Protein Tyrosine Phosphatases: Methods and Protocols (Methods in Molecular Biology, 2743) [2 ed.] 1071635689, 9781071635681

Citation preview

Methods in Molecular Biology 2743

Damien Thévenin · Jörg P. Müller  Editors

Protein Tyrosine Phosphatases Methods and Protocols Second Edition

METHODS

IN

MOLECULAR BIOLOGY

Series Editor John M. Walker School of Life and Medical Sciences University of Hertfordshire Hatfield, Hertfordshire, UK

For further volumes: http://www.springer.com/series/7651

For over 35 years, biological scientists have come to rely on the research protocols and methodologies in the critically acclaimed Methods in Molecular Biology series. The series was the first to introduce the step-by-step protocols approach that has become the standard in all biomedical protocol publishing. Each protocol is provided in readily-reproducible step-bystep fashion, opening with an introductory overview, a list of the materials and reagents needed to complete the experiment, and followed by a detailed procedure that is supported with a helpful notes section offering tips and tricks of the trade as well as troubleshooting advice. These hallmark features were introduced by series editor Dr. John Walker and constitute the key ingredient in each and every volume of the Methods in Molecular Biology series. Tested and trusted, comprehensive and reliable, all protocols from the series are indexed in PubMed.

Protein Tyrosine Phosphatases Methods and Protocols Second Edition

Edited by

Damien Thévenin Department of Chemistry, Lehigh University, Bethlehem, PA, USA

Jörg P. Müller Center for Molecular Biomedicine, Friedrich Schiller University Jena, Jena, Germany

Editors Damien The´venin Department of Chemistry Lehigh University Bethlehem, PA, USA

Jo¨rg P. Mu¨ller Center for Molecular Biomedicine Friedrich Schiller University Jena Jena, Germany

ISSN 1064-3745 ISSN 1940-6029 (electronic) Methods in Molecular Biology ISBN 978-1-0716-3568-1 ISBN 978-1-0716-3569-8 (eBook) https://doi.org/10.1007/978-1-0716-3569-8 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Humana imprint is published by the registered company Springer Science+Business Media, LLC, part of Springer Nature. The registered company address is: 1 New York Plaza, New York, NY 10004, U.S.A. Paper in this product is recyclable.

Preface We are excited to introduce the second edition of this book to the scientific community. This collaborative effort showcases the latest advancements in protein tyrosine phosphatase (PTP) research and equips readers with cutting-edge methods to investigate these essential enzymes and discover new inhibitors. By precisely counterbalancing the transient activity of protein kinases through dephosphorylation, PTPs play critical roles in regulating cellular signaling pathways. They regulate many biological processes, including cell cycle control, growth, proliferation, differentiation, transformation, and synaptic plasticity. On the other hand, dysregulation of PTPs has been linked to major diseases such as metabolic diseases, autoimmune disorders, and cancer, making many members of the PTP family attractive therapeutic targets. All PTPs are defined by a conserved active site motif C(X)5R (where C is the catalytic Cysteine, X is any amino acid, and R is an Arginine) that specifies a common two-step, nucleophile-based catalytic mechanism. Classical PTPs dephosphorylate specific phosphotyrosine residues from protein substrates, while dual-specific phosphatases (DUSP) act on phospho-tyrosine, phospho-serine, and phospho-threonine residues. Classical PTPs are further divided into single-pass receptor and non-receptor phosphatases. Their activity and specificity are tightly regulated by their expression, localization, protein-protein complex formation, and reversible oxidation of their active site. Getting further insights into these processes in mammalian cells and animal models is necessary for an in-depth understanding of cell function and controlling malign signal transduction. The emergence of new techniques, such as proteomics, genomics, and structural biology, has enabled the field to better appreciate the complexity and diversity of these enzymes in vitro, in cells, and in animal models. Moreover, advances in pharmacology and drug design have led to the developing of novel therapeutics targeting PTPs. This new edition reflects these exciting developments and presents comprehensive and up-to-date methods from experts in the field. We have been fortunate to assemble a team of such outstanding authors and are thankful for their contributions. We hope this book will be a valuable resource for both seasoned researchers and newcomers to the field and will inspire discoveries and accelerate progress in the field of PTP, signal transduction, and drug development. Bethlehem, PA, USA Jena, Germany

Damien The´venin ¨ ller Jo¨rg P. Mu

v

Contents Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Contributors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1 Induction of Translational Readthrough on Protein Tyrosine Phosphatases Targeted by Premature Termination Codon Mutations in Human Disease. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Leire Torices, Caroline E. Nunes-Xavier, Janire Mingo, Sandra Luna, Asier Erramuzpe, Jesu´s M. Corte´s, and Rafael Pulido 2 Understanding Pseudophosphatase Function Through Biochemical Interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Shanta´ D. Hinton 3 CRISPR/Cas9-Mediated Modification of PTP Expression . . . . . . . . . . . . . . . . . . . Carolin Lossius, Anne Kresinsky, Laura Quiet, ¨ ller and Jo¨rg P. Mu 4 Osteoclast Methods in Protein Phosphatase Research . . . . . . . . . . . . . . . . . . . . . . . Nina Reuven, Maayan Barnea-Zohar, and Ari Elson 5 OT-I TCR Transgenic Mice to Study the Role of PTPN22 in Anti-cancer Immunity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Rebecca J. Brownlie, Rose Zamoyska, and Robert J. Salmond 6 Protein Tyrosine Phosphatase Studies in Zebrafish . . . . . . . . . . . . . . . . . . . . . . . . . . Danie¨lle T. J. Woutersen, Jisca Majole´e, and Jeroen den Hertog 7 Examining Phosphatases Through Immunofluorescent Microscopy . . . . . . . . . . . Caroline N. Smith and Jessica S. Blackburn 8 Identification of Protein Tyrosine Phosphatase (PTP) Substrates. . . . . . . . . . . . . . Sravan Perla, Bin Qiu, Sam Dorry, Jae-Sung Yi, and Anton M. Bennett 9 Kinase-Catalyzed Biotinylation to Identify Phosphatase Substrates (K-BIPS). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Hannah J. Bremer and Mary Kay H. Pflum 10 System-Level Analysis of the Effects of RPTPs on Cellular Signaling Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jacqueline Gerritsen, Sophie Rizzo, Damien The´venin, and Forest M. White 11 Detection of Protein Tyrosine Phosphatase Interacting Partners by Mass Spectrometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Martina Samiotaki, George Panayotou, and Panagiotis Chandris 12 Detecting PTP Protein–Protein Interactions by Fluorescent Immunoprecipitation Analysis (FIPA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Natalia Kruglova and Alexander Filatov

vii

v ix

1

21 43

57

81 93 111 123

135

153

165

181

viii

13

14

15

16

17

18

19

Contents

Identifying Transmembrane Interactions in Receptor Protein Tyrosine Phosphatase Homodimerization and Heterodimerization . . . . . . . . . . . . . . . . . . . . Sophie Rizzo and Damien The´venin Preparation of Oxidized and Reduced PTP4A1 for Structural and Functional Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ganesan Senthil Kumar Measuring the Reversible Oxidation of Protein Tyrosine Phosphatases Using a Modified Cysteinyl-Labeling Assay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Avinash D. Londhe and Benoit Boivin Identification and Optimization of Protein Tyrosine Phosphatase Inhibitors Via Fragment Ligation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Markus Tiemann and Jo¨rg Rademann Targeting Nonconserved and Pathogenic Cysteines of Protein Tyrosine Phosphatases with Small Molecules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Anthony C. Bishop and Anna Serbina In Vitro Phosphatase Assays for the Eya2 Tyrosine Phosphatase . . . . . . . . . . . . . . Christopher Alderman, Aaron Krueger, John Rossi, Heide L. Ford, and Rui Zhao High-Throughput Discovery and Characterization of Covalent Inhibitors for Protein Tyrosine Phosphatases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Zihan Qu, Aaron D. Krabill, and Zhong-Yin Zhang

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

195

211

223

239

271 285

301 317

Contributors CHRISTOPHER ALDERMAN • Department of Biochemistry and Molecular Genetics, University of Colorado Anschutz Medical Campus, Aurora, CO, USA MAAYAN BARNEA-ZOHAR • Department of Molecular Genetics, The Weizmann Institute of Science, Rehovot, Israel ANTON M. BENNETT • Department of Pharmacology, Yale School of Medicine, New Haven, CT, USA ANTHONY C. BISHOP • Department of Chemistry, Amherst College, Amherst, MA, USA JESSICA S. BLACKBURN • Molecular and Cellular Biochemistry Department, University of Kentucky, Lexington, KY, USA BENOIT BOIVIN • Department of Nanobioscience, College of Nanoscale Science and Engineering, SUNY Polytechnic Institute, Albany, NY, USA; Department of Nanoscale Science and Engineering, University at Albany, Albany, NY, USA HANNAH J. BREMER • Department of Chemistry, Wayne State University, Detroit, MI, USA REBECCA J. BROWNLIE • Leeds Institute of Medical Research at St James’s, University of Leeds, Wellcome Trust Brenner Building, St James’s University Hospital, Leeds, UK PANAGIOTIS CHANDRIS • Institute for Bioinnovation, Biomedical Sciences Research Center “Alexander Fleming”, Vari, Greece; Department of Cellular and Molecular Neurobiology, Hellenic Pasteur Institute, Athens, Greece JESU´S M. CORTE´S • Biobizkaia Health Research Institute, Barakaldo, Spain; Ikerbasque, The Basque Foundation for Science, Bilbao, Spain; Cell Biology and Histology Department, University of the Basque Country (UPV/EHU), Leioa, Spain JEROEN DEN HERTOG • Hubrecht Institute-KNAW and University Medical Center Utrecht, Utrecht, The Netherlands; Institute Biology Leiden, Leiden University, Leiden, The Netherlands SAM DORRY • Department of Pharmacology, Yale School of Medicine, New Haven, CT, USA ARI ELSON • Department of Molecular Genetics, The Weizmann Institute of Science, Rehovot, Israel ASIER ERRAMUZPE • Biobizkaia Health Research Institute, Barakaldo, Spain; Ikerbasque, The Basque Foundation for Science, Bilbao, Spain ALEXANDER FILATOV • National Research Center, Institute of Immunology of Federal Medical Biological Agency of Russia, Moscow, Russia HEIDE L. FORD • Department of Pharmacology, University of Colorado Anschutz Medical Campus, Aurora, CO, USA JACQUELINE GERRITSEN • Department of Biological Engineering, Massachusetts Institute of Technology, Cambridge, MA, USA; Koch Institute for Integrative Cancer Research, Massachusetts Institute of Technology, Cambridge, MA, USA SHANTA´ D. HINTON • Department of Biology, College of William and Mary, Williamsburg, VA, USA AARON D. KRABILL • Department of Medicinal Chemistry and Molecular Pharmacology, Purdue University, West Lafayette, IN, USA ANNE KRESINSKY • Institute for Molecular Cell Biology, CMB - Center for Molecular Biomedicine; University Hospital Jena, Jena, Germany; Regeneration of Hematopoiesis , Leibniz Institute on Aging, Fritz Lipmann Institute, Jena, Germany

ix

x

Contributors

AARON KRUEGER • Department of Biochemistry and Molecular Genetics, University of Colorado Anschutz Medical Campus, Aurora, CO, USA; KBI Biopharma, Inc., Boulder, CO, USA NATALIA KRUGLOVA • Cell and Gene Technology Group, Center for Precision Genome Editing and Genetic Technologies for Biomedicine, Institute of Gene Biology RAS, Moscow, Russia; National Research Center, Institute of Immunology of Federal Medical Biological Agency of Russia, Moscow, Russia GANESAN SENTHIL KUMAR • Integrative Structural Biology Laboratory, National Institute of Immunology, New Delhi, India AVINASH D. LONDHE • Department of Nanobioscience, College of Nanoscale Science and Engineering, SUNY Polytechnic Institute, Albany, NY, USA CAROLIN LOSSIUS • Institute for Molecular Cell Biology, CMB - Center for Molecular Biomedicine; University Hospital Jena, Jena, Germany SANDRA LUNA • Biobizkaia Health Research Institute, Barakaldo, Spain JISCA MAJOLE´E • Hubrecht Institute-KNAW and University Medical Center Utrecht, Utrecht, The Netherlands JANIRE MINGO • Biobizkaia Health Research Institute, Barakaldo, Spain JO¨RG P. MU¨LLER • Institute for Molecular Cell Biology, CMB - Center for Molecular Biomedicine; University Hospital Jena, Jena, Germany CAROLINE E. NUNES-XAVIER • Biobizkaia Health Research Institute, Barakaldo, Spain; Institute for Cancer Research, Oslo University Hospital, The Norwegian Radium Hospital, Oslo, Norway GEORGE PANAYOTOU • Institute for Bioinnovation, Biomedical Sciences Research Center “Alexander Fleming”, Vari, Greece SRAVAN PERLA • Department of Pharmacology, Yale School of Medicine, New Haven, CT, USA MARY KAY H. PFLUM • Department of Chemistry, Wayne State University, Detroit, MI, USA RAFAEL PULIDO • Biobizkaia Health Research Institute, Barakaldo, Spain; Ikerbasque, The Basque Foundation for Science, Bilbao, Spain BIN QIU • Department of Pharmacology, Yale School of Medicine, New Haven, CT, USA ZIHAN QU • Department of Chemistry, Purdue University, West Lafayette, IN, USA LAURA QUIET • Institute for Molecular Cell Biology, CMB - Center for Molecular Biomedicine; University Hospital Jena, Jena, Germany ¨ JORG RADEMANN • Institute of Pharmacy, Freie Universit€ at Berlin, Berlin, Germany NINA REUVEN • Department of Molecular Genetics, The Weizmann Institute of Science, Rehovot, Israel SOPHIE RIZZO • Department of Chemistry, Lehigh University, Bethlehem, PA, USA JOHN ROSSI • Department of Biochemistry and Molecular Genetics, University of Colorado Anschutz Medical Campus, Aurora, CO, USA ROBERT J. SALMOND • Leeds Institute of Medical Research at St James’s, University of Leeds, Wellcome Trust Brenner Building, St James’s University Hospital, Leeds, UK MARTINA SAMIOTAKI • Institute for Bioinnovation, Biomedical Sciences Research Center “Alexander Fleming”, Vari, Greece ANNA SERBINA • Department of Chemistry, Amherst College, Amherst, MA, USA CAROLINE N. SMITH • Molecular and Cellular Biochemistry Department, University of Kentucky, Lexington, KY, USA DAMIEN THE´VENIN • Department of Chemistry, Lehigh University, Bethlehem, PA, USA at Berlin, Berlin, Germany MARKUS TIEMANN • Institute of Pharmacy, Freie Universit€

Contributors

xi

LEIRE TORICES • Biobizkaia Health Research Institute, Barakaldo, Spain FOREST M. WHITE • Department of Biological Engineering, Massachusetts Institute of Technology, Cambridge, MA, USA; Koch Institute for Integrative Cancer Research, Massachusetts Institute of Technology, Cambridge, MA, USA DANIE¨LLE T. J. WOUTERSEN • Hubrecht Institute-KNAW and University Medical Center Utrecht, Utrecht, The Netherlands JAE-SUNG YI • Alphina Therapeutics, New Haven, CT, USA ROSE ZAMOYSKA • Institute of Immunology and Infection Research, University of Edinburgh, Ashworth Laboratories, Edinburgh, UK ZHONG-YIN ZHANG • Department of Chemistry, Purdue University, West Lafayette, IN, USA; Department of Medicinal Chemistry and Molecular Pharmacology, Purdue University, West Lafayette, IN, USA; Institute for Cancer Research, Purdue University, West Lafayette, IN, USA; Institute for Drug Discovery, Purdue University, West Lafayette, IN, USA RUI ZHAO • Department of Biochemistry and Molecular Genetics, University of Colorado Anschutz Medical Campus, Aurora, CO, USA

Chapter 1 Induction of Translational Readthrough on Protein Tyrosine Phosphatases Targeted by Premature Termination Codon Mutations in Human Disease Leire Torices, Caroline E. Nunes-Xavier, Janire Mingo, Sandra Luna, Asier Erramuzpe, Jesu´s M. Corte´s, and Rafael Pulido Abstract Nonsense mutations generating premature termination codons (PTCs) in various genes are frequently associated with somatic cancer and hereditary human diseases since PTCs commonly generate truncated proteins with defective or altered function. Induced translational readthrough during protein biosynthesis facilitates the incorporation of an amino acid at the position of a PTC, allowing the synthesis of a complete protein. This may evade the pathological effect of the PTC mutation and provide new therapeutic opportunities. Several protein tyrosine phosphatases (PTPs) genes are targeted by PTC in human disease, the tumor suppressor PTEN being the more prominent paradigm. Here, using PTEN and laforin as examples, two PTPs from the dual-specificity phosphatase subfamily, we describe methodologies to analyze in silico the distribution and frequency of pathogenic PTC in PTP genes. We also summarize laboratory protocols and technical notes to study the induced translational readthrough reconstitution of the synthesis of PTP targeted by PTC in association with disease in cellular models. Key words Protein translation, Translational readthrough, Nonsense mutation, Premature termination codon, PTEN, Laforin

1

Introduction Nonsense mutations by single-nucleotide substitution, generating premature termination codons (PTCs) in the protein-coding region of the targeted genes, are frequently found in association with human disease. These include germline mutations linked to hereditary diseases and tumor somatic mutations associated with cancer [1, 2]. For instance, the APC gene, coding for the adenomatous polyposis coli tumor suppressor protein, is heavily targeted by PTC in the germline of patients with familial adenomatous polyposis and sporadic colorectal cancers [3]. Another emblematic example is the DMD gene, which encodes the dystrophin protein

Damien The´venin and Jo¨rg P. Mu¨ller (eds.), Protein Tyrosine Phosphatases: Methods and Protocols, Methods in Molecular Biology, vol. 2743, https://doi.org/10.1007/978-1-0716-3569-8_1, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

1

2

Leire Torices et al.

and whose alterations cause Duchenne muscular dystrophy. Many of the disease-associated DMD mutations found are nonsense mutations generating truncated dystrophin proteoforms [4]. The high potential pathogenicity of PTC is explained by the generation of truncated proteins and by the degradation of the PTC-targeted mRNA by nonsense-mediated mRNA decay (NMD) [5]. In the case of cancer-related diseases, nonsense mutations are relatively abundant in most tumor suppressor genes, frequently due to the combination of the mutation’s pathogenicity and the presence of DNA sequences prone to mutation [6, 7]. Translational readthrough of termination codons is the biosynthetic incorporation of an amino acid in the position corresponding to a termination codon (TAA, TAG, and TGA). In mammals, translational readthrough occurs because of competition to enter the ribosome acceptor site (A-site) between near-cognate tRNAs and the release factors eRF1 and eRF3. These near-cognate tRNAs have a sufficiently stable interaction to allow the translation machinery to continue reading the next in-frame codon, incorporating a non-random amino acid and facilitating translation until the natural termination codon, thus generating a full-length protein. Under normal conditions, basal readthrough of natural termination codons in mammalian cells takes place with a very low frequency, within the range of 0.001–0.1%, mainly due to the influence of downstream molecular signals in the mRNA that mark bona fide natural termination of protein translation. The basal readthrough of PTC, caused by nonsense mutations in gene coding sequences, is higher than the readthrough of natural termination codons. Readthrough inducers can efficiently increase it without significantly increasing natural termination codons readthrough. Several readthrough inducer compounds directly bind to the A-site of the ribosomes and induce conformational changes in the ribosome decoding center, which facilitates the incorporation of nearcognate tRNAs. This is the case with some aminoglycoside antibiotics such as G418/geneticin and gentamicin [8– 12]. Non-aminoglycoside compounds can also facilitate translational readthrough, although their mechanisms of action are less understood [13]. Translational readthrough of PTC results in the biosynthesis of a full-length non-truncated protein with variable efficiency, mainly depending on the type of PTC and its nucleotide context, which can rescue the pathogenicity of the PTC mutation. This makes the experimental analysis of the readthrough responses of genes of interest targeted by PTC of special importance. Since the possibility of generation of non-functional full-length proteins by translational readthrough exists due to the incorporation of nonwild-type residues at the PTC site [14], it is also essential to verify the reconstitution of the function of the readthrough protein products experimentally [15, 16]. Thus, the potential benefits of induced readthrough as a therapeutic alternative for specific genetic diseases must be explored for each PTC mutation [17, 18].

Translational Readthrough of Protein Tyrosine Phosphatases

3

Protein tyrosine phosphatases (PTPs) are major regulators of cell and tissue homeostasis and development, and their involvement in human disease has been widely documented [19–21]. Table 1 contains information on PTP genes targeted by PTC germline mutations, as reported in the ClinVar database (https://www. ncbi.nlm.nih.gov/clinvar/), with an indication of the disease conditions associated with the mutations. Figure 1 illustrates the abundance of different PTCs found in these PTPs. Several PTPs from the dual-specificity phosphatase (DUSP) subfamily display a relatively high number of disease-associated PTCs. These include some myotubularins (MTM1 and SBF2, causing X-linked myotubular myopathy and Charcot-Marie-Tooth disease 4B2, respectively), laforin, and the PTEN tumor suppressor. Laforin is a glycogen phosphatase encoded in the EPM2A gene whose mutations cause Lafora disease, a progressive myoclonus epilepsy [22]. PTEN is the PTP gene more abundantly targeted by PTC in human diseases, both in the germline of patients causing PTEN hamartoma tumor syndrome (PHTS) and in sporadic tumors [16, 23]. PTPN11 is the only gene from the classical PTP subfamily showing a relevant number of residues targeted by PTC, which are associated with metachondromatosis. The presence of PTC in FIG4 (associated with Charcot-Marie-Tooth disease 4J) and EYA1 (associated with branchio-oto-renal syndrome) and EYA4 (associated with nonsyndromic hearing loss) stand out from the SAC and EYA PTP subfamilies, respectively. In this chapter, we show methodologies to analyze the collection of PTC (PTCome) associated with disease from PTP genes of interest and laboratory protocols to study the induction of translational readthrough of PTP targeted by PTC, taking PTEN and laforin as two representative examples.

2

Materials All solutions are prepared in double-distilled, deionized MilliQ filtered water. Plasticware is autoclaved or sterilized by ethylene oxide. Cell culture and transfection procedures require sterile conditions.

2.1 In Silico Analysis of the PTCome Distribution on PTP

1. An updated web browser, such as Google Chrome, Mozilla Firefox, or Safari. 2. Access to HGMD (Human Gene Mutation Database), ClinVar, or COSMIC mutation databases (see Note 1). 3. PTCMAKER program (generated from us [16]), suitable for Microsoft Windows (https://github.com/translationalreadthrough-on-ptps/ptc-maker) (see Note 2). 4. R and RStudio programs, version 1.1.463+. 5. stringr, dplyr, and stats R packages.

4

Leire Torices et al.

Table 1 PTP targeted by premature termination codons in human disease

a

Genea

Proteinb

Number of amino acidsc

Number of different PTCsd

PTPRA

RPTPα

1637

2

–

PTPRC

CD45

1306

4

Immunodeficiency

PTPRQ

PTPRQ

2999

6

Hearing loss

PTPN4

PTP-MEG1

926

3

Neurodevelopmental disorder

PTPN11

SHP2

593

12

PTPN23

HDPTP

802

4

Neurodevelopmental alterations

EPM2A

Laforin

331

12

Lafora disease, Progressive myoclonic epilepsy

CDC14A

CDC14A

623

7

PTEN

PTEN

403

106

TNS2

C1-TEN

1419

1

–

DNAJC6

Auxilin

998

7

Juvenile onset Parkinson’s disease

MTM1

Myotubularin

603

48

Severe X-linked myotubular myopathy

MTMR2

MTMR2

1882

19

Charcot-Marie-tooth disease type 4B1

SBF1

MTMR5

1893

9

Charcot-Marie-tooth disease type 4B3

MTMR10

MTMR10

777

7

Karyomegalic interstitial nephritis

SBF2

MTMR13

643

25

MTMR14

MTMR14

650

2

FIG4

SAC3

907

27

Charcot-Marie-tooth disease type 4 J

SYNJ1

INPP5G

1612

12

Developmental and epileptic encephalopathy, early-onset Parkinson’s disease

INPP4A

INPP4A

972

1

Intellectual disability

ACP1

LMW-PTP

159

1

–

EYA1

EYA1

592

33

Melnick-Fraser syndrome, branchio-oto-renal syndrome

EYA4

EYA4

640

17

Dilated cardiomyopathy, nonsyndromic hearing loss

ACP4

ACPT

426

1

Condition(s)e

Metachondromatosis

Hearing loss PTEN hamartoma tumor syndrome, Autism spectrum disorder

Charcot-Marie-tooth disease type 4B2 Centronuclear myopathy

–

PTP are grouped in subfamilies and ordered according to [35]. Official gene names are provided b Alternative protein names are provided c Amino acid number of the longer isoform is provided d Data are for germline PTC from ClinVar database (https://www.ncbi.nlm.nih.gov/clinvar/). The number of different PTCs along the protein coding sequence is provided. Note that ClinVar is not a comprehensive database e As indicated in ClinVar database

Translational Readthrough of Protein Tyrosine Phosphatases

5

110

PTC number

50 40 30 20 10 0 28 26

% PTC

12 10 8 6 4 2

PTPRA PTPRC PTPRQ PTPN4 PTPN11 PTPN23 EPM2A CDC14A PTEN TNS2 DNAJC6 MTM1 MTMR2 SBF1 MTMR10 SBF2 MTMR14 FIG4 SYNJ1 INPP4A ACP1 EYA1 EYA4 ACP4

0

Classical PTP

MTM DUSP

SAC

EYA

Fig. 1 Representation of the number of different PTCs identified in PTP genes in association with human disease, as reported in ClinVar database (https://www. ncbi.nlm.nih.gov/clinvar). The top panel shows the total number of different PTCs annotated for each gene. The bottom panel shows the percentage of different PTCs with respect to the total number of amino acids from each protein. Note that these are qualitative data, and they do not represent the number of times each PTC has been annotated or reported 2.2 Assessing the Induction of Translational Readthrough on PTP

1. Tissue-culture plates. 2. Simian kidney COS-7 cells are suitable for transfection and transient overexpression of recombinant proteins (see Note 3). 3. Complete medium: Dulbecco’s Modified Eagle Medium (DMEM) containing high glucose supplemented with 5% heat-inactivated FBS, 1 mM L-glutamine, 100 U/mL penicillin, and 0.1 mg/mL streptomycin. 4. Trypsin-EDTA solution. 5. cDNAs of PTP wild-type and PTC mutations of interest cloned into suitable mammalian expression vectors (see Note 4). 6. Transfection reagents (see Note 5). 7. Translational readthrough inducers (see Note 6). 8. Lysis buffer: Mammalian Protein Extraction Reagent (M-PER, Thermo Fisher Scientific) supplemented with protease and phosphatase inhibitors (cOmplete EDTA-free and PhosSTOP (Roche)).

6

Leire Torices et al.

9. Sample buffer: NuPAGE™ LDS (4X) (Thermo Fisher Scientific), containing 5% 2-mercaptoethanol. 10. Polyvinylidene fluoride (PVDF) protein transfer membranes: Immobilon-FL Transfer Membrane (Merck Millipore), compatible with the detection system used (see Note 7). 11. Transfer buffer: 48 mM Tris base, 39 mM glycine, 0.037% SDS, and 20% methanol. 12. Prestained molecular weight standard protein markers. 13. Primary antibodies against the recombinant protein under study or recognizing a suitable artificial tag, and primary antibodies against an endogenous reference protein (see Note 8). 14. Fluorochrome-conjugated secondary antibody, compatible with the detection system used (see Note 7). 15. Immunoblot blocking buffer: Odyssey Blocking Buffer (OBB buffer, LI-COR Biosciences) diluted 1:1 in PBS. 16. Immunoblot washing buffer: 50 mM Tris–HCl, pH 7.5, 150 mM NaCl, 5 mM EDTA, 0.05% Triton X-100, and 0.25% gelatin.

3

Methods The incorporation in the clinic routine of genetic studies as standard procedures for patient diagnosis provides a wealth of information on the molecular basis of human disease. Many genetic alterations associated with diseases, including alterations targeting the PTP gene family, consist of nonsense single-nucleotide substitutions generating premature termination codons (PTCs) in the coding region of specific proteins. This makes the experimental analysis of these PTCs important to develop efficient precision therapies for carrier patients. The first method (see Subheading 3.1) that we present in this chapter consists of in silico methodology to visualize and analyze the qualitative and quantitative distribution of the collection of PTC (PTCome) found on genes of interest in association with disease (PTEN and laforin dualspecificity phosphatases are used as examples), as annotated in the literature and in gene mutation and gene variant databases. The qualitative analysis of PTC distribution along the domains of the protein under study may provide predictive genotype/phenotype information and potential attributes of pathogenicity to the distinct protein regions. In contrast, the quantitative analysis of PTC informs on the presence of mutational hotspots, which are influenced by both the functional output of the PTC mutation and the mutagenic chemical properties of specific nucleotide sequences (see Fig. 2 and comments in the legend). In many cases, proteoforms generated by PTC are unstable, and their functional output is

Translational Readthrough of Protein Tyrosine Phosphatases

7

A Germline PTCome Potential PTCome

Germline PTCome Potential PTCome

PTEN N 1 7 exons 1

B

Laforin PTP

2 3 4

5

C 353 403

C2

185 6

7

8

9

N 1 exons

35

R335X

R130X

10

25 20 15 10

1

124

157 2

PTP 3

C 326 331 4

12

R233X

PTC number

PTC number

30

CBM

R241X

8 6 4

5

2

0

0

Fig. 2 Representation of qualitative and quantitative PTEN and laforin (EPM2A gene) PTCome distribution. (a) Qualitative kernel density plots represent the density of PTC along the protein amino acid sequence. This type of representation is qualitative and does not take into consideration the number of times each PTC has been annotated. The black line represents potential PTCome, and the blue line represents germline-associated PTCome. At the bottom of the figure, a schematic depiction of each protein is shown, with indication of amino acid numbering (PTEN, accession NP_000305.3; laforin, accession NP_005661.1). Structured domains are represented as boxes, according to [23] and [36], and disordered regions are represented as lines (PTP, protein tyrosine phosphatase domain; C2, C2-membrane binding domain; CBM, carbohydrate-binding module; N, N-terminus; C, C-terminus). A schematic of PTEN and laforin (EPM2A gene) exons is also shown. (b) Quantitative barplots indicate the number of times each PTC has been reported in disease. Results for PTEN are from ClinVar database, and results for laforin are from ClinVar and from literature search. The Arg (R) residues more targeted by PTC are indicated, corresponding to CGA to TGA substitutions with a high mutation rate

similar to gene loss. In addition, PTC truncated proteoforms can exist in the cell displaying pathogenic altered functions. In conclusion, qualitative and quantitative distribution of the PTCome associated with disease display gene-specific patterns, which need to be analyzed individually in the context of pathogenicity caused by loss or alteration of specific protein functions. The second method (see Subheading 3.2) illustrates the methodology to study the induction of biosynthetic translational readthrough of genes targeted by PTC, as monitored by immunoblot with specific antibodies of cell lysates from mammalian cells transiently transfected with cDNAs encoding the PTC gene variants of interest. During translational readthrough, the incorporation of an amino acid in the position of the PTC is achieved, which evades premature translation termination and permits the biosynthesis of a full-length protein. The efficiency of induced readthrough in these

Leire Torices et al.

C

EV -

+

PTEN-HA WT R233X + +

PTEN WT R233X + +

Erythromycin

Tobramycin

anti-PTEN G418

anti-PTEN

anti-GAPDH

PTEN R11X R15X + +

D EV -

WT +

-

+

R130X +

G418 anti-PTEN

anti-HA

*

anti-GAPDH

Readthrough efficiency (%)

B

-

Amikacin

-

-

R233X Gentamicin

C2

C 353 403

G418

185

WT

EV

Erythromycin

PTP M35

PTEN

R233

Tobramycin

N 1 7

R130

Amikacin

PTEN R11 R15

Gentamicin

A

G418

8

-

G418

R11X R15X R130X

Fig. 3 Translational readthrough induction of PTEN PTC mutations. (a) Schematic of PTEN domains, with indication of the residues targeted by PTC and tested for readthrough in (b–d). The methionine residue is also indicated at position 35 (M35), which is used as an alternative initiation methionine in the presence of upstream PTC, generating the PTEN-N-terminal-truncated protein PTEN M35 [16]. (b) Readthrough-induced expression of the disease-associated PTEN R233X [p.(Arg233Ter), TGA] mutation. COS-7 cells were transfected with pRK5 plasmids containing untagged PTEN (PTEN) or C-terminal hemagglutinin-tagged PTEN (PTEN-HA), wild-type (WT) or the R233X mutation, and 24 h after transfection cells were kept untreated or incubated in the presence of G418/geneticin (200 μg/mL) for an additional 24 h, as indicated. pRK5 PTEN has been described in [37]. pRK5 PTEN-HA was generated from pRK5 PTEN adding the HA epitope at PTEN C-terminus by PCR. Mutagenesis to create the PTC mutations was performed as described in [38]. EV, empty vector. Proteins were resolved in 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gels and were detected by immunoblot using the anti-PTEN 6H2.1 mAb, which recognizes a PTEN C-terminal epitope (residues 392–398) [39] (top panel), or the anti-HA 12CA5 mAb (bottom panel). The arrow (→) indicates the migration of full-length PTEN or PTEN-HA. (c) Induced expression of PTEN R233X mutation by different readthrough inducer compounds. COS-7 cells were processed as in A and treated for 24 h with the indicated compounds (-, no treatment; G418, 200 μg/mL; gentamicin, 800 μg/mL; amikacin, 2 mg/mL; tobramycin, 800 μg/mL; erythromycin, 175 μg/mL). Proteins were detected by immunoblot using the antiPTEN 6H2.1 mAb and an anti-GAPDH antibody. The arrow (→) indicates the migration of full-length PTEN. (d) Readthrough-induced expression of the disease-associated PTEN R11X [p.(Arg11Ter), TGA], R15X [p.(Arg15Ter), TGA], and R130X [p.(Arg130Ter), TGA] N-terminal mutations. COS-7 cells were processed as in (a) and treated for 24 h with G418 (200 μg/mL), as indicated. Proteins were detected by immunoblot as in (b). In the left panel, a representative immunoblot is shown. The arrow (→) indicates the migration of full-length PTEN. The asterisk (*) indicates the migration of PTEN M35. In the right panel, the quantification of the readthroughinduced expression of the PTEN PTC mutations is shown. The readthrough efficiency is represented as the mean ± SD of the percentage of full-length PTEN expression from each PTEN PTC mutation with respect to PTEN wild type, from at least two different experiments. PTEN amino acid numbering is according to NP_000305

experiments mainly depends on the efficient translation of the protein product, the properties of the readthrough inducer, and the type of PTC and its nucleotide context (see Figs. 3, 4, and 5). This protocol is intended to detect the translation, upon

Translational Readthrough of Protein Tyrosine Phosphatases

A

C

Laforin R241

Laforin

WT +

-

+

Y86X +

R241X +

-

G418

-

PTC124

EV -

G418

Laforin

WT Amikacin

R241X

Gentamicin

B

C 326 331

G418

PTP

157

PTC124

124

Amikacin

CBM

Gentamicin

Y86

N 1

9

antilaforin

*

* antiGAPDH

Fig. 4 Translational readthrough induction of laforin (EPM2A gene) PTC mutations. (a) Schematic of laforin domains, with indication of the residues targeted by PTC and tested for readthrough in (b) and (c). (b) Readthrough-induced expression of the disease-associated laforin Y86X [p.(Tyr86Ter), TAG] and R241X [p. (Arg241Ter), TGA] mutations. COS-7 cells were transfected with pRK5 laforin, processed as in Fig. 3 and treated for 24 h with G418 (200 μg/mL), as indicated. Proteins were detected by immunoblot using anti-laforin antibody recognizing laforin residues 131–144 (Sigma Aldrich, SAB2500581), followed by anti-GAPDH antibody, as shown. The arrow (→) indicates the migration of full-length laforin. The asterisk (*) indicates the migration of laforin R241X truncated proteoform. pRK5 laforin was obtained by removing the HA epitope from pCMV HA-laforin [40] by PCR followed by subcloning into pRK5. Mutagenesis to create the PTC mutations was performed as described in [38]. (c) Induced expression of laforin R241X mutation by different readthrough inducer compounds. COS-7 cells were processed as in a and treated for 24 h with the indicated compounds (-, no treatment; G418, 200 μg/mL; gentamicin, 800 μg/mL; amikacin, 2 mg/mL; PTC124/Ataluren, 1 μg/ mL). Proteins were detected by immunoblot using the anti-laforin and anti-GAPDH antibodies, as in (b). The arrow (→) indicates the migration of full-length laforin. The asterisk (*) indicates the migration of laforin R241X truncated proteoform. Laforin amino acid numbering is according to NP_005661

readthrough induction, of the full-length protein under study, in contrast to the protocols that detect the activity of a reporter enzyme (such as luciferase) fused with the peptide containing the PTC of interest. The protocol can also detect the truncated proteoforms generated from the PTC. Our method has the advantage of determining the readthrough efficiency in the presence of the complete protein coding nucleotide sequence and the absence of secondary effects of the readthrough inducers on the reporter protein [24]. This facilitates the testing of the functional properties of the reconstituted protein (see Note 9). As cDNA is used, which lacks exon–intron junction nucleotide sequences, the effect of NMD is not manifested in this protocol, providing reliable information on the efficiency of the induced readthrough reconstitution of the full-length protein, independently of PTC-dependent mRNA decay.

10

Leire Torices et al.

A

PTEN EV -

C

PTEN R233X

WT R233X pRc/CMV pcDNA3 pcDNA6 pRK5 - + - + - + + - + - +

+

pSG5 - +

pMT2 - + G418

-

WT +

-

Y76X (TAG) +

Y76X (TAA) +

S59X (TAA) - +

WT +

G418

S59X (TGA) +

anti-PTEN

antiPTEN

anti-GAPDH

antiGAPDH

D

WT -

+

G165X (TGAG)

G165X (TGAC)

-

-

+

B

+

G418 anti-PTEN

+

-

R11X +

+

*

DEAE-Dextran

GeneJet

-

WT

R15X +

R130X +

L146X +

anti-GAPDH G418 anti-PTEN anti-GAPDH

anti-PTEN anti-GAPDH

E

Readthrough efficiency (%)

EV

40 30 20 10 0

TGA TAG TAA

A/G/T C A/G/T C A/G/T C +4

TGA

TAG

TAA

Fig. 5 Factors influencing readthrough efficiency. (a) Suitability of different expression plasmids to asses G418-induced readthrough. COS-7 cells were transfected with pRK5 plasmids (EV, empty vector; PTEN wild type [WT]; PTEN R233X) or cotransfected with pRK5 PTEN R233X and the indicated empty vectors (pRc/CMV and pCDNA3 [neomycin resistant gene]; pCDNA6 [blasticidin resistant gene]; pRK5, pSG5, and pMT2 [no mammalian antibiotic resistance gene]); 24 h after transfection cells were kept untreated or were treated with G418 (200 μg/mL), as indicated. Cells were transfected and processed by immunoblot as in Fig. 3, using anti-PTEN 6H2.1 and anti-GAPDH antibodies. The arrow (→) indicates the migration of full-length PTEN. (b) Comparison of transfection methods to monitor PTC readthrough efficiency. COS-7 cells were transfected using GeneJet™ (SignaGen Laboratories) or the diethylaminoethyl (DEAE)-dextran method [34] with pRK5 PTEN plasmids, as indicated (EV, empty vector). Readthrough was induced with G418 (200 μg/mL), and cells were processed for immunoblot as in (a). The arrow (→) indicates the migration of full-length PTEN. The asterisk (*) indicates the migration of PTEN M35. pRK5 PTEN M35 has been described in [16]. (c) Influence of the PTC identity on PTC readthrough efficiency. COS-7 cells were transfected with pRK5 PTEN plasmids containing nucleotide substitutions rendering two different PTCs from the same wild-type codon [p.(Tyr76Ter), c.228T>G (TAG), and c.228T>A (TAA); p.(Ser59Ter), c.176C>A (TAA), and c.176C>G (TGA)] and readthrough was assessed as in Fig. 3. (d) Influence of the nucleotide context on PTC readthrough efficiency. COS-7 cells were transfected with pRK5 PTEN plasmids containing the G165X PTC [p.(Gly165Ter), c.493G>T (TGAG)] or the G165X PTC with a nucleotide change in +4 position [p.(Gly165Ter), c.493G>T + c.496G>C (TGAC)] and readthrough was assessed as in Fig. 3. (e) Quantification of the readthrough efficiency of the diseaseassociated PTEN PTCome depending on PTC identity (TGA [13.15 ± 8.27], TAG [5.5 ± 4.93], TAA [2.99 ± 2.58]) (**P values from Student’s t test Import Dataset > From text (readr) or using the read_csv function. 4. Run the following code in RStudio (https://github.com/ translational-readthrough-on-ptps/kernel-plots, section #1) to obtain the list of potential PTC-mutated residues in a vector format. 3.1.2 GermlineAssociated PTCome

PTC generated by single-nucleotide substitution mutations found in the germline. 1. To obtain the germline-associated PTCome, browse for the desired PTP in the HGMD (https://www.hgmd.cf.ac.uk; https://digitalinsights.qiagen.com/products-overview/clini cal-insights-portfolio/human-gene-mutation-database/) database. Click on “Missense/Nonsense,” which displays a dataframe with missense and nonsense mutations together. Germline-associated PTCome can also be obtained from ClinVar (NCBI, https://www.ncbi.nlm.nih.gov/clinvar) database (see Note 1). 2. Import the downloaded file in RStudio by File > Import Dataset > From Excel or using the read_excel function. Additional PTCs described in the literature or other databases should be added manually if the number of different PTCs in HGMD is low. 3. Run the following code in RStudio (https://github.com/ translational-readthrough-on-ptps/kernel-plots, section #2) to obtain unique PTC-mutated residues in a vector format.

3.1.3 Cancer-Associated PTCome

PTC generated by single-nucleotide substitution somatic mutations found in tumors, of utility for PTP associated with tumor suppression of oncogenesis. 1. To obtain the cancer-associated PTCome, browse for the desired PTP in the COSMIC database (Wellcome Trust Sanger Institute; https://cancer.sanger.ac.uk/cosmic). Go to the “Filters” section, and in “Mutation type,” “Substitution” subheading, click on “Nonsense.” Click on “Apply filters.” Go to the “Variants” section to see the list of PTC-mutated residues (see Note 1). 2. Download/export the desired dataset in a comma-delimited (. csv) format. Import the downloaded file in RStudio by File > Import Dataset > From text (readr) or using the read_csv function, specifying a comma as a delimiter. Reliable cancer-

12

Leire Torices et al.

associated PTCome can be obtained only for PTP whose function is related to cancer and displays a significant number of PTC mutations distributed along their sequence. 3. Run the code in RStudio (https://github.com/translationalreadthrough-on-ptps/kernel-plots, section #3) to obtain unique PTC-mutated residues in a vector format. 3.1.4 PTCome Qualitative Representation: Kernel Density Plot

1. Insert in lines 55 and 56 of the code in RStudio (https:// github.com/translational-readthrough-on-ptps/kernel-plots, section #4) the name and the length in amino acidse of your protein of interest, respectively. (PTEN protein name and amino acid length are given as examples.) 2. Run RStudio with the code to obtain a Kernel density plot representation. PTCome vectors (potential PTCome, germline-associated PTCome, and cancer PTCome) are rescaled to 100 to facilitate comparisons between proteins. Mirror vectors are generated to avoid bias at boundaries, and a density is calculated as the sum of the parental vector and the mirror vectors. The final plot contains curves representing the potential PTCome (in black), the germline-associated PTCome (in blue), and the cancer-associated PTCome (in red) (see Note 10).

3.1.5 PTCome Quantitative Representation: Barplot of PTC Frequency (See Note 11)

1. To obtain the germline-associated PTC frequency, browse for the desired PTP in the ClinVar database (the HGMD database does not provide quantitative data on the mutations). Go to the “Mutation type,” “Substitution” section and click on “Nonsense.” 2. Click on the desired mutational variant and look for the number of submissions. 3. Represent it in a barplot. Examples of in silico analysis of the PTCome from the dualspecificity phosphatases PTEN and laforin (EPM2A gene) are provided in Fig. 2. More PTC mutations are found in the PTEN gene than in the EPM2A gene, which gives more robustness to the interpretation of the PTEN plots. In both cases, the kernel plots (see Fig. 2a) show a relative enrichment in the density of different PTC mutations in the N-terminal portion of the proteins and diminished density at their C-terminal regions. This suggests that these C-terminal regions are not essential for the pathogenic-related functions of the proteins. The barplots of PTC frequency (see Fig. 2b) highlight the existence of hotspot PTC mutations targeting Arg residues and resulting from high-frequency CpG to TpG transitions [25]. These methylation-dependent transitions depend on local GC content and the genomic region [26, 27].

Translational Readthrough of Protein Tyrosine Phosphatases

3.2 Assessing the Induction of Translational Readthrough on PTP

13

1. Plate cells in six well-culture plates (1.5 × 104 cells/cm2; about 1.5 × 105 cells/well) (see Notes 3 and 12). 2. After 24 h in culture, transfect each well with 1 μg of plasmid containing cDNA coding for the PTP under study, wild-type or PTC mutations, or with empty vector (see Notes 4 and 5). When designing the experiment, consider that you will always need a control well for each cDNA that will not be treated with the readthrough inducer. 3. After 24 h in culture, add the readthrough inducer (see Note 6). 4. After 24–48 h in culture, lyse the cells directly, adding 150 μL of ice-cold lysis buffer to each well. Keep the plate on ice for 10 min under balancing and transfer the lysate from each well to an Eppendorf tube. 5. Centrifuge the lysate 10,000 g, 10 min, 4 °C. Transfer the supernatant, containing the cellular protein extract, to a new Eppendorf tube. Cell lysates can be frozen at this step. 6. Mix 45 μL of cell lysate with 15 μL sample buffer (see Note 13). 7. Boil for 5 min, spin, and load the mix in a 10% SDS-PAGE gel. We routinely run 10% or 12% SDS-PAGE gels, depending on the size of the proteins to be resolved. Include a lane with prestained molecular weight standard protein markers. 8. Run the gel and transfer it to a PVDF membrane. Cut a piece of the membrane in the range of the migration of your full-length protein, the PTC-generated truncated protein, and the reference endogenous protein (see Note 14). 9. Perform standard immunoblot procedure, incubating the membrane with primary and secondary antibodies according to the specifications of the antibodies used (see Notes 7 and 8). 10. Quantify the intensity of the full-length band of the PTC protein variant under study, upon the readthrough induction conditions, in comparison with the wild-type protein. The readthrough efficiency can be represented as the percentage of expression with respect to the wild-type protein (see Notes 7 and 15). Examples of induction of translational readthrough of PTC from PTEN and laforin, and its monitoring by immunoblot, are provided in Figs. 3 and 4. In our experiments, G418 is the best readthrough inducer out of the compounds we have tested, followed by gentamicin (Figs. 3c and 4c). The use of C-terminal tagged PTEN (PTEN-HA) to monitor the readthrough (Fig. 3b) illustrates an alternative to the use of antibodies against the PTP under study (see Note 16). Figure 3d shows the generation of N-terminal truncated proteoforms (PTEN M35) by nonsense-

14

Leire Torices et al.

mediated translation reinitiation. This process occurs in the presence of PTC upstream of a methionine residue, which is different from the canonical initiation methionine [16, 28, 29]. In addition, Fig. 3d shows examples of quantitative representation of the readthrough efficiency of specific PTEN PTC. The generation of C-terminal truncated proteoforms caused by the PTC and its detection using antibodies against the PTP under study is illustrated in Fig. 4 with laforin (see Note 17). Additional technical aspects when monitoring PTC readthrough by ectopic expression of the PTP of interest are illustrated in Fig. 5, using PTEN as an example. The importance of using the appropriate expression plasmids, specially when antibiotic compounds are used as readthrough inducers, is shown in Fig. 5a. Note that the expression of antibiotic resistance genes can impair the effect of the antibiotic readthrough inducer (see Note 4). Figure 5b illustrates different transfection procedures using the pRK5 high-efficiency mammalian expression plasmid. Figure 5c and d show examples of the influence of the PTC type and the nucleotide at position +4 (the first nucleotide of the PTC being position +1) on the readthrough efficiency, using G418 as an inducer. Figure 5e recompiles the G418-induced readthrough efficiency of the PTEN PTCome, comparing the three possible PTC and the influence of a cytosine (C) nucleotide at position +4. When compared globally, the readthrough efficiency of the three possible PTC follows the order TGA > TAG > TAA, and a C at position +4 usually improves the readthrough efficiency. However, significant variations between individual PTC occur, making it necessary to test the readthrough efficiency for each PTC and readthrough inducer individually [16].

4

Notes 1. Human Gene Mutation Database (HGMD; not an open database; [30]) and ClinVar (open database; [31]) are two general databases annotating germline mutations found in association with human disease. COSMIC (open database; [32]) annotates somatic mutations found in human tumors and cancer-derived cell lines. Data can be imported from databases, but it is convenient to revise the annotations manually. HGMD provides datasets containing both missense and nonsense mutations. ClinVar provides datasets containing both nonsense and frameshift mutations, and the same mutations are named with distinct numbering corresponding to the protein isoforms annotated for the gene. In addition, there are disease-specific databases that can be of utility. A literature search is recommended, specially when the number of annotated mutations is low (see Fig. 2).

Translational Readthrough of Protein Tyrosine Phosphatases

15

2. PTCMAKER lists the PTC, which can be generated by singlenucleotide substitutions from a protein-coding nucleotide sequence, which we call the potential PTCome [16]. PTCMAKER also provides standardized mutagenic primers to generate PTC mutations by polymerase chain reaction (PCR) site-directed mutagenesis [33]. In general, the potential PTCome shows a relatively homogeneous distribution (straight line in kernel density plot representation) of potential PTC along the sequence of interest. In some cases (for instance, laforin in this report, see Fig. 2a), the distribution is not homogeneous, which should be taken into account for the interpretation of genotype/phenotype relations. 3. Simian kidney COS-7 cells are suitable for high transfection efficiency and high overexpression levels. Still, other mammalian or human cell lines can be used if transfection and overexpression conditions are optimized. 4. A high-efficiency mammalian expression vector is recommended. We routinely use the pRK5 mammalian expression vector, which contains the SV40 origin of replication and works efficiently in COS-7 cells expressing the SV40 large T-antigen. In the case of using antibiotics as readthrough inducers, note that the expression vector cannot contain the corresponding antibiotic resistance gene. Accordingly, when using the mammalian selectable antibiotic G418/geneticin as the inducer, the commonly used neomycin resistance gene (whose protein product is active against geneticin) cannot be carried in the expression vector (see Fig. 5a). 5. Different protocols for transfecting adherent mammalian cells exist, but they must be optimized for specific cell lines. In our experience, GenJet™ (SignaGen® Laboratories) works efficiently as a transfection reagent with several cell lines, but any other commercial lipid transfection reagent can also be used. In addition, COS-7 cells are transfected with good efficiency using the DEAE-dextran/DMSO/chloroquine method [34] (see Fig. 5b). 6. A wide variety of aminoglycoside and non-aminoglycoside readthrough inducers exist. G418/geneticin is widely documented as a high-efficiency aminoglycoside readthrough inducer in a 100–200 μg/mL (about 150–300 μM) concentration range, and it is recommended as an inducer in the first tests. G418 toxicity excludes its use in long-term experiments or its potential therapeutic use. 7. We use the Odyssey® CLx Imaging System (LI-COR Biosciences) fluorescence detection system, which permits reliable quantification of the protein bands from the immunoblot experiments. We use the Image Studio™ (LI-COR Biosciences) software for band quantification.

16

Leire Torices et al.

8. Dilution and washing conditions need to be individually tested for each antibody. High-affinity antibodies are recommended to detect the recombinant protein of interest upon translational readthrough efficiently. Consider that the location of the epitope recognized by the antibody determines the possibility of detecting truncated stable proteins generated by the PTC (see Figs. 3 and 4). The use of artificial epitope tags (such as Hemagglutinin or Flag epitope tags), which can be located N-terminal or C-terminal in the recombinant protein sequence, is helpful to visualize truncated and non-truncated protein forms under readthrough induction. Note that when a C-terminal tag is incorporated into the recombinant protein under study, the reconstitution of a full-length protein upon readthrough induction is monitored both by the size of the protein and by the recognition by the anti-tag antibody (see Fig. 3). The addition of artificial tags can produce variations in the wild-type basal protein expression, which should be monitored before performing readthrough induction experiments. As endogenous reference antibodies, we routinely use antiGAPDH or anti-β-actin antibodies. 9. During translational readthrough, near-cognate tRNAs incorporate variable amino acids in the position of the PTC [14], potentially obtaining protein variants with altered functions. This is an important issue in the case of PTP, and the functional testing of the proteins obtained by induced readthrough is recommended [16]. 10. A minimal number of different PTCs is needed to obtain a representative kernel density plot. A low number of different PTCs will probably render a kernel density plot with unreliable information regarding the potential pathogenicity of PTC along the protein sequence. 11. COSMIC database directly provides quantitative bar plots of mutations. 12. Optimal cell confluency for transfection depends on each cell line, but in our experience, it is not convenient to transfect cells with more than 50% confluency. For example, COS-7 cells are optimally transfected at about 20–25% confluency. 13. Final loading volumes of about 40–80 μL are convenient, depending on the size of your SDS-PAGE wells. 14. Do not cut away the bottom part of the membrane to detect truncated forms of the protein under study caused by the PTC. 15. The efficiency of the readthrough induction depends on several factors (see Subheading 3). A 0–40% efficiency range can be achieved using G418/geneticin as the readthrough inducer

Translational Readthrough of Protein Tyrosine Phosphatases

17

[16]. Although the readthrough of each PTC needs to be evaluated individually, the readthrough efficiency of the three types of PTC, when considered globally, follows the order TGA > TAG > TAA (see Fig. 5e). 16. C-terminal tagging allows the monitoring of translation of the full-length protein and N-terminal truncations created using alternative initiation methionines but does not allow for the monitoring of C-terminal truncations generated by the PTC. N-terminal tagging enables the monitoring of translation of the full-length protein and C-terminal truncations caused by the PTC but does not allow the tracking of N-terminal truncations. 17. To detect C-terminal truncations using antibodies against the PTP of interest, the epitope recognized by the antibody needs to be preserved in the truncated proteoform.

Acknowledgments This study was partially supported by grant SAF2016-79847-R (to RP). LT has received a predoctoral fellowship from Asociacio´n ˜ ola Contra el Ca´ncer (AECC, Junta Provincial de Bizkaia, Espan Spain). JM has received a predoctoral fellowship (PRE_2014_1_285) from Gobierno Vasco, Departamento de Educacio´n (Basque Country, Spain). AE is funded by The Spanish Ministry of Science and Innovation, grant RYC2021-032390-I, and Ikerbasque, the Basque Foundation for Science, Spain. CEN-X is the recipient of a Miguel Servet research contract (CP20/00008) from Instituto de Salud Carlos III (ISCIII, Spain, co-funded by European Union). JMC and RP are funded by Ikerbasque, the Basque Foundation for Science, Spain. The plasmid pCMV HA-laforin was kindly provided by Dr. Pascual Sanz (Institute of Biomedicine of Valencia (CSIC), Valencia, Spain). References 1. Gold B (2017) Somatic mutations in cancer: stochastic versus predictable. Mutat Res Genet Toxicol Environ Mutagen 814:37–46. https:// doi.org/10.1016/j.mrgentox.2016.12.006 2. Mort M, Ivanov D, Cooper DN, Chuzhanova NA (2008) A meta-analysis of nonsense mutations causing human genetic disease. Hum Mutat 29(8):1037–1047. https://doi.org/ 10.1002/humu.20763 3. Zhang L, Shay JW (2017) Multiple roles of APC and its therapeutic implications in colorectal cancer. J Natl Cancer Inst 109(8). https://doi.org/10.1093/jnci/djw332

4. Aartsma-Rus A, Ginjaar IB, Bushby K (2016) The importance of genetic diagnosis for Duchenne muscular dystrophy. J Med Genet 53(3): 1 4 5 – 1 5 1 . h t t p s : // d o i . o r g / 1 0 . 1 1 3 6 / jmedgenet-2015-103387 5. Hug N, Longman D, Caceres JF (2016) Mechanism and regulation of the nonsense-mediated decay pathway. Nucleic Acids Res 44(4): 1483–1495. https://doi.org/10.1093/nar/ gkw010 6. Nesta AV, Tafur D, Beck CR (2021) Hotspots of human mutation. Trends Genet: TIG 37(8):

18

Leire Torices et al.

717–729. https://doi.org/10.1016/j.tig. 2020.10.003 7. Zhang M, Yang D, Gold B (2021) Origins of nonsense mutations in human tumor suppressor genes. Mutat Res 823:111761. https:// doi.org/10.1016/j.mrfmmm.2021.111761 8. Dabrowski M, Bukowy-Bieryllo Z, Zietkiewicz E (2015) Translational readthrough potential of natural termination codons in eucaryotes-the impact of RNA sequence. RNA Biol 12(9): 9 5 0 – 9 5 8 . h t t p s : // d o i . o r g / 1 0 . 1 0 8 0 / 15476286.2015.1068497 9. Keeling KM, Xue X, Gunn G, Bedwell DM (2014) Therapeutics based on stop codon readthrough. Annu Rev Genomics Hum Genet 15:371–394. https://doi.org/10. 1146/annurev-genom-091212-153527 10. Morais P, Adachi H, Yu YT (2020) Suppression of nonsense mutations by new emerging technologies. Int J Mol Sci 21(12). https://doi. org/10.3390/ijms21124394 11. Nagel-Wolfrum K, Moller F, Penner I, Baasov T, Wolfrum U (2016) Targeting nonsense mutations in diseases with translational read-through-inducing drugs (TRIDs). BioDrugs 30(2):49–74. https://doi.org/10. 1007/s40259-016-0157-6 12. Palma M, Lejeune F (2021) Deciphering the molecular mechanism of stop codon readthrough. Biol Rev Camb Philos Soc 96(1): 310–329. https://doi.org/10.1111/brv. 12657 13. Dabrowski M, Bukowy-Bieryllo Z, Jackson CL, Zietkiewicz E (2021) Properties of non-aminoglycoside compounds used to stimulate translational readthrough of PTC mutations in primary ciliary dyskinesia. Int J Mol Sci 2 2 ( 9 ) . h t t p s : // d o i . o r g / 1 0 . 3 3 9 0 / ijms22094923 14. Roy B, Friesen WJ, Tomizawa Y, Leszyk JD, Zhuo J, Johnson B, Dakka J, Trotta CR, Xue X, Mutyam V, Keeling KM, Mobley JA, Rowe SM, Bedwell DM, Welch EM, Jacobson A (2016) Ataluren stimulates ribosomal selection of near-cognate tRNAs to promote nonsense suppression. Proc Natl Acad Sci U S A 113(44):12508–12513. https://doi.org/10. 1073/pnas.1605336113 15. Dabrowski M, Bukowy-Bieryllo Z, Zietkiewicz E (2018) Advances in therapeutic use of a drug-stimulated translational readthrough of premature termination codons. Mol Med 24(1):25. https://doi.org/10.1186/s10020018-0024-7 16. Luna S, Torices L, Mingo J, Amo L, Rodriguez-Escudero I, Ruiz-Ibarlucea P, Erramuzpe A, Cortes JM, Tejada MI,

Molina M, Nunes-Xavier CE, Lopez JI, Cid VJ, Pulido R (2021) A global analysis of the reconstitution of PTEN function by translational readthrough of PTEN pathogenic premature termination codons. Hum Mutat 42(5):551–566. https://doi.org/10.1002/ humu.24186 17. Bordeira-Carrico R, Pego AP, Santos M, Oliveira C (2012) Cancer syndromes and therapy by stop-codon readthrough. Trends Mol Med 18(11):667–678. https://doi.org/10.1016/j. molmed.2012.09.004 18. Martins-Dias P, Romao L (2021) Nonsense suppression therapies in human genetic diseases. Cell Mol Life Sci 78(10):4677–4701. https://doi.org/10.1007/s00018-02103809-7 19. Hendriks WJ, Elson A, Harroch S, Pulido R, Stoker A, den Hertog J (2013) Protein tyrosine phosphatases in health and disease. FEBS J 280(2):708–730. https://doi.org/10.1111/ febs.12000 20. Hendriks WJ, Pulido R (2013) Protein tyrosine phosphatase variants in human hereditary disorders and disease susceptibilities. Biochim Biophys Acta 1832(10):1673–1696. https:// doi.org/10.1016/j.bbadis.2013.05.022 21. Pulido R, Stoker AW, Hendriks WJ (2013) PTPs emerge as PIPs: protein tyrosine phosphatases with lipid-phosphatase activities in human disease. Hum Mol Genet 22(R1): R66–R76. https://doi.org/10.1093/hmg/ ddt347 22. Nitschke F, Ahonen SJ, Nitschke S, Mitra S, Minassian BA (2018) Lafora disease - from pathogenesis to treatment strategies. Nat Rev Neurol 14(10):606–617. https://doi.org/10. 1038/s41582-018-0057-0 23. Pulido R (2015) PTEN: a yin-yang master regulator protein in health and disease. Methods 77–78:3–10. https://doi.org/10.1016/j. ymeth.2015.02.009 24. Auld DS, Inglese J (2004) Interferences with luciferase reporter enzymes. In: Markossian S, Grossman A, Brimacombe K et al (eds) Assay guidance manual, Bethesda 25. Misawa K, Kikuno RF (2009) Evaluation of the effect of CpG hypermutability on human codon substitution. Gene 431(1–2):18–22. https://doi.org/10.1016/j.gene.2008. 11.006 26. Fryxell KJ, Moon WJ (2005) CpG mutation rates in the human genome are highly dependent on local GC content. Mol Biol Evol 22(3):650–658. https://doi.org/10.1093/ molbev/msi043

Translational Readthrough of Protein Tyrosine Phosphatases 27. Zhao Z, Jiang C (2007) Methylationdependent transition rates are dependent on local sequence lengths and genomic regions. Mol Biol Evol 24(1):23–25. https://doi.org/ 10.1093/molbev/msl156 28. Cohen S, Kramarski L, Levi S, Deshe N, Ben David O, Arbely E (2019) Nonsense mutationdependent reinitiation of translation in mammalian cells. Nucleic Acids Res 47(12): 6330–6338. https://doi.org/10.1093/nar/ gkz319 29. Torices L, de Las Heras J, Arango-Lasprilla JC, Cortes JM, Nunes-Xavier CE, Pulido R (2021) MMADHC premature termination codons in the pathogenesis of cobalamin D disorder: potential of translational readthrough reconstitution. Mol Genet Metab Rep 26:100710. https://doi.org/10.1016/j.ymgmr.2021. 100710 30. Stenson PD, Mort M, Ball EV, Evans K, Hayden M, Heywood S, Hussain M, Phillips AD, Cooper DN (2017) The Human Gene Mutation Database: towards a comprehensive repository of inherited mutation data for medical research, genetic diagnosis and nextgeneration sequencing studies. Hum Genet 136(6):665–677. https://doi.org/10.1007/ s00439-017-1779-6 31. Landrum MJ, Kattman BL (2018) ClinVar at five years: delivering on the promise. Hum Mutat 39(11):1623–1630. https://doi.org/ 10.1002/humu.23641 32. Forbes SA, Beare D, Boutselakis H, Bamford S, Bindal N, Tate J, Cole CG, Ward S, Dawson E, Ponting L, Stefancsik R, Harsha B, Kok CY, Jia M, Jubb H, Sondka Z, Thompson S, De T, Campbell PJ (2017) COSMIC: somatic cancer genetics at high-resolution. Nucleic Acids Res 45(D1):D777–D783. https://doi.org/10. 1093/nar/gkw1121 33. Mingo J, Erramuzpe A, Luna S, Aurtenetxe O, Amo L, Diez I, Schepens JT, Hendriks WJ, Cortes JM, Pulido R (2016) One-tube-only standardized site-directed mutagenesis: an alternative approach to generate amino acid

19

substitution collections. PLoS One 11(8): e0160972. https://doi.org/10.1371/journal. pone.0160972 34. Ausubel FM (2002) Short protocols in molecular biology: a compendium of methods from current protocols in molecular biology, 5th edn. Wiley, New York 35. Alonso A, Pulido R (2016) The extended human PTPome: a growing tyrosine phosphatase family. FEBS J 283(8):1404–1429. https://doi.org/10.1111/febs.13600 36. Gentry MS, Roma-Mateo C, Sanz P (2013) Laforin, a protein with many faces: glucan phosphatase, adapter protein, et alii. FEBS J 280(2):525–537. https://doi.org/10.1111/j. 1742-4658.2012.08549.x 37. Torres J, Navarro S, Rogla I, Ripoll F, Lluch A, Garcia-Conde J, Llombart-Bosch A, Cervera J, Pulido R (2001) Heterogeneous lack of expression of the tumour suppressor PTEN protein in human neoplastic tissues. Eur J Cancer 37(1): 114–121 38. Luna S, Mingo J, Aurtenetxe O, Blanco L, Amo L, Schepens J, Hendriks WJ, Pulido R (2016) Tailor-made protein tyrosine phosphatases: in vitro site-directed mutagenesis of PTEN and PTPRZ-B. Methods Mol Biol 1447:79–93. https://doi.org/10.1007/9781-4939-3746-2_5 39. Mingo J, Luna S, Gaafar A, Nunes-Xavier CE, Torices L, Mosteiro L, Ruiz R, Guerra I, Llarena R, Angulo JC, Lopez JI, Pulido R (2019) Precise definition of PTEN C-terminal epitopes and its implications in clinical oncology. NPJ Precis Oncol 3:11. https://doi.org/ 10.1038/s41698-019-0083-4 40. Moreno D, Towler MC, Hardie DG, Knecht E, Sanz P (2010) The laforin-malin complex, involved in Lafora disease, promotes the incorporation of K63-linked ubiquitin chains into AMP-activated protein kinase beta subunits. Mol Biol Cell 21(15):2578–2588. https:// doi.org/10.1091/mbc.e10-03-0227

Chapter 2 Understanding Pseudophosphatase Function Through Biochemical Interactions Shanta´ D. Hinton Abstract Pseudophosphatases have been solidified as important signaling molecules that regulate signal transduction cascades. However, their mechanisms of action remain enigmatic. Reflecting this mystery, the prototypical pseudophosphatase STYX (phospho-serine-threonine/tyrosine-binding protein) was named with allusion to the river of the dead in Greek mythology to emphasize that these molecules are “dead” phosphatases. Although proteins with STYX domains do not catalyze dephosphorylation, this does not preclude their having other functions, including as integral elements of signaling networks. Thus, understanding their roles may mark them as potential novel drug targets. This chapter outlines common strategies used to characterize the functions of pseudophosphatases, using as an example MK-STYX [MAPK (mitogenactivated protein kinase) phospho-serine-threonine/tyrosine-binding], which has been linked to tumorigenesis, hepatocellular carcinoma, glioblastoma, apoptosis, and neuronal differentiation. We start with the importance of “restoring” (when possible) phosphatase activity in a pseudophosphatase, so the active mutant may be used as a comparison control throughout immunoprecipitation and mass spectrometry analyses. To this end, we provide protocols for site-directed mutagenesis, mammalian cell transfection, co-immunoprecipitation, phosphatase activity assays, and immunoblotting that we have used to investigate MK-STYX and the active mutant MK-STYXactive. We also highlight the importance of utilizing RNA interference (RNAi) “knockdown” technology to determine a cellular phenotype in various cell lines. Therefore, we outline our protocols for introducing short hairpin RNA (shRNA) expression plasmids into mammalian cells and quantifying knockdown of gene expression with real-time quantitative PCR (qPCR). We also provide a bioinformatic approach to investigating MK-STYX and MK-STYX(active mutant). These bioinformatic approaches can stand alone experimentally but also complement and enhance “wet” bench approaches such as binding assays and/or activity assays. A combination of cellular, molecular, biochemical, proteomic, and bioinformatic techniques has been a powerful tool in identifying novel functions of MK-STYX. Likewise, the information provided here should be a helpful guide to elucidating the functions of other pseudophosphatases. Key words Pseudophosphatases, STYX domains, MAPK phosphatases, MK-STYX, Immunoprecipitation, shRNA knockdown, Bioinformatics

Damien The´venin and Jo¨rg P. Mu¨ller (eds.), Protein Tyrosine Phosphatases: Methods and Protocols, Methods in Molecular Biology, vol. 2743, https://doi.org/10.1007/978-1-0716-3569-8_2, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

21

22

1

Shanta´ D. Hinton

Introduction Bioinformatics was used to initially classify pseudophosphatases. Proteins having any mutation within positions that are critical for a protein phosphatase to perform catalysis are considered pseudophosphatases [1, 2]. For example, mutations within the catalytic active site (HCX5R) or other sites, such as the WPD (tryptophan, proline, and aspartic acid) loop, are thought to render the protein tyrosine phosphatases (PTPs) catalytically inactive. Thus, pseudophosphatases are catalytically inactive because they lack critical residues within their protein tyrosine phosphatase (PTP) active site signature motif (HCX5R), which is essential for catalytic activity [3]. Many pseudophosphatases have been identified [4], yet their modes of action remain elusive. Importantly, however, many possess the characteristic three-dimensional PTP fold. Consequently, though catalytically inactive, they maintain their ability to bind phosphorylated proteins, which makes them ideal for investigating the dynamics of protein–protein interactions and their effects on pathways such as Ras and apoptotic signaling. What are the functions of pseudophosphatases, and how can these proteins be investigated? Identifying PTP substrates was essential for understanding the roles of these active phosphatases in signal transduction [5]. A key to identifying PTP substrates was the development of “substratetrapping,” a powerful technique in which mutations of invariant residues of the active site render an active enzyme a “substratetrap.” The mutated PTP binds a phosphorylated residue but does not hydrolyze it [5–7]. The discovery of naturally occurring proteins structurally related to PTPs that resemble “substratetrapping” mutants led to many questions. Why do they exist? Are they simply dominant-negative proteins, functioning like substrate traps, similar to artificial mutants? Just as identifying substrates of PTPs led to many discoveries of their central role in signaling pathways, identifying interacting partners of pseudophosphatases, such as the myotubularins, Caenorhabditis elegans EGGs, prototypical STYX, and MK-STYX, has led to defining the molecular details of their functions. The myotubularins (MTM) include several pseudophosphatases [8]. Six of the 14 human MTM genes encode pseudophosphatases [9] that function as scaffolds to form complexes with their active homologs to regulate catalytic function and subcellular localization of the active phosphatase. For example, the pseudophosphatase MTMR13 (MTM-related protein 13) binds the active enzyme MTMR2 [9]. A mutation in the MTMR13 pseudophosphatase or the active MTMR2 is associated with Charcot-Marie-Tooth disease, a neuropathy characterized by abnormal nerve myelination [8]. The C. elegans pseudophosphatases EGG4 and EGG5 trap MBK-2 (mini brain kinase 2) to

Pseudophosphatase Function Through Biochemical Interactions

23

regulate the oocyte-to-zygote transition through DYRK (dualspecificity tyrosine phosphorylated and regulated kinase) [10, 11]. The prototypical STYX competes for ERK binding and modulates cell fate decisions and cell migration as a spatiotemporal regulator of ERK1/2 signaling [12]. MK-STYX is structurally related to DUSPs (dual-specificity phosphatases). In particular, the bioinformatic analysis revealed homology to MKP (MAP kinase phosphatase), which dephosphorylates both threonine and tyrosine residues in the activation loop of members of the MAPK family [5]. It does not possess the active site signature motif HC(X5)R that is essential for phosphatase activity; instead, it has the sequence IFSTQGISRS, which renders it catalytically inactive [5, 13, 14]. For many years, little was known about MK-STYX, except that it is highly expressed in Ewing’s sarcoma family tumors [14, 15]. More recently, significant advances have been made in understanding its functional role. It interacts with G3BP-1 and decreases stress granule assembly [6, 14], thereby affecting the stress response pathway. In addition, it regulates mitochondria-dependent apoptosis [7]. It interacts with PTPM1 (PTP localized to the mitochondrion 1) and inhibits its catalytic activity, regulating cell viability [16]. MK-STYX also has implications in neuronal development [17], in which it induces neurite outgrowth and decreases RhoA activation. Furthermore, it affects a downstream player of RhoA, the actin-binding protein cofilin. Overexpression of MK-STYX decreases the phosphorylation of cofilin in non-NGF-stimulated cells but increases its phosphorylation in NGF-stimulated cells, whereas knocking it down has the opposite effect [17]. In addition, MK-STYX interacts with several other cytoskeleton proteins, such as myosin, vimentin, and spectrin [18]. Evolutionary studies show that pseudophosphatases have very distinctive roles from each other throughout history. Prototypical STYX is very conserved and resists change, which is required for its role in the well-established MAPK signaling [19]. However, MK-STYX changes throughout evolution. Perhaps this plasticity is necessary for a signaling molecule to have roles in numerous pathways [19]. An essential facet of understanding the function of pseudophosphatases is identifying their interacting (binding) partners. Interactions between a pseudophosphatase and its binding partner may reveal the signaling cascade and/or cellular processes in which the pseudophosphatase plays a significant role. Bioinformatics, used initially to characterize pseudoenzymes [1, 2], is also helpful in predicting their capacity for interacting with other molecules. Because MK-STYX is a member of the MKP family, the original investigations pursued whether MK-STYX interacts with MKP substrates and MAPKs, such as the extracellular regulated kinases (ERK1 and 2). To date, we have not found that MK-STYX interacts with ERK1/2, consistent with the reports of others [7, 16]. The

24

Shanta´ D. Hinton

Fig. 1 The kinase interaction motif (KIM) of MK-STYX and its catalytically active homologs, MAP kinase phosphatase (MKP)-1 and MKP-3. The KIM regulates substrate specificity and substrate docking in the CH2 domain of classical MKPs. The KIM requires two to three consecutive arginine residues to create a consecutive positively charged region that interacts with the negatively charged aspartic acid residues on the target MAPK substrate for docking it. Unlike its active homologs, MKP-1 and MKP-3, the KIM consensus motif of MK-STYX contains only a single arginine. This may explain why MK-STYX does not bind and regulate MAPKs in the same manner as typical MKPs. The sequence logo was made using WebLogo 3.7.4 and has a 2.0-bit scale (positive residues are shown in blue, neutral residues are shown in gray, and no negative residues are present). (This figure was adapted and modified from Hepworth, E. M. W. H. and Hinton, S. D. 2021 manuscript [37])

bioinformatics analysis demonstrates that mutations of MK-STYX within its kinase interaction motif (KIM), which binds MAPKs, prevent MK-STYX from interacting with MAPKs (Fig. 1). These studies also show that the mutations within the active signature motif that render it catalytically inactive do not change its structure, implying that MK-STYX maintains its ability to bind phosphorylated residues (Fig. 1). These reports illustrate that the pseudophosphatases are key molecules in signaling pathways, highlighting the increasing importance of investigating the function of such pseudoenzymes. Some reports describe methods of analyzing the physiological functions of pseudophosphatases [20]. This chapter presents methodologies for designing catalytically active mutants, characterizing interacting partners, and assessing the function of pseudophosphatases through classical wet bench studies and bioinformatic analysis. There is a plethora of diseases linked to pseudophosphatase dysregulation, such as cancers (breast, glioblastoma, colorectal, and ovarian) and neuropathies (Charcot-Marie-Tooth diseases and epilepsy) [21–24]. This highlights the important roles of pseudophosphatases and why it is imperative to continue investigating them. While the molecular mechanisms of pseudophosphatases remain relatively uncharacterized, we expect strategies to tackle these catalytically inactive members of the PTP family will lead to many novel discoveries.

Pseudophosphatase Function Through Biochemical Interactions

2

25

Materials

2.1 Conversion of a Pseudophosphatase STYX Domain to an Active Signature Motif (HCX5R)

All solutions are prepared in deionized MilliQ-filtered H2O. Cell culture and transfection procedures require sterile conditions. 1. Recombinant cDNA of the tagged pseudophosphatase of interest cloned into a mammalian expression vector). 2. Design and order primers through Oligo, Primers, Probes, and Genes (Invitrogen) software (or equivalent software for primer design) available through the Thermo Fisher’s website, which will substitute the residues of the pseudophosphatase to residues that resemble a PTP signature active site (HCX5R). For example, the PTP signature site of MK-STYX consists of FSTQGIRS. Thus, we designed primers that substituted HC for the FS to generate the sequence HCTQGIRS (see Note 1). 3. Site-directed mutagenesis kit (see Note 2), and cloned tagged pseudophosphatase of interest cloned into a mammalian expression vector (see Note 3). 4. Appropriate reagents to perform PCR (see Note 4). 5. PTP activity assay kit.

