Genotype Phenotype Coupling. Methods and Protocols 978-1-4939-9852-4, 978-1-4939-9853-1

Table of contents :
Preface......Page 6
Contents......Page 8
Contributors......Page 10
1 Introduction......Page 14
2.2 In Vitro Selection......Page 16
2.6 Detection of TOB......Page 17
3.2 In Vitro Selection......Page 18
3.3 Agarose Gel Electrophoresis Verification......Page 19
3.5 Molecular Cloning......Page 20
3.6 Determination of Dissociation Constants......Page 21
3.7 Specificity Analysis......Page 22
3.9 Aptamer Truncation......Page 23
3.10 Preparation of Gold Nanoparticles (AuNPs)......Page 24
3.12 Detection of Tobramycin Using AuNP-Based Colorimetric Assay......Page 25
4 Notes......Page 27
References......Page 30
1 Introduction......Page 32
2.2.1 Primers Used for Generation of the 5′-Flanking Region of the Ribosome Display Construct and the Randomized Positions of .........Page 34
2.3 PCR......Page 35
2.7 Selection......Page 36
3 Methods......Page 37
3.1.1 Production of the Aho7c Library Fragments......Page 38
pFP RDV1......Page 40
pFP RDV2......Page 41
3.1.5 Subcloning for Sequencing......Page 42
3.2 In Vitro Transcription......Page 43
3.3 Ribosome Display Selection in a 96-Well Microtiter Plate......Page 44
3.4 Additional Selection Rounds......Page 46
3.5 Selection Using Magnetic Beads......Page 47
3.6 Follow-Up of the Selection......Page 48
3.7 Screening of Clones and Sequence Analysis......Page 49
4 Notes......Page 50
References......Page 53
1 Introduction......Page 55
2.2 In Vitro Transcription......Page 59
2.3 Synthesis of Puromycin-Linker (See Note 2)......Page 60
2.5.3 Reverse Transcription (Synthesis of mRNA/cDNA-POI Fusion Molecule)......Page 61
3 Methods......Page 62
3.3.1 Reduction of the Puromycin Segment......Page 63
3.3.4 HPLC Purification of the Puromycin-Linker......Page 64
3.5.1 In Vitro Translation (Synthesis of mRNA-POI Fusion Molecule)......Page 65
3.5.5 His-Tag Affinity Purification of mRNA/cDNA-POI Fusion (i.e., cDNA Display) Molecules......Page 66
4 Notes......Page 67
References......Page 68
1 Introduction......Page 69
2.1 Three-Finger (3-F) Peptide Library......Page 72
2.2 Transcription and Translation......Page 74
2.4 Selection by Binding......Page 75
3.1.2 Assembling Oligonucleotides......Page 76
3.2.1 3-F mRNA Library Generation......Page 77
3.2.3 Translation and Protein Fusion to Linker......Page 79
3.3 cDNA Display: Immobilization and Reverse Transcription......Page 81
3.4.2 Target Selection......Page 82
3.5 Preparation of AChBP-Conjugated Magnetic Beads......Page 83
4 Notes......Page 84
References......Page 88
Abbreviations......Page 90
1 Introduction......Page 91
2 Materials......Page 93
2.3 S. aureus Immunoprecipitation and Flow Cytometry......Page 94
3.1 Preparation of Electrocompetent S. aureus Cells......Page 95
3.3 Electroporation of Competent S. aureus......Page 96
3.4 S. aureus Surface Display and Immunoprecipitation in a Plate Format......Page 97
3.5 S. aureus Surface Display and Immunoprecipitation in a Tube Format......Page 98
3.6 S. aureus Surface Display in a Flow Cytometry Format......Page 99
4 Notes......Page 100
References......Page 103
1 Introduction......Page 106
2.2 Linearization of the Phagemid Over PCR......Page 108
2.5 Library Quality Control......Page 109
2.8 Coating of Microtiter Plate Wells, Panning, and Phage Titration......Page 110
3.1 Design of the Peptide Library......Page 111
3.2 Amplification of the Library Oligomer......Page 113
3.4 Hot Fusion Reaction......Page 114
3.6 Library Quality Control......Page 115
3.7 Library Packaging......Page 116
3.9 Panning Procedure for a Biotinylated Protein......Page 117
4 Notes......Page 121
References......Page 124
1 Introduction......Page 125
2.1 Error-Prone PCR......Page 127
2.2 Mutant Library Construction......Page 128
2.3 Library Rescue and Selection......Page 129
2.5 Reformatting of Selected Clones for Transient Expression in Eukaryotic Cells......Page 130
2.7 Sequence Determination and Large-Scale Expression and Purification of Lead Clones......Page 131
2.10 Analytical Size Exclusion Chromatography......Page 132
3.2 Mutant Library Construction......Page 133
3.3.1 Library Phage Rescue......Page 134
3.3.3 Phage Display Selection (See Note 16)......Page 135
3.4 Periprep ELISA......Page 137
3.5 Reformatting of Selected Clones for Transient Expression in Eukaryotic Cells......Page 138
3.6 Surface Plasmon Resonance (SPR) Off-Rate Selection......Page 139
3.7 Sequence Determination, Large-Scale Plasmid Preparation, Transient Expression, and Purification of Lead Clones......Page 140
3.9.1 Immobilization of Human Serum Albumin on CM5 Chip......Page 142
3.9.2 Affinity Measurement of Humanized Anti-HSA VNARs......Page 143
3.10 Analytical SEC......Page 144
4 Notes......Page 145
References......Page 151
1 Introduction......Page 153
2.2 Panning......Page 155
2.5 ELISA of Soluble Monoclonal Antibody Fragments......Page 156
3.2 Panning......Page 157
3.4 Production of Soluble Monoclonal Antibody Fragments in Microtiter Plates......Page 159
3.5 ELISA of Soluble Monoclonal Antibody Fragments......Page 160
4 Notes......Page 161
References......Page 163
1 Introduction......Page 166
2.1 Isolation of IgG from Human Serum......Page 167
2.3 Purification and Amplification of the Nucleic Acid from the Selected Library Members (See Note 4)......Page 168
2.4 Preparation of Samples for Deep Sequencing......Page 169
3.1 Isolation of IgG from Human Serum......Page 170
3.2 Selection of a Library Against Isolated IgG......Page 171
3.4 Generating cDNA of Selected Library Members......Page 173
3.6 Preparation of Samples for Deep Sequencing......Page 174
4 Notes......Page 176
References......Page 180
1 Introduction......Page 181
2.2 Plasmids......Page 183
2.3 Media......Page 184
2.7 Gene-Specific Amplification of VHH Regions......Page 186
3 Methods......Page 187
3.1 PBMC Isolation by Density Gradient Centrifugation......Page 188
3.2 Total RNA Isolation from PBMCs......Page 189
3.5 Destination Vector (pDest) Digestion......Page 190
3.6 Yeast Transformation for Library Establishment......Page 191
3.9 Induction of VHH Expression for Antibody Surface Display......Page 192
3.10.1 Antibody Display Control......Page 193
4 Notes......Page 194
References......Page 196
1 Introduction......Page 198
2.2 Subcloning of Anti-idiotypic vNAR Variants......Page 201
2.4 Characterization of Recombinant Anti-idiotypic vNAR Variants......Page 202
3.1 Induction of vNAR Expression......Page 203
3.2 Immunofluorescence Staining of Yeast Cells......Page 204
3.3 Verification of Anti-idiotypic Binding on the Surface of Yeast......Page 205
3.4 Reformatting and Recombinant Expression of Anti-idiotypic vNAR Domains......Page 206
3.4.1 Isolation of vNAR DNA from Yeast Cells and Subcloning into the pEXPR Plasmid......Page 207
3.4.2 Transfection of Expi293F Cells......Page 209
3.4.3 Characterization of Recombinant and Anti-idiotypic vNARs via Bio-Layer Interferometry (BLI)......Page 210
3.4.4 Characterization of Recombinant and Anti-idiotypic vNARs via Ligand Binding Assays (ELISA)......Page 211
4 Notes......Page 213
References......Page 214
1 Introduction......Page 217
2.1 Single Clone Construction for Yeast Surface Display......Page 218
2.2 Oligonucleotides......Page 219
3.2 Single Clone Construction: Backbone......Page 220
3.3 Single Clone Construction: Transformation......Page 221
3.4 Single Clone Flow Cytometry Analysis......Page 223
4 Notes......Page 225
References......Page 227
1.1 Structure and Function of TCRs......Page 229
1.3 Yeast Display of TCR......Page 230
1.3.1 Preferred Methods for Construction of TCR Yeast Display Libraries......Page 232
1.3.2 Gap Repair-Driven Homologous Recombination in TCR Library Construction......Page 235
1.3.4 Detection Agents for Labeling of Yeast Display TCR Libraries......Page 237
2.2 Solutions and Buffers......Page 238
2.3 Media......Page 239
2.6 Yeast Strains......Page 240
3.1.1 TCR Yeast Library Construction Using a Bicistronic Vector......Page 241
3.1.2 TCR Yeast Display Library Construction Using Yeast Mating......Page 245
3.1.3 TCR Yeast Display Libraries Constructed with an Integrative Vector......Page 247
3.2 Induction of TCR Expression in Yeast Cells......Page 248
3.3 Selection of TCR-Displaying Yeast Cells with MACS (Magnetic-Activated Cell Sorting)......Page 249
4 Notes......Page 251
References......Page 252
1 Introduction......Page 255
2.1 Cells and Plasmids......Page 258
2.2 Yeast Media and Reagents......Page 259
2.3 Reagents and Equipment for FACS......Page 260
3.2 Induction of Expression for the Display of Antibody Fab Fragments on Yeast Cells......Page 261
3.3.1 Determination of Saturating Antigen Concentrations......Page 262
3.3.2 Labeling Strategy......Page 263
Preparing the Antigen Binding Control Sample......Page 264
3.4 Fluorescence-Activated Cell Sorting......Page 265
3.5.2 Epitope Binning ELISA......Page 267
4 Notes......Page 269
References......Page 271
1 Introduction......Page 273
2.1 Yeast Media......Page 275
2.2 Reagents......Page 277
3.1.1 BSA-Hapten Conjugate for Immunization......Page 278
3.1.2 Fc-Hapten Conjugate for Monitoring and Screening Purposes......Page 280
3.2.2 Serum Titer ELISA......Page 281
3.4 Fluorescence-Activated Cell Sorting......Page 282
3.4.1 Preparing the Sort Sample......Page 283
3.4.3 Preparing the Antigen-Binding Control Sample......Page 284
3.5 Characterization of Hapten Binders......Page 285
4 Notes......Page 288
References......Page 292
1 Introduction......Page 294
2.2 cDNA Synthesis and Amplification of VH and VL Genes......Page 295
2.3 Construction of Yeast Surface Display Library......Page 296
3.1 Chicken Immunization and Isolation of Total Spleen RNA......Page 297
3.2 cDNA Synthesis and Amplification of VH and VL Genes......Page 298
3.3 Generation of Chicken scFv Library in Yeast......Page 299
3.3.2 Yeast Transformation......Page 300
3.4.1 Induction of scFv Display......Page 301
3.4.2 Analysis of Surface Presentation......Page 302
Second Sorting Round......Page 303
3.6 Sequencing of Antigen-Binding scFvs......Page 304
4 Notes......Page 305
References......Page 306
1 Introduction......Page 308
2.1.1 Media, Buffers, and Reagents......Page 312
2.1.2 Equipment and Consumables......Page 313
3.1 Cell Panning of Yeast Surface-Displayed Libraries......Page 314
3.1.2 Yeast Cell Preparation......Page 315
3.2.1 Target-Negative Mammalian Cell Preparation......Page 316
3.3.1 Titratable Avidity Reduction......Page 318
3.3.3 Generating a Fluorescence Calibration Curve to Assess Ligand Expression......Page 319
3.3.4 Analysis of Ligand Expression Data......Page 320
4 Notes......Page 322
References......Page 323
1 Introduction......Page 326
2.1 Strains and Plasmids......Page 328
2.4 Yeast Transformation and Culture......Page 329
2.6 SDS-PAGE and Immunoblotting......Page 330
3.1 Cloning the Gene of Interest into pCT-CT-F2A or pCT-NT-F2A......Page 331
3.3 Induction of Protein Secretion and Purification......Page 332
3.4 Detection of Secreted Protein by SDS-PAGE and Western Blotting......Page 333
3.5 Detection of Secreted Protein by Flow Cytometry......Page 334
3.7 Preparing Secreted Protein for Use in Functional Assays......Page 335
4 Notes......Page 336
References......Page 338
1 Introduction......Page 340
2.2 Plasmid and Primer Design......Page 342
2.4 Reagents......Page 343
3 Methods......Page 344
3.1 CDR-H3 Library Amplification and Cloning......Page 345
3.3 Yeast Library Generation......Page 346
3.4 Single-Clone Transformation......Page 347
3.5 Library Mating......Page 348
3.6 Cell Surface Manipulation and IgG Display......Page 349
3.7.2 Preparation of Library Cells......Page 350
3.8 Single-Clone Analysis......Page 351
4 Notes......Page 352
References......Page 353
1 Introduction......Page 355
2.1 Sleeping Beauty, GFP Expression Plasmids......Page 357
2.3 Transient and/or Stable Transfection......Page 358
3.1 Plasmid Preparation and Purification......Page 359
3.2.1 Determination of the Optimal PEI to pDNA ratio (See Note 3)......Page 360
3.2.2 Stable Transfection of 293F Cells......Page 362
4 Notes......Page 363
References......Page 364
1 Introduction......Page 366
2 Materials......Page 369
2.1 Equipment......Page 370
2.2 Software......Page 371
2.4 Molecular Tools for Working with Peptide Libraries......Page 372
2.6 Cell Sorting by MACS and Flow Cytometry......Page 375
3.2 Library Design for Optimized Cloning into SDGF......Page 376
3.3 Cloning Pooled Oligonucleotide Libraries into SDGF......Page 377
3.4 Target Protein Design and Labeling......Page 379
3.6 Cell Staining for a Large-Scale, High-Diversity Screen Using an autoMACS Pro Sorter......Page 380
3.7 Medium-Scale FACS Staining for Low Diversity (

Citation preview

Methods in Molecular Biology 2070

Stefan Zielonka · Simon Krah Editors

Genotype Phenotype Coupling Methods and Protocols

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.

Genotype Phenotype Coupling Methods and Protocols

Edited by

Stefan Zielonka and Simon Krah Protein Engineering and Antibody Technologies, Merck KGaA, Darmstadt, Germany

Editors Stefan Zielonka Protein Engineering and Antibody Technologies Merck KGaA Darmstadt, Germany

Simon Krah Protein Engineering and Antibody Technologies Merck KGaA Darmstadt, Germany

ISSN 1064-3745 ISSN 1940-6029 (electronic) Methods in Molecular Biology ISBN 978-1-4939-9852-4 ISBN 978-1-4939-9853-1 (eBook) https://doi.org/10.1007/978-1-4939-9853-1 © Springer Science+Business Media, LLC, part of Springer Nature 2020 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express 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: 233 Spring Street, New York, NY 10013, U.S.A.

Preface In the last decades, the field of protein engineering has made tremendous progress with respect to complexity. Nowadays, through rational design, evolutionary approaches (i.e., randomized libraries) or via a combination of both, proteins can be tailor-made for specific applications such as target binding or pharmacokinetics. In this context, genotype–phenotype coupling has enabled the isolation of a desired protein candidate with prescribed properties from unprecedented diverse protein repertoires. The principle of genotype–phenotype coupling relies on the linkage of a protein of interest (POI; the phenotype) to its respective genetic information (the genotype). This allows for barcoding of billions of different variants, from which specific proteins can be identified after a selection process by DNA isolation and subsequent sequencing. One of the first methods developed in this field was the display of peptides, followed by the display of antibody fragments on filamentous phage particles. For this, antibody presenting phages are incubated with recombinant antigen or antigen expressing cells, unbound variants are washed away, and bound particles are used for the infection of E. coli cells. In an iterative process, antigen-specific variants are enriched, and a large initial diversity can be reduced to only a small fraction of antigen-specific antibody fragments. Compared to hybridoma approaches, which rely on a fusion of murine B-cells after an immunization with immortalized tumor cells, phage display enabled the identification of human antibody variable regions. Fully human antibodies should display better pharmacokinetic profiles when applied as therapeutics in humans, as the risk of generating an immune response against the drug might be reduced. Moreover, phage display drastically increased the throughput during drug discovery and enabled an optimization of antibodies with respect to different parameters (e.g., affinity, stability). For their pioneering work in directed evolution, George Smith (inventor of phage display) and Sir Gregory Winter (one of the main inventors of antibody phage display) have been awarded with the Nobel Prize in chemistry in 2018. Since the development of phage display in the 1980s, several other methodologies emerged that follow the strategy of genotype–phenotype coupling, such as ribosomal display, mRNA and DNA display, as well as cellular display on bacteria, yeast, and mammalian cells. All of these have their own advantages and limitations. As an example, Yeast Surface Display (YSD), published by Boder and Wittrup in 1997, makes use of the baker’s yeast S. cerevisiae as a host for protein expression and display. The larger size of yeast cells compared to phage particles allows the selection of YSD libraries with fluorescence activated cell sorting (FACS). This allows a better control and analysis of binding-signals during the sorting process compared to panning of phage display libraries. Moreover, yeast cells should be able to process complex proteins like antibodies more efficiently than E. coli because of their higher biological complexity (e.g., protein folding machinery and posttranslational modifications). However, as of now, the size of YSD-libraries is limited to a maximum of 109 variants, while phage display libraries can extend this number by several orders of magnitude. Along with this, especially naı¨ve antibody or synthetic protein libraries are better suited for phage display approaches. The principle of genotype–phenotype coupling has also been applied to DNA and RNA molecules, referred to as aptamers. With unique 3D structures, such molecules can bind

v

vi

Preface

their targets with high affinities and specificities. Aptamers are typically selected in a process called systematic evolution of ligands by exponential enrichment (SELEX). In this edition of Methods in Molecular Biology we have summarized protocols that cover different aspects of genotype–phenotype coupling technologies. Therefore, we distinguished four different categories, starting with in vitro methods over prokaryotic display systems, lower eukaryotes, and finally mammalian cells. We want to thank all contributors that helped us to make a handbook that aims at giving an overview of today’s technologies in this exciting and continuously evolving field. Darmstadt, Germany

Stefan Zielonka Simon Krah

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

1 Screening, Post-SELEX Optimization and Application of DNA Aptamers Specific for Tobramycin. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Nandi Zhou, Rongfeng Cai, and Xuyan Han 2 Affitins: Ribosome Display for Selection of Aho7c-Based Affinity Proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Valentina Kalichuk, Stanimir Kambarev, Ghislaine Be´har, Benjamin Chalopin, Axelle Renodon-Cornie`re, Barbara Mouratou, and Fre´de´ric Pecorari 3 cDNA Display: A Stable and Simple Genotype–Phenotype Coupling Using a Cell-Free Translation System. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Hidenao Arai, Shigefumi Kumachi, and Naoto Nemoto 4 cDNA Display of Disulfide-Containing Peptide Library and In Vitro Evolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tai Kubo and Mohammed Naimuddin 5 Rapid Antigen and Antibody-Like Molecule Discovery by Staphylococcal Surface Display . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Marco Cavallari 6 Restriction-Free Construction of a Phage-Presented Very Short Macrocyclic Peptide Library . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Valentin Jakob, Saskia Helmsing, Michael Hust, and Martin Empting 7 In Vitro Maturation of a Humanized Shark VNAR Domain to Improve Its Biophysical Properties. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . John Steven, Obinna C. Ubah, Magdalena Buschhaus, Marina Kovaleva, Laura Ferguson, Andrew J. Porter, and Caroline J. Barelle 8 Antibody Phage Display: Antibody Selection in Solution Using Biotinylated Antigens. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ¨ hner, Esther V. Wenzel, Kristian D. R. Roth, Giulio Russo, Viola Fu Saskia Helmsing, Andre´ Frenzel, and Michael Hust 9 Assessing Antibody Specificity in Human Serum Using Deep Sequence-Coupled Biopanning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Kathryn M. Frietze, Susan B. Core, Alexandria Linville, Bryce Chackerian, and David S. Peabody 10 Isolation of Antigen-Specific VHH Single-Domain Antibodies by Combining Animal Immunization with Yeast Surface Display . . . . . . . . . . . . . ¨ nther, Lukas Roth, Simon Krah, Janina Klemm, Ralf Gu Lars Toleikis, Michael Busch, Stefan Becker, and Stefan Zielonka 11 Selection and Characterization of Anti-idiotypic Shark Antibody Domains . . . . . Doreen Ko¨nning, Stefan Zielonka, Anna Kaempffe, Sebastian J€ a ger, Harald Kolmar, and Christian Schro¨ter

vii

v ix

1

19

43

57

79

95

115

143

157

173

191

viii

12

13

14

15

16

17

18

19

20

21

22 23

Contents

Simplifying the Detection of Surface Presentation Levels in Yeast Surface Display by Intracellular tGFP Expression . . . . . . . . . . . . . . . . . . . . Steffen C. Hinz, Adrian Elter, Julius Grzeschik, Jan Habermann, Bastian Becker, and Harald Kolmar Methods for Construction of Yeast Display Libraries of Four-Domain T-Cell Receptors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Fla´vio Sa´dio, Gerhard Stadlmayr, Katja Eibensteiner, ¨ ker, and Gordana Wozniak-Knopp Katharina Stadlbauer, Florian Ru Isolation of Tailor-Made Antibody Fragments from Yeast-Displayed B-Cell Receptor Repertoires by Multiparameter Fluorescence-Activated Cell Sorting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Anna Kaempffe, Sebastian J€ a ger, Doreen Ko¨nning, Harald Kolmar, and Christian Schro¨ter Isolation of Anti-Hapten Antibodies by Fluorescence-Activated Cell Sorting of Yeast-Displayed B-Cell Receptor Gene Repertoires. . . . . . . . . . . . Sebastian J€ a ger, Simon Krah, Doreen Ko¨nning, Janis Rosskopf, Stephan Dickgiesser, Nicolas Rasche, Harald Kolmar, Stefan Hecht, and Christian Schro¨ter Rapid Generation of Chicken Immune Libraries for Yeast Surface Display . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jan P. Bogen, Julius Grzeschik, Simon Krah, Stefan Zielonka, and Harald Kolmar Ligand Engineering via Yeast Surface Display and Adherent Cell Panning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Lawrence A. Stern, Patrick S. Lown, and Benjamin J. Hackel Simultaneous Soluble Secretion and Surface Display of Proteins in Saccharomyces cerevisiae Using Inefficient Ribosomal Skipping . . . . . . . . . . . . . Carlos A. Cruz-Teran, Kaitlyn Bacon, and Balaji M. Rao Chemical Modification of the Yeast Cell Surface Allows the Switch Between Display and Soluble Secretion of Full-Length Antibodies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ¨ nther, Stefan Becker, Stefan Zielonka, Simon Krah, Ralf Gu and Laura Rhiel Advanced Establishment of Stable Recombinant Human Suspension Cell Lines Using Genotype–Phenotype Coupling Transposon Vectors. . . . . . . . . Karen Berg, Vanessa Nicole Sch€ a fer, Natalie Tschorn, and Jo¨rn Stitz Mammalian Surface Display Screening of Diverse Cystine-Dense Peptide Libraries for Difficult-to-Drug Targets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Zachary R. Crook, Gregory P. Sevilla, Andrew J. Mhyre, and James M. Olson Engineering Antibodies on the Surface of CHO Cells . . . . . . . . . . . . . . . . . . . . . . . Annalee W. Nguyen, Kevin Le, and Jennifer A. Maynard Single B Cell Cloning and Production of Rabbit Monoclonal Antibodies . . . . . . Juliet Rashidian and Joshua Lloyd

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

211

223

249

267

289

303

321

335

351

363

397 423

443

Contributors HIDENAO ARAI  Epsilon Molecular Engineering, Inc., Saitama, Japan KAITLYN BACON  Department of Chemical and Biomolecular Engineering, North Carolina State University, Raleigh, NC, USA CAROLINE J. BARELLE  Elasmogen Ltd., Aberdeen, UK BASTIAN BECKER  Institute for Organic Chemistry and Biochemistry, Technische Universit€ at Darmstadt, Darmstadt, Germany STEFAN BECKER  Protein Engineering and Antibody Technologies (PEAT), Merck KGaA, Darmstadt, Germany GHISLAINE BE´HAR  CRCINA, INSERM, CNRS, Universite´ d’Angers, Universite´ de Nantes, Nantes, France KAREN BERG  Pharmaceutical Biotechnology, Faculty of Applied Natural Sciences, STEPs Institute, TH Ko¨ln—University of Applied Sciences, Leverkusen, Germany; Research Group Translational Hepatology and Stem Cell Biology, Cluster of Excellence REBIRTH, Hannover Medical School, Hannover, Germany JAN P. BOGEN  Institute for Organic Chemistry and Biochemistry, Technische Universit€ at Darmstadt, Darmstadt, Germany MICHAEL BUSCH  Discovery Pharmacology, Merck KGaA, Darmstadt, Germany MAGDALENA BUSCHHAUS  Elasmogen Ltd., Aberdeen, UK RONGFENG CAI  The Key Laboratory of Carbohydrate Chemistry and Biotechnology, Ministry of Education, School of Biotechnology, Jiangnan University, Wuxi, China MARCO CAVALLARI  BIOSS Centre for Biological Signalling Studies, University of Freiburg, Freiburg, Germany BRYCE CHACKERIAN  Department of Molecular Genetics and Microbiology, School of Medicine, University of New Mexico Health Sciences, Albuquerque, NM, USA BENJAMIN CHALOPIN  CRCINA, INSERM, CNRS, Universite´ d’Angers, Universite´ de Nantes, Nantes, France SUSAN B. CORE  Department of Molecular Genetics and Microbiology, School of Medicine, University of New Mexico Health Sciences, Albuquerque, NM, USA ZACHARY R. CROOK  Clinical Research Division, Fred Hutchinson Cancer Research Center, Seattle, WA, USA CARLOS A. CRUZ-TERAN  Department of Chemical and Biomolecular Engineering, North Carolina State University, Raleigh, NC, USA STEPHAN DICKGIESSER  Antibody-Drug Conjugates and Targeted NBE Therapeutics, Merck KGaA, Darmstadt, Germany KATJA EIBENSTEINER  Christian Doppler Laboratory for Innovative Immunotherapeutics, Department of Biotechnology, University of Natural Resources and Life Sciences (BOKU), Vienna, Austria ADRIAN ELTER  Institute for Organic Chemistry and Biochemistry, Technische Universit€ at Darmstadt, Darmstadt, Germany MARTIN EMPTING  Department of Drug Design and Optimization (DDOP), HelmholtzInstitute for Pharmaceutical Research Saarland (HIPS)—Helmholtz Centre for Infection Research (HZI), Saarbru¨cken, Germany; Department of Pharmacy, Saarland University, Saarbru¨cken, Germany

ix

x

Contributors

LAURA FERGUSON  Elasmogen Ltd., Aberdeen, UK ANDRE´ FRENZEL  Institut fu¨r Biochemie, Biotechnologie und Bioinformatik, Abteilung Biotechnologie, Technische Universit€ a t Braunschweig, Braunschweig, Germany; YUMAB GmbH, Braunschweig, Germany KATHRYN M. FRIETZE  Department of Molecular Genetics and Microbiology, School of Medicine, University of New Mexico Health Sciences, Albuquerque, NM, USA; Clinical and Translational Science Center, University of New Mexico Health Sciences, Albuquerque, NM, USA VIOLA FU¨HNER  Institut fu¨r Biochemie, Biotechnologie und Bioinformatik, Abteilung Biotechnologie, Technische Universit€ a t Braunschweig, Braunschweig, Germany JULIUS GRZESCHIK  Institute for Organic Chemistry and Biochemistry, Technische Universit€ at Darmstadt, Darmstadt, Germany RALF GU¨NTHER  Protein Engineering and Antibody Technologies (PEAT), Merck KGaA, Darmstadt, Germany JAN HABERMANN  Institute for Organic Chemistry and Biochemistry, Technische Universit€ at Darmstadt, Darmstadt, Germany BENJAMIN J. HACKEL  Department of Chemical Engineering and Materials Science, University of Minnesota—Twin Cities, Minneapolis, MN, USA XUYAN HAN  The Key Laboratory of Carbohydrate Chemistry and Biotechnology, Ministry of Education, School of Biotechnology, Jiangnan University, Wuxi, China STEFAN HECHT  Antibody-Drug Conjugates and Targeted NBE Therapeutics, Merck KGaA, Darmstadt, Germany SASKIA HELMSING  Institut fu¨r Biochemie, Biotechnologie und Bioinformatik, Abteilung Biotechnologie, Technische Universit€ a t Braunschweig, Braunschweig, Germany STEFFEN C. HINZ  Institute for Organic Chemistry and Biochemistry, Technische Universit€ at Darmstadt, Darmstadt, Germany MICHAEL HUST  Institut fu¨r Biochemie, Biotechnologie und Bioinformatik, Abteilung Biotechnologie, Technische Universit€ a t Braunschweig, Braunschweig, Germany; YUMAB GmbH, Science Campus Braunschweig Su¨d, Braunschweig, Germany SEBASTIAN J€AGER  Antibody-Drug Conjugates and Targeted NBE Therapeutics, Merck KGaA, Darmstadt, Germany; Institute for Organic Chemistry and Biochemistry, Technische Universit€ a t Darmstadt, Darmstadt, Germany VALENTIN JAKOB  Department of Drug Design and Optimization (DDOP), HelmholtzInstitute for Pharmaceutical Research Saarland (HIPS)—Helmholtz Centre for Infection Research (HZI), Saarbru¨cken, Germany; Department of Pharmacy, Saarland University, Saarbru¨cken, Germany ANNA KAEMPFFE  Antibody-Drug Conjugates and Targeted NBE Therapeutics, Merck KGaA, Darmstadt, Germany; Institute for Organic Chemistry and Biochemistry, Technische Universit€ a t Darmstadt, Darmstadt, Germany VALENTINA KALICHUK  CRCINA, INSERM, CNRS, Universite´ d’Angers, Universite´ de Nantes, Nantes, France STANIMIR KAMBAREV  CRCINA, INSERM, CNRS, Universite´ d’Angers, Universite´ de Nantes, Nantes, France JANINA KLEMM  Protein Engineering and Antibody Technologies (PEAT), Merck KGaA, Darmstadt, Germany HARALD KOLMAR  Institute for Organic Chemistry and Biochemistry, Technische Universit€ at Darmstadt, Darmstadt, Germany

