of % UV absorbance at 260 nm for the filtered fractions with respect to the absorption values before filtration.

Name

Sequence

FA-2

5’-ACGAGCCTTTGACGCTTGGA-TT-TAGTGCGTAACATAGGCTAC-TTCTGAAATTATGATAAAGA-3’

E’F’

5’-ATTTACCTGGAAGCAGCCAC-TT-TCCAAGCGTCAAAGGCTCGT-3’

ED

5’-GTGGCTGCTTCCAGGTAAAT-TT-CACTATGTAACTGGTCTCTTA-3’

D’C’

5’-TAGAGACCAGTTACATAGTG-TT-TGACCTCAGTCGCAAGGCTG-3’

CB

5’-CAGCCTTGCGACTGAGGTCA-TT-TCGGGTCAACGAATGGCTGC-3’

Pseudo-hexagonal structure D

E D⬘

C

E⬘

C⬘

F⬘ B⬘

B

B’A’

5’-GCAGCCATTCGTTGACCCGA-TT-GTAGCCTATGTTACGCACTA-3’

B’X

5’-GCAGCCATTCGTTGACCCGA-TT-CCCCCCCCCTTTTTTTTTTT-3’

F

A⬘ 5⬘

A

2⬘ 3⬘ Open structure

2

A

F⬘

E⬘

D⬘

C⬘

B’

F

E

D

C

B

X

Martina Banchelli et al., Scheme 13.2 DNA nanostructure composing strands. Adapted from Baldelli Bombelli et al. (2008).

12

Efficiency

10 Scattering Absorption

8 6 4 2 0 400

600 800 Wavelength (nm)

1000

Guohui Wu et al., Figure 14.2 Theoretical absorption and scattering efficiency for HGN with a diameter of 33 nm and shell thickness of 1.75 nm in water.

14

12 300 10 200 8 100

Absorption cross-section (cm2)

Absorption efficiency

400 ⫻ 10−12

6 20

40

60 80 HGN diameter (nm)

100

Guohui Wu et al., Figure 14.4 Absorption efficiency (red) and cross section (blue) at 800 nm of HGNs with optimized geometry as a function of diameter.

1.4

Absorbance (a.u.)

1.2 1.0 0.8

Experiment Simulation

0.6 0.4 0.2 0.0 400

500

600 700 800 Wavelength (nm)

900

1000

Guohui Wu et al., Figure 14.6 Measured (red) and calculated (blue) extinction spectra of the experimental distribution of HGN diameters and shell thicknesses from Fig. 14.5.

2 NIR

Extinction (a.u.)

1.5

1

0.5

0

400

600 800 Wavelength (nm)

1000

Guohui Wu et al., Figure 14.7 Extinction spectra (normalized at 350 nm) of: (blue) HGN suspension; (red) HGN suspension laser-irradiated at a power density of 16.1 mJ/cm2 for 8 min (120 fs pulses at 1-kHz repetition rate). Below the peaks of extinction curves are photographs of the corresponding dispersions. The spectra are consistent with the collapse and annealing of the HGN into solid gold/silver alloy nanoparticles.

100

50

80

40

60

30

40

20

20

10

0

0 Au NS freely outside liposomes

80

40

60

30

40

20

20

10

0

Derivative

Dye release (%)

Au NS inside liposomes

0 0

4 6 8 10 12 14 2 Laser power density (mJ/cm2)

16

Guohui Wu et al., Figure 14.12 Effect of energy density on CR release from liposomes with HGNs encapsulated inside and suspended outside (black data points). The solid red curves are sigmoidal fits to the data; the blue curves are the derivative of the fit. The peak in the derivative gives the threshold energy for CF release, which is the same for both samples. The maximum release is quite different due to the different average distances between HGNs and the liposome bilayers.

A

B

+

Pi

C

−

Pi F = 1–10 pN

GUV

VB VA

MLV ftube ~ 100 nm fGUV ~ 10-100 mm D

Pi

VC

E

F

VA

ΨB/A =

3rtube VB = VA 2rGUV(A)

VB

Aldo Jesorka and Owe Orwar, Figure 15.1 Schematic representation of a fabrication strategy for complex nanofluidic networks composed of buffer filled giant unilamellar phospholipid vesicles (GUV) and nanotubes. (A) Penetration of a GUV with a glass injection pipette after electroporation with a DC pulse. The injection pipette is backfilled with a solution that can be the same as or differ from the liquid content of the vesicle, for example, a buffered enzyme solution, here indicated by a different color. Attached to the GUV is an MLV, which serves as a lipid reservoir for subsequent network building. (B) By retracting the needle from the vesicle, using a micromanipulation device, a nanotube is drawn from the GUV. (C) Injecting the content of the injection pipette with positive backpressure Pi leads to formation of a second, the so-called daughter vesicle, at the end of the nanotubes. (D) The procedure can be repeated with more differently backfilled injection pipettes, leading to a contentdifferentiated complex nanofluidic network. (E) For the creation of a new daughter vesicle, the ratio C of the two volumes involved is dependent only on the dimensions of the nanotubes connecting them and the size of the mother vesicle.

A

C

B

D

Aldo Jesorka and Owe Orwar, Figure 15.4 Examples of internal and membrane functionalization of NVNs. (A, B) Fluorescence micrograph of a four-vesicle network with reconstituted membrane proteins (fluorescently labeled AE1 ion channels). The center vesicle is also schematically represented. The lower two vesicles in this panel are additionally filled with small unilamellar vesicles, containing the same labeled protein. (C) Fluorescence micrograph of a three vesicle network with different fluorescent dyes in each container. (D) Compartmentalization of a GUV by means of a hydrogel derived from a thermoresponsive LCST polymer hydrogel. The diameter of a GUV in each panel is  25 mm. Reprinted in permission of the American Chemical Society.

+p

A

rGUV

Solvent B

Solvent A

2rtube

3 mm 5 %A 15 mm 1 %A 30 mm 0.5 %A

Membrane reservoir

B

C D1

D2

MV

2.56 1.92 1.28 0.64 0 (s)

D 10 mm

Aldo Jesorka and Owe Orwar, Figure 15.5 Mixing and diffusional transport in NVNs. (A) Schematic drawing of mixing driven by membrane flow. Solvent A (blue arrows) from the left-hand vesicle is transported by viscous drag through the interconnecting nanotube as the right-hand vesicle grows during injection of solvent B (green arrows). The red figures represent the increasing diameter of the inflated vesicle, the blue figures are the corresponding percentages of solvent A in that vesicle. Membrane material is supplied from a membrane reservoir (red arrows). (B) Fluorescence micrograph of a three vesicle system constructed by micropipette injection according to panel (A). Mixing in a chain of consecutively microinjected vesicles leads to a dilution series that is represented by the decreasing fluorescence of the fluorescein solution originating from the mother vesicle (MV), as it is increasingly diluted with the injected buffer solution in daughter vesicles D1 and D2. (C) Schematic representation of two vesicle network interconnected by a nanotube. The red enclosure is the area of the nanotube where diffusive transport of a 30 nm single latex particle has been followed over a period of 2.56 s. (D) Time-dependent positional information of the particle in the enclosed area from panel (C). The color represents the point in time the particle was detected in the particular position. Reprinted in permission of the American Chemical Society.

Josemar A. Castillo and Mark A. Hayes, Figure 16.5 Fluorescence image of a lipid tubule formed by applying an electric field (50 V/cm) to a 1:9 NBD-PA:PC liposome in solution. The approximate size of the structure is 5  50 mm (Hayes et al., 2007).