2.2 ImmunoprecipitationImmunoprecipitations

1. Mammalian cell lines are suitable for transfection and transient overexpression of recombinant proteins (see Note 5). 2. Cell culture medium and any required supplements. 3. Lipofectamine 2000 and Opti-MEM Reduced Serum Medium or the lab’s preferred transfection reagent. 4. 10X Phosphate-buffered saline (PBS): 0.2 M KH2PO4 and 1.5 M NaCl, pH 7.2. Dilute as required to 1X (specifically to wash cells for Lipofectamine transfection); alternative for general washing is 1X PBS (1.85 mM NaH2PO4, 8.4 mM NaHPO4, and 150 mM NaCl, pH 7.2). 5. Lysis buffer: 50 mM HEPES, pH 7.2, 150 mM NaCl, 10% glycerol, 10 mM NaF, 1% Nonidet P-40 alternate (e.g., IGEPAL), and Roche protease inhibitor cocktail tablets; or other appropriate protease and phosphatase inhibitors (see Note 6). 6. Cell lifters or scrapers. 7. Bradford protein concentration assay reagent (or the lab’s preferred method to determine protein concentration). 8. Antibodies against the pseudophosphatase of interest (or an epitope tag in a mammalian expression vector) (see Note 7). 9. Immunoglobulin (Ig) G beads (GE) (or protein A beads; dependent on primary antibody) (see Note 8). 10. 5X Laemmli sample loading buffer: 10% sodium dodecyl sulfate, 30% glycerol, 250 mM Tris–HCl, pH 6.8, and 0.02% bromophenol blue.

26

Shanta´ D. Hinton

Fig. 2 Identification of binding partner(s) of a pseudophosphatase. Protein A or G beads are used to interact with antibodies (Y symbol) that are bound to specific proteins from cell lysates. These proteins are released from beads and analyzed through SDS PAGE followed by mass spectrometry (curved arrow) of a specifically excised protein (band with the red dotted rectangle). Or the released protein is directly analyzed through mass spectrometry, which provides a global analysis of interactors of the protein of interest, in this case, the pseudophosphatase

11. Dithiothreitol (DTT) or β-mercaptoethanol. 12. 1.5-mL centrifuge tubes). 13. Access to mass spectrometry facilities (see Fig. 2). 2.3 Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis (SDS-PAGE)

1. H2O. 2. 30% acrylamide-bisacrylamide mix (29:1). 3. 1.5 M Tris–HCl (pH 8.8). 4. 1.0 M Tris–HCl (pH 6.8). 5. 10% SDS. 6. 10% ammonium persulfate. 7. Tetramethylethylenediamine (TEMED). 8. Protein standard molecular weight markers (pre-stained or unstained). 9. Protein stain (Coomassie blue and destain, or silver stain).

2.4

Immunoblotting

1. 10% SDS-PAGE gel (see Note 9). 2. Polyvinyl difluoride (PVDF) membrane or nitrocellulose membrane (see Note 10). 3. Blotting paper. 4. Transfer buffer: 48 mM Tris base, 39 mM glycine, 1.3 mM SDS, and 20% methanol (see Note 10). 5. Blocking reagent (powdered milk, bovine serum albumin (BSA), or commercial blocking reagent) (see Note 11). 6. Antibodies against the pseudophosphatase of interest (or an epitope tag used in a mammalian expression vector) (see Note 3).

Pseudophosphatase Function Through Biochemical Interactions

27

7. Anti-rabbit or anti-mouse antibodies conjugated to horse radish peroxidase (HRP), dependent on the source of the primary antibody. 8. 10X Tween 20-Tris saline buffer (TTBS): 1.0 M Tris–HCl pH 7.5, 1.5 M NaCl, and 0.1% Tween 20 (v/v). Dilute as required to 1X. 9. Enhanced chemiluminescence (ECL) Prime, or another chemiluminescence detection reagent. 10. Mild stripping buffer: 200 mM glycine, 3.5 mM SDS, and 1% Tween 20 (see Note 12). 2.5

Knockdown

1. Short hairpin RNA (shRNA) expression plasmids specific for pseudophosphatase of interest (see Notes 13–16). 2. Lipofectamine 2000 transfection reagent and Opti-MEM (or other transfection reagents). 3. Aurum™ Total RNA Mini Kit (or the lab’s preferred RNA isolation kit) (see Note 13). 4. SuperArray RT2 First Stand Kit (Bioscience) or the lab’s preferred cDNA kit. 5. StepOne™ Real-Time PCR [Applied Biosystems] or a lab-preferred real-time quantitative polymerase chain reaction (qPCR) apparatus. 6. RT2SYBR Green/Fluorescein qPCR Master Mix or reagent designed for the lab’s specific qPCR apparatus. 7. Primers for the pseudophosphatase of interest.

3

Methods

3.1 Creation of an Active PTP from a Pseudophosphatase

1. Use the custom-designed primers that convert the residues of the pseudophosphatase to residues that resemble a PTP signature active site (HCX5R), which was developed in Subheading 2.1 of this chapter (see Note 1). 2. Follow the protocol of the site-directed mutagenesis kit and PCR to construct a plasmid that contains the signature active site (see Note 2). 3. Sequence plasmid to confirm that the newly generated residues are equivalent to the PTP-active site signature motive sequence (I/VHCXXGXXR[S/T]), in which the residues in bold play a critical role in catalysis. 4. Perform a PTP activity assay to determine whether the substituted residues of the pseudophosphatase confer PTP activity (see Note 17).

28

Shanta´ D. Hinton

3.2 Identifying Interacting Partners of Pseudophosphatases

If the mutated pseudophosphatase is catalytically active, it can serve as a powerful tool for functional studies of the wild-type pseudophosphatase. Therefore, when an active mutant is available, it should also be used for immunoprecipitation studies. The wildtype pseudophosphatase may bind different proteins than the active mutant (see Notes 1 and 18). 1. Transfect cells of interest with a mammalian expression plasmid containing the tagged pseudophosphatase (see Note 3). We transfected cells seeded on a 10-cm plate with 2 or 4 μg plasmid following the Lipofectamine 2000 protocol. 2. 24 or 48 h post-transfection, remove the medium (see Note 19), wash with 1X PBS, and remove PBS. 3. Add 1 mL (for 10-cm plates) lysis buffer with protease and phosphatase inhibitors (when warranted; see Note 6), and make sure all cells are covered by the buffer. Incubate cells (in plate) on ice for 3 min (see Note 20). When warranted, add appropriate phosphatase ihibitors to the buffer and washes (see Note 21). 4. Use a cell lifter to remove cells from the plate, and pipet ~1 mL lysate into a pre-chilled 1.5-mL microcentrifuge tube. Incubate on ice for 5–10 min. Time is dependent on the viscosity of samples; samples should not become too viscous. 5. Centrifuge for 15 min at 14,000 × g at 4 °C. Samples may be stored at -20 °C or protocol continued for immunoprecipitation. 6. Determine the protein concentration with Bradford reagent or another quantitative protein assay. The minimum amount of protein required for immunoprecipitation is 500 μg/mL. We prefer to use 1 mg/mL or up to 2 mg/mL (see Note 22). Dilute samples with lysis buffer to a total volume of 1 mL. 7. Pre-clear lysate (see Note 23) by adding 20 μL protein A or G beads (dependent on antibody; see Note 9). Incubate on ice for 30 min, and centrifuge for 5 min at 14,000 × g, at 4 °C. Transfer lysate (not beads) to a new 1.5-mL centrifuge tube (see Note 24). 8. Save 40–50 μL lysate to analyze by immunoblotting to ensure that the immunoprecipitations were performed with equal amounts of protein. 9. Add 5 μL antibody probing for the pseudophosphatase of interest, and incubate on ice for 1 h. Make sure the antibody is mixed well with the lysate by inverting the tube. 10. Add 25 μL of well-suspended protein A or G beads to the lysate (sample) and antibody mixture. Gently pipette twice to release excess beads into solution, invert the tube, and place it on ice.

Pseudophosphatase Function Through Biochemical Interactions

29

Place tubes containing the lysates (samples), antibodies, and beads on a rocker in a cold room (at 4 °C) for 1 h. Be careful to ensure tubes are tightly sealed. We prefer an adjustable test tube rocker. 11. Centrifuge for 15 min at 14,000 × g at 4 °C; protein A or G agarose beads should be visible as a pellet. 12. Aspirate supernatant with a Pasteur pipette that has a gel loading (flat) tip connected to it. The extra tip minimizes the force of the aspiration and prevents the aspiration of the beads (the immunoprecipitated sample). Be careful not to disturb beads; we suggest not aspirating all of the supernatant until the final wash (see Note 25). 13. Wash three to four times with 1 mL ice-cold lysis buffer by centrifuging at 10,000 × g at 4 °C for 30–60 s. If complexes are stable, the washes may be done at room temperature. 14. After the final wash, aspirate as much as possible of the supernatant. Be consistent with the volume of beads (samples) left in each tube because this is the sample that will be analyzed for complexes. 15. Elute the complexes from the beads with 50 μL sample loading buffer and 100 mM DTT. Heat to 85 °C (or boil) for 5 min, centrifuge briefly (~1 min), and load the sample onto SDS-PAGE. Samples may also be stored at -20 °C for electrophoresis at a later time. Samples may be aliquoted into two separate tubes for staining and immunoblotting purposes. 16. Resolve samples on a 10% SDS-PAGE or appropriate percentage gel for the size of pseudophosphatase of interest. 17. Stain gel with Coomassie Blue or silver stain to visualize proteins (see Note 26). 18. To identify the proteins of interest, cut out the bands (proteins), and send them to a mass spectrometry facility for identification of the peptides (see Note 27). When a possible candidate is identified as an interacting partner of the pseudophosphatase, an immunoblot may be performed on the immunoprecipitated complex. 19. Resolve samples (20 μL from step 15) with 10% SDS-PAGE (or desired percentage gel), making sure samples are heated and centrifuged before loading. 20. Transfer samples from SDS-PAGE to a PVDF membrane using desired transfer methods (see Note 11). 21. Detect the candidate determined to be the possible interacting partner by standard immunoblotting and chemiluminescence detection techniques, using the antibody against the candidate and the appropriate secondary antibody.

30

Shanta´ D. Hinton

22. Detect the pseudophosphatase by standard immunoblotting and chemiluminescent detection techniques, using the antibody against the pseudophosphatase or the tag and appropriate secondary antibody. If the pseudophosphatase and possible interacting candidate are of sufficiently different molecular masses, cut the PVDF membrane so that each protein is separate and may be probed separately with appropriate antibodies. However, if the proteins are close in size, the membrane must be stripped (see Note 12). 23. Strip the membrane by warming the stripping buffer to 50 °C, and add enough buffer to cover the membrane placed in a small plastic container with a lid. Or use the lab’s established stripping protocol. 24. Add the membrane, and incubate at 50 °C while agitating for 45 min. 25. Dispose of the buffer in a β-mercaptoethanol waste container or as required by the institution. 26. Rinse the membrane under running tap water for 1–2 h. 27. Wash two times for 5 min with 1X TBST before performing the immunoblotting procedure, blocking first. 3.3 Investigating the Biological Significance of Pseudophosphatases by Knockdown

1. Seed cells of interest at 6 × 105 cells per 10-cm plate. 2. 24 h post-transfection with the pseudophosphatase expression plasmid, transfect cells with 10 μg of the pseudophosphatase shRNA expression plasmid (see Notes 14–16). 3. Incubate plates for 4–6 h, and replace medium with fresh medium. The success of transfection may be visualized with a fluorescence microscope when using green fluorescent protein (GFP)-tagged shRNA plasmids. 4. 24 h post-transfection, purify RNA with the Aurum™ Total RNA Mini Kit (or the lab’s established RNA purification compatible with the qPCR apparatus; see Note 13). 5. Determine the concentration of RNA; only use RNA with the following ratios: A260/280 > 2.0 and A260/230 > 1.7 for qPCR. 6. Synthesize cDNA from 0.735 μg RNA with the SuperArray Bioscience’s RT2 First Strand Kit or the lab’s established cDNA protocol. 7. Use a 48-well plate to set up reactions; one reaction (total volume 25 μL) should contain 12.5 μL RT2 SYBR Green/ Fluorescein qPCR Master Mix specifically designed for the lab’s preferred qPCR apparatus, 11.5 μL RNase-free water, 1.5 μL cDNA template, and 1 μL appropriate primer. A reaction should also be set up with a housekeeping gene (dependent on cell type) for each experimental sample to normalize

Pseudophosphatase Function Through Biochemical Interactions

31

the raw data. “No reverse transcriptase” controls should be included for each gene of interest and housekeeping gene to detect any genomic contamination. 8. Centrifuge the 48-well plate for 90 s at 400 × g. 9. Load the Applied Biosystems’ StepOne™ Real-Time PCR apparatus or the lab’s preferred apparatus. 10. Run the qPCR at the following parameters: 10 min at 95 °C (for enzyme activation) and 40 cycles of 15 s at 95 °C, 35 s at 55 °C, and 30 s at 72 °C. The SYBR Green fluorescence should be detected and recorded during the annealing step of each cycle. 11. Perform a melting curve as a quality control measure, and analyze the qPCR data. When using the Applied Biosystems’ StepOne™ Real-Time PCR apparatus, the ΔΔCt (Livak) method is used with ABI StepOne software. 3.4 Validating Knockdown of the Pseudophosphatase

1. Validation of knockdown of protein expression may be performed by immunoblotting (see Note 16). 2. Transfect cells of interest with 10 μg of various confirmed pseudophosphatase shRNA plasmid according to normal transfection procedure. 3. Incubate plates for 4–6 h, and replace transfection medium with fresh, complete medium. 4. 24 h post-transfection, observe cells with phase and fluorescence microscopy for any noticeable morphological changes (such changes may be noticed sooner than 24 h, so it is advisable to monitor cells over time). These morphological changes may suggest possible biological functions and proteins involved in these functions that the pseudophosphatase may regulate (see Note 28). 5. 24 h post-transfection, stimulate cells with the appropriate stimulus or inhibitor for the desired biological functional assay (see Note 29). 6. Perform immunoblot with the antibodies against the desired protein of interest (see Subheading 3.2). Analyze whether knocking down the pseudophosphatase changes the expression pattern of the chosen protein of interest (see Note 30).

3.5 Analyzing Potential Interactions of Pseudophosphatases Through Bioinformatic Structural Analysis

1. Compare the consensus sequence of the pseudophosphatase with the active homologs (see Note 31 and Fig. 1). 2. Create a sequence logo with Weblogo 3.7.4 to determine the conservation of amino acids between the pseudophosphatase and active homologs (see Note 32 and Fig. 1).

32

Shanta´ D. Hinton

Fig. 3 Bioinformatic structural analysis of pseudophosphatase MK-STYX. Computational mutagenesis of the KIM of MK-STYX affects the size and shape of a predicted binding pocket. To probe for possible explanations behind why MK-STYX lacks the ability to regulate MAPK signaling, a predictive model of the macromolecular structure of MK-STYX was generated by Iterative Threading ASSEmbly Refinement (I-TASSER). The best I-TASSER model (validated through MolProbity with a score of 3.03) was refined using DeepRefiner. The best quality refined model (predicted global quality score of 0.142 and MolProbity score of 2.14) was mutated using the mutagenesis function in the PyMOL Molecular Graphics System 4.6.0 to create the V53R mutant, which restores the consecutive arginine residues in the KIM. The structures of both wild-type (WT) MK-STYX and the V53R mutant were submitted to Pocket Cavity Search Application (POCASA) to probe the surface of the protein and predict possible binding sites (1.0 Å grid size and 2 Å probe radius). When the predicted pockets were compared between the WT MK-STYX and the V53R mutant, the V53R mutant displayed a smaller and differently shaped predicted binding pocket in the area of the KIM (rank 3 out of 13 predicted pockets, decrease in volume from 98 to 96, and decrease in volume depth value from 271 to 261). The area of change (shown by the black arrow) was in the immediate area of the mutated residue (WT residue shown in yellow and V53R mutation in red). The WT model also was submitted to Missense3D, which also predicted that the V53R mutation altered a cavity and led to the contraction of cavity volume (predicted volume contraction of 89.424 Å^3). For validation, the WT model was submitted to predict the effect of an S246C mutation in the signature active site motif; there was no predicted structural damage (data not shown). This was expected because most pseudophosphatases maintain their three-dimensional fold. This may indicate that, while MK-STYX does not bind MAPKs at this site, this pocket may allow MK-STYX to bind a novel set of binding partners. (This figure was adapted and modified from Hepworth, E. M. W. H. and Hinton, S. D. 2021 manuscript [37])

3. To generate a predicted structure of the protein of interest, if one is not available, submit the amino acid sequence to Iterative Threading ASSEmbly Refinement (I-TASSER) [25] (see Note 33). The mutated sequence for the “restored” consensus should also be submitted. 4. To validate the predicted structures and determine the most sterically favorable model from I-TASSER, use the software MolProbity, which determines the steric hindrance of a model [26]. 5. Submit the most sterically favorable model for the pseudophosphatase or “restored” mutant to DeepRefiner (see Note 34) so that a refined model is obtained.

Pseudophosphatase Function Through Biochemical Interactions

33

6. To “restore” the consensus sequence of the pseudophosphatase (wild-type) to that of its active phosphatase homolog, mutate the residue of interest to that of the active homolog in PyMOL. 7. Submit both the wild-type model of pseudophosphatase and the newly generated mutant model Protein Data Bank (PDB) files to POcket-CAvity Search Application (POCASA) to identify and map potential binding pockets of the surface of the protein to determine whether the mutagenesis changed the binding affinity (number) or shape of the pocket [27] (see Note 35 and Fig. 3). 8. Submit only the wild-type pseudophosphatase model to Missense3D [28] and use the platform to predict any changes in the structure based on single amino acid mutations (see Note 36).

4

Notes 1. Not all pseudophosphatases may be able to have their activity “restored.” However, it is worthwhile to test whether, as in the case of the prototypical STYX [3] or MK-STYX [14], mutations could be introduced to generate a catalytically competent form. A comparison of the effects of the catalytically active and inactive forms of a pseudophosphatase may reveal different functions on signaling cascades, which may provide insight into a specific role of a pseudophosphatase. 2. Substituting amino acid residues is no longer labor-intensive. Mutations to change nucleotide codons for amino acid residues may be accomplished by purchasing site-directed mutagenesis kits from companies such as Agilent Technologies (QuikChange II) or New England Bio Labs (site-directed mutagenesis Kit). Furthermore, there are companies that will synthesize a DNA construct with the desired sequence (e.g., GeneArt, Life Technologies). Considering that there may be other important domains responsible for the function of PTPs, such as the CH2 (cdc25 homology) in the MKP family, we suggest that the sequence of such domains be closely examined for any possible substitution as well for possible future investigations. 3. MK-STYX was flanked by two FLAG epitopes and then cloned into the vector of interest. We suggest using a tag such as a FLAG expression system. The FLAG expression system will allow numerous applications with the pseudophosphatase, such as protein purification, immunoblotting, immunoprecipitation, and so on. FLAG is a small hydrophilic peptide that is

34

Shanta´ D. Hinton

unlikely to interfere with protein folding or alter the function of the pseudophosphatase. In addition, the FLAG epitope may be detected with a commercially available anti-FLAG antibody. 4. PCR is a common technique widely used in cellular and molecular and biochemical laboratories. Therefore, the details of PCR are not discussed in this chapter. 5. MK-STYX has been successfully overexpressed in mammalian cell lines such as HEK-293, HeLa, Cos-1, and PC-12. The appropriate media, such as Dulbecco’s Modified Eagle Medium (DMEM), Minimum Essential Medium (MEM), or Roswell Memorial Institute (RPMI) medium, are required to culture each cell line successfully. All these media can be purchased from Life Technologies. 6. No one buffer is universal to sufficiently lyse all cells. Therefore, it is important to choose the appropriate buffer that allows the release of an antigen (pseudophosphatase) recognizable by the antibody. Various buffers such as NP-40 Lysis: 150 mM NaCl, 1% NP-40, and 50 mM Tris–HCl, pH 8.0 (non-ionic detergent) or RIPA: 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS (combination of non-ionic and ionic detergents, thus more denaturing), and 50 mM Tris–HCl, pH 8.0 should be tested to determine the most efficient way to release the pseudophosphatase and other proteins. Researchers may build their own lysis buffer by altering salt concentration, type of detergent, pH, and the presence of divalent cations, which are variables that may drastically alter the release of proteins in the cells. 7. Although an antibody for MK-STYX (anti-STYXL1) is now available, we prefer to use anti-FLAG for immunoprecipitating over-expressed Flag-tagged MK-STYX for the following reasons: (1) reliability, (2) specificity to the over-expressed MK-STYX, and (3) cost. Therefore, we recommend when immunoprecipitation does not require pulling down the endogenous protein of interest, an antibody against the tag epitope is used. Furthermore, if an active mutant phosphatase is used as a comparison tool of the pseudophosphatase, then an antibody against the tag epitope should be used. The success of immunoprecipitation is dependent on the purification of the antigen, which is affected by the abundance of antigen in the sample and the affinity of the antibody for the antigen. 8. Purifying the immune complex (pseudophosphatase-antibody and possible unidentified binding partners) relies on the appropriate secondary reagent that binds the antibody (IgA or IgG). Therefore, the appropriate beads (recombinant protein A or G) should be chosen), based on the affinity of the antibody for

Pseudophosphatase Function Through Biochemical Interactions

35

protein A or G. For example, the anti-FLAG antibody has a greater affinity for protein G; therefore, IgG beads were used to precipitate MK-STYX-fused to FLAG. 9. SDS-PAGE gels may be made at various concentrations (6–12%) dependent on the molecular mass (size) of the protein of interest. Typically, a 10% gel is used for resolving proteins of a broad range of molecular weights. When the protein of interest resolves at a large molecular weight, lower gel percentages are used, and higher gel percentages are used for proteins that resolve at smaller molecular weights. A gradient gel with various acrylamide percentages may also be used. Instead of making SDS-PAGE gels, they may also be purchased commercially at the desired percentages or gradients. 10. We have provided a protocol for the traditional way to transfer proteins to membranes. However, PVDF or nitrocellulose membranes may be purchased as stacks (membrane, blotting paper, cathode, anode, and the required buffers) for quick and reliable transfer. These stacks require the purchase of a specialized gel transfer device such as the iBlot (GE, Health Care) or Trans-Blot Turbo System (BioRad). We use both traditional and newer techniques and have found, with low protein concentrations, that the iBlot provides the best results; we have not tested the Trans-Blot Turbo System for these purposes. 11. Various blocking reagents may be used, dependent on the antibody required for the immunoblotting. Notably, 5% non-fat dry milk is a commonly used blocking reagent. To substitute for milk, many labs also use a commercial blocking reagent that accompanies the secondary antibodies and may be purchased from companies such as GE Healthcare. The antibody product sheet will provide specifications for antibody concentrations and/or ratios and blocking conditions to be used; some antibodies may require the use of BSA as a blocking agent. 12. When detecting (probing) proteins that migrate at or near each other, the membrane must be stripped and re-probed with the other primary antibody of interest. The reagents provided in this chapter are for a mild stripping buffer (Abcam). However, a more stringent stripping buffer (when residual is seen from the first blot) may be used. 13. High-quality RNA is important for gene expression experiments and is essential for accurate, good-quality, and consistent real-time quantitative PCR analysis [29]; a BioAnalyzer (Agilent) may also be used to quantitate the RNA as well as DNA or proteins.

36

Shanta´ D. Hinton

14. RNA interference (RNAi) is commonly used to examine intracellular protein functions; it allows sequence-specific repression of gene expression through mRNA degradation or translational repression [30]. RNAi may be accomplished by introducing small interfering RNAs (siRNAs) directly into cells or by introducing plasmid expression vectors for short hairpin RNAs (shRNAs) [31, 32], or viral vectors for short hairpins (more stable conditions) [33]. Methods in this chapter describe transfection techniques using a commercially available plasmid that co-expresses GFP and shRNA against the pseudophosphatase MK-STYX. The advantage of the co-expression of GFP is that transfection efficiency can be validated by viewing cells under a fluorescence microscope. 15. More than one shRNA should be included in the knockdown experiment to ensure that any results are not due to non-specific, off-target effects. shRNAs with different efficacies of knockdown also provide an additional level of control and verify for dose dependency of functional effects [33]. 16. RNAi experiments are only significant when the knockdown of the specific gene is confirmed; therefore, knockdown must be validated. Knockdown of the transcript is confirmed by qPCR, and the knockdown of the protein product may be confirmed by immunoblotting. Immunoblotting provides visualization of the protein knockdown (and may also provide semiquantitative validation), whereas qPCR allows highly accurate quantification of gene expression levels [29, 34]. We recommend using both methods in combination. A more recent gene editing technique, clustered regularly interspaced short palindromic tandem repeats (CRISPR), may also be used for deletion or addition of a specific gene of interest [35, 36]. 17. A purified protein is required for PTP activity assay. When purified protein is unavailable, cells may be transfected with plasmids encoding the pseudophosphatase, and immunoprecipitation is performed to obtain purified protein. Then, a PTP activity assay may be performed. PTP catalytic activity is very substrate-specific; therefore, commercial PTP activity kits may not accurately test for putative “restored” activity, and it may be necessary that specific substrates be radiolabeled with [γ-P32]ATP. 18. In the enzymology field, the term mutant is normally associated with creating an inactive version of an enzyme. Therefore, we coined the term “active mutant” to avoid confusion. 19. The protocol provided in this chapter describes non-stimulated conditions. Allow more culturing time for cells that require stimulation for activation or inactivation. Time requirements for stimulations are dependent on the stimulus, protein of

Pseudophosphatase Function Through Biochemical Interactions

37

interest, and cell type. Stimulating the MAPK signaling with epidermal growth factor in 293 (human embryonic kidney) cells requires early time points such as 0, 1, 3, or 5 min. 20. Immunoprecipitation should be done on ice; therefore, buffers, centrifuges, and microcentrifuge tubes should be at 4 °C. It is not necessary for the experiment to be done in a cold room; however, the 1.5-mL microcentrifuge tubes should be pre-chilled, and buffers stored on ice during the procedure. 21. The addition of phosphatase inhibitors or other inhibitors to the lysis buffer is dependent on the design of the experiment. When studying an active phosphatase, we suggest including phosphatase inhibitors in the lysis buffer. We add all inhibitors to the buffer immediately before lysing cells, providing enough time to dissolve in buffer on ice. 22. Another constraint of immunoprecipitation is determining the antibody–antigen ratio (how much antibody should be used to interact efficiently with the antigen). The protocol provided has been optimized for 1 mg/mL protein lysate. Occasionally, we have used 2 mg/mL; however, when a larger amount of protein lysate is preferred, the amount of antibody, as well as protein A or G agarose beads, should be optimized. 23. Pre-clearing lowers the background and improves the signalto-noise ratio by minimizing non-specific binding (proteins in the lysates that bind non-specifically are removed). Because of the commercial availability of protein A or G beads and the thought that these products are immediately ready for immunoprecipitation, many investigators think they can skip pre-clearing. However, we suggest pre-clearing the lysate (samples), at least with the beads. Pre-clearing may also be done by adding an irrelevant antibody to the lysate and following the immunoprecipitation procedure. We found that adding beads alone suffices and provides an efficient control “mock” immunoprecipitation. 24. It is helpful to pre-label tubes ahead of time with the correct lysate (sample) and antibody used. When we are immunoprecipitated with two different antibodies, we prefer to use two different colors of 1.5-mL centrifuge tubes. 25. The most common concern about immunoprecipitation is background (non-specific binding). However, it is important to note that the background may be specific when antigens are recognized by spurious antibodies used for the experiment, which gives rise to specific background bands. Methods to solve this problem are as follows: (1) remove the contaminated antibody by using an antigen affinity column to purify the antigen-specific antibody and (2) block the contaminated antibody’s activity by saturating it with the proteins that it binds.

38

Shanta´ D. Hinton

Non-specific background results from numerous sources. This problem is often solved by using more stringent washes, alternating wash buffers from high salt to low salt, and increasing the number of washes or length of wash. 26. Both Coomassie Blue and silver staining are used to visualize proteins. However, silver staining is 30-fold more sensitive and is compatible with mass spectrometric analysis. We use Coomassie Blue staining for the initial analysis of complexes. When the decision is made to identify a specific protein, silver staining is used so that mass spectrometry may be performed. Coomassie Blue and silver stains may be purchased commercially. The manufacturer’s protocol should be followed. 27. Collaboration should be established with a mass spectrometry facility during initial experiments so that the correct parameters for buffers and elution procedure may be established with the mass spectrometry facility performing the analysis. The decision may be not to elute samples from the beads with sample loading buffer and/or not to use Protein A or protein G beads to interact with the antibody. When mass spectrometry was performed on the immunoprecipitate for MK-STYX, the immunoprecipitation was performed with EZview™ RED ANTI-FLAGR M2 Affinity Gel, and the mass spectrometry facility performed the elution. 28. Whether overexpressing or knocking down MK-STYX in PC-12 cells, a morphological difference was immediately noticed. Overexpressing MK-STYX induced these cells to form neurite-like outgrowths, whereas knockdown prevented any outgrowth. Furthermore, RhoA activity decreased in the presence of MK-STYX and increased when MK-STYX was knocked down [17]. 29. To determine the role of MK-STYX in neuronal differentiation, we knocked down MK-STYX and stimulated cells with 100 ng/mL NGF [17]. 30. MK-STYX decreased RhoA activation; therefore, we evaluated MK-STYX’s effect on the RhoA downstream effector, cofilin. Knocking down MK-STYX increased cofilin phosphorylation in unstimulated cells but decreased it 24 h post nerve growth factor stimulation [17]. MK-STYX’s effects on RhoA and cofilin dynamics imply that this pseudophosphatase may have a role in biological functions where cytoskeletal reorganization is important. 31. The consensus sequence may be of the DUSP domain and/or of a motif, such as the kinase interaction motifs in MKPs (Fig. 1).

Pseudophosphatase Function Through Biochemical Interactions

39

32. Weblogo 3.7.4 is the current version to perform this analysis. However, it is important that the most recent version is always used. 33. A structure is needed to perform the structural analysis of how the mutations of the pseudophosphatase, in this case, MK-STYX, impact its structure and, therefore, its function. If a structure resolved through X-ray crystallography, nuclear magnetic resonance, and cryo-EM is available, it may be obtained through the PDB website. The 3D structure of MK-STYX has not been obtained; therefore, computational software was used to generate a structure. We used I-TASSER; however, the structure may be obtained through other software such as AlphaFold. 34. DeepRefiner is a deep learning-model-based protein structure refinement server for model refinement [25]. 35. POCASA demonstrated that there was no difference in the volumes, volume depth (VD) values, or shape of the predicted pockets when comparing the active signature motif of MK-STYX and its active mutant [37]. As many pseudophosphatases maintain their three-dimensional shape and have been reported to bind other proteins [4, 13, 24, 37, 38], this was expected. However, there was a change in the volume, and the binding pocket decreased in the KIM mutant compared with wild type (Fig. 3), supporting that MK-STYX does not bind MAPKs [7, 37, 39]. 36. To validate mutagenesis and POCASA results with a different computational approach, Missense3D was used. Only the wildtype MK-STYX was used, and single acid mutations for both the active site F245H and S246C and the KIM V53R were used (Fig. 3) [37]. These data supported the POCASAgenerated data; mutations in the KIM domain of MK-STYX may be why this pseudophosphatase doesn’t bind MAPK, although it binds several other molecules.

Acknowledgments We are grateful for researchers such as Jack E. Dixon, who dared to investigate catalytically inactive members of the PTP family, and current researchers pursuing pseudophosphatases, such as Hesso Farhan’s team. Furthermore, we would like to thank Nicholas K. Tonks for introducing us to the pseudophosphatase MK-STYX; it is a fascinating protein to study. We would also like to thank Lizabeth A. Allison for her interest and enthusiasm in our pursuit of the function of a “dead” phosphatase and Vincent Roggero for designing and formatting images. Finally, I thank my doctoral advisor, William R. Eckberg, for introducing me to

40

Shanta´ D. Hinton

research and all my former undergraduates and Master’s students who have tirelessly assisted with the projects related to MK-STYX. In particular, we thank the previous Master’s student, Emma Marie Wilber Hepworth, who performed the computational analysis of the kinase interaction motif of MK-STYX. This work was supported in part by Jeffress Memorial Trust Award J-931 (to S.D.H), National Science Foundation grants MCB1113167 and MCB 1909316 both to S.D.H., National Institutes of Health (NIH): NIH Research 1 R15 NS115074, the Howard Hughes Medical Institute Undergraduate Science Education Grant awarded to the College of William and Mary, Reves Faculty Fellowship (2019 and 2020) at the College of William and Mary to S.D.H., and the Virginia Space Grant to E.M.W.H. References 1. Todd AE, Orengo CA, Thornton JM (2002) Sequence and structural differences between enzyme and nonenzyme homologs. Structure 10:1435–1451 2. Murphy JM, Farhan H, Eyers PA (2017) Bio-zombie: the rise of pseudoenzymes in biology. Biochem Soc Trans 45:537–544 3. Wishart MJ, Denu JM, Williams JA, Dixon JE (1995) A single mutation converts a novel phosphotyrosine binding domain into a dualspecificity phosphatase. J Biol Chem 270: 26782–26785 4. Tonks NK (2009) Pseudophosphatases: grab and hold on. Cell 139:464–465 5. Tonks NK (2006) Protein tyrosine phosphatases: from genes, to function, to disease. Nat Rev Mol Cell Biol 7:833–846 6. Barr JE, Munyikwa MR, Frazier EA, Hinton SD (2013) The pseudophosphatase MK-STYX inhibits stress granule assembly independently of Ser149 phosphorylation of G3BP-1. FEBS J 280:273–284 7. Niemi NM, Lanning NJ, Klomp JA, Tait SW, Xu Y, Dykema KJ, Murphy LO, Gaither LA, Xu HE, Furge KA, Green DR, MacKeigan JP (2011) MK-STYX, a catalytically inactive phosphatase regulating mitochondrially dependent apoptosis. Mol Cell Biol 31:1357–1368 8. Begley MJ, Dixon JE (2005) The structure and regulation of myotubularin phosphatases. Curr Opin Struct Biol 15:614–620 9. Robinson FL, Dixon JE (2005) The phosphoinositide-3-phosphatase MTMR2 associates with MTMR13, a membrane-associated pseudophosphatase also mutated in type 4B Charcot-Marie-Tooth disease. J Biol Chem 280:31699–31707

10. Cheng KC, Klancer R, Singson A, Seydoux G (2009) Regulation of MBK-2/DYRK by CDK-1 and the pseudophosphatases EGG-4 and EGG-5 during the oocyte-to-embryo transition. Cell 139:560–572 11. Parry JM, Velarde NV, Lefkovith AJ, Zegarek MH, Hang JS, Ohm J, Klancer R, Maruyama R, Druzhinina MK, Grant BD, Piano F, Singson A (2009) EGG-4 and EGG-5 link events of the oocyte-to-embryo transition with meiotic progression in C. elegans. Curr Biol 19:1752–1757 12. Reiterer V, Fey D, Kolch W, Kholodenko BN, Farhan H (2013) Pseudophosphatase STYX modulates cell-fate decisions and cell migration by spatiotemporal regulation of ERK1/2. Proc Natl Acad Sci U S A 110:E2934–E2943 13. Wishart MJ, Dixon JE (1998) Gathering STYX: phosphatase-like form predicts functions for unique protein-interaction domains. Trends Biochem Sci 23:301–306 14. Hinton SD, Myers MP, Roggero VR, Allison LA, Tonks NK (2010) The pseudophosphatase MK-STYX interacts with G3BP and decreases stress granule formation. Biochem J 427:349– 357 15. Siligan C, Ban J, Bachmaier R, Spahn L, Kreppel M, Schaefer KL, Poremba C, Aryee DN, Kovar H (2005) EWS-FLI1 target genes recovered from Ewing’s sarcoma chromatin. Oncogene 24:2512–2524 16. Niemi NM, Sacoman JL, Westrate LM, Gaither LA, Lanning NJ, Martin KR, MacKeigan JP (2014) The pseudophosphatase MK-STYX physically and genetically interacts with the mitochondrial phosphatase PTPMT1. PLoS One 9:e93896

Pseudophosphatase Function Through Biochemical Interactions 17. Flowers BM, Rusnak LE, Wong KE, Banks DA, Munyikwa MR, McFarland AG, Hinton SD (2014) The pseudophosphatase MK-STYX induces neurite-like outgrowths in PC12 cells. PLoS One 9:e114535 18. St-Denis N, Gupta GD, Lin ZY, GonzalezBadillo B, Veri AO, Knight JDR, Rajendran D, Couzens AL, Currie KW, Tkach JM, Cheung SWT, Pelletier L, Gingras AC (2016) Phenotypic and interaction profiling of the human phosphatases identifies diverse mitotic regulators. Cell Rep 17:2488–2501 19. Qi Y, Kuang D, Kelley K, Buchser WJ, Hinton SD (2022) Evolutionary genomic relationships and coupling in MK-STYX and STYX pseudophosphatases. Sci Rep 12:4139 20. Kharitidi D, Manteghi S, Pause A (2014) Pseudophosphatases: methods of analysis and physiological functions. Methods 65:207–218 21. Mattei AM, Smailys JD, Hepworth EMW, Hinton SD (2021) The roles of pseudophosphatases in disease. Int J Mol Sci 22 22. Tomar VS, Baral TK, Nagavelu K, Somasundaram K (2019) Serine/threonine/tyrosineinteracting-like protein 1 (STYXL1), a pseudo phosphatase, promotes oncogenesis in glioma. Biochem Biophys Res Commun 515:241–247 23. Wu JZ, Jiang N, Lin JM, Liu X (2020) STYXL1 promotes malignant progression of hepatocellular carcinoma via downregulating CELF2 through the PI3K/Akt pathway. Eur Rev Med Pharmacol Sci 24:2977–2985 24. Reiterer V, Pawlowski K, Desrochers G, Pause A, Sharpe HJ, Farhan H (2020) The dead phosphatases society: a review of the emerging roles of pseudophosphatases. FEBS J 287:4198–4220 25. Shuvo MH, Gulfam M, Bhattacharya D (2021) DeepRefiner: high-accuracy protein structure refinement by deep network calibration. Nucleic Acids Res 49:W147–W152 26. Davis IW, Leaver-Fay A, Chen VB, Block JN, Kapral GJ, Wang X, Murray LW, Arendall WB 3rd, Snoeyink J, Richardson JS, Richardson DC (2007) MolProbity: all-atom contacts and structure validation for proteins and nucleic acids. Nucleic Acids Res 35:W375–W383 27. Yu J, Zhou Y, Tanaka I, Yao M (2010) Roll: a new algorithm for the detection of protein pockets and cavities with a rolling probe sphere. Bioinformatics 26:46–52

41

28. Ittisoponpisan S, Islam SA, Khanna T, Alhuzimi E, David A, Sternberg MJE (2019) Can predicted protein 3D structures provide reliable insights into whether missense variants are disease associated? J Mol Biol 431:2197– 2212 29. Bustin SA, Benes V, Garson JA, Hellemans J, Huggett J, Kubista M, Mueller R, Nolan T, Pfaffl MW, Shipley GL, Vandesompele J, Wittwer CT (2009) The MIQE guidelines: minimum information for publication of quantitative real-time PCR experiments. Clin Chem 55:611–622 30. Obbard DJ, Gordon KH, Buck AH, Jiggins FM (2009) The evolution of RNAi as a defence against viruses and transposable elements. Philos Trans R Soc Lond Ser B Biol Sci 364: 99–115 31. Aagaard L, Rossi JJ (2007) RNAi therapeutics: principles, prospects and challenges. Adv Drug Deliv Rev 59:75–86 32. Paddison PJ, Caudy AA, Bernstein E, Hannon GJ, Conklin DS (2002) Short hairpin RNAs (shRNAs) induce sequence-specific silencing in mammalian cells. Genes Dev 16:948–958 33. Moore CB, Guthrie EH, Huang MT, Taxman DJ (2010) Short hairpin RNA (shRNA): design, delivery, and assessment of gene knockdown. Methods Mol Biol 629:141–158 34. Derveaux S, Vandesompele J, Hellemans J (2010) How to do successful gene expression analysis using real-time PCR. Methods 50: 227–230 35. Deltcheva E, Chylinski K, Sharma CM, Gonzales K, Chao Y, Pirzada ZA, Eckert MR, Vogel J, Charpentier E (2011) CRISPR RNA maturation by trans-encoded small RNA and host factor RNase III. Nature 471:602–607 36. Jinek M, Chylinski K, Fonfara I, Hauer M, Doudna JA, Charpentier E (2012) A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity. Science 337: 816–821 37. Hepworth EMW, Hinton SD (2021) Pseudophosphatases as regulators of MAPK signaling. Int J Mol Sci 22:12595 38. Hinton SD (2016) Analyzing pseudophosphatase function. Methods Mol Biol 1447:139– 153 39. Hinton SD (2020) Pseudophosphatase MK-STYX: the atypical member of the MAP kinase phosphatases. FEBS J 287:4221–4231

Chapter 3 CRISPR/Cas9-Mediated Modification of PTP Expression Carolin Lossius, Anne Kresinsky, Laura Quiet, and Jo¨rg P. Mu¨ller Abstract Alteration of protein tyrosine phosphatase (PTP) gene expression is a commonly used approach to experimentally analyze their function in the cell physiology of mammalian cells. Here, exemplified for receptor-type PTPRJ (Dep-1, CD148) and PPTRC (CD45), we provide the CRISPR/Cas9-mediated approaches for their inactivation and transcriptional activation using genome editing. These methods are generally applicable to any other protein of interest. Key words Protein tyrosine phosphatase, CRISPR, Cas9, Nonhomologous end joining, Gene expression, Flow cytometry, Immune detection, Synergistic activation mediators

1

Introduction To study the role of protein tyrosine phosphatases (PTPs) in cell physiology, CRISPR/Cas9-mediated inactivation or overexpression is the method of choice. A prerequisite of a functional knockout is the nonessential function in standard cell function. Therefore, activation of gene expression is an excellent alternative approach to examining the function of an essential gene. Systems for CRISPR/Cas9-mediated alteration of gene expression evolved in the last decade to the standard technique to study their role in eukaryotes, in particular in mammalian cells [1]. By introducing RNA-guided DNA endonuclease Cas9 (for CRISPRassociated) and a guide RNA, any site at the chromosome can be reached and subsequently edited. Native Cas9 requires two separate RNA molecules—a crRNA complementary to the DNA target site and a trans-activating crRNA (tracrRNA). In contrast, CRISPR/ Cas9 applied in molecular biology has been simplified using a single chimeric guide RNA (here called sgRNA) [2]. The chosen targeting sequence for the 20-bp-long crRNA must be localized 5′ of an

Carolin Lossius and Anne Kresinsky contributed equally with all other contributors. Damien The´venin and Jo¨rg P. Mu¨ller (eds.), Protein Tyrosine Phosphatases: Methods and Protocols, Methods in Molecular Biology, vol. 2743, https://doi.org/10.1007/978-1-0716-3569-8_3, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

43

44

Carolin Lossius et al.

NGG PAM (protospacer adjacent motif) sequence. The PAM is required for the Cas9 to reach its target site [3, 4]. While the PAM for Cas9 derived from Streptococcus pyogenes is NGG, it differs for other Cas nucleases. After the nuclease induces double-strand breaks, the cellular repair system eliminates the Cas9-mediated chromosomal lesions by non-homologous end-joining (NHEJ). This imprecise repair leads to the loss of nucleotides, which can result in a frameshift mutation of exon structures, consequently resulting in allele-specific gene inactivation [5]. Before CRISPR/Cas9-mediated genome inactivation was applicable, RNA interference (RNAi) approaches were used. Cytosolic mRNA of particular genes was downregulated by transient transfection of small interfering double-stranded RNA molecules or stable expression of genes encoding small hairpin or modified microRNA molecules [6, 7]. A disadvantage of RNAi applications is their off-target effects affecting other genes besides the targeted gene expression. The so-called smart pools of siRNA molecules containing several targets at low molarity were designed to circumvent off-targeting. In addition to the off-target effects, RNAi only downregulates cytosolic mRNA levels, and low-level gene expression of the target gene remains [8]. Thus, the absence of a clear cell physiological response to RNAi might be due to the residual protein synthesis and activity. If permanent downregulation is not tolerated by cell physiology, inducible synthesis of interfering RNA molecules is also a way to use RNA interference for essential genes. Thus, the separation of cell amplification and downregulation of gene expression should be uncoupled. RNAi application is usually carried out by simultaneous manipulation of bulk populations of cells. CRISPR/Cas9-mediated gene inactivation is based on the mutation of the chromosomal alleles. Thus, it provides a neat background to study gene function. If NHEJ results in deletions of 3n base pairs, proteins with small deletions but no gene inactivation occur. In addition, because Cas9 nuclease acts site-specific, biallelic out-of-frame mutations are the prerequisite for gene inactivation. Documentation of abrogation of protein expression is necessary to ensure gene inactivation. Due to allele-specific NHEJ and short in-frame deletions, particular cell clones need to be established and characterized. Alternatively, the CRISPR/Cas9 system can be used for specific induction of gene expression, commonly called CRISPRa (for activation). Here, the two nuclease domains of Cas9 are inactivated by introducing the point mutations D10A and H840A. The so-called enzymatically inactive dCas9 is fused to a transcription factor like VP64, an RNA-guided transcriptional activator. The chosen sgRNA guides dCas9-VP64 to activate the target gene’s promoter. Besides, in the CRISPRa SAM (for synergistic activation mediator) system, additional domains in the tracrRNA regions are

CRISPR/Cas9-Mediated Modification of PTP Expression

45

used to recruit other transcription factors such as NFκB transactivating subunit 5 (p65) and HSP1 (activation domain of human heat-shock factor 1) to stimulate gene expression further [9]. To introduce Cas9 nuclease and gRNA to the cells, CRISPR/ Cas9 systems can be applied in several ways. Besides the direct introduction of the Cas9 nuclease protein and the respective sgRNA, genes encoding both components can be transfected (see Chapter 4 in this issue) or transduced to the cell. While transfection results in the transient synthesis of the components, virus-mediated transduction provides a stable ongoing expression of Cas9 and gRNA. For genome editing, a transient action of Cas9 and the respective sgRNA is usually sufficient if cells are transfectable with reasonable efficiency. For stable gene expression induction, permanent CRISPRa components synthesis is necessary. Systems applied for genome editing or gene activation follow the same experimental methodology described below. These lentivirus-mediated approaches alter the cellular PTP protein level but can also target other mammalian genes. While the gRNA-mediated targeting of nuclease Cas9 consequently results in genome edition with subsequent genetic inactivation, the use of enzymatically inactive dCas9, fused to transcriptional activator VP64, can be used for the induction of gRNA-targeted gene expression.

2

Materials

2.1 Production of Lentiviral Particles for Genome Editing

1. Derivative transfer vector lentiCRISPRv2 [10] (https://www. addgene.org/52961/) encoding sgRNA specific for the targeted gene. This plasmid encodes the Cas9 nuclease, a puromycin resistance cassette, and a U6 RNA polymerase III promoter controlling the expression of the encoded sgRNA. It is a generation III transfer vector (see Note 1). 2. Oligonucleotides to serve as sgRNA according to the following principle: FWD: 5′ CACCNNNNNNNNNNNNNNNNNNNN 3′ REV: 3′ CNNNNNNNNNNNNNNNNNNNNCAAA 5′ Where the end sequences (in bold) ensure that the oligonucleotides are compatible with the overhangs of the BsmBIdigested lentiCRISRPv2 vector, and the Ns correspond to the targeted sequence (see Note 2). 3. A GFP (green fluorescense protein)-encoded lentiviral control vector for cell marking (preferable leGO-G2, https://www. addgene.org/25917) to monitor transfection and transduction efficiency (see Note 3). 4. Packaging plasmids (for generation III transfer vector, two plasmids, such as pMDLg/pRRE and pRSV-Rev, are used).

46

Carolin Lossius et al.

5. Envelope plasmid (such as pVSVg for amphotropic or pEcoenv for ecotropic viral particles) for lentivirus production. 6. Vectors for gene activation are listed under Note 4. 7. HEK293T cells. 8. 100-mm Cell culture petri dish. 9. 6-Well tissue culture plates. 10. Dulbecco’s Modified Eagle Medium (DMEM), supplied with 10% defined fetal calf serum and penicillin/streptomycin. 11. PEI (polyethylenimine) stock solution 10 μg/μL in H2O. 12. EndoFree Plasmid Maxi Kit for plasmid isolation. 13. 0.2-μm Low-protein binding syringe filter holders. 14. Amicon Ultra-15 centrifugal filter unit 30 k. 15. Tabletop centrifuge to spin the 50-mL Amicon filter units Ultra-15. 16. Humidified incubator in an atmosphere of 5% CO2. 2.2 Transduction of Target Cells and Selection of Transduced Cells

1. 12-well tissue culture plates. 2. 96-well tissue culture plates. 3. Appropriate cell culture medium for targeted cells. The murine 32D and human MV4-11 cells used here are cultivated in RPMI medium supplemented with 10% heat-inactivated fetal calf serum and 1 mM sodium pyruvate. To cultivate factordependent 32D cells, murine recombinant IL3 (2.5 ng/mL) must be applied to the growth medium. 4. Polybrene (1,5-dimethyl-1,5-diazaundecamethylene polymethobromide). 5. Puromycin (10 mg/mL stock solution). 6. Blasticidin (10 mg/mL stock solution). 7. Hygromycin (50 mg/mL stock solution). 8. Zeomycin (50 mg/mL stock solution).

2.3 Characterization of Cell Clones with Altered Gene Expression of PTPRC or PTPRJ

1. 5-mL polystyrene round bottom tube 2. Phosphate-buffered saline (PBS): 137 mM NaCl, 2.7 mM KCl, 8 mM Na2HPO4, and 2 mM KH2PO4, pH 7.4. 3. Antibodies for cell sorting (Table 1). 4. Tube rotator. 5. Flow cytometer (e.g., BD Celesta) with associated software. 6. Fixation/permeabilization buffer (e.g., Cytofix/Cytoperm solution (BD, Heidelberg, Germany)). 7. Permeabilization/wash buffer (e.g., Cytofix/Cytoperm solution (BD, Heidelberg, Germany)).

CRISPR/Cas9-Mediated Modification of PTP Expression

47

Table 1 Antibodies for cell sorting Antibody

Clone

Specificity

Label

CD148 (PTPRJ)

MO86-5

Mouse

PE

CD45 (PTPRC)

30-F11

Mouse

PE

CD148 (PTPRJ)

143-41

Human

Nonea

CD45 (PTPRC)

30-F11

Human

APC/Cy7

a

The primary antibody is getting labeled with a Cy3-labeled secondary antibody (e.g., anti-Cy3 (A-6) from Santa Cruz)

3

Methods

3.1 Production of Lentiviral Pseudoparticles Encoding Cas9 and sgRNA

The production and subsequent transduction of cDNA-encoding lentiviral pseudoparticles is a standard method for the stable manipulation of mammalian cells. For safety reasons, produced viral particles are replication incompetent. Any lentiviral transfer vector can be packed into viral particles. The DNA sequences transferred into the mammalian cell genome are flanked by long terminal repeat (LTR) sequences, which facilitate the integration of the transfer plasmid sequences into the host cell genome. Generation II transfer plasmids have a wild-type 5′-LTR and are packaged using plasmid psPAX2, which encodes the Tat protein. Generation III transfer plasmids have a chimeric 5′-LTR that removes the requirement for the Tat protein. They are packaged using packaging plasmids pMDLg/pRRE and pRSV-Rev. The envelope plasmid defines the host spectrum for transduction. While amphotropic lentiviral particles created by using envelope plasmid VSVg allow the infection of all mammalian cells, using an MLV (eco) envelope encoding vector results in the production of ecotropic virus particles, which can infect cells derived from rodents only. As the production of viral particles and the efficiency of transduction of target cells cannot be monitored directly, the production of viral particles encoding a constitutively expressed fluorescence protein like the GFP gene should be included. This allows monitoring of the efficiency of transfection and subsequent transduction. GFP-positive cells can be easily monitored using a fluorescence microscope or flow cytometry. 1. Seed 2 × 106 HEK293T cells in a 100-mm cell culture petri dish (note: seeding should result in a 20–25% confluence. Make sure that cells get evenly distributed over the dish. 2. Incubate the plate at 37 °C in a humidified incubator in an atmosphere of 5% CO2 overnight.

48

Carolin Lossius et al.

3. Prewarm DMEM medium without fetal calf serum to room temperature. 4. Dilute 10 μg transfer vector (sgRNA-encoding lentiCRISPRv2-derivative plasmid) and lentiviral packaging plasmids pMDLg/pRRE (10 μg) and pRSV-Rev (5 μg) and envelope pVSVg (2 μg) or pEcoEnv (3 μg) plasmids in 0.25 mL DMEM. 5. In parallel, prepare lentiCRISPRv2 control plasmid not encoding a sgRNA and a GFP-encoding vector for cell marking (i.e., LeGO-G2). For the application of the CRISPRa SAM system, prepare transfer vectors lenti_sgRNA(MS2)_Zeo, lenti_dCASVP64_Blast, or lenti_MS2-p65-HSF1_Hygro, the lentiviral packaging plasmid psPAX2 (10 μg each), and the envelope encoding plasmid pVSVg (2 μg) or pEcoEnv (3 μg) in 0.25 mL DMEM (see Note 5). 6. Dilute PEI stock solution in 0.25 mL DMEM medium, using 2.5 μg PEI per 1 μg DNA (see Note 6). 7. Add diluted PEI solution to the DNA solution and vortex immediately. 8. Incubate the mixture at room temperature for 20–30 min. Fast and efficient homogenization of plasmids and PEI is critical for efficient transfection. 9. Replace FCS-containing DMEM medium from the packaging cell line and carefully replace it with 2 mL prewarmed DMEM medium without FCS. 10. Add DNA-PEI-solution dropwise to the cells. Carefully spread the solution over the entire surface of the well. Handle cells with care to prevent detachment. 11. Incubate for 4–8 h at 37 °C in a humidified incubator in an atmosphere of 5% CO2. 12. Replace PEI-containing DMEM medium with DMEM supplemented with 10% FCS. 13. Incubate cells for 24–72 h at 37 °C in a humidified incubator in an atmosphere of 5% CO2. 14. After 24–72 h post-transfection, collect the cell media to harvest the viral particles. 15. Filter the virus-containing medium into collection tubes through a sterile low-protein binding 0.2-μm filter unit. 16. Repeat harvesting virus-containing media over 3 days. 17. After removing media from the cells, carefully add prewarmed fresh FCS-containing DMEM medium. Handle cells with care to prevent detachment.

CRISPR/Cas9-Mediated Modification of PTP Expression

49

18. To raise the titer of lentiviral particles, centrifuge filtered viruscontaining culture medium using Amicon filter units at 3000 × g for 15 min (liquid flow: 1 mL/min). Notably, 10 mL virus-containing culture medium should be reduced to 0.5 mL (see Note 7). 3.2 Transduction of Target Cells and Selection of Transduced Cells

Lentiviral transduction is a powerful tool to stably modify mammalian cells. Lentivirus-mediated gene editing allows the infection of non-proliferating cells. Still, the efficiency of transduction is enhanced mainly in proliferating cells. After virus infection, regions localized between LTR domains of the transfer vector are reversely transcribed and integrated into the host cell genome. Permanent production of Cas9 is not problematic as the activity of Cas9 is impaired after site-specific mutation of the targeted sequence. It has been previously demonstrated that PTPRJ and PTPRC have overlapping functions in cell physiology [11]. Thus, it might be necessary to simultaneously inactivate both RPTP (receptor protein tyrosine phosphatases) genes. LentiCRISPRv2-encoded cDNAs are integrated into the host cell genome by virus-mediated transduction. Subsequently, transduced cells are selected by LentiCRISPRv2-encoded antibiotic resistance. Thus, an alternative antibiotic resistance marker needs to be introduced for a second transduction round. In particular, at lentiCRISPRv2, we replaced the puromycin resistance cassette with an alternative resistance gene. Thus, after inactivating PTPRC using a puromycin-resistant lentiCRISPRv2, PTPRJ was inactivated using a lentiCRISPRv2derivative containing a blasticidin cassette. Alternatively, multiple gene inactivation can be achieved by simultaneous transduction of transfection vectors encoding different sgRNA targeting sequences. Here, cell clones must be characterized separately for the inactivation of particular PTPs using flow cytometry or immune blotting. 1. Seed the targeted cell line into a 12-well plate. Cells should be seeded sparsely with about 2–5 × 104 cells per well. Use a medium for optimal growth conditions of the recipient cells (see Note 8). 2. Add polybrene (final concentration 8 μg/mL) to each well. Gently swirl the plate to distribute it evenly (see Note 9). 3. Add lentiviral particles to the wells. Gently distribute in the medium. The amount of virus-containing supernatant depends on viral titer. 4. Centrifuge plates at 500 × g for 60 min and further incubate plates at 37 °C in a humidified incubator in an atmosphere of 5% CO2.

50

Carolin Lossius et al.

5. After 8–16 h post-infection, remove polybrene and viruscontaining medium and replace it with a fresh growth medium optimized for the cells used (see Notes 9–11). 6. Incubate cells at 37 °C in a humidified incubator in an atmosphere of 5% CO2. 7. After 48 h post-transduction, replace the cell growth medium and add the appropriate antibiotic whose resistance gene is encoded by the transfer vector (see Note 12). 8. Maintain selection with antibiotics for 5–10 days until all un-transduced cells are inactivated. Due to stable virusmediated manipulation, selection can be omitted after that (see Note 13). 3.3 Selection of Cells with Altered Receptor PTP Expression

CRISPR/Cas9-mediated genome editing acts allele-specific. Therefore, independent cell clones need to be characterized separately. Thus, not all transduced puromycin-resistant cells will necessarily show abrogation of target protein production. In-frame deletions do not alter protein expression. Note that genome editing must not alter the transcription of the targeted gene. Thus, the mRNA pool will usually remain unchanged. In the case of the inactivation of receptor PTP, we can take advantage of the cell surface expression of these proteins. Available antibodies recognizing the extracellular exposed N-terminal part of the PTP allow the staining of viable cell populations (see Fig. 1 and Note 14). 1. After selection with puromycin, dilute cells to a concentration of 5 cells/mL (200 μL/well) in the appropriate growth medium and seed cells in 96-well plates. 2. Incubate cells for clonal amplification for several days. 3. Monitor daily to ensure the amplification of single-cell clones by bright field microscopy. 4. After clonal amplification of cells, transfer cell clones to 12-well plates after amplification to a cell count of 105 cells. 5. Harvest an aliquot of clonally amplified cell populations, centrifuge cells at 500 × g for 5 min, and wash them with PBS. Include a reference cell population transduced with the lentiCRISPRv2 control vector and an un-transduced parental cell population (see Note 15). 6. Resuspend cells in 48 μL PBS containing 2 μL of PE-labeled antibodies recognizing CD148 (for cells with targeted PTPRJ inactivation) or CD45 (for cells with targeted PTPRC inactivation). 7. Incubate cells at a tube rotator at 4 °C for 30 min. Centrifuge at 500 × g for 5 min.

CRISPR/Cas9-Mediated Modification of PTP Expression

51

Fig. 1 Flow cytometric analysis of cells inactivated for PTPRJ or PTPRC. Human MV4-11 cells (a, b) or murine 32D cells (c, d) transduced with lentiCRISPRv2 encoding sgRNA targeting PTPRJ (a, c) or PTPRC (b, d) were stained with PE-labeled CD148 antibody MO86-5 (a), PE-labeled CD45 antibody 30-F11 (b), APC-Cy7-labeled CD45 antibody 30-F11 (d), or labeled with primary anti-CD148 antibody and co-stained with Cy3 anti-mouse secondary antibody (c). Blots show unstained (left) and stained mock transduced wild-type cells (middle left) and two individual RPTP knock-out clones (middle right and right). Note: Clone MV4-11 PTPRC-/- clone 1 ((b), center right) showed partial PTPRC inactivation only

52

Carolin Lossius et al.

8. Resuspend the cell pellet with 300 μL PBS to remove unbound antibodies. 9. Centrifuge at 500 × g for 5 min. 10. Resuspend in 300 μL PBS. 11. Stained wild-type cells (transduced with lentiCRISPRv2 not encoding a sgRNA or mock transduced cells) are used as positive and unstained ctrl cells as negative control (see Fig. 1). 12. Analyze living cells for their PTP-staining with fluorescencecoupled antibodies by flow cytometry (see Note 16). If no efficient fluorescence-labeled antibody recognizing the extracellular domain of a receptor-type PTP is available or if a non-receptor PTP is targeted, detection of altered protein production must be carried out using fixed and permeabilized cells. Here, we combine a PTP-binding primary antibody and a fluorescencelabeled secondary antibody. The method is also applicable when using a fluorescence-labeled primary antibody. 1. Harvest cells and wash cells with PBS. 2. Fix and permeabilize cells using ice-cold fixation/permeabilization buffer for 20 min. 3. Centrifuge fixed cells at 500 × g for 5 min. 4. Resuspend cells in 100 μL permeabilization/wash buffer containing antibody. 5. Incubate for 30 min on ice. 6. Centrifuge 500 × g for 5 min. 7. Resuspend cells pellet in 100 μL permeabilization/wash buffer. 8. Repeat steps 6 and 7 one more time. 9. Carefully resuspend the cells in 48 μL permeabilization/wash buffer. 10. Add 2 μL fluorescence-labeled secondary antibody recognizing the primary antibody. 11. Incubate for 30 min on ice. 12. Centrifuge 500 × g for 5 min. 13. Resuspend cells pellet in 100 μL permeabilization/wash buffer. 14. Repeat steps 12 and 13 one more time. 15. Centrifuge cells at 500 × g for 5 min and resuspend in 300 μL PBS (see Note 17). 16. Analyze stained cells using flow cytometry. Robust detection of gene inactivation should be possible (Fig. 1).

CRISPR/Cas9-Mediated Modification of PTP Expression

4

53

Notes 1. 3′ of the sgRNA target sequence, an NGG PAM sequence is the prerequisite for activating Cas9. It must not be included in the target sequence. 2. To select appropriate sgRNA sequences, use sequence information from Addgene: For murine targets, use https://www.addgene.org/pooledlibrary/ zhang-mouse-gecko-v2/. For human targets, use https://www.addgene.org/pooledlibrary/zhang-human-gecko-v2/. Three or six different targets are presented for each gene. 3. For the inactivation of PTPRJ and PTPRC genes, the following target sequences were used: muPTPRJ: GGCCGCGGCCTTAGACTCGCA muPTPRC: CCACAAACCCATGGTCATATC huPTPRJ: CAGGCTCTAACCCGACAAGC huPTPRC: TTACCACATGTTGGCTTAGA 4. For the activation of gene expression, the following vectors have to be used: • Transfer vector Lenti_sgRNA(MS2)_zeo: This plasmid is the backbone for inserting sgRNAs. It encodes a zeomycin resistance cassette. The sgRNA sequences binding to the promoter region of human or murine genes are available at: https://www.addgene.org/pooled-library/zhanghuman-sam-v2/ or https://www.addgene.org/pooled-library/ zhangmouse-sam-v1/ • Inserting sgRNA-encoding sequences for articular genes into lenti_sgRNA(MS2)_zeo is identical to the way described above: Forward and complementary reverse oligonucleotides get annealed and subsequently inserted into lenti_sgRNA(MS2)_zeo hydrolyzed with BsmBI. For the activation of human PTPRJ, the following target sequences were used: huPTPRJ I: GCCGCAGCCGCCGGGCCGGCC huPTPRJ II: GCGGCCGCGGATCCCCTCCCG huPTPRJ III: GCTCGGAGGGGGCGGGGGCAG It is recommended to use several sgRNA sequences as different proximities to the promoter result in different efficiencies of transcription induction.

54

Carolin Lossius et al.

• Transfer vector lenti_dCAS-VP64_Blast: Plasmid encodes an enzymatically inactive dCas9 fused to VP64 and a blasticidin resistance cassette. • Transfer vector lenti_MS2-p65-HSF1_Hygro: Plasmid encodes MS2-p65-HSF1 (bacteriophage coat proteins MS2 fused with NFκB transactivating subunit p65) and HSF-1 (activation domain of human heat-shock factor 1) and a hygromycin resistance cassette. Lentiviral transfer vectors used for gene activation are generation II vectors. These plasmids necessarily need packaging plasmid psPAX2 for packaging into viral particles. 5. In parallel, produce a lentivirus sample derived from a lentiCRISPRv2 control vector containing no or a scrambled gRNA sequence. In addition, besides target vectors, a lentivirus particle containing a GFP should be transfected and transduced in parallel, allowing easy monitoring of the efficiency of infection. 6. Different transfection procedures can be used. PEI-mediated transfection results in efficient transfection rates of 80–90%. 7. The viral particles can be stored at 4 °C for a short time (up to several days). However, as the particles are relatively unstable, aliquot samples and freeze at -80 °C for long-term storage. 8. Although lentiviral particles can infect non-proliferating cells, exponentially growing cultures result in higher transduction rates. Seed cells sparsely to have a high ratio between virus particles and cells to be infected. 9. Polybrene enhances virus infection efficiency but is detrimental to particular cell types. Determine cell sensitivity if you work with a cell line for the first time. If cells lose viability, titrate down the concentration of polybrene. If polybrene is not tolerated well, shorten incubation time with polybrene. 10. If non-adherent cells are used, transfer cells to a polypropylene tube and spin them at 500 × g for 5 min. Subsequently, resuspend the cell pellet in a fresh growth medium and transfer cells back into wells. 11. To raise the efficiency of transduction, second and third infection rounds should be carried out by adding new virus particles to the cells and spin plates again. 12. To monitor selection efficiency, leave one well uninfected and select with the same amount of antibiotic. Resistance level against antibiotics is cell-specific. If a cell line is used for the first time, determine the resistance level by titrating antibiotic concentration using an un-transduced cell population.

CRISPR/Cas9-Mediated Modification of PTP Expression

55

13. To introduce the dCas9 system for gene activation, three different lentiviral transduction cycles must be carried out. First, lenti MS2-p65-HSF1_Hygro and lenti dCAS-VP64_Blast need to be transduced and selected independently. Lentiviral particles derived from lenti sgRNA(MS2)_zeo-derivative transfer vectors (and a lenti sgRNA(MS2)_zero-derivative encoding no or the scrambled RNA) get transduced last. 14. For the enrichment of cells inactivated for receptor PTP, flow cytometric cell sorting can be used. Transduced and antibioticselected pooled cell population can be stained for the receptor PTP using fluorescence-labeled antibodies and sorted using a BD Aria™ III sorter. 15. If adherent cells are used, trypsinize, withdraw an aliquot for cell characterization, and re-seed the remaining cells. 16. For gene activation, no clonal selection is necessary. Here the bulk population of cells can be used. To select cells with maximal activation of gene expression of receptor PTP, cells can be enriched using fluorescence-labeled antibodies and their subsequent sorting using a BD Aria™ III sorter. 17. Use un-transduced control cells in parallel as a reference. Control cells should be treated with fluorescence-labeled secondary antibodies but omit the primary antibody to monitor unspecific cross-reaction. If flow cytometry does not allow efficient detection of altered protein levels, classical immune detection using Western blotting of cell lysates with subsequent immune detection would have to be used (see Figs. 2 and 3).

Ptprc Ptprj β-actin

Fig. 2 Inactivation of PTP PTPRJ and PTPRC in 32D cells. 32D wild-type cells inactivated for PTPRJ, PTPRC, or both were analyzed for their RPTP expression. Duplicated cell samples were lysed in RIPA buffer, and equivalent amounts of protein were subjected to SDS-PAGE, blotted to a PVDF membrane, and analyzed with antibodies recognizing PTPRJ, PTPRC, and ß-actin

56

Carolin Lossius et al.

Ptprj β-actin

sgRNA

huPtprj I

-

+

-

-

+

huPtprj II

-

-

+

-

+

huPtprj III

-

-

-

+

+

MV4-11dCas9p65

Fig. 3 Gene activation of PTPRJ in MV4-11 CRISPR SAM PTPRJ cells. MV4-11 cells were transduced with CRISPR SAM system encoding components dCas9, MS2-p65-HSF1, and sgRNAs (PTPRJ I, II, or III, as indicated). Cells were grown under standard conditions, lysed, and subjected to SDS-PAGE and immunoblotting. Membranes were probed with antibodies recognizing PTPRJ and ß-actin

Acknowledgments The authors thank the Deutsche Forschungsgemeinschaft for funding the presented work. The study was supported by grants Mu955/14-1 and Mu955/15-1. References 1. Deltcheva E et al (2011) CRISPR RNA maturation by trans-encoded small RNA and host factor RNase III. Nature 471(7340):602–607 2. Cong L et al (2013) Multiplex genome engineering using CRISPR/Cas systems. Science 339(6121):819–823 3. Nishimasu H et al (2014) Crystal structure of Cas9 in complex with guide RNA and target DNA. Cell 156(5):935–949 4. Sternberg SH et al (2014) DNA interrogation by the CRISPR RNA-guided endonuclease Cas9. Nature 507(7490):62–67 5. Ran FA et al (2013) Genome engineering using the CRISPR-Cas9 system. Nat Protoc 8(11):2281–2308 6. Elbashir SM et al (2001) Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature 411(6836):494–498

7. Elbashir SM, Lendeckel W, Tuschl T (2001) RNA interference is mediated by 21- and 22-nucleotide RNAs. Genes Dev 15(2): 188–200 8. Hannus M et al (2014) siPools: highly complex but accurately defined siRNA pools eliminate off-target effects. Nucleic Acids Res 42(12): 8049–8061 9. Konermann S et al (2015) Genome-scale transcriptional activation by an engineered CRISPR-Cas9 complex. Nature 517(7536): 583–588 10. Sanjana NE, Shalem O, Zhang F (2014) Improved vectors and genome-wide libraries for CRISPR screening. Nat Methods 11(8): 783–784 11. Zhu JW et al (2008) Structurally distinct phosphatases CD45 and CD148 both regulate B cell and macrophage immunoreceptor signaling. Immunity 28(2):183–196

Chapter 4 Osteoclast Methods in Protein Phosphatase Research Nina Reuven, Maayan Barnea-Zohar, and Ari Elson Abstract Osteoclasts are specialized cells that degrade bone and are essential for bone formation and maintaining bone homeostasis. Excess or deficient activity of these cells can significantly alter bone mass, structure, and physical strength, leading to significant morbidity, as in osteoporosis or osteopetrosis, among many other diseases. Protein phosphorylation in osteoclasts plays critical roles in the signaling pathways that govern the production of osteoclasts and regulate their bone-resorbing activity. In this chapter, we describe the isolation of mouse splenocytes and their differentiation into mature osteoclasts on resorptive (e.g., bone) and non-resorptive (e.g., plastic or glass) surfaces, examining matrix resorption by osteoclasts, immunofluorescence staining of these cells, and knocking out genes by CRISPR in the mouse osteoclastogenic cell line RAW264.7. Key words Tyrosine phosphatase, Osteoclast, Bone, CRISPR, Pit formation

1

Introduction The mass, shape, and physical properties of bone are determined and maintained by the activities of two major cell types: mesenchymal osteoblasts (OBs), which produce bone matrix, and hematopoietically derived osteoclasts (OCLs), which degrade it. OCLs are large multi-nucleated phagocytes that are formed through differentiation and fusion of monocytes and macrophages in a complex and multi-stage process that is induced by the cytokines M-CSF (macrophage colony-stimulating factor, CSF-1) and RANKL (receptor activator of NFkB ligand) [1–3]. OCLs sense the topography of the underlying bone surface through their specialized adhesion structures called podosomes, which assemble into a beltshaped array that confines the bone area destined for degradation [4, 5]. Secretory lysosomes then fuse with the ventral cell membrane delimited by the podosomal belt [6], thereby enriching the cell membrane with proton pumps and chloride transporters that acidify the OCL–bone interface and discharging proteolytic enzymes onto the underlying bone surface [7, 8]. The combination

Damien The´venin and Jo¨rg P. Mu¨ller (eds.), Protein Tyrosine Phosphatases: Methods and Protocols, Methods in Molecular Biology, vol. 2743, https://doi.org/10.1007/978-1-0716-3569-8_4, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

57

58

Nina Reuven et al.

of low pH and proteases locally degrades the mineral and organic components of the bone matrix, respectively, and generates a resorption pit. Degradation of bone by OCLs is an integral and essential part of bone homeostasis, for example, by removing old and damaged bone material and enabling its replacement by a new matrix. The activity of OCLs is tightly regulated; excess or deficient activity can lead to serious diseases, such as osteoporosis [9] or osteopetrosis [10], respectively. Excess OCL activity can be reduced in clinical settings with anti-RANKL antibodies, which reduce the formation of OCLs, or with bisphosphonates, which induce OCL apoptosis [11]. Phosphorylation of tyrosine residues in proteins plays a major role in regulating the formation, adhesion, and function of OCLs. For example, M-CSF signals through its receptor on the plasma membrane, the tyrosine kinase c-Fms, which triggers a cascade of downstream phosphorylation events that help determine the survival and differentiation of OCLs. Protein phosphorylation plays similarly important roles also in RANKL signaling, and the tyrosine kinase Src is key in regulating the adhesion of OCLs to the matrix [12–15]. Multiple protein tyrosine phosphatases are expressed in OCLs, and their molecular and cellular functions have been studied [16]. Among the receptor-type PTPs, PTPROt (PTP-oc) supports the production and activity of OCLs by activating Src and downstream molecules [17–20] in a manner regulated by its C-terminal phosphorylation at Y399 [21, 22]. The cytosolic isoform of the receptor-type PTP Epsilon (cyt-PTPe) activates Src downstream of integrins and regulates the structure and organization in OCLs in a manner that is also dependent on its C-terminal phosphorylation [23–25]. CD45 also promotes OCL formation and function by regulating Src activity [26]. The receptor-type PTP, PTPRJ (CD148, DEP-1), regulates the maturation of multi-nucleated OCL precursors into functional OCLs, in part by dephosphorylating the M-CSF receptor and Cbl [27]. Other PTPs with known roles in OCLs include the nonreceptor-type PTP SHP1, which downregulates signaling events induced by RANKL and Src, leading to increased osteoclastogenesis in motheaten viable mice that express a partially active mutant SHP1 [28–31]. In contrast, the related PTP SHP2 promotes OCL production, suggesting that this PTP regulates signaling events downstream of the receptors for M-CSF [32] or RANKL [33]. Another non-receptor type PTP, PTP-PEST, promotes podosomal organization, is required for optimal attachment, spreading, and activity of OCLs [34, 35], and plays a role in cellcell fusion during the formation of OCLs or the related multinucleated giant cells [36]. The dual specificity phosphatase, MKP1, downregulates the activity of MAP kinases by dephosphorylating the threonine and tyrosine residues of their Thr-x-Tyr activation

Osteoclast Methods in Protein Phosphatase Research

59

site. MKP1 is a positive regulator of RANKL-induced formation of OCLs, but it is not clear if this PTP activates [37] or inactivates [38] these cells. Studies of OCLs are required to more fully understand the molecular processes that regulate their production and activity and to gain insights that might ultimately lead to the development of novel therapies for disease. The culture of OCLs is challenging since the mature cells are large, fragile, and terminally differentiated; they cannot be passaged or frozen. Consequently, OCLs are often produced in culture by cytokine-induced differentiation of precursor cells found in bone marrow, spleen, or blood. This chapter describes the production of mouse OCLs from primary spleen cells, as well as protocols for growing and differentiating these cells on bone fragments and mineral-coated tissue-culture plates, immunofluorescence staining of cells grown on glass coverslips or bone fragments, and CRISPR-mediated knockout of genes in the RAW264.7 osteoclastogenic cell line. This chapter expands upon an earlier one [39], in which the production of OCLs from mouse bone marrow cells, expression of foreign proteins, knocking down endogenous proteins in OCLs using adenoviral and lentiviral vectors, and assorted methods for studying signal transduction in the cells were described.

2

Materials

2.1 Production of OCLs in Culture

1. Mice (see Note 1). 2. 70% ethanol. 3. Recombinant Mouse TRANCE/RANK ligand/TNFSF11 (RANKL) (Catalog # 462-TEC-110-5, R&D Systems, Minneapolis, MN, USA) (see Note 2). 4. Recombinant Murine M-CSF (Catalog # 315-02, PeproTech, Rocky Hill, NJ, USA) (see Note 2). 5. Phosphate-buffered saline (PBS) solution: 8.1 mM Na2HPO4, 1.47 mM KH2PO4, 2.67 mM KCl, and 137 mM NaCl. 6. Surgical operating tools (two to three sets of small surgical scissors and two to three forceps (5–7 cm in length) with straight or curved ends). 7. OCL medium: alpha-MEM, supplemented with 10% fetal calf serum, 2 mM glutamine, 50 U/mL penicillin, and 50 μg/mL streptomycin. 8. Tissue-culture plasticware: 6- and 10-cm plates and 6- and 24-well plates. 9. Glass coverslips (optional, if growing cells for immunofluorescence studies) (see Note 3). 10. Cell strainers—70 or 40 μm.

60

Nina Reuven et al.