Contributors

xi

DOREEN KO¨NNING  Antibody-Drug Conjugates and Targeted NBE Therapeutics, Merck KGaA, Darmstadt, Germany MARINA KOVALEVA  Elasmogen Ltd., Aberdeen, UK SIMON KRAH  Protein Engineering and Antibody Technologies (PEAT), Merck KGaA, Darmstadt, Germany TAI KUBO  Molecular Profiling Research Center for Drug Discovery, National Institute of Advanced Industrial Science and Technology (AIST), Tokyo, Japan; JapanAIST-UTokyo Advanced Operando Measurement Technology Open Innovation Laboratory, Kashiwa, Chiba, Japan; United Graduate School of Drug Discovery and Medical Information Sciences, Gifu University, Gifu, Japan SHIGEFUMI KUMACHI  Epsilon Molecular Engineering, Inc., Saitama, Japan KEVIN LE  Department of Chemical Engineering, The University of Texas at Austin, Austin, TX, USA ALEXANDRIA LINVILLE  Department of Molecular Genetics and Microbiology, School of Medicine, University of New Mexico Health Sciences, Albuquerque, NM, USA JOSHUA LLOYD  Cell Marque/MilliporeSigma, Rocklin, CA, USA PATRICK S. LOWN  Department of Chemical Engineering and Materials Science, University of Minnesota—Twin Cities, Minneapolis, MN, USA JENNIFER A. MAYNARD  Department of Chemical Engineering, The University of Texas at Austin, Austin, TX, USA ANDREW J. MHYRE  Clinical Research Division, Fred Hutchinson Cancer Research Center, Seattle, WA, USA BARBARA MOURATOU  CRCINA, INSERM, CNRS, Universite´ d’Angers, Universite´ de Nantes, Nantes, France MOHAMMED NAIMUDDIN  Molecular Profiling Research Center for Drug Discovery, National Institute of Advanced Industrial Science and Technology (AIST), Tokyo, Japan; Department of Applied Biology, Adama Science and Technology University, Adama, Ethiopia NAOTO NEMOTO  Epsilon Molecular Engineering, Inc., Saitama, Japan; Graduate School of Science and Engineering, Saitama University, Saitama, Japan ANNALEE W. NGUYEN  Department of Chemical Engineering, The University of Texas at Austin, Austin, TX, USA JAMES M. OLSON  Clinical Research Division, Fred Hutchinson Cancer Research Center, Seattle, WA, USA DAVID S. PEABODY  Department of Molecular Genetics and Microbiology, School of Medicine, University of New Mexico Health Sciences, Albuquerque, NM, USA FRE´DE´RIC PECORARI  CRCINA, INSERM, CNRS, Universite´ d’Angers, Universite´ de Nantes, Nantes, France ANDREW J. PORTER  Elasmogen Ltd., Aberdeen, UK; Scottish Biologics Facility, School of Medical Sciences, University of Aberdeen, Aberdeen, UK BALAJI M. RAO  Department of Chemical and Biomolecular Engineering, North Carolina State University, Raleigh, NC, USA; Golden LEAF Biomanufacturing Training and Education Center (BTEC), North Carolina State University, Raleigh, NC, USA NICOLAS RASCHE  Antibody-Drug Conjugates and Targeted NBE Therapeutics, Merck KGaA, Darmstadt, Germany JULIET RASHIDIAN  Cell Marque/MilliporeSigma, Rocklin, CA, USA AXELLE RENODON-CORNIE`RE  CRCINA, INSERM, CNRS, Universite´ d’Angers, Universite´ de Nantes, Nantes, France

xii

Contributors

LAURA RHIEL  Protein Engineering and Antibody Technologies (PEAT), Merck KGaA, Darmstadt, Germany JANIS ROSSKOPF  Institute for Organic Chemistry and Biochemistry, Technische Universit€ at Darmstadt, Darmstadt, Germany; Antibody-Drug Conjugates and Targeted NBE Therapeutics, Merck KGaA, Darmstadt, Germany LUKAS ROTH  Protein Engineering and Antibody Technologies (PEAT), Merck KGaA, Darmstadt, Germany KRISTIAN D. R. ROTH  Institut fu¨r Biochemie, Biotechnologie und Bioinformatik, Abteilung Biotechnologie, Technische Universit€ a t Braunschweig, Braunschweig, Germany FLORIAN RU¨KER  Christian Doppler Laboratory for Innovative Immunotherapeutics, Department of Biotechnology, University of Natural Resources and Life Sciences (BOKU), Vienna, Austria GIULIO RUSSO  Institut fu¨r Biochemie, Biotechnologie und Bioinformatik, Abteilung Biotechnologie, Technische Universit€ a t Braunschweig, Braunschweig, Germany FLA´VIO SA´DIO  Christian Doppler Laboratory for Innovative Immunotherapeutics, Department of Biotechnology, University of Natural Resources and Life Sciences (BOKU), Vienna, Austria VANESSA NICOLE SCH€AFER  Pharmaceutical Biotechnology, Faculty of Applied Natural Sciences, STEPs Institute, TH Ko¨ln—University of Applied Sciences, Leverkusen, Germany CHRISTIAN SCHRO¨TER  Antibody-Drug Conjugates and Targeted NBE Therapeutics, Merck KGaA, Darmstadt, Germany GREGORY P. SEVILLA  Clinical Research Division, Fred Hutchinson Cancer Research Center, Seattle, WA, USA KATHARINA STADLBAUER  Christian Doppler Laboratory for Innovative Immunotherapeutics, Department of Biotechnology, University of Natural Resources and Life Sciences (BOKU), Vienna, Austria GERHARD STADLMAYR  Christian Doppler Laboratory for Innovative Immunotherapeutics, Department of Biotechnology, University of Natural Resources and Life Sciences (BOKU), Vienna, Austria LAWRENCE A. STERN  Department of Hematology and Hematopoietic Cell Transplantation, Beckman Research Institute of the City of Hope, Duarte, CA, USA JOHN STEVEN  Elasmogen Ltd., Aberdeen, UK JO¨RN STITZ  Pharmaceutical Biotechnology, Faculty of Applied Natural Sciences, STEPs Institute, TH Ko¨ln—University of Applied Sciences, Leverkusen, Germany LARS TOLEIKIS  Protein Engineering and Antibody Technologies (PEAT), Merck KGaA, Darmstadt, Germany NATALIE TSCHORN  Pharmaceutical Biotechnology, Faculty of Applied Natural Sciences, STEPs Institute, TH Ko¨ln—University of Applied Sciences, Leverkusen, Germany OBINNA C. UBAH  Elasmogen Ltd., Aberdeen, UK ESTHER V. WENZEL  Institut fu¨r Biochemie, Biotechnologie und Bioinformatik, Abteilung Biotechnologie, Technische Universit€ a t Braunschweig, Braunschweig, Germany GORDANA WOZNIAK-KNOPP  Christian Doppler Laboratory for Innovative Immunotherapeutics, Department of Biotechnology, University of Natural Resources and Life Sciences (BOKU), Vienna, Austria NANDI ZHOU  The Key Laboratory of Carbohydrate Chemistry and Biotechnology, Ministry of Education, School of Biotechnology, Jiangnan University, Wuxi, China STEFAN ZIELONKA  Protein Engineering and Antibody Technologies (PEAT), Merck KGaA, Darmstadt, Germany

Chapter 1 Screening, Post-SELEX Optimization and Application of DNA Aptamers Specific for Tobramycin Nandi Zhou, Rongfeng Cai, and Xuyan Han Abstract Tobramycin (TOB) is an aminoglycoside antibiotic. The residue of TOB in animal-derived foods and environment will be harmful to human health, and therefore the specific detection of TOB residue in food and water is of great importance. Herein, through magnetic beads-based SELEX, overall 37 ssDNA aptamers specific for TOB were screened after ten rounds of selection. The affinity and specificity of these aptamers were evaluated, among which No. 32 aptamer (Ap 32) exhibits excellent performance. Then a post-SELEX optimization of Ap 32 was carried out based on rational design, through which a truncated aptamer with the length of 34 nucleotides (Ap 32-2) was identified as the best aptamer for TOB. Finally, the application of the screened aptamer was explored. A colorimetric assay of TOB was established based on the aptamer-modified gold nanoparticles (AuNPs). In the range from 100 to 1400 nM, the absorbance of AuNPs solution at 520 nm was linearly decreased with the increased concentration of TOB. The detection limit was estimated to be 37.9 nM. The assay was applied to detect TOB residue in honey samples. Key words Magnetic beads, Single-stranded DNA aptamer, Dissociation constant, SELEX, Truncation, Tobramycin, Gold nanoparticles, Colorimetric detection

1

Introduction Aptamers are mainly synthetic oligonucleotides including singlestranded DNA (ssDNA) and RNA, with unique 3D structures that can recognize and bind to their cognate targets with high affinity and specificity [1, 2]. They can specifically bind to small organic molecules, peptides, proteins, even ions, cells or viruses, and form stable complexes [3–6]. Aptamers are usually obtained via an in vitro process called systematic evolution of ligands by exponential enrichment (SELEX) which was first reported in 1990 [7]. Since then, many articles have described the use of SELEX to select aptamers targeting different substances. The principle of the magnetic beads-based aptamer screening is to use the magnetic beads conjugated with a specific target to separate ssDNA with affinity for the target from the random ssDNA pool. Then, all

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2070, https://doi.org/10.1007/978-1-4939-9853-1_1, © Springer Science+Business Media, LLC, part of Springer Nature 2020

1

2

Nandi Zhou et al.

beads-bound DNA can be separated by an external magnet and further extracted [8]. The use of magnetic beads has greatly improved the SELEX process [9]. Because of their unique properties, such as controllable modification, long-term stability, low immunogenicity, fast tissue penetration, and short blood residence time, aptamers have gained increasing popularity as recognition elements for disease diagnosis and therapy in molecular medicine, especially aptamer-based molecular imaging [10–13]. Tobramycin (TOB) is an aminoglycoside antibiotic which acts effectively against a wide spectrum of Gram-negative bacteria [14]. This mechanism of action stems from its ability to interfere with bacterial protein synthesis process by irreversibly binding to ribosome, and then causing damages to cell membrane and cell death. TOB has been widely used as veterinary drugs for therapeutic and prophylactic purposes. Because of the side effects of TOB caused from the accumulation of TOB in certain organs and the transfer through the food chain, the detection of TOB in clinical samples, animal-derived foods and environment water is of great importance. A few methods have been reported for the detection of TOB residues in food products, such as microbiological assays [15], immunoassays [16], high-performance liquid chromatography (HPLC) [17], and capillary electrophoresis (CE) [18]. Owing to the lack of unsaturated bonds in TOB molecule, the detection of TOB through equipment analysis commonly requires a derivation step (pre- or post-column), thereby leading to several disadvantages, such as complexity, time consumption, and high cost. Since the report of aptamers for antibiotics, aptamer-based assays have been used to detect antibiotics. For example, kanamycin-specific aptamer reported by Song et al. has been widely applied to construct diverse biosensors for kanamycin [19]. An RNA aptamer for TOB was reported by Jiang et al. in 1998 [20]. However, its application in detection has not been explored much due to the poor stability and high cost of RNA. On the other hand, the original aptamer sequences screened from SELEX are typically with the length of 70–130 nucleotides (nt), which are too long for their application. Truncation of the original aptamers through post-SELEX process can not only lower the cost in aptamer-based assays, but also improve the stability and affinity of the aptamers. Herein, we present a magnetic beads-based system for fast screening of ssDNA aptamers with high affinity and specificity for TOB from a randomized library by SELEX after ten rounds of screening. During the post-SELEX truncation, we finally obtained an optimal aptamer with the length of 34 nt by using a rationally designed strategy. The obtained aptamer was further applied to construct a gold nanoparticles (AuNPs)-based colorimetric assay for TOB.

DNA Aptamer Specific for Tobramycin

2

3

Materials Prepare all solutions using ultrapure water (18 MΩ cm) obtained from a Millipore water purification system and analytical grade reagents. Store all reagents at 4
C. Diligently follow all waste disposal regulations when disposing waste materials.

2.1 Modification of Epoxy Magnetic Beads

1. Epoxy magnetic beads: Epoxy functional groups-modified and superparamagnetic magnetic particles which are widely used functional biomagnetic beads for immobilization of biological ligands. 2. 100 nM phosphate buffered saline, pH 8.0: Add about 100 mL water in a 1 L glass beaker. Weigh 35.814 g Na2HPO4·12H2O and transfer to the beaker. Make up to 1 L with water to prepare 100 mM Na2HPO4 solution. Add about 100 mL water in another 1 L glass beaker. Weigh 15.601 g NaH2PO4·2H2O and transfer to the beaker. Make up to 1 L with water to prepare 100 mM NaH2PO4 solution. Take another glass beaker and mix the above prepared Na2HPO4 solution and NaH2PO4 solution at the ratio of 94.7:5.3. Then adjust pH to 8.0 with the two solutions. Store at 4
C (see Note 1). 3. 0.5 M ethanolamine: Add about 50 mL water to a 100 mL glass beaker. Weigh 3.05 g C2H7NO and transfer to the beaker. Make up to 100 mL with water. Store at 4
C (see Note 2). 4. 10 mM tobramycin: Add about 50 mL water to a 100 mL glass beaker. Weigh 0.566 g tobramycin and transfer to the beaker. Make up to 100 mL with water. Store at 4
C (see Note 3). 5. Binding buffer, pH 7.6: 20 mM Tris–HCl containing 100 mM NaCl, 1 mM CaCl2, 2 mM MgCl2, 5 mM KCl and 0.02% Tween-20, pH 7.6. Add about 100 mL water to a 1 L glass beaker. Weigh 2.423 g Tris, 5.844 g NaCl, 0.111 g CaCl2, 0.407 g MgCl2·6H2O, 0.373 g KCl and transfer to the beaker. Add 200 μL Tween-20 to the beaker. Make up to 1 L with water (see Note 4). Mix the solution adequately and adjust pH to 7.6 with HCl (see Note 5). Store at 4
C. 6. Elution buffer, pH 8.0: 40 mM Tris–HCl containing 3.5 M urea, 10 mM EDTA and 0.02% Tween-20, pH 8.0. Add about 100 mL water to a 250 mL glass beaker. Weigh 1.211 g Tris, 52.55 g urea, 0.7306 g EDTA and transfer to the glass beaker. Add 50 μL Tween-20 to the beaker. Make up to 250 mL with water. Mix and adjust pH to 8.0 with HCl. Store at 4
C (see Note 6).

2.2

In Vitro Selection

1. A synthetic ssDNA library was used as initial pool, which includes a random sequence of 35 nucleotides flanked by two primers binding sequences for PCR amplification and cloning (50 -TAGGGAATTCGTCGACGGATCC-N35-CTGCAGGTC

4

Nandi Zhou et al.

GACGCATGCGCCG-30 ). A forward primer (50 -TAGGGAA TTCGTCGACGGAT-30 ) or a biotinylated forward primer (50 -biotin-TAGGGAATTCGTCGACGGAT-30 ), and a reverse primer (50 -CGGCGCATGCGTCGACCTG-30 ) or a biotinylated reverse primer (50 -biotin-CGGCGCATGCGTCGA CCTG-30 ) were used for PCR amplification and ssDNA generation. 2. 3 M sodium acetate: Add about 5 mL water in a 10 mL glass beaker. Weigh 2.46 g sodium acetate and transfer to the beaker. Make up to 10 mL with water. Store at 4
C (see Note 7). 3. Binding and washing (B&W) buffer, pH 7.5: 10 mM Tris–HCl containing 2 M NaCl and 1 mM EDTA, pH 7.5. Add about 100 mL water in a 1 L glass beaker. Weigh 1.2114 g Tris, 116.88 g NaCl, 0.29224 g EDTA and transfer to the beaker. Make up to 1 L with water. Mix and adjust pH to 7.5 with HCl. Store at 4
C. 4. PCR reagents: 2.5 mM dNTP, 10 μM forward primer, 10 μM biotinylated reverse primer, 2.5 U μL1 Taq DNA polymerase, 10 PCR buffer, template ssDNA. Make up to 25 μL with double distilled water. 2.3 Molecular Cloning

1. pMD19-T vector plasmid: a vector specifically designed for efficient cloning of PCR products, containing the T-cloning site Eco Rv which has been cleaved.

2.4

1. TAE buffer (50): Add about 100 mL water to a 1 L glass beaker. Weigh 242 g Tris, 37.2 g Na2EDTA 2H2O and transfer to the beaker. Add 800 mL water. Add 57.1 mL acetic acid. Make up to 1 L with water. Store at room temperature (see Note 8).

Agarose Gel

2.5 Medium Preparation

1. LB liquid media: Add about 100 mL water to a 1 L glass beaker. Weigh 10 g peptone, 10 g NaCl, 5 g yeast powder and transfer to the beaker. Add 800 mL water. Stir with magnetic stirrer. Make up to 1 L with water. Sterilize 20 min at 121
C. 2. LB solid media: Add about 100 mL water to a 1 L glass beaker. Weigh 10 g peptone, 10 g NaCl, 5 g yeast powder, 1.5 g agar and transfer to the beaker. Add 800 mL water. Stir with magnetic stirrer. Make up to 1 L with water. Sterilize 20 min at 121
C.

2.6

Detection of TOB

1. 1% chloroauric acid: Weigh 1 g chloroauric acid and transfer to a 100 mL glass beaker. Make up to 100 mL with water. Keep away from light and store at 4
C (see Note 9). 2. 1% trisodium citrate: Weigh 1 g trisodium citrate and transfer to a 100 mL glass beaker. Make up to 100 mL with water. Store at 4
C.

DNA Aptamer Specific for Tobramycin

3

5

Methods

3.1 Modification of Epoxy Magnetic Beads

1. Mix the epoxy magnetic beads. Add 200 μL of 1.0 mg epoxy magnetic beads into centrifuge tube. 2. Place the centrifuge tube on the magnetic separator. Remove the solution with pipette carefully. Add 1 mL PBS into the centrifuge tube and mix it with the epoxy magnetic beads. Then place the centrifuge tube on the magnetic separator and remove the solution again. Repeat the step and rinse the epoxy magnetic beads with PBS for five times (see Note 10). 3. Add 200 μL tobramycin into the centrifuge tube (see Note 11). 4. Incubate the centrifuge tube at 30
C for 16 h with mild shaking in a shaker to modify tobramycin on the magnetic beads (see Note 12). 5. Repeat step 2 to wash the tobramycin-coated beads five times with PBS. Add 200 μL ethanolamine and incubate at 30
C for 8 h with mild shaking in a shaker to completely block the residual active groups on the beads. 6. Wash the beads with PBS for five times. Re-suspend the tobramycin-coated beads in 200 μL of PBS and store at 4
C prior to use (see Note 13). 7. Immobilize streptomycin, neomycin, kanamycin, gentamicin, and tetracycline onto the surface of epoxy magnetic beads respectively as described in steps 1–6 to produce corresponding other antibiotic-coated magnetic beads.

3.2

In Vitro Selection

1. Wash the tobramycin-coated magnetic beads for five times with binding buffer (see Note 14). 2. Add 300 pmol initial ssDNA library and 200 μL binding buffer into a 1.5 mL centrifuge tube. Place the centrifuge tube at 90
C for 10 min, cool on ice for 10 min and stand at room temperature for 5 min. 3. Incubate the ssDNA library with the tobramycin-coated magnetic beads for 1 h with mild shaking at room temperature (see Note 15). 4. Wash the magnetic beads five times with binding buffer (see Note 16). 5. Add 150 μL of elution buffer into the centrifuge tube to elute the ssDNA bound to the magnetic beads. Incubate the tube at 80
C for 20 min with mild shaking. Place the tube on the magnetic separator. Collect the eluent with pipette carefully. Repeat the elution for three times (see Note 17).

6

Nandi Zhou et al.

6. Add 500 μL eluent and 500 μL phenol/chloroform/isoamylol (25:24:1) into a 2.0 mL centrifuge tube. Centrifuge at 10,854  g for 10 min at 4
C. Transfer the upper layer of the liquid into another 2.0 mL centrifuge tube (see Note 18). 7. Add 1.25 mL of 95% cold ethanol and 50 μL sodium acetate into the tube. Freeze it for 2 h at 20
C. 8. Centrifuge at 10,854  g for 10 min at 4
C. Discard the supernatant. 9. Rinse the precipitation with 75% ethanol and centrifuge at 10,854  g for 10 min at 4
C (see Note 19). Discard the supernatant and add 10 μL of sterile water to dissolve the precipitated ssDNA. 10. Detect the concentration of ssDNA by NanoDrop 2000 spectrophotometer. 11. Mix 2 μL dNTP (2.5 mM), 1 μL forward primer (10 μM), 1 μL biotinylated reverse primer (10 μM), 0.1 μL Taq DNA polymerase (2.5 U μL1), 1 μL template ssDNA, 2.5 μL 10  PCR buffer and 17.4 μL double distilled water in a 0.2 mL centrifuge tube. 12. Set the PCR conditions as follows: Pre-denaturation at 95
C for 5 min, denaturation at 95
C for 30 s, annealing at 55
C for 1 min and elongation at 72
C for 1 min. The final elongation step is extended to 5 min at 72
C. 13. The PCR product can be stored at 4
C (see Note 20). 3.3 Agarose Gel Electrophoresis Verification

1. Place the comb fix on the gel sheet in advance (see Note 21). 2. Add 20 mL TAE buffer in a 50 mL conical flask. Weigh 0.6 g agarose and transfer to the conical flask. Heat to boil with microwave oven (see Note 22). 3. Lower the temperature to 60
C. Add 1 μL Goldview staining solution and cast the gel within a 6  6 cm gel sheet (see Note 23). 4. Place 30 min at room temperature (see Note 24). 5. Pull the comb and place the gel sheet in the electrophoresis tank (see Note 25). 6. Add 1 μL 6 Loading buffer into 5 μL samples (see Note 26). 7. Evenly add the samples along the edge of the gel hole (see Note 27). 8. Set the electrophoresis at 120 V for 35 min (see Note 28). 9. Turn off the power switch of the electrophoresis system. 10. Image the gel by a gel imaging system (see Note 29).

DNA Aptamer Specific for Tobramycin

3.4 Preparation of the Secondary Library

7

1. Add 25 μL streptavidin-modified magnetic beads in a 2 mL centrifuge tube (see Note 30). 2. Wash streptavidin-modified magnetic beads with binding and washing (B&W) buffer for three times. Suspend the beads in 80 μL B&W buffer. 3. Add 20 μL the above PCR products into the tube. Incubate at room temperature for 2 h with mild shaking. 4. Place the centrifuge tube on the magnetic separator. Transfer the solution with pipette carefully. Rinse streptavidin-modified magnetic beads with B&W buffer for five times. 5. Add 30 μL of 0.1 M NaOH and incubate in 37
C water bath for 2 h. Place the centrifuge tube on the magnetic separator. Collect the supernatant. 6. Repeat step 5 (see Note 31). 7. Detect the concentration of ssDNA in the supernatant by NanoDrop 2000 spectrophotometer. The recovery of the ssDNA in each round can be calculated by dividing the content of ssDNA in the supernatant by the total amount of ssDNA added to the reaction system. The increase of the recovery levels off when the selection reaches saturation (see Fig. 1).

3.5 Molecular Cloning

1. Cut the gel after electrophoresis to recover the PCR product of the final round by gel recovery kit (see Note 32). 2. Mix 7 μL gel recovery product, 1 μL pMD-19, 1 μL T4 buffer, 1 μL T4 DNA ligase in a 0.2 mL centrifuge tube. Place the centrifuge tube at 16
C for 8–12 h (see Note 33).

Fig. 1 Recovery of bound ssDNA after each SELEX round. (Reprinted by permission from Springer: Han X., Zhang Y., Nie J., Zhao S., Tian Y., Zhou N., Gold nanoparticle based photometric determination of tobramycin by using new specific DNA aptamers, Microchimica Acta, 2018, 185: 4)

8

Nandi Zhou et al.

3. Take 200 μL of E. coli competent cell suspension from 80
C freezer. Thaw at room temperature and place on ice immediately after thawing (see Note 34). 4. Add 10 μL pMD-19 plasmids DNA solution. Mix gently with a pipette and place on ice for 30 min (see Note 35). 5. Heat shock 90 s at 42
C and place on ice immediately for 3–5 min. 6. Add 1 mL LB liquid medium without ampicillin. Mix gently with a pipette. Culture at 37
C for 1 h at the agitation rate of 220 rpm. 7. Centrifuge the medium at 4
C for 5 min at 5427  g. Discard the supernatant and take 200 μL of the bacteria solution on LB solid medium containing ampicillin. After half an hour of positive growth, invert the culture dish and incubate for 10–16 h at 37
C (see Note 36). 8. Pick a single colony and inoculate in LB liquid medium containing ampicillin, culture at 37
C for 10–18 h at the agitation rate of 220 rpm (see Note 37). 9. Centrifuge at 4
C for 5 min at 5427  g. Discard the supernatant and take the bacteria solution. 10. Use plasmid extraction kit to extract plasmid. A suitable number of plasmids can be sent to sequencing. 11. Analyze the homology of the sequencing sequences. Open sequence results, do multiple sequence alignment. Divide the sequences into several families based on their primary sequence homology for further analysis. 12. Use Mfold online software to simulate the secondary structure of these sequences. Open the online software, choose “DNA folding form” and enter the sequence to be folded in the box. Set the folding temperature and ionic conditions according to the reaction system. Click the “folding sequence” and the folding results can be output. 3.6 Determination of Dissociation Constants

1. Wash the tobramycin-coated magnetic beads with binding buffer for five times. 2. Add different amount of FAM-labeled aptamer to 200 μL binding buffer in 0.5 mL centrifuge tubes (see Note 38). The final concentrations of the aptamer are 0, 100, 200, 300, 400, and 500 nM, respectively. Place the centrifuge tubes at 90
C for 10 min, immediately cool on ice for 15 min. 3. Incubate the aptamer with the tobramycin-coated magnetic beads for 2 h with mild shaking at room temperature. 4. Wash the tobramycin-coated magnetic beads for five times with binding buffer.

DNA Aptamer Specific for Tobramycin

9

Fig. 2 The saturation curves for the dissociation constants of aptamers. (Reprinted by permission from Springer: Han X., Zhang Y., Nie J., Zhao S., Tian Y., Zhou N., Gold nanoparticle based photometric determination of tobramycin by using new specific DNA aptamers, Microchimica Acta, 2018, 185: 4)

5. Wash the tobramycin-coated magnetic beads with 200 μL elution buffer at 80
C for 20 min with mild shaking. Place the centrifuge tube on the magnetic separator. Take the solution with pipette carefully. Add the solution in a 96-well plate. 6. Measure the fluorescence intensity by fluorescence spectrophotometer at 515 nm (excitation wavelength 490 nm). 7. Nonlinear fitting by using GraphPad Prism 5.0 to analyze the dissociation constant of each aptamer. The equation of y ¼ Bmax·[free ssDNA]/(Kd + [free ssDNA]) is used, where y represents degree of saturation, Bmax represents the maximum number of binding sites, and [free ssDNA] is the concentration of the unbound ssDNA. 8. Among all the determined aptamer sequences, the one with the lowest dissociation constant has the highest affinity for tobramycin (see Fig. 2). The aptamer with the highest affinity for tobramycin is No. 32 aptamer (Ap 32), with the sequence of 50 -TAGGGAATTCGTCGACGGATCCATGGCACGTTATGC GGAGGCGGTATGATAGCGCTACTGCAGGTCGACGCAT GCGCCG-30 , and the Kd value of 56.88  4.61 nM. 3.7 Specificity Analysis

1. Wash the magnetic beads modified with different kinds of antibiotics with binding buffer for five times before use. 2. Add 200 pmol aptamer in 200 μL binding buffer. Denature at 90
C for 10 min, immediately cool on ice for 15 min. 3. Mix antibiotics-modified magnetic beads with the aptamer for 2 h with mild shaking at room temperature. 4. Wash the antibiotics-modified magnetic beads for five times with binding buffer.