11. Pistons from 1-mL disposable syringes. 12. 20-μL, 200-μL pipettor tips, 15-mL test tubes 13. Hypotonic red blood cell lysis buffer (e.g., Sigma catalog number R7757). (All items from 3 onward are sterile). 2.2 Growth of OCLs on Bovine Bone Fragments and Visualization of Resorption Pits

1. Freshly prepared cells from mouse spleen (Subheading 3.1) or bone marrow [39]. 2. Fragments of bovine bone (prepared as described below). 3. RANKL and mouse M-CSF. 4. Phosphate-buffered saline solution (PBS): 8.1 mM Na2HPO4, 1.47 mM KH2PO4, 2.67 mM KCl, and 137 mM NaCl. 5. Forceps. 6. OCL medium: alpha-MEM, supplemented with 10% fetal calf serum, 2 mM glutamine, 50 U/mL penicillin, and 50 μg/mL streptomycin. 7. Tissue-culture plasticware (typically 6- or 24-well plates) (see Note 4). 8. 70% ethanol. 9. Horseradish peroxidase (HRP)-labeled wheat germ agglutinin (HRP-WGA). 10. 3-3′-Diaminobenzidine for HRP.

(DAB)-based

detection

reagents

(Items 1–7 should be sterile.) 2.3 Fluorescence Staining of OCLs

1. OCLs grown on glass coverslips or bone fragments (as described in Subheadings 3.1 and 3.2, respectively). 2. 24-well plates. 3. Parafilm. 4. Reagents for fluorescently staining cells, such as primary antibodies, fluorescently tagged secondary antibodies; fluorescently tagged phalloidin; Hoechst 33342, or similar fluorescent DNA-staining reagent. 5. Forceps (sharp- and blunt-ended). 6. Phosphate-buffered saline solution (PBS): 8.1 mM Na2HPO4, 1.47 mM KH2PO4, 2.67 mM KCl, and 137 mM NaCl. 7. Fixation solution: 3% paraformaldehyde (PFA) in PBS, pH 7.5. 8. Permeabilization solution: Fixation solution supplemented with 0.25% Triton X-100. 9. Mounting solution: Elvanol. 10. Glass microscope slides.

Osteoclast Methods in Protein Phosphatase Research

2.4 Growth of OCLs on CalciumPhosphate-Coated Plastic Plates

61

1. Buffer solution: 50 mM Tris base, adjusted to pH 7.4 with HCl. Prepare using pure (e.g., MilliQ) water. 2. Calcium stock solution: 25 mM CaCl2, 1.37 M NaCl, and 15 mM MgCl2, prepared in buffer solution. 3. Phosphate stock solution: 1.1 mM Na2HPO4 and 42 mM NaHCO3, prepared in buffer solution. 4. 2.5 × SBF (simulated body fluid) solution: Mix buffer solution, calcium stock solution, and phosphate stock solution at a ratio of 2:1:1. 5. CPS (calcium phosphate solution): 2.25 mM Na2HPO4, 4 mM CaCl2, and 0.14 M NaCl, prepared in buffer solution. 6. Sterile tissue-culture 24-well plates (tissue culture vessels of any format may be used). 7. Primary bone marrow or spleen cells (e.g., from Subheading 3.1). 8. RANKL. 9. M-CSF. 10. Phosphate-buffered saline solution (PBS): 8.1 mM Na2HPO4, 1.47 mM KH2PO4, 2.67 mM KCl, and 137 mM NaCl. 11. OCL medium: alpha-MEM, supplemented with 10% fetal calf serum, 2 mM glutamine, 50 U/mL penicillin, and 50 μg/mL streptomycin. 12. 70% ethanol. 13. 10% bleach (in water). (All solutions should be purchased sterile or sterilized by filtration through a 22-μm filter.)

2.5 Production of CRISPR-Mediated Knockout RAW 264.7 Cells

1. RAW264.7 cells (ATCC TIB-71™) (see Note 5). 2. Choose a suitable sgRNA (single guide RNA) to knock out your gene using any of the several free available web-based tools (see Notes 6 and 7). We use the UCSC Genome Browser (https://genome.ucsc.edu/), which provides a complete overview of the targeted genomic locus. Choose an sgRNA that targets an early coding exon in your target gene that is common to all isoforms listed. Select guides with the highest specificity scores and the fewest off-target sites. For PTPRJ, we chose a guide targeting exon 3 (guide: AGAATGTATAC GAAGTGCCT , PAM (protospacer adjacent motif): GGG). This exon is the first coding exon common to all the isoforms shown for this gene in Genome Browser. 3. Plasmid for expressing Cas9 and chosen sgRNA (see Note 8). 4. Oligonucleotides for cloning the chosen sgRNA into the expression vector (see Note 8).

62

Nina Reuven et al.

5. Oligonucleotide primers for genotyping by polymerase chain reaction (PCR). 6. PCR polymerase mix (preferably proofreading) (see Note 9). 7. Plasticware (pipette tips, PCR tubes, microcentrifuge tubes, cell culture dishes) (6-, 12-, 24-, and 96-well plates). 8. Cell transfection reagents (see Note 10). 9. Cell culture medium (DMEM supplemented with 10% fetal calf serum, 2 mM glutamine, 100 U/mL penicillin, and 100 mg/mL streptomycin) (see Note 11). 10. 50 mM NaOH. 11. 1 M Tris–HCl pH 8.

3

Methods

3.1 Production of OCLs in Culture from Unselected Spleen Cells of Mice

This protocol describes the production in vitro of OCLs from crude preparations of splenocytes of mice that are grown in the presence of the cytokines M-CSF and RANKL. This protocol reproducibly yields OCLs that may contain very small amounts of stromal and other cells; its advantages include its speed and relative ease and the ability to follow osteoclastic differentiation and manipulate it at selected time points. OCLs can also be produced in vitro from crude or purified preparations of mouse bone marrow cells, as described earlier [39]. 1. To prevent contamination of the tissue culture room, this step is performed on a clean, but not sterile, workbench in another room. Sacrifice a mouse by CO2 asphyxiation or by cervical dislocation. Wet the mouse with 70% ethanol to mat down the fur. Place the mouse on a clean dissection plate. Dissect and remove the spleen using scissors and forceps that have been immersed in 70% ethanol. Place the spleen inside a previously unopened sterile 15-mL tube. At this point, the spleens are not considered sterile. 2. If preparing cells from more than one mouse, repeat the above steps for each mouse. Sacrifice the mice only when you are ready to harvest their spleens. When collecting spleens from several mice (up to 10), collect the spleens one after the other, and only then proceed to the dissociation steps below. Several spleens can be pooled in one 15-mL tube or placed in individual tubes depending on the experiment. 3. Transfer the tubes with the spleens to a tissue culture room, but do not place them yet inside the sterile tissue culture hood. Inside the sterile hood, prepare two 3.5-cm plates for each spleen: Plate 1: 3–4 mL of sterile PBS. Plate 2: 1-mL sterile PBS and a cell strainer. Also prepare a 15-mL tube with 6–7 mL of sterile PBS.

Osteoclast Methods in Protein Phosphatase Research

63

4. While still outside the tissue culture hood, pour 1–2 mL of 70% ethanol into a 15-mL tube holding a spleen. After 5–10 s, decant the ethanol carefully and transfer the spleen into Plate 1 inside the hood to dilute and wash away the ethanol. If working with multiple spleens, perform the ethanol wash and transfer to Plate 1 for each spleen separately, making sure to limit the exposure time of each spleen to ethanol as above. From this point, the spleens and cells derived from them are considered sterile. 5. Using sterile forceps, transfer a spleen from Plate 1 to the inside of the cell strainer in Plate 2. Using the flat side of a sterile piston from a 1-mL syringe, crush the spleen onto the mesh of the cell strainer until the spleen is dissociated. The PBS present in this plate will facilitate this process. 6. Lift the cell strainer and add 1 mL fresh PBS to the inside of the strainer. Let the PBS drain into the plate below and dispose of the strainer. Pipette the PBS and the cells up and down several times to break cell clusters. Transfer the cell suspension into a 15-mL tube containing PBS. 7. Centrifuge at 350 × g for 4 min at room temperature. The cell pellet will typically appear reddish due to the presence of erythrocytes. 8. Remove the supernatant by aspiration inside the tissue culture hood. Aspirate with great care and retain a small amount of supernatant to avoid aspirating cells. 9. Lyse erythrocytes by gently resuspending the pellet in 1 mL of hypotonic red blood cell lysis buffer. Incubate at room temperature for 1 min (no longer), and then add 5 mL of PBS to halt lysis. 10. Centrifuge at 350 × g for 4 min at room temperature. The color of the cell pellet should now be lighter. 11. Carefully aspirate the supernatant inside the tissue culture hood. Resuspend cells in 2–4 mL OCL medium (without cytokines) per tube. Count the cells; expect about 40–80 × 106 cells per spleen of a young male C57Black/6 mouse. 12. Seed cells in appropriate vessels: 5 × 106 cells per well of a 6-well plate (2 mL medium), 1 × 106 cells per well of a 24-well plate (1 mL medium). Seed cells in OCL medium supplemented with 20 ng/mL M-CSF (no RANKL at this point). If seeding wells on glass coverslips, insert the sterile coverslips into the wells before adding cells. 13. Incubate at 37 °C, 5% CO2. The seeding day is referred to as Day 0.

64

Nina Reuven et al.

Fig. 1 Appearance of a culture of primary mouse splenocytes, as they differentiate into osteoclasts (OCLs) as seen by light phase microscopy. Time in parentheses indicates the time elapsed since seeding. RANKL was added at 72 h. Scale bars: 100 μm

14. The spleen contains relatively few M-CSF-responsive cells that will develop into OCLs. Consequently, on Day 1, only a small minority of the cells will have adhered to the plate. Gently remove the medium and the floating cells and replace with fresh OCL medium supplemented with 20 ng/mL M-CSF. From this point onward, the volume of medium per cell can be reduced (1.5 mL/well (6-well plate), 0.5–0.7 mL/well (24-well plate)) to conserve cytokines. Refresh the medium daily. 15. The M-CSF-responsive cells will proliferate and form associations of elongated cells (Fig. 1). On Days 2–4, depending on the cell density, initiate the osteoclastic differentiation by feeding cells with OCL medium that contains 20 ng/mL M-CSF and 20 ng/mL RANKL. Continue to replace medium daily until OCLs are formed (3–4 days after the initial addition of RANKL).

Osteoclast Methods in Protein Phosphatase Research

65

16. What to expect: Most of the cells plated are non-adherent (e.g., lymphocytes) and will be removed at the first medium change on Day 1. Relatively few adherent cells will remain, but these will proliferate massively in the presence of M-CSF. 24 h after the addition of RANKL, the cells will begin assuming a flatter shape (Fig. 1). After an additional 24–48 h with RANKL, initial cell fusion will become evident, and this will lead to the formation of large flat OCLs after an additional 24–48 h (Fig. 1). Mature mouse OCLs grown on glass/plastic have a very characteristic appearance—they are much larger than other cells and possess multiple nuclei, often in excess of 10–15. The cells will form initially in the densest regions of the well (typically the center and periphery) and then spread to other regions. Once large amounts of mature OCLs are formed, the culture will begin to deteriorate rapidly, so the cells should be used promptly. 17. Troubleshooting: The quality of the RANKL, M-CSF, and serum used is critical; If OCLs do not form, the fault is usually with one or more of these three reagents (see Note 2). 3.2 Growth of OCLs on Bovine Bone Fragments and Visualization of Resorption Pits

This protocol describes the production and growth in vitro of OCLs from crude preparations of mouse splenocytes or bone marrow cells on fragments of bovine bone and the subsequent visualization of the cells and the pits they excavate in the bone surface. OCLs naturally bind and interact with mineralized surfaces and do so in a manner that is quite distinct from their interaction with non-resorbable glass or plastic surfaces. Moreover, in vitro assays of OCL activity require the growth of these cells on a mineralized surface that they can degrade. A separate section describes the growth of OCLs on mineral-coated tissue-culture plates. 1. Preparation of bone fragments (see Note 12): Obtain pieces of fresh bovine bone (e.g., femur or tibia). Wash them in water and remove all non-bone materials, such as bone marrow. Using a diamond-tipped saw, cut the bone pieces into smaller fragments of approximately 2 × 2 mm area and 1 mm height. Take care to produce these small fragments from the cortical bone (the solid bone material that defines the boundaries of the bone) and not from the trabecular bone (internal bone structures of sponge-like appearance). 2. The small bone fragments may be air-dried and stored for future use at –20 °C. We have kept fragments for up to 1 year and expect that longer storage is possible. The original uncut bone pieces can be kept similarly for future use. 3. For use, thaw the required number of fragments. Using a pencil, mark the surface of the fragment on which the cells will NOT be seeded; this enables easy identification of the

66

Nina Reuven et al.

growth surface in case the fragments flip, e.g., when replacing medium. Sterilize the fragments by immersion in 70% ethanol for 3–5 min and then wash them several times with sterile PBS. 4. Using sterile forceps, transfer the bone fragments to the appropriate growth vessel, placing the fragments with the marked side facing down. It is possible to place several fragments together, e.g., two to four fragments, in a single well of a 24-well plate. Add OCL medium (without cytokines) and incubate in a tissue-culture incubator for 1–2 h. 5. Remove medium. Seed freshly prepared cells onto the bone fragments in OCL medium supplemented with 20 ng/mL MCSF (no RANKL). The number of cells seeded and the volume of medium should be appropriate for the size of the well, e.g., for primary spleen cells, 1 × 106 cells in 1 mL of medium/well (24-well plate) and 5 × 106 cells in 2 mL medium/well (6-well plate). 6. Replace the medium with similar M-CSF-containing medium daily; at 48–72 h after seeding, replace medium with OCL medium supplemented with 20 ng/mL M-CSF and 20 ng/ mL RANKL. Continue to replace medium daily with OCL medium containing M-CSF and RANKL until the endpoint of the experiment (see Notes 13 and 14). 7. To visualize pits formed by OCL-mediated bone resorption: remove cells from the bone surface. This can be done by brief sonication in PBS or immersion for 3–5 min in a solution of 1% bleach or 0.5 M NaOH, followed by debris removal with a cotton swab. Wash bone fragments in PBS. 8. To visualize the pits formed by active OCLs in the bone surface by HRP-WGA staining, incubate bone fragments with a solution of 2 μg/mL HRP-WGA in PBS. 9. Briefly rinse bone fragments in PBS and transfer them to a new tube/well. 10. Incubate the bone fragments with a solution of DAB, which forms a brown product in the presence of hydrogen peroxide and HRP. Continue incubation for several minutes, balancing the intensity of the pit color vs. the gradually increasing background (Fig. 2a, b). 11. Wash bone fragments in water. Stained fragments may be stored in water at 4 °C until analyzed (see Note 15). 12. Alternatively, pits may be visualized by incubating the bone fragments with a solution of hematoxylin (6 g/L) for 3–10 min, followed by washing with water. Pits stain violet (Fig. 2c, d). 13. OCLs grown on bone may be stained with fluorescently tagged antibodies or with other reagents, as described in Subheading 3.3.

Osteoclast Methods in Protein Phosphatase Research

67

Fig. 2 Staining of bovine bone fragments for visualization of resorption pits. (a, b) Fragments stained with HRP-tagged WGA. (a) Bone fragment without pits (background staining). (b) Bone fragment following culture of WT OCLs. Scale bars: 100 μm. (c, d) Fragments stained with hematoxylin: (c) WT OCLs. (d) OCLs from PTPRJKO mice. Scale bars: 200 μm

3.3 Fluorescence Staining of OCLs Grown on Glass Coverslips or Bone Slices

Like other cell types, OCLs can be stained with fluorescently tagged antibodies or other tagged molecules to visualize specific cellular proteins or structures. This method is of great importance in studies of OCLs since OCL cultures are very heterogeneous and contain cells at all stages of differentiation; visual examination enables focusing on individual cells of interest. Since OCLs interact with mineralized and non-mineralized surfaces in distinct manners, structural analysis of OCLs grown on both types of surfaces will often yield different results (Fig. 3a, b). This section describes the staining of OCLs grown on bone fragments or glass coverslips. 1. Seed and grow cells on sterile glass coverslips as described in Subheading 3.1. 2. Using forceps (see Notes 16 and 17), carefully lift the coverslips and transfer them individually to separate wells that contain PBS. 3. After 5 min, transfer the coverslips to wells that contain the fixation solution (see Note 17). Incubate at room temperature for 20 min. 4. Transfer the coverslips to new wells that contain the permeabilization solution. Incubate at room temperature for 3 min. 5. Transfer the coverslips to new wells that contain PBS. Incubate for 5 min and then transfer the coverslips to new wells containing fresh PBS. Further processing can be done immediately, or the coverslips can be stored in PBS at 4 °C until processed. 6. Optional: Before staining, block the cells by carefully transferring the coverslips to wells containing 1% BSA or serum in PBS, incubate for 30 min, and then transfer back to PBS-containing wells.

68

Nina Reuven et al.

Fig. 3 OCLs stained for actin and DNA. (a) Primary WT splenocytes were seeded on fragments of bovine bone differentiated into OCLs with M-CSF/RANKL. Cells were fixed and stained with TRITC-labeled phalloidin (to visualize actin) and Hoechst 33342 (to visualize DNA/nuclei). (b) A similar experiment, except cells were plated on glass coverslips. Cells grown on bone and other mineralized resorbable surfaces are smaller, and their actin-rich podosomes are organized in internal belts, sometimes several per cell. Cells on glass, plastic, and other non-resorbable surfaces are larger and contain a single podosomal belt at the cell periphery. Scale bars: 100 μm. (c) WT bone-marrow-derived cells were differentiated into OCLs on bovine bone fragments and stained with FITC-tagged phalloidin and Hoechst 33342. (d) Similar to (c), using cells from PTPRJ-KO mice. Note that the KO cells contain actin belts that are smaller, more compact, and less organized than the WT cells, as noted in [27]. Scale bars: 100 μm

7. In preparation for coverslip staining, attach a strip of Parafilm to the surface of a workbench. 8. Place 30 μL of the primary labeling solution (e.g., primary antibody or fluorescently tagged phalloidin, suitably diluted in PBS) on the Parafilm strip. 9. Carefully lift each coverslip from its well using forceps, and use it to overlay the drop of labeling solution with the cell-covered side facing down. Incubate for 60 min. If the labeling reagent is fluorescent, incubate it in the dark or cover it with aluminum foil or another non-transparent material. 10. Using forceps, carefully lift the coverslips from the Parafilm, place them - with cells facing upwards - individually in wells containing PBS, and incubate for 5 min. Repeat by moving each coverslip to a new PBS-containing well. 11. If required, label cells with a secondary reagent (e.g., fluorescently tagged secondary antibody) spotted on Parafilm as above. Incubate for 60 min in the dark. 12. Repeat washes in PBS as described in Item 10 above. A suitably diluted DNA stain can be added to any incubation step to visualize nuclei. Alternatively, nuclei can be stained by separately incubating the coverslips in the diluted DNA stain for 5 min and subsequent washing with PBS. 13. Place a drop of mounting medium on a glass microscope slide and overturn the coverslip with cells facing down into the mounting medium. Let dry and store at 4 °C until examined (see Fig. 3b).

Osteoclast Methods in Protein Phosphatase Research

69

14. The above procedure can result in significant contact between the cells and the Parafilm, which can damage or even remove some of them. To avoid this, one may perform the labeling steps (steps 9, 11, and 12) by transferring each coverslip, cellside up, to an individual well of a 24-well plate that already contains the diluted labeling solution. This procedure requires significantly more labeling solution (300 μL vs. 30 μL in the Parafilm-based method, per coverslip) but reduces possible damage to the cells. 15. OCLs grown on bone fragments can be stained according to the protocol described above (Fig. 3a, c, and d) with several small modifications: Due to their thickness, we prefer to stain the bone fragments in wells of tissue culture plates rather than on sheets of parafilm. When examined at the wavelength used to detect GFP, some auto-fluorescence occurs in bone. Typically, this is not severe enough to prevent the use of GFP-tagged reagents, but clearer images may be obtained using fluorescent markers at other wavelengths (see Note 14). Like glass coverslips, cell-carrying bone fragments may be stored in PBS at 4 °C after fixation, before staining, or after examination. 3.4 Growth of OCLs on CalciumPhosphate-Coated Plastic Plates

A possible alternative to growing OCLs on fragments of bone or dentine is to grow them on plastic plates coated with a thin layer of OCL-degradable mineral, such as calcium phosphate. Plates of this type can be purchased or produced in the lab using protocols such as the one presented here, which is based on [40, 41]. The calcium phosphate layer consists solely of mineral and differs in this respect from the natural bone matrix, which also contains proteins. Moreover, the mineral layer is quite thin, and active OCLs can degrade it and expose the underlying plastic material quite readily. This results in the OCLs re-organizing their podosomal adhesion structures from the closed inner rings characteristic of cells grown on resorbable mineralized surfaces to a single large peripheral podosomal ring found in OCLs grown on plastic or glass surfaces (e.g., compare Fig. 3a, b). Nonetheless, mineral-coated plates offer a useful and convenient alternative to bone fragments, and their flat surfaces are typically more amenable to documentation and quantification. 1. Incubate plates with 2.5 × SBF solution (1 mL/well) for 3 days at room temperature with daily changes. Perform this step in a tissue culture hood. 2. Aspirate 2.5 × SBF and add CPS (0.5 mL/well) for 3 days at room temperature with daily changes. Perform this step in a tissue culture hood. 3. Wash thoroughly with water, and dry at room temperature overnight.

70

Nina Reuven et al.

Fig. 4 Images of homemade mineral-coated plates before (a) and after (b) growth of osteoclasts (OCLs). Cells were removed with bleach, washed with water, and dried before photography. Note pits (closed circular structures) formed by the OCLs in (b). Scale bars: 200 μm

4. Sterilize plates by washing with 70% ETOH for 30 min, and dry in a biological hood. 5. Dried plates can be wrapped in Parafilm and stored at 4 °C until use. We have kept plates this way for over 2 years. 6. Before use, wash wells with 1 mL OCL medium (including serum, not including cytokines) three times. Incubate for 1 h in 1 mL OCL medium as above and discard before cell plating. 7. Seed cells at the appropriate density. Grow cells in the presence of M-CSF and RANKL as required (e.g., as described in Subheading 3.1). Refresh the medium every 1–2 days. 8. Remove cells by rinsing wells with 10% bleach. Wash wells three times with water, and air dry. 9. What to expect: As the OCLs mature and become active, they will degrade the mineral coat at their location and expose the plastic below. This can be observed and documented using phase light microscopy as the culture matures, and more so after cells are removed, as per item 8 above (Fig. 4). 3.5 Production of CRISPR-Mediated Knockout RAW 264.7 Cells

This protocol describes the CRISPR-mediated knockout of genes in the mouse osteoclastogenic RAW264.7 cell line, using the knock-out of PTPRJ as an example. The strategy described here uses one single guide RNA (sgRNA) to create a double-strand break (DSB) in an early exon of the targeted gene. Repair of the DSB is predominantly mediated by the non-homologous end-joining DNA repair pathway. This typically results in small insertions/ deletions (indels) at the break site, which may cause a frameshift and effectively knock out the expression of the protein product. Often, due to nonsense-mediated decay, the mRNA is also degraded. Cutting by the Cas9/sgRNA depends on a precise match between the sgRNA and the genomic sequence and the presence of the PAM in the genomic sequence. Therefore, one

Osteoclast Methods in Protein Phosphatase Research

71

must confirm by sequencing that the chosen sgRNA and PAM sequences are present in the genomic sequence of the RAW264.7 cells first. 1. Select primer sequences to amplify by PCR a 450–700 bp fragment from RAW264.7 genomic DNA that contains the guide sequence for comparison with the database sequence. Ensure that the guide sequence is located at least 200 bp from the ends of the fragment. NCBI primer blast (https:// www.ncbi.nlm.nih.gov/tools/primer-blast/) is a useful tool for choosing PCR primers. 2. Prepare crude template genomic DNA from RAW264.7 cells: Pellet 5 × 104–1 × 105 cells in a 1.5-mL microcentrifuge tube. Aspirate the supernatant and resuspend the pellet in 18 μL 50 mM NaOH with gentle pipetting. Incubate at 95 °C for 10 min, centrifuge for 1 min at 13,000 × g, and add 2 μL of 1 M Tris–HCl pH 8. 3. PCR-amplify the targeted genomic region using the primers chosen in item 1 and the genomic DNA of the wild-type (WT) RAW264.7 cells from item 2. Use 1 μL of the undiluted genomic DNA per 20 μL reaction, but note that it may be necessary to dilute the genomic DNA 2–50-fold. Adjust the conditions of the PCR reaction to achieve a clear, single PCR product. If this is not possible, choose a different primer set. 4. Sequence the PCR product. It is recommended to do this by directly sequencing an aliquot of the diluted PCR reaction since it greatly simplifies the processing of these samples. This will become more critical when screening the putative knockout clones by DNA sequencing (item 12). If this method does not give good results, purify the PCR product to prepare it for sequencing. Verify that the sequence matches the database, particularly for the guide/PAM. If the sequences do not match, select a different guide sequence that matches the RAW264.7 DNA sequence. The sequence file of this genomic region will be used later as a standard for comparison with the sequences of the putative knock-out clones, so the sequence must be clear and of good quality. 5. Clone the guide into the chosen Cas9/guide plasmid (see Note 8), verify the sequence, and prepare plasmid DNA. 6. Seed RAW264.7 cells in a 24-well plate, 50,000 cells/well, on the day before transfection (see Note 18). 7. Transfect the cells with 1 μg DNA (see Note 10) according to the manufacturer’s protocol. If the Cas9/sgRNA plasmid does not have its own selection marker, co-transfect with a plasmid that provides selection, using 700 ng Cas9/sgRNA plasmid and 300 ng of the selectable plasmid. To generate “wild-type”

72

Nina Reuven et al.

control cells, transfect with the Cas9/sgRNA plasmid bearing a non-targeting guide. A negative control for the selection should also be prepared: cells transfected with a control plasmid without the antibiotic resistance marker. 8. Replace the medium 6 h post-transfection. 9. At 36–48 h post-transfection, remove the cells from the plate by dislodging them using a cell scraper and/or pipetting. Re-seed the cells into a larger well (12- or 6-well plate) using medium with the selective antibiotic. Replate a small sample of cells into a small well with no antibiotics as a control for selection. The cells will generally be more sensitive to the antibiotic upon re-seeding, particularly if they were dense before re-seeding. 10. After 24 h, change the medium to medium without selection (see Note 19). Replace medium (without selection) every 2–3 days, as needed, to remove dead cells. FACS sorting may be used to recover the transfected cells for transfections employing a fluorescent marker. 11. When the targeted cells have recovered and are growing as single cells or as small colonies (4–5 days post-selection), and the negative control cultures have no surviving cells, harvest the targeted cells, and count them. 12. Dilute the cells and plate them in several 96-well plates to deposit a single cell per well. As it is difficult to achieve this level of precision, plate several plates, varying the dilutions two to fourfold, so that ultimately there will be enough single cell colonies to assay (see Note 20). To help the cells grow under these conditions, plate them in growth medium supplemented with 20% conditioned medium (medium from a 2–3-day-old culture of RAW264.7 cells that has reached approximately 70% confluency, filtered through a 0.22-μm filter before use). 13. Replate the remainder of the selected cells as a pool. Use some cells to prepare genomic DNA and amplify the targeted region using the primers and protocols described above in items 1–4. Continue to maintain the pool of edited cells or freeze them for use if the first round of single-cell plating does not yield sufficient knock-out cell clones. 14. Sequence the amplified genomic region of the pool of targeted cells. If the editing has been successful, the sequence should contain overlapping sequences that will diverge, beginning close to the location of the cleavage site dictated by the sgRNA. Useful tools for deconvoluting these sequences are Synthego ICE (Inference of CRISPR Edits; https://ice. synthego.com/#/) or TIDE (https://tide.nki.nl/). Upon uploading the raw sequence files of the WT control and the

Osteoclast Methods in Protein Phosphatase Research

73

edited pool, as well as the sequence of the sgRNA, the analysis tool displays the different deconvoluted sequences and determines what percentage of the sequences have been edited. This analysis helps to determine what percentage of the edited cells are likely to be knockouts and, thus, how many colonies will have to be screened. With a good guide RNA, at least 50% of the sequence reads should display editing. 15. When the colonies (item 12) have grown (approximately 2 weeks), they can be transferred to a 24-well dish for expansion. 16. Prepare genomic DNA from samples of the individual colonies, and analyze the targeted genomic region by PCR, sequencing, and ICE analysis (see Note 21). 17. Verify the knockout status of selected clones by Western blotting and/or quantitative PCR (Fig. 5).

Fig. 5 Targeting of PTPRJ in RAW264.7 cells by CRISPR. (a) Top: Nucleic acid and protein sequences in exon 3 at the targeted site. The sgRNA (green) and PAM sequence (pink) target the opposite strand and are shown in their anti-sense orientation. Bottom: Similar sequences of the targeted clone 4C. This clone yielded a single sequence with a 4 bp deletion, which resulted in premature termination of the protein product; this is due either to both alleles being targeted identically or to a massive deletion in the other allele that prevented its PCR amplification and sequencing. (b, c) Images of WT (b) and PTPRJ-KO (c, Clone 4C), stained for actin (red, podosomes) and DNA (blue, nuclei). Note reduced differentiation of PTPRJ-KO cells, as noted in [27]. Scale bar: 200 μm. (d) Protein blot documenting lack of PTPRJ protein expression in two PTPRJ-KO clones. Molecular mass marker is in kD

74

4

Nina Reuven et al.

Notes 1. The strain of mice used can influence the yield of OCLs. In our experience, mice of the strains C57Black/6 and FVB/N yield good cultures, especially when grown in 24- or 6-well plates. Strain 129 SvEv mice yield excellent cultures that grow well in larger plates. 2. M-CSF and RANKL can be produced and purified in-house using bacterial or cellular systems that overexpress their respective cDNAs. In some cases, M-CSF can be used as conditioned media from overexpressing cells. Alternatively, these cytokines can be purchased from commercial suppliers. The source chosen should balance between the need for cytokines of consistently high quality and their relatively high cost. We have had consistently good experiences with the commercial cytokines noted in the text, but additional commercial sources exist. 3. The make of the glass can affect OCL differentiation. We have had good results with coverslips manufactured by MenzelGlaser, Braunschweig, Germany. 4. The cells are grown on bone fragments placed in plastic dishes or wells. Non-tissue culture-grade (e.g., bacteriological) plasticware may therefore be used. 5. All manipulations of the RAW264.7 cells should be performed under sterile conditions, using standard tissue culture equipment and protocols. 6. Avoid knockout strategies that retain alternative start sites (in-frame ATGs) downstream of your planned cleavage site. If the mutated allele produces mRNA, translation starting from such sites could produce a truncated protein which may confound the results. 7. A gene can be targeted with two sgRNAs to engineer a large deletion. However, in RAW264.7 cells, we find that most of the repair of these dual DSBs results in small indels at each cleavage site rather than the deletion of the DNA between them. This makes it challenging to isolate cells with the desired deletions and more complicated to analyze the putative knockout clones. 8. There are many options for plasmids expressing Cas9 and its associated sgRNA. We use those based on the plasmids developed in the Zhang lab at MIT, such as pX330 (Addgene # 42230), which expresses SpCas9 from a Cbh promoter and the sgRNA from a U6 promoter or other plasmids based on this format. Consult the Addgene website for tips on using these plasmids and for the sgRNA cloning protocol (https://www. addgene.org/crispr/zhang/). For selecting the transfected

Osteoclast Methods in Protein Phosphatase Research

75

cells, one may use a plasmid expressing Cas9-T2A-puro (pX459, Addgene# 62988), or Cas9-T2A-mCherry (Addgene #64324), to select the transfected cells with either puromycin or by FACS sorting, respectively. Other similarly constructed plasmids are available with different selectable markers from Addgene or commercial vendors. Alternatively, the basic Cas9/ sgRNA plasmids can be co-transfected with a plasmid expressing a desired selection marker. 9. Because some PCR polymerases or their formulations may be sensitive to the crude genomic DNA preparation used in this protocol, it may be necessary to try a few from different vendors before finding a suitable polymerase. Since the amplified fragments will be sequenced to determine the specific mutations generated in the cell lines, a proofreading polymerase is recommended to ensure an accurate sequence. We have successfully used the 2× ready mix PCRBIO VeriFi (PCR Biosystems). 10. We have used jetPEI (Polyplus) to transfect the RAW264.7 cells, with an efficiency of 20–40%. 11. RAW264.7 cells are grown without M-CSF or RANKL. These cytokines are added to the medium only when osteoclastogenic differentiation is initiated. RAW264.7 cells produce M-CSF, so in some protocols, osteoclastogenic differentiation is induced in these cells using only 20 ng/mL RANKL. We typically add 20 ng/mL M-CSF to the differentiation medium to ensure sufficient M-CSF. 12. We purchase pieces of bovine bone from a local meat shop and cut them to size as described here. Alternatives for bovine bone include fragments or discs of dentin, which can be purchased commercially, or pieces of bones from other animal species. 13. Unless they express fluorescent markers, it is very difficult to see the living cells while they are growing on bone. We, therefore, seed several bone fragments per experimental point and process them individually around the expected endpoint, considering that OCL production is somewhat slower, and the cells survive longer on bone than on glass or plastic growth surfaces. 14. Since bone is not transparent, care has to be taken to ensure that the correct surface of the bone is exposed to the microscope. Inverted microscopes require gently flipping the fragment, so the cell-bearing surface faces downward. This results in the growth surface contacting the bottom of the well, but the cells are typically unaffected if done gently. 15. Note that the pits visualized at a particular time point represent the cumulative bone-degrading activity of cells on the bone fragment up to that point.

76

Nina Reuven et al.

16. A coverslip can be lifted from a well by using a pair of sharpended forceps to gently push it against the side of the well and gently pry it upward from the bottom of the well. While holding up the coverslip, use a second pair of forceps to grab it and transfer it elsewhere. The presence of liquid in the well facilitates this process. 17. We prefer to add coverslips to wells that already contain liquid instead of liquid to wells that contain coverslips. This reduces shear forces that may harm or even dislodge the larger and more fragile OCLs. 18. This quantity of cells is sufficient, although one may begin with more cells (and scale up the plate size and transfection components), as many cells will be lost during the selection. After selection, the surviving cells are few and dilute. Nevertheless, plate the cells for single-cell colonies early after selection to avoid multiple isolations of the same clone. 19. The ideal knock-out cell lines will be mutated only at the targeted locus. They will not have incorporated any plasmids used to engineer the mutations into their genomes. The selection should be applied as a strong pulse for a short time to select the cells that have been productively transfected transiently. If the selection is applied later, or for a longer period, many of the transiently transfected cells that have lost the selective plasmids will die, and the selection will favor cells that have randomly incorporated the selection plasmid into their genomic DNA. After the short 24-h pulse with the selection reagent, cells will continue to die even after the medium is changed to medium without selection. The surviving selected cells will recover, which may take a few days. If the selection has worked properly, the negative control transfected cells should all be dead at this point, while the culture transfected with the plasmid with the resistance marker should have viable cells. The surviving cells may be transferred to a smaller well to encourage their growth. 20. The single cells will take approximately 2 weeks to form clear colonies. One must inspect the wells during this time to identify wells containing a single colony. 21. Sequencing of the single-cell clones should give two overlapping sequences representing the two alleles of the gene. Most often, the two chromosomes will be repaired independently, and thus each will have its unique indel. It is essential to verify the sequence suggestions presented by the analysis tool by manually examining the peaks of the Sanger sequencing to be sure that they match the computed output. Some clones may display only a single sequence that is mutated. This can arise from identical repair of both alleles or a large deletion in one

Osteoclast Methods in Protein Phosphatase Research

77

allele that eliminates the primer binding site(s) used to amplify the genomic region by PCR. Such a large deletion should typically also lead to a functional knockout of this allele (Fig. 5). For a more careful analysis of the targeted locus, other primer sets can be employed for PCR, or Southern blot analysis can be considered. In addition, to verify the sequence identity of each allele, one option is to subclone the PCR products of the targeted genomic region into a convenient vector, such as pBluescript. Each bacterial clone will contain the genomic sequence of only one of the alleles; by sequencing the cloned PCR product from several bacterial colonies, the sequence identities of the two alleles can be confirmed.

Acknowledgments This study was supported by grants from the Israel Science Foundation (1734/20) and from the Kekst Family Institute for Medical Genetics at the Weizmann Institute of Science, Israel. References 1. Feng X, Teitelbaum SL (2013) Osteoclasts: new insights. Bone Res 1(1):11–26 2. Teitelbaum SL (2007) Osteoclasts: what do they do and how do they do it? Am J Pathol 170(2):427–435 3. Tsukasaki M, Huynh NC, Okamoto K, Muro R et al (2020) Stepwise cell fate decision pathways during osteoclastogenesis at single-cell resolution. Nat Metab 2(12):1382–1390 4. Geblinger D, Zink C, Spencer ND, Addadi L et al (2012) Effects of surface microtopography on the assembly of the osteoclast resorption apparatus. J R Soc Interface 9(72): 1599–1608 5. Georgess D, Machuca-Gayet I, Blangy A, Jurdic P (2014) Podosome organization drives osteoclast-mediated bone resorption. Cell Adhes Migr 8(3):191–204 6. Ng PY, Brigitte Patricia Ribet A, Pavlos NJ (2019) Membrane trafficking in osteoclasts and implications for osteoporosis. Biochem Soc Trans 47(2):639–650 7. Novack DV, Teitelbaum SL (2008) The osteoclast: friend or foe? Annu Rev Pathol 3:457– 484 8. Teitelbaum SL (2011) The osteoclast and its unique cytoskeleton. Ann N Y Acad Sci 1240: 14–17

9. Compston JE, McClung MR, Leslie WD (2019) Osteoporosis. Lancet 393(10169): 364–376 10. Palagano E, Menale C, Sobacchi C, Villa A (2018) Genetics of osteopetrosis. Curr Osteoporos Rep 16(1):13–25 11. Pavone V, Testa G, Giardina SMC, Vescio A et al (2017) Pharmacological therapy of osteoporosis: a systematic current review of literature. Front Pharmacol 8:803 12. Wada T, Nakashima T, Hiroshi N, Penninger JM (2006) RANKL-RANK signaling in osteoclastogenesis and bone disease. Trends Mol Med 12(1):17–25 13. Ross FP (2006) M-CSF, c-Fms, and signaling in osteoclasts and their precursors. Ann N Y Acad Sci 1068:110–116 14. Nedeva IR, Vitale M, Elson A, Hoyland JA et al (2021) Role of OSCAR signaling in osteoclastogenesis and bone disease. Front Cell Dev Biol 9:641162 15. Izawa T, Zou W, Chappel JC, Ashley JW et al (2012) c-Src links a RANK/αvβ3 integrin complex to the osteoclast cytoskeleton. Mol Cell Biol 32(14):2943–2953 16. Shalev M, Elson A (2018) The roles of protein tyrosine phosphatases in bone-resorbing osteoclasts. Biochim Biophys Acta 1866(1): 114–123

78

Nina Reuven et al.

17. Lau KH, Stiffel V, Amoui M (2012) An osteoclastic protein-tyrosine phosphatase regulates the beta3-integrin, syk, and shp1 signaling through respective src-dependent phosphorylation in osteoclasts. Am J Physiol Cell Physiol 302(11):C1676–C1686 18. Amoui M, Sheng MH, Chen ST, Baylink DJ et al (2007) A transmembrane osteoclastic protein-tyrosine phosphatase regulates osteoclast activity in part by promoting osteoclast survival through c-Src-dependent activation of NFkappaB and JNK2. Arch Biochem Biophys 463(1):47–59 19. Yang JH, Amoui M, Lau KH (2007) Targeted deletion of the osteoclast protein-tyrosine phosphatase (PTP-oc) promoter prevents RANKL-mediated osteoclastic differentiation of RAW264.7 cells. FEBS Lett 581(13): 2503–2508 20. Lau KH, Wu LW, Sheng MH, Amoui M et al (2006) An osteoclastic protein-tyrosine phosphatase is a potential positive regulator of the c-Src protein-tyrosine kinase activity: a mediator of osteoclast activity. J Cell Biochem 97(5): 940–955 21. Roth L, Wakim J, Wasserman E, Shalev M et al (2019) Phosphorylation of the phosphatase PTPROt at Tyr(399) is a molecular switch that controls osteoclast activity and bone mass in vivo. Sci Signal 12(563):eaau0240 22. Wakim J, Arman E, Becker-Herman S, Kramer MP et al (2017) The PTPROt tyrosine phosphatase functions as an obligate haploinsufficient tumor suppressor in vivo in B-cell chronic lymphocytic leukemia. Oncogene 36(26):3686–3694 23. Levy-Apter E, Finkelshtein E, Vemulapalli V, Li SS et al (2014) Adaptor protein GRB2 promotes Src tyrosine kinase activation and podosomal organization by protein-tyrosine phosphatase in osteoclasts. J Biol Chem 289(52):36048–36058 24. Granot-Attas S, Luxenburg C, Finkelshtein E, Elson A (2009) PTP epsilon regulates integrinmediated podosome stability in osteoclasts by activating Src. Mol Biol Cell 20(20): 4324–4334 25. Chiusaroli R, Knobler H, Luxenburg C, Sanjay A et al (2004) Tyrosine phosphatase epsilon is a positive regulator of osteoclast function in vitro and in vivo. Mol Biol Cell 15(1):234–244 26. Shivtiel S, Kollet O, Lapid K, Schajnovitz A et al (2008) CD45 regulates retention, motility, and numbers of hematopoietic progenitors, and affects osteoclast remodeling of metaphyseal trabecules. J Exp Med 205(10): 2381–2395

27. Shalev M, Arman E, Stein M, Cohen-Sharir Y et al (2021) PTPRJ promotes osteoclast maturation and activity by inhibiting Cbl-mediated ubiquitination of NFATc1 in late osteoclastogenesis. FEBS J 288(15):4702–4723 28. Aoki K, Didomenico E, Sims NA, Mukhopadhyay K et al (1999) The tyrosine phosphatase SHP-1 is a negative regulator of osteoclastogenesis and osteoclast resorbing activity: increased resorption and osteopenia in me(v)/ me(v) mutant mice. Bone 25(3):261–267 29. Zhang Z, Jimi E, Bothwell AL (2003) Receptor activator of NF-kappa B ligand stimulates recruitment of SHP-1 to the complex containing TNFR-associated factor 6 that regulates osteoclastogenesis. J Immunol 171(7): 3620–3626 30. Ke K, Sul OJ, Choi EK, Safdar AM et al (2014) Reactive oxygen species induce the association of SHP-1 with c-Src and the oxidation of both to enhance osteoclast survival. Am J Physiol Endocrinol Metab 307(1):E61–E70 31. Umeda S, Beamer WG, Takagi K, Naito M et al (1999) Deficiency of SHP-1 protein-tyrosine phosphatase activity results in heightened osteoclast function and decreased bone density. Am J Pathol 155(1):223–233 32. Bauler TJ, Kamiya N, Lapinski PE, Langewisch E et al (2011) Development of severe skeletal defects in induced SHP-2-deficient adult mice: a model of skeletal malformation in humans with SHP-2 mutations. Dis Model Mech 4(2):228–239 33. Zhou Y, Mohan A, Moore DC, Lin L et al (2015) SHP2 regulates osteoclastogenesis by promoting preosteoclast fusion. FASEB J 29(5):1635–1645 34. Chellaiah MA, Schaller MD (2009) Activation of Src kinase by protein-tyrosine phosphatasePEST in osteoclasts: comparative analysis of the effects of bisphosphonate and protein-tyrosine phosphatase inhibitor on Src activation in vitro. J Cell Physiol 220(2):382–393 35. Chellaiah MA, Kuppuswamy D, Lasky L, Linder S (2007) Phosphorylation of a WiscottAldrich syndrome protein-associated signal complex is critical in osteoclast bone resorption. J Biol Chem 282(13):10104–10116 36. Rhee I, Davidson D, Souza CM, Vacher J et al (2013) Macrophage fusion is controlled by the cytoplasmic protein tyrosine phosphatase PTP-PEST/PTPN12. Mol Cell Biol 33(12): 2458–2469 37. Valerio MS, Herbert BA, Griffin AC 3rd, Wan Z et al (2014) MKP-1 signaling events are required for early osteoclastogenesis in lineage

Osteoclast Methods in Protein Phosphatase Research defined progenitor populations by disrupting RANKL-induced NFATc1 nuclear translocation. Bone 60:16–25 38. Carlson J, Cui W, Zhang Q, Xu X et al (2009) Role of MKP-1 in osteoclasts and bone homeostasis. Am J Pathol 175(4):1564–1573 39. Finkelshtein E, Levy-Apter E, Elson A (2016) Production of osteoclasts for studying protein tyrosine phosphatase signaling. Methods Mol Biol 1447:283–300

79

40. Patntirapong S, Habibovic P, Hauschka PV (2009) Effects of soluble cobalt and cobalt incorporated into calcium phosphate layers on osteoclast differentiation and activation. Biomaterials 30(4):548–555 41. Maria SM, Prukner C, Sheikh Z, Mueller F et al (2014) Reproducible quantification of osteoclastic activity: characterization of a biomimetic calcium phosphate assay. J Biomed Mater Res B Appl Biomater 102(5):903–912

Chapter 5 OT-I TCR Transgenic Mice to Study the Role of PTPN22 in Anti-cancer Immunity Rebecca J. Brownlie, Rose Zamoyska, and Robert J. Salmond Abstract Phosphotyrosine phosphatase non-receptor type 22 (PTPN22) is a key regulator of immune cell activation and responses. Genetic polymorphisms of PTPN22 have been strongly linked with an increased risk of developing autoimmune diseases, while analysis of PTPN22-deficient mouse strains has determined that PTPN22 serves as a negative regulator of T cell antigen receptor signaling. As well as these key roles in maintaining immune tolerance, PTPN22 acts as an intracellular checkpoint for T cell responses to cancer, suggesting that PTPN22 might be a useful target to improve T cell immunotherapies. To assess the potential for targeting PTPN22, we have crossed Ptpn22-deficient mice to an OT-I TCR transgenic background and used adoptive T cell transfer approaches in mouse cancer models. We provide basic methods for the in vitro expansion of effector OT-I cytotoxic T lymphocytes, in vitro phenotypic analysis, and in vivo adoptive T cell transfer models to assess the role of PTPN22 in anti-cancer immunity. Key words PTPN22, T cell receptor, OT-I, Adoptive cell transfer, Immunotherapy, Ovarian carcinoma

1

Introduction Phosphotyrosine phosphatase non-receptor type 22 (PTPN22) is a cytoplasmic tyrosine phosphatase predominantly expressed in cells of hematopoietic origin. Interest in the cellular functions of PTPN22 has been driven by studies showing that single-nucleotide polymorphisms in PTPN22 are significant risk factors for the development of autoimmune diseases such as rheumatoid arthritis and type 1 diabetes (reviewed in [1–3]). Consistent with a central role for PTPN22 in immunity, over the past two decades, key functions for PTPN22 in regulating both adaptive and innate immune cell function have been defined. Studies using Ptpn22-deficient mouse strains have determined that PTPN22 is a negative regulator of T cell antigen receptor (TCR) signaling [4, 5] and, in particular, plays a key role in limiting responses to low-affinity self-antigens [6]. While this inhibitory function is likely important in preventing

Damien The´venin and Jo¨rg P. Mu¨ller (eds.), Protein Tyrosine Phosphatases: Methods and Protocols, Methods in Molecular Biology, vol. 2743, https://doi.org/10.1007/978-1-0716-3569-8_5, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

81

82

Rebecca J. Brownlie et al.

inappropriate T cell activation to self-antigens, we and others have recently shown that PTPN22 is also a negative regulator of protective T cell responses to cancer [7–11]. These data suggest that pharmacological inhibition or deletion of PTPN22 might represent a useful approach to enhancing the efficacy of cancer immunotherapies. CD8+ T cells are thought to be the major target of immunotherapies such as immune checkpoint inhibitors and play a key role in controlling malignancy by direct target cell killing and through the production of inflammatory cytokines. Our studies have determined that, in the absence of PTPN22 expression, these CD8+ T cell effector functions are enhanced. To model adoptive CD8+ T cell cancer therapies, we have taken advantage of the OT-I TCR transgenic mouse strain and syngeneic mouse cancer lines [7, 8]. OT-I CD8+ T cells express an H-2Kb-restricted TCR specific for a peptide derived from chicken ovalbumin (OVA257–264/SIINFEKL) [12]. A useful feature of this model lies in the capacity to assess the responses of the transgenic T cells to a range of well-defined variants of the SIINFEKL peptide that differ in affinity for the OT-I TCR [13]. When the OT-I strain is further crossed with knock-out, knock-in, or transgenic mouse strains, this enables researchers to test the impact of genetic modifications (e.g., PTPN22 deficiency) on both high- and low-affinity antigen-specific CD8+ T cell responses. This is particularly useful in modeling responses to tumor antigens, which may vary from high-affinity, immunogenic neoantigens to weakly stimulatory, low-affinity selfantigens. Here we describe basic protocols for in vitro expansion and differentiation of control and Ptpn22-/- OT-I effector cytotoxic T lymphocytes (CTLs). Furthermore, we describe approaches to assess CTL effector function in vitro and adoptive cell transfer approaches to assess anti-cancer functions of expanded T cell populations using the syngeneic ID8-OVA model of ovarian carcinoma. The protocols described here apply not only to the study of PTPN22 in anti-cancer immune responses but potentially to the analysis of any genetically modified mouse strain that has been crossed to the OT-I transgenic background.

2

Materials

2.1 CTL Activation and Differentiation

1. Rag1-/- OT-I or Rag1-/- Ptpn22-/- OT-I mice (see Note 1). 2. Centrifuge. 3. Cell culture incubator (37 °C/5% CO2). 4. Cell counter or hemocytometer. 5. 15- and 50-mL Falcon tubes

Mouse Models for the Study of PTPN22 Function in T Cells

83

6. 70-μm cell strainer 7. Plunger of a 5-mL syringe. 8. Six-well plates. 9. T25, T75, and T150 cell culture flasks. 10. Iscove’s Modified Eagle’s Medium, supplemented with 5% bovine serum albumin (BSA), 200 mM L-glutamine, penicillin/streptomycin, and 50 μM 2-mercaptoethanol. 11. OVA257–264 peptide (SIINFEKL). 12. Recombinant human interleukin-2 (see Note 2). 13. (Optional) Red blood cell lysis buffer: 0.15 M NH4Cl, 8.29 g/ L, 10 mM KHCO3, 1 g/L, EDTA 0.1 mM, 0.037 g/L in 1 L H2O. Sterile-filtered with a 0.22-μm filter and stored at 4 °C. 2.2 CTL Restimulation

1. Centrifuge. 2. Cell culture incubator (37 °C/5% CO2). 3. 15-mL or 50 mL Falcon tubes 4. 48-well or 24-well cell culture plates 5. Iscove’s Modified Eagle’s Media, supplemented with 5% BSA, 200 mM L-glutamine, penicillin/streptomycin, and 50 μM 2-mercaptoethanol. 6. Brefeldin A solution (stock solution 5 mg/mL in ethanol). 7. OVA257–264 peptide and variants (e.g., SIINFEKL, SIITFEKL, and SIIGFEKL).

2.3 Flow Cytometry Analysis

1. Refrigerated centrifuge. 2. Fluorescent-activated cell sorter. 3. Flow cytometry sample tubes. 4. Cell fixation and permeabilization reagents (e.g., eBioscience FoxP3 transcription factor staining buffer set). 5. Antibodies against T cell surface antigens and cytokines (e.g., CD8β, CD25, CD44, IFNγ, and TNF) coupled to fluorescent conjugates (e.g., PE-Cy7, PE, APC-Cy7, Alexa-Fluor-488, and PerCP Cy5.5). 6. Live/dead discrimination dye (e.g., Thermo Fisher Live/Dead Aqua fixable dye). 7. Phosphate buffered saline (PBS)-0.5% BSA.

2.4 ID8 Ovarian Cancer Cell Culture

1. Centrifuge. 2. Cell culture incubator (37 °C/5% CO2). 3. T75 and T150 cell culture flasks. 4. ID8-OVA-firefly luciferase (fLuc) cell line.

84

Rebecca J. Brownlie et al.

5. RPMI-1640 media, supplemented with 5% bovine serum albumin, 200 mM L-glutamine, and penicillin/streptomycin. 6. 0.05% Trypsin-EDTA solution 7. Sterile PBS. 2.5 In Vivo ID8 Cancer Model

1. Female C57BL/6 mice (7–10 weeks old). 2. 1-mL syringes 3. 27 or 29 gauge needles 4. Luciferin. 5. Small animal optical imaging system (e.g., IVIS imaging system).

3

Methods

3.1 In Vitro Activation and Expansion of OT-I T Cells

The key steps in Subheading 3.1 are illustrated in Fig. 1. 1. Dissect lymph nodes ± spleen from Rag1-/- OT-I or Rag1-/Ptpn22-/- OT-I mice (see Note 1), carefully avoiding excessive amounts of fat, and transfer to a 15-mL Falcon tube containing IMDM medium (see Note 3). 2. (Work is performed in rupt tissues using the disrupted tissues with placed on a 50-mL suspension.

a biosafety/tissue culture hood.) Displunger from a 5-mL syringe. Wash IMDM through a 40-μM cell filter Falcon tube to create a single-cell

3. Pellet cells by centrifugation at room temperature (400 g/ 5 min) and discard the supernatant. 4. Optional: Lyse red blood cells by resuspending cells in red blood cell lysis buffer for 2 min. Wash in fresh medium. 5. Resuspend cells in fresh IMDM and count using a hemocytometer or cell counter. 6. Adjust cell concentration to 1.5 × 106/mL for lymph node cells or 2 × 106/mL for spleen cells. 7. Culture cells in six-well plates, T25 or T75 flasks in the presence of 10-9 M OVA257–264 (SIINFEKL) peptide in a cell culture incubator at 37 °C/5% CO2. 8. Following 2 days of culture, transfer cell suspension to 15-mL Falcon tube(s). Add fresh IMDM to a final volume of 12 mL and pellet by centrifugation at room temperature (400 g/ 5 min). 9. Resuspend the pellet and repeat cell wash in fresh IMDM, to remove excess OVA257–264 peptide. Pellet by centrifugation as described above.

Mouse Models for the Study of PTPN22 Function in T Cells

85

Fig. 1 Overview of protocol steps for in vitro generation of effector OT-I CTLs. ACT, adoptive T cell transfer. (The figure was created on Biorender.com)

10. Resuspend cells in fresh IMDM and count using a hemocytometer or cell counter. Adjust cell concentration to 2 × 105/mL. 11. Culture OT-I T cells in T75 or T150 cell culture flasks, as appropriate, in the presence of 20 ng/mL recombinant human IL-2 in an incubator at 37 °C/5% CO2 (see Notes 2 and 4). 12. Following a further 2 days of culture, repeat steps 8–11. 13. By day 6 of culture, OT-I T cells will have differentiated to an effector CTL phenotype (see Note 5) and can be used for in vitro (see Subheading 3.2) and in vivo (see Subheading 3.5) functional analyses. 3.2 Re-stimulation of Effector CTLs

1. (Work is performed in a biosafety/tissue culture hood.) Transfer day 6 OT-I CTL cell suspensions to 15 or 50 mL Falcon tubes. Pellet by centrifugation at room temperature (400 g/ 5 min). 2. Resuspend cells in fresh IMDM and count using a hemocytometer or cell counter. Adjust cell concentration to 1 × 106/mL. 3. Add Brefeldin A (see Note 6) to cells at a final concentration of 2.5 μg/mL to prevent the release of cytokines.

86

Rebecca J. Brownlie et al.

4. Stimulate 5 × 105 OT-I CTL in 48-well or 24-well plates with a titration of OVA257–264 and variant peptides, typically ten-fold dilutions from 10-6 to 10-10 M, for 4 h in a cell culture incubator (37 °C/5% CO2) (see Note 7). For both wild-type and Ptpn22-/- cells, additional control samples should be left unstimulated (see Note 8). 3.3 Cell Staining and Flow Cytometry Analysis

1. Transfer 5 × 105 control and stimulated OT-I CTLs (see Subheading 3.2) to flow cytometry tubes (see Note 9) and wash in PBS-0.5% BSA. Pellet cells by centrifugation at 4 °C. 2. Prepare master-mix of fluorophore-conjugated antibodies for self-surface antigens and live/dead dye by diluting in chilled PBS-0.5% BSA. Prepare 100 μL of antibody master-mix per sample plus an additional 100–200 μL to ensure sufficient volume for all samples. Antibody dilutions are typically in the region of 1:100–1:1000 but should be titrated for each antibody/fluorophore. We use fixable Live/Dead Aqua dye from Thermo Fisher at a 1:500 dilution; other similar reagents can be used depending on the specific antibody panel and fluorophores used. 3. To enable compensation to be set on the flow cytometer or during data analysis, additional control samples should be stained as single-color controls for each antibody–fluorophore combination. For this purpose, resuspend cells in 100 μL of PBS-BSA containing one antibody conjugate or live/dead dye diluted to the same concentration as for the master-mix (see step 2). A further essential control is an unstained sample while using isotype control antibodies, and fluorescence minus one stained controls should be considered. 4. Stain experimental cell samples and controls in 100 μL/tube for 20 min at 4 °C in the dark. 5. Wash cells by adding 1–2 mL PBS per tube. Pellet samples by centrifugation at 4 °C (400 g/5 min). 6. Fix and permeabilize samples using the manufacturer’s instructions for a kit of choice. If using the eBioscience FoxP3 staining kit, fix and permeabilize samples in 50 μL of fix/perm buffer for 20 min at 4 °C, followed by washing in 1 mL chilled permeabilization buffer and pelleting by centrifugation at 4 ° C (400 g/5 min). 7. Prepare master-mix of fluorophore-conjugated antibodies for intracellular antigens (e.g., cytokines) by diluting in a chilled permeabilization buffer. Prepare 50 μL of antibody master-mix per sample plus an additional 100 μL to ensure sufficient volume for all samples. As described above, antibody dilutions should be titrated for each antibody/fluorophore, and singlecolor controls should be prepared (see step 3).

Mouse Models for the Study of PTPN22 Function in T Cells

87

8. Stain experimental cell samples and controls in 50 μL/tube for 30 min at 4 °C in the dark (see Note 10). 9. Wash cells by adding 1–2 mL permeabilization buffer per tube. Pellet samples by centrifugation at 4 °C (400 g/5 min). 10. Resuspend cells in 350 μL PBS-0.5% BSA. If clumps or debris are present, cell suspensions should be run through a 40-μm cell strainer prior to analysis to prevent blocking cytometer tubing. 11. Using single-color control and unstained samples, set compensation on a flow cytometer. Analyze the experimental cell suspensions and collect at least 104 live cell events per sample. 3.4 Culture of ID8 Ovarian Carcinoma Cells

1. (Work is performed in a biosafety/tissue culture hood.) Upon receipt of frozen vials of ID8-OVA-fLuc cells (see Note 11), defrost by quickly warming in the hands and transfer cell suspension to a 15 mL Falcon. 2. Wash cells by adding 10 mL pre-warmed (37 °C) complete RPMI-1640 (see Note 3). Pellet by centrifugation at room temperature (400 g/5 min). 3. Resuspend cells in 2 mL fresh RPMI-1640 and count using a hemocytometer or cell counter. Adjust cell concentration to 1 × 106/mL. 4. Transfer 5 × 105 or 1 × 106 cells to a T75 or T150 flask, respectively. Add RPMI-1640 to flasks in sufficient volume to cover the surface area of the flask bottom. 5. Growth of ID8 cells should be checked daily. Once adherent ID8 cells reach 70–80% confluency, remove RPMI-1640 using a vacuum pump or pipettor. Wash the remaining RPMI-1640 from the flask and remove non-adherent, dead cells by adding 10 mL pre-warmed sterile PBS and remove using a pipettor. 6. Detach adherent cells by adding 0.5 or 1 mL Trypsin-0.05% EDTA per T75 or T150 flask, respectively. Incubate for up to 5 min at 37 °C, checking halfway through to determine if cells have begun to detach. Add 10 mL RPMI-1640, mix thoroughly, use a pipettor to ensure that all cells are resuspended, and transfer to a 15-mL Falcon tube. 7. Pellet by centrifugation at room temperature (400 g/5 min). 8. If continued propagation of cells is required, repeat steps 3–7. If cells are required for in vivo experiments, resuspend them in sterile PBS and count using a hemocytometer or cell counter. Repeat the wash step using sterile PBS and pellet by centrifugation. 9. Prepare cells for injection by resuspending at 5 × 107/mL in sterile PBS and retain them on ice.

88

Rebecca J. Brownlie et al.

Fig. 2 Overview of key protocol steps for assessment of in vivo OT-I CTL anti-tumor activity using ID8-OVAfLuc ovarian carcinoma model. (The figure was created on Biorender.com) 3.5 In Vivo ID8 Cancer Model

The key steps in the in vivo ID8 cancer model are illustrated in Fig. 2. 1. In the animal facility, inject 100 μL of ID8 cell suspension intraperitoneally into recipient female C57BL/6 mice (5 × 106 cells/mouse) (see Note 12). 2. At least 7 days after ID8 cell injection, tumor burden should be assessed in mice by injecting luciferin intraperitoneally and performing intravital imaging under general anesthetic (see Note 13). Mice are assigned to experimental groups, ensuring that all groups have similar averages and ranges of tumor burden, as determined by analysis of luciferase activity, before treatment. 3. For the preparation of wild-type and Ptpn22-/- OT-I CTLs for adoptive T cell transfer, follow Subheading 3.1, steps 1–10. 4. Inject wild-type or Ptpn22-/- OT-I CTLs intravenously into recipient mice (see Note 14). A third control group should receive a mock adoptive cell transfer (PBS alone).

Mouse Models for the Study of PTPN22 Function in T Cells

89

5. The impact of OT-I cell adoptive cell transfer on tumor growth is assessed by repeated luciferin injection and intravital imaging of luciferase activity at multiple time points post-injection (see Note 15). 6. At experimental endpoints, mice are sacrificed. Ex vivo analysis (e.g., flow cytometric analysis) of transferred OT-I T cell population present in peritoneal washes can be performed if required (see Note 16).

4

Notes 1. Our control and Ptpn22-/- OT-I mouse lines have been backcrossed to a Rag1-/- background. This ensures that the development of B cells and non-TCR transgenic T cells is blocked. Consequently, OT-I T cells represent >90% of all cells in lymph nodes but are less pure in spleens due to the high proportions of myeloid cells. It is anticipated that the described cell culture and activation conditions will also be applicable or can be easily adapted to OT-I mouse strains on a Rag1-sufficient background. 2. The use of human IL-2 is recommended; however, recombinant mouse IL-2 is a suitable replacement. The concentrations of mouse IL-2 to be used in cell culture must be titrated by individual researchers. 3. The use of IMDM for the culture of OT-I T cells is recommended; however, alternative media preparations such as RPMI-1640 may serve as an alternative. ID8 cells will grow well in most common media preparations, including IMDM, RPMI-1640, and DMEM. Of note, irrespective of media choice, it is essential to include 2-mercaptoethanol in culture to sustain primary T cell proliferation, but not ID8 cells. 4. Control OT-I CTLs grown under the described culture conditions expand exponentially between days 2 and 6. Large cell clumps will be visible by eye during this time and are a normal feature of highly proliferative T cells in culture. Researchers should expect at least a ten-fold expansion between days 2 and 4. IMDM may “turn yellow” due to rapid OT-I cell proliferation, nutrient consumption, and lactate secretion; transfer cells promptly to fresh medium and IL-2 when this happens. When cell clumps begin to fall apart, this is a sign that nutrients and/or IL-2 are running low. Ideally, researchers should check OT-I cell cultures daily to ensure optimal growth. 5. The protocol given here generates highly inflammatory effector CTLs that have potent anti-tumor responses but are relatively short-lived in vivo [8]. To generate longer-lived, memory

90

Rebecca J. Brownlie et al.

phenotype CTLs, researchers should include soluble antiCD28 antibody (1 μg/mL) during days 0–2 of culture and replace human IL-2 with recombinant mouse IL-15 during days 4–6 of the culture period. 6. For intracellular cytokine staining, the use of Brefeldin A is recommended, but the use of other protein transport inhibitors may also be applicable (e.g., monensin/golgi-stop). 7. It is strongly recommended that, for CTL re-stimulation experiments, researchers test recall responses to several different ovalbumin peptides of varying affinity for the OT-I TCR (e.g., SIINFEKL (high affinity), SIITFEKL (intermediate affinity), and SIIGFEKL (low affinity) [12]) at several different concentrations. Our studies have shown that Ptpn22-/- OT-I CTLs have an enhanced capacity to secrete cytokines in responses to low-affinity but not high-affinity antigens, as compared with Ptpn22+/+ controls [6]. The use of several ovalbumin variant peptides, therefore, maximizes the opportunity to distinguish differences between wild-type and Ptpn22-/- OT-I CTL responses. 8. An alternative to intracellular cytokine staining is the analysis of secreted cytokines by supernatant ELISA or multiplex reagents. In this case, OT-I CTL cultures are set up as described without the addition of Brefeldin A, supernatant collected at 24 h and stored at –20 °C/–70 °C before ELISAs are performed. 9. The use of flow cytometry sample tubes has been suggested; however, all staining steps can also be performed in 96-well V-bottom or U-bottom plates, with a corresponding adjustment of antibody mix and washing buffer volumes. 10. Staining times of 20 min for surface markers and 30 min for intracellular markers have been found to be suitable and convenient for the basic cell surface marker panel and specific cytokines described here, but optimal staining times may vary for other markers. Of note, a recent methods paper has evaluated the efficacy of overnight staining and the use of lower antibody concentrations for flow cytometry analysis of intracellular proteins in particular [14]. The impact of reduced antibody concentrations and longer staining times should be determined for each antibody. 11. We have described culture conditions and the in vivo use of ID8 cells expressing ovalbumin and firefly luciferase (ID8-OVA-fLuc) generated in our lab [7]. Variants of this cell line that express ovalbumin variants of varying affinity for the OT-I TCR are available and are useful for testing OT-I T cell responses to lower affinity tumor-associated antigens

Mouse Models for the Study of PTPN22 Function in T Cells

91

(described in [7, 8]). All methods described here are applicable for using those ID8 variant cell lines. Investigators should allow approximately 7 days for the expansion of ID8 cells in vitro before commencing injections and in vivo tumor models. 12. To replicate the sex-specific hormonal environment, female C57BL/6 mice should be used exclusively as recipients for the in vivo ovarian cancer models. We recommend using 5 × 106 ID8 cells to establish reproducible tumors in mice, but researchers may wish to perform pilot studies with a titration of cancer cell numbers (e.g., 0.1–1 × 107). The number of mice required to detect reproducible statistically significant differences in tumor luminescence between experimental groups should be determined by using pilot studies and power calculations, e.g., using the NC3Rs Experimental Design Assistant (https://eda.nc3rs.org.uk/experimentaldesign-group) or similar. The researcher performing cell injections and intravital imaging should be “blinded” as to which group receives the various treatments. 13. When using C57BL/6 mice as ID8 tumor cell recipients, we recommend shaving the abdomen before and improving the quality of intravital imaging. 14. For efficient clearance/control of established ID8 tumors, the number of OT-I T cells transferred should be between 0.5 and 1 × 107 cells/mouse. 15. The ID8 tumor model is relatively slow growing, with mice typically surviving until ~day 100 post-tumor cell injection [15]. At late stages of the disease, mice will present with bloody, malignant ascites in the peritoneal cavity and should be humanely culled. The presence of ascites is likely to impact upon accurate detection of luciferase activity in the abdomen/ peritoneal cavity. Variants of the ID8 line in which tumor suppressor genes have been deleted (e.g., Tp53 and Brca2) have also been reported [16]. Deletion of tumor suppressor genes typically results in more aggressive and rapid in vivo tumor growth. 16. To enable identification and ex vivo analysis of OT-I T cells at experimental endpoints, the use of congenic mouse strains is recommended. Our control (Ptpn22+/+) OT-I mouse strain expresses the CD45.1 allelic variant, and therefore, T cells can be distinguished in adoptive cell transfer experiments based on CD45.1 (donor) and CD45.2 (host) expression. For the same reasons, in some experiments, we have generated and used CD45.1/CD45.2 heterozygous Ptpn22-/- OT-I T cells [7]. Alternative approaches include using C57BL/6-Ubc-

92

Rebecca J. Brownlie et al.

GFP transgenic mice, which have a ubiquitous expression of GFP, as hosts [7]. Without available transgenic or congenic mouse strains, the OT-I TCR can be identified using TCR Vα2 and Vβ5 antibodies, although a proportion of endogenous host T cells will express the same TCR chains. References 1. Bottini N, Peterson EJ (2014) Tyrosine phosphatase PTPN22: a multifunctional regulator of immune signaling, development, and disease. Annu Rev Immunol 32:83–119 2. Brownlie RJ, Zamoyska R, Salmond RJ (2018) Regulation of autoimmune and anti-tumor T-cell responses by PTPN22. Immunology 154:377–382 3. Mustelin T, Bottini N, Stanford SM (2019) The contribution of PTPN22 to rheumatic disease. Arthritis Rheumatol 71:486–495 4. Hasegawa K, Martin F, Huang G, Tumas D, Diehl D, Chan AC (2004) PEST domainenriched tyrosine phosphatase (PEP) regulation of effector/memory T cells. Science 303: 685–689 5. Brownlie RJ, Miosge LA, Vassilakos D, Svensson LM, Cope A, Zamoyska R (2012) Lack of the phosphatase PTPN22 increases adhesion of murine regulatory T cells to improve their immunosuppressive function. Sci Signal 5:ra87 6. Salmond RJ, Brownlie RJ, Morrison VL, Zamoyska R (2014) The tyrosine phosphatase PTPN22 discriminates weak self peptides from strong agonist TCR signals. Nat Immunol 15: 875–883 7. Brownlie RJ, Garcia C, Ravasz M, Zehn D, Salmond RJ, Zamoyska R (2017) Resistance to TGFβ suppression and improved antitumor responses in CD8+ T cells lacking PTPN22. Nat Commun 8:1343 8. Brownlie RJ, Wright D, Zamoyska R, Salmond RJ (2019) Deletion of PTPN22 improves effector and memory CD8+ T cell responses to tumors. JCI Insight 5:e127847 9. Cubas R, Khan Z, Gong Q, Moskalenko M, Xiong H, Ou Q, Pai C, Rodriguez R, Cheung J, Chan AC (2020) Autoimmunity linked protein phosphatase PTPN22 as a target for cancer immunotherapy. J Immunother Cancer 8:e001439

10. Ho WJ, Croessmann S, Lin J, Phyo ZH, Charmsaz S, Danilova L, Mohan AA, Gross NE, Chen F, Dong J, Aggarwal D, Bai Y, Wang J, He J, Leatheman JM, Yarchoan M, Armstrong TD, Zaidi N, Fertig EJ, Denny JC, Park BH, Zhang ZY, Jaffee EM (2021) Systemic inhibition of PTPN22 augments anticancer immunity. J Clin Invest 131:e146950 11. Orozco RC, Marquardt K, Mowen K, Sherman LA (2021) Proautoimmune allele of tyrosine phosphatase, PTPN22, enhances tumor immunity. J Immunol 207:1662–1671 12. Hogquist KA, Jameson SC, Heath WR, Howard JL, Mevan MJ, Carbone FR (1994) T cell receptor antagonist peptides induce positive selection. Cell 76:17–27 13. Stepanek P, Prabhakar AS, Osswald C, King CG, Bulek A, Naeher D, Beaufils-Hugot M, Abanto ML, Galati V, Hausman B, Lang R, Cole DK, Huseby ES, Sewell AK, Chakraborty AK, Palmer E (2014) Coreceptor scanning by the T cell receptor provides a mechanism for T cell tolerance. Cell 159:333–345 14. Whyte CE, Tumes DJ, Liston A, Burton OT (2022) Do more with less: improving high parameter cytometry through overnight staining. Curr Protoc 2:e589 15. Roby KF, Taylor CC, Sweetwood JP, Cheng Y, Pace JL, Tawfik O, Persons DL, Smith PG, Terranova PF (2000) Development of a syngeneic mouse model for events related to ovarian cancer. Carcinogenesis 21:585–591 16. Walton J, Blagih J, Ennis D, Leung E, Dowson S, Farquharson M, Tookman LA, Orange C, Athineos D, Mason S, Stevenson D, Blyth K, Strathdee D, Balkwill FR, Vousden K, Lockley M, McNeish IA (2016) CRISPR/Cas9-mediated Trp53 and Brca2 knockout to generate improved models of ovarian high-grade serous carcinoma. Cancer Res 76:6118–6129

Chapter 6 Protein Tyrosine Phosphatase Studies in Zebrafish Danie¨lle T. J. Woutersen, Jisca Majole´e, and Jeroen den Hertog Abstract The zebrafish is an ideal model for functional analysis of genes at the molecular, protein, cell, organ, and organism levels. We have used zebrafish to analyze the function of members of the protein tyrosine phosphatase (PTP) superfamily for more than two decades. The molecular genetic toolbox has significantly improved over the years. Currently, generating mutant lines that lack the function of a PTP gene is relatively straightforward by CRISPR/Cas9 technology-mediated generation of insertions or deletions in the target gene. In addition, generating point mutations using CRISPR/Cas9 technology and homology-directed repair (HDR) is feasible, albeit the success rate could be higher. Here, we describe the methods, including the tips and tricks, that we have used to generate knock-out and knock-in zebrafish lines in PTP genes successfully. Key words Protein tyrosine phosphatases, PTP, Zebrafish, PTEN, SHP2, CRISPR/Cas9, Genome editing

1

Introduction Intracellular processes are controlled by post-translational modifications (PTMs) of signaling proteins, enabling cells to act and respond to their environment. Phosphorylation and dephosphorylation of tyrosine residues in signaling proteins are orchestrated by the opposing actions of protein tyrosine kinases (PTKs) and protein tyrosine phosphatases (PTPs), which catalyzes the addition or removal of phosphate groups to or from tyrosine residues in proteins, respectively [1, 2]. Maintaining a balance between these phosphotyrosine regulators controls important processes, including proliferation, differentiation, and cell–cell adhesion. Aberrant intracellular signaling by altered PTP function has been linked to many human diseases, including developmental disorders, cardiovascular disease, and cancer [3, 4]. Over the years, the zebrafish (Danio rerio) has become an increasingly important vertebrate model for studying developmental biology and modeling human disease because of its many

Damien The´venin and Jo¨rg P. Mu¨ller (eds.), Protein Tyrosine Phosphatases: Methods and Protocols, Methods in Molecular Biology, vol. 2743, https://doi.org/10.1007/978-1-0716-3569-8_6, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

93

94

Danie¨lle T. J. Woutersen et al.

favorable characteristics [5]. For example, the zebrafish has emerged as an essential model for studying cardiovascular diseases. Although the human heart is morphologically distinct from the zebrafish heart, with a single circulatory system and only one atrium and ventricle, similarities are evident with respect to cardiac development, heart looping structure, and physiology [6]. The high degree of sequence conservation between human and zebrafish genes offers great potential for modeling human cardiovascular diseases [7]. Additionally, the external fertilization, rapid development, and embryonic translucency of zebrafish allow for a unique opportunity to visualize a broad range of biological processes by in vivo microscopy analysis of transgenic markers [8]. Since the development of the first transgenic zebrafish line containing a green fluorescent protein (GFP) sequence fused to the GATA-1 promotor sequence [9], numerous reporter lines have been generated that specifically stain a particular tissue or cell type (see www. zfin.org). An example of how fluorescent imaging techniques were instrumental for PTP studies in zebrafish was given by Stumpf et al. [10] by investigating the function of the tumor suppressor gene, phosphatase, and tensin homolog (PTEN). Confocal live imaging of homozygous ptena/ptenb double knock-out embryos in the transgenic Tg(kdrl:eGFP) background revealed hyperbranching of the intersegmental vessels in the absence of functional PTEN. Micro-injection of exogenous PTEN, but not PTEN phosphatase mutants, suppressed the hyperbranching phenotype emphasizing the requirement of PTEN phosphatase activity for normal embryonic angiogenesis [10, 11]. The ptena/ptenb double mutant described above was derived by target-selected gene inactivation (i.e., random mutagenesis), followed by resequencing of the gene of interest [12, 13]. Targetselected gene inactivation was effective but laborious, and since then, more directed technology has been developed to inactivate specific genes. Current genome editing strategies are based on nuclease systems, including zinc finger nucleases (ZFNs), transcription activator-like effector nucleases (TALENs), and CRISPR/ Cas9. All these technologies provoke a permanent mutagenic outcome. The CRISPR/Cas9-based method depends on an RNA– DNA interaction which provides multiple advantages over ZFNs and TALENs, including higher efficiency, more straightforward design for any genomic target, easier prediction of off-target sites, and the possibility to modify several genomic sites simultaneously [14, 15]. For these reasons, CRISPR/Cas9 technology is emerging as the superior technology to generate knock-out animal models. The advancements in CRISPR/Cas9 gene editing techniques in zebrafish include the generation of knock-in models, which enables functional analysis of proteins at the single amino acid level in embryonic development and disease. Generation of the first

Protein Tyrosine Phosphatase Studies in Zebrafish

95

CRISPR/Cas9-based zebrafish knock-in model was reported by Irion and colleagues, who applied homology-directed repair (HDR) of the albb4 mutation by co-injection of Cas9 mRNA, in vitro synthesized single-guide RNA (sgRNA), and a circular template vector containing 875-bp homology arms [16]. More recently, we and others generated knock-in zebrafish models using single-stranded oligonucleotide HDR templates [17– 20]. HDR templates diffuse faster into the nucleus, speeding up integration into the DNA, which is particularly important for zebrafish models because the blastomeres divide extremely fast, causing declined germline transmission [21, 22]. Although introducing specific mutations in the zebrafish genome has great potential for studying the functional effects of human disease-associated genetic variants, only a few studies have employed the CRISPR-Cas9 system to model human disease [23]. The first and foremost requirement when modeling human diseases is the availability of the gene in the model system. Homologs of all human PTPs have been identified in zebrafish except for PTPN7 and PTPN14 [24]. A search of the zebrafish genome (http://www.ENSEMBL.org) led to the identification of a small part of the ptpn12 gene, which may suggest either that the rest of this gene is present but has not been annotated properly yet, or that the rest of the homolog of PTPN12 is missing from the zebrafish genome altogether. In early vertebrate evolution, zebrafish and other teleost fish species have undergone a whole genome duplication [25, 26]. Since then, some, but not all, of the duplicated chromosomes have been lost from the zebrafish genome, yielding two paralogs of many genes, including genes encoding PTPs [24, 27]. Duplicated genes were identified for 14 PTPs providing an opportunity to study the sub-functionalization of the gene. Bonetti et al. demonstrated that duplication of the PTPN11 gene leads to zebrafish ptpn11a and ptpn11b encoding Shp2a and Shp2b. Whereas knock-out of the ptpn11a gene causes severe developmental defects and is embryonically lethal, the ptpn11b gene is dispensable throughout development. Yet, rescue experiments indicate functional redundancy for the two paralogs because injection of ptpn11a or ptpn11b rescues defects in ptpn11a mutants. Ptpn11a is expressed constitutively during development, whereas ptpn11b is expressed at very low levels during early development. Endogenous Shp2b levels are insufficient to compensate for the loss of Shp2a, but the expression of exogenous Shp2b does rescue the phenotype [28]. Loss of SHP2 function is lethal in mice and zebrafish. Elevated activity of SHP2 also induces developmental defects. Therefore, broadening our understanding of the SHP2 function is essential. SHP2 is dynamically involved in multiple human diseases, including Noonan syndrome (NS), Noonan Syndrome with multiple lentigines (NSML), juvenile myelomonocytic leukemia (JMML),

96

Danie¨lle T. J. Woutersen et al.

and metachondromatosis (MC) [17, 29]. NS is an autosomal dominant disorder characterized by over 100 gain-of-function (GOF) variants in PTPN11 and other RAS/MAPK pathway genes. Initially, we modeled NS transiently by micro-injection at the one-cell stage of synthetic mRNA encoding Shp2 variants with mutations identified in human patients [28, 30, 31]. Subsequently, our laboratory developed and characterized a novel genetic zebrafish model with the NS Shp2-D61G mutation. Micro-injection of the CRISPR/Cas9 complex and single-stranded oligonucleotide HDR templates at the one-cell stage led to the identification of founder fish and the establishment of a stable line. These Shp2 mutant zebrafish recapitulate multiple typical NS features, including short stature, craniofacial anomalies, and JMML-like myeloproliferative neoplasia [17]. This zebrafish line provides a valuable preclinical model for the development of future therapies. Considering the elegance by which the Shp2-D61G genetic model recapitulates human disease features, our laboratory is eager to explore CRISPR/Cas9 technology further to establish additional zebrafish lines, which will facilitate the analysis of genotype–phenotype correlations associated with NS. Accordingly, we have successfully generated multiple knock-out and knock-in models currently being characterized. In this chapter, we provide details of the method that we have used to generate stable genetic knock-out and knock-in zebrafish models.