10

Nandi Zhou et al.

5. Elute the bound aptamer with 200 μL elution buffer at 80
C for 20 min with mild shaking. 6. Place the centrifuge tube on the magnetic separator. Take the solution with pipette carefully. Repeat step 5 for three times to elute all bound ssDNA. 7. Determine the concentration of ssDNA by NanoDrop 2000 spectrophotometer. 8. Calculate the recovery and use it to analyze the binding specificity of the aptamers (see Note 39). 3.8 Molecular Docking

1. Draw the structure of tobramycin with Chemdraw and convert it into 3D format with Chem 3D. 2. Treat the structure of tobramycin with Autodock 4.0 software by hydrogenation, calculation of charge, and incorporation of non-polar hydrogen. 3. Obtain the 3D conformation of the aptamer by homology modeling. 4. Treat the conformation of the aptamer with Autodock 4.0 software by hydrogenation, calculation of charge, and incorporation of non-polar hydrogen. 5. Construct a 100 A˚  100 A˚  100 A˚ docking box around tobramycin and the aptamer and center on the active sites respectively. Generate active sites by Autogrid operation. 6. By using the default parameters of the software, adopt the structure of tobramycin a flexible method, and run Autodock program based on Lamarckian Genetic Algorithm for 60 times to finish the molecular docking.

3.9 Aptamer Truncation

Basically, the truncation of the aptamer is based on the structural analysis. The stem or loop modules irrelevant with the binding sites are considered to be removed to shorten the aptamer sequence (see Note 40). 1. Based on the screened Ap 32, some modules in the secondary structure irrelevant with the binding sites are removed in turn (see Fig. 3). Synthesize the truncated sequences (Ap 32-1 and Ap 32-2). 2. Determinate the dissociation constants of the truncated aptamer sequence (see Subheading 3.6) and compare it with that of the original sequence (see Note 41). Finally, a 34 nt aptamer (Ap 32-2) 50 -CGTCGACGGATCCATGGCACGTT ATAGGTCGACG-30 with the Kd value of 48.40  6.63 nM was obtained, which can be used in the detection of tobramycin.

DNA Aptamer Specific for Tobramycin

11

Fig. 3 The secondary structure of the aptamer with the highest affinity from SELEX and its truncation design: (a) the original aptamer, (b) the truncated 49 nt aptamer, (c) the truncated 34 nt aptamer. The red boxes indicate the removed modules

3.10 Preparation of Gold Nanoparticles (AuNPs)

1. Immerse the glassware used in the experiment in newly prepared aqua regia for 30 min. Then soak in ultrapure water overnight (see Note 42). 2. Add 1 mL of 1% chloroauric acid in a 100 mL glass beaker (see Note 43). Make up to 100 mL with water. Formulate as 0.01% chloroauric acid solution. 3. Add 100 mL of 0.01% chloroauric acid to a round bottom flask (see Note 44). 4. Continuously stir the solution at 300 rpm and heat until the solution boils (see Note 45). 5. Add 3.5 mL of 1% trisodium citrate solution quickly (see Note 46). 6. Continuously stir the solution at 300 rpm and heat for 15 min (see Note 47). 7. Stop heating and continuously stir at 300 rpm for another 30 min. 8. Continuously stir at 300 rpm and heat until the volume of the remaining solution is 20 mL (see Note 48). 9. Cool the solution to room temperature. 10. Load it into a brown bottle and wrap in tin foil, store at 4
C.

12

Nandi Zhou et al.

3.11 Optimization of the Conditions for Colorimetric Detection of Tobramycin 3.11.1 Optimization of NaCl Concentration.

1. Add 50 μL AuNPs solution in a series of 0.5 mL centrifuge tubes respectively (see Note 49). 2. Add 100 μL of 150 nM aptamer to each tube (see Note 50). 3. Keep the tubes in the dark at room temperature for 1 h. 4. Add 50 μL of 800 nM tobramycin in each tube. Incubate in the dark at room temperature for 50 min. 5. Add 50 μL of 80, 100, 120, 140, and 160 mM NaCl to the respective tubes. 6. Observe the color change in the centrifuge tubes. When the concentration of the salt is higher than 120 mM, the color of AuNPs solution no longer changes. Therefore, 120 mM is the optimal concentration of NaCl in the colorimetric detection.

3.11.2 Optimization of Aptamer Concentration

1. Add 50 μL AuNPs solution in a series of 0.5 mL centrifuge tubes respectively. 2. Add 100 μL of 0, 50, 100, 150, and 200 nM aptamer to the respective tubes (see Note 50). 3. Keep the tubes in the dark at room temperature for 1 h. 4. Add 50 μL of double distilled H2O to each tube. Keep the tubes in the dark at room temperature for 50 min. 5. Add 50 μL of 120 mM NaCl. 6. Observe the color change in the centrifuge tubes. When the concentration of the aptamer is higher than 150 nM, the color of AuNPs solution no longer changes. Therefore, 150 nM is the optimal concentration of the aptamer in the colorimetric detection.

3.12 Detection of Tobramycin Using AuNP-Based Colorimetric Assay

1. Add 50 μL AuNPs solution in a series of 0.5 mL centrifuge tubes respectively. 2. Add 100 μL of 150 nM aptamer to each tube. 3. Keep the tubes in the dark at room temperature for 1 h. 4. Add 50 μL of 0, 50, 100, 200, 400, 600, 800, 1000, 1200, and 1400 nM tobramycin to the respective tubes. 5. Keep the tubes in the dark at room temperature for 50 min. 6. Add 50 μL of 120 mM NaCl to each tube. 7. Observe the color change in the centrifuge tubes. 8. Record the UV-visible spectra of the solution in each tube (see Fig. 4). 9. Plot the absorbance at 520 nm to the concentration of tobramycin to obtain the calibration curve, which is y ¼ 2.599e  4x + 1.02186, R2 ¼ 0.99047, where y represents the absorbance at 520 nm, x represents the concentration of tobramycin (nM) (see Fig. 5). The linear detection range is

DNA Aptamer Specific for Tobramycin

13

Fig. 4 UV-Vis spectra of AuNPs in the presence of different concentrations of tobramycin. (Reprinted by permission from Springer: Han X., Zhang Y., Nie J., Zhao S., Tian Y., Zhou N., Gold nanoparticle based photometric determination of tobramycin by using new specific DNA aptamers, Microchimica Acta, 2018, 185: 4)

Fig. 5 The linear relationship between the absorbance of AuNPs at 520 nm and the concentration of tobramycin. (Reprinted by permission from Springer: Han X., Zhang Y., Nie J., Zhao S., Tian Y., Zhou N., Gold nanoparticle based photometric determination of tobramycin by using new specific DNA aptamers, Microchimica Acta, 2018, 185: 4)

within 100–1400 nM, and the detection limit is 37.90 nM (S/N ¼ 3). 10. To detect kanamycin in real samples, repeat steps 1–8. Add 50 μL real sample in step 4. Record the absorbance at 520 nm and calculate the concentration of tobramycin using the calibration curve.

14

4

Nandi Zhou et al.

Notes 1. Buffer must be filtered with 0.22 μm membrane before storage. 2. Acrylamide solution can be stored at 4
C for 1 month. 3. Tobramycin solution is best dispensed into 1.5 mL centrifuge tubes to avoid contamination. 4. Having water at the bottom of the glass beaker helps to dissolve Tris, allowing the magnetic stir bar to work immediately. Tris can be dissolved faster if the water is heated to about 37
C. However, the downside is that care should be taken to bring the solution to room temperature before adjusting pH. 5. Concentrated HCl can be used at first to narrow the gap from the starting pH to the required pH. From then on it would be better to use a series of HCl with lower ionic strengths to avoid a sudden drop in pH below the required pH. 6. If the refrigerator is not clean enough, the microorganisms in the refrigerator will grow in the buffer, resulting in changes in the buffer composition. Need to keep the refrigerator clean. 7. Sodium acetate solution is used during multiple rounds of screening, so sodium acetate solution is dispensed into 1.5 mL centrifuge tubes to avoid contamination. 8. TAE buffer is diluted 50 times before use. 9. Chloroauric acid solution can be stored in the refrigerator for a few months. 10. The magnetic beads need to be fully mixed to ensure that the magnetic beads to be taken each time are equal. 11. Use a pipette to blow the beads to suspension so that tobramycin is evenly distributed around the beads. 12. Place the beads in a shaker and incubate them. The rotation speed of the shaker is adjusted in a way that the beads are in the suspension. The general rotation speed is 200 rpm. 13. Do not invert or tilt the centrifuge tube holding the magnetic beads. Keep the beads in liquid environment. 14. When washing the magnetic beads, the supernatant should be carefully removed to avoid sucking the magnetic beads to reduce the loss of magnetic beads and improve the screening efficiency. 15. Place the centrifuge tube in a shaker and incubate them. The rotation speed of the shaker is adjusted so that the beads are in suspension to increase the binding of ssDNA with tobramycin. 16. Rinsing the magnetic beads with binding buffer will help remove a bulk of the nonspecific ssDNA.

DNA Aptamer Specific for Tobramycin

15

17. Mix the eluent from the three repeated elution step. If no ssDNA is eluted, consider increasing the elution time. 18. Organic solvents are highly toxic and should be used with special care. Wear double gloves and masks when taking organic solvents. After use, the gloves, masks, and tips must be properly handled and cannot be mixed with ordinary waste. 19. Can be washed once more with 75% alcohol to remove residual phenol. 20. The PCR product cannot be placed more than 24 h, otherwise the electrophoresis verification will appear false negative. 21. Check the electrophoresis tank and replace the buffer according to the situation, ensure buffer capacity, reduce pollution. 22. When heated to produce a large amount of bubbles in the gel, shake it out, continue heating until it is completely dissolved, shake it out, and heat it until it boils. Be careful of hot hands and be careful not to overheat the gel out of the bottle. Therefore, be careful to choose a bottle that is at least two times of the gel volume. Ensure that the gel is mixed and completely dissolved; reducing the possibility that the resulting non-homogeneous pore size affects the separation effect. 23. Slowly pour in one side along the side of the gel sheet. Note that the number of comb holes can be all the points below. Use the pipette tip to remove the air bubbles. The tabletop is relatively horizontal. When pouring glue, minimize the generation of bubbles. 24. Do not touch the comb during the process; try to keep the gel position from moving. The time should not be too long, leading to deformation of the gel; it should not be too short and affect the formation of pores inside the gel. 25. Gradually pull the comb vertically out of the comb hole so that the comb is pulled out of each hole as much as possible. Temporarily unused gel is best placed in TAE buffer for immersion. TAE should be submerged in 1 mm. 26. Pay attention to mixing. The concentration of loading buffer should not be too low and the sample can’t sink well in the gel hole when spotted; it should not be too high, and it is easy to form band-shaped deformation during electrophoresis. 27. Try to avoid damaging the rubber hole. Do not breathe too much gas on the tip of the pipette, and do not overspill the sample when pulled up. One tip at a time for each change. The quicker the better, the better the quality. The amount of spotting should not be too large. On the one hand, it will spill and contaminate the adjacent sample; on the other hand, it will easily lead to tailing and blurring.

16

Nandi Zhou et al.

28. The glue hole and the electrode are leveled to prevent the sample from running. Pay attention to the observation during the run to prevent the accidental occurrence of the sample running out of the glue. 29. At the early SELEX rounds, due to the large capacity of different sequences in the initial library, the oligonucleotides with weak affinity also bind to the tobramycin-conjugated magnetic beads. The electrophoresis bands of PCR products are scattered and blurred. With the increased round of screening, ssDNAs with high affinity for tobramycin were gradually enriched, whereas those with low affinity were gradually eliminated. Thus, the electrophoresis bands of PCR products were concentrated and bright. 30. Several centrifuge tubes may be needed to meet the need of next round screening. 31. The ssDNA solution obtained at this time can be used as the library for the next round of screening. 32. The recovered ssDNA from the last round of SELEX is amplified by PCR with non-biotinylated primers. 33. Connection time should not be too short. 34. The competent cell cannot be used repeatedly, otherwise it will lead to a decrease in conversion rate. 35. Competent cell is very fragile, gently pipette and mix. 36. In addition to some special temperature requirements, other operations are paid attention to operating on ice, centrifuges, etc., to be pre-cooled. 37. Take care to avoid contamination between colonies when picking single colonies. Pick a single colony under aseptic conditions. 38. The FAM-labeled aptamer was synthesized and should be stored in the dark at 20
C. During the experiment, exposure should also be avoided as much as possible. Use a foil to wrap around the tube to avoid light. 39. Compare the recovery of the aptamers combined with tobramycin to those of with other antibiotics. If the recovery of the aptamers bound to tobramycin is significantly higher than those to other antibiotics, the specificity of the aptamers is satisfactory. 40. When the aptamer interacts with tobramycin, the binding sites can be predicted by molecular docking. 41. The smaller the Kd value, the higher the affinity of the sequence to tobramycin. If the Kd value of the truncated sequence reduces or keeps unchanged compared to the original

DNA Aptamer Specific for Tobramycin

17

aptamer, the truncation is valid. Otherwise the truncation is invalid. 42. The aqua regia is very corrosive. During the process of preparation and use, protective measures should be taken with double gloves and masks. Since hydrogen chloride is volatile, aqua regia needs to be ready for use. 43. It is found that 1% of chloroauric acid does not change significantly at 4
C in half a year. 44. Round bottom flask is marked with 20 mL before use. 45. The round bottom flask is placed in a pot and heated in a water bath. During the heating process, the round bottom flask is sealed with cling film to prevent water evaporation. 46. When add 3.5 mL trisodium citrate, it should be quickly added into round bottom flask. Take 3.5 mL of trisodium citrate directly with a 5 mL pipette and add it to the flask. 47. The water in the pot needs to be heated to boiling. 48. The water in the pot does not need to be heated to boiling. 49. AuNPs need to be protected from light during use. 50. Aptamers need to be diluted with double distilled water to a specific concentration.

Acknowledgments This work was supported by the National Natural Science Foundation of China (no. 31271860). References 1. Yuan BY, Jiang XC, Chen YY, Guo QP, Wang KM, Meng XX et al (2017) Metastatic cancer cell and tissue-specific fluorescence imaging using a new DNA aptamer developed by cellSELEX. Talanta 170:56–62 2. Bing T, Zhao HM, Lei D, Gan XR, Xie Q (2016) A versatile fluorescent biosensor based on target-responsive graphene oxide hydrogel for antibiotic detection. Biosens Bioelectron 83:267–273 3. Kong RM, Zhang XB, Zhang L, Jin XY, Huan SY, Shen GL et al (2009) An ultrasensitive electrochemical "turn-on" label-free biosensor for Hg2+ with AuNP-functionalized reporter DNA as a signal amplifier. Chem Commun (Camb) 37:5633–5635 4. Kuang H, Chen W, Xu DH, Xu LG, Zhu YY, Liu LQ et al (2011) Fabricated aptamer-based

electrochemical "signal-off" sensor of ochratoxin a. Biosens Bioelectron 26:710–716 5. Wilson DS, Szostak JW (2003) In vitro selection of functional nucleic acids. Annu Rev Biochem 68:611–647 6. Ho H-A, Leclerc M (2004) Optical sensors based on hybrid aptamer/conjugated polymer complexes. J Am Chem Soc 126:1384–1387 7. Tuerk C, Gold L (1990) Systematic evolution of ligands by exponential enrichment: RNA ligands to bacteriophage T4 DNA polymerase. Science 249:505–510 8. Huang CJ, Lin HI, Shiesh SC, Lee GB (2010) Integrated microfluidic system for rapid screening of CRP aptamers utilizing systematic evolution of ligands by exponential enrichment (SELEX). Biosens Bioelectron 25:1761–1766 9. Darmostuk M, Rimpelova S, Gbelcova H, Ruml T (2015) Current approaches in

18

Nandi Zhou et al.

SELEX: an update to aptamer selection technology. Biotechnol Adv 33:1141–1161 10. Cho EJ, Lee JW, Ellington AD (2009) Applications of aptamers as sensors. Annu Rev Anal Chem 2:241–264 11. Xing H, Wong NY, Xiang Y, Lu Y (2012) DNA aptamer functionalized nanomaterials for intracellular analysis, cancer cell imaging and drug delivery. Curr Opin Chem Biol 16:429–435 12. Iliuk AB, Hu L, Tao WA (2011) Aptamer in bioanalytical applications. Anal Chem 83:4440–4452 13. Wochner A, Menger M, Orgel D, Cech B, Rimmele M, Erdmann VA et al (2008) A DNA aptamer with high affinity and specificity for therapeutic anthracyclines. Anal Biochem 373:34–42 14. Manyanga V, Elkady E, Hoogmartens J, Adams E (2013) Improved reversed phase liquid chromatographic method with pulsed electrochemical detection for tobramycin in bulk and pharmaceutical formulation. J Pharm Anal 3:161–167

15. Shanson DC, Hince CJ, Daniels JV (1976) Rapid microbiologic assay of tobramycin. J Infect Dis 134:S104–S109 16. Darwish IA (2003) Development of generic continuous-flow enzyme immunoassay system for analysis of aminoglycosides in serum. J Pharm Biomed Anal 30:1539–1548 17. Feng CH, Lin SJ, Wu HL, Chen SH (2002) Trace analysis of tobramycin in human plasma by derivatization and high-performance liquid chromatography with ultraviolet detection. J Chromatogr B 780:349–354 18. Fonge H, Kaale E, Govaerts C, Desmet K, Van Schepdael A, Hoogmartens J (2004) Bioanalysis of tobramycin for therapeutic drug monitoring by solid-phase extraction and capillary zone electrophoresis. J Chromatogr B 810:313–318 19. Song KM, Cho M, Jo H, Min K, Jeon SH, Kim T et al (2011) Gold nanoparticle-based colorimetric detection of kanamycin using a DNA aptamer. Anal Biochem 415:175–181 20. Jiang L, Patel DJ (1998) Solution structure of the tobramycin-RNA aptamer complex. Nat Struct Biol 5:769–774

Chapter 2 Affitins: Ribosome Display for Selection of Aho7c-Based Affinity Proteins Valentina Kalichuk, Stanimir Kambarev, Ghislaine Be´har, Benjamin Chalopin, Axelle Renodon-Cornie`re, Barbara Mouratou, and Fre´de´ric Pecorari Abstract Engineered protein scaffolds have made a tremendous contribution to the panel of affinity tools owing to their favorable biophysical properties that make them useful for many applications. In 2007, our group paved the way for using archaeal Sul7d proteins for the design of artificial affinity ligands, so-called Affitins. For many years, Sac7d and Sso7d have been used as molecular basis to obtain binders for various targets. Recently, we characterized their old gifted protein family and identified Aho7c, originating from Acidianus hospitalis, as the shortest member (60 amino-acids) with impressive stability (96.5
C, pH 0–12). Here, we describe the construction of Aho7c combinatorial libraries and their use for selection of binders by ribosome display. Key words Ribosome display, In vitro selection, Sul7d, Aho7c, Sac7d, Sso7d, Affitin

1

Introduction Monoclonal antibodies are still the most widely used class of affinity proteins. Though they have many attractive properties, they also have their limitations. Depending on the application, some of their features can be impractical. A protocol that requires high temperature or extreme pH, radiolabeling or cleaning-in-place procedures for affinity chromatography columns for instance, will reduce drastically the chance of obtaining functional antibodies at the end. Their large size, complex molecular structure with four polypeptide chains and disulfide bridges, and their high production costs are also criteria to consider before starting a project involving a standard antibody. On the other hand, engineered protein scaffolds have been designed to possess as many as possible favorable properties: stability, structural simplicity with one polypeptide chain and no disulfide

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2070, https://doi.org/10.1007/978-1-4939-9853-1_2, © Springer Science+Business Media, LLC, part of Springer Nature 2020

19

20

Valentina Kalichuk et al.

Fig. 1 Model of wild type Aho7c, based on its primary sequence. The model has been created by I-TASSER server [23, 24] and drawn using PyMOL software (www.pymol.org). Randomized residues in library L5 are represented as green sticks

bridges, small size, ease of engineering, and cost-effective production. In fact, engineered protein scaffolds have been demonstrated to possess the main advantages of monoclonal antibodies, high affinity and specificity, without being burdened with their inconvenient features. We have previously described the application of archaeal proteins to design artificial binders with a certain specificity by using the Sac7d protein (Fig. 1) from Sulfolobus acidocaldarius [1, 2]. By coupling combinatorial libraries of Sac7d and ribosome display selection against a chosen target, we have generated the so-called Affitins. Sac7d-based Affitins were shown to generally inherit the remarkable properties of their parent protein: they are highly thermostable (up to 90
C), pH-stable (from 0 up to 13), highly soluble, and well-produced in Escherichia coli grown in flasks (up to 200 mg/L culture) [1]. The first crystallographic structure of an Affitin showed that the native fold is not altered, although 21% of the sequence was randomized [3]. Their interaction with targets can be designed using either a flat surface of interaction or a flat surface and a loop. The crystal structures of two glycosidase/ Affitin complexes validated our library design for the two modes of binding [4]. We have repeatedly generated numerous Sac7d-based Affitins which show specificity and affinity in nanomolar and subnanomolar ranges [1–10]. We also recently demonstrated that a binding site from a Sac7d-based Affitin can be transferred onto a Sso7d scaffold [9] and that a CDR3 from an antibody can be grafted into Sac7d to confer affinity for the cognate ligand [8]. Sac7d-based Affitins can be used for detection-, capture-, and inhibition-based applications, such as one-step ELISA and Western blots [1], immunolocalization [6], biosensors [7], protein-chip

Selection of Aho7c-Based Affitins

21

array [11], affinity chromatography [12], magnetic fishing [13], enzymatic [4] and intracellular inhibition [1], and imaging [14]. Recently, in a study assessing whether another member of the Sul7d family could show more interesting properties than Sac7d and Sso7d, we identified Aho7c, originating from Acidianus hospitalis [15]. This protein is the shortest known member of this family (60 amino acids compared to 66 for Sac7d and 64 for Sso7d) and it was shown to possess thermal stability higher than the one of Sac7d (89.6
C) and comparable with Sso7d (96.8
C versus 96.5
C). As a showcase, we generated Aho7c-based Affitins specific for the extracellular domain of epithelial cell adhesion molecule (EpCAM) with picomolar affinities (KD ¼ 110 pM) [10]. This was the highest affinity ever obtained for a Sul7d-based binder validating Aho7c as a molecular basis for the generation of Affitins. So far, all Affitins from our group have been created using one of the most powerful selection techniques: ribosome display [16]. Other groups working with Sso7d use also yeast display [17] or phage display [18]. However, ribosome display has the advantage of being performed entirely in vitro, which avoids the limiting step of having cells transformed with the DNA of the libraries, thus allowing access to diversity of at least 1012 sequence variants. In the ribosome display system, the link between the phenotype and the genotype is ensured by the ribosome [19]. In the following sections, we describe the generation of an Aho7c library, denoted L5 [10], in a format adapted to ribosome display and its use for selections against a target of interest.

2 2.1

Materials Strains

1. DH5α cells (Invitrogen or another supplier). 2. DH5αF0 IQ cells (Invitrogen).

2.2

Oligonucleotides

2.2.1 Primers Used for Generation of the 50 -Flanking Region of the Ribosome Display Construct and the Randomized Positions of the Gene Encoding Aho7c

AF-Lib5.1: 50 -CTACAAAGATGACGATGACAAAGGATCCGC GACCAAAGTAAAATTC-30 . AF-Lib5.2: 50 -TCCGCGACCAAAGTAAAATTCAAANHKNHK GGTGAGGAAAAAGAGGTGGATATTAGCAAGATC-30 . AF-Lib5.3: 50 -CTTGCCGTTGTCGTCGTASNNAAASNNGAT CATTTTGCCGACACGSNNCACSNNSNNGATCTTGCT AATATCCACCTC-30 . AF-Lib5.4: 50 -GCAGTTCTTTCGGGGCGTCTTTTTCGGAAA CSNNACCSNNGCCSNNCTTGCCGTTGTCGTCGTA-30 . AF-Lib5.5: 50 -GAATTCGGCCCCCGAGGCCATATAAAGCTT CAGTTTCTCCAGCAGTTCTTTCGGGGCGTC-30 . T7C: 50 -ATACGAAATTAATACGACTCACTATAGGGAGACCA CAACGGTTTCCCTC-30 .

22

Valentina Kalichuk et al.

SDA_FLAG: 50 -AGACCACAACGGTTTCCCTCTAGAAATAA TTTTGTTTAACTTTAAGAAGGAGATATATCTATGGACT ACAAAGATGACGATGACAAA-30 . SDA_RGS: 50 -AGACCACAACGGTTTCCCTCTAGAAATAA TTTTGTTTAACTTTAAGAAGGAGATATATCCATGAGAGG ATCG-30 . 2.2.2 Primers for the Amplification of tolA Linker Encoded by pFP-RDV1 2.2.3 Primers for the Final Assembly of the 50 -Construct and tolA Linker 2.2.4 Primers for RT-PCR and PCR After Selection

AF-link-F: 50 -AAGCTTTATATGGCCTCGGGGGCCGAATTC-30 . TolAkurz: 50 -CCGCACACCAGTAAGGTGTGCGGTTTCAGTT GCCGCTTTCTTTCT-30 . T7B: 50 -ATACGAAATTAATACGACTCACTATAGGGAGACCA CAACGG-30 . TolAkurz: 50 -CCGCACACCAGTAAGGTGTGCGGTTTCAG TTGCCGCTTTCTTTCT-30 . RDV2-RT: 50 -GATGACGATGACAAAGGATCC-30 . AF-link-R: 50 -GAATTCGGCCCCCGAGGCCATATAAAGC-30 . TolAext: 50 -CCGCACACCAGTAAGGTGTGCGGTTTCAGTTG CCGCTTTCTTTCTTGCTTCAGCTGCAGCTGCTTC-30 . RDV2-Famp: 50 -TGAGTAGAAGCTTTATATGGCC-30 . RDV2-Ramp: 50 - TCATTTGGATCCTTTGTCATCGTCATC-30 .

2.2.5 Primers for Construction of pFP RDV1 and pFP RDV2

RDV1-P18-F: 50 -TGCGGTAAACGCGTGCCTGGGGTGCCT AATGAGTG-30 . RDV1-P18-R: 50 -ATTTCGTATCCATGGTTAAGCCAGCCCCG ACACC-30 . K7-RDV2-F: 50 -P-CTAGAAATAATTTTGTTTAACTTTAAG AAGGAGATATATCTATGGACTACAAAGATGACGATGAC AAAG-30 . K7-RDV2-R: 50 -P-GATCCTTTGTCATCGTCATCTTTGTAGT CCATAGATATATCTCCTTCTTAAAGTTAAACAAAATTA TTT-30 .

2.2.6 Primers for Sequencing

2.3

PCR

Qe30for: 50 -CTTTCGTCTTCACCTCGAG-30 . Qe30rev: 50 -GTTCTGAGGTCATTACTGG-30 . 1. UHP water (various suppliers). 2. Phusion Hot Start II DNA Polymerase (Thermo Fisher Scientific). 3. dNTP solution: 10 mM of each dNTP.

Selection of Aho7c-Based Affitins

23

4. Exonuclease I (Thermo Fisher Scientific). 5. 1 kb Plus DNA ladder. 6. 6 Orange DNA loading dye or equivalent. 7. Agarose. 8. GelGreen nucleic acid gel stain (Biotium). 9. 1 TAE: 40 mM Tris base, 20 mM acetic acid, 1 mM EDTA. 10. Wizard SV Gel and PCR Clean-Up System (Promega). 2.4

Ligation

1. T4 DNA ligase (New England Biolabs). 2. BamHI and HindIII restriction enzymes (Thermo Fisher Scientific). 3. FastAP thermosensitive alkaline phosphatase (Thermo Fisher Scientific). 4. pFP RDV1 or pFP RDV2.

2.5 In Vitro Transcription

1. TranscriptAid T7 High Yield Transcription kit (Thermo Fisher Scientific). 2. DNase I, RNase free (Thermo Fisher Scientific). 3. 6 M LiCl. 4. 70% (v/v) Ethanol. 5. 100% (v/v) Ethanol. 6. 3 M sodium acetate. 7. High-Pure RNA purification kit (Macherey-Nagel).

2.6 In Vitro Translation

1. In vitro translation system (various suppliers or lab-made).

2.7

1. TBS: 20 mM Tris–HCl pH 7.4, 150 mM NaCl.

Selection

2. 200 mM L-methionine (Sigma). Aliquot and store at 20
C.

2. 20 μM solution of streptavidin or NeutrAvidin (Thermo Fisher Scientific) prepared in 1 TBS. Aliquot and store at 20
C. 3. Tween-20 (Calbiochem). 4. BSA (Merck). 5. Washing Buffer (WB): 50 mM Tris–acetate (pH 7.4), 150 mM NaCl, 50 mM magnesium–acetate. 6. WBT: WB containing 0.1% Tween-20. 7. WBT-BSA: WBT containing 0.5% BSA. 8. WBTH-BSA: WBT-BSA containing 20 mg/mL Heparin. 9. Elution Buffer: 50 mM Tris–acetate (pH 7.4), 150 mM NaCl, 20 mM EDTA.

24

Valentina Kalichuk et al.

10. 25 μg/μL solution of S. cerevisiae RNA (Merck) prepared in UHP water. Aliquot and store at 20
C. 11. RNeasy MinElute Cleanup Kit (Qiagen). 12. Reverse transcriptase (Thermo Fisher Scientific or another supplier). 13. RiboLock RNase Inhibitor (Thermo Fisher Scientific). 14. MaxiSorp plates (Nunc). 15. Streptavidin or NeutrAvidin coated magnetic beads, 300 nm (Ademtech or another supplier). 2.8

ELISA

1. Anti-FLAG-HRP conjugate (Merck). 2. SigmaFAST OPD (o-phenylenediamine dihydrochloride) tablets (Merck). 3. TBS-Tween: 1 TBS containing 0.1% Tween-20. 4. Bugbuster 10 Protein Extraction Reagent (Thermo Fisher Scientific).

2.9 Screening of Clones and Sequence Analysis

1. Luria-Bertani (LB) agar Petri plates containing bacteriological agar (1.5%), ampicillin (100 μg/mL) and kanamycin (25 μg/ mL). 2YT growth medium containing ampicillin (100 μg/ mL), kanamycin (25 μg/mL) and 1% glucose. 2. Lysis buffer: TBS containing 1 Bugbuster solution and 5 μg/ mL DNase I. 3. IPTG: 1 M isopropyl β-D-1-thiogalactopyranoside solution in water.

3

Methods The Aho7c library L5 is obtained via randomization of ten amino acid residues (Y9, K10, K22, K23, W25, S32, T34, T41, R43, A45). These positions were chosen among those expected to participate in the interaction between DNA and Aho7c (see Fig. 1 and Note 1). Based on sequencing of numerous Affitins obtained after selection, we picked the positions, validated as essential for the interaction of Affitins with their targets [4]. As the gene encoding Aho7c is quite short (174 bp), it was randomized at the mentioned positions by PCR using a combination of four non-degenerated sequences and three degenerated oligonucleotides that include NNS or NHK triplets. These triplets encode for 20 and 16 different amino acids, respectively. The NHK triplets were used to exclude tryptophan at positions 9 and 10 to avoid oligomerization (unpublished observation). A final PCR allows the complete assembly of L5 into the ribosome display format with flanking regions, including a T7 promoter, a ribosome binding site, a tolA linker and stem

Selection of Aho7c-Based Affitins

25

loops at the ends to stabilize the construct, necessary for selection by ribosome display (see Fig. 2 and Notes 2 and 3). In order to generate a reliable source of tolA spacer, we have designed a plasmid derived from pFP1001 which encodes the part of the tolA gene that is needed for the ribosome display construct [20]. For that purpose, the tolA gene was amplified from E. coli DH5α cells and ligated into pFP1001. In vitro transcription of the final PCR product yields mRNA used for in vitro translation. The absence of stop codon in the mRNA transcript has been shown to cause the so-called ribosome stalling [21]. Hence, in the ribosome display setting, this phenomenon stabilizes the ternary complex nascent polypeptide-ribosomemRNA and creates a stable phenotype–genotype link. These ternary complexes are then used for selections. The number of selection rounds depends on the desired characteristics for binders. However, usually three to five rounds are necessary. 3.1 Production of Input Library 3.1.1 Production of the Aho7c Library Fragments

1. The DNA product containing the modified aho7c gene is obtained by PCR using a combination of four standard (T7C, SDA_FLAG, AF-Lib5.1, AF-Lib5.5) and three degenerated oligonucleotides encoding NNS and NHK triplets (AF-Lib5.2, AF-Lib5.3, AF-Lib5.4) (see Notes 4 and 5). For this, prepare a series of 50 μL PCR mixtures in PCR tubes containing 2 pmol of each internal primer (0.2 μL of 10 μM primer SDA_FLAG, AF-Lib5.1, AF-Lib5.2, AF-Lib5.3, AF-Lib5.4), 10 pmol of each external primer (1 μL of 10 μM primer T7C and AF-Lib5.5), 1 μL of dNTPs mix (containing 10 mM of each dNTP), 10 μL of 5 Phusion-HF buffer, and 1.5 U of Phusion Hot Start II DNA polymerase. 2. Use a thermocycler to perform the following PCR program: an initial denaturation step at 98
C for 30 s, followed by 5 cycles of 98
C for 30 s, 44
C for 30 s, 72
C for 22 s, followed by 30 cycles of 98
C for 30 s, 64
C for 30 s, 72
C for 22 s with a final elongation step of 72
C for 5 min. 3. Prepare a 1.5% agarose gel. Mix 5 μL of the PCR reaction with 1 μL of 6 Orange DNA loading dye buffer and run on the gel at 110 V for at least 40 min. On an adjacent lane, run 5 μL of 1 kb Plus DNA Ladder. 4. Image the gel and if the correct product corresponding to the expected size of 326 bp is observed, load the rest of the PCR mixture on a 1% agarose gel and excise the band (see Note 6). 5. Purify the DNA product using Promega Wizard SV PCR and Gel purification kit according to the manufacturer’s specifications. Elute DNA with 30 μL UHP water. 6. Load 5 μL of the library fragment on a 1.5% agarose gel for quantification. About 1 μg of purified DNA should be obtained if four reactions are performed (see Note 7).

26

Valentina Kalichuk et al.

Fig. 2 Schemes of the ribosome display construct used for selection. (a) Vector pFP RDV2. It contains a β-lactamase gene (ampicillin resistance) as a selection marker (1727–2584) and a 484-bp region (152–633)

Selection of Aho7c-Based Affitins

27

3.1.2 Plasmid Construction

The plasmid pFP RDV1 encodes the ribosome display construct in which displayed proteins are tagged with RGS-His6 on their N-terminus (Fig. 2b). In pFP RDV2, this tag is replaced with FLAG-tag to use an alternative detection system.

pFP RDV1

pUC18 (ATCC 37253) is amplified with primers (RDV1-P18-F and RDV1-P18-R: amplicon size 2317 bp) followed by digestion with NcoI and MluI. Gel-purified product is then ligated into a similarly digested synthetic DNA product containing the complete ribosome display construct (see Note 8). For this: 1. Prepare a 50 μL PCR reaction mix in PCR tubes containing 2.5 ng of pUC18, 25 pmol of each primer (2.5 μL of 10 μM primer RDV1-P18-F and RDV1-P18-R), 1 μL of dNTPs mix (containing 10 mM of each dNTP), 10 μL of 5 Phusion-HF buffer, and 1.5 U of Phusion Hot Start II DNA polymerase. 2. Place the tubes on a thermocycler and preheat the reaction to 98
C for 5 min. Perform 25 cycles of PCR of 98
C for 30 s, 68
C for 30 s, 72
C for 70 s, followed by 1 cycle of 72
C for 5 min. 3. Mix 1 μL of 6 Orange DNA loading dye with 5 μL of the PCR product and run on a 1% agarose gel to be sure that the PCR reaction gives a band corresponding to the expected size of 2317 bp. 4. Purify the PCR product using the Promega Wizard SV PCR Clean-up kit. 5. Digest the purified DNA with NcoI and MluI enzymes for 2 h at 37
C. 6. Mix the digested product with the appropriate volume of 6 Orange DNA loading dye and gel-purify on a 1% agarose gel. 7. Ligate the purified DNA product with the synthetic DNA which corresponds to the sequence necessary for ribosome display including tolA (see Note 8) predigested with NcoI

ä Fig. 2 (continued) possessing all functional regions required for successful cloning, in vitro transcription, translation and selection of a potential DNA sequence. A BamHI/HindIII cloning site is positioned at location 269 and 295, respectively, and four stop codons (274, 277, 287, 291) are located between the two restriction sites. Upstream of the BamHI restriction site are located a FLAG-tag (244–267), a ribosome binding site (RBS, 227–232), a hairpin-loop (178–198), and a T7 promoter (162–177). Immediately after the HindIII restriction site are located a linker sequence (300–356) which are followed by a 276-bp “tether” region, TolA (357–632) ending with a hairpin loop region (610–632). In pFP RDV1, the FLAG-tag is replaced with a RGS-His6 tag. (b) DNA construct used for in vitro transcription: T7p: T7 promoter, aho7c∗: randomized gene of Aho7c, tolA: TolA spacer. Standard and degenerated oligonucleotides used to generate the library in this format are indicated by black and gray horizontal arrows, respectively. The amino acid residues 9, 10, 22, 23, 25, 32, 34, 41, 43, and 45 were randomized (Fig. 1). To generate a DNA product with an encoded RGS-His6-tag (as in pFP RDV1), the primer SDA_RGS is used instead of SDA_FLAG

28

Valentina Kalichuk et al.

and MluI (499 bp). The reaction contains 60 ng of the synthetic DNA product, 100 ng of the pUC18 vector, 1 μL of 10 T4 DNA ligase buffer and UHP water up to 10 μL. Add 2.5 U of T4 DNA ligase and incubate overnight at room temperature. 8. Inactivate the ligase for 10 min at 70
C. 9. Transform E. coli DH5α cells to obtain individual clones and sequence some of them. pFP RDV2

pFP RDV2 plasmid is generated by subcloning a cassette encoding the FLAG-tag into pFP RDV1 via XbaI and BamHI. The cassette is prepared by hybridization of two 50 -phosphorylated oligonucleotides (K7-RDV2-F and K7-RDV2-R). The following protocol was used: 1. Prepare a 6 μL mix in a PCR tube containing 0.45 pmol of each oligonucleotide (0.45 μL of 1 μM primer K7-RDV2-F and K7-RDV2-R). 2. Place the tube in a beaker containing 500 mL of boiling water and let it cool down to 25
C (see Note 9). 3. Ligate the cassette (69 bp) with pFP RDV1 predigested with XbaI and BamHI (2711 bp). The reaction contains 2.55 ng of the cassette and 100 ng of pFP RDV1 vector (molar ratio 1:1), 1 μL of 10 T4 DNA ligase buffer and UHP water up to 10 μL. Add 2.5 U of T4 DNA ligase and incubate overnight at room temperature. A negative control can also be prepared by replacing the cassette in the reaction with UHP water. 4. Inactivate the ligase for 10 min at 70
C. 5. Control the ligation by performing a PCR. In a PCR tube, mix 1 μL of the ligation reaction, 10 pmol of each primer (0.1 μL of 100 μM primer SDA-FLAG and TolAkurz), 0.4 μL of dNTPs mix (containing 10 mM of each dNTP), 4 μL of 5 PhusionHF buffer, 0.2 U of Phusion Hot Start II DNA polymerase and UHP water till final volume of 20 μL. 6. Use a thermocycler to perform the following PCR program: an initial denaturation step at 98
C for 30 s, followed by 25 cycles of 98
C for 10 s and 72
C for 30 s, with a final elongation step of 72
C for 5 min. 7. Mix 1 μL of 6 Thermo Fisher Scientific loading buffer with 5 μL of the PCR product and run on a 1.5% agarose gel to be sure that the PCR reaction gives a band corresponding to the expected size of 452 bp. 8. Transform E. coli DH5αF’IQ cells according to the manufacturer’s instructions to obtain individual clones and sequence some of them.

Selection of Aho7c-Based Affitins 3.1.3 Production of the tolA Fragment

29

1. Large quantities of tolA spacer can be obtained quickly via PCR amplification from pFP-RDV1 vector. Prepare two tubes with 50 μL PCRs containing 25 pmol of each primer (0.25 μL of 100 μM primer AF-link-F and TolAkurz), 25 ng of pFP-RDV1 vector (0.25 μL of 100 ng/μL), 1 μL of dNTPs mix (containing 10 mM of each dNTP), 10 μL of 5 Phusion HF buffer, and 1 U of Phusion Hot Start II DNA polymerase. 2. Place the tubes on a thermocycler and use the following PCR program: first an initial denaturation at 98
C for 30 s, then 30 cycles of 98
C for 10 s, 69
C for 30 s, 72
C for 25 s, and final elongation at 72
C for 5 min. Control the product on 1.5% agarose gel. The PCR should give an amplicon of 339 bp. 3. Purify the tolA linker with Promega Wizard SV Gel and PCR Clean-up kit. 4. Determine the concentration of the tolA linker by UV absorbance and store the product at 20
C.

3.1.4 Production of the Ribosome Display Construct

1. Prepare a series of PCR reaction mixes in PCR tubes (up to 10) containing 50 pmol of each primer (0.5 μL of 100 μM primer T7B and TolAkurz), 40 ng of Aho7c library, 53 ng of tolA linker, 1 μL of dNTPs mix (containing 10 mM of each dNTP), 10 μL of 5 Phusion HF buffer, and 1.5 U of Phusion Hot Start II polymerase then add UHP water to 50 μL. 2. Place the tubes in a thermocycler and use the following PCR program: an initial denaturation step at 98
C for 30 s, followed by 8 cycles of 98
C for 10 s, 45
C for 30 s, 72
C for 30 s, and then 30 cycles of 98
C for 10 s, 55
C for 30 s, 72
C for 30 s with a final elongation step of 72
C for 5 min. 3. Check on a 1.5% agarose gel that there is only one PCR product of 635 bp. If there are additional products, perform gel extraction of the band with the right size. Each μg of the obtained library is equivalent to about 1.46  1012 molecules (see Note 10). The library can be stored at 80
C for several months or years.

3.1.5 Subcloning for Sequencing

Clone some of the library into the pFP1001 vector using BamHI and HindIII restriction enzymes in order to sequence individual library members (see Note 11). 1. Prepare a PCR reaction containing 10 ng of the purified library from Subheading 3.1.4 as a template, 25 pmol of each primer (0.25 μL of 100 μM primer RDV2-RT, introducing BamHI restriction site and AF-link-R, introducing HindIII restriction site), 1 μL of dNTPs mix (containing 10 mM of each dNTP), 2 μL of DMSO, 10 μL of 5 Phusion HF buffer, and 0.5 U of Phusion Hot Start II polymerase then add UHP water up to 50 μL.

30

Valentina Kalichuk et al.

2. Place the tube on a thermocycler and preheat the reaction to 98
C for 30 s. Perform 25 cycles of PCR of 98
C for 10 s, 60
C for 30 s, 72
C for 20 s with a final elongation step of 72
C for 5 min. 3. Run a 1.5% agarose gel to be sure that the PCR reaction gives a band corresponding to the expected size of 225 bp. 4. Purify the product with Promega Wizard SV Gel and PCR Clean-up kit and determine the concentration of the purified PCR product by UV absorbance. 5. Digest about 200 ng of PCR DNA with BamHI and HindIII enzymes for 2 h at 37
C. 6. Run a 1.5% agarose gel to be sure that the PCR reaction gives a band corresponding to the expected size of 180 bp. 7. Purify the digested product with Promega Wizard SV Gel and PCR Clean-up kit and determine the concentration by UV absorbance. 8. Digest 500 ng of pFP1001 DNA with BamHI and HindIII for 2 h at 37
C. Dephosphorylate 50 and 30 ends of the digested DNA fragments with FastAP phosphatase for 30 min at 37
C and inactivate enzymes for 15 min at 65
C. This preparation of digested vector is ready to be used for ligation without further purification. 9. Ligate the purified DNA product with the pFP1001 DNA predigested with BamHI and HindIII. The reaction contains 100 ng of the pFP1001 fragment, 10 ng of the library PCR, 1 μL of 10 T4 DNA ligase buffer and UHP water up to 10 μL. Add 2.5 U of T4 DNA ligase and incubate overnight at room temperature. 10. Inactivate the ligase for 10 min at 70
C. 11. Transform E. coli DH5αF’IQ cells with the ligation product, and plate on LB/agar/ampicillin/kanamycin Petri plates to obtain individual clones. 12. Sequence several clones using primer Qe30for (or Qe30rev) to ensure that the library was synthesized as designed and there is no strong bias in the nucleotide composition (see Note 12). 3.2 In Vitro Transcription

1. Use the TranscriptAid T7 High Yield Transcription Kit from Thermo Fisher. Prepare the transcription reaction following the provider protocol with reagents warmed at room temperature (except enzyme) using 1.2 μg of the prepared DNA library (6 μL of 200 ng/μL concentrate). Incubate at 37
C for 2 h (see Notes 13–15). The transcription mixture should start becoming turbid after about 10 min of incubation. This step is crucial. Lack of turbidity indicates failure of the transcription reaction.

Selection of Aho7c-Based Affitins

31

2. Treat the transcription reaction (20 μL) with 0.5 μL RNA-free DNase I (Thermo Fisher Scientific, 50 U/μL) in order to eliminate the DNA input. 3. Inactivate DNase I with 2 μL of 0.5 M EDTA. 3.2.1 Subsequently, Isolate the RNA Using the Following Standard LiCl Precipitation Method

1. Add 180 μL of ice-cold UHP water and 200 μL of 6 M LiCl, vortex and incubate on ice for at least 30 min. 2. Centrifuge at 20,000  g for 30 min at 4
C and discard the supernatant. 3. Wash the pellet two times with 500 μL 70% ice-cold ethanol and dry it for 5 min in a vacuum concentrator. 4. Dissolve the pellet with 200 μL of ice-cold UHP water by pipetting up and down. 5. Centrifuge at 20,000  g for 5 min at 4
C. 6. Transfer 180 μL of the supernatant to a new tube, add 20 μL of 3 M sodium acetate, 500 μL 100% ice-cold ethanol, vortex and incubate at 20
C for minimum 30 min or overnight. 7. Centrifuge at 20,000  g for 30 min at 4
C and discard the supernatant. 8. Wash the pellet two times with 500 μL 70% ice-cold ethanol. 9. Centrifuge at 20,000  g for 5 min at 4
C and discard remaining ethanol with a pipette. 10. Dry the pellet for at least 5 min in a vacuum concentrator. 11. Dissolve the pellet in 52 μL of ice-cold UHP water by pipetting up and down, the concentration should be around 2.5 μg/μL. For long storage keep at 80
C. 12. Determine the concentration of the obtained mRNA by UV absorbance and control the quality on agarose gel using the 2 RNA Loading Dye and RiboRuler High Range RNA Ladder from the TranscriptAid T7 High Yield Transcription Kit. 13. Purify the mRNA using a NucleoSpin RNA XS Kit in combination with NucleoSpin RNA/DNA Buffer Set (MachereyNagel) according to the manufacturer’s specifications and measure the concentration. Dilute the mRNA with UHP water to a final concentration of 2.5 μg/μL. Aliquot and store at 80
C. This step ensures template-free RNA which is crucial for efficient progression of the selection. 14. Check the removal of template by performing a PCR as in Subheading 3.3, step 12 in a volume of 20 μL using 0.5 ng of purified mRNA and 25 cycles of amplification.

3.3 Ribosome Display Selection in a 96-Well Microtiter Plate

1. Add 100 μL of a 66 nM solution of streptavidin or neutravidin in TBS into the wells of a MaxiSorp microtiter plate (see Notes 16 and 17). Incubate overnight at 4
C, or for 1 h at room temperature with shaking.

32

Valentina Kalichuk et al.

2. Wash three times with 300 μL of 1 TBS. Add 300 μL of 1 TBS containing 0.5% BSA to block any remaining surfaces of the plate and incubate at room temperature for 1 h or overnight at 4
C. 3. Discard the remaining buffer from the plate, beat it dry, add 100 μL of a 150 nM solution of biotinylated target in 1 TBS-BSA 0.5% and incubate the plate at room temperature for 1 h or overnight at 4
C (see Note 18). Remove the solution; wash with 4  300 μL of TBS and once with 300 μL of WBT. Fill the wells with WBT and keep the plate at 4
C till needed. 4. Prepare the translation reaction. Each 27.5 μL reaction contains 0.5 μL of 200 mM methionine, 12.6 μL of Premix, 12.5 μL of S30 extract, and 5 μg of mRNA from Subheading 3.2 (see Note 19). Translate for 8 min at 37
C (see Note 20). Add 120 μL of ice-cold WBTH-BSA to stop the translation reaction. Then centrifuge at 20,000  g for 5 min at 4
C, and place on ice till use. 5. For the selection of binders, remove the WBT only from the wells lacking the target protein and add 140 μL of the stopped translation reaction in its place to perform the pre-panning step. Incubate for 60 min under gentle shaking at 4
C. 6. Further transfer the 140 μL of the stopped translation reaction to the target-coated wells to perform the selection step. Incubate for 60 min under gentle shaking at 4
C. 7. Dry the plate and wash eight times for few seconds to remove unbound complexes. The first wash is done with 300 μL of ice-cold WBT-BSA, the following ones with 300 μL of ice-cold WBT. 8. To elute the bound mRNA, add twice 100 μL of ice-cold Elution Buffer containing 50 μg/mL S. cerevisiae RNA. Incubate for 10 min under gentle shaking at 4
C. 9. Purify the mRNA using RNeasy MinElute Cleanup Kit (Qiagen, elute mRNA with 14 μL Elution Buffer) according to the manufacturer’s specifications. 10. Prepare a PCR tube per eluted well containing 12.8 μL of eluted purified mRNA (from step 9) and 20 pmol primer AFlink-R (0.2 μL of 100 μM). Incubate at 70
C for 5 min and, immediately after denaturation, chill on ice. 11. Prepare the reverse transcription mix; per reaction add 4 μL of 5 Reverse Transcription buffer, 2 μL of dNTPs mix (containing 10 mM of each dNTP), 0.5 μL of Ribolock (40 U/μL), 1 μL of RevertAid Minus H reverse transcriptase. Add the 13 μL denatured mRNA (from step 10) to the 7.5 μL reverse transcription mix and incubate at 42
C for 1 h in a PCR

Selection of Aho7c-Based Affitins

33

machine. Stop the reaction by heating at 70
C for 10 min. Chill on ice. 12. To amplify the reverse transcription products, prepare a series of RT-PCR reaction mixtures in PCR tubes containing 15 pmol of each primer (0.15 μL of 100 μM primer RDV2RT and AF-link-R), 5 μL of the reverse transcription template (from step 11), 1 μL of dNTPs mix (containing 10 mM of each dNTP), 2 μL of DMSO, 10 μL of 5 Phusion HF buffer, and 1 U of Phusion Hot Start II polymerase then add UHP water up to 50 μL (see Notes 16, 21 and 22). 13. Place the tubes on a thermocycler and preheat the reaction to 98
C for 30 s. Perform 30 cycles of PCR of 98
C for 10 s, 61
C for 30 s, 72
C for 10 s with a final elongation step of 72
C for 5 min. 14. Run a 1.5% agarose gel to be sure that the PCR reaction gives a band corresponding to the expected size of 225 bp (see Note 23). 15. Perform an Exonuclease I treatment for 1 h at 37
C (0.5 μL from 20 U/μL directly into the 50 μL PCR reaction) to remove residual primers. 16. Purify the product with Promega Wizard SV Gel and PCR Clean-up kit. 17. Determine the concentration of the purified PCR product by UV absorbance and store the product at 20
C or at 80
C for several months. 3.4 Additional Selection Rounds

To proceed with an additional round of selection, the promoter and the spacer regions must be re-incorporated into the selection output by ligating it back into the ribosome display vector as in Subheading 3.1.4. For this purpose, enough amounts of both selection output and vector need to be generated by PCR (see Note 24). 1. In order to amplify the ribosome display vector, prepare a 250 μL PCR with the following composition: 50 μL of 5 HF buffer, 5 μL of 10 mM dNTPs, 50 pmol of each primer (RDV2-Famp and RDV2-Ramp), 11 μL of 100% DMSO and 5 U of Phusion Hot Start II DNA polymerase. According to Subheading 3.1.5 (step 8), use 10 ng of BamHI/HindIII digested vector as a template. Distribute the reaction mixture into 5 PCR tubes and amplify as follows: initial denaturation at 98
C for 30 s, 30 cycles of 98
C for 30 s, 63
C for 30 s, 72
C for 40 s and final extension 72
C for 5 min. Analyze the amplification product by pooling all 5 PCRs and loading 5 μL of product on 1% agarose gel. A single, well-defined amplicon of about 2773 bp must be observed.

34

Valentina Kalichuk et al.

2. Remove residual primers from the ribosome display vector amplicons by treating them with Exonuclease I. Add 0.5 μL of enzyme (20 U/μL) to each 50 μL of reaction mixture. Incubate at 37
C for 1 h and then desalt the PCR products using Promega Wizard SV clean up system. Proceed with next step or store at 20
C until needed. 3. Digest about 1 μg of both selection output from Subheading 3.3 and ribosome display vector using BamHI and HindIII restriction enzymes according to Subheading 3.1.5 (steps 5–8). Perform a reaction clean up using Promega Wizard SV clean up system and estimate the concentration of the purified products. Proceed with next step or store at 20
C until needed. 4. Ligate about 100 ng of digested vector to 30 ng of digested selection output in a 20 μL ligation reaction at 37
C for 1 h (5U T4 ligase, 2 μL 10 ligase buffer). Proceed with next step or store at 20
C until needed. 5. Use the entire ligation volume as a template for 200 μL PCR with the following composition: 40 μL of 5 HF buffer, 4 μL of 10 mM dNTPs, 40 pmol of each primer, T7C and TolAext, and 5 U of Phusion Hot Start II DNA polymerase. Again, distribute the reaction mixture into 4 PCR tubes and amplify as follows: initial denaturation at 98
C for 30 s, 25 cycles of 98
C for 30 s, 72
C for 20 s and final extension 72
C for 5 min. Analyze the amplicon by pooling the 4 PCRs and loading 5 μL of product on 1% agarose gel. A single, welldefined amplicon of about 635 bp must be observed. 6. Remove residual primers with Exonuclease I as explained in point 2 and then desalt the PCR products using Promega Wizard SV clean up system. Proceed with next step or store at 20
C until needed. 7. Concentrate the purified product with a vacuum concentrator down to 150–200 ng/μL. Use the PCR product for in vitro transcription as in Subheading 3.2. The product is stable for up to several years at 80
C. 8. Proceed with the immobilization of the biotinylated target on a microtiter plate previously coated with NeutrAvidin as in Subheading 3.3 (see Note 25). Repeat the round as described in Subheading 3.3 (see Notes 26 and 27). 3.5 Selection Using Magnetic Beads

Alternatively, the selection could be performed in solution on magnetic beads as follows: for each sample, prepare two tubes (one for the positive and one for the negative selection step), each containing 50 μL Bio-Adembeads Streptavidin or SteptaDivin.

Selection of Aho7c-Based Affitins

35

1. Place the tubes on a magnetic separator and wait 4 min or until all the beads are collected. Remove the clear supernatant. Do not dry the beads in any of the steps. 2. Wash the beads three times with 500 μL WB. Between washings, collect the beads as in step 1. 3. Resuspend the beads in 300 μL WB-BSA 0.5% and block for 1 h at 4
C with shaking. 4. Wash the beads three times with 500 μL WB and resuspend them in 200 μL WBT-BSA 0.5%. The beads are ready for use. 5. Collect the beads in the tubes for negative selection and remove the 200 μL buffer. Add 140 μL stopped and centrifuged translation reaction (see Subheading 3.3, step 4) and incubate for 1 h at 4
C with shaking. 6. Collect the beads on a magnetic stand and transfer the supernatant to a new tube. Add your biotinylated target and incubate with shaking for 2 h at 4
C. 7. Transfer the supernatant to the second tube containing 50 μL blocked beads (the 200 μL buffer should be removed) and capture the ternary complexes for 15 min at 4
C with shaking. 8. Wash the beads with 500 μL ice-cold WBT-BSA and WBT as in Subheading 3.3, step 7. Collect the beads with the magnet between each washing step. 9. Elute the ternary complexes as indicated in Subheading 3.3, step 8. Proceed with mRNA purification as already described. 3.6 Follow-Up of the Selection

Generally, a decreasing number of PCR cycles needed to obtain the same quantity of RT-PCR product from round to round is a good sign for an enrichment of the selection. Often, a decrease of 5 cycles for every round of selection is observed (see Note 27). The last selection rounds can be done in parallel in the presence and absence of the target in the selection of wells or tubes. This allows the specificity of the selected Affitins to be tested. While an RT-PCR product should be obtained in the presence of the target, no RT-PCR product should be obtained in its absence. To verify that the selection is successful, it is essential to carry out an ELISA test to detect the binding of Affitins to the target via their FLAG- or RGS-His6 tag. To do this: 1. Coat wells with biotinylated target as described in Subheading 3.3, steps 1 and 2, as well as wells for negative control with streptavidin or NeutrAvidin. 2. For each selection round, translate the pools in vitro for 1 h at 37
C (see Subheading 3.3, step 4). 3. Dilute the translations eight times with the appropriate volume of 1 TBS containing 0.1% Tween and distribute 100 μL of this mixture into each test well.