2 2.1

Materials Fish Husbandry

1. Tanks (e.g., Aqua Schwarz GmbH). 2. Mesh nets (e.g., Aqua Schwarz GmbH). 3. Butterfly net. 4. Single boxes (e.g., Aqua Schwarz GmbH SD-881). 5. Single boxes nets (e.g., Aqua Schwarz GmbH 006-0620). 6. Scalpel (e.g., Swann-Morton No. 23). 7. Tweezers. 8. Ice bucket. 9. 10-cm Petri dishes 10. Transfer pipet (e.g., Sarstedt—Ref. 86.1171). 11. Tricaine methanesulfonate (MS-222) (0.1%). 12. E3 medium: 5 mM NaCl, 0.17 mM KCl, 0.33 mM CaCl2, and 0.33 mM MgSO4. 13. Methylene blue (0.01%).

Protein Tyrosine Phosphatase Studies in Zebrafish

2.2 Genotyping by polymerase chain reaction (PCR)

97

1. PCR strip. 2. 96-well plates 3. Taq DNA polymerase. 4. dNTPs (25 mM). 5. Proteinase K (20 mg/mL). 6. Sequencing stickers. 7. SZL lysis buffer: 10 mM Tris, pH 8.3, 50 mM KCl, 2.5 mM MgCl2, 0.005% NP-40, 0.005% Tween-20, and 0.01% Gelatine. 8. 5× Green GoTaq Flexi Reaction buffer (e.g., Promega— M8911) 9. Q5® Hot Start High-Fidelity DNA Polymerase, and 10× reaction buffer (e.g., Bioke—M0493S). 10. Gene-specific primers. 11. T7 forward primer: 5′-TAATACGACTCACTATAGGG-3′. 12. M13 reverse primer: 5′-CAGGAAACAGCTATGAC-3′. 13. LB agar plates + 50 μg/mL ampicillin. 14. Isopropyl β-D-1-thiogalactopyranoside (IPTG) (20 mg/mL). 15. 5-Bromo-4-chloro-3-indolyl-beta-D-galacto-pyranoside (X-gal) (20 mg/mL).

2.3

Injection

1. Magnetic stirrer. 2. Low protein binding tubes 0.5 mL (e.g., Eppendorf— 022431064). 3. Recombinant SpCas9 (5 μg/μL). 4. sgRNA (200 ng/μL), 5. HDR template (200 ng/μL). 6. PCR purification kit (e.g., Thermo Scientific GeneJET PCR Purification Kit—K0702). 7. Protein dilution buffer: 25 mM phosphate buffer, pH 8.0, 500 mM NaCl, 1.25 mM MgCl2, RNAse free H2O, and 2 mM beta-mercapto-ethanol. 8. KCl (1 M). 9. Phenol red (50%). 10. Ethanol (70%).

2.4 Gel Electrophoresis

1. Ultrapure agarose (e.g., Invitrogen—16500-500). 2. 10× Tris-borate-EDTA (TBE) buffer: 0.89 M Tris, pH 8.3, 0.89 M H3BO3, and 20 mM EDTA

98

Danie¨lle T. J. Woutersen et al.

3. Ethidium bromide (1%). 4. 50-bp DNA ladder (5 μg/mL) (e.g., Fisher Scientific— SM0372). 2.5

Equipment

1. Brightfield stereomicroscope with an ocular micrometer (e.g., Leica M165 FC). 2. Thermomixer. 3. Transilluminator. 4. Vortex. 5. pH meter. 6. Nanodrop (e.g., Thermo Scientific Nanodrop One). 7. Thermal cycler.

3

Methods

3.1 Preparation— Design of sgRNA

For CRISPR/Cas9-mediated knock-out and knock-in approaches, we use protocols based on the publication by Gagnon et al. about CRISPR/Cas9-mediated mutagenesis [32], as outlined in Fig. 1. Selecting a suitable sgRNA near the target site depends on the location of specific recognition motifs necessary for the binding of the Cas9 protein. Since we use the Cas9 nuclease, we have designed sgRNAs containing a protospacer adjacent motif (PAM) site. 1. Go to the online tool https://chopchop.cbu.uib.no/ to select a target site for CRISPR/Cas9-directed mutagenesis. 2. Fill in the requested boxes, and hit the button “Find Target Sites!” (see Note 1). 3. Zoom into your exon of interest by scrolling the mouse (see Note 2). 4. Select a guide sequence near your desired cut site with the highest possible ranking (see Notes 3–6). 5. Select the guide sequence by clicking the colored bar (see Note 7). 6. Order the target sequence in the 5′–3′ direction (see Notes 8 and 9). 7. Dissolve the sgRNA in MilliQ water to 100 ng/μL and store in aliquots at -80 °C.

3.2 Preparation— Preparation of HDR Template

A homology-directed repair (HDR) template is injected together with the sgRNA and Cas9 protein to induce specific mutations rather than knock out the gene of interest. 1. Design an HDR template to incorporate the mutation of choice. The HDR template comprises the specific mutation

Protein Tyrosine Phosphatase Studies in Zebrafish

99

Graphical summary of the generation of a knock-in zebra sh model 1. 5'

Preparation 2-4 weeks

3'

Design components

Optimize genotyping

Test Cas9 e ciency

2. Injections 1-2 weeks Make injection mixture

Perform injections

Raise injected embryos

Raising sh 12 weeks

3.

Screening 1

Set up family crosses

Nucleotide

10

4-8 weeks

Sequencing

Screen pairs

4.

1

Nucleotide

Ligation

Transformation

Injected founder

F1

F2

Wildtype

Wildtype

Wildtype

10

Veri cation 1-2 weeks

Sequencing

5. Deriving a stable line 26 weeks

Fig. 1 Graphical overview of the steps involved in CRISPR/Cas9-based generation of zebrafish knock-in and knock-out models. Step 1. Preparation of injection mixture components and optimization of the genotyping strategy. Step 2. Composition and administration of the injection mixture. Step 3. Screening of the potential founder fish. Step 4. Verification of the mutation of interest by subcloning of the mutant transcript in pBlueScript. Step 5. Obtaining a stable line by outcrossing. (Created using BioRender (biorender.com))

you aim to introduce, two or three silent mutations (i.e., mutations that do not alter the amino acid sequence), and a left and right homology arm of >10 bases (Fig. 2). Altogether this amounts to approximately 60 bases (see Note 10). 2. Order as a standard oligo (see Note 11).

100

Danie¨lle T. J. Woutersen et al.

sgRNA

Constant oligonucleotides

1. sgRNA / Cas9 complex recognizes PA P M site

2. Double stranded break is generated

3. Repair with HDR template

HDR template Silent mutations

Fig. 2 Overview of the CRISPR/Cas9 genome editing approach for generating knock-in models. The one-cell stage eggs of the zebrafish are injected with a mixture including sgRNA, Cas9 protein, HDR template, and KCL. A complex of the sequence-specific sgRNA and the Cas9 endonuclease will recognize the target sequence and create a double-strand break. Repairing in the presence of the HDR template containing silent mutations will produce a new sequence that lacks a PAM site and hence is not recognized by the sgRNA/Cas9 complex. (Created using BioRender (biorender.com))

3. Dissolve the oligo and perform an additional PCR purification step (see Note 12). 4. Measure the concentration on a spectrophotometer and adjust it to 200 ng/μL. 3.3 Preparation— Primer Testing

To screen for the desired mutation, it is a must to have a robust and reliable genotyping pipeline for each specific mutation. The primer pairs mentioned by the ChopChop online tool work very well in our hands. They are usually about 20 bases long and have a small region of interest to amplify. Before screening the primary injected zebrafish, we determine the most optimal PCR conditions for the primers on wild-type embryos and select a suitable sequencing primer. 1. Anesthetize wild-type embryos in MS-222 and transfer the embryos to a PCR strip. 2. Remove as much E3 medium as possible from the PCR strip. 3. Prepare the SZL lysis buffer by adding 5 μL proteinase K to 1 mL of buffer (final concentration of 100 μg/mL). 4. Add 50 μL SZL lysis buffer to the embryos in the PCR strip and ensure they are submerged within the liquid (see Note 13). 5. Incubate the embryos for 1 h at 60 °C followed by 15 min at 95 °C in a thermocycler.

Protein Tyrosine Phosphatase Studies in Zebrafish

101

Table 1 PCR program for gene-specific amplification Step

Temperature

Time

1. Denaturation

94 °C

1 min

2. Denaturation 3. Annealing 4. Extension

94 °C Tm 72 °C

20 s 30 s 30 s

5. Final extension

72 °C

3 min

6. Hold

12 °C

Infinite

34 cycles

6. Primers arrive as a powder; dissolve them to a stock concentration of 100 μM by adding MilliQ. 7. Prepare working solutions by diluting an aliquot of the primers to 10 μM in MilliQ. 8. Vortex the lysate and prepare a fresh PCR strip including 1 μL lysate and 3.7 μL Green GoTaq Flexi Reaction buffer (5×), 0.2 μL dNTPs (25 mM), 0.2 μL Taq DNA polymerase, 4.7 μL MilliQ, and 0.1 μL of each primer (10 μM) (see Note 14). 9. Incubate the lysates according to Table 1 (see Note 15). 10. Prepare a 4% 1× TBE agarose gel and load 5 μL PCR product and 2 μL 50 bp ladder (5 μg/mL) to confirm the size of the product (see Note 16). 11. Send the PCR product, including forward or reverse primer, to your sequencing company of choice according to their requirements (see Note 17). 3.4 Preparation— Cas9 Concentration Test

Before the knock-out and knock-in injections, we determine the optimal concentration of the recombinant SpCas9 protein. In our experience, a low concentration is enough to induce correct cutting. However, it is essential to determine this for each new injection mixture. 1. Recombinant SpCas9 protein was provided to us by the lab of Prof. Geijsen, LUMC Leiden (http://divvly.com). We obtained the protein as a solution of 12.5 ng/μL and stored it at -80 °C (see Notes 18 and 19). 2. Prepare the protein dilution buffer according to the instructions of the manufacturer. 3. Freshly add beta mercapto-ethanol to a final concentration of 2 mM to the protein dilution buffer. 4. Thaw the SpCas9 protein stock at 22 °C in a thermoshaker.

102

Danie¨lle T. J. Woutersen et al.

Table 2 Concentrations and working solutions for preparing the injection mixture Stock concentration Final concentration Volume in μL Cas9 protein

5 ng/μL

1.25 ng/μL

1.25

sgRNA

200 ng/μL

20 ng/μL

0.50

HDR template 200 ng/μL

40 ng/μL

1.00

KCl

1M

200 mM

1.00

Phenol Red

50%

12.5%

1.25

sgRNA single-guide RNA, HDR homology-directed repair

5. Prepare a SpCas9 protein dilution range by supplementing the SpCas9 protein with the protein dilution buffer (see Note 20). 6. Aliquot the diluted SpCas9 protein in low protein binding tubes (see Note 21). 7. Freeze the aliquots at -80 °C. 3.5 Injections— Preparation of Injection Mixture and Injections

One should be skilled in injection handling to successfully microinject in one-cell stage embryos. The procedure, as well as the preparation of the injection plate, is performed according to Hale et al. [33]. 1. Thaw and incubate a Cas9 aliquot in a thermoshaker at 22 °C. 2. Prepare the injection mixture according to Table 2 in low protein binding tubes. Add the Cas9 last and homogenize the mixture. 3. Incubate for 5 min at 37 °C. 4. Homogenize the injection mixture before insertion in the needle. 5. Inject injection mixture into one-cell stage zebrafish embryos as previously described by Hale et al. [33].

3.6

Screening

Following the injections, the embryos should be raised to adulthood. Survival of the embryos highly depends on the efficiency and/or toxicity of the sgRNA and HDR template. We noticed significantly higher lethality when the HDR template was co-injected compared to the injection of only Cas9 and sgRNA. Generation of a knock-in model by CRISPR/Cas9 and HDR results in mosaic adult zebrafish with genotypic variations from cell to cell. Not every cell harbors the desired mutation. We routinely select founders based on genotyping of F1 offspring embryos (Fig. 3). Once a founder has been identified, this mosaic founder will be outcrossed with a wild-type partner, and their offspring will be raised.

Protein Tyrosine Phosphatase Studies in Zebrafish

103

A Gene-specific PCR product of primary injected outcrossed zebrafish Predicted Cas9 cleavage site Intended base mutation for amino acid change Intended silent mutation to remove PA P M B Sequencing traces of PCR products in A

WT embryo

Embryo 1

Embryo 7 C Allele separation of embryo 7 in pBlueScript

WT allele

Mutant allele

Fig. 3 Generation of knock-in zebrafish line. (a) Gene-specific PCR product of primary injected knock-in embryos and non-injected wild-type siblings. (b) Sequencing traces of the various gene-specific PCR products including a wild type, a deletion, and a correct incorporation. (c) Sequencing traces of wild-type and mutant alleles of embryo 7. (Created using BioRender (biorender.com))

1. Once the injection has been performed, distribute 60–80 injected eggs in a fresh dish with E3 and incubate at 28.5 °C until 5 dpf (see Note 22). 2. Remove unfertilized eggs at 24 hpf and check regularly for dead embryos (see Note 23). 3. Dechorionate the embryos at 48 hpf and clean the Petri dish from eggshells and other debris (see Note 24). 4. Move all healthy embryos containing a swim bladder to the nursery at 5 dpf and allow them to grow up to adulthood according to the standards of your fish facility (see Notes 25–27). 5. Determine the ratio of males and females in the injected colony at 12 weeks post-injection (see Notes 28 and 29). 6. Prepare a family cross once a week to ensure that sexual maturity is reached (see Notes 30 and 31).

104

Danie¨lle T. J. Woutersen et al.

7. Prepare single crosses of one potential founder and one easily recognizable wild-type and separate them overnight (see Notes 32–35). 8. Pool the male and female fish in the morning to start mating (see Notes 36–38). 9. Keep the pairs that produced eggs together in a single box and collect the eggs, one Petri dish per pair. 10. Collect the eggs in a fresh dish with E3 medium and count the number of eggs. 11. Remove the unfertilized eggs at 24 hpf. 12. Select a batch of healthy-looking embryos for lysis at 24 hpf or later (see Notes 39 and 40). 13. Collect the selected embryos in a Petri dish and sedate the embryos in 0.1% MS-222 before lysis. 14. Transfer the embryos to a PCR strip or 96-well plate and remove as much E3 medium as possible. 15. Lyse the embryos and perform PCR, agarose gel electrophoresis, and sequencing according to the method optimized as in Subheading 3.3 (primer testing). 16. Check the sequences (see Note 41). 3.7 Verification of the Mutation by Subcloning

Despite previous optimization steps, the sequencing results from the lysate of whole embryos may not be entirely clear. Additionally, it is crucial to confirm that, in the offspring, one allele is unaffected (derived from the wild-type fish) and the other allele (derived from the mutant founder fish) only has the desired mutation(s) and not unintended ones, such as partial deletions or multiple integrations of the HDR sequence. Therefore, we routinely analyze the sequence of each allele individually by subcloning a PCR fragment of the genomic DNA isolated from the embryo containing the mutation in pBluescript. 1. Digest pBluescript with a blunt-cutting enzyme (see Note 42). 2. Perform a Q5 PCR reaction on the lysate of embryos with suspected correct mutations using your gene-specific primers. Combine 1 μL lysate with 10 μL Q5 buffer (5×), 1 μL dNTPs (25 mM), 1 μL Q5 high-fidelity DNA polymerase (2000 units/mL), 32 μL MilliQ, and 2.5 μL of each primer (10 mM) (see Note 43). 3. Run a PCR reaction according to Table 3. 4. Perform agarose gel electrophoresis using 1 μL of the PCR product to confirm the presence and size of a single band. 5. Purify the product with the GeneJET PCR purification kit and measure the DNA concentration.

Protein Tyrosine Phosphatase Studies in Zebrafish

105

Table 3 PCR program for Q5 high-fidelity amplification Step

Temperature

Time

1. Denaturation

98 °C

30 s

2. Denaturation 3. Annealing 4. Extension

98 °C Tm 72 °C

10 s 30 s 30 s

5. Final extension

72 °C

3 min

6. Hold

12 °C

Infinite

34 cycles

6. Ligate the product into the digested pBluescript II KS+ (see Note 44). 7. Transform bacteria with the ligated product and plate on LB Agar plates containing ampicillin, X-Gal, and IPTG. Incubate overnight at 37 °C. 8. Inoculate 12 white colonies (or as many as possible) into a 10 μL PCR reaction mix and perform a colony PCR using either gene- or plasmid-specific primers (see Notes 45 and 46). 9. Perform agarose gel electrophoresis to confirm the presence of the insert in the plasmid. 10. Send the PCR product with plasmid or gene-specific primer for sequencing. 11. Analyze the sequencing result: One bacterial colony should contain either the wild-type or mutant sequence. To confirm the exclusive incorporation of the desired mutations in the target gene, we routinely analyze the sequence of the entire transcript of the target gene. To this end, mRNA is isolated from 5 dpf mutant embryos, and cDNA is generated. The entire Open Reading Frame of the target gene is amplified by RT-PCR and cloned into pBluescript, using the protocol described above. The complete sequence of at least four clones is established to verify that the desired mutations were inserted in the target gene and that no additional mutations were introduced in the process. 3.8 Deriving a Stable Line

We outcross our genetically modified founder fish at least twice before phenotypic characterization to reduce potential off-target effects caused by the CRISPR-Cas9 gene-editing procedure. 1. Once a founder fish is identified with the correct mutation, a suitable adult wild-type of similar size and age is selected (see Note 47).

106

Danie¨lle T. J. Woutersen et al.

2. Outcross the founder fish with a wild-type partner and raise the embryos (see Note 48). 3. Genotype the F1 adults at 12 weeks post-fertilization. A small piece of tissue is amputated from the caudal fin, lysed in 100 μL SZL lysis buffer with proteinase K, and further processed using the optimized genotyping protocol described in Subheading 3.3—primer testing. 4. Outcross the heterozygous adults with a wild-type partner to generate the F2 generation. 5. To ensure that the mutation is not lost over generations, every new generation is genotyped as described in Subheading 3.3— primer testing.

4

Notes 1. If your gene of interest is sequenced and well annotated, type the gene’s name. It is also possible to manually enter the (genomic) sequence. 2. Above the exon of interest, several smaller bars are indicated, displaying the guide sequences. 3. Identified guide sequences are labeled with a number and color code indicating several parameters, including Cas9 efficiency score, number of off-target effects, self-complementary regions, GC content, and location within the gene (for further information, see https://chopchop.cbu.uib.no/instructions). 4. Some DNA regions provide many sgRNA options, while others contain a scarce amount of compatible sequence motifs. It is important to find a balance between the distance of the predicted Cas9 cleavage site to your mutation location and the ranking of the sgRNA. 5. The higher the number, the better the sgRNA works. However, proximity to the intended mutation site is, in this case, the most important. 6. To minimize off-target binding of the chosen sgRNA sequence, we run a search on the entire genomic DNA of the organism using the USCS genome browser BLAT option (https://genome.ucsc.edu/). 7. A new screen pops up, including the predicted cut site and multiple primer pairs to genotype the area of interest. 8. We use the Custom synthetic guide RNA option of Integrated DNA Technologies (IDT). 9. It is also an option to generate templates for sgRNA transcription [32].

Protein Tyrosine Phosphatase Studies in Zebrafish

107

10. Silent mutations should be included in the HDR template to prevent the binding of the sgRNA after the desired sequence is incorporated. If possible, we induce silent mutations in or near the PAM site to inhibit repetitive cleavage by Cas9. 11. We routinely use the Custom DNA oligo option of IDT. 12. We use the GeneJet purification kit. 13. For 4–5 dpf embryos, a short spin ensures that the embryos are not sticking to the plastic or floating above the liquid level. 14. When multiple reactions are performed, prepare a master mix. 15. We use the NEB Tm calculator (https://tmcalculator.neb. com/#!/main) to determine the annealing temperature of the PCR reaction. 16. Different Tm temperatures and a range of cycles should be used to optimize the PCR. 17. To optimize the sequencing, send both forward and reverse primers with variations of PCR product to determine the optimal primer and amount of PCR product. 18. The recombinant SpCas9 protein precipitates at 4 °C or when stored on ice, so make sure that the protein is kept at room temperature when in use or at -80 °C when stored. 19. A high-quality preparation of SpCas9 increases the success rate of DNA cleavage. 20. We prepared and tested protein dilutions of 1.25, 2.5, and 5.0 ng/μL. 21. According to the manufacturer, aliquoting in normal tubes results in significant protein loss. 22. To check the overall quality of the offspring, you should raise some uninjected embryos from the same batch. 23. Removing dead and unfertilized eggs is important for maintaining optimal growth conditions. 24. Some injection mixture might spill into the chorion egg during retrieval of the needle from the one-cell. This spilled injection mixture might result in “dirty eggs” with unwanted developmental defects. 25. Embryos lacking a swim bladder cannot swim to food and will not survive. To raise healthy, fertile adults, make sure only to select healthy embryos. 26. The survival ratio highly varies among the injected colonies. Some losses are inevitable and should be expected. 27. If the size of the embryos/ larvae within the colony is very different, it might be helpful to split the colony into two tanks, one with smaller larvae and one with larger larvae, to ensure that the smaller individuals will also get enough food to thrive.

108

Danie¨lle T. J. Woutersen et al.

28. Sometimes injected families are small, which may yield families with a biased gender ratio containing more females. 29. If the number of males in the colony is lower than that of females, add some similar age and size males to the colony to stimulate egg production. 30. We continue to perform family crosses until the family produces fertilized eggs for at least 2 consecutive weeks. 31. If the fish do not lay eggs in a mass crossing, they will likely not produce eggs in a single cross. 32. Start with the largest animals to give the smaller animals more time to grow. 33. Place the male below the net and the female above the net and put them together in the morning. Splitting the adults overnight will enhance the cue in the morning to spawn eggs. 34. When the potential founder has a dotted pigmentation pattern, one could use striped wild types and vice versa. Another option is using albino fish. 35. Ensure that the wild type is similar in age and size to the potential founder. 36. To induce egg spawning, we lift the net and create a slope or put shallow water in the net. 37. The female swims over the slope, and this might induce spawning. 38. If the female founder does not lay eggs, adding two males instead of one in a tank and one female might induce egg spawning. 39. We always select 12 embryos, but this number is arbitrary. More than 12 embryos may be analyzed if desired. However, we do not recommend analyzing fewer embryos. 40. Select embryos without a phenotype because they will likely survive until adulthood. 41. If all 12 embryos are wild type, the potential founder parent will be discarded. 42. We use pBluescript II KS+ and perform digestion with EcoRV. 43. Another blunt-ending polymerase may be used as well. 44. We generally perform the ligation using 20 ng pBluescript II KS+ with a vector:insert ratio of 1:9 and 1:18. 45. Use the same PCR mix and program as described in Subheading 3.3, replacing the 1 μL of lysate with MilliQ. 46. We generally use the T7 forward and M13 reverse primers since these are present in pBluescript.

Protein Tyrosine Phosphatase Studies in Zebrafish

109

47. In our experience, a weak phenotype is more easily visualized in the TL-wild-type background, whereas a very strong (lethal) phenotype may be less severe in the AB-wild-type background. 48. We usually find 1 or 2 embryos out of 12 with the correct mutation. Therefore, consider the number of F1 fish that you raise into adulthood. We typically grow up 120 embryos to adulthood. References 1. Tonks NK (2006) Protein tyrosine phosphatases: from genes, to function, to disease. Nat Rev Mol Cell Biol 7:833–846. https://doi. org/10.1038/NRM2039 2. Yu ZH, Zhang ZY (2018) Regulatory mechanisms and novel therapeutic targeting strategies for protein tyrosine phosphatases. Chem Rev 118:1069–1091. https://doi.org/10.1021/ A C S . C H E M R E V. 7 B 0 0 1 0 5 / A S S E T / IMAGES/MEDIUM/CR-2017-00105F_ 0014.GIF 3. Hendriks WJAJ, Elson A, Harroch S et al (2013) Protein tyrosine phosphatases in health and disease. FEBS J 280:708–730. https://doi. org/10.1111/FEBS.12000 4. Tautz L, Pellecchia M, Mustelin T (2006) Targeting the PTPome in human disease. Expert Opin Ther Targets 10:157–177. https://doi. org/10.1517/14728222.10.1.157 5. Choi TY, Choi TI, Lee YR et al (2021) Zebrafish as an animal model for biomedical research. Exp Mol Med 53(3):310–317. https://doi. org/10.1038/s12276-021-00571-5 6. Bowley G, Kugler E, Wilkinson R et al (2022) Zebrafish as a tractable model of human cardiovascular disease. Br J Pharmacol 179:900– 917. https://doi.org/10.1111/BPH.15473 7. Barbazuk WB, Korf I, Kadavi C et al (2000) The syntenic relationship of the zebrafish and human genomes. Genome Res 10:1351–1358. https://doi.org/10.1101/GR.144700 8. Høgset H, Horgan CC, Armstrong JPK et al (2020) In vivo biomolecular imaging of zebrafish embryos using confocal Raman spectroscopy. Nat Commun 11. https://doi.org/10. 1038/S41467-020-19827-1 9. Komori T, Yagi H, Nomura S et al (1997) Targeted disruption of Cbfa1 results in a complete lack of bone formation owing to maturational arrest of osteoblasts. Cell 89:755–764. https://doi.org/10.1016/S0092-8674(00) 80258-5 10. Stumpf M, Blokzijl-Franke S, Den Hertog J (2016) Fine-tuning of Pten localization and phosphatase activity is essential for zebrafish

angiogenesis. PLoS One 11:e0154771. h t t p s : //d o i . o r g / 10 . 1 3 7 1 / J O U R N A L . PONE.0154771 11. Stumpf M, Den Hertog J (2016) Differential requirement for Pten lipid and protein phosphatase activity during zebrafish embryonic development. PLoS One 11. https://doi.org/ 10.1371/JOURNAL.PONE.0148508 12. Faucherre A, Taylor GS, Overvoorde J et al (2008) Zebrafish pten genes have overlapping and non-redundant functions in tumorigenesis and embryonic development. Oncogene 27(8):1079–1086. https://doi.org/10.1038/ sj.onc.1210730 13. Wienholds E, Schulte-Merker S, Walderich B, Plasterk RHA (2002) Target-selected inactivation of the zebrafish rag1 gene. Science (1979) 297:99–102. https://doi.org/10.1126/SCI ENCE.1071762/SUPPL_FILE/ 1071762S.PDF 14. Gupta RM, Musunuru K (2014) Expanding the genetic editing tool kit: ZFNs, TALENs, and CRISPR-Cas9. J Clin Invest 124:4154– 4161. https://doi.org/10.1172/JCI72992 15. Sertori R, Trengove M, Basheer F et al (2016) Genome editing in zebrafish: a practical overview. Brief Funct Genomics 15:322–330. https://doi.org/10.1093/BFGP/ELV051 16. Irion U, Krauss J, Nu¨sslein-Volhard C (2014) Precise and efficient genome editing in zebrafish using the CRISPR/Cas9 system. Development 141:4827–4830. https://doi.org/10. 1242/DEV.115584 17. Solman M, Blokzijl-Franke S, Piques F et al (2022) Inflammatory response in hematopoietic stem and progenitor cells triggered by activating SHP2 mutations evokes blood defects. Elife 11. https://doi.org/10.7554/ELIFE. 73040 18. Tessadori F, Roessler HI, Savelberg SMC et al (2018) Effective CRISPR/Cas9-based nucleotide editing in zebrafish to model human genetic cardiovascular disorders. Dis Model Mech 11. https://doi.org/10.1242/DMM. 035469

110

Danie¨lle T. J. Woutersen et al.

19. Armstrong GAB, Liao M, You Z et al (2016) Homology directed knockin of point mutations in the zebrafish tardbp and fus genes in ALS using the CRISPR/Cas9 system. PLoS One 11:e0150188. https://doi.org/10. 1371/JOURNAL.PONE.0150188 20. Prykhozhij S V., Fuller C, Steele SL, et al (2018) Erratum: Optimized knock-in of point mutations in zebrafish using CRISPR/Cas9 (Nucleic Acids Res (gky512). https://doi. org/10.1093/nar/gky512). Nucleic Acids Res 46:9252. https://doi.org/10.1093/ NAR/GKY674 21. Olivier N, Luengo-Oroz MA, Duloquin L et al (2010) Cell lineage reconstruction of early zebrafish embryos using label-free nonlinear microscopy. Science (1979) 329:967–971. h t t p s : // d o i . o r g / 1 0 . 1 1 2 6 / S C I E N C E . 1189428/SUPPL_FILE/OLIVIER_ SOM.PDF 22. Kane DA, Kimmel CB (1993) The zebrafish midblastula transition. Development 119: 447–456. https://doi.org/10.1242/DEV. 119.2.447 23. de Vrieze E, de Bruijn SE, Reurink J et al (2021) Efficient generation of Knock-in zebrafish models for inherited disorders using CRISPR-Cas9 ribonucleoprotein complexes. Int J Mol Sci 22:9429. https://doi.org/10. 3390/IJMS22179429 24. van Eekelen M, Overvoorde J, van Rooijen C, den Hertog J (2010) Identification and expression of the family of classical protein-tyrosine phosphatases in zebrafish. PLoS One 5: e12573. https://doi.org/10.1371/JOUR NAL.PONE.0012573 25. Wittbrodt J, Meyer A, Schartl M (1998) More genes in fish? https://doi.org/10.1002/ (SICI)1521-1878(199806)20:6 26. Taylor JS, Van de Peer Y, Braasch I, Meyer A (2001) Comparative genomics provides evidence for an ancient genome duplication event in fish. Philos Trans R Soc Lond Ser B

Biol Sci 356:1661–1679. https://doi.org/10. 1098/RSTB.2001.0975 27. Jatllon O, Aury JM, Brunet F et al (2004) Genome duplication in the teleost fish Tetraodon nigroviridis reveals the early vertebrate proto-karyotype. Nature 431(7011): 9 4 6 – 9 5 7 . h t t p s : // d o i . o r g / 1 0 . 1 0 3 8 / nature03025 28. Bonetti M, Rodriguez-Martinez V, Overman JP et al (2014) Distinct and overlapping functions of ptpn11 genes in zebrafish development. PLoS One 9:e94884. https://doi.org/ 10.1371/JOURNAL.PONE.0094884 29. Solman M, Woutersen DTJ, den Hertog J (2022) Modeling (not so) rare developmental disorders associated with mutations in the protein-tyrosine phosphatase SHP2. Front Cell Dev Biol 0:2210. https://doi.org/10. 3389/FCELL.2022.1046415 30. Jopling C, Van Geemen D, Den Hertog J (2007) Shp2 knockdown and Noonan/LEOPARD mutant Shp2–induced gastrulation defects. PLoS Genet 3:e225. https://doi.org/ 10.1371/JOURNAL.PGEN.0030225 31. Paardekooper Overman J, Yi J-S, Bonetti M et al (2014) PZR coordinates Shp2 Noonan and LEOPARD syndrome signaling in zebrafish and mice. Mol Cell Biol 34:2874–2889. https://doi.org/10.1128/MCB.00135-14/ ASSET/B54 9E 4F1- 4E1 B-4 C3 4-8 69 7DB76E04EE77B/ASSETS/GRAPHIC/ ZMB9991005210014.JPEG 32. Gagnon JA, Valen E, Thyme SB et al (2014) Efficient mutagenesis by Cas9 proteinmediated oligonucleotide insertion and largescale assessment of single-guide RNAs. PLoS One 9:e98186. https://doi.org/10.1371/ JOURNAL.PONE.0098186 33. Hale AJ, den Hertog J (2016) Studying protein-tyrosine phosphatases in zebrafish. Methods Mol Biol 1447:351–372. https:// doi.org/10.1007/978-1-4939-3746-2_19/ COVER

Chapter 7 Examining Phosphatases Through Immunofluorescent Microscopy Caroline N. Smith and Jessica S. Blackburn Abstract Immunofluorescent microscopy enables the examination of cellular expression and localization of proteins. Cellular localization can often impact protein function, as certain molecular interactions occur in specific cellular compartments. Here we describe in detail the processes necessary for identifying phosphatases in the cell through immunofluorescent microscopy. Identification of phosphatase expression and localization could lead to the discovery of protein function in disease states along with potential substrates and binding partners. Key words Immunofluorescence, Microscopy, Confocal, Antibody, Protein expression, Localization

1

Introduction Fluorescence microscopy has continued to be an ever-evolving and powerful tool that allows researchers to examine the inner workings of the cell at the subcellular level. Over time, this technique has enabled the study of diverse cellular processes, including but not limited to protein localization and trafficking, protein–protein interactions, cell motility, ion transport, and metabolism [1]. Many of the first instances utilizing fluorescent microscopy in the mid-1900s focused on fluorescently labeled antibodies and fluorescent protein tags, such as GFP [2]. While fluorescent microscopy has continued to evolve new ways to permit cell imaging, fixed-cell analysis utilizing antibodies remains one of the most commonly used techniques in the scientific literature. In this chapter, we describe how to examine any phosphatase of interest through fluorescent microscopy, including the detection of endogenous protein or by exogenous expression of the phosphatase of interest. We provide two approaches based on the number of

Damien The´venin and Jo¨rg P. Mu¨ller (eds.), Protein Tyrosine Phosphatases: Methods and Protocols, Methods in Molecular Biology, vol. 2743, https://doi.org/10.1007/978-1-0716-3569-8_7, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

111

112

Caroline N. Smith and Jessica S. Blackburn

samples to be analyzed. For small-scale experiments involving the examination of a single phosphatase with a small number of antibodies, follow the coverslip protocol. If utilizing a variety of antibodies, cDNA expression constructs, or examining a variety of phosphatases, we recommend following the 96-well glass bottom plate protocol. The plate format will allow for the analysis of many samples while avoiding the need to mount each sample to a microscopy slide. Many similar approaches have been designed for assessing protein fluorescence on coverslips [3–5], and here we expand upon these methods and examine high-throughput approaches to examine phosphatase expression and localization through microscopy. Additionally, this chapter details how to best analyze images and prepare figures once staining and microscopy procedures are complete.

2

Materials

2.1 Plating Cells of Interest

1. 96-Well Black Glass Bottomed Imaging Plates with Sterile Lid (e.g., PorVair from Thomas Scientific Cat No. 1167W30) for high-throughput screening. 2. 18 × 18-1 Microscope Cover Glass (e.g., Fisher Scientific Cat. No. 12542A) for smaller-scale experiments. 3. Six-well cell culture dish (e.g., Thermo Fisher Scientific Cat. No. 130184). 4. Cells of interest. In this protocol, we used HCT116 cells (e.g., Sigma-Aldrich Cat. No. 91091005) which have large cytoplasm, are easily transfected, and attach to plates well. 5. DMEM high glucose media (e.g., Life Technologies Cat. No. 11965118) supplemented with 10% FBS for proper HCT116 growth. Store at 4 °C. 6. Dulbecco’s PBS (e.g., Caisson Labs Cat. No. PBL01) to wash cells. PBS can be made in-house but must be sterilized before use to reduce the risk of contamination of cells. Store at room temperature. 7. >95% Ethanol for sterilizing coverslips (e.g., Fisher Scientific Cat. No. BP2818100).

2.2 Plasmid Transfection (If Necessary)

1. Transfection reagent (e.g., Lipofectamine 3000, Life Technologies Cat. No. L3000015) to transfect plasmid expressing phosphatase of interest. Store at 4 °C. 2. Opti-MEM (e.g., Life Technologies Cat. No. 31985070). Store at 4 °C.

Examining Phosphatases Through Immunofluorescent Microscopy

2.3 Preparation of Fixed Cells

113

1. Dulbecco’s PBS (e.g., Caisson Labs Cat. No. PBL01) to wash cells. PBS can be made in-house but must be sterilized before use to reduce the risk of contamination of cells. Store at room temperature. 2. 4% paraformaldehyde in PBS (e.g., VWR Cat. No. AAJ61899AK) to fix cells. Store at 4 °C. 3. Triton X-100 (e.g., Sigma-Aldrich Cat. No. X100-100). Store at room temperature. 4. 0.1% Triton 100X solution in PBS. Store at room temperature. 5. 2% Bovine serum albumin (BSA) solution in PBS. Store at 4 °C. 6. Primary antibody (dependent on the phosphatase of interest). 7. Secondary antibody (dependent on the primary antibody). 8. Hoechst 33342 (e.g., Thermo Fisher Cat. No. H3570). Store at 4 °C. 9. Charged microscope slides (if using coverslips) with a permanent positive charge, frosted on one end, 25 × 75 mM, 1–1.2 mM (e.g., MTC Bio Cat. No. M7100-90, or Thomas Scientific Cat. No. 1144T49). 10. Antifade mounting media (e.g., VECTASHILED, NOVUS Biologicals Cat. No. H-1000-NB).

3

Methods Carry out all methods at room temperature unless otherwise specified.

3.1 Plating Cells of Interest in a 96-Well Dish

3.2 Plating Cells of Interest Using Coverslips

1. Split cells of interest into black 96-well glass bottom plate to be 40–50% confluent the next day (see Notes 1 and 2). 2. Incubate at the proper temperature and CO2 levels overnight, allowing cells to settle on the plate. HCT116 cell conditions are 5% CO2 at 37 °C (see Note 3). 1. Sterilize coverslips in six-well cell culture dishes in a tissue culture hood by washing the coverslips with 500 μL of >95% ethanol, followed by 3 × 1 mL washes with PBS. 2. Split cells of interest into the experimental dish(es) with the sterilized coverslips to be 40–50% confluent the next day. 3. Incubate at the proper temperature and CO2 levels overnight, allowing cells to settle on the plate (see Note 3).

114

Caroline N. Smith and Jessica S. Blackburn

3.3 Transfection with Phosphatase of Interest Plasmid

1. If it is necessary to exogenously express the phosphatase of interest, follow your general lab protocol, scaling it down to a 6-well or 96-well plate. The recommended protocol for a 1× reaction with examples is detailed in the Subheading 4 (see Notes 4 and 5). 2. Leave transfection mixture on cells for at least 24 h, up to 72 h in optimal growth conditions. Leaving cells longer than 72 h could cause cells to overgrow in the 96-well plate, making imaging difficult.

3.4 Fixation and Permeabilization of Cells

For all of the following sections, any details regarding the use of a 96-well glass bottom plate will be labeled with “96:” and details regarding the use of coverslips will be labeled using “CV:”. 1. Aspirate media from all experimental wells with a vacuum trap and aspirating pipet, connected to a 200-μL pipet tip. When aspirating liquid from wells, place the aspirating tip in the corner of each well so as not to disturb cells, before and after fixation. This will allow for quick aspiration without disturbing the cells of interest. 2. Wash all wells once with (96:100 μL; CV: 1 mL) of PBS (see Note 6). When washing wells in a 96-well plate, we recommend using a multichannel pipet and leaning tips against the side of the well when dispensing liquid to not disturb cells and limit lifting from the plate. Utilize the same technique in a six-well dish with either a P1000 or serological pipet, dispensing liquid slowly along the side of the dish. After 1 min of incubation, aspirate the wash in the same way as Subheading 3.3, step 1 (see Note 7). 3. Fix cells to the glass bottom plate with (96: 50 μL; CV: 500 μL) of 4% paraformaldehyde for 15 min. Administer the paraformaldehyde with a multi-channel or P1000 pipet, pipetting along the side of the well, to ensure proper incubation time without disturbing the cells. 4. Remove 4% paraformaldehyde from wells with the multichannel or P1000 pipet and dispense in a container that can be moved to the proper waste bin in accordance with laboratory chemical waste management plans. 5. Wash all experimental wells three times with (96:100 μL; CV: 1 mL) of PBS in the same fashion as Subheading 3.3, step 2. 6. Permeabilize the cells with (96: 50 μL; CV: 500 μL) of 0.1% Triton 100X for 10 min. 7. Remove 0.1% Triton 100X from cells by aspiration, as described. 8. Wash all wells three times with (96: 100 μL; CV: 1 mL) of PBS.

Examining Phosphatases Through Immunofluorescent Microscopy

115

At this point, continue with immunofluorescence staining or store the plate, with the lid in place, at 4 °C for up to a week. If storing a 96-well glass bottom plate, leave wells in 150 μL of PBS to account for potential evaporation. If using coverslips in a six-well dish, place the entire dish into a humidified chamber and store at 4 °C overnight or continue with the staining protocol. A humidified chamber is used to limit evaporation and stop samples from drying out. An example of a humidified chamber is a pipette box with deionized water-soaked paper towels placed in the bottom of the box. 3.5 Immunofluorescence Labeling of Phosphatase of Interest

If using coverslips in this section, perform all steps in the humidified chamber to limit antibody evaporation. 1. Aspirate PBS from all wells. 2. Add (96: 50 μL; CV: 120 μL directly to coverslip) of blocking solution (2% BSA) for at least 30 min at room temperature (see Note 8). 3. Dilute the primary antibody in 2% BSA while the blocking solution is incubating. The concentration of primary antibody used is antibody-dependent (see Note 9). Prepare (96: 50 μL; CV: 120 μL) of antibody dilution per well/coverslip to be treated. 4. Aspirate the blocking solution from the wells/coverslips. 5. Add (96: 50 μL; CV: 120 μL) of primary antibody dilution per well/coverslip. For any control wells/coverslips that are not receiving any primary antibody, add (96: 50 μL; CV: 120 μL) of 2% BSA. 6. Incubate for 1 h at room temperature with the lid on the plate (see Note 10). 7. After 1 h, aspirate liquid from step 4 from all wells/coverslips. 8. Wash all wells/coverslips with (96: 100 μL; CV: 200 μL) of PBS three times, aspirating between each wash step. 9. Leave experimental wells/coverslips in their final wash step while preparing the secondary antibody cocktail. The secondary antibody used in this step will depend on the species the primary antibody was raised in and the lasers that are available on the microscope (see Note 11). 10. Dilute the secondary antibody in 2% BSA. 11. Add 1:1000 Hoechst stain to the diluted secondary antibody to stain all nuclei. Prepare enough secondary antibody/ Hoechst cocktail to plate (96: 50 μL per well; CV: 120 μL per coverslip), plus an extra 50 μL of the solution to account for pipetting error.

116

Caroline N. Smith and Jessica S. Blackburn

12. For the buffer-only and primary-only control wells, prepare a Hoechst-only solution that is 1:1000 Hoechst in 2% BSA and plate (96: 50 μL per well; CV: 120 μL per coverslip). 13. Aspirate the final wash step from all wells/coverslips. 14. Add (96: 50 μL; CV: 120 μL) of desired antibody cocktail or Hoechst-only control to wells. 15. Incubate for 1 h at room temperature with the lid on the plate (see Note 12). Place the plate or humidified chamber in a dark place or wrap it in aluminum foil to limit secondary antibody exposure to the light. 16. Aspirate liquid from step 12 from all wells/coverslips. 17. Wash all experimental wells with (96: 100 μL; CV: 200 μL) of PBS five times, aspirating between each wash step. At this point, if using a glass bottom plate, continue with immunofluorescence imaging or store the plate, with the lid secured, at 4 °C overnight. If storing at 4 °C, leave wells in 150 μL of PBS to account for potential evaporation. If using coverslips, they must be mounted to microscopy slides. To do so, place a drop of VECTASHIELD mounting media on the center of the slide, and place the coverslip face down on the droplet. Seal the coverslip with two to three dots of black or clear nail polish and wait for 2 min. Then, cover all of the edges of the coverslip with nail polish. Let the coverslip dry, covered at room temperature for at least 30 min. Store the slide in a microscope box at -20 °C. 3.6 Confocal Imaging with Nikon A1R Microscope

1. After turning on the microscope, Open NIS-Elements. 2. Press “Escape Z’ on the computer or microscope prior to mounting the 96-well plate (see Note 13) or slide so the objective is moved back and will not be scratched while placing the sample on the microscope. 3. Set the magnification based on experimental parameters. Examples in this chapter use the 40× water objective. The Nyquist XY digital zoom function can be used for high magnification without needing an oil objective. 4. In the optical configuration (OC) panel, click “DAPI” and then the X-cite controller in the EYES configuration. This allows scanning of the sample by identifying Hoechst nuclear staining and places the sample in focus. 5. Once the sample is in focus, click “Confocal Ready” on the OC panel. This will allow scanning of the sample in four channels, or the channels that the microscope offers. These include DAPI (405 nM), FITC (488 nM), TRITC (561 nM), and FarRed (640 nM). In the control panel, turn off any of the individual laser lines that are not being used to speed up image scanning (see Note 13).

Examining Phosphatases Through Immunofluorescent Microscopy

117

6. Select image scanning speed and pinhole size (see Note 14). 7. To begin imaging, press the SCAN button; a live image window will appear showing the image scanning progress, and once the image is finished, press the SCAN button again to stop (see Note 15). 8. After scanning, save the image as a .ND2 file to store image metadata. Then export as a .tiff file, saving each image as a separate channel at 8 bit for bit depth. This will allow for analysis of each individual channel in Adobe Photoshop prior to merging. 9. Following the imaging of one well, switch back to the EYES configuration and use the joystick to move to the next well and focus on your next sample. Repeat for all samples. Save all files to an external flash drive to continue analysis with Adobe Photoshop and Illustrator on a separate computer. 3.7 Image Processing and Analysis

1. Open all .tiff channel files for a single image in Adobe Photoshop 2022 or the latest version of Adobe Photoshop. 2. The DAPI layer will act as the “Background” layer. For the FITC channel image, select Command+A to select the entire image, Command+C to copy, and then Command+V on the DAPI file to add the FITC channel as “Layer 1” on top of the DAPI “Background” channel. Repeat for any other channels. 3. For each layer, choose Image > Adjustments > Levels in order to amplify the fluorescent signal while reducing noise. Using the Auto tool can aid in the adjustment. Repeat for each layer separately. 4. After auto-adjusting the levels for each channel, draw a square with the Rectangular Marquee tool (e.g., 9.5 × 9.5 inches) on the area of the image to be used for the final figure. Copy each layer in Grayscale to Adobe Illustrator for side-by-side grayscale images of each channel. 5. To prepare a pseudo-colored merged image, first, change the image mode. Select Image > Mode > RGB Color. Do not merge the image at this point. 6. Double-click each individual Layer to open up the “Layer Styles” window. First, select the “Blend Mode” drop-down menu and select Screen, with 100% opacity. This will allow for the overlay of each channel. 7. Second, select the blending channels to include to pseudocolor each channel. The following are recommended: Hoechst: B only; FITC: G only; TRITC: R only; FarRed: R + B only.

118

Caroline N. Smith and Jessica S. Blackburn

8. To merge the final image, select Layer > Flatten Image. Now, when copying the 9.5 × 9.5 inch square, all layers will be combined. Before saving the final image, Command+Z to undo image flattening. This will allow opening of the combined image later to change the pseudo-color selection or select other regions of the image for figure construction.

4

Notes 1. For a 96-well plate using HCT116 cells, 40–50% confluency is seeding 5000 cells per well in 100 μL of media per well. 2. The best way to plate cells for this experiment is to calculate the number of cells needed and the volume of media needed for the number of wells being plated, plus an extra five wells to account for pipetting error. For example, to determine the required number of cells: 5000 cells per well × (100 wells + 5 extra wells) = 525,000 cells. To determine the volume with which to dilute cells: 100 μL media per well × (100 wells + 5 extra wells) = 10,500 μL media. Dilute cell stock to 525,000 cells in 10.5 mL of media, and then plate 100 μL of cells + media per well. 3. If transfection efficiency is low following the above protocol, either increase the amount of Lipofectamine 3000 being used and/or the amount of DNA being transfected. 4. These are the ratios of reagents we recommend for the Thermo Fisher Lipofectamine 3000 transfection kit for a six-well dish using coverslips. Tube 1: 125 μL of Opti-MEM and 5.62 μL of Lipofectamine 3000; Tube 2: 125 μL of Opti-MEM, 5 μL of P3000, and 2 μg of phosphatase vector. After mixing reagents and waiting for 15 min, plate 260 μL of reaction mixture per well. Scale up these values based on the number of reactions. Do not change the cell media prior to transfection, as it could disturb the settled cell population. 5. These are the ratios of reagents we recommend for the Thermo Fisher Lipofectamine 3000 transfection kit for a 96-well dish. Tube 1: 5 μL of Opti-MEM and 0.22 μL of Lipofectamine 3000; Tube 2: 5 μL of Opti-MEM, 0.2 μL of P3000, and 100 ng of phosphatase vector. Scale up these values based on the number of reactions. After mixing reagents and waiting for 15 min, plate 10.4 μL of reaction mixture per well. Do not change the cell media prior to transfection, as it could disturb the settled cell population.

Examining Phosphatases Through Immunofluorescent Microscopy

119

6. There are many PBS washes in this protocol. Be sure to use a cell line that does not lift from the plate easily with multiple PBS washes. If a cell line that lifts easily is necessary, do not wash cells prior to fixation. Instead, aspirate media and immediately add 4% paraformaldehyde. Then, wash only once following fixation prior to permeabilization. This should limit the number of cells that are lost prior to cell fixation. 7. If cells are lifting from the plate or coverslip during wash steps, do not wash cells prior to fixation. Instead, aspirate media and immediately add 4% paraformaldehyde. Then, wash only once following fixation prior to permeabilization. This should limit the number of cells that are lost prior to cell fixation. If the cells are still lifting, reduce vacuum suction during aspiration and eliminate all wash cells prior to permeabilization. 8. Make 2% BSA in PBS fresh each time this experiment is run to reduce background signal during staining. 9. Our studies focus on the phosphatase PRL-3 and utilize an anti-PRL-3 alpaca-derived nanobody made in-house. If using a nanobody in studies, dilute the nanobody to 1 mg/mL and then dilute that stock solution 1:100, both steps in 2% BSA. If using a commercially available antibody, follow the manufacturer’s instructions for dilution concentrations. 10. If a high background signal is detected following staining, start by changing the primary antibody incubation from 1 h at room temperature to overnight at 4 °C. 11. Recommended secondary antibodies and concentrations for staining are shown in Table 1. 12. If changing the incubation for the primary antibody does not reduce the background signal, reduce the concentration of the secondary antibody being used. 13. Prior to confocal microscopy, be sure to identify a clamp that will hold the 96-well plate, we used a plate stage incubator [6], and all Nikon A1R clamps are described by Dailey et al. [6]. We used a Nikon A1R confocal microscope, and all of the experimental steps described in Subheading 3.5 relate to this microscope. We recommend the following power and gain (HV) for each laser: 405 nM: Gain 150, Power 6.00; 488 nM: Gain 30, Power 2.00; 561 nM: Gain 60, Power 2.00; 640 nM: Gain 60, Power 2.00.

120

Caroline N. Smith and Jessica S. Blackburn

Table 1 Secondary antibodies successfully used in immunofluorescence microscopy with recommended dilutions

Application

Antibody

Supplier

Catalog number

Concentration for IF

Nanobody detection (TRITC laser)

Alexa Fluor® 594 AffiniPure Goat Anti-Alpaca IgG, VHH domain

Jackson ImmunoResearch

128-585232

1:400 in 2% BSA

Anti-rabbit detection (FITC laser)

Goat anti-Rabbit IgG (H + L) Cross-Adsorbed Secondary Antibody, Alexa Fluor™ 488

Invitrogen

A11008

1:500 in 2% BSA

Anti-rabbit detection (TRITC laser)

Goat anti-Rabbit IgG (H + L) Cross-Adsorbed Secondary Antibody, Alexa Fluor™ 555

Invitrogen

A21428

1:500 in 2% BSA

Anti-mouse detection (FarRed laser)

Anti-rabbit IgG (H + L), F(ab’) 2 Fragment (Alexa Fluor® 647 Conjugate)

Cell Signaling Technology

4414S

1:500 in 2% BSA

Anti-mouse detection (FITC laser)

Goat anti-Mouse IgG (H + L) Cross-Adsorbed Secondary Antibody, Alexa Fluor™ 488

Invitrogen

A11001

1:500 in 2% BSA

Anti-mouse detection (FarRed laser)

Goat anti-Mouse IgG (H + L) Cross-Adsorbed Secondary Antibody, Alexa Fluor™ 647

Invitrogen

A21235

1:500 in 2% BSA

14. Selecting the speed and pinhole of images will impact image clarity. Representative images shown in Fig. 1 were taken at a speed of 2048 in fast mode with a pinhole of 0.8. Follow the pinhole size recommended by the microscope at the time of use. 15. Long exposure to each laser during image scanning can increase the chance of photobleaching; therefore, stop scanning as soon as the entire image has been processed.

Examining Phosphatases Through Immunofluorescent Microscopy

121

Fig. 1 Phosphatase detection and localization assessed by nanobodies. Immunofluorescence of HCT116 cells transfected with CMV-PRL-3. Cells were stained with an anti-PRL-3 nanobody (1 mg/mL diluted 1:100 in 2% BSA) followed by an anti-alpaca VHH coupled to Alexa594 secondary antibody (1:400 in 2% BSA) for visualization. Top panel: nanobody-only control, stained with anti-PRL-3 nanobody and 1:1000 Hoechst; middle panel: secondary-only control, stained with anti-alpaca VHH antibody and 1:1000 Hoechst; bottom panel: experimental well stained with anti-PRL-3 nanobody, anti-alpaca VHH antibody, and 1:1000 Hoechst. Primary-only and secondary-only controls should show no staining in the experimental (red)

122

Caroline N. Smith and Jessica S. Blackburn

References 1. Combs CA (2010) Fluorescence microscopy: a concise guide to current imaging methods. Curr Protoc Neurosci Chapter 2:Unit2.1. https:// doi.org/10.1002/0471142301.ns0201s50 2. Hickey SM, Ung B, Bader C, Brooks R, Lazniewska J, Johnson IRD, Sorvina A, Logan J, Martini C, Moore CR, Karageorgos L, Sweetman MJ, Brooks DA (2021) Fluorescence microscopy-an outline of hardware, biological handling, and fluorophore considerations. Cell 11(1). https://doi.org/10. 3390/cells11010035 3. Fischer AH, Jacobson KA, Rose J, Zeller R (2008) Preparation of slides and coverslips for microscopy. CSH Protoc 2008:pdb.prot4988. https://doi.org/10.1101/pdb.prot4988

4. Bhattacharyya D, Hammond AT, Glick BS (2010) High-quality immunofluorescence of cultured cells. Methods Mol Biol 619:403– 410. https://doi.org/10.1007/978-1-60327412-8_24 5. Im K, Mareninov S, Diaz MFP, Yong WH (2019) An introduction to performing immunofluorescence staining. Methods Mol Biol 1897: 299–311. https://doi.org/10.1007/978-14939-8935-5_26 6. Dailey ME, Focht DC, Khodjakov A, Rieder CL, Spring KR, Claxton NS, Olenych SG, Griffin JD, Davidson MW (2014) Maintaining live cells on the microscope stage. MicroscopyU. https:// www.microscopyu.com/applications/live-cellimaging/maintaining-live-cells-on-the-micro scope-stage.

Chapter 8 Identification of Protein Tyrosine Phosphatase (PTP) Substrates Sravan Perla, Bin Qiu, Sam Dorry, Jae-Sung Yi, and Anton M. Bennett Abstract Protein tyrosine phosphorylation and dephosphorylation are key regulatory mechanisms in eukaryotes. Protein tyrosine phosphorylation and dephosphorylation are catalyzed by protein tyrosine kinases (PTKs) and protein tyrosine phosphatases (PTPs), respectively. The combinatorial action of both PTKs and PTPs is essential for properly maintaining cellular functions. In this unit, we discuss different novel methods to identify PTP substrates. PTPs depend on specific invariant residues that enable binding to tyrosinephosphorylated substrates and aid catalytic activity. Identifying PTP substrates has paved the way to understanding their role in distinct intracellular signaling pathways. Due to their high specific activity, the interaction between PTPs and their substrates is transient; therefore, identifying the physiological substrates of PTPs has been challenging. To identify the physiological substrates of PTPs, various PTP mutants have been generated. These PTP mutants, named “substrate-trapping mutants,” lack catalytic activity but bind tightly to their tyrosine-phosphorylated substrates. Identifying the substrates for the PTPs will provide critical insight into the function of physiological and pathophysiological signal transduction. In this chapter, we describe interaction assays used to identify the PTP substrates. Key words Protein tyrosine phosphatases, Substrate-trapping mutants, Substrate identification, Interacting proteins

1

Introduction Tyrosine phosphorylation is a post-translational modification of proteins that regulates catalytic activity, protein subcellular localization, and interactions with other proteins. Tyrosine phosphorylation is involved in a breadth of cellular functions, including gene expression, cell differentiation, migration, and survival [1]. Furthermore, dysregulation of tyrosine phosphorylation has been linked to the pathogenesis of several diseases, such as cancer, metabolic dysfunction, and immunological diseases [2]. Tyrosine phosphorylation is mediated by tyrosine kinases that catalyze the transfer of a phosphate group from adenosine triphosphate (ATP) to the hydroxyl group on a target tyrosine residue. In

Damien The´venin and Jo¨rg P. Mu¨ller (eds.), Protein Tyrosine Phosphatases: Methods and Protocols, Methods in Molecular Biology, vol. 2743, https://doi.org/10.1007/978-1-0716-3569-8_8, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

123

124

Sravan Perla et al.

contrast, the reverse reaction is catalyzed by protein tyrosine phosphatases (PTPs) [3, 4]. The classical PTPs are divided into two subgroups: 21 receptor protein tyrosine phosphatases (RPTPs) and 16 non-receptor protein tyrosine phosphatases (NRPTPs) [5]. All classical PTPs have a signature catalytic PTP domain containing a phosphate-binding loop (p-loop) with a conserved C(X)5R motif, as well as a spatially adjacent WPD loop, named after its invariant tryptophan-proline-aspartate sequence [6]. Both the p-loop and the WPD loop are essential for catalysis. RPTPs have a variable extracellular domain, a single transmembrane domain, and a cytoplasmic domain containing at least one PTP domain [6]. The extracellular domain drives dimerization upon ligand binding, often inhibiting the catalytic activity of the cytoplasmic domain. Like RPTPs, NRPTPs contain a PTP domain but lack extracellular and transmembrane domains. Instead, they have variable non-catalytic domains that coordinate conformational changes, protein binding, and phosphatase activity. The phosphatase activity of both RPTPs and NRPTPs is initiated by the conserved C(X)5R motif in their p-loop of the PTP domains. The cysteine residue of the motif performs a nucleophilic attack on the phosphoryl group of a target pTyr substrate, forming a covalent phosphothionate intermediate, which is stabilized by the proximal arginine residue [6]. Then the conserved aspartate residue in the WPD loop mediates the hydrolysis of the phosphothionate, releasing free phosphate [6]. 1.1 PTP Substrate Identification by In Vitro Substrate Trapping with PTP Active Site Mutants

Given the critical role of PTPs in regulating several cell signaling pathways, there is a great demand to identify novel PTP substrates. However, the transient interaction between PTPs and their substrates during dephosphorylation can make this challenging. Substrate trapping mutants (STMs) are PTPs with mutations in key catalytic residues that impair their phosphatase activity but retain their Km for substrates, thus allowing for more stable interactions between PTP and substrate, providing an opportunity for the proteins to be isolated (Fig. 1). A common STM is the Cys to Ser mutant, which has a cysteine to serine mutation in the C(X)5R motif of the p-loop. This mutation still allows the substrate to bind the PTP active site but prevents the cysteine-driven nucleophilic attack of the target pTyr, dephosphorylation, and substrate release [7]. Another STM is the D-A mutant, which replaces the conserved aspartate residue of the WPD loop with an alanine [8]. This mutant can form the phosphothionate intermediate in the first step of catalysis, but the absence of aspartate prevents its hydrolysis and substrate release (see Note 1). STMs can form stable complexes with their substrates, which allows the proteins to be isolated using common techniques such as co-immunoprecipitation and subsequently identified by western blotting or mass spectrometry.

Identification of Protein Tyrosine Phosphatase (PTP) Substrates

125

Fig. 1 (a) Wild-type PTPs transiently interact with their substrates to dephosphorylate target pTyr residues. (b) PTP substrate-trapping mutants bind phosphorylated substrate, forming a stable complex that can be isolated and identified

2 Materials Prepare all reagents using ultrapure water (prepared by purifying deionized water to attain a sensitivity of 18 MΩ-cm at 25 °C) and analytical grade reagents. Prepare and store all solutions at room temperature (unless mentioned otherwise). Strictly follow all waste disposal regulation protocols when disposing of any waste materials. 1. GST, GST-STM-PTP, and GST-WT-PTP cloned in a bacterial expression vector. 2. WT and STM-PTP (for interaction assay in cellular context). 3. Luria broth (LB) medium. 4. Isopropyl β-D-1-thiogalactopyranoside (IPTG). 5. Phosphate-buffered saline (PBS). 6. Protease inhibitor cocktail (sodium orthovanadate, phenylmethylsulfonyl fluoride, sodium fluoride, benzonase, pepstatin A, aprotinin, and leupeptin). 7. Triton X-100. 8. Pierce™ bicinchoninic acid (BCA) protein assay kit. 9. Glutathione agarose beads. 10. Dithiothreitol (DTT). 11. 2× loading buffer: 120 mM Tris (pH 6.8), 4% SDS, 20% glycerol, 10% β-mercaptoethanol, and 0.01% bromophenol blue.

126

Sravan Perla et al.

12. Coomassie brilliant blue (CBB) staining solution: 0.05% CBB R-250, 50% methanol, and 10% acetic acid. 13. CBB destaining solution: 5% methanol and 7% acetic acid. 14. Media and reagents for culturing mammalian cells. 15. Pervanadate (PV) solution preparation: Dilute 30% H2O2 with 20 mM HEPES to get 0.3% H2O2 solution. Mix 10 μL of 100 mM sodium orthovanadate gently and 940 μL of 0.3% H2O2 solution to get 1 mM PV stock for 5 min. Add 1–10 μL of catalase (10 mg/mL) and mix into the solution. Release the air pressure generated by O2 bubbles by opening the lid. This PV stock must be used within a few hours. 16. Lysis buffer: 150 mM NaCl, 1.5 mM MgCl2, 1 mM EDTA, 50 mM HEPES, 1% Triton X-100, and 10% glycerol, protease inhibitor cocktail (add protease inhibitor cocktail just before use and use within 1 day) (see Note 2). 17. Wash buffer: 150 mM NaCl, 10% glycerol, 50 mM HEPES, 0.1% Triton X-100, and protease inhibitors (add protease inhibitor cocktail just before use and use within 1 day) (see Note 2). 18. Elution buffer: 5 mM glutathione made in 50 mM Tris–HCl (pH 9.0). 19. Anti-phosphotyrosine antibodies, antibodies to GFP, control IgG, anti-p-Tyr antibodies, or other required primary and secondary antibodies. 20. Reagents for SDS-PAGE and immunoblotting. 21. Enhanced chemiluminescence kit. 22. Nitrocellulose or PVDF membranes for immunoblot. 23. Phosphatase inhibitors (iodoacetamide/iodoacetic acid), NaF, and Na3VO4. 24. Phosphate-buffered saline (PBS): 1.85 mM NaH2 PO4, 137 mM NaCl, 8.4 mM NaHPO4, and 2.7 mM KCl, pH 7.4. 25. DNA construct of full-length substrate trapping PTP and fulllength wild-type PTP. 26. Transfection reagents (Lipofectamine, calcium phosphate, polyethylenimine, and FuGENE HD). 27. Protein A/G Plus Agarose Beads or GFP agarose conjugated beads (see Note 3). 28. Immunoprecipitation (IP) buffer: 20 mM Tris pH 7.4, 1% Triton X-100; 5 mM EDTA, 0.1% BSA; 150 mM NaCl; 1 mM PMSF, 1 mM Na Orthovanadate or 1 mM IAA, and EDTA-free protease inhibitor (see Note 4). 29. Ponceau S solution: 30% TCA, 10× stock containing 2% Ponceau S and 30% sulfosalicylic acid in H2O.

Identification of Protein Tyrosine Phosphatase (PTP) Substrates

127

30. Destaining solution: 10% glacial acetic acid and 40% methanol. 31. Staining solution: 0.1% CBB R-250, 50% methanol, and 10% glacial acetic acid. 32. 1× SDS sample buffer: 2% SDS, 60 mM Tris pH 6.8, 10% glycerol, 5% β-mercaptoethanol, and 0.005% bromophenol blue. 33. Transfer buffer: 0.0375% SDS, 39 mM glycine, 48 mM Tris– HCl, and 20% methanol. 34. Deprobing buffer: 2% SDS, 62.5 mM Tris pH 6.8 and 100 mM β-mercaptoethanol in H2O. 35. TBST: 150 mM NaCl, 0.1% Tween, 10 mM Tris, pH 8.0. 36. Blocking solution: 2% BSA in TBST.

3

Method Carry out all experimental procedures at room temperature unless otherwise specified. In Vitro Interaction Assay In this approach, GST-tagged substrate-trapping PTP mutant is expressed using a bacterial expression system and is incubated with cell lysates treated with PV, which inhibits PTP activity, thereby stabilizing the levels of cellular protein tyrosine phosphorylation levels in the cells. The tyrosinephosphorylated substrates can be detected by western blotting using anti-phosphotyrosine antibodies or identified by mass spectroscopy (MS) [9, 10] (Fig. 2).

3.1 Prepare the GSTTagged Proteins

1. Grow the Escherichia coli expressing GST, GST-STM-PTP, or GST-WT-PTP overnight. 2. The cultures are diluted 1:20 with LB medium and incubated until an OD of ~0.4 at 600 nM is reached. 3. The IPTG at a final concentration of 1 mM is added to the growing cultures for 3–4 h at 37 °C (see Note 5). 4. The bacteria pellet is acquired by centrifuge at 10,000 × g for 15 min. 5. The pellet is suspended with 1 mL of cold PBS containing protease inhibitors and sonicated with 5 s bursts on ice at 30 s intervals. 6. The 10 μL of Triton X-100 (final concentration at 1%) is added and incubated on ice for 20 min. 7. Centrifuge at 10,000 × g for 20 min and transfer the supernatant into new tubes. 8. The protein concentration is determined by a BCA method.

128

Sravan Perla et al.

Fig. 2 Flow chart of in vitro substrate-trapping experiment

9. The 2 μg of GST-tagged proteins (GST, GST-STM-PTP, or GST-WT-PTP) is mixed with 20 μL (50% slurry) of glutathione agarose beads and incubated with slow tumbling for 1 h at 4 ° C. 10. Centrifuge at 1000 × g for 2 min to collect the beads conjugated with GST-tagged proteins. 11. Wash the beads with cold PBS and centrifuge at 1000 × g for 2 min. Repeat this twice. 12. The beads conjugated with GST-tagged proteins (GST, GST-STM-PTP, or GST-WT-PTP) can be stored in PBS containing protease inhibitors, 1 mM DTT, and 10% glycerol at 4 ° C. 13. Mix 5 μL of beads with the same volume of 2× loading buffer, and boil the samples at 100 °C for 5 min. Protein expression is confirmed by SDS-PAGE, followed by CBB staining. 3.2 Prepare the Cell Lysates

1. The target cells plated in a 100-mm dish with 70% confluent are treated with 1 mM PV for 30 min. 2. After being washed with cold PBS, the cells are lysed by adding 1 mL lysis buffer on ice for 30 min with 15 s vortex every 10 min.

Identification of Protein Tyrosine Phosphatase (PTP) Substrates

129

3. Centrifuge at 14,000 × g for 20 min at 4 °C and transfer the supernatant into new tubes. 4. The protein concentration is determined by the BCA method. 5. Add a final concentration at 1 mM of DTT and incubate on ice for 10 min to inactivate phosphatase inhibitors (see Note 6). 3.3 Combination of GST-Tagged Proteins and Their Substrates

1. The 500 μg of supernatants are added to the above three tubes containing GST-tagged proteins (GST, GST-STM-PTP, or GST-WT-PTP) and incubated with slow tumbling for 4 h at 4 °C. 2. The bead-conjugated complexes are washed four times with 1 mL of wash buffer by centrifugation at 1000 × g for 1 min. 3. The complexes containing GST-tagged proteins (GST, GST-STM-PTP, or GST-WT-PTP) and their substrates are eluted from beads by shaking with 50 μL of elution buffer for 2 min (see Notes 7 and 8). 4. Centrifuge at 1000 × g for 2 min to collect the eluted proteins present in the supernatant with buffer containing 50% glycerol and 1 mM DTT.

3.4 Identification of PTP Substrates

1. The eluted tyrosine-phosphorylated substrates are confirmed by SDS-PAGE followed by CBB staining and western blot detected by anti-phosphotyrosine antibodies (see Notes 9–12). 2. The tyrosine-phosphorylated substrates which bind to PTP in the pure protein can be identified by MS (see Note 13).

3.5 Substrate Trapping in the Cellular Context

3.5.1

Method

Even though in vitro substrate trapping is an efficient method to identify novel PTP substrates, sometimes it can produce falsepositive results because compartmentalization and cell integrity are compromised before the PTP and substrate interaction occurs. Therefore, to identify physiologically relevant PTP substrates, we need to utilize substrate trapping in cells. Substrate trapping in cells involves overexpressing the tagged STM-PTP, followed by IP, and then identifying the associated proteins with antibodies against the PTP or its tag. By using a full-length PTP, one can establish a physiologically relevant context between the PTP and its substrates. Expression of STM-PTP will prevent the dephosphorylation of its substrates. Therefore, the proteins in the cells need not necessarily be artificially phosphorylated by PV treatment or expression of a specific kinase [10]. 1. Transfect mammalian cells (70–80% confluent) of interest plated in 60-mm Petri dishes with WT or STM-PTP (see Note 14).

130

Sravan Perla et al.

2. After 24–48 h of treatment with or without agonist/PV, lyse the cells in 600 μL of IP buffer. 3. Vortex the cell lysates and keep them on ice for 10–15 min before centrifugation at 15,000 rpm for 15 min at 4 °C. 4. Collect the supernatant in a new Eppendorf tube, estimate protein concentration, and save a small amount (1–2%) boiled in sample buffer to be used later as whole-cell lysate. 5. Transfer (2–6 mg) of cell lysate equally to two tubes and incubate with an equal amount (2 μg) of the tag antibody or IgG control at 4 °C in a cold room for 4–17 h. 6. Add 20 μL of protein A/G plus agarose beads and incubate for 30 min to 1 h. 7. Wash the agarose beads three times with 1 mL of wash buffer. 8. Centrifuge at 2000 rpm for 1 min at 4 °C and remove the supernatant. 9. Recover the precipitated sample by lysing in SDS sample buffer and subject to SDS-PAGE and immunoblotting with anti p-Tyr antibodies (see Note 15). 10. The blots are stripped before incubation with other antibodies. The efficiency of IP is examined by blotting with anti-tag or anti-PTP antibodies and comparing levels between immune complexes and whole cell lysates (see Notes 16–18). 11. The lysates can be precleared by incubating with 20 μL of protein A-sepharose for 30 min at 4 °C to avoid too much background. Sometimes WT enzymes alter total tyrosine phosphorylation levels on many cellular proteins. When examining specific target protein dephosphorylation, it is important to note that the WT enzyme does not significantly alter total tyrosine phosphorylation levels on many cellular proteins by examining whole cell lysate by immunoblotting or by indirect immunofluorescence [11]. In these circumstances, overexpression can be regulated to prevent non-specific dephosphorylation of target proteins. The GFP-STM trap method utilizes GFP fusion proteins to identify interacting partners. The pull-down achieved by this method is quick and efficient. This method does not require the use of any anti-GFP antibodies. Therefore, this method has the advantage over other methods because the light and heavy chains of IgG antibodies do not appear in the pull-down. Camel-derived antibodies that bind with high efficiency to GFP are coupled to agarose beads and can be used as a trap. Based on a similar principle, the GST-Trap method is also available for the pull-down of GST-FPs [10]. The cell lysate is incubated with 15–20 μL of equilibrated agarose beads for 1 h at 4 °C and recovered by centrifugation and washing with buffer. The complexes are dissociated

Identification of Protein Tyrosine Phosphatase (PTP) Substrates

131

Fig. 3 Flow chart of in vivo substrate-trapping experiment in cells

from beads by incubating for 8–10 min at 95 °C in 60 μL of buffer without protease inhibitors. The beads are precipitated and removed by centrifugation. Collect the supernatant and use this supernatant for MS analysis and SDS-PAGE. The methods mentioned here depend on the identification of the co-purified proteins. Occasionally, this can be done by estimating the molecular weight of the phosphoprotein, understanding the signaling events in which the enzyme is involved, and subcellular localization of the protein and tyrosine-phosphorylated substrates and validating these by using specific antibodies. Unbiased approaches depend on the identification of the interacting partners using a mass spectrometry [12, 13] (Fig. 3).

4 Notes 1. Be sure to use a mutant PTP with a confirmed loss of phosphatase activity. The catalytic cysteine residue is typically replaced with an alanine residue, but other non-reactive residues can also be used. 2. The protease inhibitor cocktail should be added to lysis buffer and wash buffer just before use and use within 1 day.

132

Sravan Perla et al.

3. The protein A/G beads should be pre-washed and resuspended in the lysis buffer before use to ensure efficient binding of the immune complexes. 4. The PV solution should be prepared and used within a few hours. 5. In case protein forms inclusion bodies or induction is poor, cells could be induced overnight at 28 °C with 0.1 mM IPTG. 6. DTT will prevent the catalytic cysteine from getting oxidized and disable the PTP from interacting with its substrates. 7. Care should be taken when eluting the immune complexes from the beads to avoid the loss of the trapped substrates. 8. The substrate mixture should be prepared fresh and include a phosphatase inhibitor cocktail to prevent dephosphorylation of trapped substrates. 9. Air bubbles between gel and filter paper are to be avoided. 10. Longer transfer time is critical for high-molecular-weight proteins. 11. Biochemical assays or cellular studies should be performed to validate the functions and regulatory roles of the identified substrates. 12. All the immunoblotting steps followed for the detection of phospho antibodies require 2% BSA as a blocking buffer with phosphatase inhibitors (1 mM Na3VO4 and 10 mM NaF). 13. When IP samples are being prepared for MS analysis, replace Triton in the IP lysis buffer with lauryl maltoside. Lauryl maltoside is an alkyl disaccharide and polar surfactant that solubilizes protein complexes well and is easier to remove before MS analysis. 14. In the case of GFP-tagged PTP, one can use GFP-coupled agarose beads. 15. The incubation times and temperatures for each step should be optimized to ensure efficient binding of substrates to the mutant PTP and capture by the antibody. 16. The anti-PTP antibody should be specific for the mutant PTP used in the experiment to avoid cross-reactivity with other PTPs or proteins. 17. It is important to confirm the purity and activity of the mutant PTPs and validate the identified substrates using in vitro dephosphorylation assays with both wild-type and mutant PTPs. 18. Bands are detected by chemiluminescence imaging method or by exposure to X-ray film.

Identification of Protein Tyrosine Phosphatase (PTP) Substrates

133

Acknowledgments Work in A.M.B. lab is supported by NIH grants R01 AR080152 and R01 HL158876. References 1. Soulsby M, Bennett AM (2009) Physiological signaling specificity by protein tyrosine phosphatases. Physiology (Bethesda) 24:281–289. https://doi.org/10.1152/physiol.00017. 2009. 24/5/281 [pii] 2. Hendriks WJ, Elson A, Harroch S, Pulido R, Stoker A, den Hertog J (2013) Protein tyrosine phosphatases in health and disease. FEBS J 280(2):708–730. https://doi.org/10.1111/ febs.12000 3. Tonks NK (2013) Protein tyrosine phosphatases--from housekeeping enzymes to master regulators of signal transduction. FEBS J 280(2):346–378. https://doi.org/10.1111/ febs.12077 4. Hubbard SR, Till JH (2000) Protein tyrosine kinase structure and function. Annu Rev Biochem 69:373–398. https://doi.org/10.1146/ annurev.biochem.69.1.373 5. Lee H, Yi JS, Lawan A, Min K, Bennett AM (2015) Mining the function of protein tyrosine phosphatases in health and disease. Semin Cell Dev Biol 37:66–72. https://doi.org/10. 1016/j.semcdb.2014.09.021 6. Tautz L, Critton DA, Grotegut S (2013) Protein tyrosine phosphatases: structure, function, and implication in human disease. Methods Mol Biol 1053:179–221. https://doi.org/10. 1007/978-1-62703-562-0_13 7. Flint AJ, Tiganis T, Barford D, Tonks NK (1997) Development of “substrate-trapping” mutants to identify physiological substrates of protein tyrosine phosphatases. Proc Natl Acad Sci U S A 94(5):1680–1685

8. Tiganis T, Bennett AM (2007) Protein tyrosine phosphatase function: the substrate perspective. Biochem J 402(1):1–15. https://doi. org/10.1042/BJ20061548 9. Mercan F, Bennett AM (2010) Analysis of protein tyrosine phosphatases and substrates. Curr Protoc Mol Biol Chapter 18:Unit 18.16. https://doi.org/10.1002/0471142727. mb1816s91 10. Radha V (2016) Use of dominant-negative/ substrate trapping PTP mutations to search for PTP interactors/substrates. Methods Mol Biol 1447:243–265. https://doi.org/10. 1007/978-1-4939-3746-2_14 11. Mitra A, Kalayarasan S, Gupta V, Radha V (2011) TC-PTP dephosphorylates the guanine nucleotide exchange factor C3G (RapGEF1) and negatively regulates differentiation of human neuroblastoma cells. PLoS One 6(8): e23681. https://doi.org/10.1371/journal. pone.0023681 12. Chang YC, Lin SY, Liang SY, Pan KT, Chou CC, Chen CH, Liao CL, Khoo KH, Meng TC (2008) Tyrosine phosphoproteomics and identification of substrates of protein tyrosine phosphatase dPTP61F in Drosophila S2 cells by mass spectrometry-based substrate trapping strategy. J Proteome Res 7(3):1055–1066 13. Liang F, Kumar S, Zhang ZY (2007) Proteomic approaches to studying protein tyrosine phosphatases. Mol BioSyst 3(5):308–316. https://doi.org/10.1039/b700704n

Chapter 9 Kinase-Catalyzed Biotinylation to Identify Phosphatase Substrates (K-BIPS) Hannah J. Bremer and Mary Kay H. Pflum Abstract Phosphorylation is a reversible post-translational modification that alters the functions of proteins to govern various cellular events, including cell signaling. Kinases catalyze the transfer of a phosphoryl group onto the hydroxyl residue of serine, threonine, and tyrosine, while phosphatases catalyze the removal. Unregulated kinase and phosphatase activity have been observed in various cancers and neurodegenerative diseases. Despite their importance in cell biology, the role of phosphatases in cellular events has yet to be fully characterized, partly due to the lack of tools to identify phosphatase–substrate pairs in a biological context. The method called kinase-catalyzed biotinylation to identify phosphatase substrates (K-BIPS) was developed to remedy the lack of information surrounding phosphatase biology, particularly focused on substrate identification. In the K-BIPS method, the γ-phosphoryl modified adenosine 5′-triphosphate (ATP) analog, ATP-biotin, is used by kinases to biotin-label phosphoproteins. Because phosphatases must initially remove a phosphoryl group for subsequent biotinylation by ATP-biotin, phosphatase substrates are identified in K-BIPS by comparing biotinylated proteins in the presence and absence of active phosphatases. K-BIPS has been used to discover novel substrates of both serine/threonine and tyrosine phosphatases. This chapter describes the K-BIPS method to enable the identification of substrates to any phosphatases of interest, which will augment studies of phosphatase biology. Key words Phosphatase, Phosphatase substrate, Substrate identification, Proteomics

1

Introduction Protein kinases and phosphatases are enzymes that manage the reversible phosphorylation of proteins [1]. By influencing protein structure and function, phosphorylation regulates a variety of cellular events, including cell signaling, cell growth, and differentiation [2]. When phosphorylation is uncontrolled, various cancers and neurodegenerative diseases can arise [3]. Due to the potential for disease, dynamic phosphorylation is highly regulated. Kinases catalyze the transfer of the γ-phosphoryl of adenosine 5′-triphosphate (ATP) onto the hydroxyl group of specific serine, threonine, or tyrosine amino acids of substrates, and phosphatases

Damien The´venin and Jo¨rg P. Mu¨ller (eds.), Protein Tyrosine Phosphatases: Methods and Protocols, Methods in Molecular Biology, vol. 2743, https://doi.org/10.1007/978-1-0716-3569-8_9, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

135

136

Hannah J. Bremer and Mary Kay H. Pflum

Fig. 1 Kinase-catalyzed phosphorylation and the K-BIPS method. (a) Kinases transfer the γ-phosphoryl of ATP or an ATP analog onto the hydroxyl group of specific serine, threonine, or tyrosine amino acids on their substrate. (b) Active phosphatases in the cell lysate dephosphorylate substrates, which can subsequently undergo kinase-catalyzed biotinylation with ATP-biotin (top row). In lysates where phosphatases were inactivated through knockdown, substrates remain phosphorylated and unaffected by ATP-biotin (bottom row). After avidin enrichment of biotinylated proteins in both samples, comparison of proteins after LC-MS/MS analysis and label-free quantitation will identify potential substrates (yellow) of the inactivated phosphatase, without observing non-substrates (blue)

catalyze the removal of the phosphoryl, yielding the unmodified hydroxyl residue (Fig. 1a) [4]. With a critical role in the development of disease states, kinases are established targets of drug development efforts. Many kinase inhibitors are in clinical use, such as lapatinib and neratinib to treat breast cancers or tofacitinib to treat rheumatoid arthritis [5]. In contrast, only a few phosphatase inhibitors are in clinical trials. Foundational cell biology studies of phosphatases have lagged far behind kinases [6], which has stalled progress toward drug development. Protein phosphatases are divided into two main classes based on residue specificity. Protein tyrosine phosphatases (PTPs) act on phosphotyrosine and contain both catalytic and regulatory domains. In contrast, serine/threonine phosphatases (PPs) act on phosphoserine and phosphothreonine and typically contain only a catalytic domain, with the regulatory domain dictating substrate specificity provided by a second protein [7]. Given the different substrate specificities of phosphatases, several methods have been established to identify phosphatase substrates. The most widely used is substrate trapping, which utilizes an inactive mutant phosphatase that can stably bind its substrate for enrichment and liquid chromatography–tandem mass spectrometry (LC/MS-MS) analysis [8, 9]. However, substrate trapping of PPs is complicated by the lack of a regulatory domain, which could decrease phosphatase activity or specificity. As an alternative, peptide libraries containing phosphoserine, phosphothreonine, or phosphotyrosine have been developed to identify phosphatase–substrate pairs [10]; however, peptide sequences might poorly mimic the structure of a full-length

Kinase-Catalyzed Biotinylation to Identify Phosphatase Substrates (K-BIPS)

137

protein substrate, leading to false positive or negative hits. To characterize phosphatase biology more thoroughly, additional tools and methods must be developed that identify phosphatase– substrate pairs. To expand the methods used to study phosphoproteins, γ-phosphoryl modified ATP analogs have been developed as chemical tools to study kinases and phosphatases. Kinases utilize γ-modified ATP analogs as cosubstrates to label phosphoproteins. Various tags have been conjugated to the γ-phosphoryl of ATP to create a library of analogs, including ATP-biotin [11], ATP-benzophenone [13], ATP-arylazide [12], ATP-methylacrylamide [14], ATP-aryl fluorosulfate, and ATP-hexanoyl bromide [15]. A useful feature of kinase-catalyzed labeling is that the labeled phosphoproteins are resistant to phosphatase activity [16], allowing the phosphoprotein to remain labeled until subsequent analysis. Multiple methods have been developed using γ-phosphoryl modified ATP analogs that enable the study of phosphoproteins [17–22], which have led to the identification of novel kinase and phosphatase substrates. Among the methods utilizing γ-phosphoryl modified ATP analogs, phosphatase substrate discovery is the focus of a method called kinase-catalyzed biotinylation to identify phosphatase substrates (K-BIPS) [20, 21]. K-BIPS relies on the dynamic nature of kinases and phosphatases to mediate phosphorylation. If a protein has been phosphorylated by a kinase and endogenous ATP, the phosphorylated residue cannot be biotinylated by ATP-biotin until the phosphoryl group is removed by a phosphatase (Fig. 1b). As a result, active phosphatases are required for robust substrate labeling. Based upon the phosphatase dependence of kinase-catalyzed biotinylation, K-BIPS takes advantage of the fact that the loss of a single phosphatase activity will alter ATP-biotinmediated biotinylation of only the substrates of that inactive phosphatase. As a result, two sets of samples with different phosphatase activities are compared in K-BIPS (Fig. 1b). In one sample, the active phosphatase will generate free hydroxyl residues on its substrates, which can be biotinylated by ATP-biotin (Fig. 1b, top). In the second sample, the phosphatase is inactivated, typically by knockdown, which results in stable phosphorylation that prevents biotinylation by ATP-biotin (Fig. 1b, bottom). After the biotinylation reactions, enrichment using avidin resin will isolate biotinylated proteins. The bound biotinylated proteins can then be eluted, separated by sodium dodecyl sulfate – polyacrylamide gel electrophoresis(SDS-PAGE), and analyzed by Western blot to monitor the presence of a selected substrate of interest. The expectation is elevated substrate levels in samples with active versus inactive phosphatase. This way, K-BIPS validates a candidate substrate of a chosen phosphatase. In addition to gel-based analysis of a specific substrate, K-BIPS can be used to discover new substrates of

138

Hannah J. Bremer and Mary Kay H. Pflum

the phosphatase of interest by analyzing samples with LC/MS-MS. In this case, candidate substrates are expected to be enriched in samples with active versus inactive phosphatase. K-BIPS represents a powerful tool by offering both phosphatase substrate validation and discovery. In prior work, K-BIPS was established with the well-studied PTP, Protein Tyrosine Phosphatase 1B (PTP1B), and its substrate Janus Kinase 2 (JAK2) [23] using siRNA knockdown to reduce PTP1B activity. PTP1B knockdown resulted in decreased biotinylation of JAK2 without altering the presence of JAK2 in the lysate. Once the method was validated, K-BIPS was then used to discover novel substrates of PTP1B using LC-MS/MS) analysis, with Llactate dehydrogenase (LDHA) identified and validated as a novel substrate of PTP1B [21]. K-BIPS has also been applied to study the Ser/Thr phosphatase PP1. Because a regulatory subunit is needed together with the catalytic subunit to dictate substrate specificity, the small molecule guanabenz was used to decrease phosphatase activity by disrupting the complex between PP1 and one of its regulatory subunits, Gadd34. After proteomics analysis, COPS5 was identified as a candidate substrate. Subsequent validation of COPS5 as a PP1-Gadd34 substrate included siRNA-mediated knockdown of Gadd34 in K-BIPS, which established K-BIPS as a validation tool for PP-regulatory complexes [20]. Recently, K-BIPS was also applied to discover the substrates of the PP1-PPP1R12A complex using conditional knockdown of the PPP1R12A regulatory subunit [24]. This previous work established K-BIPS as an effective tool to identify and validate phosphatase substrates for any phosphatase of interest. The first step in K-BIPS is to create lysates with and without the activity of a chosen phosphatase (Fig. 1b). Small molecule inhibitors were previously used to inactivate phosphatases [20]. However, due to the lack of highly selective phosphatase inhibitors, siRNA is a more general means to achieve phosphatase inactivation. Given the broad utility of knockdown methods to any phosphatase or phosphatase complex of interest, the first protocol (Subheading 3.1) describes the siRNA knockdown of a phosphatase of interest, with knockdown of PTP1B phosphatase in HEK293 cells as an example. Samples without treatment or treated with a scrambled, non-targeting siRNA control are also included to confirm that the decrease in phosphatase levels is not a result of off-target effects from the siRNA or transfection reagent. After the generation of lysates with and without a chosen phosphatase activity, the second protocol (Subheading 3.2) describes kinase-catalyzed biotinylation, which is the next step in K-BIPS. Initially, kinase-catalyzed biotinylation of the selected lysates with gel analysis ensures the presence of active kinases, which is a prerequisite for K-BIPS. As an example experiment, kinase-catalyzed labeling of HEK293 lysates after PTP1B knockdown is described, which showed robust labeling by

Kinase-Catalyzed Biotinylation to Identify Phosphatase Substrates (K-BIPS)

139

ATP-biotin and active kinases. The final protocol (Subheading 3.3) completes the full K-BIPS method with avidin enrichment of the biotinylated proteins after kinase-catalyzed biotinylation, which can be used to confirm suspected substrates or discover new ones. The complete K-BIPS example experiment provided here focuses on validating the known PTP1B substrate JAK2. Tips are also provided to apply these methods to any chosen phosphatase in a cell line of interest.

2

Materials

2.1 siRNA Knockdown

1. Complete growth medium: DMEM containing 10% fetal bovine serum. Store at 4 °C. 2. Serum-free medium DMEM. Store at 4 °C. 3. T75 flasks. 4. Water-jacketed CO2 incubator. 5. PTP1B-targeting siRNA. Store at -20 °C. 6. Non-targeting pool of siRNA. Store at -20 °C. 7. siRNA transfection reagent. Store at 4 °C. 8. Trypsin-EDTA solution 0.25% with phenol red. Store at 4 °C. 9. Dulbecco’s phosphate-buffered saline (DPBS). Store at 4 °C. 10. Lysis buffer: 50 mM Tris–HCl, pH 7.5, 150 mM NaCl, 0.5% Triton X-100, and 10% glycerol. Store at 4 °C for up to 1 year. 11. Tube rotator. 12. Protease inhibitors: Added to the lysis buffer with final concentrations of one or more of the following: 50 μM phenylmethanesulfonyl fluoride (PMSF), 10 μM pepstatin A, 20 μM leupeptin, 100 μM benzamidine, 50 mM bestatin. Store at -20 °C. 13. Bradford reagent. Store at 4 °C. 14. SYPRO Ruby Protein Gel Stain (or another protein stain). 15. Anti-PTP1B antibody. Store at -20 °C. 16. Anti-rabbit HRP-conjugated antibody. Store at -20 °C. 17. Enhanced chemiluminescence (ECL) substrate. 18. Imaging and analyzing software. 19. 10% SDS-polyacrylamide gel: Buy commercial gels or make gels as previously published [25]. The choice of gel percentage depends on the mass of the protein of interest [25]. 20. Electrophoresis apparatus. 21. PVDF membrane. 22. Electrotransfer apparatus.

140

Hannah J. Bremer and Mary Kay H. Pflum

2.2 KinaseCatalyzed Biotinylation

1. 1.5-mL microcentrifuge tubes. 2. Benchtop microcentrifuge. 3. Temperature-controlled tube shaker. 4. Rocking platform. 5. Chemiluminescent and fluorescent imaging instrument. 6. ATP-biotin synthesized as previously published [16]. ATP-biotin is stored for up to 1 year as a dry lyophilized solid at -80 °C. After resuspension in water, keep at -80 °C in single-use aliquots, which will avoid freeze/thaw cycles. The concentration after resuspending lyophilized ATP-biotin in water typically ranges from 30 to 200 mM, which might need to be diluted before use in kinase reactions (see Note 1). 7. Cell lysate stock solution: A stock solution of 10 mg/mL or greater is preferred to maintain small reaction volumes (see Note 2). 8. 10× Protein Kinase (PK) buffer: 50 mM Tris–HCl, 10 mM MgCl2, 0.1 mM EDTA, 2 mM DTT, and 0.01% Brij 35 (New England Biolabs). 9. β-Mercaptoethanol. 10. 4× Gel-loading buffer, such as 4× Laemmli buffer: 277.8 mM Tris–HCl, pH 6.8, 44.4% (v/v) glycerol, 4.4% lithium dodecyl sulfate, 0.02% bromophenol blue, containing fresh β-mercaptoethanol reducing agent (10% w/v). 11. Tris-buffered saline with Tween (TBST): First, prepare a 10× TBS solution containing 200 mM Tris-base and 1500 mM NaCl. Adjust pH to 7.6 with either HCl or NaOH as needed. The solution can be stored at room temperature. To make TBST, dilute 10× TBS with water to a final concentration of 1× and add Tween to 0.1%. 12. Blocking solution: Use 5% (w/v) nonfat dry milk or 5% BSA in 1× TBST or a commercially available blocking solution (see Note 3). 13. Cy5-labeled streptavidin.

2.3 Avidin Enrichment

1. Pierce Spin Columns-Screw Cap (Thermo Fisher Scientific). 2. SpeedVac concentrator. 3. 0.5-mL 3-kDa centrifugal filter units. 4. Foam floating test tube rack. 5. Phosphate binding buffer (PBB): 28 mM sodium phosphate monobasic monohydrate, 72 mM sodium phosphate dibasic heptahydrate, pH 7.2, and 150 mM NaCl.

Kinase-Catalyzed Biotinylation to Identify Phosphatase Substrates (K-BIPS)

141

6. Avidin resin: Either Streptavidin or NeutrAvidin resins may be used (see Note 4). 7. SDS elution buffer: 2% SDS in water. 8. Anti-JAK2 antibody. Store at -20 °C.

3

Methods

3.1 siRNA Knockdown with Verification by Gel Analysis

1. Culture desired cell line in appropriate growth medium (10 mL) in three T75 flasks at 37 °C in 5% CO2 environment until 50–70% confluency. Use one flask for each cell condition: targeting siRNA, non-targeting siRNA control, and non-treated control. For the example experiment, culture HEK293 cells in DMEM complete growth medium. 2. Separately, prepare the knockdown reagents. Dilute pools of targeting and non-targeting siRNA separately to 1 mL with serum-free media. Create a negative control with only 1 mL of serum-free media for the non-treated sample. Incubate the three tubes for 5 min at room temperature. For the example experiment, use serum-free DMEM. 3. Separately, prepare three tubes, each containing 50 μL of transfection reagent in 1 mL of serum-free media. Incubate the three tubes for 5 min at room temperature. 4. Combine each sample from step 2 with one diluted transfection reagent solution from step 3 and incubate for 20 min at room temperature. Then, dilute each sample further with 8 mL of serum-free media to give a final volume of 10 mL. Finally, remove the media from each cell growth in step 1, and add one solution from step 3 to each flask. 5. After a 72-h incubation at 37 °C in a 5% CO2 environment, harvest cells by removing the media, washing with 3 mL of DPBS, and incubating with 2 mL of trypsin-EDTA for 5 min at 37 °C to release the adherent cells from the flask. 6. Dilute the trypsin-EDTA solution with 6 mL of media to stop the trypsinization and pellet the cells by centrifugation at 1000 rpm for 5 min at 4 °C. Remove media, wash cells with 1 mL of cold DPBS, and pellet the washed cells by centrifugation at 1000 rpm in a microcentrifuge for 5 min at 4 °C. 7. Store cell pellets at -80 °C or lyse immediately. 8. For cell lysis, add 500 μL of lysis buffer and 5.2 μL of protease inhibitor cocktail to each cell pellet (see Note 5) and pipette up and down rapidly ten times to mix thoroughly. Then, rotate samples for 30 min at 4 °C and collect soluble fraction by centrifugation at 13,200 rpm in a microcentrifuge for 20 min at 4 °C.

142

Hannah J. Bremer and Mary Kay H. Pflum

9. Determine the protein concentration in the lysate by Bradford assay according to the manufacturer’s instructions. Store the lysates at -80 °C in single-use aliquots or use them immediately. 10. For gel analysis of knockdown, add enough 4× gel-loading dye to 50 μg total protein from each lysate sample from step 9 to make a final concentration of 1×. Incubate samples at 95 °C for 1 min to denature proteins. 11. Prepare SDS-polyacrylamide gels according to published protocols [25] or use commercially available gels. In this example of PTP1B knockdown in HEK293 lysate, two 10% gels were used to resolve PTP1B (50 kDa) for subsequent immunoblotting and SYPRO Ruby staining. 12. Load the two 10% SDS-PAGE gels with the denatured samples (50% of each sample in separate wells of each gel). Separate proteins using 100 V for 10 min (or until the dye front has entered the stacking layer) and then increase the voltage to 180 V for 45 min (or until the dye front has run off the gel). 13. Electrotransfer the separated proteins from one gel onto a PVDF membrane at 90 V for 2 h using an electroblotting apparatus [26]. 14. During electrotransfer in step 13, as a loading control, stain the second gel with SYPRO Ruby according to the manufacturer’s procedure. Skip to step 20 for protein visualization. 15. After electrotransfer in step 13, incubate the membrane with the manufacturer-recommended dilution of the primary antibody in 5% (w/v) BSA in 1× TBST for 1 h at room temperature or overnight at 4 °C with rocking. In the example of PTP1B, use a 1:1000 dilution of PTP1B primary antibody in 5% BSA and 1× TBST and incubate for 1 h at room temperature. Primary antibody dilutions may be reused for 1 month if stored at 4 °C. 16. After the primary antibody incubation in step 15, wash the membrane three times with 15 mL of 1× TBST for 5 min with rocking. 17. Incubate the membrane with the manufacturer-recommended dilution of the secondary antibody in 5% (w/v) nonfat milk in 1× TBST for 1 h at room temperature with rocking. In the example of PTP1B, use 1:10,000 dilution of anti-rabbit HRP secondary antibody in 5% nonfat milk and 1× TBST, and incubate for 1 h at room temperature with rocking. 18. After the secondary antibody incubation, wash the membrane three times with 15 mL of 1× TBST for 5 min with rocking.

Kinase-Catalyzed Biotinylation to Identify Phosphatase Substrates (K-BIPS)

143

Fig. 2 PTP1B knockdown in HEK293 cells. Samples were generated following the protocol in Subheading 3.1 with PTP1B knockdown in HEK293 cells. Western blot with a PTP1B antibody (α-PTP1B) was used to assess knockdown levels, and Sypro Ruby total protein stain was used as a load control. Quantification of the targeting siRNA (SR, lane 1), scrambled non-targeting siRNA (SC, lane 2), and nontreated (NT, lane 3) samples using ImageJ revealed 91% knockdown of PTP1B (50 kDa) with targeted siRNA, but only 16% knockdown with nontargeting siRNA, compared with the NT control in this illustrative example (see Note 6). Molecular weight markers (kDa) are shown to the left of each gel image

19. To observe proteins with the HRP in the secondary antibody, incubate the membrane with enhanced chemiluminescence (ECL) substrate following the manufacturer’s instructions. 20. Visualize proteins on the Western blot membrane from step 19 with an imaging instrument reading chemiluminescence and the SYPRO Ruby stained gel from step 14 with an imaging instrument reading fluorescence. Example gel images are seen in Fig. 2. 21. Quantify the intensity of the PTP1B protein bands in the Western blot gel. To determine the percentage knockdown value, divide the intensity values of the PTP1B bands in the targeting and nontargeting siRNA-treated samples by the nontreated sample, subtract that calculated ratio from 1, and multiply by 100 (see Note 6). Example knockdown percentages are provided in the legend of Fig. 2. 3.2 KinaseCatalyzed Labeling with ATP-Biotin

1. Label four 1.5-mL Eppendorf tubes (Table 1). 2. Add 10× PK buffer into the labeled tubes to achieve a final concentration of 1× and dilute with ultrapure water, so all samples have the same final volume. The volumes of 10× PK buffer and water used in the illustrative example with PTP1B knockdown lysates are provided in Table 1, where a final volume of 30 μL was used.

144

Hannah J. Bremer and Mary Kay H. Pflum

Table 1 Kinase-catalyzed biotinylation reactions for gel analysis Reactions

1

2

3

4

Lysate from targeting siRNA transfection (SR, 150 μg)

15 μL

–

–

–

Lysate from nontargeting siRNA transfection (SC, 150 μg)

–

15 μL

–

–

Nontreated lysate (NT, 150 μg)

–

–

15 μL

15 μL

ATP-biotin (final concentration of 2 mM)

6 μL

6 μL

6 μL

–

PK buffer (final concentration of 1×)

3 μL

3 μL

3 μL

3 μL

Water

6 μL

6 μL

6 μL

12 μL

Total volume

30 μL

30 μL

30 μL

30 μL

Lysate stock concentrations were 10 μg/μL. An ATP-biotin solution of 10 mM was used, which was a dilution of the 50 mM stock solution

3. Thaw lysate from the siRNA treatment protocol (Subheading 3.1) on ice and then add the appropriate lysates to each labeled tube. For the example reactions, lysates containing 150 μg total protein were used, which required 15 μL of the 10 μg/μL lysate stocks (Table 1). 4. Thaw ATP-biotin on ice. Initiate kinase reactions by adding ATP-biotin to respective tubes. In this example, a final concentration of 2 mM ATP-biotin was used (see Note 7). Also note that a negative control reaction without ATP-biotin was included (Table 1, reaction 4) to observe endogenously biotinylated proteins in the lysates. 5. Vortex samples for 5 s to mix and centrifuge the samples at 10,000 rpm in a microcentrifuge for 5 s to combine reagents. 6. Incubate samples in the temperature-controlled tube shaker at 31 °C for 2 h with 300 rpm shaking. 7. After reaction incubation, add 4× gel-loading dye to each sample to achieve a final concentration of 1×. In this example, 7.5 μL of gel-loading dye was added to each sample. Samples were incubated at 95 °C for 1 min to denature proteins. 8. Prepare two SDS-PAGE gels, load samples onto the gels, and run SDS-PAGE as outlined in Subheading 3.1, steps 11 and 12. Electrotransfer proteins from one gel onto a membrane as outlined in Subheading 3.1, step 13. In this example, two 10% gels were used. 9. During electrotransfer, stain the second gel with SYPRO Ruby stain according to the manufacturer’s procedure.

Kinase-Catalyzed Biotinylation to Identify Phosphatase Substrates (K-BIPS)

145

10. After electrotransfer in step 8, block the membrane as needed with a blocking solution (see Note 3). For this example, the membrane was blocked with 1× TBST containing 5% BSA for 1 h at room temperature with rocking. 11. Wash the membrane three times with 15 mL of 1× TBST for 5 min with rocking at room temperature. 12. Incubate the membrane with the manufacturer’s recommended amount of streptavidin-Cy5 in 10 mL 1× TBST containing 2% BSA for 1 h at room temperature with rocking while protected from light to avoid Cy5 fluorophore quenching (see Note 8). 13. After incubation, rinse the membrane three times with 15 mL 1× TBST for 5 min with rocking. Visualize proteins in both gels with an imaging instrument reading fluorescence. Representative gel images for the biotinylation of PTP1B knockdown lysates are shown in Fig. 3.

Fig. 3 Kinase-catalyzed biotinylation of lysates after PTP1B knockdown in HEK293 cells. Samples were generated following the protocol in Subheading 3.2. Staining with a streptavidin-Cy5 (SA-Cy5) conjugate was used to assess biotinylated proteins, whereas Sypro Ruby total protein stain was used as a load control. The presence of biotinylated proteins after ATP-biotin incubation documents that active kinases are available, which is a requirement of K-BIPS. Note that only a slight reduction (or no reduction) in biotinylated proteins is typically observed in lysates with targeting siRNA (SR, lane 1) compared to lysates from scrambled non-targeting siRNA (SC, lane 2) or nontreated (NT, lane 3) due to inactivation of only a single phosphatase. The negative control sample without ATP-biotin shows the presence of endogenously biotinylated proteins (arrow). Molecular weight markers (kDa) are shown to the left of each gel image

146

Hannah J. Bremer and Mary Kay H. Pflum

Table 2 Kinase-catalyzed biotinylation reactions for avidin enrichment Reactions

1

2

3

Lysate from targeting siRNA transfection (SR, 500 μg)

50 μL

–

–

Lysate from nontargeting siRNA transfection (SC, 500 μg)

–

50 μL

–

Nontreated lysates (NT, 500 μg)

–

–

50 μL

ATP-biotin (final concentration of 2 mM)

2.8 μL

2.8 μL

2.8 μL

PK buffer (final concentration of 1×)

7 μL

7 μL

7 μL

Water

10.2 μL

10.2 μL

10.2 μL

Total volume

70 μL

70 μL

70 μL

Lysate stock concentrations were 10 μg/μL. ATP-biotin stock solution was 50 mM

3.3 Avidin Enrichment

1. Repeat Subheading 3.2, steps 1–6 except using a larger amount of total lysate proteins. For the PTP1B K-BIPS example provided here, 500 μg of total proteins in each lysate was used, with the required amount of all reagents shown in Table 2. 2. While the kinase-catalyzed biotinylation reactions are incubating (step 1), label three Amicon Ultra-0.5-mL 3-kDa centrifugal filters and two Eppendorf tubes that come with each filter. Then, equilibrate each filter by placing it into one Eppendorf tube, adding 400 μL of PBB, centrifuging at 14,200 × g for 10 min, and emptying any remaining buffer from the filter (see Note 9). 3. After incubating the kinase-catalyzed biotinylation reactions in step 1, add each reaction sample to its labeled filter unit, dilute to 400 μL with PBB, centrifuge samples at 14,200 × g for 30 min at 4 °C, and remove the collected filtrate. Repeat these dilution and centrifugation steps for a second time (see Note 10). 4. During the second 30-min centrifugation in step 3, prepare the avidin resin. For each reaction sample, obtain one spin column, cap, plug, and 2-mL Eppendorf tube. Place each spin column (without plug) into the 2-mL Eppendorf tube. Thoroughly mix the avidin resin to create a homogenized slurry. Add 400 μL of resin slurry to each spin column and centrifuge at 500 × g for 1 min at room temperature. Add 400 μL of PBB to the resin in each column and centrifuge at 500 × g for 1 min. Repeat the PBB wash step three more times (for a total of five times), with removal of the flow through waste from the 2-mL Eppendorf as needed (see Note 11).

Kinase-Catalyzed Biotinylation to Identify Phosphatase Substrates (K-BIPS)

147

5. After both filter spins from step 3, inverted filter into the second labeled Eppendorf tube and centrifuge at 3200 rpm in a microcentrifuge for 2 min at 4 °C to collect the biotinylated samples. Remove 50 μg of collected lysate as input controls for later gel analysis (steps 15–17); input samples can be stored at -80 °C until gel loading, if needed. For this example reaction, save 7 μL of lysate from each sample. 6. After avidin resin washing (step 4), attach spin column plugs to the columns containing the avidin resin, and add each filtered lysate from step 5 to one filter containing resin. Dilute the lysate-resin mixture to 400 μL with PBB. 7. Place top caps on spin columns containing the lysate–resin mixture. Place sealed spin columns on a rocker and rotate for 1 h at room temperature (see Note 12). 8. Label new 1.5-mL Eppendorf tubes for each sample. After incubation in step 7, remove the caps and plugs from the spin columns and place them in the new 1.5-mL Eppendorf tubes. Centrifuge samples at 500 × g for 1 min at room temperature (see Note 13). 9. To the protein-bound resin in the spin columns, add 400 μL of PBB and centrifuge at 500 × g for 1 min. Repeat for a total of 10 PBB washes. Then, repeat this washing step with 400 μL of water for a total of 10 water washes (see Note 14). Discard filtrates after each wash. 10. Return plugs to spin columns and add 200 μL of SDS elution buffer to each column. Replace top caps but only loosely tighten to prevent pressure build-up while heating in the next step. 11. Put sample in foam floater and place in 95 °C water bath for 7 min. Water from the bath must not enter columns, which will dilute samples and prevent effective elution. 12. Label new 1.5-mL Eppendorf tubes for each sample. Remove spin columns from water bath and dry any excess water on the outsides. Remove plugs and top caps from the spin columns, and place columns into the newly labeled Eppendorf tubes. Centrifuge at 500 × g for 2 min to collect the eluate. 13. Dry eluate samples (step 12) in the speedvac concentrator. Once dried, store samples at -20 °C until gel analysis or use immediately. 14. Resuspend the eluate samples in a water volume equal to the input lysate samples from step 5. In this example, use 7 μL of water. If the volume is too low to resuspend the dried samples fully, more water may be added (see Note 15). However, adjust the input samples such that the final volumes of all samples are equal before loading gel (see Note 16).

148

Hannah J. Bremer and Mary Kay H. Pflum

Fig. 4 K-BIPS enrichment of JAK2, a known PTP1B substrate. Samples were generated following the protocol in Subheading 3.3. Western blot visualization with PTP1B (α-PTP1B) and JAK2 (α-JAK2) antibodies is indicated. Eluate samples after enrichment are analyzed in the top image, while input lysate samples are visualized in the bottom images. The input lysate samples show PTP1B knockdown in the targeting siRNA (SR) compared with nontreated (NT) or scrambled non-targeting siRNA (SC) samples and equal levels of JAK2 in all samples. In eluate samples, decreased JAK2 (131 kDa) is observed in targeting siRNA (SR) samples compared with nontreated (NT) or scrambled non-targeting siRNA (SC) samples, as expected. Molecular weight markers (kDa) are shown to the left of each gel image

15. Add 4× gel-loading dye to a final concentration of 1×. For this example experiment, 3.5 μL of gel-loading dye was added to each sample. Briefly vortex to mix, and then centrifuge for 10 s at 10,000 × g to collect samples at the bottom of the tubes. 16. Incubate samples for 2 min at 95 °C. Collect samples by centrifugation. Slowly pipet up and down or briefly vortex until proteins have resuspended. 17. Prepare SDS-PAGE gels, load samples onto the gels, run SDS-PAGE, and electrotransfer as outlined in Subheading 3.1, steps 11–13. Visualize proteins with an antibody for substrate of interest, as outlined in Subheading 3.1, steps 15–20 (see Note 17). In the example of PTP1B, the change in enrichment due to phosphatase activity of known substrate JAK2 was analyzed via western blot (Fig. 4). For the western blot, use a 1:1000 dilution of JAK2 primary antibody in 10 mL of 1× TBST containing 5% BSA, and incubate for 1 h at room temperature with rocking.

4

Notes 1. After resuspending lyophilized ATP-biotin in water (200 μL is typical), determine the concentration using a UV–Vis instrument by taking absorbance from 230 to 280 nM. Using the wavelength with the highest absorbance, calculate

Kinase-Catalyzed Biotinylation to Identify Phosphatase Substrates (K-BIPS)

149

concentration with Beer’s law (A = εlC, where A is the absorbance at maximal wavelength, l is the cuvette pathlength (typically 1 cm), ε is the molar absorptivity (15.4 × 103 M-1 cm-1 for adenosine), and C is the calculated concentration). Note that dilutions of the stock ATP-biotin solution (1:1000, for example) will likely be needed to obtain the most accurate measurements, which should be in the 0.2–0.8 absorbance range. ATP-biotin purity can be assessed via thin layer chromatography with silica and a solvent system consisting of 3:1.5:0.5 isopropanol:ammonium hydroxide:water, with visualization of ATP-biotin using UV light at 365 nM (Rf of ATP-biotin is 0.4). Minimal degraded ATP (Rf of ATP is 0) should be present to prevent nonspecific labeling by the biotin-amine degradation product [27, 28]. 2. General kinase-catalyzed biotinylation reactions with gel analysis require 50–250 μg of lysate per sample, depending on the abundance of the substrate of interest. 3. Using nonfat dry milk to block the membrane before streptavidin-Cy5 staining to assess kinase-catalyzed biotinyation is not recommended due to the presence of biotin in milk, which might lead to non-specific binding. 4. The enrichment step of K-BIPS can be completed with either Streptavidin or NeutrAvidin resin (Subheading 3.3). NeutrAvidin and Streptavidin both have a lower pI than Avidin, which helps to reduce non-specific binding to the beads. To further reduce non-specific binding, NeutrAvidin may be used due to its deglycosylated state and lack of off-target binding sequences [29, 30]. 5. To maintain small volumes in kinase-catalyzed labeling reactions, the volume of lysis buffer may be decreased to increase lysate concentrations. 6. For K-BIPS, knockdown of the phosphatase of interest by 75% or greater is expected after siRNA transfection. If siRNA transfection results in less than 75% knockdown, optimization of the transfection conditions, such as higher concentrations of siRNA, should be attempted; however, increased siRNA and transfection reagent can be toxic to cells, resulting in lower protein concentration. Note also that multiple companies sell siRNA reagents, giving more optimization options. Alternatively, a different cell line can be used to optimize knockdown. K-BIPS has been performed in HeLa and HEK293 cell lines, but any cell line of interest can be used. If siRNA-based knockdown proves challenging, lysates derived from viral or CRISPR-mediated knockdown/knockout should also be compatible with K-BIPS.

150

Hannah J. Bremer and Mary Kay H. Pflum

7. ATP-biotin concentrations for kinase-catalyzed biotinylation vary from 2 to 10 mM and can be optimized for robust labeling. 8. If the biotin signal is low using streptavidin-Cy5, streptavidinHRP, followed by incubation with an HRP chemiluminescent substrate (e.g., enhanced Chemiluminescence (ECL) substrate), can be used as an alternative for visualization of biotinylated proteins after kinase-catalyzed biotinylation. Streptavidin-HRP is more sensitive due to the multiple turnovers of HRP substrates. However, Streptavidin-Cy5 might give cleaner gel images if over-saturation from the HRP enzyme is observed. 9. Filters should be equilibrated with PBB before adding lysates to reduce background binding that could overwhelm the biotinylation signal and interfere with the analysis. 10. The filtration step after kinase-catalyzed biotinylation reactions will remove excess ATP-biotin from the samples before avidin enrichment. Excess biotin could prevent the binding and enrichment of biotinylated proteins from the lysates. 11. Keep resin on ice until ready to be used. The resin should always be stored in buffer; do not allow the resin to dry. 12. The incubation time for binding biotinylated proteins to avidin resin may need to be optimized depending on lysate and protein abundance. For example, the incubation time can be extended if few biotinylated proteins are observed by gel analysis after avidin enrichment. 13. Unbound proteins in the flow through filtrate can be kept for gel analysis to assess the success of avidin enrichment. If interested in analyzing the unbound proteins, store the filtrate on ice and then process it like the elution samples in steps 13–17. 14. If high background is observed by gel analysis, which prevents clear visualization of change in biotinylation or protein levels between samples (Fig. 4), the avidin resin washing conditions can be optimized. To remove nonspecific binding to the resin after enrichment, a high salt wash can be used where 500 mM NaCl replaces the 150 mM NaCl in PBB. 15. After drying the eluate in the speedvac, resuspending the dried proteins into solution can be difficult, and precipitates can persist. If precipitates are observed, add more water or sonicate proteins in water or gel-loading dye until fully resuspended. 16. When loading input and elution samples onto a gel, ensure protein levels are equal in each lane. Excess protein might cause gel lanes to run unevenly. Protein concentration might need to be optimized to ensure enough signal is seen in every lane.

Kinase-Catalyzed Biotinylation to Identify Phosphatase Substrates (K-BIPS)

151

17. K-BIPS samples can be analyzed alternatively by LC-MS/MS to discover unanticipated substrates of the chosen phosphatase. LC-MS/MS after K-BIPS is previously described [21]. References 1. Wang Z, Cole PA (2014) Catalytic mechanisms and regulation of protein kinases. Methods Enzymol 548:1–21. https://doi.org/10. 1016/b978-0-12-397918-6.00001-x 2. Hunter T (1995) Protein kinases and phosphatases: the Yin and Yang of protein phosphorylation and signaling. Cell 80(2):225–236. https://doi.org/10.1016/0092-8674(95) 90405-0 3. Cohen P, Alessi DR (2013) Kinase drug discovery – what’s next in the field? ACS Chem Biol 8(1):96–104. https://doi.org/10. 1021/cb300610s 4. Ubersax JA, Ferrell JE Jr (2007) Mechanisms of specificity in protein phosphorylation. Nat Rev Mol Cell Biol 8(7):530–541. https://doi. org/10.1038/nrm2203 5. Roskoski R Jr (2022) Properties of FDA-approved small molecule protein kinase inhibitors: a 2022 update. Pharmacol Res 175:106037. https://doi.org/10.1016/j. phrs.2021.106037 6. Mullard A (2018) Phosphatases start shedding their stigma of undruggability. Nat Rev Drug Discov 17(12):847–849. https://doi.org/10. 1038/nrd.2018.201 7. Jackson MD, Denu JM (2001) Molecular reactions of protein phosphatases – insights from structure and chemistry. Chem Rev 101(8): 2313–2340. https://doi.org/10.1021/ cr000247e 8. Blanchetot C, Chagnon M, Dube N, Halle M, Tremblay ML (2005) Substrate-trapping techniques in the identification of cellular PTP targets. Methods 35(1):44–53. https://doi.org/ 10.1016/j.ymeth.2004.07.007 9. Flint Andrew J, Tiganis T, Barford D, Tonks Nicholas K (1997) Development of “substratetrapping” mutants to identify physiological substrates of protein tyrosine phosphatases. Proc Natl Acad Sci 94(5):1680–1685. https://doi.org/10.1073/pnas.94.5.1680 10. Ren L, Chen X, Luechapanichkul R, Selner NG, Meyer TM, Wavreille A-S, Chan R, Iorio C, Zhou X, Neel BG, Pei D (2011) Substrate specificity of protein tyrosine phosphatases 1B, RPTPα, SHP-1, and SHP-2. Biochemistry 50(12):2339–2356. https:// doi.org/10.1021/bi1014453

11. Green KD, Pflum MH (2007) Kinasecatalyzed biotinylation for phosphoprotein detection. J Am Chem Soc 129(1):10–11 12. Suwal S, Pflum MH (2010) Phosphorylationdependent kinase-substrate cross-linking. Angew Chem Int Ed Engl 49(9):1627–1630. https://doi.org/10.1002/anie.200905244 13. Garre S, Senevirathne C, Pflum MK (2014) A comparative study of ATP analogs for phosphorylation-dependent kinase-substrate crosslinking. Bioorg Med Chem 22(5): 1620–1625. https://doi.org/10.1016/j.bmc. 2014.01.034 14. Fouda AE, Gamage AK, Pflum MKH (2021) An affinity-based, cysteine-specific ATP analog for kinase-catalyzed crosslinking. Angew Chem Int Ed Engl 60(18):9859–9862. https://doi. org/10.1002/anie.202014047 15. Beltman RJ, Herppich AA, Bremer HJ, Pflum MKH (2023) Affinity-based kinase-catalyzed crosslinking to study kinase–substrate pairs. Bioconjugate Chem 34(6):1054–1060. https://doi.org/10.1021/acs.bioconjchem. 3c00131 16. Senevirathne C, Pflum MK (2013) Biotinylated phosphoproteins from kinase-catalyzed biotinylation are stable to phosphatases: implications for phosphoproteomics. Chembiochem 14(3):381–387. https://doi.org/10.1002/ cbic.201200626 17. Ramanayake-Mudiyanselage V, Embogama DM, Pflum MKH (2021) Kinase-catalyzed biotinylation to map cell signaling pathways: application to epidermal growth factor signaling. J Proteome Res 20(10):4852–4861. https://doi.org/10.1021/acs.jproteome. 1c00562 18. Song H, Kerman K, Kraatz HB (2008) Electrochemical detection of kinase-catalyzed phosphorylation using ferrocene-conjugated ATP. Chem Commun (Camb) 4:502–504. https://doi.org/10.1039/b714383d 19. Embogama DM, Pflum MK (2017) K-BILDS: a kinase substrate discovery tool. Chembiochem 18(1):136–141. https://doi.org/10. 1002/cbic.201600511 20. Dedigama-Arachchige PM, Acharige NPN, Pflum MKH (2018) Identification of PP1-Gadd34 substrates involved in the

152

Hannah J. Bremer and Mary Kay H. Pflum

unfolded protein response using K-BIPS, a method for phosphatase substrate identification. Mol Omics 14(2):121–133. https://doi. org/10.1039/c7mo00064b 21. Acharige NPN, Pflum MKH (2021) l-lactate dehydrogenase identified as a protein tyrosine phosphatase 1B substrate by using K-BIPS. Chembiochem 22(1):186–192. https://doi. org/10.1002/cbic.202000499 22. Garre S, Gamage AK, Faner TR, DedigamaArachchige P, Pflum MKH (2018) Identification of kinases and interactors of p53 using kinase-catalyzed cross-linking and immunoprecipitation. J Am Chem Soc 140(47): 16299–16310. https://doi.org/10.1021/ jacs.8b10160 23. Johnson TO, Ermolieff J, Jirousek MR (2002) Protein tyrosine phosphatase 1B inhibitors for diabetes. Nat Rev Drug Discov 1(9):696–709. https://doi.org/10.1038/nrd895 24. Dedigama-Arachchige PM, Acharige NPN, Zhang X, Bremer HJ, Yi Z, Pflum MKH (2023) Identification of PP1c-PPP1R12A substrates using kinase-catalyzed biotinylation to identify phosphatase substrates. ACS Omega 8 (39):35628–35637. https://doi.org/10. 1021/acsomega.3c01944

25. He F (2011) Laemmli-SDS-PAGE. Bio-protocol 1(11):e80. https://doi.org/10. 21769/BioProtoc.80 26. Goldman A, Ursitti JA, Mozdzanowski J, Speicher DW (2015) Electroblotting from polyacrylamide gels. Curr Protoc Protein Sci 82: 10.17.11–10.17.16. https://doi.org/10. 1002/0471140864.ps1007s82 27. Arora DP, Boon EM (2013) Unexpected biotinylation using ATP-gamma-Biotin-LC-PEOamine as a kinase substrate. Biochem Biophys Res Commun 432(2):287–290. https://doi. org/10.1016/j.bbrc.2013.01.115 28. Senevirathne C, Embogama DM, Anthony TA, Fouda AE, Pflum MK (2016) The generality of kinase-catalyzed biotinylation. Bioorg Med Chem 24(1):12–19. https://doi.org/10. 1016/j.bmc.2015.11.029 29. Nguyen TT, Sly KL, Conboy JC (2012) Comparison of the energetics of avidin, streptavidin, neutrAvidin, and anti-biotin antibody binding to biotinylated lipid bilayer examined by second-harmonic generation. Anal Chem 84(1):201–208. https://doi.org/10.1021/ ac202375n 30. NeutrAvidin Protein (2022) ThermoFisher. https://www.thermofisher.com/order/cata log/product/31000. Accessed 22 Nov 22

Chapter 10 System-Level Analysis of the Effects of RPTPs on Cellular Signaling Networks Jacqueline Gerritsen, Sophie Rizzo, Damien The´venin, and Forest M. White Abstract Tyrosine phosphorylation regulates signaling network activity downstream of receptor tyrosine kinase (RTK) activation. Receptor protein tyrosine phosphatases (RPTPs) serve to dephosphorylate RTKs and their proximal adaptor proteins, thus serving to modulate RTK activity. While the general function of RPTPs is well understood, the direct and indirect substrates for each RPTP are poorly characterized. Here we describe a method, quantitative phosphotyrosine phosphoproteomics, that enables the identification of specific phosphorylation sites whose phosphorylation levels are altered by the expression and activity of a given RPTP. In a proof-of-concept application, we use this method to highlight several direct or indirect substrate phosphorylation sites for PTPRJ, also known as DEP1, and show their quantitative phosphorylation in the context of wild-type PTPRJ compared to a mutant form of PTPRJ with increased activity, in EGF-stimulated cells. This method is generally applicable to define the signaling network effects of each RPTP in cells or tissues under different physiological conditions. Key words Signaling networks, Tyrosine phosphorylation, Phosphatase activity

1

Introduction Protein phosphorylation is a reversible post-translational modification that is critical for the regulation of cellular signaling networks and cellular response to intra- and extra-cellular perturbations. Phosphorylation of given sites is mediated by competing activities of kinases and phosphatases that can phosphorylate and dephosphorylate proteins, respectively [1–4]. In response to extracellular signals, receptor tyrosine kinases (RTKs) such as the epidermal growth factor receptor (EGFR), or insulin receptor, among many others, become activated, auto/cross-phosphorylate, and drive signaling networks leading to cell response. These signaling networks are controlled, in part, by receptor protein tyrosine phosphatases

Damien The´venin and Jo¨rg P. Mu¨ller (eds.), Protein Tyrosine Phosphatases: Methods and Protocols, Methods in Molecular Biology, vol. 2743, https://doi.org/10.1007/978-1-0716-3569-8_10, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

153

154

Jacqueline Gerritsen et al.

(RPTPs), several of which have been shown to dephosphorylate RTKs and proximal adaptor proteins. RPTPs are a subtype of phosphatases that span the cell membrane and are active as monomers, and dimerization of RPTPs leads to inactivity due to lack of binding ability to their substrates [5, 6]. PTPRJ, also known as DEP1, has been shown to regulate the C-terminal phosphorylation sites on EGFR, an established oncogenic RTK [7]. As EGFR is a frequent driver of cancers, PTPRJ and other RPTPs have gained interest as potential therapeutic targets to indirectly regulate EGFR activity [8]. To increase the activity of PTPRJ and further decrease EGFR activity, previous work has identified several transmembrane domain mutations preventing dimerization and hence disfavoring the inactive conformation of PTPRJ [9]. Mutation of G983L was particularly effective at shifting equilibrium toward active monomer rather than inactive dimer [10]. Moreover, transmembrane peptides have been used to block dimerization and thereby increase the activity of the enzyme, with the ultimate goal of controlling RTK signaling [11]. For PTPRJ and other RPTPs, establishing the range of direct and indirect substrates is important for defining the therapeutic implications of activating or inhibiting these enzymes. Advances in enrichment strategies and instrumentation in the field of phosphoproteomics now enable the study of tyrosine-phosphorylation signaling networks under a variety of cellular contexts, allowing for quantification of specific tyrosine phosphorylation sites during activation or inhibition of RPTPs, thus defining their impact on RTKs and downstream signaling. In this chapter, we describe a method by which protein tyrosine phosphorylation levels may be quantified in a multiplexed analysis with site-specific resolution to gain biological insight regarding the mechanisms and signaling axes connecting RPTPs and RTKs, using PTPRJ and EGFR as specific examples. In this method, biological samples are lysed by a denaturant, and proteins are enzymatically digested. The resulting peptides are subsequently labeled with a stable isotope-coded reagent (TMT) to enable the quantification of peptides across multiple biological samples in a single analysis. Following the mixing of labeled peptide samples, tyrosine phosphorylated peptides are immunoprecipitated with a pan-specific phosphotyrosine antibody, and phosphopeptides are further enriched through immobilized metal affinity chromatography using Fe-NTA. Samples are passed through a reverse-phase column and analyzed by liquid chromatography–tandem mass spectrometry (LC–MS/MS). The use of this method highlights selected sites on EGFR and downstream signaling components that are altered by PTPRJ expression and activity (Fig. 1). These data implicate proteins and phosphorylation sites as direct or indirect substrates of

System-Level Analysis of the Effects of RPTPs on Cellular Signaling Networks

155

7

Log2FC over EV-no EGF

6 5 4 3 2 1 0

EGFR -Y1148 EGFR -Y1173

MAPK1

MAPK3

SHB

HSP90AAI

7

Log2FC over EV-no EGF

6 5 4 3 2 1 0

AHNAK

KRT19

7

Log2FC over EV-no EGF

6 5 4

EV +EGF WT +EGF

3

M +EGF

2 1 0

ANXA2

EPB41L2

PLCG1

Fig. 1 Direct and indirect targets of PTPRJ revealed by pTyr phosphoproteomics. SCC2 cells were transfected to express empty vector (EV), wild-type PTPRJ (WT), or the G983L dimerization-disfavored mutant form of PTPRJ (M). Cells were stimulated with DMSO or 2 nM EGF for 1 min, then processed for LC–MS/MS) analysis using the described protocol. Selected phosphorylation sites were extracted to highlight the effects of PTPRJ mutation on the EGFR pathway. Data are represented as Log2 fold-change of EGF-treated compared to DMSO control for each cell line (EV, WT, or M)

156

Jacqueline Gerritsen et al.

PTPRJ. Although we illustrate this method here by analyzing the temporal dynamics of tyrosine phosphorylation following EGF stimulation in the presence or absence of activating PTPRJ mutations, the general approach is readily applicable to a broad variety of systems.

2

Materials

2.1 Cell Culture and Lysis

1. DMEM supplemented with 100 U/mL penicillin, 100 μg/mL streptomycin, and 1% (v/v) fetal bovine serum. SCC2 cells are cultured at 37 °C in a 5% CO2 incubator. 2. Serum-free DMEM medium, supplemented with the same components as described above minus fetal bovine serum. 3. Phosphate-buffered saline (PBS). 4. 0.05% Trypsin. 5. Lysis buffer consisting of 8 M urea. 6. Cell lifters. 7. BCA Protein Assay Reagent Kit.

2.2 Protein Digestion and Sample Preparation

1. Dithiothreitol 2. Iodoacetamide (IAc) (light sensitive) 3. Sequencing grade-modified trypsin 4. Trypsin digestion buffer: 100 mM ammonium acetate, pH 8.9 5. C18 Sep-Pak Plus cartridges (Waters, Milford, MA) 6. Glacial acetic acid 7. 0.1% acetic acid solution 8. 0.1% acetic acid and 25% acetonitrile solution 9. 0.1% acetic acid and 40% acetonitrile solution 10. 0.1% acetic acid and 90% acetonitrile solution

2.3 Tandem Mass Tag (TMT) Labeling and Phosphotyrosine Peptide Immunoprecipitation

1. 50 mM HEPES, pH 8.5 in HPLC (high-peformance liquid chromatography) grade water. 2. Anhydrous acetonitrile. 3. 25% Acetonitrile and 0.1% acetic acid. 4. 5% Hydroxylamine: 50%, dilute 10× in HPLC water. 5. TMT 6-plex kit (Thermo Fisher Scientific). 6. Protein G Agarose Beads (Calbiochem IP08). 7. Immunoprecipitation (IP) buffer: 100 mM Tris–HCl, pH 7.4, 100 mM NaCl, and 0.3% NP40. 8. 0.5 M Tris–HCl buffer, pH 8.5.

System-Level Analysis of the Effects of RPTPs on Cellular Signaling Networks

157

9. Rinse buffer: 100 mM Tris–HCl, pH 7.4, and 100 mM NaCl. 10. Elution buffer: 0.2% trifluoroacetic acid (TFA). 11. Pan-specific anti-phosphotyrosine 4G10 antibody and store at 4 °C (Millipore). 12. Pan-specific anti-phosphotyrosine PT66 (Sigma). 2.4 Immobilized Metal Chelate Affinity Chromatography (IMAC) and LC–MS/MS

1. High-Select Fe-NTA columns (A32992, Thermo Fisher Scientific). 2. Tryptic-digested bovine serum albumin (BSA) for elution tube coating. 3. 0.1% acetic acid and 90% acetonitrile solution. 4. 0.1% acetic acid solution. 5. Resuspension buffer: 5% acetonitrile in 0.1% formic acid. 6. HPLC solvent A: H2O/acetic acid, 99/1 (v/v). 7. HPLC solvent B: H2O/MeCN/acetic acid, 29/70/1 (v/v). 8. Fused silica capillary (360 μm outer diameter [O.D. × 50 μm inner diameter [I.D.]), (360 μm O.D. × 100 μm I.D.), and (360 μm O.D. × 200 μm I.D.) (Polymicro Technologies, Phoenix, AZ). 9. YMC ODS-AQ 5 μm packing material (Waters).

3

Methods

3.1 Cell Culture and EGF Stimulation

1. Passage human squamous cell carcinoma SCC2 cells at 80–90% confluence with 0.05% trypsin on 10-cm culture plates. 2. For sample generation, cells were serum-depleted over the course of 3 days.

3.2 Preparation of Samples for Phosphotyrosine Peptide Immunoprecipitation

1. Following cell culture and stimulation with 2 nM agonist (EGF), place cultures on ice, wash with PBS, and lyse with 0.5 mL of lysis buffer. 2. Collect lysates into 1.5 mL aliquots and centrifuge at 200 g for 5 min. 3. Take a 20 μL aliquot from each sample to measure protein concentration using the BCA assay according to the manufacturer’s protocol. 4. Incubate sample lysates with dithiothreitol for 1 h at 56 °C to reduce disulfide bonds. 5. Incubate samples with IAc for 1 h at room temperature in the dark (IAc is sensitive to light) to alkylate free cysteines. 6. Following alkylation, dilute samples 4× by the addition of 10 mL of trypsin digestion buffer.

158

Jacqueline Gerritsen et al.

7. Digest samples overnight with 40 μg of trypsin (~1:100 trypsin:substrate ratio). 8. Terminate digestion by acidifying the solution to pH 3.0 with glacial acetic acid. 9. Pre-condition Sep-Pak Plus cartridges by sequentially flowing the following solutions, each at a rate of 2 mL/min: (i) 10 mL of 0.1% acetic acid solution; (ii) 10 mL of 0.1% acetic acid and 90% acetonitrile solution; and (iii) 10 mL of 0.1% acetic acid solution. 10. Once the Sep-Pak Plus cartridges have been conditioned, load sample lysates at a rate of 1 mL/min. 11. Rinse Sep-Pak Plus cartridges with 10 mL of 0.1% acetic acid solution (flow rate 2 mL/min). 12. Elute peptides with 10 mL of 0.1% acetic acid and 40% acetonitrile solution. 13. Concentrate eluted peptides and divide them into 200 μg aliquots. 14. Dry peptides to 100–200 μL using a vacuum centrifuge prior to overnight lyophilization. Dried aliquots can be stored at – 80 °C until needed. 3.3 TMT Labeling and Phosphotyrosine Peptide Immunoprecipitation

1. Solubilize one aliquot of labeled peptides from each sample (e.g., empty vector, wild type, mutant, treated, or untreated) with 30 μL 50 mM HEPES (pH 8.5) in HPLC grade water. To ensure complete solubilization, vortex samples until the solution is clear and then spin down for 1 min at 16,500 g. 2. Thaw a set of TMT labeling reagents (stored at –80 °C) to room temperature. 3. Each TMT vial contains 800 μg TMT, suitable for labeling two samples (see Note 1). Solubilize each of the TMT reagents with 32 μL of anhydrous acetonitrile. Vortex for 20 s until dissolved and spin down 1 min at 16,500 g (see Note 2). 4. Add 15 μL of TMT label reagent to corresponding sample (see Fig. 2 for a general workflow including labeling schematic). Vortex 15 s and centrifuge. Reaction proceeds at room temperature for 1 h on a shaker at 15 g. 5. After 1 h, add 3.2 μL 5% hydroxylamine in HPLC water to each reaction tube to quench the reaction. Incubate samples 15 min on a shaker at RT at 400 rpm. 6. Combine all TMT-labeled samples in a single tube. To remove any residual sample, wash each sample reaction tube twice with 40 μL 25% acetonitrile and 0.1% acetic acid; add the resulting solution to the tube containing the combined samples. Reduce the resulting combined sample plus wash solution to complete dryness in a vacuum centrifuge and store at –80 °C.

System-Level Analysis of the Effects of RPTPs on Cellular Signaling Networks

159

Fig. 2 General workflow along with TMT labeling scheme. SCC2 cells from each condition were lysed and processed using the described protocol. Samples resulting from each cell condition were labeled with a selected TMT channel and then mixed, enabling the identification of signaling network alterations across the different conditions in a single LC–MS/MS analysis

7. To prepare for phosphotyrosine IP, wash 60 μL of agarose beads with 300 μL IP buffer. Spin down at 3900 g for 1 min, and then remove supernatant. Add 300 μL fresh IP buffer to the beads along with 12 μg of 4G10 and 6 μg of PT66. Incubate antibody and bead mixture for 6–8 h at 4 °C. 8. Spin down the antibody and bead mixture at 3900 g, and remove the supernatant. Wash beads with 400 μL IP buffer for 5 min on a rotator at 4 °C. Spin down at 3900 g for 1 min and remove the supernatant. 9. Solubilize the TMT-labeled sample in 400 μL IP buffer using vortex.

160

Jacqueline Gerritsen et al.

10. Adjust the pH of the TMT-labeled sample to pH 7.4 with 0.5 M Tris buffer, pH 8.5. 11. Add to the bead–antibody mixture. 12. Incubate the sample and antibody mixture overnight at 4 °C while rotating. 13. Spin down antibody beads at 3900 g for 1 min. 14. Remove the supernatant and store at –80 °C. 15. Rinse antibody beads three times with 400 μL and rinse buffer at 4 °C for 5 min. 16. Elute peptides with 25 μL of 0.2% TFA for 10 min while rotating, followed by centrifugation at 200 g for 20 s. 17. Collect the supernatant containing eluted peptides. 18. Repeat the elution step. 3.4 Phosphopeptide Enrichment by Immobilized Metal Affinity Chromatography

1. Prepare Fe-NTA spin columns by centrifugation for 30 s at 200 g. 2. Flow through 2 × 200 μL binding buffer (e.g., add 200 μL binding buffer, centrifuge to flow through, and repeat). 3. Suspend Fe-NTA beads in 25 μL binding buffer and cap column with Luer plug before adding the solution containing eluted peptides. 4. Incubate at room temperature for 30 min. Gently tap the column every 5 min to mix the resin. 5. Precondition elution tube by mixing 1 μL tryptic digested BSA (1 μg/μL) with 39 μL 0.1% acetic acid. 6. Vortex, centrifuge, and speedvac down to dryness. 7. Rinse the tube with 40 μL elution buffer, 2 × 40 μL 10% acetonitrile, 0.1% acetic acid, and 40 μL 0.1% acetic acid. Remove the liquid after each wash. 8. Remove Luer plug, and centrifuge FE-NTA spin column for 30 s at 200 g. 9. Rinse Fe-NTA resin with 3 × 150 μL binding buffer followed by 200 μL HPLC grade water, and each step should be performed by adding the appropriate solution followed by centrifugation at 200 g for 30 s. 10. Discard flowthrough from each wash step. 11. Place Fe-NTA spin column in preconditioned tube, and elute the sample using 2 × 20 μL Elution buffer, with each elution centrifuged at 200 g for 30 s. 12. Vacuum centrifuge sample to dryness. 13. Resuspend the sample in 10 μL of 5% acetonitrile in 0.1% formic acid.

System-Level Analysis of the Effects of RPTPs on Cellular Signaling Networks

3.5 Liquid Chromatography– Tandem Mass Spectrometry Analysis (LC-MS/MS)

161

1. Analyze the peptides eluted from Fe-NTA IMAC spin columns by LC-MS/MS using reverse-phase chromatography performed in line with an Exploris 480 quadrupole-Orbitrap mass spectrometer. 2. Load sample onto reverse-phase analytical column (see Note 3) using pressure-based column packing device or autosampler. 3. Elute peptides from the analytical column using an organic gradient from aqueous to organic (see Note 4). 4. Acquire data using the mass spectrometer in data-dependent acquisition mode using dynamic exclusion (see Notes 5 and 6). 5. To assess the consistency of peptide loading across the multiplexed samples, perform data-dependent LC-MS/MS analysis of an aliquot of the peptide IP supernatant sample. (i) Dilute 1 μL of the phosphotyrosine peptide IP supernatant 1000× and load 5 μL of the resulting diluted sample onto a fused silica capillary precolumn. (ii) Acquire LC-MS/MS data using the above settings. (iii) Chromatographic gradient can be altered to decrease analysis time (e.g., by decreasing the gradient run time) or depth of coverage (e.g., by increasing the gradient run time).

3.6 Data Analysis and Validation

1. Extract MS/MS spectra and search against human protein database (NCBI) using MASCOT, Sequest, or other search engines as desired. Set enzyme choice to trypsin; set fixed modifications to cysteine carbamidomethylation, TMT-labeled lysine, and TMT-labeled peptide N-termini. Set dynamic modifications to allow for methionine oxidation and phosphorylation of serine, threonine, and tyrosine residues. Set mass tolerance for database search as appropriate depending on instrument choice and calibration. Quantify TMT reporters using Proteome Discoverer (Thermo Fisher Scientific), which performs isotope corrections as needed. 2. To remove false positive, poor quality peptide assignments, and peptides with low signal intensity reporter ions, filter peptide spectrum matches. We tend to use the following parameters: rank = 1, mascot ion score >15, isolation interference 1000. 3. Manually validate phosphotyrosine-containing peptides identified from the database search by confirming the assignment of y-, b-, and a-type ions, as well as neutral loss (of H2O (typically from S, T, D, E residues), NH3 (typically from Q, N residues), H3PO4 (phosphoserine- or phosphothreonine-containing fragments), or HPO3 (from phosphotyrosine containing fragments)). Accept peptide sequence assignments only when all major peaks in the MS/MS spectra can be assigned.

162

Jacqueline Gerritsen et al.

4. To correct for sample labeling and mixing, search MS/MS data from the LC–MS/MS analysis of the supernatant using the same parameters as described in the previous step. Select non-phosphorylated peptide hits from abundant proteins for calculating relative protein loading for each TMT channel. 5. Calculate mean phosphorylation, standard deviation, and pvalues to estimate statistical significance for differential phosphorylation between different experimental conditions using Excel, Matlab, and Graphpad Prism.

4

Notes 1. Each tube of TMT labeling reagent can label up to 200 μg of peptide. Incomplete labeling could occur if an excess amount of peptide is used. The sample amount should be kept fairly constant for labeling with each of the TMT isoforms. 2. TMT rapidly degrades in the presence of water due to reaction with hydroxyl ions; hence, it is crucial to work fast when resuspending TMT for addition to the sample for labeling to optimize labeling efficiency. 3. Analytical HPLC columns are generated in-house and contain an integrated ESI emitter tip. ESI emitter tips are generated on these columns using a Sutter P-2000 laser puller and typically range from 1 to 2 μm, providing optimal flow rates at 20–50 nL/min. 4. The composition and timing of the organic gradient can be modified to best suit the experiment. For pTyr phosphoproteomics analyses, we tend to use the following gradient (A = 0.1% acetic acid and B = 70% acetonitrile in 0.1% acetic acid): 10 min from 0% to 15% buffer B (starting at 100% buffer A), 75 min from 15% to 40% buffer B, and 15 min from 40% to 70% buffer B with a column flow rate of approximately 50–100 nL/min. 5. Dynamic exclusion can improve data quality by reducing the number of times a given mass-to-charge ratio is selected for MS/MS within a given time period. Dynamic exclusion time and number of MS/MS repeats should be set according to chromatographic peak widths to allow for one to four MS/MS spectra for each ion species. 6. The number of MS/MS scans per cycle should be adjusted based on chromatographic peak widths and sample complexity, but commonly a “top 15” method (e.g., 15 MS/MS spectra per full scan mass spectrum) will work well.

System-Level Analysis of the Effects of RPTPs on Cellular Signaling Networks

163

Acknowledgments This work was supported in part by grants from the National Institutes of Health, including U01CA238720 and R01GM139998, as well as support from the Center for Precision Cancer Medicine in the Koch Institute for Integrative Research at MIT. References 1. Bixby JL (2001) Ligands and signaling through receptor-type tyrosine phosphatases. IUBMB Life 51:157–163. https://doi.org/ 10.1080/152165401753544223 2. Barr AJ, Ugochukwu E, Lee WH et al (2009) Large-scale structural analysis of the classical human protein tyrosine phosphatome. Cell 136:352–363. https://doi.org/10.1016/J. CELL.2008.11.038 3. Tonks NK (2013) Protein tyrosine phosphatases - from housekeeping enzymes to master regulators of signal transduction. FEBS J 280: 346–378. https://doi.org/10.1111/FEBS. 12077 4. Tonks NK (2006) Protein tyrosine phosphatases: from genes, to function, to disease. Nat Rev Mol Cell Biol 7:833–846. https://doi. org/10.1038/NRM2039 5. Nunes-Xavier CE, Martı´n-Pe´rez J, Elson A, Pulido R (2013) Protein tyrosine phosphatases as novel targets in breast cancer therapy. Biochim Biophys Acta Rev Cancer 1836:211–226. https://doi.org/10.1016/J.BBCAN.2013. 06.001 ¨ stman A, Hellberg C, Bo¨hmer FD (2006) 6. O Protein-tyrosine phosphatases and cancer. Nat Rev Cancer 6:307–320. https://doi.org/10. 1038/NRC1837

7. Tarcic G, Boguslavsky SK, Wakim J et al (2009) An unbiased screen identifies DEP-1 tumor suppressor as a phosphatase controlling EGFR endocytosis. Curr Biol 19:1788–1798. https://doi.org/10.1016/J.CUB.2009. 09.048 8. Xu Y, Tan LJ, Grachtchouk V et al (2005) Receptor-type protein-tyrosine phosphatase-κ regulates epidermal growth factor receptor function. J Biol Chem 280:42694–42700. https://doi.org/10.1074/jbc.M507722200 9. Bilwes AM, Den Hertog J, Hunter T, Noel JP (1996) Structural basis for inhibition of receptor protein-tyrosine phosphatase-α by dimerization. Nature 382:555–559. https://doi. org/10.1038/382555A0 10. Bloch E, Sikorski EL, Pontoriero D et al (2019) Disrupting the transmembrane domain-mediated oligomerization of protein tyrosine phosphatase receptor J inhibits EGFR-driven cancer cell phenotypes. J Biol Chem 294:18796–18806. https://doi.org/ 10.1074/JBC.RA119.010229 11. Bennasroune A, Fickova M, Gardin A et al (2004) Transmembrane peptides as inhibitors of ErbB receptor signaling. Mol Biol Cell 15: 3464–3474. https://doi.org/10.1091/MBC. E03-10-0753

Chapter 11 Detection of Protein Tyrosine Phosphatase Interacting Partners by Mass Spectrometry Martina Samiotaki, George Panayotou, and Panagiotis Chandris Abstract Unraveling interacting partners of protein tyrosine (Tyr) phosphatases is considered a key aspect in resolving the regulation of signaling cascades either in a pathological or in developmental context. Mass spectrometry (MS)-based protein identification has emerged as the major approach in this arena, complemented by the development of novel biochemical methodologies for sample preparation. In this chapter, we highlight two methods that, combined with mass spectrometry, may help the investigator create an interactome map for the phosphatase of interest within a specific biological context. Key words Tyrosine phosphatases, Mass spectrometry, Interactome, Proximity labeling, APEX2, Protein–protein interactions

1

Introduction The specificity toward substrates displayed by most protein tyrosine phosphatases within a cellular context is widely attributed not to their intrinsic enzymatic specificity, which is rather weak in vitro, but to their interactions with distinct regulatory subunits that modulate their activity and specificity. Scaffold proteins may also serve this purpose by assembling complexes of phosphatases with their substrates and other regulatory proteins. Moreover, various adaptor proteins may link phosphatases to other components of a signaling cascade, for example, through SH2 or PTB domain interactions. Interactions with substrates can be direct or indirect, often regulated by other signaling components. Therefore, identifying these crucial interactions between protein tyrosine phosphatases and other proteins is of paramount significance for understanding the function of these enzymes in health and disease. Concerted efforts toward that goal have been supported significantly by the advent of contemporary mass spectrometry–based methodologies for protein identification. Mass spectrometry has

Damien The´venin and Jo¨rg P. Mu¨ller (eds.), Protein Tyrosine Phosphatases: Methods and Protocols, Methods in Molecular Biology, vol. 2743, https://doi.org/10.1007/978-1-0716-3569-8_11, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

165

166

Martina Samiotaki et al.

been used extensively to identify protein–protein interactions, either after direct isolation of a native complex (e.g., immunoprecipitation) or through the use of cell-permeable cross-linkers to ensure the integrity of the complex through the purification steps. The latter approach has many advantages, such as a plurality of the reactive agents that can be used (thiol, amine, carboxyl reactive, etc.), along with the ability to perform quantitative peptide analysis by using isotope-labeled spacers in the structure of the cross-linker and also to infer information regarding protein targets of natural compounds [1–4]. Furthermore, the use of cross-linkers may be exploited to differentiate between interacting and noninteracting domains of a protein and to infer structural information. Although these methods are very useful regarding the identification of interacting partners and give structural insight, they do require more sophisticated data processing, which includes deconvolution and advanced structural analysis. Plain use of cell-permeable cross-linkers and pull down of a protein of interest either with an antibody or with a tag-specific bait (in case of exogenously added protein) may also be used to gain insight into a protein’s interactome. It should be kept in mind, though, that cross-linkers act in a nonspecific manner, and labeling requires a rather extended period of incubation (usually more than 30 min), which is likely to pose additional stress to the cells and yield artificial interactions. In this chapter, among a plethora of different available approaches, we focus on two major methods that can be used to characterize potential tyrosine phosphatase interacting partners. Given that dual specificity phosphatases (DUSPs) also act upon tyrosine residues, some additional tips will be included that may help distinguish between their serine/threonine (Ser/Thr) and Tyr phosphatase activity that may guide interaction with partner proteins. The methods described below are complementary to each other: The first one is a classic immunoprecipitation approach, and the second involves proximity labeling (PL) of an exogenously added phosphatase. In both the cases, the identification of potential interacting partners is accomplished using liquid chromatography (LC) followed by mass spectrometry and appropriate data analysis. 1.1 Immunoprecipitation

Immunoprecipitation is the classic approach for detecting protein– protein interactions and remains a popular method, being simpler than the one described in the next section. It can be applied to the endogenous protein of interest without overexpressing an engineered chimera. It should be kept in mind, though, that phosphatases may be expressed at very low, almost undetectable levels under steady-state conditions or within the context of a disease, and under these conditions, immunoprecipitation results should be interpreted with caution. A critical parameter in this approach is the specificity of the antibody that is used to bring down the protein of interest and also its compatibility with the lysis method. Although,

Detection of Protein Tyrosine Phosphatase Interacting Partners by Mass. . .

167

in most cases, the detergents included in the lysis solution are nonpolar and presumably have minimal effect on the conformation of the protein of interest, it is hard to predict antibody specificity but also to exclude protein partners that will coimmunoprecipitate with the bait either as a result of nonspecific interactions or as part of a larger supramolecular complex, within which the protein of interest might not make direct contacts. The proteins that have been pulled down can be further analyzed either by classic methods such as immunoblotting, aiming at specific proteins, or by mass spectrometry to characterize the full spectrum of interactors. 1.2 ProximityDependent Labeling with an Engineered Ascorbic Acid Peroxidase 2 (APEX2)

To assess direct contact at higher precision, complementary approaches have been developed that rely on introducing a chimera of the protein of interest tagged with an enzyme capable of modifying proximal protein targets. The method with all its variations is called proximity labeling (PL), and the enzyme chosen to escort the protein of interest defines, to a great extent, spatial and temporal resolution [5–10]. Proximity labeling has been used extensively to map protein–protein interactions, and the toolkit is rather broad. One critical parameter, though, is the incubation time needed to monitor biochemical events using interactome in “real-time.” Various enzymes serve this purpose, with a common feature being the biotinylation of proximal proteins. The labeling efficiency can be tested by classic methods such as protein electrophoresis and detection of biotinylated proteins using avidin- or streptavidin-based detection schemes. One may also use immunofluorescence approaches to spot the localization of biotinylated proteins, albeit cross-reactivity with endogenous biotinylated proteins might obscure subcellular compartment specificity. Ultimately, the interacting (or proximal) partners of the protein of interest can be analyzed downstream by mass spectrometry. In this chapter, we describe the use of an ascorbic acid peroxidase 2 (APEX2) that, when supplied with biotin-phenol in the presence of hydrogen peroxide (H2O2), converts biotin-phenol into short-lived biotin phenoxyl radicals, resulting in the biotinylation of proximal electron-rich amino acids. Proximal targets are within a range of 10–20 nm. Biotinylated proteins can then be pulled down using streptavidin bound to magnetic beads and processed for mass spectrometry. One important issue for consideration when isolating protein phosphatase complexes is the utilization of phosphatase inhibitors during cell lysis. While their presence is desirable to detect interactions involving phosphorylated proteins (e.g., SH2–pTyr interactions), it may interfere with interactions that require the phosphatase bait in an enzymatically active state. Therefore, comparing results with and without phosphatase inhibitors and using catalytically inactive mutants, if available, may be necessary to detect all relevant interactions.