36

Valentina Kalichuk et al.

4. Incubate for 1 h at room temperature with agitation; wash with 6 300 μL of 1 TBS-Tween. 5. Distribute 100 μL of anti-RGS-His6-HRP conjugate (1/10,000), or anti-FLAG-HRP conjugate (1/10,000), in each well; incubate for 1 h at room temperature with agitation. 6. Wash with 6  300 μL of TBS with Tween 0.1% and distribute 100 μL of OPD substrate 1 mg/mL and read the extinction value at 450 nm (see Note 28). 3.7 Screening of Clones and Sequence Analysis

Sequences which have been selected can be analyzed as follows: 1. Subclone the RT-PCR product from a selection round into pFP1001 via BamH1 and HindIII restriction sites according to Subheading 3.1.5 (steps 5–10). 2. Transform E. coli DH5αF’IQ cells with the ligation product, and plate on LB/agar/ampicillin/kanamycin Petri plates to obtain individual clones. Incubate overnight at 37
C. 3. Affitins from isolated clones are then produced in a 96-well Deep Well Culture Plate as follows: 4. Distribute with a step-pipette 1.4 mL/well of 2YT containing ampicillin/kanamycin/glucose. 5. Inoculate each well with a colony from the plate. 6. Seal the Deep-Well with a gas permeable adhesive and incubate overnight at 37
C with shaking at 750 rpm. 7. Add glycerol to a final concentration of 16% to each well, homogenize with shaking. 8. This master plate can be kept for long-term storage at 80
C. 9. For the production of Affitins, distribute with a step-pipette 1.2 mL/well of 2YT containing ampicillin/kanamycin/glucose in a new 96-well Deep Well Culture Plate. 10. Inoculate with a multi-channel pipette each well with 200 μL from the master plate. 11. Incubate 3 h at 37
C with shaking at 750 rpm (the medium should become turbid). 12. Induce the production of Affitins with 0.5 mM IPTG and incubate 3 h or overnight at 30
C with shaking at 750 rpm. 13. Centrifuge the Deep-Well plate for 20 min at 2000  g. 14. Discard supernatants by flicking quickly the Deep-Well on top of a trash. 15. Add 50 μL per well of lysis buffer and shake the plate for 30 min at 1000 rpm at room temperature. 16. Add 250 μL per well of TBS and shake the plate for 30 min at 1000 rpm at room temperature.

Selection of Aho7c-Based Affitins

37

17. Centrifuge the deep-well plate for 20 min at 2000  g at room temperature. 18. Supernatants from these crude extracts can be then screened by an ELISA test as in Subheading 3.5, step 3 by using 100 μL per well. 19. Use Qe30for and Qe30rev primers to sequence positive clones.

4

Notes 1. This mutagenesis scheme targets amino acids in the DNA-binding area of the Sul7d family. This surface is large enough and slightly concave to match the globular shape of most proteins. 2. The tolA spacer allows the displayed proteins to properly fold away from the ribosome and to have enough degree of freedom to interact with the target. 3. We use the construction described by Amstutz et al. [22] but with modifications to meet our needs, such as different detection tags and restriction sites. This system uses prokaryotic in vitro translation with an E. coli S30 extract for ribosome display (Fig. 2). 4. For the construction of high-quality libraries, it is recommended to use highly purified oligonucleotides (HPLC or gel purification). This is to avoid as much as possible undesirable sequences due to n  1 products. 5. NNS (N ¼ A, C, T or G and S ¼ C or G) and NHK (H ¼ A, C or T and K ¼ G or T) codons used for the mutagenesis encode all amino acids or a subset of 16 amino acids (i.e., excluding W, C, G, and R), respectively, while minimizing the number of stop codons. We use NHK codons for positions where W induces multimerization (unpublished observation). 6. This DNA product corresponds to the gene of Aho7c, flanked by the 50 sequence necessary for ribosome display and an additional 30 sequence necessary for subsequent PCR assembly step with the tolA spacer (Fig. 2). 7. The upper limit for the size of the library at this step can be estimated to 1012 variants when manipulating 1 μg of DNA at this step. 8. The sequence of the synthetic DNA product is (NcoI and MluI restriction sites are in bold): 50 -CCATGGATACGAAATTAATACGACTCACTATAGGGA GACCACAACGGTTTCCCTCTAGAAATAATTTTGTTT AACTTTAAGAAGGAGATATATCCATGAGAGGATCG CATCACCATCACCATCACGGATCCTAATGAGGTA

38

Valentina Kalichuk et al.

CCCTGAGTAGAAGCTTTATATGGCCTCGGGGGCC GAATTCTCTGAGCTCTCTACCGGTTCCGGCGGTT CTGGCCAGAAGCAAGCTGAAGAGGCGGCAGCGAAA GCGGCGGCAGATGCTAAAGCGAAGGCCGAAGCAG ATGCTAAAGCTGCGGAAGAAGCAGCGAAGAAAGC GGCTGCAGACGCAAAGAAAAAAGCAGAAGCAGAA GCCGCCAAAGCCGCAGCCGAAGCGCAGAAAAAAG CCGAGGCAGCCGCTGCGGCACTGAAGAAGAAAGC GGAAGCGGCAGAAGCAGCTGCAGCTGAAGCAAG AAAGAAAGCGGCAACTGAAACCGCACACCTTACT GGTGTGCGGTAAACGCGT-30 . 9. The cassette does not need to be predigested for subcloning by XbaI and BamHI as hybridized 50 phosphorylated oligonucleotides generate cohesive ends. 10. The quality of the PCR product is crucial for the efficiency of the subsequent in vitro transcription. When a single sharp band without any smear appears on agarose electrophoresis, use this product without further gel-extraction as the template for in vitro transcription. 11. pFP1001 is an expression vector derived from the Qiagen vector pQE30 in which we have replaced the suppressible stop of the original vector with two non-suppressible stops. This allows expression in the DH5αF’IQ. pQE30 strain, or any other vector with unique restriction sites BamHI and HindIII, can be used instead of pFP1001. 12. Usually the sequencing of some dozens of randomly picked clones confirms that the observed residue frequency is similar to the predicted one. Alternatively, one can use next generation sequencing to evaluate quality of the library. From about 500,000 Affitins sequenced for library L5, we determined that, by following this protocol for library preparation, ~90% of them were different and that the largest cluster of sequences was composed of no more than 6 members (unpublished observation). 13. Calculate the volume of the transcription reaction required to retain the library complexity. 14. Incubations can be performed from 1 h to overnight. Generally, most of the product is generated within 1–2 h; however, longer incubations will yield more product if necessary. 15. Several basic precautions including wearing gloves, using disposable materials and filter tips and tubes purchased RNase/ DNase-free are required when working with RNA. Water can be DEPC-treated. A water purification system producing RNase-free water can also be used. 16. We usually perform the first round with four wells per target and combine the resulting pools for the second round of

Selection of Aho7c-Based Affitins

39

selection. The buffer used for immobilization must be adapted depending on target and its eventual specific requirements. 17. A non-biotinylated target can be immobilized directly without using streptavidin with risks that this can lead to at least partial denaturation of the target molecule and to selection against epitopes which are not accessible in the native form of the target. Whenever possible, prefer to use indirect immobilization by biotinylation/streptavidin couple. 18. Consider preparing wells that are identically coated but lack the target to perform a negative selection step (pre-panning) to reduce risks to isolate sequences with affinity for streptavidin or matrix. 19. We use the prokaryotic in vitro translation system with an E. coli S30 extract and Premix prepared according to the procedure of Amstutz et al. [22] though other commercially available translation mixes can be used when only a few translation reactions are performed. 20. For each new batch of E. coli S30 extract and Premix, the duration of the translation should be optimized. Usually, translation times between 5 and 10 min give optimal yields. 21. Perform the RT-PCR reaction at least in quadruplicate in order to maintain variability of selected sequences. It is important to prepare negative control reaction by replacing the mRNA template with water to detect a contamination of reagents; this control will be checked by PCR in the next step (see Note 22). 22. Prepare negative control reaction by replacing the RT template with water to detect contamination of PCR solutions. Test also the RT negative control prepared in the previous step (see Note 21). If a band is observed in the negative reaction, discard all PCR solutions and repeat the selection round. 23. When a diffuse band or other side-products appear on agarose electrophoresis, gel-purify the band of interest and use this product to initiate a second PCR. This will normally yield a high-quality DNA. 24. The procedure is given for pFP RDV2, oligonucleotides used to amplify pFP RDV1 are different. 25. It is preferable to alternate the use of streptavidin and NeutrAvidin from round to round in order to avoid the generation of binders against both molecules. 26. The stringency of the selection can be increased each round by: (a) decreasing the concentration of the target protein (200 nM, 100 nM, 50 nM and finally 10 nM, for instance), and/or (b) increasing the number of washing steps or the duration of the washing steps (6 1 min, 6 5 min, 6 10 min, 6 20 min, for instance).

40

Valentina Kalichuk et al.

27. The number of RT-PCR cycles needs to be adjusted for each selection round. If too many cycles are done, this will normalize the relative proportions of different pool members, reducing the selective enrichment due to binding. A general rule is to reduce by about 5 cycles per round. 28. The ratio of (OD450 for the target)/(OD450 for the negative control) should be at least ten for a successful selection.

Acknowledgments The authors thank all previous members of the laboratory who helped to develop this protocol. In memoriam of Ghislaine Be´har. References 1. Mouratou B, Schaeffer F, Guilvout I, TelloManigne D, Pugsley AP, Alzari PM, Pecorari F (2007) Remodeling a DNA-binding protein as a specific in vivo inhibitor of bacterial secretin PulD. Proc Natl Acad Sci U S A 104 (46):17983–17988 2. Pecorari F, Alzari PM (2008) OB-fold used as scaffold for engineering new specific binders. Patent Publication Nos PCT/IB2007/ 004388 3. Be´har G, Bellinzoni M, Maillasson M, PaillardLaurance L, Alzari PM, He X, Mouratou B, Pecorari F (2013) Tolerance of the archaeal Sac7d scaffold protein to alternative library designs: characterization of antiimmunoglobulin G Affitins. Protein Eng Des Sel 26(4):267–275 4. Correa A, Pacheco S, Mechaly Ariel E, Obal G, Be´har G, Mouratou B, Oppezzo P, Alzari PM, Pecorari F (2014) Potent and specific inhibition of glycosidases by small artificial binding proteins (Affitins). PLoS One 9(5):e97438 5. Krehenbrink M, Chami M, Guilvout I, Alzari PM, Pecorari F, Pugsley AP (2008) Artificial binding proteins (Affitins) as probes for conformational changes in secretin PulD. J Mol Biol 383(5):1058–1068 6. Buddelmeijer N, Krehenbrink M, Pecorari F, Pugsley AP (2009) Type II secretion system secretin PulD localizes in clusters in the Escherichia coli outer membrane. J Bacteriol 191 (1):161–168 7. Miranda FF, Brient-Litzler E, Zidane N, Pecorari F, Bedouelle H (2011) Reagentless fluorescent biosensors from artificial families of antigen binding proteins. Biosens Bioelectron 26(10):4184–4190

8. Pacheco S, Behar G, Maillasson M, Mouratou B, Pecorari F (2014) Affinity transfer to the archaeal extremophilic Sac7d protein by insertion of a CDR. Protein Eng Des Sel 27 (10):431–438 9. Be´har G, Pacheco S, Maillasson M, Mouratou B, Pecorari F (2014) Switching an anti-IgG binding site between archaeal extremophilic proteins results in Affitins with enhanced pH stability. J Biotechnol 192:123–129 10. Kalichuk V, Renodon-Corniere A, Behar G, Carrion F, Obal G, Maillasson M, Mouratou B, Preat V, Pecorari F (2018) A novel, smaller scaffold for Affitins: showcase with binders specific for EpCAM. Biotechnol Bioeng 115(2):10 11. Cinier M, Petit M, Williams MN, Fabre RM, Pecorari F, Talham DR, Bujoli B, Tellier C (2009) Bisphosphonate adaptors for specific protein binding on zirconium phosphonatebased microarrays. Bioconjug Chem 20 (12):2270–2277 12. Be´har G, Renodon-Cornie`re A, Mouratou B, Pecorari F (2016) Affitins as robust tailored reagents for affinity chromatography purification of antibodies and non-immunoglobulin proteins. J Chromatogr A 1441:44–51 13. Fernandes CSM, Rd S, Ottengy S, Viecinski AC, Be´har G, Mouratou B, Pecorari F, Roque ACA (2016) Affitins for protein purification by affinity magnetic fishing. J Chromatogr A 1457:50–58 14. Vukojicic P, Be´har G, Tawara MH, FernandezVillamarin M, Pecorari F, Fernandez-Megia E, Mouratou B (2019) Multivalent Affidendrons with High Affinity and Specificity toward as

Selection of Aho7c-Based Affitins Versatile Tools for Modulating Multicellular Behaviors. ACS Appl Mater Interfaces 11(24):21391–21398 15. Kalichuk V, Behar G, Renodon-Corniere A, Danovski G, Obal G, Barbet J, Mouratou B, Pecorari F (2016) The archaeal “7 kDa DNA-binding” proteins: extended characterization of an old gifted family. Sci Rep 6:37274 16. Mattheakis LC, Bhatt RR, Dower WJ (1994) An in vitro polysome display system for identifying ligands from very large peptide libraries. Proc Natl Acad Sci U S A 91 (19):9022–9026 17. Gera N, Hussain M, Wright RC, Rao BM (2011) Highly stable binding proteins derived from the hyperthermophilic Sso7d scaffold. J Mol Biol 409(4):601–616 18. Zhao N, Schmitt MA, Fisk JD (2016) Phage display selection of tight specific binding variants from a hyperthermostable Sso7d scaffold protein library. FEBS J 283(7):1351–1367 19. Hanes J, Jermutus L, Weber-Bornhauser S, Bosshard HR, Plu¨ckthun A (1998) Ribosome display efficiently selects and evolves high-

41

affinity antibodies in vitro from immune libraries. Proc Natl Acad Sci U S A 95 (24):14130–14135 20. Mouratou B, Be´har G, Paillard-Laurance L, Colinet S, Pecorari F (2012) Ribosome display for the selection of Sac7d scaffolds. Methods Mol Biol 805:315–331 21. Hanes J, Pluckthun A (1997) In vitro selection and evolution of functional proteins by using ribosome display. Proc Natl Acad Sci U S A 94 (10):4937–4942 22. Amstutz P, Binz HK, Zahnd C, Pluckthun A (2006) Ribosome display: in vitro selection of protein-protein interactions. In: Celis J (ed) Cell biology—a laboratory handbook, vol 1. Elsevier Academic Press, Cambridge, pp 497–509 23. Roy A, Kucukural A, Zhang Y (2010) I-TASSER: a unified platform for automated protein structure and function prediction. Nat Protoc 5(4):725–738 24. Zhang Y (2008) I-TASSER server for protein 3D structure prediction. BMC Bioinformatics 9:40

Chapter 3 cDNA Display: A Stable and Simple Genotype–Phenotype Coupling Using a Cell-Free Translation System Hidenao Arai, Shigefumi Kumachi, and Naoto Nemoto Abstract A cDNA display method was developed based on the mRNA display method to increase its stability and efficiency for the directed evolution of various kinds of peptides and proteins. In this method, the puromycin-linker is a key molecule to realize smart genotype–phenotype coupling. A recently improved puromycin-linker and its use were explained in detail for the in vitro selection of peptides and proteins using the cDNA display method. Key words cDNA display, mRNA display, In vitro selection, Directed evolution, Puromycin-linker, cnvK, Photocrosslinking

1

Introduction Although a ribozyme includes a genotype and its phenotype, genotype–phenotype coupling is indispensable for the directed evolution of peptides or proteins (enzymes) because a protein genotype cannot be bound with its encoded DNA without a cell or virus. Thus, genotype–phenotype coupling with a cell-free translation system in a test tube takes a little ingenuity. Using a puromycinlinker is one of the few excellent ideas (e.g., ribosome display, CIS display) for genotype–phenotype coupling. In vitro virus [1] and mRNA display [2] enable an mRNA to fuse with its encoded protein using a puromycin-linker with a cell-free translation system. The library size for these screening methods is 10,000 times larger than that for phage display (Fig. 1a). However, the kinds of target molecules for screening become restricted because mRNA is very unstable (e.g., degradation by ribonuclease contamination) during the selection process. Thus, a cDNA display (Fig. 1b) was developed to link cDNA with its encoded protein by designing a novel puromycin-linker, which enables the mRNA display to convert the cDNA display using reverse transcription on a bead (Fig. 1c). Some puromycin-linkers were gradually improved to increase

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2070, https://doi.org/10.1007/978-1-4939-9853-1_3, © Springer Science+Business Media, LLC, part of Springer Nature 2020

43

44

Hidenao Arai et al.

(a)

cDNA

mRNA display

mRNA (b)

Puromycin

cDNA

cDNA display

mRNA

Puromycin

(c) Elution & PCR DNA library

DNA sequencing Affinity selection

Transcription

mRNA

Puromycin linker Magnetic beads immobilizing target molecule

Ligation

Selection cycle cDNA display

Enzymatic digestion

mRNA-linker

in vitro translation

cDNA

mRNA-peptide/protein

Reverse transcription

Immobilization on beads & Purification Strepavidin beads

Fig. 1 (a) Schematic illustration of mRNA display, (b) cDNA display, (c) and in vitro selection cycle using the cDNA display. (a, b) POI is covalently linked with mRNA and cDNA in the mRNA and cDNA displays, respectively. (c) The preparation procedure for the cDNA display molecule was constituted from each step: i.e., transcription (mRNA preparation), ligation (photocrosslinking), in vitro translation (formation of mRNA display), immobilization on beads, purification (buffer exchange), reverse transcription (formation of cDNA display), and enzymatic digestion. The cDNA display molecule was purified using the Ni-NTA magnetic beads and the cDNA display was then brought to the affinity selection against the target molecule

productivity and overall efficiency (Fig. 2). The mRNA display was originally prepared by ligation reaction of mRNA with a puromycin-DNA fragment with T4 RNA ligase [1] or T4 DNA ligase using a splint DNA [2]. The ligation efficiency of both methods was very low despite the excess input from the puromycin-linker (200 times more than mRNA) (Fig. 2a). To overcome the mRNA display weakness (i.e., low ligation efficiency and mRNA instability), a cDNA display was developed by introducing a hybridization region with mRNA and biotin into the puromycin-linker

Genotype-Phenotype Linking by cDNA Display

Linking method

45

mRNA : Linker

Reaction time

1 : 200 ~ 400

Over night

(a) The linker used for mRNA display T4 DNA Ligase or T4 RNA Ligase

Enzymatic linking with T4 DNA Ligase

mRNA

or T4 RNA Ligase Puromycin

Splint DNA

(b) Long biotin segment puromycin linker (LBP) Puromycin

Pvu II digestion site

Enzymatic linking with T4 RNA Ligase

mRNA

1:4

~ 60 min

Biotin T4 RNA Ligase

(c) Short biotin segment puromycin linker (SBP) Puromycin

RNase T1 digeston site Enzymatic linking

rG

mRNA

rG

1 : 1 ~ 1.5

~ 60 min

1 : 1 ~ 1.5

~ 1 min

with T4 RNA Ligase Biotin

T4 RNA Ligase

(d) cnvK puromycin linker UV (366 nm)

Puromycin mRNA

cnv

K

Photo-crosslinking by irradiation of

rG Biotin

UV (366 nm)

RNase T1 digeston site

Fig. 2 Comparison of preparation methods for the mRNA–puromycin-linker conjugate. (a) Ligation of mRNA and puromycin-linker by splinted ligation, (b) long-biotin-segment puromycin-linker (LBP) linker, (c), shortbiotin-segment puromycin-linker (SBP) linker, and (d) cnvK puromycin-linker by photocrosslinking. In the case of photocrosslinking of mRNA with cnvK puromycin-linker, the mixture ratio of mRNA and cnvK puromycinlinker is 1:1–1.5, and the required time for ligation is about 1 min

(Fig. 2b). The hybridization of the puromycin-linker to mRNA can increase the efficiency of the ligation reaction by T4 RNA ligase because the 50 -terminal of puromycin-linker and 30 -terminal of mRNA are vicinally oriented [3]. Biotin is required to immobilize the mRNA display molecule on a magnetic bead for the following reverse transcription reaction and chemical modification of its displayed polypeptide. In addition, the puromycin-linker contains a reverse transcription primer region; therefore, addition of primers for the reverse transcription reaction is not necessary. The immobilization of mRNA on the magnetic bead is important to wash the bead and exchange buffers easily with an ethylenediaminetetraacetic acid (EDTA)-containing buffer using a magnetic separation apparatus because some ribosomes that interfere with reverse transcription reaction on an mRNA should be removed. Thus, the

46

Hidenao Arai et al. (a)

3’ CC(PEG)(PEG)CTCTCTs

Primer region for RT

Puromycin

EMCS NH

Fluorescein

RNase T1 cleavage site

5’ 3’-TCCTGCCCCCCGCCGKACCTTTAA(rG)AA gaagcaggacggggggcggcguggaaa-3’ Biotin mRNA

5’

(b)

5’

3’ Fig. 3 (a) Structure of cnvK puromycin-linker and (b) cnvK residue. The cnvK puromycin-linker was hybridized with mRNA (indicated by small letters) at the puromycin-linker hybridization region. cnvK (indicated with red background) in puromycin-linker and the uracil residue in mRNA are photocrosslinked by UV irradiation (λ ¼ 366 nm). In the cell-free translation system, the C-terminus of the nascent peptide or protein was conjugated with mRNA via puromycin-linker (indicated by green circle). Biotin moiety (indicated by magenta circle) was utilized for immobilization on SA beads. Reverse transcriptase recognized the 30 -terminal CCT region in the biotin segment and began reverse transcription by using mRNA as a template on the beads. The mRNA/cDNA–POI fusion molecule was released from the beads by cleavage of the 30 -terminus of guanosine (rG) with RNase T1

cDNA display can be rapidly and completely transformed from the mRNA display [4]. Moreover, long-segment-biotin-puromycin (LBP) linker is replaced by downsizing the DNA fragment (i.e., short-segment–biotin-puromycin (SBP) linker) because the cost of DNA synthesis is reduced and the ligation efficiency between an mRNA and the linker is increased (Fig. 2c) [5]. The most substantial recent improvement was achieved by introducing the cnvK base to the linker (Figs. 2d and 3), which decreased the mRNA degradation effectively because the T4 RNA ligase reaction buffer containing cations that cause activation of the contaminated ribonuclease is unnecessary [6]. In addition, the ligation reaction time dramatically decreased after hybridization between mRNA and the puromycin-linker.

Genotype-Phenotype Linking by cDNA Display

47

The puromycin-linker for the cDNA display method could become one of the most powerful tools for directed evolution of functional peptides and proteins. Herein, we describe a method for the preparation of cDNA display molecules for the selection of peptides or proteins of interest (POI).

2

Materials All reagents should be of molecular biology grade to avoid ribonuclease contamination.

2.1 DNA Construct for cDNA Display

1. The DNA construct for the preparation of cDNA display is shown in Fig. 4. The DNA template sequence is 50 -GA TCCCGCGAAATTAATACGACTCACTATAGGGGAAGTA TTTTTACAACAATTACCAACAACAACAACAAACAACAA CAACATTACATTTTACATTCTACAACTACAAGCCACC ATG-[POI]-GGAGGAGGCAGCCATCATCATCATCATCA CGGCGGAAGCAGGACGGGGGGCGGCGTGGAAA-30 , which is composed of a T7 promoter, omega enhancer sequence, Kozak consensus sequence, start codon, gene of POI, GGGS spacer, 6 His-tag, GGS spacer, and puromycin-linker hybridization region from the 50 - to 30 -terminals (see Note 1). 2. Primer 1: 50 -GATCCCGCGAAATTAATACGACTCACTAT AGGG-30 . 3. Primer 2: 50 -TTTCCACGCCGCCCCCCGTCCT-30 . 4. PrimeSTAR® HS DNA Polymerase (TaKaRa). 5. 5 PrimeSTAR® Buffer (Mg2+ plus) (TaKaRa). 6. 2.5 mM for each dNTP mix (TaKaRa). 7. Agencourt AMPure XP (A63882; Beckman Coulter).

2.2 In Vitro Transcription

1. dsDNA (PCR product) prepared in Subheading 2.1. 2. T7 RiboMAX™ Express Large Scale RNA Production System (P1320; Promega). Store at 20  C. 3. Heat block or thermal cycler (various suppliers). 4. Agincourt RNA Clean XP (A66514; Beckman Coulter).

Primer 1 Ω enhancer

ATG Kozak

T7 promoter

Fig. 4 DNA construct for cDNA display

Gene of POI

6 x His

Primer 2

48

Hidenao Arai et al.

2.3 Synthesis of Puromycin-Linker (See Note 2)

The construction of the puromycin-linker is shown in Fig. 3a. 1. Puromycin segment: the DNA oligomer was composed of the following sequence, 50 -(Thiol-Modifier C6S-S)-TC-(Fluorescein-dT)-CTC-(Spacer 18)  2-CC-puromycin-30 . This oligomer can be obtained from a custom DNA synthesis service. The synthesis should be performed on a scale of 1 μmol or more and purification should be done by high-performance liquid chromatography (HPLC). 2. Biotin–cnvK segment: the DNA oligomer was composed of the following sequence, 50 -biotin-AA-(RiboG)-AATTTCCA(cnvK)-GCCGCCCCCCG-(Amino Modifier C6 dT)-CCT-30 . This oligomer can be obtained from a custom DNA synthesis service (Hokkaido System Science). The synthesis should be performed on a scale of 1 μmol and purification should be done by HPLC. 3. 1 M Na2HPO4, pH 9.0: 1 M Na2PO4 in ddH2O, pH rises to 9.0. 4. 200 mM phosphate buffer, pH 7.2. 5. 20 mM phosphate buffer, pH 7.2. 6. 1 M DTT: 1 M dithiothreitol in ddH2O. Store at 20  C. 7. NAP-5 column (GE Healthcare). 8. 100 mM EMCS: dissolve 2 mg of N-(6-maleimidocaproyloxy) succinimide in 65 μL of dimethylformamide (see Note 3). 9. 99.5% Ethanol. 10. 70% Ethanol: prepared by dilution of 99.5% ethanol with ddH2O. 11. Quick-Precip™ Plus Solution (70437; Edge Biosystems) for ethanol precipitation. 12. SYBR™ Gold Nucleic Acid Gel Strain (S11494; Invitrogen). Other reagents can be used. 13. HPLC column: Symmetry 4.6  250 mm (Waters).

300C18

Column,

5

μm,

14. 100 mM TEAA: prepared by dilution of 2 M triethylammonium acetate, pH 7.0 (HPLC grade, various suppliers) with ddH2O (see Note 4). 15. 80% Acetonitrile: prepare by dilution of 100% acetonitrile (HPLC grade, various suppliers) with ddH2O. 16. Vortex mixer (various suppliers). 17. Vacuum centrifuge (e.g., SpeedVac; Thermo Fisher Scientific). 18. Nuclease-free water (various suppliers).

Genotype-Phenotype Linking by cDNA Display

2.4 Photocrosslinking Between mRNA and Puromycin-Linker

49

1. 1 M NaCl prepared with nuclease-free water. 2. 0.25 M Tris–HCl, pH 7.5 prepared with nuclease-free water. 3. mRNA (transcript of DNA in Subheading 2.2). Store at 80  C (see Note 5). 4. Puromycin-linker from Subheading 2.3 (see Note 6). 5. Nuclease-free water. 6. Thermal cycler (various suppliers). 7. CL-1000 ultraviolet (UV) Crosslinker (UVP; Analytik Jena). 8. SYBR™ Gold Nucleic Acid Gel Strain (S11494; Invitrogen). Other reagents can be used. 9. Fluorescence scanner (e.g., Amersham Typhoon Biomolecular Imager; GE Healthcare).