168

2

Martina Samiotaki et al.

Materials

2.1 Materials for Cell Culture, Protein Isolation, and Fluorescent Labeling

1. 10 cm sterile tissue culture dish. 2. Tissue culture treated sterile 12-well plates. 3. An easy-to-transfect cell line (e.g., HEK 293, human embryonic kidney cells) that has been used extensively as a platform for this type of experiment. 4. Dulbecco’s Modified Eagle Medium (DMEM) cell culture medium (high glucose), supplemented with pyruvate and glutamine. 5. Fetal bovine serum (FBS). Add up to 10% in cell culture medium. 6. Tissue culture incubator with temperature maintained at 37 ° C, supplied with 5% carbon dioxide (CO2). 7. Competent bacterial strains suitable for plasmid deoxyribonucleic acid (DNA) replication and isolation (such as DH5α). 8. 10 cm dishes with Luria–Bertani medium OR Lysogeny Broth (LB) agar (LB plus 15 g agar per liter, autoclaved) with a suitable antibiotic for the selection of transformed bacteria. 9. Erlenmeyer flasks for bacterial growth in liquid media. 10. High-purity plasmid isolation kit (such as the PureLink™ HiPure Plasmid from Thermo or another similar system). 11. Shaker with temperature regulation suitable for growing liquid bacterial cultures. 12. Autoclaved LB medium for bacterial growth (10 g tryptone, 10 g sodium chloride [NaCl], and 5 g yeast extract in 1 L of distilled water). 13. Refrigerated benchtop centrifuge that can accommodate 1.5 mL tubes with lid. 14. Tube revolver rotator that can accommodate the 1.5 mL tubes. 15. Nonidet P-40 (or its substituent IGEPAL CA-630) or TX-100 (see Note 1). 16. Complete protease inhibitors with ethylenediaminetetraacetic acid (EDTA). 17. Complete protease inhibitors without EDTA for APEX2 labeling and streptavidin pull-down. 18. Serine/threonine phosphatase Inhibitor cocktail that contains cantharidine, (-)-p-bromolevamisole oxalate, and calyculin A. 19. 0.1 M sodium orthovanadate solution (pH-regulated and boiled). 20. Complete phosphatase inhibitor cocktail (for both Ser/Thr and tyrosine phosphatases) (100×).

Detection of Protein Tyrosine Phosphatase Interacting Partners by Mass. . .

169

21. Transfection reagent suitable for the target cell line. 22. Biotin-phenol (biotin tyramide). 23. Hydrogen peroxide (H2O2) solution. 24. Protein A/G magnetic beads or protein A/G agarose beads. 25. Streptavidin magnetic beads. 26. Cold 1× phosphate-buffered saline (PBS) buffer: 137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, and 1.8 mM KH2PO4. 27. Lysis buffer (for immunoprecipitation): 50 mM Tris–HCl, pH 7.4, 150 mM NaCl, plus complete protease inhibitors, including 1 mM EDTA containing 0.5–1% IGEPAL 630 (or TX-100) (see also Note 1). A complete set of (1×) phosphatase inhibitors should also be included. The detergent (s) included in the buffer and the final concentration may vary as described below in Subheading 3.2 (see Note 1). 28. Modified radioimmunoprecipitation assay (RIPA) buffer (for APEX2 biotinylation studies): 50 mM Tris–HCl, pH 7.4, 150 mM NaCl, plus complete protease inhibitors (without EDTA), phosphatase inhibitors, (6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid) (Trolox) (5 mM), L-ascorbate (10 mM) and sodium azide (100 μM), and 0.5–1% Nonidet P-40 (or its substituent IGEPAL CA-630) or TX-100 (see Note 1). 29. (+/-)-6-Hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox). 30. Stock saline sodium citrate (SSC) buffer (20×) (sodium chloride 3 M, sodium citrate 0.3 M, pH 7). 31. Sodium L-ascorbate. 32. Streptavidin Alexa Fluor 568. You may use any other desirable excitation wavelength (e.g., 488 and 647), accompanied by a suitable excitation light source and detection filter sets. 33. 3% H2O2 solution. 34. Ampules of high-grade aqueous paraformaldehyde solution (32% or 16%). 35. APEX2 biotinylation rinsing buffer: PBS supplemented with 5 mM Trolox, 10 mM L-ascorbate, and 100 μM sodium azide (NaN3). 36. Permeabilization buffer for APEX2 biotinylation labeling: PBS, 2% bovine serum albumin (BSA), plus L-ascorbate 10 mM, Trolox 5 mM, and NaN3 100 μM plus 0.2% TX-100. 37. Blocking buffer for APEX2 biotinylation labeling: 10% FBS plus L-ascorbate 10 mM, Trolox 5 mM, and NaN3 100 μM in (1×) PBS.

170

Martina Samiotaki et al.

38. Streptavidin dilution buffer for APEX2 biotinylation labeling: (1×) PBS, 2% BSA, plus L-ascorbate 10 mM, Trolox 5 mM, and NaN3 100 μM. 39. Liquid mountant to limit photobleaching (commercially available or homemade). 40. Coverslips (No. 1.5). 2.2 Materials for Sample Preparation for Mass Spectrometry

1. Sample lysis buffer (based on Laemmli buffer): Tris–HCl, pH 6.8, sodium dodecyl sulfate (SDS) 4%, dithiothreitol (DTT) 1 mM, and glycerol 10%. 2. Magnetic beads—carboxylate functionalized beads with a hydrophilic surface (e.g., Sera-Mag Speed Bead; Cytiva, Catalogue Numbers: 45152105050350 and 65,152,105,050,350) freshly prepared. 3. C18 analytical column (25 cm × 75 μm internal diameter (ID), particle size 1.9 μm, 120 Å, PepSep Bruker). 4. NanoSpray emitters suitable for stable electrospray ionization. 5. Magnetic rack for 1.5 mL tubes. 6. 96-well V-bottom plates. 7. 1.5 mL Eppendorf tubes, protein LoBind. 8. Ethanol 100%. 9. 100 mM DTT. 10. 20 mM Iodoacetamide solution (freshly prepared). 11. Freshly made 200 mM ammonium bicarbonate solution. The pH should be around 7.8. 12. Trypsin or trypsin/LysC mix, MS grade, diluted in 100 mM ammonium bicarbonate just before adding to the sample. 13. Buffer A: 2% (v/v) acetonitrile (ACN) in 0.1% (v/v) formic acid (both liquid chromatography–mass spectrometry [LC-MS] grade). 14. Buffer B: 80% (v/v) acetonitrile (ACN) in 0.1% (v/v) formic acid (both LC-MS grade). 15. Water, LC-MS grade. 16. Autosampler polypropylene vial with 300 μL insert with Snap caps. 17. Nitrogen gas (N50). 18. Sonicating water bath. 19. Magnetic rack suitable for 1.5 mL tubes. 20. Heating and shaking device. Type: Eppendorf Thermomixer. 21. SpeedVac concentrator. instrument 22. NanoDrop determination.

for

protein

concentration

Detection of Protein Tyrosine Phosphatase Interacting Partners by Mass. . .

171

23. Nano LC-MS/MS instrument suitable for proteomics. 24. DIA-NN software 1.8.1 (freely available). 25. Perseus (freely available). 26. Spectronaut software (Biognosys).

3

Methods

3.1 Preparation of Antibody–Beads Coupling

1. You may have the antibody precomplexed with protein A/G beads as follows: Initially, take the recommended from the manufacturer volume of protein A/G slurry and transfer it to a 1.5 mL polypropylene tube and pull down beads either by a small magnet placed on the side of the tube or by mild centrifugation on a small benchtop centrifuge at 400× g. 2. Remove the supernatant and add 200 μL of the lysis buffer. 3. Add 1–2 μg of the primary antibody (see Note 2). 4. Incubate on a rotator for 1–2 h at room temperature (RT). 5. Pull down the bead–antibody complex as above. Remove the supernatant and add 50 μL lysis buffer. Normally, the beads are in excess, so the beads should have captured all the antibodies. However, as a precaution, the supernatant is removed to eliminate unbound antibodies that will interfere with the downstream process.

3.2 Immunoprecipitation

1. Place the tissue culture plates with the cells on ice. 2. Remove tissue culture medium, and rinse twice with ice-cold (1×) PBS before harvesting. 3. Aspirate PBS and add 1 mL lysis solution (see Note 3). 4. Scrape cells and transfer them to 1.5 mL polypropylene tubes. 5. Incubate on ice for 20 min with occasional tapping. 6. Spin down in a refrigerated centrifuge (4 °C) at 13,000 g for 15 min. 7. Take the supernatant and add the 50 μL slurry of magnetic beads with the immunoglobulin G (IgG) bound on them. Mix gently with a blue tip and place on the rotator in the cold room. 8. Rotate at a regular speed for 3 h. 9. Stop the rotation/mixing and pull down beads, preferably using a small magnet or low-speed centrifugation (800× g). 10. Remove the supernatant and rinse beads with 500 μL lysis buffer. Repeat the process three times. 11. Pull down the beads and proceed with the elution using a competitive peptide or Laemmli buffer (see Note 4).

172

Martina Samiotaki et al.

3.3 Proximity Labeling Using Ascorbic Acid Peroxidase 2 (APEX2)

1. Transform competent bacterial strains (such as DH5α) with the phosphatase construct of interest (see Notes 5–8). Incubate at 37 °C without antibiotic so that the resistance gene will be expressed. Following the 1-h incubation, spin down bacteria at 4000× g, remove most of the supernatant, resuspend the pellet, streak in suitable LB agar plates with the appropriate antibiotic for selection, and grow overnight at 37 °C unless otherwise specified. 2. The next day, isolate a single colony and inoculate a 150 mL liquid culture with the appropriate antibiotic. Grow overnight with constant shaking at 37 °C unless otherwise specified. Plasmids may be isolated using commercially available kits that yield high-purity plasmids. This protocol is for transient transfections. An alternative approach involves using stable cell lines or lentiviral particles (see Notes 5 and 13–16). 3. Plate HEK293 cells and grow in a 10 cm tissue culture dish for 48 h to reach 80% confluency on the day of transfection using the plasmid of interest (see Notes 6 and 13). 4. An hour before transfection, you may change the medium with fresh DMEM plus 5% FBS. Do not use antibiotics, as this will result in excessive cell death (see also Note 13). 5. Use only the highest purity plasmid for transfection experiments; otherwise, transfection efficiency and cell viability will be severely compromised. 6. For each 10 cm tissue culture dish, prepare the DNA liposomemix: In 300 μL plain DMEM (without serum), add 2 μg of DNA and then mix gently with a 1000 μL pipette. Gently tap the tube with the TurboFect reagent and add 12 μL of the lipids in the DMEM/plasmid solution. Mix three to four times gently using large tips (1000 μL pipette). Let the mixture stand at room temperature for 20 min. Avoid mixing up and down and vortexing the mixture. 7. After 20 min, with the gentle use of a 1000 μL pipette, pick up the DNA/lipid solution and spread it over the medium of the 10 cm dish dropwise. Return the cells to the incubator and let them stand for 24–48 h. 8. On the day of harvesting, add biotin-phenol to the cell culture medium for 30 min at a concentration of 0.5–2.5 mM. For cells that have reached confluency, a lower concentration of biotinphenol is needed (see Note 9). 9. After the 30-min incubation, add H2O2 to the cells at a concentration of 0.1–0.5 mM. Return the cells to the incubator and incubate for 1 min (see Note 10).

Detection of Protein Tyrosine Phosphatase Interacting Partners by Mass. . .

173

10. Take the cells out of the incubator and quench the reaction by adding Trolox and sodium L-ascorbate to a final concentration of 5 and 10 mM and sodium azide (NaN3) at 100 μM. 11. Aspirate medium and carefully rinse once with ice-cold APEX2 biotinylation rinsing buffer. Spin down cells at 300× g using a refrigerated benchtop centrifuge. 12. Lyse cells in ice-cold modified RIPA buffer. 13. Lyse for 30 min on ice with occasional tapping of the tube. 14. Centrifuge at 12000× g. Collect supernatant, add streptavidin beads (20 μL of slurry), and incubate on a rotating wheel in the cold room for 2–3 h. Alternatively, you may proceed at room temperature for 30–45 minutes. However, this might increase proteolysis. 15. Bring down the beads, preferably with a magnet. This will reduce the nonspecific trapping of foreign material. Alternatively, you may bring down the beads by centrifuging at 800× g. 16. Wash the beads two to three times with lysis buffer to remove as much as possible of nonspecific binders. 17. Either elute the bound material by adding 10 mM d-biotin or boil it in lysis buffer (Laemmli type, see Subheading 2) and proceed for biochemical analysis either by immunoblotting or by mass spectrometry (see Notes 11, 12, 17, and 18). 3.4 Fluorescent Detection of APEX2 Labeling

1. Plate HEK293 cells as above in sterilized round-glass coverslips in 12-well plates. 2. Culture for 48 h. 3. Transfect with the APEX2-tagged construct(s) using 250 ng of plasmid DNA per well diluted in 100 μL of plain warm DMEM. 4. Add 3–4 μL of TurboFect and again use a 1000 μL pipette to mix the solution three to four times. Do not vortex. 5. Incubate at room temperature for 20 min. 6. Bring the cells to the hood and add the liposome/DNA mixture to the cells dropwise using a 1000 μL pipette. 7. Bring the cells back to the incubator for another 24–48 h. 8. Label the cells exactly as described in the previous section. 9. Rinse mildly with APEX2 biotinylation rinsing buffer. 10. Rinse once mildly with PBS. 11. Fix cells using formaldehyde (4%) in PBS (1× final concentration without any scavenger) for 5–10 min. 12. Remove the fixative and rinse with (1×) PBS.

174

Martina Samiotaki et al.

13. Perforate plasma membrane in permeabilization buffer for 5 min. 14. Block in APEX2 biotinylation blocking buffer for at least an hour (see Notes 19–23). 15. Label with fluorescent streptavidin diluted at the recommended concentration in streptavidin dilution buffer (see Note 24) for 1 h. 16. Rinse twice with PBS and mount using antifade mountant. 3.5 Sample Preparation for Mass Spectrometry

Here we describe the identification and quantification of proteins present in the samples of phosphatase interactomes using dataindependent acquisition (DIA) as the mass spectrometry acquisition method [11, 12]. 1. Have your sample harvested before proteomics analysis in Laemmli-based lysis buffer and heat for 3 min at 99 °C. 2. Sonicate in a water bath for 15 min. 3. Centrifugate for 10 min at 13000× g in a benchtop centrifuge accommodating 1.5 mL tubes with lid and use supernatant for proteomic analysis. 4. Rinse a 1:1 mixture of two different types of Sera-Mag magnetic beads twice with water using a magnetic rack before processing. More specifically, pool 50 μL from each stock solution (50 μg/μL) in a tube and place on the magnetic rack for 2 min. 5. Carefully remove the supernatant and rinse the beads with 200 μL water three times off-rack (water is added to the beads off the magnetic rack, and after gentle mixing, the supernatant is removed after placement on the rack). 6. After the final wash, resuspend the beads in 500 μL of water at a final concentration of 10 μg beads/μL water. 7. Label two sets of Eppendorf tubes for all samples, one for protein digestion and peptide extraction and the other for storing the extracted peptides.

3.6 Cysteine Carbamidomethylation

1. Mix the samples with an equal volume of the iodoacetamide solution at a final concentration of 100 mM. 2. Incubate the mixture for 30 min at room temperature (RT) in total darkness.

3.7 SP3 Peptide Extraction

1. Add the magnetic beads to the protein mixtures from step 3 to achieve an estimated concentration ratio of 1:10 μg of SP3 beads/μg of protein. 2. Add an equal volume of 100% ethanol to initiate binding at a final concentration of 50% v/v ethanol.

Detection of Protein Tyrosine Phosphatase Interacting Partners by Mass. . .

175

3. Incubate the tubes for 5 min at room temperature with mixing at a speed of 1000 rpm. 4. Place the tubes on a magnetic rack and incubate for 2 min at RT. 8. Discard the supernatant and rinse the beads three times off-rack with 180 μL of 80% ethanol. 9. For the elution step, remove the tubes from the magnetic rack and resuspend the beads in 100 μL 0.1 M ammonium bicarbonate solution containing MS-grade protease (e.g., trypsin/ LysC mix) (1:50 enzyme to protein concentration). 10. Sonicate the samples briefly for 30 s in a bath sonicator. 11. Incubate overnight at 37 °C in a ThermoMixer at 1200 rpm. 12. The next day, place the tubes on a magnetic rack and recover the supernatant for further processing. 3.8 SP3-Mediated Peptide Clean-Up

1. Transfer the peptidic solutions to clean vials containing 1 mL of 100% acetonitrile and 20 μL of magnetic bead slurry and incubate for 30 min at room temperature with mild mixing. 2. Place the samples onto the magnetic stand for 2 min and carefully remove the supernatants. 3. Wash the beads twice with 200 μL of 100% acetonitrile. 4. Add 50 μL of buffer A to the beads for 5 min for peptide elution. 5. Transfer the eluates to high-performance liquid chromatography (HPLC) polypropylene autosampler vials. 6. Dry down completely in a SpeedVac concentrator.

3.9 Peptide Solubilization

1. Reconstitute the dried pellets in 10 μL of buffer A and incubate for 3 min in a sonication water bath. 2. Determine the peptide concentration of the samples by NanoDrop absorbance measurement at 280 nm (where necessary, the samples are diluted so that the absorbance readings fall within the linear range of the NanoDrop). 3. Adjust the sample volume to a maximum concentration of 1 μg/μL.

3.10 Liquid Chromatography– Mass Spectrometry

1. Place the vials in the cold autosampler inserts. 2. Directly inject a volume from each sample containing a maximum of 0.5 μg peptides into a 25 cm-long C18 column. The column is operated at room temperature and is mounted near the stainless steel emitter (zero dead volume). 3. Separate the peptides using a linear gradient of 7% to 35% buffer B in 40 min at a flow rate of 400 nL/min, followed by

176

Martina Samiotaki et al.

an increase to 45% in 5 min at a flow rate of 250 nL/min, then to 99% in 0.5 min at a flow rate of 400 nL/min, and then kept constant for equilibration for 14.5 min. 4. The eluted peptides are ionized by a NanoSpray source equipped with a nano-emitter and detected by a high-end mass spectrometer operating preferentially in a dataindependent acquisition (DIA) mode. Acquire full-scan MS spectra in an m/z range of 375–1400 using high resolving power and high accuracy. The choice of the mass spectrometer and the mode of data acquisition depend on the available options. Data Analysis

The raw files generated from the mass spectrometer are analyzed using DIA-NN software (Version 1.8.1) searched against the complete UniProt SWISS reviewed entries specific for the species under investigation [13, 14]. The search is set to library-free mode and double-pass search mode. Search parameters are strict trypsin specificity, allowing up to two missed cleavages and a minimum peptide length of seven amino acids. The main search tolerance is set to 20 ppm. Oxidation of methionines and N-terminal acetylation are set as variable modifications (common). Cysteine carbamidomethylation is set as a fixed modification.

3.12 Statistics and Data Visualization

Perseus software (Max Planck) or other statistical packages can be used for data analysis aiming at the quantitative comparison of control samples versus the bait-enriched samples. Data can be visualized as heat maps (grouped representation) or volcano plots (pairwise comparisons). Typically, volcano plots are generated using false discovery rate (FDR) of 5% and an S0 ~ 0.1.

3.11

4

Notes 1. Nonionic detergents are selected based on their critical micellar concentration (CMC)—lower values of CMC point to a more potent solubilizing agent. TX-100 has a CMC of 0.23 mM, while Nonidet P-40 and its substitute IGEPAL 630 have values in the range of 0.059–0.083 mM. When used at the higher end of concentration (0.5–1%), these detergents behave in a similar manner. An important feature is the purity of the reagent. 2. Protein A and protein G differ in their optimum pH for antibody binding. Protein A exhibits a maximum binding capacity for IgG at the pH of 8.2 and protein G at the pH of 5.2. Although magnetic beads that carry both protein A and protein G avoid the use of either protein A or protein G on a per-case basis, it should be kept in mind that optimum conditions for pull-downs are critical, especially if short incubation times are desired (30 min–1 h).

Detection of Protein Tyrosine Phosphatase Interacting Partners by Mass. . .

177

3. Despite the use of phosphatase inhibitors in the lysis buffer, it should be kept in mind that, upon cell lysis, phosphatases may act in a promiscuous manner, and incubation time should be kept as low as possible. Incubation at room temperature should be avoided. 4. Alternatively to elution by competitive peptide, the investigator may use glycine-based acidic solution (glycine solution 0.2 M, pH 2.5) or boil the beads in Laemmli buffer (Tris–HCl pH 6.8, SDS 2–4%, DTT 1 mM, glycerol 10%). The last option gives a higher yield, but it also contaminates the solution with antibodies that may interfere with some downstream applications. 5. Although, with the use of stable cell lines expressing the tagged tyrosine phosphatase of interest, it will be easier to get higher consistency between experiments, investigators may also use transient transfection, employing a cell line and a transfection reagent that can yield an efficiency of at least 40–50% to be able to perform biochemical analysis. 6. You will need to clone the phosphatase of interest in a suitable expression vector (such as pcDNA4) carrying APEX2 at the c-terminus or the n-terminus. It is advisable that both cloning strategies are followed, and the study of the behavior of phosphatase starts from its proper localization. A small tag (V5, hemagglutinin [HA], or other similar one) should be implemented between the gene of interest and APEX2 to perform immunofluorescence studies to validate the proper localization of the chimera. 7. Between the phosphatase of interest and the APEX2 tag, there should be a flexible linker sequence, usually carrying glycines and serines (such as [GGGGS]n or [G2S4]n). These are short, unstructured, and flexible sequences that also exhibit high solubility due to the polar serine [17, 18]. The linker used to tag a phosphatase of interest with APEX2 should be flexible enough; hence, it is recommended to use a sequence that carries glycine and serine residues (GGGGS motif). 8. When designing APEXX2 constructs, it should be taken into account that lengthy repetitive linker sequences ([GGGGS]n) will increase the probability of identifying pseudo-target proteins as potential substrates. 9. Further optimization might be needed for the APEX system depending on the cell line used. In general, proliferating cells take up biotin-phenol much easily compared with quiescent ones. Incubation time should be kept at 30 min [15, 16]. 10. The treatment with hydrogen peroxide should not exceed 1 min, and its concentration should be kept at a minimum for two reasons: First, it is toxic to the cells, and second, the basic idea is to label protein at proximity and in particular after a

178

Martina Samiotaki et al.

specific treatment involving the pathway of interest. Extended periods of incubation with hydrogen peroxide and high concentrations of biotin-phenol result in spurious artifacts and false-positive readouts. 11. It should be kept in mind that both immunoprecipitation and proximity labeling can help identify proteins in proximity and/or in the same supramolecular complex. The substrate of the phosphatase, though, is the tyrosine residue of an interacting protein that will be exposed to the catalytic center at the core of the phosphatase and not at the periphery. 12. In order to distinguish between interacting partners and potential substrates, it is advisable to use a catalytically inactive (dead) chimera of the phosphatase of interest, tagged in the exact same way as the active variant. This should be complemented with the use of a mutant that exhibits constitutive activity. The interactome of all three APEX2 constructs could give important insights regarding specific interactors. 13. Transient transfections are always escorted by elevated autophagy due to the use of lipids and excess DNA that the cell is trying to degrade. It is advisable to create stable cell lines with the constructs of interest. This will greatly facilitate the interpretation of results between experiments. In this case, a bicistronic vector could be used that expressed green fluorescent protein (GFP) as a marker of plasmid insertion. Doubletagging of the phosphatase with APEX2 and GFP is not recommended. 14. The plasmids used for generating stable cell lines should be of the highest purity. This will markedly improve transfection efficiency. 15. Although single-gene vectors can be used for stable cell lines, a bicistronic system that also expresses GFP as a reporter might be used. This strategy helps sort the positive cells using a flow sorter instead of the plain use of antibiotics. 16. In case there is a preference for harder-to-transfect cell lines, which the investigator chooses to use as a more appropriate biological platform for the system under study, it might be necessary to implement lentiviral vectors and raise lentiviral particles. 17. Preparing and analyzing samples with the selective omission of classes of phosphatase inhibitors for immunoprecipitation (Ser/Thr versus Tyr-specific phosphatase inhibitors) can be used to obtain information regarding the need for the presence of a phosphorylated residue at a target protein so as to interact with the tyrosine phosphatase under study. This should be analyzed in combination with the results from the APEX2 system (data from normal protein tyrosine phosphatase (PTP)

Detection of Protein Tyrosine Phosphatase Interacting Partners by Mass. . .

179

under study, the phosphatase-dead mutant, and the constitutively active mutant). The above set of experiments will help to distinguish between protein–protein interactions of the tyrosine phosphatase under study and its interactions with potential substrates. 18. Regarding the APEX2 labeling, it is advisable to include as a negative control plain APEX2 expression in the same subcellular compartment as the tyrosine phosphatase construct. 19. The abovementioned concentrations of biotin-phenol and H2O2 can be used as general guidelines and starting points. There might be a need for fine-tuning. Increasing the concentration of biotin-phenol will result in nonspecific labeling. Also, the treatment with H2O2 should be kept at a minimum to minimize its toxic effects. 20. When performing fluorescent detection of biotinylated proteins using fluorescent streptavidin, a negative control should always be included as there are endogenous biotinylated proteins. 21. To minimize nonspecific labeling by fluorescent streptavidin, one may alter the ionic strength of the buffer used for fluorescent labeling by using increasing amounts of NaCl up to 0.5 M. Alternatively, one may use the classic saline sodium citrate (SSC) buffer, 150 mM NaCl, 150 mM sodium citrate) that is used for hybridizations, up to 5× of its ionic strength (5× SSC). 22. The abovementioned concentrations of Trolox and L-ascorbate are at the highest end. One might titrate down to 0.5 to 1 mM each. 23. The addition of NaN3 is necessary to inactivate the APEX2 activity completely. 24. In case of high background, switch to SSC buffer (2–5× strength). References 1. Matzinger M, Mechtler K (2021) Cleavable cross-linkers and mass spectrometry for the ultimate task of profiling protein-protein interaction networks in vivo. J Proteome Res 20(1): 78–93. https://doi.org/10.1021/acs. jproteome.0c00583 2. Wright MH, Sieber SA (2016) Chemical proteomics approaches for identifying the cellular targets of natural products. Nat Prod Rep 33(5):681–708. https://doi.org/10.1039/ c6np00001k 3. Rappsilber J (2011) The beginning of a beautiful friendship: cross-linking/mass spectrometry and modelling of proteins and multi-

protein complexes. J Struct Biol 173(3): 530–540. https://doi.org/10.1016/j.jsb. 2010.10.014 4. Iacobucci C, Gotze M, Ihling CH, Piotrowski C, Arlt C, Schafer M, Hage C, Schmidt R, Sinz A (2018) A cross-linking/ mass spectrometry workflow based on MS-cleavable cross-linkers and the MeroX software for studying protein structures and protein-protein interactions. Nat Protoc 13(12):2864–2889. https://doi.org/10. 1038/s41596-018-0068-8 5. Nguyen TMT, Kim J, Doan TT, Lee MW, Lee M (2020) APEX proximity labeling as a

180

Martina Samiotaki et al.

versatile tool for biological research. Biochemistry 59(3):260–269. https://doi.org/10. 1021/acs.biochem.9b00791 6. Markmiller S, Soltanieh S, Server KL, Mak R, Jin W, Fang MY, Luo EC, Krach F, Yang D, Sen A, Fulzele A, Wozniak JM, Gonzalez DJ, Kankel MW, Gao FB, Bennett EJ, Lecuyer E, Yeo GW (2018) Context-dependent and disease-specific diversity in protein interactions within stress granules. Cell 172(3):590–604 e513. https://doi.org/10.1016/j.cell.2017. 12.032 7. Bosch JA, Chen CL, Perrimon N (2021) Proximity-dependent labeling methods for proteomic profiling in living cells: an update. Wiley Interdiscip Rev Dev Biol 10(1):e392. https://doi.org/10.1002/wdev.392 8. Chen CL, Perrimon N (2017) Proximitydependent labeling methods for proteomic profiling in living cells. Wiley Interdiscip Rev Dev Biol 6(4). https://doi.org/10.1002/ wdev.272 9. Yang R, Meyer AS, Droujinine IA, Udeshi ND, Hu Y, Guo J, McMahon JA, Carey DK, Xu C, Fang Q, Sha J, Qin S, Rocco D, Wohlschlegel J, Ting AY, Carr SA, Perrimon N, McMahon AP (2022) A genetic model for in vivo proximity labelling of the mammalian secretome. Open Biol 12(8):220149. https://doi.org/10. 1098/rsob.220149 10. Pfeiffer CT, Paulo JA, Gygi SP, Rockman HA (2022) Proximity labeling for investigating protein-protein interactions. Methods Cell Biol 169:237–266. https://doi.org/10. 1016/bs.mcb.2021.12.006 11. Hughes CS, Moggridge S, Muller T, Sorensen PH, Morin GB, Krijgsveld J (2019) Single-pot, solid-phase-enhanced sample preparation for proteomics experiments. Nat Protoc 14(1): 68–85. https://doi.org/10.1038/s41596018-0082-x 12. Hughes CS, Foehr S, Garfield DA, Furlong EE, Steinmetz LM, Krijgsveld J (2014)

Ultrasensitive proteome analysis using paramagnetic bead technology. Mol Syst Biol 10(10):757. https://doi.org/10.15252/msb. 20145625 13. Demichev V, Messner CB, Vernardis SI, Lilley KS, Ralser M (2020) DIA-NN: neural networks and interference correction enable deep proteome coverage in high throughput. Nat Methods 17(1):41–44. https://doi.org/10. 1038/s41592-019-0638-x 14. Demichev V, Szyrwiel L, Yu F, Teo GC, Rosenberger G, Niewienda A, Ludwig D, Decker J, Kaspar-Schoenefeld S, Lilley KS, Mulleder M, Nesvizhskii AI, Ralser M (2022) Dia-PASEF data analysis using FragPipe and DIA-NN for deep proteomics of low sample amounts. Nat Commun 13(1):3944. https:// doi.org/10.1038/s41467-022-31492-0 15. Peterson CWH, Deol KK, To M, Olzmann JA (2021) Optimized protocol for the identification of lipid droplet proteomes using proximity labeling proteomics in cultured human cells. STAR Protoc 2(2):100579. https://doi.org/ 10.1016/j.xpro.2021.100579 16. Tan B, Peng S, Yatim S, Gunaratne J, Hunziker W, Ludwig A (2020) An optimized protocol for proximity biotinylation in confluent epithelial cell cultures using the peroxidase APEX2. STAR Protoc 1(2):100074. https:// doi.org/10.1016/j.xpro.2020.100074 17. van Dongen EM, Evers TH, Dekkers LM, Meijer EW, Klomp LW, Merkx M (2007) Variation of linker length in ratiometric fluorescent sensor proteins allows rational tuning of Zn (II) affinity in the picomolar to femtomolar range. J Am Chem Soc 129(12):3494–3495. https://doi.org/10.1021/ja069105d 18. van Rosmalen M, Krom M, Merkx M (2017) Tuning the flexibility of glycine-serine linkers to allow rational design of multidomain proteins. Biochemistry 56(50):6565–6574. https://doi.org/10.1021/acs.biochem. 7b00902

Chapter 12 Detecting PTP Protein–Protein Interactions by Fluorescent Immunoprecipitation Analysis (FIPA) Natalia Kruglova and Alexander Filatov Abstract Identifying protein–protein interactions is crucial for revealing protein functions and characterizing cellular processes. Manipulating PPIs has become widespread in treating human diseases such as cancer, autoimmunity, and infections. It has been recently applied to the regulation of protein tyrosine phosphatases (PTPs) previously considered undruggable. A broad panel of methods is available for studying PPIs. To complement the existing toolkit, we developed a simple method called fluorescent immunoprecipitation analysis (FIPA). This method is based on coimmunoprecipitation followed by protein gel electrophoresis and fluorescent imaging to visualize components of a protein complex simultaneously on a gel. The FIPA allows the detection of proteins expressed under native conditions and is compatible with mass spectrometry identification of protein bands. Key words Phosphatases, CD45, Immunoprecipitation, Fluorescent labeling, Mass spectrometry

1

Introduction Protein tyrosine phosphatases (PTPs) dephosphorylate proteins on tyrosine amino acid residues, counteracting the activity of protein tyrosine kinases (PTKs). Balance in the activity of PTKs and PTPs plays a key role in cellular signaling processes. PTKs initiate signaling and determine the magnitude of a cellular response to a stimulus, whereas PTPs inhibit signaling and thus regulate the duration of the response. The dampening of cellular activation pathways by PTPs is required for normal cell growth and proliferation. Aberrant PTP activity has been associated with tumor progression, autoimmunity, diabetes, and other diseases [1]. Some PTPs, such as SHP2 and PTP1B, have been suggested as good candidates for drug targets [2, 3]. However, numerous attempts to find an effective PTP inhibitor have failed, making PTPs considered undruggable. This is because the highly conserved and positively charged active center of the PTP catalytic domain makes selecting inhibitors

Damien The´venin and Jo¨rg P. Mu¨ller (eds.), Protein Tyrosine Phosphatases: Methods and Protocols, Methods in Molecular Biology, vol. 2743, https://doi.org/10.1007/978-1-0716-3569-8_12, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

181

182

Natalia Kruglova and Alexander Filatov

extremely challenging [1]. Whereas more than a hundred protein kinase inhibitors have received the US Food and Drug Administration (FDA) approval, no inhibitors have been approved for protein phosphatases [3]. Fortunately, over the last several years of work, the focus of research has shifted toward PTP allosteric inhibitors demonstrating promising results [4]; several of these are being evaluated in clinical trials [3]. Since allosteric inhibitors binding outside of the active center of an enzyme can effectively block PTP activity, researchers have directed their attention to PTP regulation. Accumulating evidence suggests that the activity of PTPs is modulated by intramolecular and intermolecular bonds, and it is now an established notion that the area of interaction between protein partners can be targeted by drugs [5]. Several such inhibitors have already been identified [5]. Importantly, manipulating the activity of PTPs via their interactions with other proteins helps achieve what was unthinkable when using conventional inhibitors. With these compounds, it might be possible to block PTPs and stimulate their activity, which may be a helpful strategy for controlling PTPs with tumor suppressor functions. In addition, this could act as a vehicle for targeting PTPs with a dual role in the cellular signaling [6]. To conclude, identifying and studying protein partners of PTPs represent an important area of research because it can help find novel drug targets for many human diseases. Over more than a hundred years of research, different methods have been developed to identify and study protein–protein interactions (PPIs) [7]. Affinity chromatography of proteins with peptide tags, tandem affinity purification, and the yeast two-hybrid method were among the first such tools [8]. Later, coimmunoprecipitation coupled with mass spectrometric analysis has become one of the most widely used approaches to studying protein interactome. Chemical cross-linkers were subsequently introduced to stabilize weak interactions [9]. Furthermore, blue-native polyacrylamide gel electrophoresis (BN-PAGE) emerged as an electrophoretic method for PPI interrogation [10]. In recent years, with the development of detection devices, fluorescence-based methods have gained prominence, such as bimolecular fluorescence complementation (BiFC) [11] and fluorescence (Fo¨rster) resonance energy transfer (FRET) [12]. We propose a method that we named fluorescent immunoprecipitation analysis (FIPA), which is based on coimmunoprecipitation, a gold standard for the identification of protein interactions [13]. This simple method can be used to study protein complexes of the protein of interest, identify its protein partners, and roughly estimate the stoichiometry of the complexes. We will now outline the basic principle of the method. Live cells are stained in the phosphate-buffered saline (PBS) solution with a fluorescent dye

PPI Identification with Fluorescent Immunoprecipitation Analysis

183

that covalently binds to amino (NH2-) groups of lysine residues or sulfhydryl (SH-) groups of cysteine residues. Cells are washed and lysed, and the lysates are immunoprecipitated with beads bound to monoclonal antibodies against the protein of interest. Precipitates are eluted and resolved with sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE). Gels are scanned for the fluorescent visualization of protein bands, which can then be excised and identified using mass spectrometry. The FIPA method was first described in our paper as a simpler alternative to the radioimmunoprecipitation assay (RIPA) for studying membrane proteins [14]. We validated the method on a panel of 29 monoclonal antibodies with confirmed specificity and successfully used it to identify an antigen that is detected by antibodies with unknown specificity. Later, we applied FIPA to characterize monoclonal antibodies against the Tn antigen associated with many types of tumors [15, 16]. Our work was the first to demonstrate that the Tn antigen was present on the PTP CD45 and the molecule CD44. Apart from identifying surface antigens, the FIPA proved to be suitable for studying protein complexes, as described in our paper, where we characterized the protein complex of the CD4 molecule in the T cell line CCRF-CEM [17]. In coprecipitates of CD4 and CD45, we discovered a poorly investigated protein lymphocyte phosphatase-associated phosphoprotein (LPAP). We used the FIPA to examine the binding partners of LPAP, revealing the phosphatase CD45 as the only reliable partner [18]. Then, using a panel of cell clones and populations with the knockout of either LPAP or CD45, we observed a correlation between the level of LPAP and CD45. We hypothesized that one of the functions of LPAP might be to control the cellular level of CD45 [19]. In conclusion, the FIPA is a simple method that can be applied to detect and investigate protein partners under native expression conditions. The method has several advantages. First, the FIPA does not necessitate the handling of radioactivity. Second, its sensitivity reaches around 0.3 ng/band. Third, compared with biotin labeling, fluorescent protein labeling, in most cases, preserves epitopes recognized by antibodies. Fourth, the FIPA allows for the identification of proteins present in either high or low amounts in the cell. Fifth, fluorescent labeling does not hinder trypsin digestion, and the FIPA is compatible with subsequent Western blotting and mass spectrometry. Except for the mass spectrometry stage, the FIPA method is simple and fast: The results can be obtained in two days. Compared with coimmunoprecipitation with subsequent Western blotting, the FIPA can visualize all protein complex components simultaneously. Furthermore, the stoichiometry of the components can be roughly estimated with a correction for unequal labeling of proteins with different lysine content [14].

184

2 2.1

Natalia Kruglova and Alexander Filatov

Materials Cell Culture

1. Cells: Jurkat cells or any other cell line or primary cells in which the protein of interest is expressed. 2. Cell media: RPMI 1640 medium supplemented with 10% fetal calf serum, 2 mM L-glutamine, 24 μg/mL gentamicin, 1 mM sodium pyruvate, and 10 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES).

2.2 Fluorescent Staining

1. PBS: 0.137 M NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.8 mM KH2PO4, pH 7.4. 2. Fluorescent dye R6G or another dye such as fluorescein isothiocyanate (FITC), Cy3, Cy5, or Dy547 (see Note 1). The dye must be in the active form of an N-hydroxysuccinimide (NHS) ester for coupling via a side-chain amine group of lysine residues or in the maleimide form for coupling via thiol groups of cysteine residues (see Note 2). Prepare stock solutions (10 mg/ mL) in dimethyl sulfoxide (DMSO), aliquot, and freeze. The frozen solution is stable for several weeks at -20 °C.

2.3 Cell Lysis and Immunoprecipitation

1. Cell lysis buffer with stringent detergent: 50 mM Tris–HCl, рH 8.2, 0.15 M NaCl, 5 mM ethylenediaminetetraacetic acid (EDTA), 1% Triton X-100, 0.02% NaN3, 1 mM phenylmethylsulfonyl fluoride (PMSF), 0.01 M NaF, and 1 mM Na3VO4 (see Note 3). 2. Cell lysis buffer with mild detergent: 50 mM Tris–HCl, рH 8.2, 0.15 M NaCl, 5 mM EDTA, 1% Brij97, 0.02% NaN3, 1 mM PMSF, 0.01 M NaF, and 1 mM Na3VO4 (see Note 4). 3. Primary antibodies against the protein of interest (see Notes 5–7). In addition, primary antibodies against binding partners of the protein of interest can be used for validation experiments. As a negative control, choose antibodies against an irrelevant antigen or a normal immunoglobulin from the same animal species used to generate the primary antibodies. 4. Appropriate secondary antibodies. 5. Affinity medium for antibody immobilization: Protein A or Protein G sepharose or agarose (see Note 8). Beads with immobilized secondary antibodies or ready-to-use beads with immobilized primary antibodies can be a more convenient alternative (see Note 9). 6. Tube rotator suitable for 1.5 mL tubes.

2.4

SGS-PAGE

1. Laemmli sample buffer (4×): 0.25 M Tris–HCl, pH 6.8, 50% glycerol, 8% SDS, 0.08% bromophenol blue, 4% β-mercaptoethanol.

PPI Identification with Fluorescent Immunoprecipitation Analysis

185

2. Resolving gel for Laemmli SDS-PAGE (see Note 10): 0.375 M Tris–HCl, pH 8.8, 10% acrylamide/bis-acrylamid solution (T = 30%, С = 2.75%), 0.1% SDS, 0.05% ammonium persulfate, 0.08% (vol:vol) tetramethylethylenediamine (TEMED). 3. Stacking gel for Laemmli SDS-PAGE: 0.125 M Tris–HCl, pH 6.8, 5% acrylamide/bis-acrylamid solution (T = 30%, С = 2.75%), 0.1% SDS, 0.05% ammonium persulfate, 0.08% (vol:vol) TEMED. 4. Laemmli electrophoresis running buffer: 0.025 M Tris–HCl, 0.192 M glycine, 0.0035 M SDS. 5. Equipment for protein electrophoresis. 6. Fluorescent protein markers (optional) (see Note 11). 7. Gel imager with filters compatible with fluorescent dyes. 2.5 Mass Spectrometry

3 3.1

1. Access to mass spectrometry facility (see Note 12). 2. 1% acetic acid.

Methods Protein Labeling

1. Culture cells until they reach the required amount (see Note 13). Stimulate cells if needed for analysis (see Note 14). Harvest cells (see Note 15). 2. Wash cells two times with cold PBS, and resuspend in 1 mL of cold PBS (see Note 16). Use tubes and a microfuge prechilled to 4 °C. 3. Add 30 μL of dye stock solution (final concentration = 0.3 mg/mL) to cells dropwise with constant stirring of the suspension by vortex and incubate for 20 min on ice. 4. Remove the unreacted dye by washing cells two times with cold PBS.

3.2

Cell Lysis

1. Supplement the required amount of lysis buffer with a mixture of inhibitors, including 1 mM PMSF (see Note 17), add to cells (1 mL of lysis buffer per 20–40 million cells), and incubate for 30 min at 4 °C. 2. Pellet debris by centrifugation at 20,000× g for 15 min at 4 °C.

3.3 Sorbent Preparation

Perform this step during cell lysis (see Note 18). 1. Transfer beads for lysate preclearance or beads for precipitation (see Note 19) into tubes and wash two times with cold lysis buffer. Prepare a number of tubes equal to the number of samples. Before transferring the sorbent suspension, cut the pipette’s tip to widen the hole. When pipetting suspension,

186

Natalia Kruglova and Alexander Filatov

always carefully mix it before each collection by inverting the bottle several times and checking that all sorbent from the bottom has moved to suspension. Otherwise, the number of beads in tubes will vary significantly. 2. Wash beads by centrifugation at 3000× g for 30 s. 3. If the beads are used with immobilized primary antibodies, the sorbent is ready for immunoprecipitation. 4. If the beads are coupled with secondary antibodies, add primary antibody at the dilution recommended by the manufacturer. 5. Incubate for 1–2 h by rotation at 4 °C. 6. Wash beads three times with cold lysis buffer by centrifugation at 3000× g for 30 s. 7. The sorbent is ready for immunoprecipitation. 8. If the beads are used with Protein A/G, add secondary antibody at the dilution recommended by the manufacturer. 9. Incubate for 1–2 h by rotation at 4 °C. 10. Wash beads three times with cold lysis buffer by centrifugation at 3000× g for 30 s. 11. Add primary antibody at the recommended dilution. 12. Incubate for 1–2 h by rotation at 4 °C. 13. Wash beads three times with cold lysis buffer by centrifugation at 3000× g for 30 s. 14. The sorbent is ready for immunoprecipitation. 3.4 Immunoprecipitation

1. Preclear cell lysates by rotation at 4 °C with empty sorbent, sorbent with bound secondary antibodies, or sorbent with secondary antibodies bound to a normal immunoglobulin instead of specific antibodies (see Note 20). This step can be completed in 2 h or left overnight. 2. Transfer precleared lysates to prepared beads. 3. Incubate samples by rotation for 2 h at 4 °C. 4. Wash the beads three times with cold lysis buffer (see Note 21). Always leave some liquid above the beads while washing to avoid disturbing them. 5. Aspirate the wash buffer almost entirely and remove residual liquid carefully with the thin pipette tip touching the bottom of the tube. Add 30 uL of 4× SDS sample buffer to completely cover the beads and elute proteins by heating the beads for 5 min at 80 °C (see Note 22). 6. Centrifuge samples and collect eluates (see Note 23).

PPI Identification with Fluorescent Immunoprecipitation Analysis

3.5 Protein Electrophoresis

187

1. Cast gel for the Laemmli SDS–PAGE and resolve sample eluates until the tracer stain reaches the edge of the gel but does not leave it (see Note 24). 2. Before opening the glass cassette, place it on a tray, put it in the imager, and scan it. Cut the desired bands (see Note 25) and put them into 1.5 mL tubes with 1% acetic acid. 3. Send gel pieces to mass spectrometry analysis.

4

Notes 1. A broad panel of fluorescent dyes can be used in FIPA, such as FITC, FAM, JOE, HEX, R6G, TAMRA, ROX, Cy3, Cy5, and others. The spectral properties of an available imaging system primarily determine the choice of a fluorophore. Also, consider the hydrophobic/hydrophilic characteristics of a fluorescent probe because it affects the ability of a probe to penetrate cell membranes. Fluorescein and its derivatives (FITC, FAM, JOE, and HEX) represent hydrophilic fluorescent dyes that do not pass through the cell membrane [20] and stain predominantly surface proteins. More hydrophobic rhodamine-based dyes (R6G, TAMRA, and ROX) penetrate cell membranes [21] and stain both surface and cytoplasmic proteins. We have not studied the possibility of staining nuclear proteins, but the FIPA may also be suitable for proteins of this type. R6G is a universal dye that produces the best results in our hands. It is convenient to use a pair of cyanine dyes Cy3/Cy5 when comparing subproteomes in different functional states [22]. 2. N-hydroxysuccinimide (NHS) ester is the predominantly used form of fluorescent dye for covalent protein labeling. The NHS group effectively reacts with side-chain amine groups of lysine residues. The isothiocyanate group of the FITC molecule has the same activity, but it is rarely used with other dyes. An alternative labeling approach relies on the maleimide reaction group of a dye, which enables coupling via thiol groups of cysteine residues. Since lysine residues are more abundant in proteins, lysine labeling typically produces a brighter signal than labeling cysteine residues. Larger proteins have more amino acid residues available for labeling and are thus stained better than smaller proteins. Nevertheless, even a small protein like CD59 was detected by the FIPA with a good signal despite its mature form having a molecular weight (MW) of less than 9 kDa and only six lysine residues. Apart from side-chain amine groups of lysine residues, the NHS group reacts with the N-terminal NH2 group. For instance, the FIPA detected the small protein LPAP (MW = 21 kDa) lacking lysine residues [18]. The labeling by side-chain SH groups of cysteine residues

188

Natalia Kruglova and Alexander Filatov

can be applied when protein complexes need to be stabilized by covalent cross-linking before cell lysis. The reagent dithiobis (succinimidyl propionate) (DSP) is a cross-linker that couples proteins via NH2 groups, thus making NHS fluorescent dyes unsuitable [18]. 3. Lysis buffer with the detergent Triton X-100 serves as a control buffer that destroys most protein interactions. Samples lysed in this buffer should not contain coprecipitation bands. 4. The strength of the interaction between components in protein complexes can vary significantly. The detection of a protein– protein interaction may require experimental determination of a buffer with a suitable mild detergent that, on the one hand, sufficiently solubilizes the proteins of interest but, on the other hand, keeps the complexes intact. Instead of 1% Triton X-100, which retains only fairly strong interactions, try milder detergents, such as 1% Brij58, Brij97, 3-[(3-cholamidopropyl) dimethylammonio]-1-propanesulfonate (CHAPS), or 3-[(3cholamidopropyl)dimethylammonio]-2-hydroxy-1-propanesulfonate (CHAPSO). It is worth noting that nuclear proteins are solubilized only in stringent buffers like RIPA and that partners of these proteins can be exclusively detected with preliminary protein cross-linking followed by lysis in RIPA buffer. This protocol involving protein cross-linking and RIPA lysis is also applicable when stabilization of weak and/or temporary interactions is required [18]. Most cytoskeleton proteins resist lysis even with stringent detergents [23, 24], which makes the FIPA inappropriate for studying these proteins. 5. Unlike biotin labeling, previously popular for surface labeling of proteins [25–28], fluorescent dye labeling does not disrupt antibody epitopes, which was validated on seven different proteins using flow cytometry. We showed that when proteins were labeled with TAMRA dye and FITC-conjugated antibodies were used for flow cytometry, the signal was slightly lower than for the unlabeled proteins. In contrast, when TAMRA was replaced with ROX, the signal remained unchanged. We concluded that antibody binding was not affected by labeling but decreased due to energy transfer from FITC to TAMRA that did not occur when using ROX. Only one of our antibodies tested in FIPA, an antibody against CD7, did not work with the FIPA method despite producing a good signal with the RIPA method [14]. In this case, we recommend selecting a different primary antibody (see Note 6) or a dye with a different reactive group that binds to other amino acid residues (see Note 2).

PPI Identification with Fluorescent Immunoprecipitation Analysis

189

6. The primary antibody against the protein of interest used for immunoprecipitation must be validated by Western blotting using the same antibody for membrane staining. It should be ensured that the target band is correctly positioned and that there are no extraneous bands. An antibody with irrelevant specificity or a normal immunoglobulin should be used as a negative control in immunoprecipitation. An additional negative control would be a cell line with the knockout of the protein of interest. Our data show that proteins with both high and low abundance can be detected by the FIPA. Signal intensity is highly dependent on the affinity of the antibody used. An antibody must have a dissociation constant in the low molar—picomolar—range for successful immunoprecipitation. We could not precipitate several monoclonal antibodies, perhaps due to the low affinity of the monoclonal antibodies used, which precluded their use in FIPA. In most cases, labeling proteins with fluorescent dyes does not disrupt the epitope recognized by the antibody (see Note 5). Both monoclonal and polyclonal antibodies can be used for immunoprecipitation. Our experiments were conducted with monoclonal antibodies, giving a lower background signal. However, in some cases, a particular antibody clone may not be suitable for the detection of PPI due to the overlap of its epitope with the surface of interaction between the protein of interest and its partners. In such cases, we recommend trying a different antibody clone or using polyclonal antibodies. 7. We have used antibodies against endogenous proteins without peptide tags to study protein complexes under physiological conditions. If such antibodies are unavailable, antibodies against peptide tags (e.g., Flag or HA) can be used, keeping in mind that these tags can disrupt native protein interactions. In addition, when dealing with tagged proteins, special attention should be paid to the choice of a model cell line. The most convenient method involves taking an irrelevant cell line and transfecting it with plasmids that code for potential interacting partners. When using this method, it is essential to consider whether an endogenous protein of interest is expressed in these cells, even at a low level. A more complex approach is to choose a relevant cell line with the knockout of the corresponding protein stably transduced with the construct coding for the tagged protein of interest. However, transfection or stable transduction often results in the overexpression of a protein, which can significantly affect PPI. Endogenous expression can be preserved using the third, most complex model where a peptide tag is introduced on the N- or C-terminus of the protein of interest by CRISPR-Cas-mediated knocking of a small construct [29]. Typically, a population of CRISPR-Cas-

190

Natalia Kruglova and Alexander Filatov

edited cells undergoes some enrichment or cloning. If cell clones are used for experiments, it is important to account for their variability, which stems from the heterogeneity of the parental population or different off-target effects. This problem can be solved by selecting an entire population of edited cells rather than individual clones, for example, using the Surface Oligopeptide knock-in for Rapid Target Selection (SORTS) method, which is suitable even for intracellular proteins [30]. 8. Protein A and Protein G have varying binding strengths to immunoglobulins of different subtypes and from different animal origins. 9. If a large amount of the primary antibody against the protein of interest is accessible, the antibody can be covalently immobilized to an affinity medium. Good results are achieved with the medium Affi-Gel Hz (BioRad), which allows for oriented coupling of an antibody. Using directly conjugated sorbents is beneficial in several ways. First, it helps save time and resources for routine experiments. Second, it helps reduce a background signal. Third, it prevents the elution of antibodies from the sorbent and the proteins of interest, which may interfere with identifying proteins with a molecular weight close to the weight of IgG heavy chains. Light chains enter the eluate even with immobilized primary antibodies. Our work used the affinity medium Affi-Gel Hz with immobilized monoclonal antibodies against the LPAP protein. Using 6 mg of the antibody, we obtained about 4 mL of the 25% sorbent suspension, which is sufficient for 100–150 individual immunoprecipitation reactions. 10. The percentage of the resolving gel should be determined experimentally. As a starting point, try 10%. 11. The interpretation of the FIPA results is easier with fluorescent molecular weight markers. Choose markers with the same fluorophore used for sample protein labeling or at least a fluorophore of similar size. A rough estimate of the molecular weights of coprecipitated proteins can facilitate their identification and verify mass spectrometry data. 12. We do not perform a mass spectrometry analysis ourselves but send samples to a specialized facility. Samples can be shipped as fixed gel pieces or as a mixture of peptides after trypsin digestion. The best way for shipment should be determined with the operator of the mass spectrometer. 13. The number of cells sufficient for detecting the protein of interest and its partners should be determined experimentally. It depends on the target protein’s expression and the primary antibody’s affinity. We successfully detected bands of the LPAP and CD45 proteins using 5 × 106 cells per lane [18]. The

PPI Identification with Fluorescent Immunoprecipitation Analysis

191

sensitivity of the FIPA is about 0.3 ng per protein band for the R6G-labeled bovine serum albumin (BSA). For comparison, applying the same test, the sensitivity for SYPRO Red was 10 ng per band and for Coomassie 100 ng per band [14]. 14. Incubate cells with a specific reagent before analysis if the proteins of interest are supposed to interact only after a particular stimulation. Take unstimulated cells as a negative control. 15. Cell lines with the knockout of one of the protein partners under study can act as an additional negative control. For instance, in cases where it is necessary to test the possibility of direct interaction between two proteins without an intermediate protein, a cell line with the knockout of this intermediate protein will facilitate data interpretation. It may also help when the protein of interest forms a permanent tight complex with one of its partners. It is required to investigate interactions between the protein of interest and other proteins independently of its main partner. The CRISPR-Cas technology can produce knockout cell lines, and knockouts of intracellular proteins are easily obtained by using the SORTS method [30]. 16. Labeling can be done at room temperature if the proteins under study are not prone to degradation, but this needs to be verified. 17. Additional protease inhibitors like pepstatin A (1 μM) and leupeptin (10 μM) may be added when degradation is observed. If phosphorylation is involved or suspected in protein interaction, the lysis buffer should be supplemented with phosphatase inhibitors, 10 mM NaF and 1 mM Na3VO4. 18. Better results are achieved when lysates are incubated with beads prebound to antibodies than when lysates are mixed sequentially with primary antibodies, then secondary antibodies, and, finally, beads [17]. Therefore, we recommend preparing a sorbent with noncovalently coupled antibodies in advance or using beads with covalently immobilized primary antibodies (see Note 9). 19. The required number of beads depends on the affinity of the antibody and the antibody/beads ratio, as well as on the cell lysate volume. We used 30 μL of the 25% sorbent suspension per sample. The amount should be tested in a titration experiment, which can be set up as follows: (1) Prepare a cell lysate volume sufficient for 4–5 samples. (2) Add an increasing number of beads into separate tubes, wash, aliquot an equal volume of the lysate, and proceed with the experiment as described in the following steps. (3) Monitor the level of precipitated protein by the FIPA or Western blotting, and choose the second sample once saturation is observed. (4) The number of beads for preclearance may be equal to or higher than those for precipitation.

192

Natalia Kruglova and Alexander Filatov

20. Lysate preclearance drastically reduces a background signal. An ideal preclearing medium is the same sorbent used for immunoprecipitation bound to a normal immunoglobulin, though the empty sorbent itself will suffice. 21. The level of background can be significantly reduced with alternative washes. Instead of washing beads with the same lysis buffer, try the following three buffers with varying ionic strength and detergent concentration: Solution 1: 50 mM Tris–HCl, рH 7.4, 500 mM NaCl, 5 mM EDTA, 0.5% Triton X-100, 0.02% NaN3. Solution 12: 50 mM Tris–HCl, рH 7.4, 150 mM NaCl, 5 mM EDTA, 0.5% Triton X-100, 0.02% NaN3. Solution 33: 10 mM Tris–HCl, рH 8.0, 0.1% Triton X-100. If cells are lysed with a mild detergent, the same detergent must be used in all buffers for alternative washes instead of Triton X-100. 22. Avoid boiling samples, as it will lead to the aggregation of membrane proteins. 23. Traces of beads can be collected, so it is necessary to spin samples and pipette carefully before loading them on a gel. 24. Labeling can slightly modify the electrophoretic mobility of proteins. Although the change is generally in the range of 3 kDa, it is advisable to use protein markers with the same fluorescent dye used for sample protein labeling (see Note 11). 25. Adjust the resolution and signal acquisition time. Take photographs. Determine which bands you want to excise. To simplify manual band excision, you can mark the bands on the glass with a marker and check the position of the mark by taking a photograph. The desired band can be precisely positioned via sequential marking and scanning. Open the glass plates while leaving the gel on the marked side, cut the band within the defined borders, and ensure that the cut is complete with a final scan. References 1. Ko¨hn M (2020) Turn and face the strange: } a new view on phosphatases. ACS Cent Sci 6: 467–477 2. He RJ, Yu ZH, Zhang RY, Zhang ZY (2014) Protein tyrosine phosphatases as potential therapeutic targets. Acta Pharmacol Sin 35:1227 3. Turdo A, D’Accardo C, Glaviano A, Porcelli G, Colarossi C, Colarossi L, Mare M, Faldetta N, Modica C, Pistone G, Bongiorno MR, Todaro M, Stassi G (2021) Targeting

phosphatases and kinases: how to checkmate cancer. Front Cell Dev Biol 9:2979 4. Chen YNP, Lamarche MJ, Chan HM, Fekkes P, Garcia-Fortanet J, Acker MG, Antonakos B, Chen CHT, Chen Z, Cooke VG, Dobson JR, Deng Z, Fei F, Firestone B, Fodor M, Fridrich C, Gao H, Grunenfelder D, Hao HX, Jacob J, Ho S, Hsiao K, Kang ZB, Karki R, Kato M, Larrow J, La Bonte LR, Lenoir F, Liu G, Liu S, Majumdar D, Meyer

PPI Identification with Fluorescent Immunoprecipitation Analysis MJ, Palermo M, Perez L, Pu M, Price E, Quinn C, Shakya S, Shultz MD, Slisz J, Venkatesan K, Wang P, Warmuth M, Williams S, Yang G, Yuan J, Zhang JH, Zhu P, Ramsey T, Keen NJ, Sellers WR, Stams T, Fortin PD (2016) Allosteric inhibition of SHP2 phosphatase inhibits cancers driven by receptor tyrosine kinases. Nature 535:148–152 5. Hendriks W, Bourgonje A, Leenders W, Pulido R (2018) Proteinaceous regulators and inhibitors of protein tyrosine phosphatases. Molecules 23:395 6. Rhee I, Veillette A (2012) Protein tyrosine phosphatases in lymphocyte activation and autoimmunity. Nat Immunol 13:439–447 7. Rao VS, Srinivas K, Sujini GN, Kumar GNS, Rao VS, Srinivas K, Sujini GN, Kumar GNS (2014) Protein-protein interaction detection: methods and analysisю Int. J Proteome 2014: 147648 8. Braun P, Gingras A-C (2012) History of protein-protein interactions: from egg-white to complex networks. Proteomics 12:1478– 1498 9. Kim KM, Yi EC, Kim Y (2012) Mapping protein receptor-ligand interactions via in vivo chemical crosslinking, affinity purification, and differential mass spectrometry. Methods 56: 161–165 10. Schamel WW (2008) Two-dimensional blue native polyacrylamide gel electrophoresis. Curr Protoc Cell Biol Chapter 6:Unit 6.10 11. Kerppola TK (2006) Design and implementation of bimolecular fluorescence complementation (BiFC) assays for the visualization of protein interactions in living cells. Nat Protoc 1:1278–1286 12. Brzostowski JA, Meckel T, Hong J, Chen A, Jin T (2009) Imaging protein-protein interactions by Fo¨rster resonance energy transfer (FRET) microscopy in live cells. Curr Protoc Protein Sci Chapter 19(Unit19):5 13. Fluorescent immunoprecipitation analysis. https://www.protocols.io/view/fluorescentimmunoprecipitation-analysis-ewov1zzpgr24/ v1. Accessed 16 Dec 2022 14. Filatov AV, Krotov GI, Zgoda VG, Volkov Y (2007) Fluorescent immunoprecipitation analysis of cell surface proteins: a methodology compatible with mass-spectrometry. J Immunol Methods 319:21–33 15. Blixt O, Lavrova OI, Mazurov DV, Clo´ E, Kracˇun SK, Bovin NV, Filatov AV (2012) Analysis of Tn antigenicity with a panel of new IgM and IgG1 monoclonal antibodies raised against leukemic cells. Glycobiology 22:529–542

193

16. Shuvalova ML, Kopylov AT, Mazurov DV, Pichugin AV, Bovin NV, Filatov AV (2020) CD44-Associated Tn antigen as a new biomarker of tumor cells with aberrant glycosylation. Biochemistry (Mosc) 85:1064–1081 17. Krotov GI, Krutikova MP, Zgoda VG, Filatov AV (2007) Profiling of the CD4 receptor complex proteins. Biochemistry (Mosc) 72:1216– 1224 18. Kruglova NA, Kopylov AT, Filatov AV. (2019) [Identification of the Molecular Partners of Lymphocyte Phosphatase-Associated Phosphoprotein (LPAP) that are involved in human lymphocyte activation]. Mol Biol (Mosk) 53:838–848, Russian 19. Kruglova NA (2019) Role of lymphocyte phosphatase associated phosphoprotein (LPAP) in T and B lymphocyte activation. Unpublished doctoral dissertation. Lomonosov Moscow State University 20. Nitsch A, Haralambiev L, Einenkel R, Muzzio DO, Zygmunt MT, Ekkernkamp A, Burchardt M, Stope MB (2019) Determination of in vitro membrane permeability by analysis of intracellular and extracellular fluorescein signals in renal cells. In Vivo 33:1767–1771 ˜ es GM, Ramos C, Samelo J, 21. Magalha˜es N, Simo Oliveira AC, Filipe HAL, Ramalho JPP, Moreno MJ, Loura LMS (2022) Interactions between rhodamine dyes and model membrane systems—insights from molecular dynamics simulations. Molecules 27:1420 22. Kruglova NA, Meshkova TD, Kopylov AT, Mazurov DV, Filatov AV (2017) Constitutive and activation-dependent phosphorylation of lymphocyte phosphatase-associated phosphoprotein (LPAP). PLoS One 12:e0182468 23. Choi S, Kelber J, Jiang X, Strnadel J, Fujimura K, Pasillas M, Coppinger J, Klemke R (2014) Procedures for the biochemical enrichment and proteomic analysis of the cytoskeletome. Anal Biochem 446:102–107 24. Filatov AV, Shmigol IB, Kuzin II, Sharonov GV, Feofanov AV (2003) Resistance of cellular membrane antigens to solubilization with Triton X-100 as a marker of their association with lipid rafts – analysis by flow cytometry. J Immunol Methods 278:211–219 25. Hurley WL, Finkelstein E, Holst BD (1985) Identification of surface proteins on bovine leukocytes by a biotin-avidin protein blotting technique. J Immunol Methods 85:195–202 26. Cole SR, Ashman LK, Ey PL (1987) Biotinylation: an alternative to radioiodination for the identification of cell surface antigens in immunoprecipitates. Mol Immunol 24:699–705

194

Natalia Kruglova and Alexander Filatov

27. K€ahne T, Ansorge S (1994) Non-radioactive labelling and immunoprecipitation analysis of leukocyte surface proteins using different methods of protein biotinylation. J Immunol Methods 168:209–218 28. Schuberth HJ, Kroell A, Leibold W (1996) Biotinylation of cell surface MHC molecules: A complementary tool for the study of MHC class II polymorphism in cattle. J Immunol Methods 189:89–98 29. Dewari PS, Southgate B, Mccarten K, Monogarov G, O’Duibhir E, Quinn N, Tyrer A, Leitner MC, Plumb C, Kalantzaki M,

Blin C, Finch R, Bressan RB, Morrison G, Jacobi AM, Behlke MA, von Kriegsheim A, Tomlinson S, Krijgsveld J, Pollard SM (2018) An efficient and scalable pipeline for epitope tagging in mammalian stem cells using Cas9 ribonucleoprotein. elife 7:e35069 30. Zotova A, Pichugin A, Atemasova A, Knyazhanskaya E, Lopatukhina E, Mitkin N, Holmuhamedov E, Gottikh M, Kuprash D, Filatov A, Mazurov D (2019) Isolation of gene-edited cells via knock-in of short glycophosphatidylinositol-anchored epitope tags. Sci Rep 9:3132

Chapter 13 Identifying Transmembrane Interactions in Receptor Protein Tyrosine Phosphatase Homodimerization and Heterodimerization Sophie Rizzo and Damien The´venin Abstract Receptor protein tyrosine phosphatases (RPTPs) are one of the key regulators of receptor tyrosine kinases (RTKs) and therefore play a critical role in modulating signal transduction. While the structure–function relationship of RTKs has been widely studied, the mechanisms modulating the activity of RPTPs still need to be fully understood. On the other hand, homodimerization has been shown to antagonize RPTP catalytic activity and appears to be a general feature of the entire family. Conversely, their documented ability to physically interact with RTKs is integral to their negative regulation of RTKs, but there is a yet-tobe proposed common model. However, specific transmembrane (TM) domain interactions and residues have been shown to be essential in regulating RPTP homodimerization, interactions with RTK substrates, and activity. Therefore, elucidating the contribution of the TM domains in RPTP regulation can provide significant insights into how these receptors function, interact, and eventually be modulated. This chapter describes the dominant-negative AraC-based transcriptional reporter (DN-AraTM) assay to identify specific TM interactions essential to homodimerization and heteroassociation with other membrane receptors, such as RTKs. Key words Receptor protein tyrosine phosphatase, Receptor protein tyrosine kinase, Dimerization, Transmembrane domain interaction, Bacterial transcription-based reporter assay

1

Introduction Twenty-two human protein tyrosine phosphatases (PTPs) are classified as receptor protein tyrosine phosphatases (RPTPs). They all share the same architecture: an extracellular region, a single-pass transmembrane segment, and one or two cytosolic and highly conserved PTP domains [1]. By being some of the most important regulators of receptor tyrosine kinases (RTKs), RPTPs play critical signaling regulatory roles in development, homeostatic control, health, and disease progression. They can be tumor-suppressors of many cancers, including colon, lung, breast, and thyroid. Still, they

Damien The´venin and Jo¨rg P. Mu¨ller (eds.), Protein Tyrosine Phosphatases: Methods and Protocols, Methods in Molecular Biology, vol. 2743, https://doi.org/10.1007/978-1-0716-3569-8_13, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

195

196

Sophie Rizzo and Damien The´venin

can also lead to cancer progression when deregulated through reduced expression or loss-of-function mutations [2–6]. Therefore, RPTPs are crucial for controlling cell growth and have long been viewed as potential therapeutic targets. Yet, the detailed structural basis underpinning the regulation of their catalytic activity remains unclear. However, the activity of many RPTPs has been reported to be suppressed by homodimerization, which may prevent RPTP access to their RTK substrates. The transmembrane (TM) domains for RPTPs are proposed to mediate homodimerization, although there is no clear structure-based model for this mechanism [7– 10]. Understanding the influence of RPTP TM domains in oligomerization and activity could advance our basic biological understanding of how RPTPs regulate cell signaling pathways in cells and develop new therapeutic approaches to promote RPTP activity against their oncogenic RTK substrates. We previously reported that specific TM interactions and residues regulate the homodimerization of a member of the RPTP family (protein tyrosine phosphatase receptor J [PTPRJ]) [11]. To do so, we used the dominant-negative AraC-based transcriptional reporter (DN-AraTM) assay, which reports on the propensity of TM domains to self-associate and heterodimerize in cell membranes. Disrupting these interactions, through point mutations in the TM domains of the full-length receptor expressed in mammalian cells, disrupted PTPRJ homodimerization, reduced access to one of its substrates estimated glomerular filtration rate (eGFR) [12–14], led to reduced eGFR phosphorylation, and antagonized eGFR-driven cell phenotypes [11]. Similar results have been obtained in acute myeloid leukemia cell models expressing the PTPRJ mutant and the oncogenic FLT3 mutant (with internal tandem duplications) [15]. In this chapter, we describe the use of the DN-AraTM assay to readily identify TM–TM interactions mediating RPTP homodimerization and heteroassociation with other single-pass membrane receptors, like RTKs. Briefly, the assay relies on protein chimeras containing the receptor TM domains of interest fused either to the Escherichia coli transcription factor AraC (active at the araBAD promoter as a homodimer) or to its inactive R210A mutant (namely AraC*) (Fig. 1a) [16, 17]. Both chimeras include an N-terminal maltose-binding protein (MBP) fusion to promote unidirectional insertion in the inner membrane of AraC-deficient E. coli (with MBP residing in the periplasm). The AraC and AraC* also include an HA-tag and a myc-tag, respectively, to determine their expression levels. TM-mediated homodimerization of AraC induces the transcription of the green fluorescent protein (GFP) reporter gene under the control of the araBAD promoter, enabling quantification of fluorescence measurements on whole cells directly from culture [16–19]. On the other hand, preferential

Assessing Transmembrane Interaction of RPTPs

197

Fig. 1 (a) Overview of the DN-AraTM assay. Chimeras containing N-terminal MBP fused to either an in-frame receptor A fragment (Domain A; dark blue) and C-terminal AraC or an in-frame receptor B fragment (Domain B; light blue) and C-terminal disabled AraC unable to activate transcription at the araBAD promoter (AraC*) are expressed by the regulator plasmids (pAraTMwt, ampicillin-resistant; pAraTMDN, kanamycin-resistant). Once expressed, MBP directs the integration of chimera in the inner membrane of E. coli. Homodimerization of Domain A (AraC-AraC) brings the AraC transcription factors in close proximity and activates the araBAD promoter to produce GFP. If Domain A has a higher affinity to heterodimerize with Domain B versus to homodimerize with Domain A, a reduction in GFP will be observed (AraC–AraC*) due to the inability of heterodimers containing receptor B-fused AraC* to activate transcription at the araBAD promoter. (This figure was created with BioRender.com.) (b) Representative example of results after correction for bacterial growth rate and chimera expression levels if needed (bars are color-coded as in (a))

TM-mediated AraC–AraC* heterodimerization reduces the level of GFP transcription. Therefore, the level of GFP fluorescence intensity, when normalized for cell density and expression level, is a measure of receptor domain dimerization (Fig. 1b). Depending on which TM domain is in fusion with AraC or AraC*, the DN-AraTM assay enables simultaneous measurement of homodimerization and heterodimerization. Notably, the effect of point mutations on both the processes can be evaluated.

2 Materials 2.1 Subcloning the Transmembrane Domain of Interest into AraTM Plasmids

1. AraTM Plasmids: pAraTMwt (AraC plasmid; Addgene, catalog number 47514) and pAraTMDN (AraC* plasmid; Addgene, catalog number 47515). 2. Oligonucleotide primers to amplify and subclone the deoxyribonucleic acid (DNA) sequence coding TM domains of interest into the AraTM plasmids (AraC/AraC*). The final insert should include SacI and KpnI restriction sites at the 5′ and 3′, respectively (see Notes 1 and 2). 3. DNA polymerase. 4. SacI and KpnI restriction enzymes. 5. T4 DNA ligase with appropriate buffer.

198

Sophie Rizzo and Damien The´venin

6. E. coli DH5α (or equivalent).

competent

cells

for

transformation

7. Sequencing primer to confirm subcloning (see Note 3). 8. Lysogeny broth (LB) medium with appropriate selective antibiotic (100 μg/mL ampicillin [Amp] or 50 μg/mL kanamycin [Kan]). 9. LB agar plate with appropriate selective antibiotic (100 μg/mL ampicillin or 50 μg/mL kanamycin) (see Note 4). 10. Plasmid DNA purification kit. 11. Plasmid DNA gel extraction kit. 12. Thermocycler. 13. 0.8% agarose gel. 14. Agarose gel electrophoresis apparatus. 15. Ultraviolet (UV) light transilluminator. 2.2 Cotransformation of AraC, AraC*, and Reporter Plasmids into E. coli

1. AraC-deficient E. coli strain SB1676 (The E. coli Genetic Stock Center at Yale University). 2. Empty vectors: pAraTMwt(empty) and pAraTMDN(empty) (not available on Addgene). 3. pAraGFPCDF (GFP reporter plasmid; Addgene, catalog number 47516). 4. LB medium and agar plates with appropriate selective antibiotics.

2.3 DN-AraTM Assay and Analysis

1. 24-well deep-well plate (10 mL) with U-bottom for bacterial growth and autoclave safety (see Note 5). 2. Adhesive and porous film for culture plates (e.g., VWR, catalog number 60941-086). 3. LB medium (100 μg/mL Amp, 50 μg/mL Kan, Specpure (Spec) 100 μg/mL). 4. LB agar plate (100 μg/mL Amp, 50 μg/mL Kan, Spec 100 μg/mL). 5. ZY media: 1% (w/v) tryptone, 0.5% (w/v) yeast extract. 6. 5052 (50×): 25% glycerol, 2.5% (w/v) glucose, 10% (w/v) ɑ-lactose. 7. P(NPS) (20×): 1 M sodium phosphate dibasic Na2HPO4, ~ pH 6.75. 8. Autoinduction media (100 mL): 92.8 mL of ZY media, 1× 5052, 1× P(NPS), and 2 mM MgSO4. 9. Assay plate, 96-well, black with clear and flat-bottom, polystyrene (e.g., Corning, catalog number 3631).

Assessing Transmembrane Interaction of RPTPs

199

10. Plate shaker with temperature adjustment. 11. Plate reader with filters for measuring absorbance at 600 nm and GFP fluorescence emission (i.e., 488 nm and 510 nm for excitation and emission, respectively). 2.4 Confirming Protein Expression Level by Immunoblotting

1. 8 M urea. 2. 4× sodium dodecyl sulfate (SDS) loading dye: 14 mM Tris– HCl, pH 6.8, 17.2 mM β-mercaptoethanol, 5% glycerol, 8% (w/v) SDS, 0.015% (w/v) bromophenol blue (see Note 6). 3. Heat block. 4. Polyacrylamide gel electrophoresis (PAGE) apparatus. 5. 8% SDS–polyacrylamide gel (homemade or from a commercial source). 6. SDS–PAGE running buffer (1 L): 2.5 mM Tris base, pH 8.6, 19.2 mM glycine, 0.1% (w/v) SDS. 7. Transfer buffer (2 L): 5 mM Tris base, pH 8.6, 38.4 mM glycine, 20% methanol. 8. Nitrocellulose blotting membrane (e.g., Cytiva, catalog number 1060003). 9. Transfer apparatus. 10. 5% (w/v) bovine serum albumin (BSA) solution. 11. Rocking platform. 12. Tris-buffered saline (TBS; 1×): 1 mM Tris base, pH 7.6, 15 mM NaCl. 13. TBS containing 0.1% Tween-20 (TBS-T). 14. Blocking solution: 5% milk in TBS-T or commercially available blocking solution. 15. Antibodies: Horseradish peroxidase (HRP)-conjugated antiMyelin Basic Protein (MPB) antibody (1:10000; New England Biolabs, catalog number E8038, RRID:AB_1559738), HRP-conjugated anti-Myc (1:5000; Cell Signaling Technology, catalog number 2040, RRID:AB_2148465), and antihemagglutinin (HA) (1:5000; Thermo Fisher Scientific, catalog number 26183, RRID:AB_10978021) followed by secondary anti-mouse HRP-conjugated (1:5000; Cell Signaling Technology, catalog number 7076, RRID:AB_330924). 16. Chemiluminescent horseradish peroxidase (HRP) substrate (e.g., Bio-Rad, catalog number 170-5061). 17. Gel imaging system capable of detecting chemiluminescence. 18. Image analysis software (e.g., ImageJ).

200

Sophie Rizzo and Damien The´venin

2.5 Confirming Chimera Insertion Topology by Maltose Complementation Assay

1. MBP-deficient MM39 E. coli strain [20, 21]. 2. 5× M9 Minimal Salts: 210 mM Na2HPO4, 110 mM KH2PO4, 42.8 mM NaCl, 93 mM NH4Cl. 3. Agar. 4. 1 M MgSO4 (filter-sterilized). 5. 1 M CaCl2 (filter-sterilized). 6. 50% glucose (filter-sterilized). 7. 1000× ampicillin solution: 100 mg/mL (filter-sterilized). 8. 1000× kanamycin solution: 50 mg/mL (filter-sterilized).

2.6 Confirming Chimera Insertion Topology by Spheroplast Protection Assay

1. Periplasting buffer: 200 mM Tris–HCl, pH 7.5, 20% sucrose, 1 mM ethylenediaminetetraacetic acid (EDTA), and 30 U/μL Ready-Lyse Lysozyme Solution (e.g., VWR, catalog number 76081-780). 2. Lysis buffer: 10 mM Tris–HCl, pH 7.5, 50 mM KCl, 1 mM EDTA, and 0.1% deoxycholate. 3. DNA/ribonucleic acid (RNA) nonspecific endonuclease (e.g., OmniCleave Endonuclease: 50,000 U). 4. Ultra-pure water. 5. 1.0 M MgCl2 solution. 6. 500 mM EDTA solution. 7. Proteinase K (e.g., number 8107).

New

England

Biolabs,

catalog

8. Nonidet P-40 (NP-40) (e.g., Thermo Scientific, catalog number 28324).

3

Methods

3.1 Subcloning the Transmembrane Domain of Interest into AraTM Plasmids

1. Amplify the transmembrane domains of interest with designed primers following standard polymerase chain reaction (PCR) protocol (see Note 7). 2. Digest both transmembrane domains of interest alongside the AraC and AraC* plasmid backbones with SacI and KpnI restriction enzymes. 3. Run the digested product on an 0.8% agarose gel by electrophoresis (see Note 8). 4. Gel-purify the appropriate bands using a DNA gel extraction kit and following the manufacturer’s protocol. 5. Ligate the purified insert and vectors using T4 DNA ligase, following standard molecular biology techniques (see Note 9).

Assessing Transmembrane Interaction of RPTPs

201

Table 1 Setup for cotransformation of SB1676 E. coli with the GFP reporter and AraC and AraC* plasmids

Homodimerization domain A

Heterodimerization A+B

Homodimerization domain B

1

2

3

AraC

Domain A

Domain A Empty vector Domain B

AraC*

Empty vector Domain A Domain A

Empty vector Domain B Domain B

Domain B Domain A

GFP reporter

+

+

+

+

+

4

5

6

7

8

Domain B Empty vector Domain A Domain B

+

+

+

Table 2 Example of cotransformation setup to study the effect of point mutation on TM self-association Homodimerization domain A 1

2

3

AraC

Wild-type

TM mutant 1

TM mutant 2

GFP reporter

+

+

+

6. Transform in E. coli DH5α competent cells and grow using appropriate antibiotics (see Note 4). 7. Confirm via sequencing using the AraC-seq primer. 3.2 Cotransformation of AraC and AraC* Plasmids into E. coli

Specific pairs of cotransformations are required depending on whether homo- or heteroassociation is being assessed (Tables 1 and 2) (see Note 10). 1. Cotransform SB1676 cells with 200–500 ng of pAraGFPCDF, AraC, and AraC* plasmids according to Tables 1 and 2, following standard transformation protocol (see Note 11). 2. Grow on an LB selective plate (100 μg/mL Amp, 50 μg/mL Kan, Spec 100 μg/mL) (see Note 12).

3.3 DN-AraTM Assay and Analysis

The overall protocol is represented in Fig. 2. 1. Add 2 mL of LB selective medium to the appropriate number of wells in a 24-well deep-well plate (see Note 13). 2. Pick individual colonies with a pipette tip. 3. Place each tip directly into individual wells. 4. Cover the plate with adhesive film. 5. Incubate at 30 °C for 8 h with shaking (50 rpm on a plate shaker).

202

Sophie Rizzo and Damien The´venin

Fig. 2 Overview of the protocol’s key steps. (Created on Biorender.com)

6. In a new 24-well deep-well plate, add 2 mL of autoinduction media to the correct number of wells. 7. Take 20 μL of culture from each well of the 24-well growth plate, and transfer to individual wells on the 24-well plate. 8. Cover the 24-well deep-well plate and incubate overnight at 30 °C with continuous shaking at 50 rpm (see Note 14). 9. In a 96-well assay plate (black with clear flat-bottom), perform twofold serial dilutions (100 μL total volume) for each culture using ZY media (undiluted, 1:2, 1:4, and 1:8) (see Notes 15 and 16). 10. Measure the absorbance at 600 nm and GFP fluorescence emission at 510 nm using a plate reader. 11. For each sample, subtract the absorbance and fluorescence values obtained with media alone. 12. Graph fluorescence emission to absorbance for each sample and dilution (Fig. 3a). 13. Determine the slope for each sample. 14. Represent slopes as a bar graph (Fig. 3b). 3.4 Confirming Protein Expression Level by Immunoblotting

It is important to ensure that expression levels of both AraC and competitor AraC* fusions are similar to have an accurate measure of homodimer versus heterodimer formation. Expression levels of AraC and AraC* fusions can be detected by immunoblotting against the HA-tag and myc-tag, respectively. In addition, comparing expression levels for coexpression of both AraC and AraC* chimera versus expression of either AraC or AraC* chimera alone can be done by blotting against MBP.

Assessing Transmembrane Interaction of RPTPs

203

Fig. 3 (a) Fluorescence intensity at 510 nm of different pair combinations from serial dilutions of bacterial cultures plotted against the corresponding cell density (absorbance at 600 nm). The results are shown as mean ± SE (n = 9). Solid lines represent the best-fit line through the experimental data, and the shaded areas represent the upper and lower bound of the 95% confidence interval for each fit. (b) GFP signal is normalized to the signal of the sample (1), and the results are shown as mean ± SE (n = 9). The results were normalized, taking into account the slight differences in expression level measured by immunoblotting. The level of significance (one-way analysis of variance [ANOVA] with Tukey’s multiple comparisons correction, P-value 80% of the ligation product was obtained when mixing one equivalent of peptide with two equivalents of an aryl amine. Subsequently, the researchers investigated whether this new method can be used for site-directed screening by combining the formylglycine hexapeptide with a small library of different

258

Markus Tiemann and Jo¨rg Rademann

Fig. 8 (a) Amino acids 988–993 of the epidermal growth factor receptor in its phosphorylated form. (b) Hexapeptide with reactive formylglycine and five flanking amino acids important for binding to PTP1B. (c) Illustration of a primary site screening via the aldehyde-containing hexapeptide and aryl amines [43]

nucleophiles. Here, the peptide and the nucleophile were tested against PTP1B, both individually and in combination. As the five site-directing amino acids ensured some binding to PTP1B, a base level of inhibition was observed with the formylglycine peptide alone. It gave a Ki value of 484 μM for PTP1B, and 341 μM for SHP2. Upon the addition of the nucleophile, in theory, three different outcomes were conceivable (Fig. 10). First, the tested nucleophile might be no mimetic of pTyr. In this case, its addition should not lower the observed Ki value beyond the base level of the hexapeptide. This was, for instance, the case for the SF5-containing aryl amine fragment F7 (Fig. 11a). As established earlier, uncharged fragments are not active as pTyr mimetics since the surface of target phosphatases is lined with positive charges (Fig. 10), and the active site of PTPs contains a positively charged arginine binding strongly to negatively charged moieties. Second, the tested nucleophile might inhibit the phosphatase but is not a pTyr mimetic or is no longer a pTyr mimetic when ligated with the formylglycine peptide. In this case, no “over-additive effect” should be observed upon the combination of the peptide and the fragment, as could be expected for allosteric PTP inhibitors and was observed for the sulfonate fragment F3. Third, the tested nucleophile might be a pTyr mimetic, in which case an over-additive effect should be observed. This was the case for several of the tested aryl amines, of which the best are shown in Fig. 12. Since there were aryl amine hits with different properties, this allowed for further investigations to study the capabilities of this new screening method for pTyr-mimetic fragments. The NMR studies under assay conditions revealed that the reactions between the formylglycine hexapeptide and the aryl amines were fast but time-dependent. This means that the ligation assay gave lower Ki values for the ligation experiment over time. As depicted in Fig. 13, this was the case for aryl amine F1 indicating the phenyl acetic acid

Identification and Optimization of Protein Tyrosine Phosphatase Inhibitors. . .

259

Fig. 9 Different NMR studies confirmed the formation of the enamine product of the formylglycine hexapeptide and aryl amines (a and b), as well as the hydrozone product formed with hydrazines (c). Further timedependent studies (d and e) gave further insights into the reaction time as well as the position of the equilibrium between the product and starting materials [43]

moiety as a pTyr mimetic. Another hit was the triflyl-containing aryl amine F2. This moiety has been considered in the past as an “uncharged” pTyr analog, and indeed, while the sulfonamide fragment showed little activity against PTP1B (IC50: 13 mM), it became active when linked to the formylglycine peptide leading to strong over-additive effects for both PTP1B and SHP2 (Fig. 12).

260

Markus Tiemann and Jo¨rg Rademann

Fig. 10 Electrostatic surface of PTP1B and SHP2 [43]

Fig. 11 Bioassay results for the ligation of the formylglycine hexapeptide 3 with fragments F7 and F3 [43]

The low affinity of the sulfonamide fragment F2 allowed for the variation in its concentration in the assay together with the formylglycine peptide and determining the equilibrium between the reactants and the ligation product in the assay. An increase in the aryl amine concentration resulted in a decrease in the observed Ki value, and due to the low affinity of the fragment itself, this decrease can be attributed to the saturation and shift of the equilibrium reaction toward the side of the ligation product. To further confirm that the screening delivered viable pTyrmimetic fragments, the three most active moieties were incorporated into more drug-like structures and tested in enzyme assays, as well as in HeLa cells. It was found that all three product molecules inhibited SHP2 with Ki values as low as 32 nM and were able to successfully shut down p-ERK production in HeLa cells (Fig. 14). 2.3 Optimization of pTyr Mimetics Via Fragment Ligation

As shown before, fragment ligation can be a valuable tool in the identification of potent pTyr-biomimetic fragments via active sitedirected screening. Some pTyr mimetics have already shown a certain selectivity toward individual PTPs, but the extreme homology of the phosphotyrosine substrate pocket remains an important

Identification and Optimization of Protein Tyrosine Phosphatase Inhibitors. . .

261

Fig. 12 Assay results for the best identified pTyr mimetics [43]

Fig. 13 Time- and ratio-dependent bioassay results for the ligation of the formylglycine hexapeptide 3 with fragments F1 and F2 [43]

issue in modern drug development. A potent but unselective inhibitor always bears the risk of inhibiting not only the targeted but also other phosphatases. Inhibiting these undesired, so-called “off-targets” can lead to severe side effects of potential drugs. To address this selectivity problem, the group of Knapp and coworkers compared 22 crystal structures of human PTPs in 2009. They found that, even though their active sides are highly conserved, adjacent neighboring sides are less homologous [44]. These additional protein pockets present potential “secondary binding sides” that could

262

Markus Tiemann and Jo¨rg Rademann

Fig. 14 Cell assay data for the tested pyrazolone-based PTP inhibitors [43]

help to bring much needed selectivity into PTP inhibitors and, ultimately, potential therapeutics. Low-affinity fragments that bind to these secondary sites have been successfully identified in the past via NMR or X-ray crystallography studies. These methods, however, are expensive and time-consuming. Here, fragment ligation can be applied again in the form of a secondary-site-directed screening approach. In 2011, Schmidt et al. utilized a generic PTP substrate in the form of 4-formyl-phenylphosphate 47 as a screening tool to identify new fragments that specifically bind to these secondary sites. In an approach known as dynamic substrate enhancement, amine fragments can react with the aldehyde functionality of substrate 47, forming an imine as a fragment ligation product. If the fragment is recognized by a binding site adjacent to the active pocket occupied by the substrate, cooperative binding of the fragment ligation product can occur that enhances the binding and, thus, the turnover of enzymatic dephosphorylation. The general idea of this screening approach is depicted in Fig. 15. Different PTPs recognized substrate 47 and/or the formed ligation products with specific Km values and started to dephosphorylate them. After a defined amount of time, the enzyme reaction was terminated. The amount of free phosphate (Pi) released during this time was subsequently detected via the addition of malachite green, which formed a green phosphomolybdate complex with the phosphate anion. Here, three different outcomes were conceivable and are shown in Fig. 15. In the first outcome, the quantity of Pi decreased as a result of fragment addition. This means that the added amine fragment inhibited the PTP, potentially binding to the primary instead of a secondary site, as intended. In the second outcome, the amount of released Pi remained constant. In these cases, the fragment did not block the active site of the phosphatase, and the fragment ligation product

Identification and Optimization of Protein Tyrosine Phosphatase Inhibitors. . .

263

Fig. 15 Strategy of a secondary site screening approach using substrate 50 [65]

did not bind the enzyme with increased affinity. In a third outcome, the quantity of released Pi increased, meaning that the fragment enhances the binding of the substrate, most likely via forming the fragment ligation product and subsequent binding to a secondary site. To test this new screening approach, Schmidt et al. investigated four PTPs that had emerged as promising drug targets. These included the earlier discussed PTP1B and SHP2, which have been gaining increased attention as potential targets for the treatment of type II diabetes, obesity, and different forms of cancer [45]. Another human PTP that was tested is the protein tyrosine phosphatase N7, also known as PTPN7 or hematopoietic PTP (HePTP). This PTP has been shown to be overexpressed in patients that suffer from acute myeloblastic leukemia [46]. Furthermore, the bacterial PTP Mycobacterium tuberculosis protein tyrosine phosphatase A (MptpA) that is believed to mediate the survival of M. tuberculosis in host cells [47] was tested, as well.

264

Markus Tiemann and Jo¨rg Rademann

Table 11 Ki values for the identified and synthesized pTyr mimetics

Compound Chemical structure

Ki values (μM) for respective PTP

Reference

52

48 ± 12 (MptpA) 110 ± 22 (PTPN7) 12 ± 6 (PTPN11) 15 ± 8 (PTP1B)

[48, 65]

48

>500 (MptpA) >500 (PTPN7) >500 (PTPN11) >500 (PTP1B)

[65]

53

318 ± 122 (MptpA) 305 ± 102 (PTPN7) 328 ± 99 (PTPN11) 311 ± 132 (PTP1B)

[65]

54

13 ± 6 (MptpA) >300 (PTPN7) >300 (PTPN11) >300 (PTP1B)

[65]

55

11 ± 7 (MptpA) >300 (PTPN7) >300 (PTPN11) >300 (PTP1B)

[65]

Initially, the Km values of all four PTPs were determined with substrate 47, and a concentration of 250 μM was set for the screening. Each of the PTPs was then investigated together with the substrate against a small library of 110 diverse primary amines in a concentration of 500 μM. This led to the identification of structurally diverse, inhibiting as well as enhancing fragments. To exclude that the enhancing fragments increased the turnover by themselves, rather than through ligation with substrate 47, they were additionally tested in an assay with p-nitrophenylphosphate (pNPP) or 6,8-difluoro-4-methylumbelliferyl phosphate (DiFUMP) as a substrate. None of the identified hits increased the turnover of these substrates by the PTPs, excluding potential allosteric activation.

Identification and Optimization of Protein Tyrosine Phosphatase Inhibitors. . .

265

Fig. 16 Results of kinetics studies performed for the ligation assay of 47 and 48 [65]

A noteworthy result of this screening approach was the identification of compound 52 (Table 11) [65], which inhibited all four PTPs. Such 2,4-thiazolidinones have also been described as PTP inhibitors before [48]. Since amine 48 only showed activity toward MptpA, that is to say, only enhanced the substrate affinity of 47 toward MptpA, it was chosen for further validation of the method. To better understand the mechanism of the substrate enhancement, it was important to exclude that fragment 48 simply accelerated the enzymatic reaction itself. Schmidt et al., therefore, systematically varied the concentrations of both 47 and 48 in the assay and monitored the release of Pi. The results of these measurements are shown in Fig. 16. In Fig. 16a, the initial velocities v0 of enzymatic reactions were plotted against the logarithmic concentration of the substrate 47 for each concentration of 48. The apparent Km values that were obtained from this were plotted against the concentration of 48 in Fig. 16b. While the observed Km of MptpA with the substrate alone was determined to be 259 μM, increasing amounts of 48 lowered the value down to about 104 μM (Fig. 16b). At the same time, increasing concentrations of 48 did not significantly modify the observed Vmax values. This led to the conclusion that 47 and 48 bind cooperatively in the form of their fragment ligation products and increase substrate affinity while not accelerating the catalytic reaction itself.

266

Markus Tiemann and Jo¨rg Rademann

Fig. 17 Docking results for 53 and 55 [65]

To further prove the success of the method, the postulated ligation product of 47 and 48 was converted into a covalently linked inhibitor. The imine moiety was replaced by either an amide or a reduced amine, while the phenyl phosphate group was replaced by the pTyr mimetic fragment 4-((trifluoromethyl)sulfonamido)benzoic acid 53 (Table 11). Compounds 48, 53, 54, and 55 were then tested once more in an established pNPP and DiFUMP assay against all four PTPs. Both 54 and 55 yielded Ki values of 13 μM and 11 μM against MptpA, respectively, while showing no activity against the other three PTPs at 300 μM. Additional molecular modeling experiments further supported these findings (Fig. 17). Fragment 53 was able to bind in the substrate pocket of MptpA. Compound 55 was also able to bind there while at the same time accessing an adjacent, secondary site. In summary, fragment ligation screening has proven successful both for the discovery of phosphotyrosine-mimetic fragments of PTP and for the identification of second-site-specific fragments of PTP, enabling the preparation of cellular active PTP inhibitors.

Acknowledgements The authors acknowledge funding by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) – Project-ID 387284271 – SFB 1349 (fluorine-specific interactions, project A3). References 1. Edwards DR (1994) Cell signalling and the control of gene transcription. Trends Pharmacol Sci 15(7):239–244. https://doi.org/10. 1016/0165-6147(94)90318-2 2. Tarrant MK, Cole PA (2009) The chemical biology of protein phosphorylation. Annu

Rev Biochem 78(1):797–825. https://doi. org/10.1146/annurev.biochem.78.070907. 103047 3. Ardito F, Giuliani M, Perrone D, Troiano G, Lo Muzio L (2017) The crucial role of protein phosphorylation in cell signaling and its use as

Identification and Optimization of Protein Tyrosine Phosphatase Inhibitors. . . targeted therapy (review). Int J Mol Med 40(2):271–280. https://doi.org/10.3892/ ijmm.2017.3036 4. Johnson TO, Ermolieff J, Jirousek MR (2002) Protein tyrosine phosphatase 1B inhibitors for diabetes. Nat Rev Drug Discov 1(9):696–709. https://doi.org/10.1038/nrd895 5. Combs AP (2010) Recent advances in the discovery of competitive protein tyrosine phosphatase 1B inhibitors for the treatment of diabetes, obesity, and cancer. J Med Chem 53(6):2333–2344. https://doi.org/10.1021/ jm901090b 6. Morlacchi P, Robertson FM, Klostergaard J, McMurray JS (2014) Targeting SH2 domains in breast cancer. Future Med Chem 6(17): 1909–1926. https://doi.org/10.4155/fmc. 14.120 7. Tautz L, Critton DA, Grotegut S (2013) Protein tyrosine phosphatases: structure, function, and implication in human disease. In: Milla´n JL (ed) Phosphatase modulators. Humana Press, Totowa, pp 179–221. https://doi.org/10. 1007/978-1-62703-562-0_13 8. Johnson DE, O’Keefe RA, Grandis JR (2018) Targeting the IL-6/JAK/STAT3 signalling axis in cancer. Nat Rev Clin Oncol 15(4): 2 3 4 – 2 4 8 . h t t p s : // d o i . o r g / 1 0 . 1 0 3 8 / nrclinonc.2018.8 9. Neel BG, Gu H, Pao L (2003) The ‘Shp’ing news: SH2 domain-containing tyrosine phosphatases in cell signaling. Trends Biochem Sci 28(6):284–293. https://doi.org/10. 1016/S0968-0004(03)00091-4 10. Noble MEM, Endicott JA, Johnson LN (2004) Protein kinase inhibitors: insights into drug design from structure. Science 303(5665):1800–1805. https://doi.org/10. 1126/science.1095920 11. Bossemeyer D (1995) Protein kinases – structure and function. FEBS Lett 369(1):57–61. https://doi.org/10.1016/0014-5793(95) 00580-3 12. Hubbard SR, Wei L, Hendrickson WA (1994) Crystal structure of the tyrosine kinase domain of the human insulin receptor. Nature 372(6508):746–754. https://doi.org/10. 1038/372746a0 13. Mohammadi M, Schlessinger J, Hubbard SR (1996) Structure of the FGF receptor tyrosine kinase domain reveals a novel autoinhibitory mechanism. Cell 86(4):577–587. https://doi. org/10.1016/S0092-8674(00)80131-2 14. Xu W, Harrison SC, Eck MJ (1997) Threedimensional structure of the tyrosine kinase c-Src. Nature 385(6617):595–602. https:// doi.org/10.1038/385595a0

267

15. McTigue MA, Wickersham JA, Pinko C, Showalter RE, Parast CV, Tempczyk-Russell A, Gehring MR, Mroczkowski B, Kan C-C, Villafranca JE, Appelt K (1999) Crystal structure of the kinase domain of human vascular endothelial growth factor receptor 2: a key enzyme in angiogenesis. Structure 7(3):319–330. https://doi.org/10.1016/S0969-2126(99) 80042-2 16. Burke TR Jr, Yao Z-J, Liu D-G, Voigt J, Gao Y (2001) Phosphoryltyrosyl mimetics in the design of peptide-based signal transduction inhibitors. Pept Sci 60(1):32–44. https://doi. org/10.1002/1097-0282(2001)60:13.0.CO;2-I 17. Al-Obeidi FA, Wu JJ, Lam KS (1998) Protein tyrosine kinases: structure, substrate specificity, and drug discovery. Pept Sci 47(3):197–223. https://doi.org/10.1002/(SICI)1097-0282 (1998)47:33.0.CO;2-H 18. Alfaro-Lopez J, Yuan W, Phan BC, Kamath J, Lou Q, Lam KS, Hruby VJ (1998) Discovery of a novel series of potent and selective substrate-based inhibitors of p60c-src protein tyrosine kinase: conformational and topographical constraints in peptide design. J Med Chem 41(13):2252–2260. https://doi.org/ 10.1021/jm9707885 19. Shoelson SE, Chatterjee S, Chaudhuri M, Burke TR (1991) Solid-phase synthesis of nonhydrolyzable phosphotyrosyl peptide analogues with Nα-Fmoc-(O,O-di-t-butyl) phosphono-p-methylphenylalanine. Tetrahedron Lett 32(43):6061–6064. https://doi. org/10.1016/0040-4039(91)80753-S 20. Domchek SM, Auger KR, Chatterjee S, Burke TR Jr, Shoelson SE (1992) Inhibition of SH2 domain/phosphoprotein association by a nonhydrolyzable phosphonopeptide. Biochemistry 31(41):9865–9870. https://doi.org/10. 1021/bi00156a002 21. Makukhin N, Ciulli A (2021) Recent advances in synthetic and medicinal chemistry of phosphotyrosine and phosphonate-based phosphotyrosine analogues. RSC Med Chem 12(1): 8 – 2 3 . h t t p s : // d o i . o r g / 1 0 . 1 0 3 9 / D0MD00272K 22. Giorgetti-Peraldi S, Ottinger E, Wolf G, Ye B, Burke TR, Shoelson SE (1997) Cellular effects of phosphotyrosine-binding domain inhibitors on insulin receptor signaling and trafficking. Mol Cell Biol 17(3):1180–1188. https://doi. org/10.1128/MCB.17.3.1180 23. Zhang ZY, Maclean D, McNamara DJ, Sawyer TK, Dixon JE (1994) Protein tyrosine phosphatase substrate specificity: size and phosphotyrosine positioning requirements in peptide

268

Markus Tiemann and Jo¨rg Rademann

substrates. Biochemistry 33(8):2285–2290. https://doi.org/10.1021/bi00174a040 24. Burke TR, Kole HK, Roller PP (1994) Potent inhibition of insulin receptor dephosphorylation by a hexamer peptide containing the phosphotyrosyl mimetic F2Pmp. Biochem Biophys Res Commun 204(1):129–134. https://doi. org/10.1006/bbrc.1994.2435 25. Atabey N, Gao Y, Yao Z-J, Breckenridge D, Soon L, Soriano JV, Burke TR, Bottaro DP (2001) Potent blockade of hepatocyte growth factor-stimulated cell motility, matrix invasion and branching morphogenesis by antagonists of Grb2 Src homology 2 domain interactions*. J Biol Chem 276(17):14308–14314. https:// doi.org/10.1074/jbc.M010202200 26. Gao Y, Wu L, Luo JH, Guo R, Yang D, Zhang Z-Y, Burke TR (2000) Examination of novel non-phosphorus-containing phosphotyrosyl mimetics against protein-tyrosine phosphatase-1B and demonstration of differential affinities toward Grb2 SH2 domains. Bioorg Med Chem Lett 10(9):923–927. https://doi.org/ 10.1016/S0960-894X(00)00124-4 27. Tong L, Warren TC, Lukas S, Schembri-King J, Betageri R, Proudfoot JR, Jakes S (1998) Carboxymethyl-phenylalanine as a replacement for phosphotyrosine in SH2 domain binding*. J Biol Chem 273(32):20238–20242. https:// doi.org/10.1074/jbc.273.32.20238 28. Szczepankiewicz BG, Liu G, Hajduk PJ, AbadZapatero C, Pei Z, Xin Z, Lubben TH, Trevillyan JM, Stashko MA, Ballaron SJ, Liang H, Huang F, Hutchins CW, Fesik SW, Jirousek MR (2003) Discovery of a potent, selective protein tyrosine phosphatase 1B inhibitor using a linked-fragment strategy. J Am Chem Soc 125(14):4087–4096. https://doi.org/10. 1021/ja0296733 29. Liu G, Xin Z, Pei Z, Hajduk PJ, AbadZapatero C, Hutchins CW, Zhao H, Lubben TH, Ballaron SJ, Haasch DL, Kaszubska W, Rondinone CM, Trevillyan JM, Jirousek MR (2003) Fragment screening and assembly: a highly efficient approach to a selective and cell active protein tyrosine phosphatase 1B inhibitor. J Med Chem 46(20):4232–4235. https:// doi.org/10.1021/jm034122o 30. Grosskopf S, Eckert C, Arkona C, Radetzki S, Bo¨hm K, Heinemann U, Wolber G, von Kries J-P, Birchmeier W, Rademann J (2015) Selective inhibitors of the protein tyrosine phosphatase SHP2 block cellular motility and growth of cancer cells in vitro and in vivo. ChemMedChem 10(5):815–826. https://doi.org/10. 1002/cmdc.201500015 31. Combs AP, Yue EW, Bower M, Ala PJ, Wayland B, Douty B, Takvorian A, Polam P,

Wasserman Z, Zhu W, Crawley ML, Pruitt J, Sparks R, Glass B, Modi D, McLaughlin E, Bostrom L, Li M, Galya L, Blom K, Hillman M, Gonneville L, Reid BG, Wei M, Becker-Pasha M, Klabe R, Huber R, Li Y, Hollis G, Burn TC, Wynn R, Liu P, Metcalf B (2005) Structure-based design and discovery of protein tyrosine phosphatase inhibitors incorporating novel isothiazolidinone heterocyclic phosphotyrosine mimetics. J Med Chem 48(21):6544–6548. https://doi.org/10. 1021/jm0504555 32. New inhibitors of SH2-mediated processes (1998) Expert Opin Ther Patents 8(3): 3 3 3 – 3 3 5 . h t t p s : // d o i . o r g / 1 0 . 1 5 1 7 / 13543776.8.3.333 33. Huang Y-Y, Kuo C-C, Chu C-Y, Huang Y-H, Hu Y-L, Lin J-J, Lo L-C (2010) Development of activity-based probes with tunable specificity for protein tyrosine phosphatase subfamilies. Tetrahedron 66(25):4521–4529. https://doi. org/10.1016/j.tet.2010.04.065 34. van Ameijde J, Overvoorde J, Knapp S, den Hertog J, Ruijtenbeek R, Liskamp RMJ (2013) Real-time monitoring of the dephosphorylating activity of protein tyrosine phosphatases using microarrays with 3-nitrophosphotyrosine substrates. ChemPlusChem 78(11):1349–1357. https://doi. org/10.1002/cplu.201300299 35. Horatscheck A, Wagner S, Ortwein J, Kim BG, Lisurek M, Beligny S, Schu¨tz A, Rademann J (2012) Benzoylphosphonate-based photoactive phosphopeptide mimetics for modulation of protein tyrosine phosphatases and highly specific labeling of SH2 domains. Angew Chem Int Ed 51(37):9441–9447. https:// doi.org/10.1002/anie.201201475 36. Wong EL, Nawrotzky E, Arkona C, Kim BG, Beligny S, Wang X, Wagner S, Lisurek M, Carstanjen D, Rademann J (2019) The transcription factor STAT5 catalyzes mannich ligation reactions yielding inhibitors of leukemic cell proliferation. Nat Commun 10(1):66. https://doi.org/10.1038/s41467-01807923-2 37. Wagner S, Schu¨tz A, Rademann J (2015) Light-switched inhibitors of protein tyrosine phosphatase PTP1B based on phosphonocarbonyl phenylalanine as photoactive phosphotyrosine mimetic. Bioorg Med Chem 23(12): 2839–2847. https://doi.org/10.1016/j.bmc. 2015.03.074 38. Accorsi M, Tiemann M, Wehrhan L, Finn LM, Cruz R, Rautenberg M, Emmerling F, Heberle J, Keller BG, Rademann J (2022) Pentafluorophosphato-phenylalanines: amphiphilic phosphotyrosine mimetics displaying

Identification and Optimization of Protein Tyrosine Phosphatase Inhibitors. . . fluorine-specific protein interactions. Angewandte Chemie International Edition. https://doi.org/10.1002/anie.202203579 39. Wagner S, Accorsi M, Rademann J (2017) Benzyl mono-P-fluorophosphonate and benzyl penta-P-fluorophosphate anions are physiologically stable phosphotyrosine mimetics and inhibitors of protein tyrosine phosphatases. Chemistry 23(61):15387–15395. https://doi. org/10.1002/chem.201701204 40. Jia Z, Barford D, Flint AJ, Tonks NK (1995) Structural basis for phosphotyrosine peptide recognition by protein tyrosine phosphatase 1B. Science 268(5218):1754–1758. https:// doi.org/10.1126/science.7540771 41. Zhang ZY, Thieme-Sefler AM, Maclean D, McNamara DJ, Dobrusin EM, Sawyer TK, Dixon JE (1993) Substrate specificity of the protein tyrosine phosphatases. Proc Natl Acad Sci 90(10):4446–4450. https://doi.org/10. 1073/pnas.90.10.4446 42. Jaegle M, Wong EL, Tauber C, Nawrotzky E, Arkona C, Rademann J (2017) Proteintemplated fragment ligations – from molecular recognition to drug discovery. Angew Chem Int Ed 56(26):7358–7378. https://doi.org/ 10.1002/anie.201610372 43. Tiemann M, Nawrotzky E, Schmieder P, Wehrhan L, Bergemann S, Martos V, Song W, Arkona C, Keller BG, Rademann J (2022) A formylglycine-peptide for the site-directed identification of Phosphotyrosine-mimetic fragments**. Chemistry 28(57):e202201282. https://doi.org/10.1002/chem.202201282 44. Barr AJ, Ugochukwu E, Lee WH, King ONF, Filippakopoulos P, Alfano I, Savitsky P, Burgess-Brown NA, Mu¨ller S, Knapp S (2009) Large-scale structural analysis of the classical human protein tyrosine phosphatome. Cell 136(2):352–363. https://doi.org/10. 1016/j.cell.2008.11.038 45. Elchebly M, Payette P, Michaliszyn E, Cromlish W, Collins S, Loy AL, Normandin D, Cheng A, Himms-Hagen J, Chan C-C, Ramachandran C, Gresser MJ, Tremblay ML, Kennedy BP (1999) Increased insulin sensitivity and obesity resistance in mice lacking the protein tyrosine phosphatase-1B gene. Science 283(5407):1544–1548. https://doi.org/10.1126/science.283.5407. 1544 46. Zanke B, Squire J, Griesser H, Henry M, Suzuki H, Patterson B, Minden M, Mak TW (1994) A hematopoietic protein tyrosine phosphatase (HePTP) gene that is amplified and overexpressed in myeloid malignancies maps to chromosome 1q32.1. Leukemia 8(2): 236–244

269

47. Bach H, Papavinasasundaram KG, Wong D, Hmama Z, Av-Gay Y (2008) Mycobacterium tuberculosis virulence is mediated by PtpA dephosphorylation of human vacuolar protein sorting 33B. Cell Host Microbe 3(5): 316–322. https://doi.org/10.1016/j.chom. 2008.03.008 48. Maccari R, Ottana` R, Ciurleo R, Paoli P, Manao G, Camici G, Laggner C, Langer T (2009) Structure-based optimization of benzoic acids as inhibitors of protein tyrosine phosphatase 1B and low molecular weight protein tyrosine phosphatase. ChemMedChem 4(6):957–962. https://doi.org/10.1002/ cmdc.200800427 49. Burke TR, Yao Z-J, Zhao H, Milne GWA, Wu L, Zhang Z-Y, Voigt JH (1998) Enantioselective synthesis of nonphosphoruscontaining phosphotyrosyl mimetics and their use in the preparation of tyrosine phosphatase inhibitory peptides. Tetrahedron 54(34): 9981–9994. https://doi.org/10.1016/ S0040-4020(98)00590-0 50. Marseigne I, Roques BP (1988) Synthesis of new amino acids mimicking sulfated and phosphorylated tyrosine residues. J Org Chem 53(15):3621–3624. https://doi.org/10. 1021/jo00250a043 51. Burke TR Jr, Russ P, Lim B (1991) Synthesis 11:1019–1020. https://doi.org/10.1055/s1991-26636 52. Lai JH, Marsilje TH, Choi SUN, Nair SA, Hangauer DG (1998) The design, synthesis and activity of pentapeptide pp60c-src inhibitors containing L-phosphotyrosine mimics. J Pept Res 51(4):271–281. https://doi.org/10. 1111/j.1399-3011.1998.tb00424.x 53. Smyth MS, Ford H, Burke TR (1992) A general method for the preparation of benzylic α,α-difluorophosphonic acids; non-hydrolyzable mimemtics of phosphotyrosine. Tetrahedron Lett 33(29):4137–4140. https://doi.org/10. 1016/S0040-4039(00)74672-7 54. Burke TR Jr, Smyth MS, Nomizu M, Otaka A, Roller PR (1993) Preparation of fluoro- and hydroxy-4-(phosphonomethyl)-D,L-phenylalanine suitably protected for solid-phase synthesis of peptides containing hydrolytically stable analogs of O-phosphotyrosine. J Org Chem 58(6):1336–1340. https://doi.org/10.1021/ jo00058a009 55. Burke TR Jr, Smyth MS, Otaka A, Nomizu M, Roller PP, Wolf G, Case R, Shoelson SE (1994) Nonhydrolyzable phosphotyrosyl mimetics for the preparation of phosphatase-resistant SH2 domain inhibitors. Biochemistry 33(21): 6490–6494. https://doi.org/10.1021/ bi00187a015

270

Markus Tiemann and Jo¨rg Rademann

56. Yao Z-J, King CR, Cao T, Kelley J, Milne GWA, Voigt JH, Burke TR (1999) Potent inhibition of Grb2 SH2 domain binding by nonphosphate-containing ligands. J Med Chem 42(1):25–35. https://doi.org/10.1021/ jm980388x 57. Ye B, Akamatsu M, Shoelson SE, Wolf G, Giorgetti-Peraldi S, Yan X, Roller PP, Burke TR Jr (1995) L-O-(2-Malonyl)tyrosine: a new phosphotyrosyl mimetic for the preparation of Src homology 2 domain inhibitory peptides. J Med Chem 38(21):4270–4275. https://doi. org/10.1021/jm00021a016 58. Gao Y, Luo J, Yao Z-J, Guo R, Zou H, Kelley J, Voigt JH, Yang D, Burke TR (2000) Inhibition of Grb2 SH2 domain binding by nonphosphate-containing ligands. 2. 4-(2-Malonyl)phenylalanine as a potent phosphotyrosyl mimetic. J Med Chem 43(5):911–920. https://doi.org/10.1021/jm9904248 59. Miller MJ, Braccolino DS, Clearly DG, Ream JE, Walker MC, Sikorski JA (1994) EPSP synthase inhibitor design IV. New aromatic substrate analogs and symmetrical inhibitors containing novel 3-phosphate mimics. Bioorg Med Chem Lett 4(21):2605–2608. https:// doi.org/10.1016/S0960-894X(01)80293-6 60. Gao Y, Burke TR Jr (2000) Stereoselective preparation of L -4-(2′-Malonyl)phenylalanine suitably protected for Fmoc-based synthesis of potent signal transduction inhibitory ligands. Synlett 1:134–136. https://doi.org/ 10.1055/s-2000-6464 61. Beaulieu PL, Cameron DR, Ferland J-M, Gauthier J, Ghiro E, Gillard J, Gorys V, Poirier M, Rancourt J, Wernic D, LlinasBrunet M, Betageri R, Cardozo M, Hickey ER, Ingraham R, Jakes S, Kabcenell A, Kirrane T, Lukas S, Patel U, Proudfoot J,

Sharma R, Tong L, Moss N (1999) Ligands for the tyrosine kinase p56lck SH2 domain: discovery of potent dipeptide derivatives with monocharged, nonhydrolyzable phosphate replacements. J Med Chem 42(10): 1757–1766. https://doi.org/10.1021/ jm980676t 62. Yao Z-J, Gao Y, Voigt JH, Ford H, Burke TR (1999) Synthesis of Fmoc-protected 4-carboxydifluoromethyl-L-phenylalanine: a phosphotyrosyl mimetic of potential use for signal transduction studies. Tetrahedron 55(10): 2865–2874. https://doi.org/10.1016/ S0040-4020(99)00076-9 63. Hellmuth K, Grosskopf S, Lum CT, Wu¨rtele M, Ro¨der N, von Kries J-P, Rosario M, Rademann J, Birchmeier W (2008) Specific inhibitors of the protein tyrosine phosphatase Shp2 identified by highthroughput docking. PNAS 105(20): 7275–7280 64. Mu¨ller G (2001) Peptidomimetic SH2 domain antagonists for targeting signal transduction. In: Waldmann H (ed) Bioorganic chemistry of biological signal transduction. Springer, Berlin/Heidelberg, pp 17–59. https://doi.org/10.1007/3-54045035-1_2 65. Schmidt MF, Groves MR, Rademann J (2011) Dynamic substrate enhancement for the identification of specific, second-site-binding fragments targeting a set of protein tyrosine phosphatases. Chembiochem 12(17): 2640–2646. https://doi.org/10.1002/cbic. 201100414 66. Accorsi M (2022) Doctoral thesis: synthesis and biological evaluation of pentafluorophosphates as lipophilic non-cleavable phosphatase inhibitors. Freie Universit€at, Berlin

Chapter 17 Targeting Nonconserved and Pathogenic Cysteines of Protein Tyrosine Phosphatases with Small Molecules Anthony C. Bishop and Anna Serbina Abstract Protein tyrosine phosphatases (PTPs) are important therapeutic targets for a range of human pathologies. However, the common architecture of PTP active sites impedes the discovery of selective PTP inhibitors. Our laboratory has recently developed methods to inhibit PTPs allosterically by targeting cysteine residues that either (i) are not conserved in the PTP family or (ii) result from pathogenic mutations. Here, we describe screening protocols for the identification of selective inhibitors that covalently engage such “rare” cysteines in target PTPs. Moreover, to elucidate the breadth of possible applications of our cysteine-directed screening protocols, we provide a brief overview of the nonconserved cysteines present in all human classical PTP domains. Key words Protein tyrosine phosphatases, Inhibitor screening, Irreversible inhibition, Nonconserved cysteine residues, SHP2

1

Introduction The protein tyrosine phosphatases (PTPs) constitute a family of cell-signaling enzymes that catalyze the dephosphorylation of phosphotyrosine in protein substrates [1–3]. Improperly regulated PTP activity has been implicated as a causative agent in a range of human diseases—including cancer, diabetes, and neurodegenerative disorders—and selective PTP inhibitors are desirable tools for signal transduction research and as potential drug candidates [4–7]. However, active-site-directed PTP inhibitor discovery is inherently difficult due to a lack of target specificity, as classical PTP active-site domains share a high degree of sequence and structural homology [8, 9]. Allosteric inhibition represents an attractive approach for circumventing the difficulties that have limited the successful development of active-site-directed PTP inhibitors, as allosteric sites are

Damien The´venin and Jo¨rg P. Mu¨ller (eds.), Protein Tyrosine Phosphatases: Methods and Protocols, Methods in Molecular Biology, vol. 2743, https://doi.org/10.1007/978-1-0716-3569-8_17, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

271

272

Anthony C. Bishop and Anna Serbina

typically much less strongly conserved than PTP-active sites. Unfortunately, few allosteric sites on PTP domains are known, and reversible allosteric PTP inhibitors that target these sites generally act with moderate potency [10–12]. Our strategies for allosteric PTP inhibitor discovery utilize the unique nucleophilic reactivity of cysteine, which potentially provides a potent handle for the development of covalent, irreversible inhibitors [13]. Specifically, our lab has focused on developing inhibitors that can engage cysteines outside the active site and are not conserved in the PTP family [14–16]. (The PTP active site also contains a completely conserved cysteine residue necessary for the catalytic mechanism [2]. Inhibitors that engage the active-site cysteine are, as a whole, not selective for individual PTPs [17].) While the focus of this chapter is on PTPs, it is important to note that nonconserved cysteines have also been targeted to achieve selective inhibition in other proteins—examples include anti-cancer drug targets, such as the protein kinases [13, 18] and KRAS [19, 20]. In seeking to target nonconserved cysteines for the potential discovery of selective PTP inhibitors, it should be clarified that no perfect definition of “nonconserved” exists. We have chosen to use the term to denote a cysteine that appears at a particular position of the conserved PTP domain in no more than five classical human PTPs. In other words, for any nonconserved cysteine targeted, no more than four off-target PTPs would have a cysteine at the corresponding position. (This is less than 15% of the off-target PTP domains encoded by the human genome [2].) We have analyzed all human classical PTPs by the “five-or-fewer” definition and found that the PTPome contains 144 nonconserved cysteine residues at 77 distinct positions in a PTP domain alignment, with every human PTP domain having at least one nonconserved cysteine [2]. As shown in Fig. 1, the locations of these nonconserved cysteine residues are distributed around much of the threedimensional structure of the conserved PTP domain fold, potentially providing ample opportunities for cysteine-directed PTP ligand discovery. Further possibilities for targeting nonconserved cysteines in PTPs can be found in cases of disease-causing cysteine mutations; when a cysteine mutation in a PTP is pathogenic, the causative agent presents a potentially targetable molecular handle [21]. In essence, pathogenic cysteines can be viewed as the limiting case of “nonconserved” in that the targeted cysteine does not even exist in the wild-type version of the targeted protein. For example, the pathogenic mutation of Y279C Src-homology-2-containing PTP 2 (SHP2), which causes the developmental disorder Noonan syndrome with multiple lentigines (NSML) [22–25], is at a position of the PTP domain that is not occupied by cysteine in any other human PTP [2].

Screen for Allosteric Cysteine-Directed PTP Inhibitors

273

Fig. 1 Locations of nonconserved cysteine residues in human classical PTP domains. The structure of a prototypical PTP domain (PTP1B, Protein Data Bank (PDB) code: 1OEM [29]) is shown as a ribbon diagram, with the signature motif of the PTP active site shown in red. Amino-acid positions that are occupied by a cysteine residue in at least one and no more than five human PTPs are shown in yellow. One example of a nonconserved cysteine that our lab has targeted previously [14, 15] is highlighted: C333 of SHP2, which occurs at a position that is occupied by proline in almost all other PTPs (e.g., P87 of PTP1B) [2]

Our laboratory has recently shown that both nonconserved (wild-type) cysteines [15] and pathogenic cysteine mutants [16] can be effectively targeted using the screening protocols detailed here (Fig. 2). As a starting point, a library of compounds that contain cysteine-reactive electrophiles must be obtained, either commercially or through small molecule–molecule synthesis. The library can first be screened for the inhibition of either the target wild-type PTP (Fig. 2a) or pathogenic mutant (Fig. 2b). Central to both the strategies is employing counter-screens on active compounds to weed out potential false-positives quickly. In the case of a nonconserved cysteine target in a wild-type PTP (e.g., C333 of SHP2 [14]), the counter-screen is run on a mutant in which the nonconserved cysteine has been replaced with the residue that appears at the corresponding position of most other PTPs (e.g., C333P SHP2, Fig. 2a) [2]. When a pathogenic cysteine is targeted, counter-screening is carried out against the corresponding wildtype enzyme (Fig. 2b). It is important to note that the screening protocols detailed in this report do not definitively establish that a selected hit compound covalently engages a target cysteine residue. However, as the screening and counter-screening PTPs differ by

274

Anthony C. Bishop and Anna Serbina

a Screen electrophile library with wildtype PTP

Counter screen with mutant PTP lacking target cysteine

Potential cysteine-directed PTP inhibitor(s)

b Screen electrophile library with pathogenic cysteine mutant PTP Counter screen with wild-type PTP

Potential specific inhibitor(s) of pathogenic cysteine mutant PTP

Fig. 2 Screening strategies for the identification of cysteine-directed PTP inhibitors. (a) Targeting a nonconserved cysteine in a wild-type PTP: A library of compounds containing cysteine-reactive electrophiles is screened for inhibition of the target wild-type PTP. Compounds that show strong (>50%) inhibition of the target are counter-screened on a mutant that lacks the target cysteine. Inhibitors of the wild-type PTP that are unable to inhibit the mutant are identified as strong candidates for cysteine-directed inhibition. (b) Targeting a pathogenic cysteine mutant PTP: A compound library is screened for the inhibition of a pathogenic mutant PTP. Compounds that show strong (>50%) inhibition of the target are counter-screened on the corresponding wildtype PTP. Inhibitors that act on the pathogenic mutant PTP but not the wild-type are identified as strong candidates for cysteine-directed inhibition

only a single residue (they either contain or lack the target cysteine), in our experience, a compound that emerges as a hit will likely show cysteine-targeted selectivity in follow-up experiments that are designed to directly measure target engagement (e.g., mass spectrometry) [15, 16]. Because we have previously used the general strategy outlined in Fig. 2a to identify a C333-directed inhibitor of wild-type SHP2 [15], the catalytic domain of SHP2 (SHPcat) and its corresponding counter-screening mutant (C333P SHP2cat) will be specifically

Screen for Allosteric Cysteine-Directed PTP Inhibitors

275

referenced in the detailed protocols that appear below. Importantly, however, the general approaches and strategies outlined in the following protocols can be applied to essentially any nonconserved or pathogenic cysteine in any target PTP of interest.

2

Materials

2.1 Electrophilic Compound Libraries

1. Commercially available compound libraries: Libraries of compounds that contain cysteine-reactive electrophiles are available from various sources. We have used a 1383-compound cysteine-focused library from Life Chemicals to identify a specific inhibitor of Y279C SHP2 [16]. In addition, a 993-compound library of cysteine-reactive fragments (enamine) has been fruitful for engaging previously untargeted cysteine residues in several proteins [26]. Individual cysteine-reactive compounds can also be purchased. Some curated lists of commercially available compounds containing cysteine-reactive electrophiles have been described in the literature (see, for example, the 63-compound library compiled by Backus et al. [27]). 2. In-house synthesized libraries: A detailed description of the organic synthesis of electrophile-containing compound libraries is beyond the scope of this report. We have found, though, that resynthesis of compound libraries that have yielded cysteine-specific inhibitors in other protein families can be a fruitful way to identify cysteine-specific PTP inhibitors. For example, we have resynthesized and “remined” a small collection of cyanoacrylamides toward discovering C333directed SHP2 inhibitors [15].

2.2 PTP Expression and Purification

1. pET-based plasmid that encodes target-cysteine-containing PTP domain as a six-His-tagged protein, for example, the SHP2 catalytic domain (SHP2cat). 2. pET-based plasmid that encodes control PTP with target cysteine removed as a six-His-tagged protein, for example, the C333P SHP2 catalytic domain (C333P SHP2cat). 3. Competent BL21(DE3) Escherichia coli. 4. Liquid Luria Broth (LB). 5. Ampicillin, dissolved in water as 100 mg/mL stock solution. 6. Isopropyl-1-thio-β-D-galactopyranoside (IPTG), dissolved in water as 1 M stock solution. 7. Bacterial Protein Extraction Reagent (BPER). 8. Thermo Scientific HisPur Ni-NTA beads. 9. Resuspension solution: 50 mM tris(hydroxymethyl)aminomethane (Tris), pH 7.8, 500 mM NaCl, 5 mM imidazole, 1 mM TCEP.

276

Anthony C. Bishop and Anna Serbina

10. Ni-NTA wash solution: 50 mM Tris, pH 7.8, 500 mM NaCl, 20 mM imidazole, 1 mM TCEP. 11. Ni-NTA elution solution: 50 mM Tris, pH 7.8, 500 mM NaCl, 250 mM imidazole, 1 mM TCEP. 12. PTP storage solution: 50 mM 3,3-dimethyl glutarate, pH 7.0, 1 mM EDTA, 1 mM TCEP. 13. Corning Spin-X UF 20-mL concentrators (30-kDa cut-off). 2.3 ElectrophileContaining Compound Screening

1. 4× PTP activity solution: 100 mM MOPS, pH 7, 200 mM NaCl, 0.2% Tween-20, 4 mM DTT 2. Dimethyl sulfoxide (DMSO). 3. Electrophilic compound Subheading 2.1).

library

to

be

screened

(see

4. para-Nitrophenyl phosphate ( pNPP), 50 mM, dissolved in water, 5. 6,8-difluoro-4-methylumbiliferyl phosphate 0.80 mM dissolved in water/DMSO (9:1).

3

(DiFMUP),

Methods

3.1 Expression and Purification of PTP Domains

1. Grow a culture of a BL21 (DE3) strain containing the appropriate PTP-expressing plasmid in liquid LB supplemented with 100 μg/mL ampicillin until an absorbance (600 nM) of 0.5 is reached. 2. Add isopropyl-1-thio-β-D-galactopyranoside (IPTG) to a final concentration of 0.2 mM and grow for 16 h at room temperature (see Note 1). Harvest the cells by centrifugation (10,000 ×g for 15 min), discard the supernatant, and store the pellets at -80 °C. 3. Lyse the cells by resuspending the pellets in BPER (10 mL per 250 mL of bacterial culture). Incubate at room temperature for 15 min with agitation. 4. Clarify the lysate by centrifugation (15,000 ×g for 5 min) and discard the pellet. 5. Incubate the supernatant with Thermo Scientific HisPur Ni-NTA beads (1 mL per 250 mL of bacterial culture) for 30 min with agitation at 4 °C. Collect the beads by centrifugation (1 min at 1000 rpm) and resuspend in Ni-NTA wash solution. Repeat the wash procedure (pelleting/resuspension) five times. 6. Elute the proteins in 2 mL of elution solution and filter off the beads, collecting the eluent.

Screen for Allosteric Cysteine-Directed PTP Inhibitors

277

7. To exchange purified proteins into PTP storage solution, place the eluted protein in the inner chamber of a Corning Spin-X UF 20-mL concentrator and fill the inner chamber to a total volume of 20 mL with PTP storage solution. Concentrate the solution to 1 mL by centrifugation at 3500 ×g. Refill the inner chamber with PTP storage solution to a total volume of 20 mL and reconcentrate to 0.5–1.0 mL of solution by centrifugation at 3500 ×g (see Note 2). 8. Aliquot the purified PTP solutions in prechilled thin-walled PCR tubes (25 μL), flash-freeze in liquid nitrogen, and store in a -80 °C freezer (see Note 3). 3.2 Screen for Target-CysteineDirected Inhibitors Using pNPP as Substrate

1. Obtain a structurally diverse library of compounds that contain cysteine-reactive electrophiles as stock solutions in DMSO (see Subheading 2.1 and Note 4). Dilute compounds to 5 mM in DMSO (see Note 5), preferably in 96-deep-well storage plates. 2. Transfer 2 μL of each compound solution (or DMSO control) solution to a clear, flat-bottomed 96-well microplate via a multichannel pipette (see Note 6). 3. Prepare a “master mix” for the pNPP reaction that contains the remaining PTP reaction components (with the exception of the pNPP substrate) multiplied by the number of reactions to be run. For each PTP reaction to being run, the mix should contain: 128 μL of deionized water, 50 μL of 4× PTP activity solution, and 10 μL of 2 μM SHP2cat (diluted in PTP storage solution, see Note 7). 4. Using a multichannel pipette, transfer 188 μL of the master mix to each well that contains a test compound (or DMSO control) in the 96-well microplate. 5. Incubate the 96-well plate at 22 °C for 1 h (see Note 8). 6. Initiate PTP reactions by adding 10 μL of 50 mM pNPP to each well with a multichannel pipette, staggering each row by 30 s, and mixing by agitation between each addition (see Note 9). 7. After 10 min, quench the PTP reactions by the addition of 40 μL of 5 M NaOH with a multichannel pipette in 30-s intervals. 8. Measure the absorbance at 405 nM (A405) of each well and correct by subtraction of the A405 of a suitable background well (see Note 10). 9. Calculate the %PTP activity for each well by dividing the corrected absorbance of each compound-containing well by the absorbance of the DMSO controls. 10. Potential hits are defined as compounds that inhibit the activity of the target PTP by greater than or equal to 50%.

278

Anthony C. Bishop and Anna Serbina

100

O

HN

1

% PTP activity

80 60

N

40 20 0

WT C333P SHP2 SHP2

Fig. 3 Representative data for a nonconserved cysteine-directed PTP inhibitor (1) [15]. PTP activities of the wild-type SHP2 (black) and C333P SHP2 (gray) catalytic domains were measured with pNPP after preincubation with compound 1 (50 μM) and normalized to vehicle-only controls, as detailed in Subheadings 3.2 and 3.3 3.3 Counter-Screen for Target-CysteineDirected Inhibitors Using pNPP as Substrate

1. Obtain potential hit compounds from Subheading 3.2 as 5 mM solutions in DMSO. 2. Repeat Subheading 3.2, steps 2–9 for potential hit compounds, replacing SHP2cat with C333P SHP2cat. 3. C333-directed SHP2 inhibitors are identified as potential hits from Subheading 3.2 that do not inhibit the activity C333P SHP2cat (greater than or equal to 80% activity is retained in the counter-screen). 4. Confirm selectivity of C333-directed SHP2 inhibitors with side-by-side assay of C333-directed SHP2 inhibitors (repeating Subheading 3.2, steps 2–9) on both wild-type SHP2cat and C3233P SHP2cat. See Fig. 3 for representative data on a previously identified C333-directed SHP2 inhibitor (compound 1).

3.4 Screen for Target-CysteineDirected Inhibitors Using DiFMUP as Substrate (See Note 11)

1. Obtain a structurally diverse library of compounds that contain cysteine-reactive electrophiles as stock solutions in DMSO (see Subheading 2.1, Note 4). Dilute compounds to 5 mM in DMSO (see Note 5), preferably in 96-deep-well storage plates. 2. Transfer 2 μL of each compound (or DMSO control) solution to a black, flat-bottomed 96-well microplate via a multichannel pipette (see Note 6). 3. Prepare a “master mix” for the DiFMUP reaction that contains the remaining PTP reaction components (with the exception of

Screen for Allosteric Cysteine-Directed PTP Inhibitors

279

the DiFMUP substrate) multiplied by the number of reactions to be run. For each PTP reaction to being run, the mix should contain 128 μL of deionized water, 50 μL of 4× PTP activity solution, and 10 μL of 250 nM SHP2cat (diluted in PTP storage solution, see Note 12). 4. Using a multichannel pipette, transfer 188 μL of the master mix to each well that contains a test compound (or DMSO control) in the 96-well microplate. 5. Incubate the 96-well plate at 22 °C for 1 h (see Note 8). 6. Initiate PTP reactions by the addition of 10 μL of 0.80 mM DiFMUP (dissolved in 9:1 H2O/DMSO) to each well with a multichannel pipette (see Note 13). 7. Quickly mix the solutions by agitation and monitor fluorescence in the wells (excitation: 360 nM, emission: 450 nM) every 5 s for 1 min in a fluorescence plate reader. Determine the slope of the resulting line (see Note 14). 8. Calculate the %PTP activity for each well by dividing the slope of the line from a compound-containing well by that of DMSO controls (see Note 15). 9. Potential hits are defined as compounds that inhibit the activity of the target PTP by greater than or equal to 50%. 3.5 Counter-Screen for Target-CysteineDirected Inhibitors Using DiFMUP as Substrate

1. Obtain potential hit compounds from Subheading 3.4 as 5 mM solutions in DMSO. 2. Repeat Subheading 3.4, steps 2–8 for potential hit compounds, replacing SHP2cat with C333P SHP2cat. 3. C333-directed SHP2 inhibitors are identified as potential hits from Subheading 3.4 that do not inhibit the activity C333P SHP2cat (greater than or equal to 80% activity is retained in the counter-screen). 4. Confirm selectivity of C333-directed SHP2 inhibitors with side-by-side assay of C333-directed SHP2 inhibitors (repeating Subheading 3.4, steps 2–8) on both wild-type SHP2cat and C3233P SHP2cat.

4

Notes 1. Expression and purification conditions vary based on the expression vector of choice, the expression level of the particular PTP, and the desired amount of purified PTP. Expression– optimization experiments should be carried out for every PTP target. Thus, the following directions are generalized and should be modified as needed.

280

Anthony C. Bishop and Anna Serbina

2. The concentration of the purified and buffer-exchanged PTP solutions will vary between preparations and should be measured using any one of a number of standard techniques for protein concentration measurement (e.g., nanodrop spectrophotometry and Bradford assay). For wild-type SHP2cat purified from 250 mL of culture, we usually obtain a final concentration of approximately 2.0 mg/mL (53 μM). 3. Alternatively, glycerol can be added to the purified protein to a final concentration of 30%, and the resulting solution can be aliquoted and stored at -20 °C. We generally prefer flashfreezing PTP aliquots (no glycerol) in liquid nitrogen and store them at -80 °C, so the presence of trace glycerol does not need to be accounted for when enzyme preparations are diluted for use in enzymatic experiments. 4. Commercial compound libraries are usually supplied as DMSO solutions. If putative inhibitors are obtained as pure compounds, stock solutions of 25–50 mM should be prepared by dissolving a known compound mass in the appropriate volume of DMSO. 5. The screening conditions in this protocol yield a final compound concentration of 50 μM in the PTP assays. Higher compound concentrations can be used but will increase the percentage of false-positive hits. Lower concentrations can also be used but may lead to “false-negatives”—cysteine-selective compounds of modest potency that could be important starting points for further optimization but are overlooked if the concentration of the screen is set too low. 6. The number of replicates that are feasible to perform in the initial compound screen is highly dependent on the size of the compound library. Performing the screen in at least triplicate is ideal, but single-point or duplicate screens can be useful for winnowing down large numbers of compounds into smaller sets of potential hits that can be investigated more closely. 7. The screening conditions in this protocol yield a final SHP2cat concentration of 100 nM in the pNPP assays. Enzyme concentrations may need to be varied to obtain data in the linear range of the assay with other PTP domains that possess different levels of kinetic activity [28]. 8. The preincubation of PTP and compound is recommended because it is expected that most cysteine-directed covalent inhibitors will exhibit time-dependent inhibitory potency. Much like the choice of compound concentration for the screen (see Note 5), preincubation times and temperatures can be varied as desired. Longer times and/or higher tempera-

Screen for Allosteric Cysteine-Directed PTP Inhibitors

281

tures may increase the percentage of false-positive hits, whereas shorter times might cause potentially interesting starting-point compounds to be overlooked. 9. The screening conditions in this protocol yield a final pNPP concentration of 2.5 mM, which is close to SHP2cat’s Michaelis constant (KM) for the substrate [14]. Substrate concentrations may need to be varied to obtain data in the linear range of the assay with other PTP domains that possess different levels of kinetic activity [28]. 10. 96-well plates can be used for background correction. Background-correction plates should be identical to test plates except for lack of enzyme. In other words, 10 μL of PTP storage solution per well should be used for preparing background-correction plates in lieu of the 10 μL of 2 μM SHP2 (diluted in PTP storage solution). 11. It is always possible that compounds may produce false-positive results through assay interference. To minimize the chances of incorrectly assigning a compound as a cysteine-directed PTP inhibitor based on a single assay, we employ two complementary PTP-inhibition screens that use different productdetection methods: pNPP assays use A405, and DiFMUP assays use fluorescence (excitation: 360 nM, emission: 450 nM). 12. The screening conditions in this protocol yield a final SHP2cat concentration of 12.5 nM in the DiFMUP assays. Enzyme concentrations may need to be varied to obtain data in the linear range of the assay with other PTP domains that possess different levels of kinetic activity [28]. 13. The screening conditions in this protocol yield a final DiFMUP concentration of 40 μM. Substrate concentrations may need to be varied to obtain data in the linear range of the assay with other PTP domains that possess different levels of kinetic activity [28]. 14. Linear fits for each slope determination should have r2 values greater than or equal to 0.98. If significant deviations from linearity are observed for either the experimental wells or the vehicle controls, substrate and/or enzyme concentrations should be adjusted accordingly. 15. Background correction of the slopes should not be needed as there should be no measurable increase in DiFMUP fluorescence in the absence of enzyme over a 1-min timescale. Nevertheless, it is a good idea to establish the stability of the no-PTP fluorescence readings by preparing wells that contain no enzyme.

282

Anthony C. Bishop and Anna Serbina

Acknowledgments The authors thank Rachel Seifert for bioinformatic analysis of nonconserved cysteines in PTP domains and HyoJeon Kim and Lynn Kao for help in the development of compound-screening protocols. Research reported in this publication was supported by the National Institute of General Medical Sciences of the National Institutes of Health under Award Number R15GM071388. Funding from Amherst College is also gratefully acknowledged. References 1. Tonks NK (2006) Protein tyrosine phosphatases: from genes, to function, to disease. Nat Rev Mol Cell Biol 7:833–846 2. Andersen JN, Mortensen OH, Peters GH, Drake PG, Iversen LF, Olsen OH et al (2001) Structural and evolutionary relationships among protein tyrosine phosphatase domains. Mol Cell Biol 21:7117–7136 3. Alonso A, Sasin J, Bottini N, Friedberg I, Friedberg I, Osterman A et al (2004) Protein tyrosine phosphatases in the human genome. Cell 117:699–711 4. Verma S, Sharma S (2018) Protein tyrosine phosphatase as potential therapeutic target in various disorders. Curr Mol Pharmacol 11: 191–202 5. Rehman AU, Rahman MU, Khan MT, Saud S, Liu H, Song D et al (2018) The landscape of protein tyrosine phosphatase (Shp2) and cancer. Curr Pharm Des 24:3767–3777 6. Bohmer F, Szedlacsek S, Tabernero L, Ostman A, den Hertog J (2013) Protein tyrosine phosphatase structure-function relationships in regulation and pathogenesis. FEBS J 280:413–431 7. Vainonen JP, Momeny M, Westermarck J (2021) Druggable cancer phosphatases. Sci Transl Med 13. https://doi.org/10.1126/ scitranslmed.abe2967 8. Blaskovich MA (2009) Drug discovery and protein tyrosine phosphatases. Curr Med Chem 16:2095–2176 9. Tsutsumi R, Ran H, Rademann J, Neel BG (2018) Off-target inhibition by active sitetargeting SHP2 inhibitors. FEBS Open Bio 8: 1405–1411 10. Wiesmann C, Barr KJ, Kung J, Zhu J, Erlanson DA, Shen W et al (2004) Allosteric inhibition of protein tyrosine phosphatase 1B. Nat Struct Mol Biol 11:730–737 11. Hansen SK, Cancilla MT, Shiau TP, Kung J, Chen T, Erlanson DA (2005) Allosteric

inhibition of PTP1B activity by selective modification of a non-active site cysteine residue. Biochemistry 44:7704–7712 12. Tautermann CS, Binder F, Bu¨ttner FH, Eickmeier C, Fiegen D, Gross U et al (2019) Allosteric activation of striatal-enriched protein tyrosine phosphatase (STEP, PTPN5) by a fragment-like molecule. J Med Chem 62:306– 316 13. Liu Q, Sabnis Y, Zhao Z, Zhang T, Buhrlage SJ, Jones LH et al (2013) Developing irreversible inhibitors of the protein kinase cysteinome. Chem Biol 20:146–159 14. Chio CM, Lim CS, Bishop AC (2015) Targeting a cryptic allosteric site for selective inhibition of the oncogenic protein tyrosine phosphatase Shp2. Biochemistry 54:497–504 15. Marsh-Armstrong B, Fajnzylber JM, Korntner S, Plaman BA, Bishop AC (2018) The allosteric site on SHP2’s protein tyrosine phosphatase domain is targetable with druglike small molecules. ACS Omega 3:15763–15770 16. Kim JY, Plaman BA, Bishop AC (2020) Targeting a pathogenic cysteine mutation: discovery of a specific inhibitor of Y279C SHP2. Biochemistry 59:3498–3507 17. Ruddraraju KV, Zhang ZY (2017) Covalent inhibition of protein tyrosine phosphatases. Mol BioSyst 13:1257–1279 18. Cohen MS, Zhang C, Shokat KM, Taunton J (2005) Structural bioinformatics-based design of selective, irreversible kinase inhibitors. Science 308:1318–1321 19. Ostrem JM, Peters U, Sos ML, Wells JA, Shokat KM (2013) K-Ras(G12C) inhibitors allosterically control GTP affinity and effector interactions. Nature 503:548–551 20. Canon J, Rex K, Saiki AY, Mohr C, Cooke K, Bagal D et al (2019) The clinical KRAS(G12C) inhibitor AMG 510 drives anti-tumour immunity. Nature 575:217–223

Screen for Allosteric Cysteine-Directed PTP Inhibitors 21. Visscher M, Arkin MR, Dansen TB (2016) Covalent targeting of acquired cysteines in cancer. Curr Opin Chem Biol 30:61–67 22. Tartaglia M, Martinelli S, Stella L, Bocchinfuso G, Flex E, Cordeddu V et al (2006) Diversity and functional consequences of germline and somatic PTPN11 mutations in human disease. Am J Hum Genet 78:279–290 23. Martinelli S, Nardozza AP, Delle Vigne S, Sabetta G, Torreri P, Bocchinfuso G et al (2012) Counteracting effects operating on Src homology 2 domain-containing protein-tyrosine phosphatase 2 (SHP2) function drive selection of the recurrent Y62D and Y63C substitutions in Noonan syndrome. J Biol Chem 287:27066–27077 24. Yu Z-H, Xu J, Walls CD, Chen L, Zhang S, Zhang R et al (2013) Structural and mechanistic insights into LEOPARD syndromeassociated SHP2 mutations. J Biol Chem 288: 10472–10482

283

25. Tajan M, de Rocca SA, Valet P, Edouard T, Yart A (2015) SHP2 sails from physiology to pathology. Eur J Med Genet 58:509–525 26. Resnick E, Bradley A, Gan JR, Douangamath A, Krojer T, Sethi R et al (2019) Rapid covalent-probe discovery by electrophile-fragment screening. J Am Chem Soc 141:8951–8968 27. Backus KM, Correia BE, Lum KM, Forli S, Horning BD, Gonzalez-Paez GE et al (2016) Proteome-wide covalent ligand discovery in native biological systems. Nature 534:570–574 28. Barr AJ, Ugochukwu E, Lee WH, King ONF, Filippakopoulos P, Alfano I et al (2009) Largescale structural analysis of the classical human protein tyrosine phosphatome. Cell 136:352– 363 29. Salmeen A, Andersen JN, Myers MP, Meng TC, Hinks JA, Tonks NK et al (2003) Redox regulation of protein tyrosine phosphatase 1B involves a sulphenyl-amide intermediate. Nature 423:769–773

Chapter 18 In Vitro Phosphatase Assays for the Eya2 Tyrosine Phosphatase Christopher Alderman, Aaron Krueger, John Rossi, Heide L. Ford, and Rui Zhao Abstract Protein tyrosine phosphatases (PTP), such as the Eyes Absent (Eya) family of proteins, play important roles in diverse biological processes. In vitro phosphatase assays are essential tools for characterizing the enzymatic activity as well as discovering inhibitors and regulators of these phosphatases. Two common types of in vitro phosphatase assays use either a small molecule substrate that produces a fluorescent or colored product, or a peptide substrate that produces a colorimetric product in a malachite green assay. In this chapter, we describe detailed protocols of a phosphatase assay using small molecule 3-O-methylfluorescein phosphate (OMFP) as a substrate and a malachite green assay using the pH2AX peptide as a substrate to evaluate the phosphatase activity of EYA2 and the effect of small molecule inhibitors of EYA2. These protocols can be easily adapted to study other protein tyrosine phosphatases. Key words Eya2, Protein tyrosine phosphatase, Phosphatase assay, OMFP, Malachite green assay

1

Introduction The Eya family of genes, Eyas1-4, were first discovered as transcriptional coactivators of the Six family of homeoprotein transcription factors [1, 2]. The Six1-Eya transcriptional complex is critical for development in species from Drosophila to mammals, where it plays a key role in organogenesis of the eye, thymus, thyroid, parathyroid, muscle, kidney, and ear [3–9]. (We use the mouse protein nomenclature such as Six1 and Eya in this section to represent these proteins from multiple species and will use the human protein nomenclature in all subsequent sections.) Six1 and Eya are typically downregulated after organ development is complete, but they can be abnormally reexpressed and reinitiate developmental programs out of context, driving tumorigenesis and metastasis in many tumor types including breast cancer, leukemia, peripheral nerve-sheath

Damien The´venin and Jo¨rg P. Mu¨ller (eds.), Protein Tyrosine Phosphatases: Methods and Protocols, Methods in Molecular Biology, vol. 2743, https://doi.org/10.1007/978-1-0716-3569-8_18, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

285

286

Christopher Alderman et al.

tumors, Wilms’ tumor, cervical, ovarian, liver, kidney, lung, and colorectal cancers [10–32]. In addition to being transcriptional coactivators, Eya family members contain tyrosine phosphatase enzymatic activity and are members of the haloacid dehalogenase (HAD) family of tyrosine phosphatases [7, 33, 34]. Eya and other HAD phosphatases use an Asp as the active site residue instead of the Cys more commonly used in cellular tyrosine phosphatases [35]. The phosphatase activity of HAD family members requires Mg++ [35]. All other known HAD phosphatases target phosphorylated Ser/Thr, while Eya is the only HAD family protein tyrosine phosphatase [36]. The tyrosine phosphatase activity of Eyas plays important roles in multiple cellular processes. Eyas have been shown to regulate transcription of a subset of Six1 target genes, making Eya the first coactivator with transcription-modifying phosphatase activity [7]. The Eya1-3 tyrosine phosphatase activity was also found to be important for transformation, migration, invasion, and metastasis in breast cancer [24, 37–40], although its precise mechanism of action is not known. In addition, Eyas1-3 direct cells to the repair instead of apoptotic pathway upon DNA damage by dephosphorylating pY142 on H2AX [36, 41]. Eya2 can also inhibit the tumorsuppressive activity of estrogen receptor β (ERβ) by dephosphorylating its pY36 residue [42]. Recently, we found that the Tyr phosphatase activity of Eya2 is critical for glioblastoma stem cell maintenance, potentially through mitotic spindle regulation [43], although a direct target for the Eya2 tyrosine phosphatase in this context has yet to be identified. The characterization of a protein tyrosine phosphatase such as Eya using an in vitro phosphatase assay is a critical first step in understanding the mechanism and regulation of these enzymes. There are two common types of in vitro phosphatase assays: One uses a small molecule, and the other uses a phosphopeptide as a substrate. In a small molecule phosphatase assay, the phosphatase dephosphorylates the small molecule substrate and generates either a fluorescent or colored product that can be monitored by fluorescence or colorimetry. Small molecule substrates commonly used for phosphatase assays include 3-O-methylfluorescein phosphate (OMFP), fluorescein diphosphate (FDP), and p-nitrophenyl phosphate (pNPP), which are dephosphorylated to fluorescent OMF, fluorescent fluorescein, and colorimetric pNP, respectively. Although these small molecules are generic substrates that do not differentiate Tyr or Ser/Thr phosphatases, this small molecule phosphatase assay is easy to set up, and this is a simple and quantitative way to evaluate the effect of potential regulators on tyrosine phosphatase activity. Furthermore, this assay can be readily adapted to a high-throughput screening format to identify potential inhibitors.

In vitro Phosphatase Assays

287

In a phosphopeptide-based phosphatase assay, the free orthophosphate released from the phosphopeptide in the phosphatase reaction forms a green complex with malachite green and molybdate, which can be detected at 620 nm on a spectrometer. Unlike the small molecule phosphatase assays, in this assay, a protein tyrosine phosphatase only acts on a peptide containing a phosphor-tyrosine, while a Ser/Thr phosphatase only acts on a peptide containing a phosphor-serine or threonine. In this chapter, we present detailed protocols of Eya2 Eya Domain (ED) expression and purification from Escherichia coli, an OMFP-based fluorescent phosphatase assay to analyze the enzyme kinetics of Eya2 phosphatase activity, use of the OMFP assay to evaluate the effect of small molecule inhibitors on the Eya2 phosphatase activity, and a pH2AX-based malachite green assay to evaluate the tyrosine phosphatase activity of Eya2 and the effect of small molecule inhibitors targeting the Eya2 tyrosine phosphatase. These protocols can be easily adapted to other Eya family members, as well as to other protein tyrosine phosphatases to evaluate their activities and identify potential regulators or inhibitors.

2

Materials

2.1 Express and Purify EYA2 ED from E. coli

1. Human EYA2 ED (residues 253-538) subcloned into the pGEX-6P1 plasmid (GE Healthcare) using the BamHI and XhoI sites. 2. E. coli strain: XA90 cell obtained from colleagues (unpublished). Other E. coli strains such as BL21 or Rosetta should also work. 3. E. coli culture media 1: autoclave LB media (10 g tryptone, 10 g NaCl, 5 g yeast extract, and q.s. 1 L with H2O) and add Ampicillin to a final concentration of 100 μg/mL. 4. E. coli culture media 2: autoclave 2xYT (16 g Tryptone, 10 g yeast extract, 5 g NaCl, adjust pH to 7.0 with 5 N NaOH, and q.s. 1 L with H2O) media and add Ampicillin to a final concentration of 50 μg/mL. 5. Isopropyl-beta-D-thiogalactoside (IPTG): dissolve 2.38 g into 10 mL H2O to make a 1000× stock solution at 1 M and store in aliquots at -20 °C. 6. Leupeptin: dissolve 10 mg/mL in dimethyl sulfoxide (DMSO) and store in aliquots at -20 °C. 7. Pepstatin A: dissolve 25 mg/mL in DMSO and store in aliquots at -20 °C. 8. Phenylmethylsulfonyl Fluoride (PMSF): dissolve 17.42 g into 100 mL DMSO to make a 100× stock solution at 1 M and store in aliquots at -20 °C.

288

Christopher Alderman et al.