2.5 Preparation of cDNA Display Molecule 2.5.1 In Vitro Translation (Synthesis of mRNA–POI Fusion Molecule)

1. Rabbit Reticulocyte Lysate System, Nuclease Treated (L4960; Promega). Store at 80  C. 2. RNasin® Ribonuclease Inhibitor (N2111; Promega). Store at 20  C. 3. mRNA–puromycin-linker conjugate (photocrosslinked product from Subheading 2.4). Store at 80  C. 4. Nuclease-free water. 5. 3 M KCl prepared with nuclease-free water. 6. 1 M MgCl2 prepared with nuclease-free water. 7. 0.5 M EDTA, pH 8.0 (various suppliers). 8. 2 B/W buffer: 20 mM Tris–HCl, pH 8.0, 2 mM EDTA, 2 M NaCl, and 0.2% Tween20 prepared with nuclease-free water. 9. Heat block (various suppliers).

2.5.2 Purification of mRNA–POI Fusion Molecule from the Lysate with SA Magnetic Beads

1. SA magnetic beads: Dynabeads™ MyOne™ streptavidin (SA) C1 (DB65002; VERITAS). 2. 1 B/W buffer prepared by dilution of 2 B/W buffer with nuclease-free water. 3. 5 RT buffer (TRT-1B; Toyobo). 4. 1 RT buffer prepared by dilution of 5 RT buffer with nuclease-free water. 5. Magnetic separator (e.g., DynaMag™ series; Invitrogen). 6. Temperature-controllable Nissin).

2.5.3 Reverse Transcription (Synthesis of mRNA/cDNA–POI Fusion Molecule)

tube

rotator

(e.g.,

SNP-24B;

1. ReverTra Ace® (TRT-101; Toyobo). Store at 20  C. 2. 2.5 mM each dNTP mix (various suppliers). 3. Nuclease-free water.

50

Hidenao Arai et al.

4. Magnetic separator (e.g., DynaMag™ series; Invitrogen). 5. Temperature-controllable Nissin). 2.5.4 Release of mRNA/ cDNA–POI Fusion Molecule from SA Magnetic Beads

tube

rotator

(e.g.,

SNP-24B;

1. RNase T1 (various suppliers). 2. Ni-NTA B/W buffer: 20 mM sodium phosphate, pH 7.4, 0.5 M NaCl, 5 mM imidazole, 0.05% Tween20. 3. Magnetic separator (e.g., DynaMag™ series; Invitrogen). 4. Temperature-controllable Nissin).

2.5.5 His-Tag Affinity Purification of cDNA Display (mRNA/cDNA–POI Fusion Molecule)

tube

rotator

(e.g.,

SNP-24B;

1. His Mag Sepharose Ni (2896390; GE Healthcare). 2. Ni-NTA B/W buffer: 20 mM sodium phosphate, pH 7.4, 0.5 M NaCl, 5 mM imidazole, and 0.05% Tween20. 3. Ni-NTA elution buffer: 20 mM sodium phosphate, pH 7.4, 500 mM NaCl, 250 mM imidazole, and 0.05% Tween20. 4. Magnetic separator (e.g., DynaMag™ series; Invitrogen). 5. Temperature-controllable Nissin).

2.5.6 Buffer Exchange of cDNA Display Solution

tube

rotator

(e.g.,

SNP-24B;

1. Micro Bio-Spin® Columns with Bio-Gel® P-6, Tris buffer: 10 mM Tris pH 7.4 (7326221; Bio-Rad). 2. Phosphate-buffered saline (PBS): 81 mM Na2HPO4, 1.47 mM KH2PO4, 2.68 mM KCl, and 137 mM NaCl, pH 7.4. 3. Tris-buffered saline (TBS): 50 mM Tris–HCl, 150 mM NaCl, pH 7.4. 4. TE buffer: 10 mM Tris–HCl, 1 mM EDTA, pH 8.0.

3

Methods A schematic illustration of the cDNA display method, i.e., the preparation of the cDNA display and its affinity selection against a target molecule, is shown in Fig. 1c. The preparation procedure for the cDNA display molecule is as follows: (1) The DNA template encoding POI is transcribed into mRNAs. (2) The mRNA is ligated by UV irradiation with the puromycin-linker to form the mRNAlinker fusion molecule. (3) The mRNA-linker fusion molecules translated to synthesize mRNA-linker-POI fusion molecules (termed as “mRNA display” molecules) and immobilized onto SA magnetic beads. (4) The mRNA moiety of the immobilized mRNA display molecules is reverse-transcribed to synthesize the mRNA/ cDNA-linker-POI fusion molecule (i.e., cDNA display molecule). The cDNA display molecule is released from the SA beads by treatment with RNase T1. (5) The cDNA display molecule is

Genotype-Phenotype Linking by cDNA Display

51

purified using the Ni-NTA magnetic bead to eliminate the mRNA/ cDNA-linker fusion molecule. The cDNA display is then brought to the affinity selection against the target molecule. 3.1 Preparation of dsDNA Construct for cDNA Display

1. Prepare the PCR solution as follows: 50–300 pg of DNA template. 1 μL of 10 μM Primer 1. 1 μL of 10 μM Primer 2. 10 μL of 5 PrimeSTAR® Buffer (Mg2+ plus). 4 μL of 2.5 mM dNTP mix (supplied with PrimeSTAR®). 1.25 U of PrimeSTAR® HS DNA Polymerase. Add ddH2O up to 50 μL. 2. The PCR program is as follows: 98  C for 2 min, repeat 25 cycles of three steps, 98  C for 10 s, 68  C for 20 s, 72  C for 1 min/kb, followed by 72  C for 5 min. 3. Purify the PCR solution with Agencourt AMPure Clean XP according to the manufacturer’s protocol and determine its concentration by absorbance at 260 nm. 4. Store the dsDNA at 20  C until use.

3.2 In Vitro Transcription

1. Prepare the transcription reaction solution as follows: 10 μL of RiboMAX™ Express T7 2 buffer. 200 ng of the dsDNA (from Subheading 3.1, the sequence shown in Subheading 2.2). 2 μL of Enzyme Mix T7 Express. Add nuclease-free water up to 20 μL (see Note 7). 2. Incubate at 37  C for 30 min (see Note 8). 3. Add 1 μL of RQ-1 RNase-free DNase and incubate at 37  C for 15 min. 4. Purify the transcript (mRNA) with Agencourt RNA Clean XP according to the manufacturer’s protocol and determine its concentration by absorbance at 260 nm. The concentration of mRNA should be over 2 pmol/μL and the ratio of 260/280 nm should be over 1.8. 5. Store the mRNA at 20  C or 80  C until use.

3.3 PuromycinLinker Synthesis 3.3.1 Reduction of the Puromycin Segment

1. Prepare a 2 mM solution of the puromycin segment in ddH2O. 2. Mix 10 μL of 2 mM puromycin segment (20 nmol) with 22.5 μL of 1 M Na2HPO4, pH 9.0, and 2.5 μL of DTT in a 1.75 mL tube. 3. React for 1 h at room temperature with agitation using a vortex mixer.

52

Hidenao Arai et al.

4. Pre-equilibrate a NAP-5 column with 9 mL of 20 mM phosphate buffer, pH 7.2. 5. Apply the reacted mixture to the NAP-5 column and discard the flow-through. 6. Add 450 μL of 20 mM phosphate buffer, pH 7.2, onto the NAP-5 column and discard the flow-through. 7. Add 1 mL of 20 mM phosphate buffer, pH 7.2, onto the NAP-5 column and collect the fluorescein fraction (see Note 9). 3.3.2 EMCS Modification of the Biotin cnvK Segment

1. Prepare a 1 mM solution of the biotin cnvK segment in ddH2O. 2. Mix 10 μL of biotin cnvK segment (10 nmol) with 50 μL of 200 mM phosphate buffer, pH 7.2, and 10 μL of 100 mM EMCS in a 1.75 mL tube. 3. Incubate the mixture for 30 min at 37  C (see Note 10). 4. Perform ethanol precipitation with the Quick-Precip™ Plus Solution. The precipitated pellets are used in the next step.

3.3.3 Cross-Linking of the Puromycin and Biotin cnvK Segments

1. Dissolve the precipitated pellet from Subheading 3.3.1 in the collected solution from Subheading 3.3.2 to crosslink the puromycin segment and biotin cnvK segment. 2. Incubate the mixture overnight at 4  C. 3. Add 1/20 volume 1 M DTT and react for 30 min at room temperature using a vortex mixer. 4. Perform ethanol precipitation with the Quick-Precip™ Plus Solution to remove the unreacted puromycin segment. Dissolve the pellet in 50 μL of ddH2O. The precipitated product should be analyzed on a denaturing 12% polyacrylamide gel electrophoresis (PAGE) containing 8 M urea by imaging the fluorescence of fluorescein isothiocyanate (FITC) derived from the puromycin segment using a fluorescence scanner and subsequently strained with the SYBR™ Gold Nucleic Acid Gel Strain and visualized by a fluorescence scanner (see Note 11).

3.3.4 HPLC Purification of the Puromycin-Linker

1. Perform the HPLC purification of the product to remove the unreacted biotin cnvK segment. The HPLC conditions are as follows: column, Waters Symmetry 300C18, 4.6  250 mm, particle size 5 μm; solvent A, 100 mM TEAA; solvent B, acetonitrile/water (80:20, v/v); gradient, B/A (15–36%, 30 min); flow rate, 0.5 mL/min; and detection, absorbance at 260 and 490 nm. 2. Dry the collected sample using a vacuum centrifuge and dissolve with the appropriate volume (30–50 μL) of nuclease-free water. The dissolved sample should be analyzed on a

Genotype-Phenotype Linking by cDNA Display

53

denaturing 12% PAGE containing 8 M urea by imaging the fluorescence of FITC derived from a puromycin segment using a fluorescence scanner and subsequently strained with the SYBR™ Gold Nucleic Acid Gel Strain and visualized by the fluorescence scanner. 3. Perform ethanol precipitation with the Quick-Precip™ Plus solution on the whole purified sample and add 50 μL of nuclease-free water. 4. Determine the concentration of the product by measuring the absorbance at 260 nm. The yield is usually about 15–20% of the biotin cnvK segment. 5. Store the prepared puromycin-linker at 20  C. 3.4 Photocrosslinking Between mRNA and the PuromycinLinker

The final yield of cDNA display harboring POI is about 10% of the photocrosslinked product that was put into the translation reaction. 1. Prepare the photocrosslinking reaction solution equal to the volume required for screening as follows: mix 4 μL of 1 M NaCl, 4 μL of 0.25 M Tris–HCl, pH 7.5, 20 pmol of puromycin-linker from Subheading 3.1, 20 pmol of mRNA from Subheading 3.2, and nuclease-free water up to 20 μL in each tube. 2. Incubate at 94  C for 1 min and then reduce to 25  C for over 45 min using a thermal cycler. 3. Irradiate with UV light at 365 nm using CL-1000 UV Crosslinker for about 2 min (405 mJ/mm). The photocrosslinked product should be analyzed on a denaturing 4% PAGE containing 8 M urea by imaging the fluorescence of FITC derived from puromycin-linker using a fluorescence scanner and subsequently strained using the SYBR™ Gold Nucleic Acid Gel Strain and visualized by the fluorescence scanner.

3.5 Preparation of cDNA Display 3.5.1 In Vitro Translation (Synthesis of mRNA–POI Fusion Molecule)

1. Prepare the photocrosslinking reaction solution equal to the volume required for screening as follows: mix 6 pmol of the photocrosslinked product from Subheading 3.4, 35 μL of the nuclease-treated rabbit reticulocyte lysate, Nuclease Treated, 0.5 μL of amino acid mixture minus leucine, 0.5 μL of amino acid mixture minus methionine, 1 μL of RNasin® Ribonuclease Inhibitor, and nuclease-free water up to 50 μL (see Note 12). 2. Incubate the mixture at 30  C for 20 min. 3. Add 24 μL of 3 M KCl and 6 μL of 1 M MgCl2 to each tube and incubate at 37  C for 60 min. 4. Add 18 μL of 0.5 M EDTA and 98 μL of 2 binding buffer and incubate at 4  C for 10 min.

54

Hidenao Arai et al.

5. Analyze the formation of mRNA–POI fusion molecules on a denaturing SDS-PAGE containing 8 M urea by imaging the fluorescence of FITC derived from the puromycin-linker using a fluorescence scanner. 3.5.2 Purification of mRNA–POI Fusion Molecule from the Lysate with SA Magnetic Beads

1. Wash 60 μL of SA magnetic beads twice with 200 μL of B/W buffer (see Notes 13 and 14). 2. Suspend the beads with the translation reaction sample. If the reaction volume is too large for one tube, this solution can be prepared in several tubes. 3. Incubate the mixture using a rotator at 25  C for 60 min (see Note 15). 4. Place the tube on a magnetic separator for 1 min and discard the supernatant, wash the beads three times with 200 μL of 1 binding buffer and once with 200 μL of 1 RT buffer.

3.5.3 Reverse Transcription (Synthesis of mRNA/cDNA-POI Fusion Molecule)

1. Prepare the reverse transcription reaction solution for 60 μL of SA beads as follows: mix 10 μL of 5 RT buffer, 20 μL of 2.5 mM each dNTP mix, 19 μL of nuclease-free water, and 1 μL of ReverTra Ace® (100 U/μL). 2. Suspend the SA beads from Subheading 3.5.2 in the above reverse transcription reaction solution. Incubate at 42  C for 60 min using a rotator (see Note 15). 3. Place the tube on a magnetic separator for 1 min and discard the supernatant, wash the beads twice with 200 μL of 1 binding buffer and once with 200 μL of Ni-NTA B/W buffer.

3.5.4 Release of mRNA/ cDNA–POI Fusion Molecule from SA Beads

1. Suspend the above beads (Subheading 3.5.3) in 99.5 μL of Ni-NTA B/W buffer nd 0.5 μL of RNase T1 (1000 U/μL). Incubate using the rotator at 37  C for 15 min. 2. Place the tube on a magnetic separator for 1 min and collect the supernatant.

3.5.5 His-Tag Affinity Purification of mRNA/ cDNA–POI Fusion (i.e., cDNA Display) Molecules

1. Add 20 μL of the His Mag Sepharose Ni into the tube and wash the beads twice with 200 μL of the Ni-NTA B/W buffer (see Note 14). 2. Suspend the beads with the mRNA/cDNA–POI fusion molecule solution from Subheading 3.5.4 and incubate using the rotator at 25  C for 60 min (see Note 15). 3. Place the tube on a magnetic separator for 1 min and discard the supernatant. Wash the beads twice with 200 μL of Ni-NTA B/W buffer. 4. Suspend the beads in 20 μL of Ni-NTA elution buffer and incubate using the rotator at 25  C for 20 min (see Note 16). 5. Place the tube on a magnetic separator for 1 min and collect the supernatant.

Genotype-Phenotype Linking by cDNA Display 3.5.6 Buffer Exchange of cDNA Display Molecules

55

1. Equilibrate Micro Bio-Spin® Columns with Bio-Gel® P-6 with the appropriate buffer (such as PBS, TBS or TE buffer) according to the manufacturer’s protocol. 2. Load the cDNA display molecules prepared in Subheading 3.5.5 and collect the cDNA display in the appropriate buffer according to the manufacturer’s protocol. 3. Store at 4  C (see Note 17).

4

Notes 1. The efficiency of cDNA display formation depends on the base length of POI; i.e., as the length of POI increases, the efficiency of cDNA display formation decreases. 2. Academic researchers can purchase the puromycin-linker from Epsilon Molecular Engineering (Japan). 3. Prepare the EMCS solution just before use because the reactive groups, i.e., maleimide and NHS ester, are easily hydrolyzed in an aqueous solution. 4. Degassing of 100 mM TEAA and 80% acetonitrile before use is required to eliminate air bubbles in the HPLC system. 5. The working solution for the mRNA can be prepared conveniently using 10–20 pmol/μL. 6. The working solution of the puromycin-linker can be prepared conveniently using 20 pmol/μL. 7. It is necessary to dissolve the precipitate in the RiboMAX™ Express T7 2 buffer sufficiently. 8. The amount of mRNA depends on the incubation time. 9. Proceed to the next step (Subheading 3.3.3) as soon as possible because the reduced thiols group in the puromycin segment can form disulfide bonds with each other. 10. Do not allow the solution to react for over 30 min because a longer reaction time causes hydrolysis of the maleimide group in EMCS. 11. The crosslinked product will migrate to the same point as 40–50 bp double-stranded DNA in a denaturing PAGE. 12. The reaction volume should be less than 50 μL because of the higher translation efficiency. If the reaction is required to be scaled up, 50 μL of the reaction solution at most should be prepared in several test tubes. 13. Ten microliters of SA magnetic beads are required for 1 pmol of the photocrosslinked product.

56

Hidenao Arai et al.

14. The beads are washed as follows: After herding the beads at the bottom of test tube with magnetic stand, remove the supernatant and add the appropriate buffer and suspend by pipetting. 15. Adjust the speed of rotation to keep the beads in suspension. 16. The volume of elution buffer can be adjusted as needed if the beads can be suspended. 17. The final product can be used for the selection against a target of interest. After the selection round, perform PCR (or Error prone PCR, if necessary) to reconstruct the DNA template using the pair of primers (Primer 2 and 3). Primer 3: 50 -GATCCCGCGAAATTAATACGACTCAC TATAGGGGAAGTATTTTTACAACAATTACCAACA-30 . References 1. Nemoto N, Miyamoto-Sato E, Husimi Y, Yanagawa H (1997) In vitro virus: bonding of mRNA bearing puromycin at the 30 -terminal end to the C-terminal end of its encoded protein on the ribosome in vitro. FEBS Lett 414:405–408 2. Roberts RW, Szostak JW (1997) RNA-peptide fusions for the in vitro selection of peptides and proteins. Proc Natl Acad Sci U S A 94:12297–12302 3. Nishigaki K, Taguchi K, Kinoshita Y, Aita T, Husimi Y (1998) Y-ligation: an efficient method for ligating single-stranded DNAs and RNAs with T4 RNA ligase. Mol Divers 4:187–190 4. Yamaguchi J, Naimuddin M, Biyani M, Sasaki T, Machida M, Kubo T, Funatsu T, Husimi Y,

Nemoto N (2009) cDNA display: a novel screening method for functional disulfide-rich peptides by solid-phase synthesis and stabilization of mRNA-protein fusions. Nucleic Acids Res 37:e108 5. Mochizuki Y, Biyani M, Tsuji-Ueno S, Suzuki M, Nishigaki K, Husimi Y, Nemoto N (2011) One-pot preparation of mRNA/cDNA display by a novel and versatile puromycin-linker DNA. ACS Comb Sci 13:478–485 6. Mochizuki Y, Suzuki T, Fujimoto K, Nemoto N (2015) A versatile puromycin-linker using cnvK for high-throughput in vitro selection by cDNA display. J Biotechnol 212:174–180

Chapter 4 cDNA Display of Disulfide-Containing Peptide Library and In Vitro Evolution Tai Kubo and Mohammed Naimuddin Abstract Directed in vitro evolution (IVE) is now a widely applied technology to obtain molecules that have designed new features of one’s demands. We describe here experimental procedures for a cDNA display IVE to select peptide aptamers from a scaffold-based random peptide library. A three-finger (3-F) peptide library is exemplified, which has been shown its pluripotency to various target molecules. Peptide scaffolds including 3-F are refined through evolution, and they are mostly stabilized by disulfide bridges. To utilize such disulfide-containing protein library in IVE, we optimized the translation and folding conditions. Co-translational folding assisted by protein disulfide isomerase was found to have better efficiency than posttranslational refolding in the IVE. Linker is also a key element to make a tight genotype–phenotype linkage. Here, we introduced a whole procedure of IVE to use a newly designed puromycin linker, which was synthesized by a novel branching strategy. The improved linker enabled rapid and highly efficient ligation of mRNA and synthesis of protein fusions. Key words Directed evolution, cDNA display, Puromycin linker, Protein disulfide isomerase (PDI), Three-finger (3-F) scaffold

1

Introduction Along with the advancement of molecular biological technologies, natural proteins and peptides have been engineered toward desired properties and functions. Among various approaches, a technology called directed protein evolution [1], or laboratory-directed protein evolution [2], has now emerged as a powerful technology platform to efficiently develop novel proteins. Starting from large libraries of random and semi-random sequences, proteins of the desired characteristics are isolated through repeated cycles of screening/selection and amplification. It enables a protein to “evolve” in desired directions and provides us with valuable molecular tools in research, high-performance enzymes in industry

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2070, https://doi.org/10.1007/978-1-4939-9853-1_4, © Springer Science+Business Media, LLC, part of Springer Nature 2020

57

58

Tai Kubo and Mohammed Naimuddin

Fig. 1 Processes for directed in vitro evolution. Schematic representation of the directed in vitro evolution processes. Three-finger peptide library is exemplified as the starting random peptide library. Elements of the process are transcription (cDNA to mRNA), linker ligation (mRNA-linker), translation (mRNA to protein), reverse transcription (cDNA display), selection (selected protein conjugated with cDNA), and amplification (selected cDNAs)

and/or bioprocessing, and diagnostic and therapeutic drugs in pharmaceutical sciences. One of the key features of the technology is the linkage between genotype (RNA/DNA) and phenotype of the protein. Proteins generated from RNA/DNA library are subjected to selection by their properties (e.g., physicochemical, biochemical, and/or pharmacological criteria), and then the genetic information corresponded to the selected proteins being proceeded to the next round of process; amplification, protein synthesis, and selection (see Fig. 1). Such genotype and phenotype linkage could be achieved by either chemical/physical conjugation of DNA/RNA to protein (e.g., puromycin [3, 4], biotin-streptavidin affinity, etc.), or by entrapping the individual DNA/RNA and its derived protein in a limited space (e.g., micro-capsule [5], emulsion [6, 7], liposome [8], etc.). Design of the initial polypeptide library is also critical due to direct influences on the peptide conformation and local chemical atmosphere to coordinately interact with target molecule(s). Linear polypeptide libraries generally do not yield high-affinity ligands to

in vitro Evolution from a Cys-Constrained Peptide Library

59

proteins [9, 10], while some exceptions are reported (polypeptide aptamers to streptavidin, Kd ¼ 5–10 nM, [11]). Contrarily, various scaffold-based random peptide libraries, which were often referred to natural polypeptide scaffolds, and rationally designed (or de novo designed) libraries have been used for in vitro evolution (see reviews [12, 13]). Among the proteins with the certain scaffold, proteins constrained by disulfide bridges (S–S) are an interesting and important class of proteins performing various functions, particularly in the offense/defense system [14, 15]. Plant and animal toxins have been in a growing interest in pharmacological uses [16, 17]. Peptide toxins consist of 10–80 amino acid residues, are resistant to proteolysis and low immunogenic and they have a knotted structure that is stabilized by three to five disulfide linkages that impart thermal stability. These properties also make them more suitable to be developed as drugs due to increased bioavailability [16]. An interesting and potentially intriguing aspect of such proteins is the diversity within the same species; for example, snake venom contains hundreds of toxins that have diverse functions [15, 18, 19]. The diversity is thought to be generated from the accelerated evolution of the binding loops of these proteins to adapt to diverse conditions, including environment and preys [15]. Therefore, we can learn from nature that such a protein/ protein scaffold could be used as library template for in vitro evolution directed to some defined purpose. For example, a three-finger (3-F) scaffold library (see Note 1), originated from a snake α-neurotoxin, was shown its pluripotency against various targets such as interleukin-6 receptor [20], serine proteases [21], vascular endothelial growth factor (VEGF) [4], PD-1 (Tada et al., personal communication), survivin (Nemoto et al., personal communication), and acetylcholine binding protein (AChBP) (Kubo et al., manuscript in preparation). Displayed proteins should be properly folded so that they show their innate/intrinsic functions. Folding and refolding of disulfiderich proteins often require chaperones such as protein disulfide isomerase (PDI) and thioredoxin. In this chapter, we present a detailed description of the current methodology for a cDNA display and in vitro evolution of disulfide-rich 3-F polypeptides. The methodology has been improved (1) by modifying a new puromycin linker-oligonucleotide for rapid and efficient generation of templates and synthesis of protein fusions (see Note 2), and (2) by optimizing PDI conditions to fold and refold for the functional integrity of this protein. The introduced method is expected to benefit for displaying disulfide-containing protein libraries to exert their potentials in directed protein evolution. Whole processes for in vitro evolution are illustrated in Fig. 1. It could be shortened less than 8 h for one cycle process (i.e., amplification, protein display, and selection).

60

2

Tai Kubo and Mohammed Naimuddin

Materials Use Milli-Q ultrapure, DNase- and RNase-free water for making all solutions.

2.1 Three-Finger (3-F) Peptide Library

Materials required to prepare a 3-F peptide library. The library is constructed by ligation of long oligonucleotides. 1. Oligonucleotides to construct a 3-F peptide library are listed in Table 1 (see also Fig. 2a). 2. T4 polynucleotide kinase (10 U/μL, Toyobo). 3. 10 mM ATP. 4. T4 DNA ligase (4 U/μL, Toyobo). 5. TE: 10 mM Tris–HCl, pH 8.0, 1 mM EDTA. 6. Phenol/chloroform (1:1). 7. Chloroform. 8. Glycogen (20 mg/mL, Wako).

Table 1 Oligonucleotides used for the construction of a 3-F cDNA library and in vitro evolution F1_S: ATGGGAGGTTCACTTGTATGTTAC(NNB)6CCTGGAACCTTAGAGACTTGTCCAGAT F2_S: GATTTCACATGTGTT(NNB)10ACCCAGTATTGTTCTCAT F3_S: GCGTGTGCGATACCG(NNB)7TGTTGCCAAACAGACAAATGCAACGGA Splint F1/F2_A: ATGTGAAATCATCTGGACAAGTC Splint F2/F3_A: TCGCACACGCATGAGAACAATA SP6-Full_S: ATTTAGGTGACACTATAGAATACAAGCTTGCTTGTTCTTTTTGCAGAAGCTCAGAA TAAACGCTCAACTTTGGCAGATCTACCATGGGAGGTTCACTTGTATGTTAC His-Full_A: TTTCCCCGCCGCCCCCCGCCCTTGTCCGCCGTGATGATGATGGTGGTGAGACCC TCCGCCTGAGCCTCCACCTCCGTTGCATTTGTCTGTTTG P_S: ATTTAGGTGACACTATAGAATACAAG P_A: TTTCCCCGCCGCCCCCCG Nucleotide sequences are shown in the 50 - to 30 -end direction. F1_S, F2_S, and F3_S are based on the cDNA sequence of the three-finger polypeptide MicTx3 [20]. Randomized sequences indicated as (NNB)s are incorporated around each fingertip. “N” denotes G, A, T, and C. “B” denotes “G”, “T,” and “C”

in vitro Evolution from a Cys-Constrained Peptide Library

61

Fig. 2 Three-finger peptide library and Puromycin linker. (a) 3-F cDNA library and the adjacent sequence elements for in vitro evolution are illustrated. A box shows the 3-F protein (mature) coding region. Three hatched boxes are randomized regions corresponding to each fingertip when the protein properly folded. Arrows are synthetic oligonucleotides used to construct the library. The oligonucleotide sequences are listed in Table 1. (b) Schematic representation of the designed linker for cDNA display and in vitro evolution. “Annealing region”, the sequences tightly anchors the linker to mRNA; “ligation site” for ligation of 50 -end of the linker to mRNA by T4 RNA ligase; “puromycin” with C18 spacers forms a flexible arm for covalent linkage of the protein; “oligo dA” for a longer puromycin arm and purification of protein fusions (optional) using oligo-dT matrices; “FITC” to track and detect the linker-conjugated entities during IVE; “RT priming site”, the sequence 50 -CCTTG-30 (blue) is a primer for reverse transcriptase to make a cDNA. The primer is branched from the linker body, including “annealing region” and “puromycin arm”, using a brancher nucleotide 5-Me-dC (indicated “C” in red). (Referred to and modified from [4])

9. 3 M Na acetate, pH 5.2. 10. Ethanol. 11. Ligation high ver. 2 (Toyobo). 12. 8 M urea-6% PAGE: per 10 mL, 4.8 g urea, 2 mL 5 TBE, 1.5 mL 38% acrylamide/2% bis-acrylamide, 50 μL 10% ammonium persulfate, water to 10 mL. Dissolve completely and

62

Tai Kubo and Mohammed Naimuddin

filtrate through 0.45 μm filter, add 5 μL TEMED, and pour into the gel casting mold (pre-assembled paired glass plates, 90  80  1 mm). Wait for polymerization for at least 2 h at room temperature (see Note 3). 13. 2 Sample loading buffer for denaturing gel: per 1 mL, 900 μL deionized formamide, 8 μL 0.5 M EDTA, pH 8.0, 92 μL 0.5% bromophenol blue. 14. 5 TBE: per 1 L, 53.9 g Tris base, 27.55 g boric acid, 20 mL 0.5 M EDTA, pH 8.0, water to 1 L. 15. SYBR Gold (10,000 concentration, ThermoFisher). 16. Filter spin cups: SPIN-X (pore size 0.22 μm, Corning). 2.2 Transcription and Translation

The following reagents are used to prepare a protein-mRNA conjugate library. 1. Oligonucleotides SP6-Full_S, His-Full_A, P_S and P_A (Table 1 and Fig. 2a). 2. Ex Taq (Takara): 5 U/μL Ex Taq, 10 Ex Taq Buffer, 2.5 mM dNTP mix. 3. 1.2% Agarose gel: 1.2 g Agarose S (Nippon Gene), 100 mL 1 TAE. 4. QIAquick PCR Purification Kit (Qiagen). 5. MEGAscript SP6 Kit (Ambion): 50 mM ATP, 50 mM CTP, 50 mM UTP, 50 mM GTP, 10 SP6 Reactions Buffer, SP6 Enzyme Mix, and 2 U/μL TURBO DNase. 6. 40 mM Cap Analog (m7G(50 )ppp(50 )G, Ambion). 7. RNeasy Mini Kit (Qiagen). 8. Puromycin linker (see Fig. 2b and Note 2). 9. T4 polynucleotide kinase (10 U/μL, Toyobo). 10. T4 RNA ligase (10 U/μL, NEB). 11. Wheat Germ Extract Translation Kit (Promega): Wheat Germ Extract, 1 mM Amino acid mix (Met), 1 mM Amino acid mix (Leu), 1 M K acetate. 12. RNasin Plus (40 U/μL, Promega). 13. High Salt Mixture: 47 μL 1 M MgCl2, 183 μL 3 M KCl. 14. 5 mg/mL Protein Disulfide Isomerase (Takara). 15. 250 mM Oxidized Glutathione. 16. 10 mM Reduced Glutathione. 17. Laemmli urea-SDS-PAGE (see Note 4): 6% resolving gel, 4% stacking gel, SDS-PAGE running buffer, Urea-containing sample loading buffer.

in vitro Evolution from a Cys-Constrained Peptide Library

2.3 Protein Trap and Reverse Transcription

63

Reagents required for trapping full-length translated proteins, and converting mRNA to DNA. 1. BSA-coated microcentrifuge tubes (see Note 5). 2. HisPur Ni-NTA Magnetic Beads (ThermoFisher). 3. Equilibration Buffer: 100 mM Na phosphate buffer, pH 8.0, 600 mM NaCl, 30 mM imidazole, 0.05% Tween 20. 4. Wash Buffer: 100 mM Na phosphate buffer, pH 8.0, 600 mM NaCl, 50 mM imidazole, 0.05% Tween 20. 5. Elution Buffer: 100 mM Na phosphate buffer, pH 8.0, 600 mM NaCl, 250 mM imidazole. 6. 10 mM dNTP mix. 7. ReverTra Ace (100 U/μL, Toyobo). 8. 5 ReverTra Ace Buffer (Toyobo). 9. 5 RT Buffer (DTT): 250 mM Tris–HCl, pH 8.0, 375 mM KCl, 15 mM MgCl2. 10. 5 RT Buffer (2:3): 2 volumes 5 ReverTra Ace Buffer, 3 volumes 5 RT Buffer (DTT). 11. RNasin Plus (40 U/μL, Promega). 12. 10 Phosphate buffered saline (PBS):1370 mM NaCl, 26.8 mM KCl, 14.7 mM KH2PO4, 81 mM Na2HPO4. 13. 5 Selection Buffer: per 50 mL, 25 mL 10 PBS, 5 mL 5 M NaCl, 1.25 mL 1 M MgCl2, 5 mL 5% Tween 20, 13.75 mL water. 14. Wash Buffer: per 50 mL, 10 mL 5 Selection Buffer, 0.5 mL 5 M NaCl, 39.5 mL water. 15. Buffer exchange spin columns: Micro Bio-Spin® Columns with Bio-Gel® P-30 (Bio-Rad).