9. NaCl: dissolve 292.2 g/L in H2O to make a stock solution at 5 M. 10. Lysis buffer: 50 mM Tris–HCL, pH 7.5, 250 mM NaCl, 5% glycerol, and 1 mM dithiothreitol (DTT). 11. Sonicator: Ultrasonic Processor purchased from Toption. 12. Glutathione resin: Glutathione-Sepharose 4B resin purchased from GE Healthcare and stored at 4 °C. 13. PreScission protease: PreScission Protease purchased from Cytiva and stored at -80 °C. 14. Elution buffer: lysis buffer plus 30 mM DTT. 15. Glass chromatography column: Glass Econo-Column purchased from Bio-Rad. 16. Protein concentrator: Pierce™ Protein Concentrators PES, 10 K MWCO, 0.5–100 mL purchased from Thermo Fisher. 17. Size exclusion column: Superdex 200 purchased from GE Healthcare. 18. Sodium dodecyl sulfate (SDS) gel: 4.1 mL H2O, 3.3 mL acrylamide/bis (30% 37.5:1; Bio-Rad), 2.5 mL Tris–HCl (1.5 M, pH 8.8), 100 μL 10% SDS, 10 μL N,N,N′,N′-tetramethylethylene-diamine (TEMED; Bio-Rad), and 32 μL 10% ammonium persulfate (APS). 19. Coomassie Brilliant Blue stain: 1 g of Coomassie Brilliant Blue (Bio-Rad), 500 mL methanol, 100 mL glacial acetic acid, and bring the final volume to 1 L with H2O. 20. 5× protein-loading buffer: 100 mg bromophenol blue, 2.5 mL 2 M Tris–HCl (pH 6.8), 10 mL glycerol, 2 g SDS, and 1.542 g DTT. 21. Destain buffer: 500 mL methanol, 100 mL glacial acetic acid, and bring the final volume to 1 L with H2O. 2.2 Analyze the Eya2 Phosphatase Kinetics Using an OMFP-Based Fluorescent Phosphatase Assay

1. 2× Assay buffer: 100 mM MES, pH 6.5, 100 mM NaCl, 5 mM MgCl2, 0.1% bovine serum albumin, 2 mM DTT. 2. Black, half-volume 96-well flat bottom plate: purchased from Greiner Bio-one. 3. Clear, 96-well round bottom plate: purchased from Greiner Bio-one or Corning. 4. OMFP: 10 mM stock, dissolved in DMSO. 5. Multichannel pipettor (8-channel). 6. Solution reservoir for multichannel pipettor: purchased from VWR. 7. Plate reader capable of recording time points automatically over the course of the assay, for example, the Promega GloMax.

In vitro Phosphatase Assays

2.3 Evaluate Small Molecule Inhibitors of the Eya2 Phosphatase Using an OMFP-Based Fluorescent Phosphatase Assay

289

1. Assay buffer: 25 mM MES, pH 6.5, 50 mM NaCl, 5 mM MgCl2, 0.33% bovine serum albumin, 5 mM DTT. 2. Eya protein: 300 nM in assay buffer. 3. Dimethyl sulfoxide (DMSO): purchased from Sigma-Aldrich. 4. OMFP: purchased from Scientific Resources Pte Ltd. 5. Disodium ethylenediaminetetraacetic acid: 0.5 M, pH 8.0. 6. Black, half-volume 96-well flat bottom plate: purchased from Greiner Bio-One. 7. Plate reader with fluorometer: Promega GloMax.

2.4 Evaluate Small Molecule Inhibitors of the EYA2 Phosphatase Using a pH2AX-Based Malachite Green Assay

1. 2× malachite green assay buffer containing EYA2: 200 mM MES, pH 6.0, 100 mM NaCl, 2 mM MgCl2, 100 μM DTT, 7.8 μM Eya2. 2. pH2AX phosphopeptide: the KATQASQEpY phospho-H2AX peptide was purchased from Abgent and dissolved in water to make a 1 mM stock solution. 3. Microplate reader: Spectramax PLUS 384 plate reader (Molecular Devices). Malachite green reagent: 0.34 mg/mL malachite green in 8.3% HCl, 3.4% ethanol, 75 mM ammonium molybdate, 0.01% Tween-20, or purchased from Sigma-Aldrich. 4. Data analysis software: Origin Pro 8.0 (OriginLab).

3

Methods

3.1 Express and Purify EYA2 ED from E. coli

Brief summary of the protocol: Human EYA2 ED is expressed and purified as a GST-fusion protein from E. coli. GST is then cleaved, and EYA2 ED is further purified using gel filtration for subsequent enzymatic assays. 1. Grow overnight starter culture of E. coli containing the EYA2ED plasmid by inoculating 10 mL of E. coli culture media 1 with a single colony or 10 μL glycerol stock (prepare 10 mL starter culture for each liter of large culture in step 2). The culture will be grown at 200 RPM, 37 °C, on a shaker for 12–16 h. 2. Inoculate each liter of E. coli culture media 2 that has been prewarmed to 37 °C with 10 mL of the overnight starter culture. 3. Grow the E. coli at 37 °C and 200 RPM on a shaker until OD600 reaches ~0.8–1.0.

290

Christopher Alderman et al.

4. Add 300 μL of IPTG stock (1 M) per liter of culture to induce EYA2 ED expression at 0.3 mM IPTG concentration. 5. Reduce temperature to 25 °C and continue to grow the bacteria at 200 RPM overnight (18–20 h). 6. Harvest bacteria culture by centrifugation at 5000 RPM, 6 °C for 10 min. 7. Bacteria pellet can be lysed immediately as in step 8, or be frozen at -80 °C for later use. If frozen, thaw the pellet at room temperature before resuspension. 8. Add approximately 10 mL of lysis buffer to pellet from 1 L of cells to resuspend the pellet (see Note 1). Take 20 μL whole cell lysate sample for SDS-PAGE in step 21 (see Note 2). 9. Add protease inhibitors to their respective final concentrations: Leupeptin: 1 μM, Pepstatin A: 1 μg/mL, PMSF: 1 mM. 10. Lyse bacteria cells via sonication (45 s × 6 at 85% power) while manually rotating on ice (to keep the solution cool) with 2 min rests between each 45 s sonication. 11. (Optional) Bring NaCl concentration up to 0.5 M for centrifugation (which helps making the pellet compact) by adding 5 M NaCl directly to the lysate. 12. Collect the supernatant after centrifugation (~18,000 g for 40 min). Take 20 μL supernatant and pellet samples for SDS-PAGE in step 21. 13. (Optional) Resuspend the pellet with 200 mL lysis buffer, sonicate, and spin down using the same condition in 12 and combine the supernatants (see Note 3). 14. (Optional) Centrifuge one or two more times (~18,000 × g for 40 min) to generate clear supernatant. 15. While centrifugation is underway, prewash glutathione resin with 2–3 bed volumes of lysis buffer. 16. Load clear supernatant onto glutathione resin-packed column. Flow through the column via gravity at 4 °C (see Note 4). 17. Collect flow through and load onto column again using gravity flow. 18. Wash column thoroughly by gravity flow with 100–150 mL of the lysis buffer (>20× bed volume). Take 20 μL of 50% resin slurry as the pre-cut sample for SDS-PAGE in step 21. 19. After wash, add 10 mL lysis buffer and 2 units of PreScission protease per 100 μg of GST-fusion proteins to resin and incubate at 4 °C overnight on an orbital shaker. 20. Elute in 5 individual 1 mL aliquots of lysis buffer with gravity flow. Take 20 μL from each elution and 50% resin slurry as the elution and post-cut samples for SDS-PAGE in step 21.

In vitro Phosphatase Assays

291

21. Boil samples mentioned in steps 9, 12, 18, and 20 with 5× protein-loading buffer for 5 min and run on SDS gel for 90 min at 100 V. Stain gel with Coomassie Brilliant Blue stain on an orbital shaker for 60 min. Destain with destain buffer on an orbital shaker for 90 min. Evaluate to ensure the EYA2 ED purification proceeded normally and evaluate the EYA2 ED purity. 22. Combine all eluates from step 20 and concentrate to 2–5 mL with a concentrator at 3000 g, 4 °C. Stop immediately if protein precipitation is observed. 23. Spin tubes for 5 min at 4 °C to remove any protein precipitation. 24. Load supernatant onto prewashed Superdex 200 column. 25. Analyze Superdex 200 fractions under the elution peak corresponding to the correct molecular weight using SDS-PAGE. If concentration of the fractions with pure EYA2 ED is sufficiently high, aliquot and freeze at -80 °C. If not, concentrate to desired concentration, aliquot and freeze at 80 °C. 3.2 Analyze the EYA2 Phosphatase Kinetics Using an OMFP-Based Fluorescent Phosphatase Assay

Brief summary of the protocol: 150 nM EYA2 ED is incubated with 16.6–1000 μM OMFP substrate. The amount of fluorescent product OMF is monitored in a time course to determine the initial velocity. Plotting of the initial velocity versus OMFP concentration generates Vmax and Km by fitting to a Michaelis-Menton equation. 1. Prepare a working stock of 300 nM EYA2 ED in 2× assay buffer in enough volume to add 25 μL for every replicate of the reaction. These assays are typically performed in triplicate. 2. Using a multichannel pipette, transfer 25 μL from this EYA2 ED stock to each well to be used in the black 96-well plate. Deposit solution down the side of the well to avoid introducing bubbles, which will interfere with the fluorescence readings. If bubbles occur, gently swirl or tap the plate to dissipate. 3. Transfer 25 μL of assay buffer to one column of the black 96-well plate to determine the background rate of background OMFP hydrolysis at each substrate concentration, serving as a no-enzyme control. 4. Prepare 2× working stock solutions of OMFP in H2O at concentrations of 2000 μM, 1000 μM, 500 μM, 250 μM, 125 μM, 62.5 μM, and 31.25 μM (see Note 5). Transfer 150 μL of each prepared dilution, in descending concentration, to a single column of the clear 96-well plate. 5. Set up plate reader to read fluorescence with excitation/emission wavelengths of 485 nm/515 nm, reading each well once every 2 min (see Note 6). The reaction will begin upon addition

292

Christopher Alderman et al. 3×10⁶

RFU

2×10⁶

1000 μM 500 μM 250 μM

1×10⁶

125 μM 62.5 μM 31.3 μM 15.6 μM

0 0

500

1000

1500

2000

2500

3000

3500

Seconds

Fig. 1 Determining the initial velocity of EYA2 ED at different substrate concentrations using an OMFP assay

Velocity, RFU/sec

2000

1500

1000

500

Vmax = 3441 RFU/sec Km = 751 nM 0 0

100

200

300

400

500

600

700

800

900

1000 1100

[Substrate], nM

Fig. 2 Determining the Vmax and Km of EYA2 ED using the Michaelis-Menten equation

of OMFP to the reaction plate, so be sure the plate reader will be ready to take readings immediately following OMFP addition. 6. Initiate reaction by adding 25 μL of prepared OMFP from the clear plate to the corresponding wells of the black reaction plate using multichannel pipette. Gently pipette to mix and avoid introduction of bubbles. Note time. At a known time interval, repeat this addition for all replicates of the reaction, then place

In vitro Phosphatase Assays

293

assay plate in plate reader and begin measurement. Allow reaction to proceed for 1 h. 7. Upon reaction completion, export data from plate reader as a table of RFU measurements versus time, in seconds. Correct the time measurement for the different replicates to account for the time interval during substrate addition in the previous step. 8. Graph RFU values versus time to generate a saturation plot for each substrate concentration (Fig. 1). Review the plots to determine which measurements were taken during the initial velocity phase of the reaction, and then fit these measurements using a linear regression. The slope of this linear regression is the initial reaction velocity for the corresponding substrate concentration in RFU/sec. 9. Determine the rate of background OMFP hydrolysis at each substrate concentration by fitting time points with linear regression. Subtract the rate of background OMFP hydrolysis at each substrate concentration from the initial velocities of Eya2 at each substrate concentration to account for hydrolysis of OMFP in assay buffer. 10. Plot initial velocity versus substrate concentration to obtain a curve that can be fit to the Michaelis-Menton equation to determine Km and Vmax values (Fig. 2, see Notes 7 and 8). 3.3 Evaluate Small Molecule Inhibitors of the EYA2 Phosphatase Using an OMFP-Based Fluorescent Phosphatase Assay

Brief summary of the protocol: 150 nM EYA2 ED is incubated with 25 μM OMFP substrate and 45.7 nM to 100 μM inhibitor for an hour. Plotting of the percentage activity versus inhibitor concentration on a log scale allows for fitting of a dose-response curve and determination of IC50. 1. Dilute EYA2 ED protein to 300 nM in assay buffer (see Note 9). 2. Prepare serial dilutions of inhibitors at 100× of the desired concentrations using DMSO (see Note 10). 3. Dilute OMFP to 50 μM in H2O to be used as substrate. 4. Dilute EDTA to 0.45 M in H2O to be used to stop the reaction. 5. Load 49 μl of buffer without protein per well as a blank control (see Note 11). 6. Load 49 μL of 300 nM EYA2 ED per well as an uninhibited control. 7. Load 49 μL of 300 nM EYA2 ED per well for all inhibitor concentrations. 8. Add 1 μL DMSO to the blank control (buffer only), 1 μL of DMSO to uninhibited control (protein only), and 1 μL of prepared inhibitor to each respective well.

294

Christopher Alderman et al.

IC50 = 3.05 ± 0.20 PM

Fig. 3 Evaluating the inhibition of Eya2 phosphatase activity by a small molecule using the OMFP assay

9. Incubate 10 min at room temperature on an orbital rotator. 10. Start the reaction by adding 50 μL of 50 μM OMFP to each well (see Note 12). Final reaction conditions are 150 nM EYA2 ED and 25 μM OMFP. 11. Incubate in dark, covered, for 1 h. 12. Stop the reaction by adding 20 μL of 0.45 M EDTA, pH 8.0 to each well (see Note 12). Final concentration of EDTA is 75 mM. 13. Read fluorescence on the plate reader, with excitation at 485 nm and emission at 515 nm. An example of the doseresponse curve for one small molecule inhibitor is shown in Fig. 3. 3.4 Evaluate Small Molecule Inhibitor of the EYA2 Phosphatase Using a pH2AX-Based Malachite Green Assay

Brief summary of the protocol: 3.9 μM EYA2 ED is incubated with 50 μM pH2AX peptide substrate, the malachite green reagent, and 45.7 nM to 100 μM inhibitor for 1 h. Plotting of the percentage activity versus inhibitor concentration on a log scale allows for fitting of a dose-response curve and determination of IC50. 1. Make a 2× malachite green assay buffer containing 7.8 μM EYA2. Keep everything on ice until it is added to the plate. 2. Add 25 μL of the above 2× malachite green assay buffer and EYA protein to wells of clear, 96-well, half-area plates (see Notes 13–15). 3. Add 22 μL H2O to each well to bring the final volume to 50 μL when all components are added. 4. Add 0.5 μL inhibitors to each well. 5. Incubate at room temperature for 10 min. 6. Add 2.5 μL pH2AX (1 mM stock) to wells to achieve a final concentration of 50 μM and to start the reaction.

Absorbance (620 nm)

In vitro Phosphatase Assays

295

1.0 0.8 0.6 0.4 0.2 0.0 -6.5

IC50 = 17.1 ± 1.3 PM -6.0

-5.5

-5.0

-4.5

-4.0

Inhibitor (Log10 M)

Fig. 4 Evaluating the inhibition of EYA2 phosphatase activity by a small molecule using the malachite green assay with the pH2AX peptide as a substrate

7. Incubate at 25 °C for 1 h. 8. Add 100 μL malachite green reagent to each well. 9. Incubate 20 min at room temperature to develop the color complex. 10. Read plate at 620 nm. 14. Doseresponse curves were generated using Origin Pro 8.0. An example of the dose-response curve for one small molecule inhibitor is shown in Fig. 4.

4

Notes 1. EYA2 ED tolerates a wide variety of buffers for purification fairly well. 2. Several samples will be taken throughout the protocol to be analyzed on a Coomassie-stained SDS gel to monitor the purification process and identify any potential issues. 3. Resuspension of the pellet after initial sonication is only required if the expression level of the protein is low. This is not typically needed with the EYA2 ED construct. 4. Protein binding to glutathione resin can also be done in batch mode. Resuspend prewashed resin in 1–2 mL of lysis buffer and add to the lysate which is then incubated on an orbital shaker at 4 °C for 2 h. Centrifuge the lysate and resin mixture at 4000 rpm, 4 °C for 10 min. Resuspend the resin in 2 mL lysis buffer and add to a column. Continue with step 18 in Subheading 3.1. 5. OMFP dilutions of 2000 μM, 1000 μM, 500 μM, 250 μM, 125 μM, 62.5 μM, and 31.25 μM can be prepared by serial

296

Christopher Alderman et al.

dilution at 1:2 ratio beginning with 2000 μM OMFP. Substrate range may need to be adjusted for other PTPs to cover from 0.2 to 5× of the Km. 6. Rate of measurement will depend on the specifications of the plate reader. A time interval should be chosen such that each well is measured once during that interval; a faster plate reader may allow for a shorter interval between measurements. 7. Due to the low solubility of OMFP, we cannot reach Vmax. Other more soluble small molecule substrates such as DFP or pNPP can be used to overcome this problem. 8. Vmax can be used to calculate Kcat using the equation Kcat = Vmax/[Eya2] after converting RFU to molar quantity of OMF using a standard curve and expressing the initial velocity in moles per unit time. 9. It is ideal to evaluate inhibitor effect at an enzyme concentration that produces >5-fold signal/noise ratio at selected time point and a substrate concentration below Km. 10. Ideal inhibitor concentrations will vary for different enzymes and inhibitors. For EYA2 ED, prepare 8 serial dilutions at a ratio of 1:3 beginning with 10 mM inhibitor. This will generate inhibitor concentrations (100×) of 10 mM, 3.33 mM, 1.11 mM, 370 μM, 123 μM, 41.2 μM, 13.7 μM, and 4.57 μM. The final inhibitor concentrations in each well are 100, 33.3, 11.1, 3.7, 1.23, 0.412, 0.137, and 0.0457 μM. Make the 100× inhibitor serial dilution stocks in DMSO so that the final DMSO concentration in each inhibitor condition remain the same. 11. Quantities listed are for one replicate. Prepare triplicate reactions for all conditions (blank control, uninhibited control, and each concentration of inhibitor). 12. OMFP and EDTA can be added to each replicate of different inhibitor concentrations simultaneously using a multichannel pipette. Addition of OMFP to different replicates should be done sequentially in 20 or 30 s intervals to ensure consistent timing for all reactions. Quenching with EDTA should then follow the same intervals, ensuring each reaction has proceeded exactly 1 h. 13. A kinetic assay similar to what is described in Subheading 3.2 (but using the pH2AX peptide and malachite green assay instead) can be performed first to determine the optimal enzyme and substrate concentration to be used in the endpoint assay for evaluating the effect of inhibitors. 14. Use caution when pipetting. Try not to create bubbles in the wells, as they will interfere with the reading.

In vitro Phosphatase Assays

297

15. Always do the following controls to check for possible free phosphate contamination: protein + peptide without inhibitor (uninhibited), protein + peptide + EDTA (fully inhibited). If the fully inhibited sample has significant absorbance reading, it indicates potential free phosphate contamination in either the protein, peptide, or reagents (buffer, DMSO, small molecule). Malachite green assay with each of these individual components alone can be carried out to identify the source of contamination and replace the reagent with one from a different source to remove the contamination.

Acknowledgments Research reported in this chapter was supported by the National Cancer Institute of the National Institute of Health under award number R03DA030559 (R.Z. and H.L.F.), R41CA180347 (H.L. F. and R.Z.), R01CA221282 (H.L.F. and R.Z.), R01CA224867 (H.L.F.), R01NS108396 (H.L.F.), Colorado Bioscience Discovery and Evaluation Grant (H.L.F. and R.Z.), and Cancer League of Colorado Grant (H.L.F. and R.Z.), as well as National Institute of General Medical Sciences of the National Institute of Health under award number R35GM145289 (R.Z.). C.A. is supported by the NIH NRSA T32CA174648 Training in Translational Research of Lung, Head and Neck Cancer. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. References 1. Zimmerman JE, Bui QT, Liu H, Bonini NM (2000) Molecular genetic analysis of Drosophila eyes absent mutants reveals an eye enhancer element. Genetics 154(1):237–246. https:// doi.org/10.1093/genetics/154.1.237 2. Xu PX, Cheng J, Epstein JA, Maas RL (1997) Mouse Eya genes are expressed during limb tendon development and encode a transcriptional activation function. Proc Natl Acad Sci U S A 94(22):11974–11979. https://doi.org/ 10.1073/pnas.94.22.11974 3. Xu PX, Zheng W, Laclef C, Maire P, Maas RL, Peters H, Xu X (2002) Eya1 is required for the morphogenesis of mammalian thymus, parathyroid and thyroid. Development 129(13): 3033–3044. https://doi.org/10.1242/dev. 129.13.3033 4. Zheng W, Huang L, Wei ZB, Silvius D, Tang B, Xu PX (2003) The role of Six1 in mammalian auditory system development. Development

130(17):3989–4000. https://doi.org/10. 1242/dev.00628 5. Ikeda K, Ookawara S, Sato S, Ando Z, Kageyama R, Kawakami K (2007) Six1 is essential for early neurogenesis in the development of olfactory epithelium. Dev Biol 311(1): 53–68. https://doi.org/10.1016/j.ydbio. 2007.08.020 6. Grifone R, Demignon J, Houbron C, Souil E, Niro C, Seller MJ, Hamard G, Maire P (2005) Six1 and Six4 homeoproteins are required for Pax3 and Mrf expression during myogenesis in the mouse embryo. Development 132(9): 2235–2249. https://doi.org/10.1242/dev. 01773 7. Li X, Oghi KA, Zhang J, Krones A, Bush KT, Glass CK, Nigam SK, Aggarwal AK, Maas R, Rose DW, Rosenfeld MG (2003) Eya protein phosphatase activity regulates Six1-Dach-Eya transcriptional effects in mammalian

298

Christopher Alderman et al.

organogenesis. Nature 426(6964):247–254. https://doi.org/10.1038/nature02083 8. Ozaki H, Nakamura K, Funahashi J, Ikeda K, Yamada G, Tokano H, Okamura HO, Kitamura K, Muto S, Kotaki H, Sudo K, Horai R, Iwakura Y, Kawakami K (2004) Six1 controls patterning of the mouse otic vesicle. Development 131(3):551–562. https://doi. org/10.1242/dev.00943 9. Pignoni F, Hu B, Zavitz KH, Xiao J, Garrity PA, Zipursky SL (1997) The eye-specification proteins So and Eya form a complex and regulate multiple steps in Drosophila eye development. Cell 91(7):881–891. https://doi.org/ 10.1016/s0092-8674(00)80480-8 10. Ford HL, Kabingu EN, Bump EA, Mutter GL, Pardee AB (1998) Abrogation of the G2 cell cycle checkpoint associated with overexpression of HSIX1: a possible mechanism of breast carcinogenesis. Proc Natl Acad Sci U S A 95(21):12608–12613. https://doi.org/10. 1073/pnas.95.21.12608 11. Reichenberger KJ, Coletta RD, Schulte AP, Varella-Garcia M, Ford HL (2005) Gene amplification is a mechanism of Six1 overexpression in breast cancer. Cancer Res 65(7): 2668–2675. https://doi.org/10.1158/ 0008-5472.CAN-04-4286 12. Coletta RD, Christensen KL, Micalizzi DS, Jedlicka P, Varella-Garcia M, Ford HL (2008) Six1 overexpression in mammary cells induces genomic instability and is sufficient for malignant transformation. Cancer Res 68(7): 2204–2213. https://doi.org/10.1158/ 0008-5472.CAN-07-3141 13. Micalizzi DS, Wang CA, Farabaugh SM, Schiemann WP, Ford HL (2010) Homeoprotein Six1 increases TGF-beta type I receptor and converts TGF-beta signaling from suppressive to supportive for tumor growth. Cancer Res 70(24):10371–10380. https://doi.org/10. 1158/0008-5472.CAN-10-1354 14. Wang CA, Jedlicka P, Patrick AN, Micalizzi DS, Lemmer KC, Deitsch E, Casas-Selves M, Harrell JC, Ford HL (2012) SIX1 induces lymphangiogenesis and metastasis via upregulation of VEGF-C in mouse models of breast cancer. J Clin Invest 122(5):1895–1906. https://doi. org/10.1172/JCI59858 15. Tan J, Zhang C, Qian J (2011) Expression and significance of Six1 and Ezrin in cervical cancer tissue. Tumour Biol 32(6):1241–1247. https://doi.org/10.1007/s13277-0110228-8 16. Wan F, Miao X, Quraishi I, Kennedy V, Creek KE, Pirisi L (2008) Gene expression changes during HPV-mediated carcinogenesis: a comparison between an in vitro cell model and

cervical cancer. Int J Cancer 123(1):32–40. https://doi.org/10.1002/ijc.23463 17. Behbakht K, Qamar L, Aldridge CS, Coletta RD, Davidson SA, Thorburn A, Ford HL (2007) Six1 overexpression in ovarian carcinoma causes resistance to TRAIL-mediated apoptosis and is associated with poor survival. Cancer Res 67(7):3036–3042. https://doi. org/10.1158/0008-5472.CAN-06-3755 18. Ng KT, Lee TK, Cheng Q, Wo JY, Sun CK, Guo DY, Lim ZX, Lo CM, Poon RT, Fan ST, Man K (2010) Suppression of tumorigenesis and metastasis of hepatocellular carcinoma by shRNA interference targeting on homeoprotein Six1. Int J Cancer 127(4):859–872. https://doi.org/10.1002/ijc.25105 19. Ng KT, Man K, Sun CK, Lee TK, Poon RT, Lo CM, Fan ST (2006) Clinicopathological significance of homeoprotein Six1 in hepatocellular carcinoma. Br J Cancer 95(8):1050–1055. https://doi.org/10.1038/sj.bjc.6603399 20. Yu Y, Khan J, Khanna C, Helman L, Meltzer PS, Merlino G (2004) Expression profiling identifies the cytoskeletal organizer ezrin and the developmental homeoprotein Six-1 as key metastatic regulators. Nat Med 10(2): 175–181. https://doi.org/10.1038/nm966 21. Sehic D, Karlsson J, Sandstedt B, Gisselsson D (2012) SIX1 protein expression selectively identifies blastemal elements in Wilms tumor. Pediatr Blood Cancer 59(1):62–68. https:// doi.org/10.1002/pbc.24025 22. Ono H, Imoto I, Kozaki K, Tsuda H, Matsui T, Kurasawa Y, Muramatsu T, Sugihara K, Inazawa J (2012) SIX1 promotes epithelialmesenchymal transition in colorectal cancer through ZEB1 activation. Oncogene 31(47): 4923–4934. https://doi.org/10.1038/onc. 2011.646 23. Robin TP, Smith A, McKinsey E, Reaves L, Jedlicka P, Ford HL (2012) EWS/FLI1 regulates EYA3 in Ewing sarcoma via modulation of miRNA-708, resulting in increased cell survival and chemoresistance. Mol Cancer Res 10(8): 1098–1108. https://doi.org/10.1158/ 1541-7786.MCR-12-0086 24. Pandey RN, Rani R, Yeo EJ, Spencer M, Hu S, Lang RA, Hegde RS (2010) The Eyes Absent phosphatase-transactivator proteins promote proliferation, transformation, migration, and invasion of tumor cells. Oncogene 29(25): 3715–3722. https://doi.org/10.1038/onc. 2010.122 25. Wang Y, Klijn JG, Zhang Y, Sieuwerts AM, Look MP, Yang F, Talantov D, Timmermans M, Meijer-van Gelder ME, Yu J, Jatkoe T, Berns EM, Atkins D, Foekens JA (2005) Gene-expression profiles to predict

In vitro Phosphatase Assays distant metastasis of lymph-node-negative primary breast cancer. Lancet 365(9460): 671–679. https://doi.org/10.1016/S01406736(05)17947-1 26. Li CM, Guo M, Borczuk A, Powell CA, Wei M, Thaker HM, Friedman R, Klein U, Tycko B (2002) Gene expression in Wilms’ tumor mimics the earliest committed stage in the metanephric mesenchymal-epithelial transition. Am J Pathol 160(6):2181–2190. https://doi.org/10.1016/S0002-9440(10) 61166-2 27. Farabaugh SM, Micalizzi DS, Jedlicka P, Zhao R, Ford HL (2012) Eya2 is required to mediate the pro-metastatic functions of Six1 via the induction of TGF-beta signaling, epithelial-mesenchymal transition, and cancer stem cell properties. Oncogene 31(5): 552–562. https://doi.org/10.1038/onc. 2011.259 28. Wang QF, Wu G, Mi S, He F, Wu J, Dong J, Luo RT, Mattison R, Kaberlein JJ, Prabhakar S, Ji H, Thirman MJ (2011) MLL fusion proteins preferentially regulate a subset of wild-type MLL target genes in the leukemic genome. Blood 117(25):6895–6905. https://doi.org/ 10.1182/blood-2010-12-324699 29. Miller SJ, Lan ZD, Hardiman A, Wu J, Kordich JJ, Patmore DM, Hegde RS, Cripe TP, Cancelas JA, Collins MH, Ratner N (2010) Inhibition of Eyes Absent Homolog 4 expression induces malignant peripheral nerve sheath tumor necrosis. Oncogene 29(3):368–379. https://doi.org/10.1038/onc.2009.360 30. Auvergne RM, Sim FJ, Wang S, ChandlerMilitello D, Burch J, Al Fanek Y, Davis D, Benraiss A, Walter K, Achanta P, Johnson M, Quinones-Hinojosa A, Natesan S, Ford HL, Goldman SA (2013) Transcriptional differences between normal and glioma-derived glial progenitor cells identify a core set of dysregulated genes. Cell Rep 3(6):2127–2141. https://doi.org/10.1016/j.celrep.2013. 04.035 31. Li Z, Qiu R, Qiu X, Tian T (2017) EYA2 promotes lung cancer cell proliferation by downregulating the expression of PTEN. Oncotarget 8(67):110837–110848. https:// doi.org/10.18632/oncotarget.22860 32. Anantharajan J, Zhou H, Zhang L, Hotz T, Vincent MY, Blevins MA, Jansson AE, Kuan JWL, Ng EY, Yeo YK, Baburajendran N, Lin G, Hung AW, Joy J, Patnaik S, Marugan J, Rudra P, Ghosh D, Hill J, Keller TH, Zhao R, Ford HL, Kang C (2019) Structural and functional analyses of an allosteric EYA2 phosphatase inhibitor that has on-target effects in human lung cancer cells.

299

Mol Cancer Ther 18(9):1484–1496. https:// doi.org/10.1158/1535-7163.MCT-18-1239 33. Rayapureddi JP, Kattamuri C, Steinmetz BD, Frankfort BJ, Ostrin EJ, Mardon G, Hegde RS (2003) Eyes absent represents a class of protein tyrosine phosphatases. Nature 426(6964): 2 9 5 – 2 9 8 . h t t p s : // d o i . o r g / 1 0 . 1 0 3 8 / nature02093 34. Tootle TL, Silver SJ, Davies EL, Newman V, Latek RR, Mills IA, Selengut JD, Parlikar BE, Rebay I (2003) The transcription factor Eyes absent is a protein tyrosine phosphatase. Nature 426(6964):299–302. https://doi. org/10.1038/nature02097 35. Jung SK, Jeong DG, Chung SJ, Kim JH, Park BC, Tonks NK, Ryu SE, Kim SJ (2010) Crystal structure of ED-Eya2: insight into dual roles as a protein tyrosine phosphatase and a transcription factor. FASEB J 24(2):560–569. https:// doi.org/10.1096/fj.09-143891 36. Krishnan N, Jeong DG, Jung SK, Ryu SE, Xiao A, Allis CD, Kim SJ, Tonks NK (2009) Dephosphorylation of the C-terminal tyrosyl residue of the DNA damage-related histone H2A.X is mediated by the protein phosphatase eyes absent. J Biol Chem 284(24): 16066–16070. https://doi.org/10.1074/jbc. C900032200 37. Blevins MA, Towers CG, Patrick AN, Zhao R, Ford HL (2015) The SIX1-EYA transcriptional complex as a therapeutic target in cancer. Expert Opin Ther Targets 19(2):213–225. https://doi.org/10.1517/14728222.2014. 978860 38. Wu K, Li Z, Cai S, Tian L, Chen K, Wang J, Hu J, Sun Y, Li X, Ertel A, Pestell RG (2013) EYA1 phosphatase function is essential to drive breast cancer cell proliferation through cyclin D1. Cancer Res 73(14):4488–4499. https:// doi.org/10.1158/0008-5472.CAN-12-4078 39. Tadjuidje E, Wang TS, Pandey RN, Sumanas S, Lang RA, Hegde RS (2012) The EYA tyrosine phosphatase activity is pro-angiogenic and is inhibited by benzbromarone. PLoS One 7(4): e34806. https://doi.org/10.1371/journal. pone.0034806 40. Krueger AB, Dehdashti SJ, Southall N, Marugan JJ, Ferrer M, Li X, Ford HL, Zheng W, Zhao R (2013) Identification of a selective small-molecule inhibitor series targeting the eyes absent 2 (Eya2) phosphatase activity. J Biomol Screen 18(1):85–96. https://doi.org/ 10.1177/1087057112453936 41. Cook PJ, Ju BG, Telese F, Wang X, Glass CK, Rosenfeld MG (2009) Tyrosine dephosphorylation of H2AX modulates apoptosis and survival decisions. Nature 458(7238):591–596. https://doi.org/10.1038/nature07849

300

Christopher Alderman et al.

42. Yuan B, Cheng L, Chiang HC, Xu X, Han Y, Su H, Wang L, Zhang B, Lin J, Li X, Xie X, Wang T, Tekmal RR, Curiel TJ, Yuan ZM, Elledge R, Hu Y, Ye Q, Li R (2014) A phosphotyrosine switch determines the antitumor activity of ERbeta. J Clin Invest 124(8): 3378–3390. https://doi.org/10.1172/ JCI74085 43. Zhang G, Dong Z, Gimple RC, Wolin A, Wu Q, Qiu Z, Wood LM, Shen JZ, Jiang L,

Zhao L, Lv D, Prager BC, Kim LJY, Wang X, Zhang L, Anderson RL, Moore JK, Bao S, Keller TH, Lin G, Kang C, Hamerlik P, Zhao R, Ford HL, Rich JN (2021) Targeting EYA2 tyrosine phosphatase activity in glioblastoma stem cells induces mitotic catastrophe. J Exp Med 218(11). https://doi.org/10.1084/ jem.20202669

Chapter 19 High-Throughput Discovery and Characterization of Covalent Inhibitors for Protein Tyrosine Phosphatases Zihan Qu, Aaron D. Krabill, and Zhong-Yin Zhang Abstract Covalent inhibition has gained increasing interest in targeting the undruggable protein tyrosine phosphatases (PTPs). However, a systematic method for discovering and characterizing covalent PTP inhibitors has yet to be established. Here, we describe a workflow involving high-throughput screening of covalent fragment libraries and a novel biochemical assay that enables the acquisition of kinetics parameters of PTP inhibition by covalent inhibitors with higher throughput. Key words Protein tyrosine phosphatase, Covalent inhibition, pNPP assay, Kinetics of covalent inhibition, High-throughput screening

1

Introduction Protein tyrosine phosphorylation is an essential process that regulates the initiation, propagation, and termination of cellular signaling cascades [1–3]. This process is strictly controlled by protein tyrosine phosphatases (PTPs) and the opposing activity of protein tyrosine kinases (PTKs). Hence, PTPs play an important role in signaling cascades and cellular processes like cell growth, proliferation, differentiation, migration, metabolism, and survival [1– 3]. The aberrant activities of PTPs are associated with human diseases, including cancers, autoimmune disorders, diabetes, and developmental disorders [4–8]. Although PTPs have emerged as a compelling therapeutic target class, they are understudied due to difficulties developing small molecule drugs/tool compounds targeting their highly conserved and positively charged active sites [9]. Covalent inhibitors react with specific residues in a protein and irreversibly or reversibly form a covalent bond to inactivate the target [10]. The process of covalent inhibition involves the reversible binding of the compound with the pocket of interest followed by a time-dependent covalent bond formation between an

Damien The´venin and Jo¨rg P. Mu¨ller (eds.), Protein Tyrosine Phosphatases: Methods and Protocols, Methods in Molecular Biology, vol. 2743, https://doi.org/10.1007/978-1-0716-3569-8_19, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

301

302

Zihan Qu et al.

electrophilic functional group of the covalent inhibitor and a proximal nucleophilic amino acid side chain at the target site. The reversible binding interaction is characterized by the dissociation constant KI, and the covalent modification reaction is characterized by the rate constant kinact. The ratio of kinact/KI is widely used to determine the efficacy of covalent inhibitors, where a higher ratio indicates more efficient overall inactivation potency. Although halfmaximal inhibitory concentration (IC50) is a more commonly utilized parameter in evaluating inhibitors, it can be misleading if used alone for covalent inhibitor characterization. Specifically, covalent inhibition is both time- and dose-dependent, so the preincubation time with the compound will significantly impact the inhibitory activity. Thus, IC50 itself does not reflect the compound’s binding affinity and/or reactivity, making it difficult to understand and optimize identified hits. [11] Traditionally, covalent inhibitors have been avoided due to their perceived promiscuous reactivity and irreversible adverse effects in vivo. However, these undesirable limitations can be offset by the benefits of covalent inhibition, such as prolonged duration of action and lower dosage and dosing frequency [12, 13]. Unlike conventional reversible inhibitors that interact with target protein by dynamically forming an inhibitor–protein complex, covalent inhibitors tend to permanently form the inhibitor–protein adduct, making them more likely to outcompete endogenous substrates of much higher concentrations. Additionally, the cells can only recover from the irreversible inhibition through resynthesis of the target protein, so the pharmacodynamics generally outlast the pharmacokinetics of covalent inhibitors unless the protein has a rapid resynthesis rate [14]. Indeed, recently disclosed covalent inhibitors show a much longer target occupancy despite the relatively faster clearance of the inhibitor in vivo [15, 16]. In some cases, the irreversible inhibition could overcome the resistance of specific mutations against reversible inhibitors [17]. For instance, the T790M mutation in eGFR significantly enhances ATP affinity and thereby abolishes the drug binding, but the irreversible inhibition was not greatly impacted due to its noncompetitive mode of action [18, 19]. The PTP-catalyzed reaction involves the nucleophilic attack of the conserved active-site thiolate on the phosphorus atom of phosphotyrosine (pTyr) [20] (Fig. 1). As such, PTPs may be considered viable targets for covalent inhibition. A reactive electrophile, also known as a covalent warhead, can be installed on a pTyr mimicking moiety that interacts with the pTyr binding pocket and facilitates the covalent modification of the active-site thiolate by the warhead (Fig. 1). Indeed, PTP-specific and pTyr mimetic activity–based probes such as α-bromobenzyl phosphonate, aryl vinyl sulfonates, and sulfones have been described [21, 22]. Unfortunately, although

High-Throughput Covalent Inhibitor Screening and Characterization

303

Fig. 1 Mechanism of PTP-catalyzed reaction and covalent inhibition targeting the active-site Cys. (a) PTP dephosphorylates substrate through the nucleophilic attack of the phosphoryl moiety by the active-site cysteine thiolate, with adjacent residues such as arginine, glutamine, and aspartate facilitating the catalytic process. (b) Proposed covalent inhibitors mimicking the substrate will reversibly interact with the active-site pocket followed by the formation of the covalent bond between the electrophilic warhead and the active-site cysteine thiolate

targeted covalent inhibition has proven successful in many fields, such as protein tyrosine kinases [10, 23, 24], no isozyme-specific covalent inhibitors have been developed for the PTPs. Although several covalent inhibitors against various PTPs have been reported [25–29], a more systematic and robust approach is still lacking to develop potent and specific covalent inhibitors for the PTPs. Here, using the Src homology 2 domain-containing phosphatase 1 (SHP1) as an example, we present a method to identify and characterize covalent PTP inhibitors. At the same time, we propose a systematic and thorough workflow to develop covalent PTP inhibitors to enable the advancement of targeting PTPs for therapeutic applications. The method involves a high-throughput screening of covalent fragment libraries followed by validation via kinetics characterization. As mentioned above, the kinact/KI second-order rate constant is preferred to analyze covalent inhibition because it describes inhibitor potency in a time-independent manner [30]. The conventional jump dilution method can be used to determine the kinact/KI ratio, which involves preincubation of the compound with the enzyme followed by 10- to 100-fold dilution into the substrate (concentration greater than Km) to dilute the unreacted inhibitor

304

Zihan Qu et al.

and monitor the residual enzymatic activity [11, 31]. Yet, the jump dilution method is unsuitable for compounds with high kinact values because the adduct formation could be much faster than the transferring process. In addition, higher enzyme concentration during preincubation is required to generate a satisfactory reaction signal after dilution, so the compounds of low KI value or high binding affinity cannot be thoroughly characterized as the measured KI cannot be lower than enzyme concentration. Besides, the preincubation process involves only the inhibitor and the enzyme, so the competition with the substrate, which is usually the case in cells or in vivo, is overlooked in jump dilution assay. To overcome the limitations associated with the jump dilution method, a time-course assay was applied to identify and characterize PTP covalent inhibitors, where a much lower concentration of protein is utilized, and the enzyme is incubated with both substrates and inhibitors, allowing for a more thorough kinetics characterization [31, 32]. The enzymatic reaction can be performed in 96- or 384-well plates, and the progress curve of substrate reaction at various inhibitor concentrations can be monitored over specified lengths of time. A pseudo first-order rate constant (kobs) can be extracted from each progress curve and plotted against inhibitor compound concentration, from which the ratio of kinact/KI or specific kinact and KI values will be calculated. The kinetic assay will reveal the time dependence, where the progress curve will reach a plateau if all proteins are inactivated, and dose dependence, where different inactivation rates at various inhibitor concentrations can be observed. Furthermore, the initial slope of each progress curve can be used to calculate the initial rate of the enzyme, which can be extracted and used to interpret if the inactivation involves significant reversible interaction, that is, noncovalent inhibition (Fig. 2).

Fig. 2 Kinetic analysis of covalent inhibitors. (a) Time-course analysis of the PTP-catalyzed reaction at constant substrate concentration (Km) and various inhibitor concentrations. (b) Nonlinear regression fit of kobs versus the concentration of the inhibitor to calculate specific kinact and KI values when compound shows reversible binding within the measured concentration range. (c) Linear regression fit of kobs versus the concentration of the inhibitor to calculate kinact/KI values when the compound does not show measurable reversible binding within the measured concentration range

High-Throughput Covalent Inhibitor Screening and Characterization

305

Fig. 3 IC50 determination of covalent inhibitors with 10-min (●) and 60-min (■) preincubation of the covalent inhibitor and the enzyme

Two time points IC50 will be determined for compounds validated by the time-course assay, where the compounds are preincubated with the protein for 10 and 60 min, followed by dilution into the substrate to wash off any unbound inhibitor. Because covalent inhibition is time-dependent, compounds showing lowered IC50 after longer preincubation are further supported as covalent inhibitors at this stage (Fig. 3). Ideally, one would expect the identified hits to modify the active-site cysteine (Cys) residue selectively based on PTP catalytic mechanism. However, modifying noncatalytic cysteine, aggregators, and protein denaturation could all result in observed inhibitory activities. To identify compounds that potentially target the active-site cysteine, a vanadate protection assay will be performed for the obtained hits. Vanadate is a reversible and pan-PTP competitive inhibitor [33, 34], so excessive vanadate can saturate the active site of all PTPs and prevent covalent modifications inside the active-site pocket (Fig. 4). Compounds showing less or no inhibition in the presence of vanadate can be considered as active site-directed [35] (Fig. 5). Determination of covalent protein adduct formation using mass spectroscopy is a critical method to examine whether the compound covalently modifies the protein and further confirms results from the vanadate protection assay [15, 32]. The hits will be incubated with both wild-type and mutant PTPs in which the active-site cysteine is replaced by serine (Ser). In this case, the presence of adduct formation for wild-type enzymes will confirm the covalent inhibition, and the absence of adduct formation for the Cys to Ser mutant will indicate that the compound is specifically modifying the active-site cysteine. Tandem mass spectrometry (MS/MS) analysis can be used to determine the site of compound modification.

306

Zihan Qu et al.

Fig. 4 Mechanism of the vanadate protection assay. (a) In the absence of vanadate (triangle), the covalent inhibitor (sphere) readily accesses the active-site pocket and covalently modifies the cysteine thiolate; (b) when the PTP is incubated with both the covalent inhibitor and vanadate, vanadate will occupy the active-site pocket, preventing the covalent inhibitor from accessing and modifying active-site cysteine thiolate

Fig. 5 Effect of vanadate on covalent inhibition of PTP. The PTP is preincubated with DMSO (■), vanadate (●), vanadate and covalent inhibitor (▼), or covalent inhibitor (~), after which the complex is diluted rapidly into saturating amount of substrate. Time-course of the PTP catalyzed hydrolysis indicates that the active site of PTP is protected from inactivation by the presence of vanadate (▼) when compared to DMSO (■) or vanadate alone (●). Little residual activity is detected when PTP is incubated with the covalent inhibitor alone (~)

Low intrinsic reactivity is essential for covalent inhibitors as higher reactivity could result in nonspecific modification of bystander proteins leading to undesirable toxicity. The relative inherent reactivity of the covalent inhibitors can be assessed through a protocol modified based on the previous reports [32, 36], where the compound is incubated with reduced glutathione (GSH) at a 1:1 ratio for 45 min and the remaining GSH

High-Throughput Covalent Inhibitor Screening and Characterization

307

Fig. 6 Comparison of the relative intrinsic reactivity of selected hits from the screening with glutathione (GSH). The compounds (100 μM) were preincubated with reduced GSH (100 μM) for 45 min at room temperature, and the reaction was quenched by the addition of DTNB (100 μM). The %GSH was quantified by OD412 and normalization to the DMSO-treated group

determined (Fig. 6). A more universal and accurate approach is to incubate the compound with GSH in great excess and examine the residual amount of the compound with liquid chromatography– mass spectrometry (LC-MS) [15, 37–39]. In the latter case, the compound’s half-life can be determined by plotting the remaining amount against time. Intrinsically less reactive compounds or those with a longer half-life will be prioritized for further optimization as covalent PTP inhibitors.

2

Materials Solutions are prepared in deionized Milli-Q filtered water or analytical-grade reagents at room temperature. Reagents for cellbased assays are prepared in sterile conditions. All reagents are prepared and stored at room temperature unless otherwise specified.

2.1 High-Throughput Screening

1. Handpicked fragment library (40 mM in DMSO in 384-well plate) and enamine cysteine–focused library (100 mM in DMSO in 384-well plate). 2. 3,3-diemthylglutaric (DMG) buffer: 50 mM 3,3-dimethylglutaric acid, 18 mM NaCl, 1 mM EDTA, pH 7.0. 3. Para-nitrophenyl phosphate (pNPP) stock: 800 mM in DMG buffer, stored at -20 °C. 4. Protein stock: SHP1 PTP domain (245–543) in Tris–HCl buffer (20 mM Tris, 150 mM NaCl, pH 8.0), stored at -80 °C. 5. Phenyl vinyl sulfonate (PVSN) stock: 100 mM in DMSO.

308

Zihan Qu et al.

6. 5 M NaOH. 7. Clear 384-well plates, flat-bottom, nontreated. 8. 50 mL falcon tubes. 2.2 Phosphatase Activity Assay

1. 20 mM compound in DMSO 2. DMG buffer: 50 mM 3,3-dimethylglutaric acid, 18 mM NaCl, 1 mM EDTA, pH 7.0 3. VO43- protection buffer: 50 mM 3,3-dimethylglutaric acid, 18 mM NaCl, pH 7.0 4. Para-nitrophenyl phosphate (pNPP) stock: 800 mM in DMG buffer, stored at -20 °C 5. 6,8-difluoro-4-methylumbelliferyl phosphate stock: 10 mM in DMG buffer, stored at -80 °C

(DiFMUP)

6. Protein stock: SHP1 PTP domain (245–543) in Tris buffer (20 mM Tris, 150 mM NaCl, pH 8.0), stored at -80 °C 7. 5 M NaOH 8. bpV(phen) stock: 20 mM in DMG buffer, stored at -80 °C 9. Sodium pervanadate stock: 100 mM in DI water, stored at -20 °C 10. Clear and black 96-well plates, flat-bottom, nontreated 11. Disposable pipette basins 12. 15 mL falcon tubes 2.3 GSH Reactivity Assay

1. 20 mM compound in DMSO 2. DMG buffer: 50 mM 3,3-dimethylglutaric acid, 18 mM NaCl, 1 mM EDTA, pH 7.0 3. GSH stock: 10 mM in DMG buffer, stored at -20 °C 4. 5,5′-dithiobis(2-nitrobenzoic acid) stock: 10 mM in DMG buffer, stored at -20 °C 5. Clear 96-well plates, flat-bottom, nontreated 6. Disposable pipette basins 7. 15 mL falcon tubes 8. 2 mL LC-MS vials 9. 200 μL flat-bottom vial inserts

3

Methods All procedures are performed at room temperature unless otherwise specified.

High-Throughput Covalent Inhibitor Screening and Characterization

3.1 High-Throughput Screening

309

1. Transfer 50 nL (for the 100 mM stock) or 125 nL (for the 40 mM stock) of library compounds to Columns 3–22 and 50 nL PVSN stock to Column 2 of the 384-well plates using an Echo 650 liquid handler so that the concentration during incubation with SHP1 PTP domain is 500 μM. 2. Dispense 10 μL of DMG buffer in Columns 23 and 24 using a Multidrop dispenser. 3. Dilute SHP1 PTP domain stock in DMG buffer to a concentration of 400 nM, and dispense 10 μL to Columns 1–22 using a Multidrop dispenser (see Note 1). Centrifuge the plates at 50 × g for 10 s to ensure that all liquids are spun down. 4. Dilute pNPP stock in DMG buffer to a concentration of 24.45 mM, and dispense 45 μL of the pNPP solution to all the wells after 30 min of incubation of compounds and protein using a Multidrop dispenser. 5. Dispense 45 μL of 5 M NaOH to all the wells after 2 min of reaction between the protein and pNPP using a Multidrop dispenser (see Note 2). Centrifuge the plates at 50 × g for 10 s to ensure that all liquids are spun down. 6. Read the absorbance of each plate at 405 nm, normalize the percent activity of the enzyme in Columns 3–22 using the equation %activity =

Average of Col: 1 - Average of Col: 23 and 24 : Well - Average of Col: 23 and 24

7. For wells showing percent activity lower than 50% or less, pick the compound and counter-screen against other PTPs in an identical manner. 8. Repurchase compounds specific toward SHP1 PTP domain. 3.2 Kinetics Characterization

1. Prepare 20 mM DMSO stock for the repurchased compounds (store the stock under 4 °C). All concentrations mentioned below are during the enzymatic reaction. 2. In a clear, flat-bottom, 96-well plate, dilute the compound (1.5×) starting from 200 μM followed by the addition of pNPP at Km value (3 mM for SHP1 PTP domain). Add SHP1 PTP domain of 100 nM, and immediately put the plate in the microplate reader. Read absorbance at 405 nm for 1 h. 3. Plot absorbance against time using GraphPad 9.0, and subtract the first time point for background subtraction. Fit each curve V 0 ð1 - e - X ∙K obs Þ to the equation Y = , where Y is the absorbance K obs at time point X (see Note 3).

310

Zihan Qu et al.

4. Plot each calculated kobs to the concentration of compound. • If a straight line is obtained, the slope of the line is the ratio of kinact to KI. • If a saturation is observed (i.e., the curve reaches plateau), fit ∙X , where Y is the Kobs at the curve to the equation Y = Kkinact I þX concentration X, to obtain specific kinact and KI values. 5. Take the compounds showing time- and dose-dependence to the next step. 3.3 Two-Time Point IC50

1. Serially dilute (1.5×) the compounds in a clear, flat-bottom, 96-well plate and add SHP1 PTP domain of 2 μM. 2. After 10 min of incubation of protein and compounds, transfer 4 μL of the mixture to another 96-well plate. Add 196 μL of 10 mM pNPP solution to all wells containing the 4 μL transferred solution. 3. After 10 min of addition of pNPP solution, quench the enzymatic reaction with 50 μL 5 M NaOH. Read absorbance at 405 nm and fit IC50 using GraphPad 9.0. 4. After 60 min of incubation of protein and compounds, repeat the above steps to obtain IC50 at 60 min (see Notes 4 and 5). For compounds showing IC50 below 10 μM after 60 min, the IC50 needs to be measured using DiFMUP. Use 200 nM protein during preincubation and 100 μM of DiFMUP for enzymatic reaction. The reaction time is 5 min, followed by a quench of 128 μM bpV(phen). Read the plate at Ex. 340/Em. 450 to determine IC50. 5. Take the compounds showing time-dependent inhibition to the next step.

3.4 Vanadate Protection

1. The reaction volume for this step is 50 μL. Calculate the volume of the compound needed based on 60-min IC50. Take one compound tested at 100 μM in triplicate as an example. 2. Dilute compound stock to fivefold of 60-min IC50 in VO43protection buffer with and without 500 μM VO43-. Prepare another solution with 500 μM VO43- and protein only. 3. Add SHP1 PTP stock to all solutions to make a final concentration of 4 μM. Pipette up and down several times to mix. 4. After 30 min of incubation, transfer 3 μL of each solution to a well in a clear, flat-bottom, 96-well plate, followed by the addition of 297 μL of pNPP in DMG buffer. Immediately read absorbance at 405 nm for 20 min.

High-Throughput Covalent Inhibitor Screening and Characterization

311

5. Plot the absorbance against time and subtract the first point for background subtraction. A protection by sodium pervanadate can be concluded if a significant difference is observed between with and without vanadate treatment (see Note 6). 6. Take the compounds that are active-site-directed to the next step. 3.5 Reactivity and Stability Assay 3.5.1 Relative GSH Reactivity Assay

1. Dispense 2 μL of compound stock to wells needed in a clear 96-well plate, and dispense 2 μL of DMSO to three wells as a negative control. 2. Dilute the GSH stock in DMG buffer to make a concentration of 204 μM, and add 98 μL of the GSH solution to all the wells. 3. Dilute the DTNB stock in DMG buffer to make a concentration of 200 μM. 4. After 45 min of incubation of GSH and compounds, add 100 μL of DTNB solution to all the wells. Shake the plate gently (~0.3 g) for 5 min to ensure mixing and reaction. 5. Read the absorbance at 412 nm. The readout for DMSO wells is considered as 100% GSH remaining and used to normalize other wells.

3.5.2 Compound Stability Assay

1. Add 100 μL of DMG buffer and 100 μL of the GSH stock to a glass insert of LC-MS vial to make 5 mM of GSH solution. 2. In a different insert, add 0.5 μL of the compound stock to 200 μL of DMG buffer. Put the inserts in the LC-MS vial and inject 2 μL of the solution to LC-MS. This will serve as the t = 0 point. 3. Add 0.5 μL of the compound stock to the 5 mM GSH solution to make a 50 μM compound concentration. Put the inserts in the LC-MS vial and inject 2 μL of the solution to LC-MS 1, 5, 10, 30, 60, and 120 min after compound addition. Use single ion monitoring (SIMS) to detect the m/z of the compound (see Note 7). 4. Integrate the peak for each spectrum and plot the area under the curve (AUC) against time. The compound’s half-life can be calculated from the curve using GraphPad 9.0. 5. The half-life and corresponding Kobs can be obtained for different GSH concentrations using an identical setup. Plotting the Kobs against GSH concentration will generate the secondorder rate constant for the reaction between the compound and GSH.

312

4

Zihan Qu et al.

Notes 1. Reducing agents such as dithiothreitol (DTT) are to be monitored for assays involving covalent inhibitors because they can efficiently quench the compounds. A previous publication mentioned the use of tris(2-carboxyethyl)phosphine (TCEP) instead [32]. However, we found that some compounds of lower reactivity can still be quenched upon testing, so no reducing agent is included during the assay to reduce falsenegatives. 2. During the HTS, ~5-fold dilution by pNPP solution may not be enough to wash off unbound inhibitors, so a short enzymatic reaction, in this case, 2 min, is needed to reduce covalent modification after dilution. The volumes of pNPP and NaOH added are identical to eliminate the time difference of volume dispensing. If these volumes are not kept identical, the time difference of enzymatic reaction between the first column to the last is great enough to interfere with the readout and cause inaccurate measurements. 3. The kinetics assay provides abundant information regarding the kinetics of covalent inhibitors, but it can be interfered with by highly reactive compounds, precipitations, or compound backgrounds. Due to technical errors such as instrument response time and pipetting time, a highly reactive compound can fully or partially inactivate the protein and change the initial slope. Therefore, compound rate constants should be evaluated carefully for reactive electrophiles combined with intrinsic reactivity. Precipitations or aggregations could result in false-positives, which can be distinguished by the progress curve: In the case of false-positives, the progress curves are not uniformly distributed, meaning that the compound does not inactivate protein at all after one specific concentration. Compounds with this property must be handled critically as they could be nonspecific. When dealing with unstable proteins like PTPs, the negative control curve will also show inactivation to an extent because of protein denaturation; this denaturation is generally much slower and tends not to affect the kinetics parameters. Additionally, when obtained kinact/KI is lower than 200 M-1 min-1, the compound could be weakly inhibitory or not inhibitory at all. Looking at the raw progress curve with aforementioned criteria could help resolve some cases. Still, a two-time point IC50 validation is necessary to assess time-dependent inhibition for hit evaluation. 4. A greater time difference between two time points of preincubation is favored to show a significant shift of IC50. The

High-Throughput Covalent Inhibitor Screening and Characterization

313

preincubation shouldn’t exceed 60 min to ensure a satisfactory readout difference between negative and positive controls. For some slow enzymes, the preincubation difference can be enlarged (e.g., measure at 1 min and 60 min) to reduce the effects of covalent inhibition during the enzymatic reaction. So far, we have not observed this issue in this 50-fold dilution setup. 5. Detergents like Triton X-100 or Tween-20 can be added during the assay (0.01–0.05%). In this case, the enzyme concentration used can be lowered. We observed that even 20 nM of SHP1 PTP domain showed linear activity within 45 min. Notably, 0.01% of detergent may not be sufficient to eliminate all aggregators. Hence, compounds that are much more potent when using less enzyme (even in the presence of detergent) need to be handled critically because aggregators’ inhibitory activity depends on enzyme concentration [40]. 6. The preincubation of vanadate protection can be performed in a 96-well plate or polymerase chain reaction (PCR) tube. When dealing with potent inhibitors, the compound could still partially inactivate the protein in the presence of vanadate, so a full recovery may not be observed. The concentration and time in this method are the preliminary condition and can be optimized based on the results. In case of long preincubation time or higher vanadate concentration, the protein could also be partially or completely inactivated as vanadate itself also shows irreversible inhibition to an extent. 7. The relative GSH reactivity assay allows a comparison of multiple compounds at a higher throughput but does not reveal the absolute stability of the compound. However, it can be used as a reference before the stability assay to determine the length of incubation of compounds and excessive GSH. In the case of weakly ionizable compounds, one can monitor the GSH adduct formation using SIMS, but the solvolysis effects will inevitably be overlooked. The second-order rate between the compound and GSH is a useful parameter to compare how much the compounds prefer targeted protein over the GSH as one can directly compare it with kinact/KI.

Acknowledgments This work was supported in part by NIH R01 CA069202 and U54 AG065181.

314

Zihan Qu et al.

References 1. Hunter T (2009) Tyrosine phosphorylation: thirty years and counting. Curr Opin Cell Biol 21:140–146. https://doi.org/10.1016/j.ceb. 2009.01.028 2. Tonks NK (2013) Protein tyrosine phosphatases: from housekeeping enzymes to masterregulators of signal transduction. FEBS J 280: 346–378. https://doi.org/10.1111/febs. 12077 3. Tonks NK (2006) Protein tyrosine phosphatases: from genes, to function, to disease. Nat Rev Mol Cell Biol 7:833–846. https://doi. org/10.1038/nrm2039 4. Frankson R, Yu Z-H, Bai Y, Li Q, Zhang R-Y, Zhang Z-Y (2017) Therapeutic targeting of oncogenic tyrosine phosphatases. Cancer Res 77:5701–5705. https://doi.org/10.1158/ 0008-5472.CAN-17-1510 5. Zhang Z-Y (2017) Drugging the Undruggable: therapeutic potential of targeting protein tyrosine phosphatases. Acc Chem Res 50: 122–129. https://doi.org/10.1021/acs. accounts.6b00537 6. Castro-Sanchez P, Teagle AR, Prade S, Zamoyska R (2020) Modulation of TCR signaling by tyrosine phosphatases: from autoimmunity to immunotherapy. Front Cell Dev Biol 8:608747. https://doi.org/10.3389/fcell. 2020.608747 7. Xie L, Lee S-Y, Andersen JN, Waters S, Shen K, Guo X-L, Moller NPH, Olefsky JM, Lawrence DS, Zhang Z-Y (2003) Cellular effects of small molecule PTP1B inhibitors on insulin signaling. Biochemistry 42:12792–12804. https:// doi.org/10.1021/bi035238p 8. Bollu LR, Mazumdar A, Savage MI, Brown PH (2017) Molecular pathways: targeting protein tyrosine phosphatases in cancer. Clin Cancer Res 23:2136–2142. https://doi.org/10. 1158/1078-0432.CCR-16-0934 9. Tautz L, Critton DA, Grotegut S (2013) Protein tyrosine phosphatases: structure, function, and implication in human disease. Methods Mol Biol Clifton NJ 1053:179–221. https:// doi.org/10.1007/978-1-62703-562-0_13 10. Sutanto F, Konstantinidou M, Do¨mling A (2020) Covalent inhibitors: a rational approach to drug discovery. RSC Med Chem 11:876– 8 8 4 . h t t p s : // d o i . o r g / 1 0 . 1 0 3 9 / D0MD00154F 11. Mons E, Roet S, Kim RQ, Mulder MPC (2022) A comprehensive guide for assessing covalent inhibition in enzymatic assays illustrated with kinetic simulations. Curr Protoc 2:e419. https://doi.org/10.1002/cpz1.419

12. Polinsky RJ (1998) Clinical pharmacology of rivastigmine: a new-generation acetylcholinesterase inhibitor for the treatment of Alzheimer’s disease. Clin Ther 20:634–647. https:// doi.org/10.1016/S0149-2918(98)80127-6 13. Chong CR, J€anne PA (2013) The quest to overcome resistance to EGFR-targeted therapies in cancer. Nat Med 19:1389–1400. https://doi.org/10.1038/nm.3388 14. Johnson DS, Weerapana E, Cravatt BF (2010) Strategies for discovering and derisking covalent, irreversible enzyme inhibitors. Future Med Chem 2:949–964 15. Lanman BA, Allen JR, Allen JG, Amegadzie AK, Ashton KS, Booker SK, Chen JJ, Chen N, Frohn MJ, Goodman G, Kopecky DJ, Liu L, Lopez P, Low JD, Ma V, Minatti AE, Nguyen TT, Nishimura N, Pickrell AJ, Reed AB, Shin Y, Siegmund AC, Tamayo NA, Tegley CM, Walton MC, Wang H-L, Wurz RP, Xue M, Yang KC, Achanta P, Bartberger MD, Canon J, Hollis LS, McCarter JD, Mohr C, Rex K, Saiki AY, San Miguel T, Volak LP, Wang KH, Whittington DA, Zech SG, Lipford JR, Cee VJ (2020) Discovery of a covalent inhibitor of KRASG12C (AMG 510) for the treatment of solid tumors. J Med Chem 63:52– 65. https://doi.org/10.1021/acs.jmedchem. 9b01180 16. Daryaee F, Zhang Z, Gogarty KR, Li Y, Merino J, Fisher SL, Tonge PJ (2017) A quantitative mechanistic PK/PD model directly connects Btk target engagement and in vivo efficacy. Chem Sci 8:3434–3443. https://doi. org/10.1039/C6SC03306G 17. Copeland RA, Pompliano DL, Meek TD (2006) Drug–target residence time and its implications for lead optimization. Nat Rev Drug Discov 5:730–739. https://doi.org/10. 1038/nrd2082 18. Yun C-H, Mengwasser KE, Toms AV, Woo MS, Greulich H, Wong K-K, Meyerson M, Eck MJ (2008) The T790M mutation in EGFR kinase causes drug resistance by increasing the affinity for ATP. Proc Natl Acad Sci 105:2070–2075. https://doi.org/10.1073/ pnas.0709662105 19. Kwak EL, Sordella R, Bell DW, GodinHeymann N, Okimoto RA, Brannigan BW, Harris PL, Driscoll DR, Fidias P, Lynch TJ, Rabindran SK, McGinnis JP, Wissner A, Sharma SV, Isselbacher KJ, Settleman J, Haber DA (2005) Irreversible inhibitors of the EGF receptor may circumvent acquired resistance to gefitinib. Proc Natl Acad Sci

High-Throughput Covalent Inhibitor Screening and Characterization 102:7665–7670. https://doi.org/10.1073/ pnas.0502860102 20. Yu Z-H, Zhang Z-Y (2018) Regulatory mechanisms and novel therapeutic targeting strategies for protein tyrosine phosphatases. Chem Rev 118:1069–1091. https://doi.org/ 10.1021/acs.chemrev.7b00105 21. Liu S, Zhou B, Yang H, He Y, Jiang Z-X, Kumar S, Wu L, Zhang Z-Y (2008) Aryl vinyl sulfonates and sulfones as active site-directed and mechanism-based probes for protein tyrosine phosphatases. J Am Chem Soc 130:8251– 8260. https://doi.org/10.1021/ja711125p 22. Kumar S, Zhou B, Liang F, Wang W-Q, Huang Z, Zhang Z-Y (2004) Activity-based probes for protein tyrosine phosphatases. Proc Natl Acad Sci 101:7943–7948. https://doi. org/10.1073/pnas.0402323101 23. Abdeldayem A, Raouf YS, Constantinescu SN, Moriggl R, Gunning PT (2020) Advances in covalent kinase inhibitors. Chem Soc Rev 49: 2617–2687. https://doi.org/10.1039/ C9CS00720B 24. Ghosh AK, Samanta I, Mondal A, Liu WR (2019) Covalent inhibition in drug discovery. ChemMedChem 14:889–906. https://doi. org/10.1002/cmdc.201900107 25. Marsh-Armstrong B, Fajnzylber JM, Korntner S, Plaman BA, Bishop AC (2018) The allosteric site on SHP2’s protein tyrosine phosphatase domain is targetable with Druglike small molecules. ACS Omega 3:15763– 15770. https://doi.org/10.1021/acsomega. 8b02200 26. Xu J, Chatterjee M, Baguley TD, Brouillette J, Kurup P, Ghosh D, Kanyo J, Zhang Y, Seyb K, Ononenyi C, Foscue E, Anderson GM, Gresack J, Cuny GD, Glicksman MA, Greengard P, Lam TT, Tautz L, Nairn AC, Ellman JA, Lombroso PJ (2014) Inhibitor of the tyrosine phosphatase STEP reverses cognitive deficits in a Mouse Model of Alzheimer’s disease. PLoS Biol 12:e1001923. https://doi. org/10.1371/journal.pbio.1001923 27. Hansen SK, Cancilla MT, Shiau TP, Kung J, Chen T, Erlanson DA (2005) Allosteric inhibition of PTP1B activity by selective modification of a non-active site cysteine residue. Biochemistry 44:7704–7712. https://doi.org/10. 1021/bi047417s 28. Karver MR, Krishnamurthy D, Kulkarni RA, Bottini N, Barrios AM (2009) Identifying potent, selective protein tyrosine phosphatase inhibitors from a library of Au(I) complexes. J Med Chem 52:6912–6918. https://doi.org/ 10.1021/jm901220m

315

29. Zhang ZY, Davis JP, Van Etten RL (1992) Covalent modification and active site-directed inactivation of a low molecular weight phosphotyrosyl protein phosphatase. Biochemistry 31:1701–1711. https://doi.org/10.1021/ bi00121a018 30. Strelow JM (2017) A perspective on the kinetics of covalent and irreversible inhibition. SLAS Discov Adv Life Sci R D 22:3–20. https://doi. org/10.1177/1087057116671509 31. Tian WX, Tsou CL (1982) Determination of the rate constant of enzyme modification by measuring the substrate reaction in the presence of the modifier. Biochemistry 21:1028– 1 0 3 2 . h t t p s : // d o i . o r g / 1 0 . 1 0 2 1 / bi00534a031 32. Resnick E, Bradley A, Gan J, Douangamath A, Krojer T, Sethi R, Geurink PP, Aimon A, Amitai G, Bellini D, Bennett J, Fairhead M, Fedorov O, Gabizon R, Gan J, Guo J, Plotnikov A, Reznik N, Ruda GF, Dı´azSa´ez L, Straub VM, Szommer T, Velupillai S, Zaidman D, Zhang Y, Coker AR, Dowson CG, Barr HM, Wang C, Huber KVM, Brennan PE, Ovaa H, von Delft F, London N (2019) Rapid covalent-probe discovery by electrophilefragment screening. J Am Chem Soc 141: 8951–8968. https://doi.org/10.1021/jacs. 9b02822 33. Huyer G, Liu S, Kelly J, Moffat J, Payette P, Kennedy B, Tsaprailis G, Gresser MJ, Ramachandran C (1997) Mechanism of inhibition of protein-tyrosine phosphatases by vanadate and pervanadate. J Biol Chem 272:843–851. https://doi.org/10.1074/jbc.272.2.843 34. Deng H, Callender R, Huang Z, Zhang Z-Y (2002) Is the PTPase-vanadate complex a true transition state analogue? Biochemistry 41: 5865–5872. https://doi.org/10.1021/ bi016097z 35. Seiner DR, LaButti JN, Gates KS (2007) Kinetics and mechanism of Protein Tyrosine Phosphatase 1B (PTP1B) inactivation by Acrolein. Chem Res Toxicol 20:1315. https://doi. org/10.1021/tx700213s 36. Naven RT, Kantesaria S, Nadanaciva S, Schroeter T, Leach KL (2013) High throughput glutathione and Nrf2 assays to assess chemical and biological reactivity of cysteine-reactive compounds. Toxicol Res 2:235–244. https:// doi.org/10.1039/C3TX50027F 37. Schwartz PA, Kuzmic P, Solowiej J, Bergqvist S, Bolanos B, Almaden C, Nagata A, Ryan K, Feng J, Dalvie D, Kath JC, Xu M, Wani R, Murray BW (2014) Covalent EGFR inhibitor analysis reveals importance of

316

Zihan Qu et al.

reversible interactions to potency and mechanisms of drug resistance. Proc Natl Acad Sci 111:173–178. https://doi.org/10.1073/ pnas.1313733111 38. Petri L, Egyed A, Bajusz D, Imre T, Hete´nyi A, ´ bra´nyi-Balogh P, Keseru ˝ GM Martinek T, A (2020) An electrophilic warhead library for mapping the reactivity and accessibility of tractable cysteines in protein kinases. Eur J Med Chem 207:112836. https://doi.org/10. 1016/j.ejmech.2020.112836

´ bra´nyi-Balogh P, Varga PR, Imre T, 39. Petri L, A ˝ GM (2020) Comparative reactivity Keseru analysis of small-molecule thiol surrogates. Bioorg Med Chem 28:115357. https://doi. org/10.1016/j.bmc.2020.115357 40. Allen SJ, Dower CM, Liu AX, Lumb KJ (2020) Detection of small-molecule aggregation with high-throughput microplate biophysical methods. Curr Protoc Chem Biol 12. https://doi. org/10.1002/cpch.78

INDEX A

G

Adoptive cell transfer .................................. 82, 88, 89, 91 Antibody ...........................................6–10, 13, 14, 16, 17, 25–31, 34, 35, 37, 38, 46, 47, 50–52, 55, 56, 58, 60, 66, 68, 83, 86, 90, 92, 111–113, 115, 116, 119–121, 126, 127, 129–131, 139, 141–143, 148, 154, 157, 159, 160, 166, 167, 171, 176, 177, 183, 184, 186, 188–191, 199, 203, 208, 224, 227, 232, 233, 235, 250, 251 APEX2 ................................167–170, 172–174, 177–179

Gene expression ............35, 36, 43–46, 53, 55, 123, 239 Genome editing ..................................45–46, 50, 94, 100

B Bioinformatics ..............................................22–24, 31–33 Biotin-labeling............................................ 183, 188, 224, 225, 227, 231–233 Bone.................................................57–63, 66–69, 74, 75

C Cas9 ................................................43–45, 47–49, 53, 61, 70–72, 74, 75, 94–96, 98, 100–102, 106, 107 CD45 ...............................4, 47, 50, 51, 58, 91, 183, 190 Clustered regularly interspaced short palindromic tandem repeats (CRISPR) .................................. 36, 43–45, 50, 56, 72, 73, 94, 96, 98–100, 102 Confocal ................................................ 94, 116–117, 119 Covalent inhibition .................... 301–303, 305, 306, 313

D Dimerization ............................................... 124, 154, 197 Disulfide............................. 157, 212, 220, 221, 224, 225

E Eya2 ...................................................................... 285–297

F Flow cytometry ...............47, 48, 52, 55, 83, 86, 90, 188 Fluorescent labeling ............................168–170, 179, 183 Fragment-based drug discovery (FBDD).................... 255 Fragment ligation................................................. 239–266

H High throughput screening.......112, 256, 303, 307, 309

I Immune detection........................................................... 55 Immunofluorescence ....................................59, 115–116, 120, 121, 130, 167, 177 Immunoprecipitation........................... 25–26, 28, 32, 34, 36–38, 126, 156–158, 166, 169, 178, 181–192 Immunotherapy .............................................................. 82 Inhibitor screening........................................................ 255 Interacting proteins......................................130, 165–179 Interactome ................................166, 167, 174, 178, 182 Irreversible inhibition .......................................... 302, 313

K Kinetics of covalent inhibition ..................................... 303

L Laforin .................................................3, 4, 6, 7, 9, 12–15 Localization ..........................................22, 111, 112, 121, 123, 131, 167, 177, 205

M Malachite green assay..........................287, 289, 294–297 MAPK phosphatases ....................................................... 39 Mass spectrometry ................................... 26, 29, 38, 124, 131, 165–179, 183, 185, 187, 190, 221, 274 Microscopy .......................... 31, 50, 64, 70, 94, 111–120 MK-STYX .................................22–25, 32, 34, 36, 38, 39

N Nonconserved cysteine residues................................... 273 Non-homologous end-joining (NHEJ) ..................44, 70 Nonsense mutations ........................................1, 2, 11, 14

Damien The´venin and Jo¨rg P. Mu¨ller (eds.), Protein Tyrosine Phosphatases: Methods and Protocols, Methods in Molecular Biology, vol. 2743, https://doi.org/10.1007/978-1-0716-3569-8, © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024

317

PROTEIN TYROSINE PHOSPHATASES: METHODS AND PROTOCOLS

318 Index O

3-O-methylfluorescein phosphate (OMFP) .......................... 286–289, 291–294, 296 Osteoclast ..............................................57, 64, 67, 68, 70 OT-I ....................................................... 82, 84–86, 88–92 Ovarian carcinoma ............................................. 82, 87, 88 Oxidation.................................................... 161, 176, 212, 213, 216, 218, 220, 221, 223–235

P Phosphatase .......................................... 3, 5, 6, 12, 22–25, 28, 33, 34, 37, 57–77, 94, 112–121, 129, 135–138, 149, 151, 153, 165–167, 169, 172, 174, 177, 178, 182, 183, 191, 211, 212, 224, 239, 240, 242, 252, 255, 258, 261 Phosphatase activity ........................................23, 94, 124, 136–138, 148, 166, 308 Phosphatase assay .......................................................... 224 Phosphatase of regenerating liver 1 (PRL-1) .............. 212 Phosphatase substrate .......................................... 135–151 Phosphotyrosine mimetics..................243, 246–250, 266 Phosphotyrosine phosphatase non-receptor type 22 (PTPN22) ................................................81–92 pNPP assay............................................................ 280, 281 Premature termination codon (PTC) ........................ 1–17 Protein expression ...................................... 16, 31, 44, 50, 73, 128, 199, 202 Protein–protein interactions (PPIs) .................... 182, 189 Protein translation............................................................. 2 Protein tyrosine phosphatase (PTP) .............. 3–7, 10–14, 16, 17, 22–25, 27, 36, 43–55, 58, 59, 93, 94, 123–127, 129, 138, 143, 145, 148, 165–179, 181–192, 195, 196, 212, 223, 224, 239–266, 271–281, 303, 304, 306–310, 313

Proteomics................................................... 138, 171, 174 Proximity labeling (PL) ......................166, 167, 172, 178 Pseudophosphatases..................................................22–39 PTEN................................. 3, 4, 6–8, 10, 12–14, 94, 212 PTP4A1 ................................................................ 211–221

R Receptor protein tyrosine phosphatase (RPTP)........ 48, 51, 55, 124, 153, 154, 195–208 Reduction .......................................... 145, 197, 213, 214, 217, 218, 221, 227, 231, 235 Reversible oxidation.....................................212, 223–235

S Screening .............................99, 102–104, 253, 255–258, 260, 262–266, 273, 274, 276, 280, 281, 307 shRNA knockdown...................................................27, 36 Signaling networks ............................................... 153–162 Src-homology-2-containing PTP 2 (SHP2)............................................ 4, 58, 95, 181, 255, 258–260, 263, 272–275, 278, 279, 281 STYX domains................................................................. 25 Substrate identification ................................................. 124 Substrate trapping mutants ................................. 124, 125 Sulfhydryl-reactive probes ................................... 225, 234 Synergistic activation mediator (SAM) ............. 44, 48, 56

T T cell receptor.................................................................. 81 Translational readthrough .......................................... 1–17 Transmembrane domain (TM) interaction ................. 196 Tyrosine phosphatase................... 81, 166, 168, 177–179 Tyrosine phosphorylation.......................... 123, 127, 130, 154, 156, 301