2.4 Selection by Binding

The following section lists the materials that is used for the selection by binding affinity to the target molecule, acetylcholine binding protein (AChBP). AChBP was labeled with biotin and immobilized on streptavidin-coated magnetic beads. 1. 10 Phosphate buffered saline (PBS) (see Subheading 2.3). 2. 5 Selection Buffer (see Subheading 2.3). 3. Wash Buffer (see Subheading 2.3). 4. AChBP (see Note 6). 5. Biotin Labeling Kit –NH2, (Dojindo): WS Buffer, NH2-Reactive Biotin, Reaction Buffer. 6. Filter spin cups: Amicon Ultra-0.5 10 kDa (Merck). 7. Streptavidin Mag Sepharose (GE Healthcare). 8. 10 mg/mL BSA. 9. 100 mM DTT.

64

3

Tai Kubo and Mohammed Naimuddin

Methods All the procedures are conducted under RNase-free conditions.

3.1 Construction of a 3-F cDNA Library

3.1.1 Phosphorylation of the 50 -End of Oligonucleotides F2_S and F3_S (See Note 8)

Three long oligonucleotides (F1_S 69-mer, F2_S 63-mer, and F3_S 63-mer nucleotides, Table 1), which are adjacent and cover the entire coding region of the 3-F mature protein, are enzymatically joined by ligase (see Note 7 and Fig. 2a). 1. Add the following per each oligonucleotide F2_S and F3_S. 100 pmol/μL 50 -OH-oligonucleotide

1 μL

10 T4 polynucleotide kinase buffer

3 μL

10 mM ATP

3 μL

10 U/μL T4 polynucleotide kinase

1 μL

Make up to 30 μL with water, and incubate for 1 h at 37  C. 2. Add 70 μL TE, and extract DNA by vigorous mixing with an equal volume (100 μL) of phenol/chloroform (1:1) and centrifuging for 5 min at room temperature. 3. Repeat the extraction by chloroform. 4. Transfer the aqueous phase (90–100 μL) to a new test tube and add 1 μL 20 mg/mL glycogen, 10 μL 3 M Na acetate, pH 5.2, 250 μL ethanol, mix, and chill. 5. Precipitate DNA by centrifugation at 16,000  g for 15 min at 4  C. 6. Rinse the pellet with 70% ethanol, and dry. 7. Dissolve the DNA in 5 μL water (∗F2_S and ∗F3_S; ∗ denotes “50 -phosphorylated”). 3.1.2 Assembling Oligonucleotides

1. Mix 50 pmol each of F1_S, ∗F2_S and ∗F3_S, and 100 pmol each of Splint F1/F2_A and Splint F2/F3_A in total volume 5 μL. 2. Heat the mixture at 70  C for 10 min in a block incubator, then switch off the incubator for gradual cooling. When it cooled down to room temperature, spin down and sit on ice. 3. Add 5 μL Ligation high ver.2 (Toyobo). 4. Incubate for 1 h at 16  C (or 4  C for overnight), and stop the reaction by incubating for 10 min at 65  C. 5. Separate the assembled ssDNA by 8 M urea-6% PAGE (see Note 3). 6. Stain the gel by SYBR Gold, and cut out the band of the expected size using clean scalpel and forceps. 7. Mash the gel pieces to paste using a disposable pestle, disperse in 300 μL TE, and mix vigorously.

in vitro Evolution from a Cys-Constrained Peptide Library

65

8. Sonicate for 5 min and mix in a microtube mixer for 10 min at 65  C. Repeat this step three times. 9. Transfer the suspension to a filter spin cup (e.g. SPIN-X) and filtrate by centrifugation at 16,000  g for 5 min. 10. From the filtrate recover the single-stranded DNA (ssDNA) by ethanol precipitation with glycogen as steps 4–6 of Subheading 3.1.1. 11. Dissolve the DNA in 10 μL water. Use 1 μL for quantification, and add 1 μL 10 TE to the remaining for storage. For longterm storage, keep the ssDNA below 20  C. 3.2 mRNA Library, Linker Ligation, and 3F Polypeptide Library

Sequences required for IVE processes, including RNA polymerase promoter and tag sequences, are introduced by PCR (primers, SP6-Full_S and His-Full_A) to the 3-F cDNA library (Table 1 and Fig. 2a).

3.2.1 3-F mRNA Library Generation

1. Template DNA (3-F cDNA library) for transcription is prepared by the following 2 steps of PCR. The first PCR Assembled ssDNA

150–300 fmol (ca. 10–20 ng for 3-F ssDNA)

2 pmol/μL SP6-Full_S 1 μL 2 pmol/μL His-Full_A 1 μL 10 Ex Taq buffer

3 μL

2.5 mM dNTP mixture 1.2 μL 5 U/μL Ex Taq

0.15 μL

Make up to 30 μL with water and proceed to the following cycle reactions. 2. Thermal cycle program: 95  C, 3 min. (94  C, 30 s; 55  C, 30 s; 72  C, 30 s)  10 cycles. 72  C, 5 min. 12  C, hold. 3. The second PCR: Add the followings to the first PCR tube. 2 pmol/μL P_S

2 μL

2 pmol/μL P_A

2 μL

10 Ex Taq buffer

2 μL

2.5 mM dNTP mixture

0.8 μL

5 U/μL Ex Taq

0.1 μL

Water

13.1 μL

66

Tai Kubo and Mohammed Naimuddin

Mix and proceed to the following cycle reactions. 4. Thermal cycle program: 95  C, 3 min. (94  C, 30 s; 55  C, 30 s; 72  C, 30 s)  20 cycles. 72  C, 5 min. 12  C, hold. 5. Check an aliquot (2–3 μL) by 1.2% agarose gel electrophoresis in TAE buffer (see Note 9). 6. Purify the PCR product by QIAquick PCR Purification Kit. 7. Take an aliquot and estimate the concentration of the template DNA solution by measuring the absorbance at 260 nm. 8. Synthesize mRNA (cRNA) using mRNA MEGAscript Kit (Ambion) following the manufacturer’s instructions with some modifications. All the reagents, once thawed, keep at room temperature and mix in the following order. Take 200–300 ng template DNA from step 7, dissolve in water to total 6 μL, incubate at 65  C for 5 min and cool down to 25  C. Add the following (∗, included in the Kit), 50 mM ATP∗

2 μL

50 mM CTP∗

2 μL

50 mM UTP∗

2 μL

10 mM GTP (dilute 50 mM GTP∗)

2 μL

40 mM cap analog

2 μL

10 SP6 reactions buffer∗

2 μL

Mix up the reagents and lightly spin down at room temperature and add SP6 Enzyme Mix∗

2 μL

Mix gently by tapping and incubate for 2 h at 37  C. 9. To terminate the reaction, add 1 μL Turbo DNase∗ and incubate for 15 min at 37  C. 10. Purify mRNA using RNeasy Mini Kit. Elute the RNA from the column twice with 50 μL water warmed up at 40  C. 11. Take an aliquot (2–3 μL) and check the quality and quantity of the RNA (see Note 10). The rest of the RNA solution (3-F mRNA library, ~100 μL) should be stored at 80  C, or for long-term storage mix the RNA solution with 10 μL 3 M Na acetate (pH 5.2) and 250 μL ethanol and store at 80  C.

in vitro Evolution from a Cys-Constrained Peptide Library 3.2.2 Linker Ligation

67

In the following steps, the puromycin-linker is annealed to the 30 -side of mRNA, and then phosphorylation and ligation of the linker with mRNA proceed simultaneously. 1. Take 40 pmol-equivalent mRNA at step 11 of Subheading 3.2.1, and precipitate the mRNA with ethanol and Na acetate. Rinse the pellet with 70% ethanol and dry. Carefully dissolve in 29 μL water. 2. Add the following to the 29 μL mRNA library (40 pmol equivalents) above. 20 pmol/μL Puromycin-linker

2 μL

10 T4 RNA ligase buffer

4 μL

10 mM ATP

1 μL

Mix well and heat the reaction mixture (total 36 μL) for 1 min at 90  C and leave it in gradual-cooling (≧30 min) to 25  C (see Note 11). 3. Add the following enzymes to start the reaction. T4 polynucleotide kinase

2 μL

T4 RNA ligase

2 μL

Incubate for 15 min at 25  C (or for 2 h at 16  C). You can leave the reaction overnight at 4  C. 4. Precipitate the mRNA-linker conjugate with 4 μL 3 M Na acetate (pH 5.2) and 100 μL ethanol, centrifuge at 16,000  g for 15 min at 4  C, rinse with 70% ethanol and dry. Dissolve the pellet in 9.5 μL water. 3.2.3 Translation and Protein Fusion to Linker

Here is a procedure to translate the puromycin-linker-conjugated 3-F mRNA library by a Wheat Germ Extract Translation Kit (Promega) (see Note 12 for comparison with a rabbit reticulocyte lysate translation system). 1. Heat 9.5 μL Linker-mRNA library (40 pmol equivalents) for 10 min at 65  C, and chill in ice. 2. Add the following (∗, included in the Kit). Wheat germ extract∗

25 μL

1 mM Amino acid mixture (–Met)∗

2 μL

1 mM Amino acid mixture (Leu)∗

2 μL

1 M K acetate∗

2.5 μL

40 U/μL RNasin plus

1 μL

5 mg/mL protein disulfide isomerase

1 μL

250 mM oxidized glutathione

2 μL

10 mM reduced glutathione

5 μL

68

Tai Kubo and Mohammed Naimuddin

Incubate the reaction mix (total 50 μL) for 15 min at 25  C (see Note 13). 3. Add 23 μL High Salts Mixture, and incubate for 1 h at 25  C (see Note 14). 4. Separate a 5 μL aliquot for checking the formation of protein fusion by Laemmli SDS-PAGE (Sodium dodecyl sulfatepolyacrylamide gel electrophoresis) containing urea (see Fig. 3a). Preparation of gel is a 2-step process containing

Fig. 3 Checking gels for protein fusion to mRNA-linker and target selection steps. Representative gels for checking the linkage of translated proteins to mRNA-linker through incorporation of puromycin (a), and for tracking the target selection steps (b). (a) Protein fusion efficiency was analyzed by comparing before and after translation of mRNA by Laemmli urea-containing SDS-PAGE (see Note 4). Gel shifts from the mRNA-linker (3-F) to the mRNA-linker-protein (3-F and PDO) are observed. The fusion efficiency is >90%. 3-F: 3-F peptide MicTx3 (61 amino acid residues). PDO: Pou-binding domain of Oct-1 (60 amino acid residues). (b) To monitor the target selection steps, cDNA aliquots recovered from each step were amplified by PCR and checked by 8 M urea-6% PAGE (see Notes 3 and 17). PCR products (350 bp, including the 3-F coding and its 50 - and 30 -peripheral sequences) are observed in the lanes 1–4, 8 and 9. 1, unbound fraction; 2–4, washing by selection buffer; 5–7, washing by wash buffer; 8 and 9, elution by DTT; M, DNA marker

in vitro Evolution from a Cys-Constrained Peptide Library

69

resolving and stacking gels (see Note 4). Follow the loading sample preparation and electrophoresis as described. 5. Total 68 μL of the remaining reaction solution is ready for the following step 2 of Subheading 3.3. 3.3 cDNA Display: Immobilization and Reverse Transcription

3-F polypeptides in the library are immobilized on Ni-NTA magnet beads through hexa-His tag at the C-terminal side of 3-F. This procedure eliminates incomplete translation products and also reading frame-shifted mutants, as well, from the library. Genetic information encoded by mRNA is quickly converted to cDNA by reverse transcriptase. The primer sequence pre-installed in the linker is utilized here (Fig. 2b). Use BSA-pretreated microcentrifuge tubes (see Note 5). 1. Take out 20 μL HisPur Ni-NTA Magnetic Beads suspension, collect the beads using a magnet stand and discard the suspension buffer. Wash the beads 3 with 300 μL Equilibration Buffer. 2. Add 232 μL Equilibrium buffer and 68 μL 3-F polypeptides (step 5 of Subheading 3.2.3). 3. Incubate for 10 min at 25  C with mixing, collect the beads and decant the supernatant. 4. Wash the beads 3 with 300 μL Wash Buffer, and once with 300 μL 1 RT Buffer (2:3). 5. Add the following to the collected beads, 5 RT Buffer (2:3)

10 μL

Water

35.5 μL

10 mM dNTP mix

2.5 μL

40 U/μL RNasin plus

1 μL

Suspend the mixture by pipetting. Start the reaction by adding 1 μL ReverTra Ace and incubate for 10 min at 42  C. 6. Collect the beads by a magnet and decant the supernatant. Wash the beads 3 with 300 μL Wash Buffer. 7. Elute the cDNA-conjugated 3-F peptide library with 50 μL Elution Buffer three times. For each elution, incubate for 5 min at 25  C. 8. Exchange buffer of the eluates to 1 Selection Buffer using Micro Bio-Spin® Column (Bio-Rad) following the manufacturer’s instructions. 9. Keep the eluates at 0–4  C, or at 20  C for long-term storage.

70

3.4

Tai Kubo and Mohammed Naimuddin

Selection

3.4.1 Negative Selection

Selection condition should be designed based on the individual criterion on demand. The following method describes a screening of a 3-F peptide library for binders to AChBP, which is homologous to the extracellular and ligand-binding domain of nicotinic acetylcholine receptor. See Subheading 3.5 for preparation of AChBPconjugated magnetic beads. To avoid nonspecific binding of the 3-F peptide library to beads, the library is exposed first to naive beads without target protein (see Note 15). Use BSA-pretreated microcentrifuge tubes (see Note 5). 1. Take 30 μL Streptavidin Mag Sepharose. Wash the beads with 500 μL 1 PBS (containing 0.1% BSA) first, and the second with 500 μL 1 Selection Buffer. 2. To the collected beads, add 50–100 μL 3-F peptide-cDNA library (from step 9 of Subheading 3.3) and 1 Selection Buffer to total 200 μL, and mix well. Incubate at 25  C with occasional agitation for 30 min. 3. Separate the supernatant from the beads. The unbound fractions are ready for step 2 of Subheading 3.4.2. Store the remaining at 0–4  C.

3.4.2 Target Selection

1. Take 50–200 pmol target-bound beads (step 18 of Subheading 3.5) and wash with 300 μL 1 Selection Buffer. 2. Collect the beads and add the pre-cleared 3-F peptide-cDNA library (step 3 of Subheading 3.4.1), mix and incubate at 25  C with occasional agitation for 30 min. 3. Collect the beads by a magnet and keep an aliquot (1/50 vol.) of the supernatant for analysis later (UB). 4. Wash the beads 3 with 300 μL 1 Selection Buffer. Keep an aliquot (1/50 vol.) of the supernatant at each step for future analysis (S1, S2, S3). 5. Wash the beads 3 with 300 μL Wash Buffer. Keep an aliquot (1/50 vol.) of the supernatant at each step for future analysis (W1, W2, W3). 6. Elute the cDNA-conjugated 3-F peptide library with 50 μL 100 mM DTT by incubating for 5 min at 25  C, twice (see Note 16). Keep an aliquot (1/50 vol.) of the supernatant at each step for future analysis (E1, E2). 7. Selection procedures above can be monitored by PCR using each sampling aliquot (1/50 vol.) as a template and P_S and P_A as a primer pair (see Note 17 and Fig. 3b). 8. Go back to Subheading 3.2 to start the next round of in vitro evolution (see Note 18).

in vitro Evolution from a Cys-Constrained Peptide Library

3.5 Preparation of AChBP-Conjugated Magnetic Beads

71

To prepare AChBP-conjugated beads, AChBP is labeled with biotin first, and mixed with streptavidin-coated magnetic beads by titration. For biotin labeling, follow the manufacturer’s instructions. Choose an appropriate cut off filter to separate the target protein and free biotin. For AChBP (ca. 26 kDa including a tag and a spacer sequences), the filter is of 10 kDa cut off. 1. Add 100 μL WS Buffer and 100 μg AChBP in a 10 kDa-cut off filter cup mounted in a microcentrifuge tube. Mix by gentle pipetting up and down. 2. Centrifuge for 10 min at 8000  g. 3. Prepare a biotin solution by adding 10 μL DMSO in the NH2Reactive Biotin tube (in the Kit). Dissolve it completely by pipetting up and down. 4. Add 100 μL Reaction Buffer to the AChBP-loaded filter cup and carefully rinse the surface of the membrane. 5. Add 8 μL DMSO-dissolved NH2-Reactive Biotin (step 3) to the cup, and mix well by pipetting up and down. Incubate for 10 min at 37  C. 6. Add 100 μL WS Buffer in the cup-tube and centrifuge for 10 min at 8000  g. Discard the filtrate at the bottom of the tube. 7. Repeat the washing step above twice by increasing the WS Buffer to 200 μL. 8. Recover the labeled protein retained on the filter by washing the membrane with 200 μL WS Buffer. Transfer the biotinyl AChBP to a micro test tube. 9. Measure absorbance at 280 nm to estimate the recovery. 10. Store the labeled protein (biotinyl AChBP) at 0–5  C. 11. Take 17.5 μL suspension of Streptavidin Mag Sepharose. 12. Wash the beads with 1 mL PBS twice by mixing on a rotator and collect beads by a magnet, and discard the supernatant. Repeat the washing procedures with 0.5 mL 0.1% BSA in PBS once, and then with 0.5 mL PBS. 13. Suspend the beads in 350 μL PBS, mix well and dispense 0, 50, 100, and 200 μL, which correspond to the original Streptavidin Mag Sepharose suspension 0, 2.5, 5, and 10 μL, respectively. 14. Collect beads by a magnet, discard the supernatant, and suspend the beads in each tube with 80 μL PBS. 15. Add 20 μL biotinyl AChBP (step 12) to each tube. Mix well and incubate the total 100 μL reaction for 30 min at 25  C.

72

Tai Kubo and Mohammed Naimuddin

16. Separate the beads and supernatant by a magnet. The supernatant is subjected to measurement of absorbance at 280 nm, and estimate the unbound protein concentration. 17. Determine the minimal mixing ratio of the Streptavidin Mag Sepharose to the biotinylated target protein (AChBP) for complete binding. Scale up the preparation by adopting the ratio for mixing Streptavidin Mag Sepharose with the biotinylated protein. Complete the conjugation by incubating them at 25  C with occasional agitation for 30 min. AChBP-bound beads are ready for “Target Selection” (Subheading 3.4). 18. Keep the remaining at 0–5  C for storage.

4

Notes 1. Three-finger (3-F) scaffold library: 3-F proteins are small (Mw 7–8 kDa) with 4–5 disulfide bonds, 2–5 β-strands and 3 protruding loops that provide the topological basis for the three-finger structure. When 3-F protein family is aligned, it is characteristic that the cysteine residues are strictly conserved and the inter-cysteine interval sequences diverge. Target and functional diversification of the 3-F family is also noticed; e.g., positive or negative modulations of ligand-gated ion channels, G-protein coupled receptors, proteinases, and so on. During evolution, 3-F proteins actively introduced mutations in fingertips to adapt to vigorous changes of target molecules, while maintaining the basic 3-F scaffold. Utilizing the scaffold thus nature selected, we constructed a 3-F cDNA library introducing random sequences in three fingertips of MicTx3, a snake α-neurotoxin [20]; 6, 10, and 7 amino acid residues in the finger loop 1, 2, and 3, respectively. Theoretical diversity in the calculation is 2023, although practically achieved 1012–14 [4, 20, 21]. 2. Puromycin linker was custom synthesized (BEX, Tokyo, Japan). The main body of the linker is 50 -CCCCCCCGCCG CCCCCCG-(5-Me-dC)-(A18)-(Spec18)-(Spec18)-(Spec18)(F-dT)-(Spec18)-CC-(Puro)-30 . The symbol 5-Me-dC denotes 50 -dimethoxytrityl-N4-(O-levulinyl-6-oxyhexyl)-50 methyl-2 -deoxycytidine (Glen Research); Spec18, C18 spacer phosphoramidite; F-dT, fluorescein-dT; and Puro, puromycin CPG. The reverse transcriptase priming sequence 50 -CCTTG30 was branched from the activated 5-Me-dC in the linker above (see Fig. 2b and [4]). 3. 8 M urea-6% PAGE: For denaturing gel electrophoresis, set the apparatus with a water-circulation system that is connected to a thermo-control bath (50–55  C). Mix sample with an equal

in vitro Evolution from a Cys-Constrained Peptide Library

73

volume of 2 Sample loading buffer, heat 10 min at 70  C, then chill in ice. Warm the gel during pre-run for 30 min, load the samples and run the gel in 1 TBE running buffer at 200 V constant voltage until the BPB dye reaches the bottom of the gel (30–40 min). After electrophoresis, carefully take out the gel and stain with 1 SYBR Gold. Take the gel image using a blue light transilluminator. 4. Laemmli urea-SDS-PAGE: 6% resolving gel: per 15 mL, 7.2 g urea, 3.75 mL 1.5 M Tris–HCl, pH 8.8, 2.25 mL 40% Acrylamide/bis-acrylamide (37.5:1), 75 μL 10% ammonium persulfate. Slightly heat for about 10 s in a microwave to dissolve urea. Bring the solution to room temperature. Then add 8 μL TEMED, mix and pour into the gel casting mold (glass plates). Carefully add water layer on the top of the gel solution. Let it stand and polymerize for 30 min. 4% stacking gel: per 5 mL, 2.4 g urea, 1.25 mL 0.5 M Tris–HCl, pH 6.8, 0.5 mL 40% Acrylamide/bis-acrylamide (37.5:1), 25 μL 10% ammonium persulfate. Remove the water layer by slanting the plates and absorb the water with a paper towel. Dissolve the urea as mentioned above. Add 3 μL TEMED and pour the above solution on the top of the lower gel. Let it stand and polymerize for 1 h. After polymerization, perform electrophoresis with SDS-PAGE running buffer (0.025 M Tris–HCl, pH 8.3, 0.192 M Glycine, 0.1% SDS). A pre-run for 10 min at 20 mA (constant current) is recommended. Sample preparation: Dissove 0.72 g of urea in 1.5 mL of 2 loading dye completely. Mix equal volumes (5 μL) of sample and urea-loading dye. DO NOT heat. Load the samples and perform electrophoresis at 20 mA constant current (for ~75 min). Gel staining and imaging are as in Note 3. 5. To avoid nonspecific binding of protein and microbeads to plastic wares, microcentrifuge tubes are precoated by BSA; incubate with 0.1% BSA in 1 PBS, rinse with 1 PBS, drain the tubes and store at 4  C until use. 6. cDNA encoding acetylcholine binding protein (AChBP) was isolated from a cDNA library prepared from the central nervous system of Aplysia kurodai (Accession number, AB709939). The AChBP cDNA subcloned into pQE32 was recombinantly expressed in E. coli, and purified by Ni-NTA affinity column chromatography. 7. Overlapping PCR vs. Splint ligation: We used to prepare a 3-F cDNA library by joining 2–4 sets of overlapping long oligonucleotides by PCR before [20]. It is convenient and relatively rapid to completion. However, this procedure requires integral

74

Tai Kubo and Mohammed Naimuddin

PCRs and it may induce biases from the original population of random sequences. Thus, to minimize the PCR frequency, we recently follow the method “Splint ligation” described here. Long oligonucleotides adjacent to each other, those junctions supported by splint oligonucleotides, are enzymatically joined. 8. Oligodeoxynucleotides synthesized by phosphoramidite method are supplied with free 50 -OH, if no special request made. To join the DNAs in 50 to 30 direction in line via phosphodiester bonds, free 50 -OH of each oligodeoxynucleotide should be monophosphorylated before T4 DNA ligase reaction. 9. A single band of the appropriate size should be observed (e.g., 350 bp for 3-F cDNA library). If not negligible extra band (s) observed, separate DNA of correct size in a preparative gel; i.e., cut out the band from the gel, and extract DNA following the method described in steps 5–9 in Subheading 3.1.2. 10. RNA gel: In vitro transcribed RNA is routinely checked its quality and quantity by running a 1.2% agarose gel electrophoresis. Each RNA (synthesized RNA or RNA marker) is mixed with 0.5–1 volume of glyoxal containing dye (e.g., NorthernMax-Gly Sample Loading Dye, ThermoFisher), incubate for 30 min at 50  C, analyze by a 1.2% agarose gel (containing a trace of ethidium bromide) in TAE buffer, and visualize under UV. 11. Alternatively, cool down the sample utilizing the ramp control of a PCR thermal cycler (e.g., 0.03  C per s from 90 to 25  C). 12. Efficiencies to produce mRNA-linker-protein fusion were compared between a rabbit reticulocyte lysate (RRL) system and a wheat germ extract (WGE) system [4]. The results show higher fusion yields in WGE (95%) than RRL (65%) for all the four peptides examined [4]. 13. We compared translation and correct folding efficiencies of the 3-F protein in the following three conditions [22]: Control, translation without PDI and GSSG/GSH; Posttranslational refolding with PDI (PTRF), refolding with PDI and GSSG/GSH after translation; Co-translational folding with PDI (CTF), simultaneous translation and folding under PDI and GSSG/GSH surveillance. We found that the relative yield of the translated and correctly folded (thus targetbound) proteins increased up to five-fold in PTRF, and eight-fold in CTF compared to the Control [22]. Apparent Kd to AChBP of the 3-F protein translated and refolded/ folded by PTRF and CTF was 95  12 nM and 54  4 nM, respectively, which is comparable to those (ca. 30 nM) expressed in E. coli and refolded. Depending on a protein to

in vitro Evolution from a Cys-Constrained Peptide Library

75

display, optimization may be required for the GSSG and GSH concentrations. 14. At the end of translation reaction, a treatment by high salts (MgCl2 and KCl) is essential for the process of fusion maturation (incorporation of puromycin at the C-terminal end and release of rRNAs). Optimization of the process is reported [4]. 15. Negative selection: For the first three rounds, we conduct a negative selection before each target selection. From the fourth round and later, we directly start the selection without negative selection. 16. DTT disrupts a 3-F scaffold that is stabilized by disulfidebridges, and thus the selected 3-F peptides bound to the target molecule are released from the beads. DNAs bound to the target, like DNA aptamers, are to be eliminated by this condition. 17. PCR conditions; 1/50 volume of each aliquot (UB, S1S3, W1W3, E1, and E2, steps 3–6 of Subheading 3.4.2) is mixed with 1 μL P_S, 1 μL P_A, 2 μL 10 PCR Buffer, 0.8 μL dNTP mix, 0.1 μL ExTaq, and water to total 20 μL. Thermal cycles: 95  C, 3 min, (95  C, 30 s; 55  C, 30 s; 72  C, 30 s)  30 cycles, 72  C, 5 min, 12  C, hold. The PCR products are analyzed by 8 M urea-6% PAGE (an example shown in Fig. 3b). For convenience, the polyacrylamide gel could be replaced by 3% NuSieve GTG agarose gel (Takara). 18. As the selection round proceed, you can properly increase the selection pressures. In case of the selection by binding affinity to the target, you can set the binding conditions more stringent; i.e., by decreasing concentration of target molecule, incubation time, and/or by increasing incubation (and washing) temperature, salt concentration [20]. It is ideal to know the sequences in the library at each end of the selection and to know how the sequences are getting converged, and decide Go/No-Go. However, it is not realistic because of time and labor consuming. We usually check the sequences at every two or three rounds, or after round seven. After target selection of each round, selected DNA pools are amplified by PCR, subcloned into a vector (e.g., pCR2.1, Invitrogen), transformed, and plasmids purified from randomly picked up colonies are subjected to sequencing. Next-generation sequencing might be an alternative, if available.

76

Tai Kubo and Mohammed Naimuddin

Acknowledgments The authors are grateful to Suzuko Kobayashi and Yoshiko Shimoyama for technical support, and to Drs. Hiroyuki Ito, Kohei Tada, and Yuki Mochizuki for discussions and contributing to optimization of some of the procedures. This work was supported in part by JSPS KAKENHI Grant Numbers JP26350978, JP16K07182, JP18K06945 to T.K. References 1. Packer MS, Liu DR (2015) Methods for the directed evolution of proteins. Nat Rev Genet 16(7):379–394. https://doi.org/10.1038/ nrg3927 2. Yuan L, Kurek I, English J, Keenan R (2005) Laboratory-directed protein evolution. Microbiol Mol Biol Rev 69(3):373–392. https://doi. org/10.1128/MMBR.69.3.373-392.2005 3. Mochizuki Y, Biyani M, Tsuji-Ueno S, Suzuki M, Nishigaki K, Husimi Y, Nemoto N (2011) One-pot preparation of mRNA/cDNA display by a novel and versatile puromycinlinker DNA. ACS Comb Sci 13(5):478–485. https://doi.org/10.1021/co2000295 4. Naimuddin M, Kubo T (2016) A high performance platform based on cDNA display for efficient synthesis of protein fusions and accelerated directed evolution. ACS Comb Sci 18 (2):117–129. https://doi.org/10.1021/ acscombsci.5b00139 5. Scott DJ, Pluckthun A (2013) Direct molecular evolution of detergent-stable G proteincoupled receptors using polymer encapsulated cells. J Mol Biol 425(3):662–677. https://doi. org/10.1016/j.jmb.2012.11.015 6. Tawfik DS, Griffiths AD (1998) Man-made cell-like compartments for molecular evolution. Nat Biotechnol 16(7):652–656. https:// doi.org/10.1038/nbt0798-652 7. Miller OJ, Bernath K, Agresti JJ, Amitai G, Kelly BT, Mastrobattista E, Taly V, Magdassi S, Tawfik DS, Griffiths AD (2006) Directed evolution by in vitro compartmentalization. Nat Methods 3(7):561–570. https:// doi.org/10.1038/nmeth897 8. Fujii S, Matsuura T, Sunami T, Nishikawa T, Kazuta Y, Yomo T (2014) Liposome display for in vitro selection and evolution of membrane proteins. Nat Protoc 9(7):1578–1591. https://doi.org/10.1038/nprot.2014.107 9. Clackson T, Wells JA (1994) In vitro selection from protein and peptide libraries. Trends Biotechnol 12(5):173–184. https://doi.org/10. 1016/0167-7799(94)90079-5

10. Katz BA (1997) Structural and mechanistic determinants of affinity and specificity of ligands discovered or engineered by phage display. Annu Rev Biophys Biomol Struct 26:27–45. https://doi.org/10.1146/ annurev.biophys.26.1.27 11. Wilson DS, Keefe AD, Szostak JW (2001) The use of mRNA display to select high-affinity protein-binding peptides. Proc Natl Acad Sci U S A 98(7):3750–3755. https://doi.org/10. 1073/pnas.061028198 12. Binz HK, Amstutz P, Pluckthun A (2005) Engineering novel binding proteins from nonimmunoglobulin domains. Nat Biotechnol 23 (10):1257–1268. https://doi.org/10.1038/ nbt1127 13. Kubo T (2016) Random peptide library for ligand and drug discovery. In: Toxins and drug discovery, pp 1–24. https://doi.org/10. 1007/978-94-007-6726-3_2-1 14. Gruber CW, Cemazar M, Anderson MA, Craik DJ (2007) Insecticidal plant cyclotides and related cystine knot toxins. Toxicon 49 (4):561–575. https://doi.org/10.1016/j. toxicon.2006.11.018 15. Kini RM, Doley R (2010) Structure, function and evolution of three-finger toxins: mini proteins with multiple targets. Toxicon 56 (6):855–867. https://doi.org/10.1016/j. toxicon.2010.07.010 16. Kolmar H (2008) Alternative binding proteins: biological activity and therapeutic potential of cystine-knot miniproteins. FEBS J 275 (11):2684–2690. https://doi.org/10.1111/j. 1742-4658.2008.06440.x 17. Brady RM, Baell JB, Norton RS (2013) Strategies for the development of conotoxins as new therapeutic leads. Mar Drugs 11 (7):2293–2313. https://doi.org/10.3390/ md11072293 18. Jeyaseelan K, Poh SL, Nair R, Armugam A (2003) Structurally conserved alphaneurotoxin genes encode functionally diverse proteins in the venom of Naja sputatrix. FEBS Lett 553(3):333–341

in vitro Evolution from a Cys-Constrained Peptide Library 19. Fry BG, Wuster W, Kini RM, Brusic V, Khan A, Venkataraman D, Rooney AP (2003) Molecular evolution and phylogeny of elapid snake venom three-finger toxins. J Mol Evol 57 (1):110–129. https://doi.org/10.1007/ s00239-003-2461-2 20. Naimuddin M, Kobayashi S, Tsutsui C, Machida M, Nemoto N, Sakai T, Kubo T (2011) Directed evolution of a three-finger neurotoxin by using cDNA display yields antagonists as well as agonists of interleukin-6 receptor signaling. Mol Brain 4:2. https://doi. org/10.1186/1756-6606-4-2

77

21. Cai W, Naimuddin M, Inagaki H, Kameyama K, Ishida N, Kubo T (2014) Directed evolution of three-finger toxin to produce serine protease inhibitors. J Recept Signal Transduct Res 34(3):154–161. https://doi. org/10.3109/10799893.2013.865747 22. Naimuddin M, Kubo T (2011) Display of disulfide-rich proteins by complementary DNA display and disulfide shuffling assisted by protein disulfide isomerase. Anal Biochem 419(1):33–39. https://doi.org/10.1016/j. ab.2011.07.034

Chapter 5 Rapid Antigen and Antibody-Like Molecule Discovery by Staphylococcal Surface Display Marco Cavallari Abstract Ever since the discovery of antibodies, they have been generated by complicated multi-step procedures. Typically, these involve sequencing, cloning, and screening after expression of the antibodies in a suitable organism and format. Here, a staphylococcal nanobody display is described that omits many the abovementioned intermediate steps and allows for simultaneous screening of multiple targets without prior knowledge nor expression of the binders. This paper reports a detailed, general step-by-step protocol to achieve nanobodies of high affinity. Apart from its focus on radioactive and fluorescent targets, it gives options for various other target formats and additional applications for the staphylococcal library; including flow cytometry and immunoprecipitation. This provides a system for antibody engineers that can be easily adopted to their specific needs. Key words Nanobody (VHH), Bacterial surface display, Target identification, Staphylococcal sortase A (SrtA), Immunoprecipitation

Abbreviations BSA Cm DTT EDTA GFP MES MSG OD PBS RT SDS Tris TSA TSB

Bovine serum albumin Chloramphenicol Dithiothreitol Ethylenediaminetetraacetic acid Green fluorescent protein 2-(N-morpholino)ethanesulfonic acid Monosodium glutamate Optical density Phosphate-buffered saline Room temperature Sodium dodecyl sulfate Tris(hydroxymethyl)aminomethane Tryptic soy agar Tryptic soy broth

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2070, https://doi.org/10.1007/978-1-4939-9853-1_5, © Springer Science+Business Media, LLC, part of Springer Nature 2020

79

80

1

Marco Cavallari

Introduction Nanobody display on Staphylococci exploits two proteins: (1) the variable region of heavy-chain-only antibodies (VHH) that originates from camelids and (2) the staphylococcal housekeeping sortase A (SrtA). These main players are briefly introduced in the next two paragraphs. VHH have first been described by Hamers-Casterman in the 1990s [1] and have increased in popularity ever since [2, 3]. Besides their advantages arising from their minimal size (~15 kD), about one-tenth of conventional antibodies, they consist of a single amino acid chain, which tremendously simplifies their molecular manipulation [4]. VHH, unlike variable regions from conventional antibodies (VH and VL), do not need to pair [5]. They possess four key amino acid substitutions in the original VH-VL interface that replace hydrophobic residues by hydrophilic ones (V37F or V37Y; G44E; L45R; and W47G) [6]. Another VHH hallmark feature is an additional, intramolecular disulfide bond in between the first and third complementarity-determining region (CDR) [7]. Although it seems to endow VHH in vivo with a net benefit, it can be omitted in VHH display systems. The Gram-positive bacterium Staphylococcus aureus decorates its lipid cell wall with proteins according to its needs in changing environments [8]. The housekeeping SrtA [9] catalyzes a transpeptidation reaction from the LPXTG motif in secreted proteins to the N-terminal amine of the lipid II pentaglycine [10]. SrtA has been evolved by protein engineers for biotechnological [11] and in vivo [12, 13] use. Also, orthogonal versions of the enzyme changing its recognition motif have been developed [14]. By equipping nanobodies with an LPXTG motif and a suitable leader sequence for secretion, VHHs can be enzymatically installed on the S. aureus surface [15]. The staphylococcal expression vector pSA-VHH-SPAXrc [15] (Fig. 1) can be introduced into the S. aureus RN4220 strain [16] by electroporation [17]. The RN4220 strain is restriction-deficient [16] and thus retains the introduced plasmid (under antibiotic selection). The electroporation efficiency of the RN4420 strain is sufficient to introduce a diverse nanobody library of synthetic, naı¨ve or immune origin (Fig. 2). Noteworthily, very few synthetic VHH libraries have been published [18, 19]. Traditionally, phage display [20] has been used to identify antibody fragments; but also other methods such as yeast, ribosome and, more recently with the advent of nextgeneration sequencing, direct B-cell screening have been developed [21–24]. Despite its advantageous growth cycles, bacterial display is less often used and mostly performed in Escherichia coli [25–31]. In these systems, fusions of antibody fragments to surface proteins are the most common format. Also in Gram-positive bacteria such as S. carnosus [32, 33] VHH-protein fusions have been

Staphylococcal VHH Display

81

Fig. 1 Map of the pSA-VHH-SPAXrc VHH expression vector. The map of pSA-VHH-SPAXrc [15] indicates the origins of replication for E. coli (Ec) and S. aureus (Sa) as well as their selection markers ampicillin (Amp) and chloramphenicol (Cm), respectively. The coding region starts with a signal peptide (S) followed by the VHH and its stem (Xrc with LPETG in the Xc region). Either SalI-BamHI or cloning with the forward (fwd) and reverse (rev) primers can be performed (see Notes 12 and 13)

Fig. 2 Flowchart of the VHH discovery procedure by staphylococcal display. Naı¨ve or immunized (1) camelids as well as synthetic DNA (2) can serve as source for the starting material to construct a plasmid-based nanobody library. Such a library is electroporated into staphylococci (3). The surface expressed VHHs are selected over one to several rounds of enrichment (4) until single VHHs can be produced from individual clones (cyan being of camelid origin and magenta from a synthetic library)

82

Marco Cavallari

Fig. 3 Versions of VHH expression constructs. The scheme depicts three different versions of VHH expression constructs. Each version integrates different parts of the staphylococcal protein A: (left) the constant Xc part encompassing the LPETG motif; (middle) the repetitive Xr and Xc parts; and (right) the immunoglobulin-binding regions D, A, B, and C—without E—plus Xrc

used. Even though the two cited papers harnessed the function of SrtA, neither one minimized the protein fusion part spanning the membrane (Fig. 3). So, the method(s) described here remain the sole ones utilizing a minimal VHH display, where only Xrc but not the full protein A (Uniprot P02976) is integrated [15]. The minimal stalk size renders the release of the VHH from the membrane by nucleophilic competition [34] or enzymatic digest [35] for downstream applications feasible. The released protein remains small (~37 kD) due to the short membrane anchor. Similar assays might be improved in the future by integrating a protease cleavage site between the VHH and the stalk if needed. Furthermore, immunoprecipitation is used in a unique way to fish for complex antigens with a diverse library; neither knowing the binder nor the target. The methods that can be applied to a pSA-VHH-SPAXrc RN4220 library are not limited to the ones described below (flow cytometry and immunoprecipitation) and should be easily transferable to techniques such as, for example, microfluidics.

2

Materials First and foremost, check with the local authorities for biosafety regulations and (waste) disposal concerning the S. aureus RN4220 strain [16, 17, 35, 36]. Ensure that all necessary permissions are granted as well as the planned experiments registered and documented according to regulatory guidelines.

Staphylococcal VHH Display

83

Common plastic material, glassware, centrifuges, shakers, etc., that are used with E. coli are suitable for use with S. aureus. Usually, any microbiology laboratory will contain the equipment required to perform the procedures described herein. 2.1 S. aureus Strain and Culture

1. Strain: S. aureus RN4220. 2. Tryptic soy broth (TSB) from Difco. Dissolve powder in ultrapure water and autoclave according to the manufacturer’s instructions. 3. For use with pSA-VHH-SPAXrc transformed RN4220, add chloramphenicol (Cm) to TSB at a final concentration of 10–20 μg/mL (see Note 1). 4. Freezing medium: 5% bovine serum albumin (BSA), 5% monosodium glutamate (MSG) in ultrapure water (sterile filtered). Take 1 mL liquid culture and mix with 1 mL freezing medium or collect plate culture in freezing medium by scraping. Store at
80  C in cryogenic screw-cap tubes.

2.2 S. aureus Competent Cell Preparation and Electroporation

1. 0.5 M sucrose in ultrapure water, sterile-filtered. 2. Gene Pulser Xcell™ with the ShockPod™ cuvette chamber and Gene Pulser cuvettes with 0.1 cm electrode gap (see Note 2). 3. TSB without antibiotics. 4. Tryptic soy agar (TSA) from Difco. Dissolve powder in ultrapure water and autoclave according to the manufacturer’s instructions. Pour plates of appropriate sizes after addition of Cm at a final concentration of 10–20 μg/mL (see Note 3).

2.3 S. aureus Immunoprecipitation and Flow Cytometry

1. Sterile (96-well plate) sealing tape. 2. NET buffer (10): 500 mM Tris, 5% NP-40, 1.5 M NaCl, 50 mM EDTA. Adjust pH to 7.4. 3. NuPAGE™ LDS sample buffer 4 (stock) or 1 (diluted with ultrapure water). 4. 1 M dithiothreitol (DTT) solution. 5. 0.2 M glycine pH 2.2. 6. 1 M Tris pH 9.1. 7. NuPAGE™ LDS loading buffer (4 or 1): Prepare 4: 270 μL 4 NuPAGE™ + 30 μL 2.5 M DTT. Prepare 1: 270 μL 4 NuPAGE™ + 30 μL 2.5 M DTT + 900 μL ultrapure water. 8. BMES buffer: 4% BSA 20 mM 2-(N-morpholino)ethanesulfonic acid (MES). Dissolve MES in ultrapure water at 0.2 M then prepare BMES buffer. 9. Protease inhibitors of your choice (f.i. cOmplete™ EDTA-free protease inhibitor cocktail).

84

Marco Cavallari

2.4 S. aureus Plasmid Library Generation, DNA Purification and Sequencing

Here, the materials for library generation by restriction cloning are given. Alternatively, Gibson or In-Fusion cloning methods can replace restriction cloning. 1. TSB with 20 μg/mL Cm. 2. TSM buffer: Sterile filtered 50 mM Tris pH 7.5, 0.5 M Sucrose, 10 mM MgCl2. 3. Lysostaphin storage buffer: 2.8 mL of 0.2 M acetic acid, 2.2 mL of 0.2 M sodium acetate, 45 mL ultrapure water. 4. 2 mg/mL Lysostaphin stock solution: Add 25 mL lysostaphin storage buffer to 50 mg lyophilized lysostaphin (AMBI PRODUCTS LLC, LSPN-50). Aliquot and store at
80  C. 5. A DNA purification kit (see Note 4). 6. Restriction enzymes: BamHI and SalI. 7. DNA ligase. 8. DNA gel extraction kit. 9. PCR clean-up kit. 10. Sequencing primers: forward 50 AAC TGA AGA ACA ACG TAA CGG 30 and/or reverse 50 CAC ACA GGA AAC AGC TAT GAC 30 . Usually, the forward sequencing primer was used.

3

Methods Carry out the following procedures at room temperature (RT) unless otherwise indicated.

3.1 Preparation of Electrocompetent S. aureus Cells

1. Inoculate 3 mL TSB with a single S. aureus RN4220 colony (see Note 5). 2. Shake overnight at 200 rpm and 37  C (see Note 6). 3. Dilute overnight culture 1:100 into 50 mL TSB (see Note 7). 4. Grow sub-culture at 200 rpm and 37  C until it reaches an optical density (OD) of 0.5 at 660 nm (see Note 8). 5. Pellet the culture at 7000 rpm for 10 min in a sterile conical tube. 6. Place the cells on ice and keep them cold from this step onwards. 7. Discard the supernatant. 8. Add 50 mL (or equal volume used for the sub-culture in step 3, see Note 7) sterile 0.5 M sucrose to the pellet. 9. Vortex the pellet until it is completely dissolved.

Staphylococcal VHH Display

85

10. Repeat steps 5–9 using 25 mL (or half the volume of the sub-culture in step 3, see Note 7) sterile 0.5 M sucrose to wash. 11. Repeat steps 5–9 using 5 mL (or 10% of the sub-culture volume in step 3, see Note 7) sterile 0.5 M sucrose to wash. 12. Aliquot electrocompetent S. aureus RN4220 cells in 100 μL batches to cryogenic screw-cap tubes (see Note 9). 13. Immediately store aliquots at
80  C (see Note 10). 3.2 Generation of S. aureus VHH Libraries

Below, the shuttling of a library into a vector for expression in Staphylococci by restriction assembly is described. If preferred, Gibson, In-Fusion, or PIPE [37, 38] cloning are suitable, tested alternatives and be conducted with the primers outlined below plus additional primers for the amplification of the backbone. 1. Linearize several micrograms of one individual staphylococcal expression vector pSA-VHH-SPAXrc (see Notes 1 and 11) with the restriction enzymes SalI and BamHI (according to the manufacturer’s instructions, see Note 12). 2. Purify the backbone by gel extraction—the band at ~6.7 kb— and store at
20  C. Neglect the ~350 bp VHH band. 3. Extend your VHH library sequences with SalI and BamHI sites (for proper primer design see Note 13) by PCR (using any high-fidelity polymerase according to the manufacturer’s instructions). 4. Run a PCR clean-up. Pool several PCRs over one column to increase DNA concentration if desired. 5. Create sticky ends on your inserts by SalI-BamHI digestion (see Note 14). 6. Perform a PCR clean-up of the digests. Several digests can be pooled over one column. Run an analytical agarose gel to check insert size if desired. 7. In a PCR machine, ligate the library into the backbone (molar ratio 3:1) at 16  C for 16 h, heat inactivate at 65  C for 10 min, cool to and keep at 4  C. Measure DNA concentration and transfer to
20  C until the electroporation (see Note 15).

3.3 Electroporation of Competent S. aureus

Work sterilely (under a flame) and with sterile materials. Importantly, keep everything except medium and plates on ice. The protocol below is for a single electroporation but includes the indications on where to loop for multiple shots. For a full library calculate 10–30 individual electroporations and then pool them before plating (to 3–9 15 cm plates). In addition, include a tenfold dilution series (10
1 to 10
9 should be sufficient; on 10 cm plates) to assess the electroporation efficiency. As a surplus, single colonies can be picked from (the appropriate plate of) the dilution series to

86

Marco Cavallari

96-well U-bottom plates, grown, and frozen for subsequent sequencing to test the diversity of the library and for screenings. 1. Retrieve competent RN4220 cells from
80  C and thaw on ice for 20 min. 2. Place electroporation cuvettes on ice for at least 10 min before use (see Note 2). 3. Distribute 1 μg of DNA in 1 μL to a 1.5 mL tube on ice (see Note 16). 4. Mix DNA with 75–100 μL of competent S. aureus cells by gently pipetting 2–3 times. 5. Incubate DNA:cell mix on ice for 5 min. 6. Place the DNA:cell suspension in an electroporation cuvette. Avoid introducing any air or bubbles (some liquid will remain above the two electrodes). 7. Take the ice bucket with the cuvettes and at least 300 μL (per shot) RT TSB to the electroporation system. 8. Set the electroporation system to 100 Ω, 25 μF, and 2500 V. 9. Take one cuvette out of the ice, wipe off any liquid on the outside of the cuvette with a Kimwipe and tap the cuvette once or twice on a hard surface to expel any air bubbles. 10. Place the cuvette in the cuvette chamber and electroporate (see Note 17). 11. Immediately retrieve the cuvette from the chamber and add 300 μL TSB without antibiotics. 12. Mix by gently pipetting up and down a few times and transfer the electroporated cells entirely to a 1.5 mL tube on ice (for multiple shots repeat steps 9–12 before advancing to step 13). 13. Recover cells standing (not shaking) at 37  C for 1 h. 14. After this hour, plate the staphylococci on TSA-plates with 20 μg/mL Cm (see Note 18). 15. Place the plates at 37  C overnight. 16. Pick colonies and scrape the plates (with freezing medium) for the bulk library. 3.4 S. aureus Surface Display and Immunoprecipitation in a Plate Format

This section describes how to use staphylococcal displayed VHH clones to fish for unknown antigens. The immunoprecipitated antigens, after elution, are ready for identification by biochemical methods, mass spectrometry, and similar techniques. The antigens can be supplied in the assay in their native state, denatured, radioactively as well as fluorescently labeled or any other chosen format. Steps 14 and 15 are for the example of radioactively labeled antigens.

Staphylococcal VHH Display

87

1. Take 96-well U-bottom plates with individual S. aureus clones from
80  C and place in shallow water to thaw—usually happens within minutes. 2. Prepare the same amount of 96-well U-bottom plates with 200 μL ice-cold 1 NET buffer per well on ice (see Note 19). 3. Transfer 50 μL thawed, VHH expressing RN4220 clonal cells to the NET buffer (see Note 20). 4. Centrifuge plate 5 min at 2500 rpm and 4  C. 5. Discard supernatant and resuspend pelleted cells in 200 μL ice-cold 1 NET buffer to wash. Repeat step 4. 6. Discard supernatant and resuspend cell pellets in 170 μL ice-cold 1 NET buffer with protease inhibitors to wash. Adjust the resuspension volume according to the antigen volume (in step 7). 7. Add 30 μL buffer or lysate/solution containing the antigen (s) and mix by pipetting 3–5 times. 8. Cover the plate with sealing tape; do not use parafilm-wrap as it will lift and cause spillover. 9. Incubate 1.5 h on a plate shaker in a 4  C room (see Note 21). 10. Centrifuge plate 5 min at 2500 rpm and 4  C, discard the supernatant and resuspend the pellets in 200 μL ice-cold 1 NET buffer to wash. 11. Repeat step 10 three additional times. After the last centrifugation and supernatant removal continue without resuspension to step 12. 12. Resuspend pellets in 100 μL 1 NuPAGE™ sample buffer and keep plate on bench at RT for 15 min (see Note 22). 13. Centrifuge plate 5 min at 2500 rpm and RT. 14. Transfer 90 μL supernatant to a fresh 96-well U-bottom plate with 2 μL 1 M DTT (see Note 23). 15. Load the desired volume on a SDS-PAGE gel. Run big gels overnight at 60–95 V (see Note 24). Solidify the gel by polyphenylene oxide, dry it and expose (few hours up to several weeks) film to it for autoradiography at
80  C in a suitable (dark) cassette. Develop the film. The dried gel can be used for repeated exposure according to the lifetime of the radioactive material used. 3.5 S. aureus Surface Display and Immunoprecipitation in a Tube Format

1. Take plates with clones from
80  C and place in shallow water to thaw (see Note 25). 2. Add 1 mL ice-cold 2% BSA in phosphate-buffered saline (PBS) to a 1.5 mL tube on ice per staphylococcal clone to be tested (see Note 26).

88

Marco Cavallari

3. Transfer 50 μL thawed, VHH expressing RN4220 clonal cells to the ice-cold 2% BSA PBS (see Note 25). 4. Centrifuge tubes 3 min at 6000 rpm and 4  C. 5. Discard supernatant and resuspend cell pellets in 500 μL ice-cold 2% BSA PBS (see Note 27). 6. Add 500 μL buffer or lysate containing the antigen(s) and mix by pipetting 3–5 times (see Note 28). 7. Place tubes on overhead rotator in 4  C room for 1 h. 8. Centrifuge tubes 3 min at 6000 rpm and 4  C, discard the supernatant and resuspend the pellets in 1 mL ice-cold 1 2% BSA PBS to wash (see Note 29). 9. Wash cells twice as in step 8 but with ice-cold PBS. 10. Transfer resuspended cells to a fresh tube and centrifuge 3 min at 6000 rpm and 4  C. 11. Carefully pipette off the supernatant and resuspend the pellet in 100 μL RT 0.2 M glycine pH 2.2 (see Note 30). 12. Keep the tube at RT for 15 min and tap to mix from time to time (2–3 times). 13. Centrifuge 3 min at 6000 rpm and RT. 14. Transfer 90 μL of the elutions (without the cells) to a fresh tube with 10 μL 1 M Tris pH 9.1 to neutralize the pH (see Note 31). 15. Mix the desired amount with 4 NuPAGE™ loading buffer (or sample buffer, see Note 23) and load onto a SDS-PAGE gel. 3.6 S. aureus Surface Display in a Flow Cytometry Format

1. Take the RN4220 VHH library stock from
80  C and place on ice to thaw (see Note 32). 2. Distribute 1 mL ice-cold, sterile BMES to a 1.5 mL tube on ice. 3. Add 300 μL thawed VHH library and mix by pipetting 2–3 times. 4. Centrifuge the tube 3 min at 6000 rpm and 4  C; discard the supernatant. 5. Resuspend the pellet in 1 mL ice-cold, sterile BMES and repeat step 4. 6. Resuspend the cells in 1 mL ice-cold, sterile BMES. 7. Transfer 50 μL cells to a fresh 1.5 mL tube and fill ad 1 mL with BMES (unstained negative control to set up the flow cytometer). 8. Add 50 μL fluorophore-labeled antigen of interest at a final concentration of 1–20 μg/mL to 950 μL cells (stained sample for measurement and/or sorting). 9. Incubate 1 h on an overhead rotator in a 4  C room.

Staphylococcal VHH Display

89

10. Centrifuge the tubes 3 min at 6000 rpm and 4  C, discard the supernatant and resuspend the pellets in 1 mL ice-cold BMES to wash. 11. Wash cells two more times as in step 9 (see Note 33). 12. Dilute 1 mL cell solution to 4 mL with 3 mL ice-cold BMES (see Note 34). 13. Set up the parameters and gates (f.i. as in [15]) on the flow cytometer with the negative control, then sort the stained sample (see Note 35). 14. Bring sorted cells close to a Bunsen burner and conduct the following steps 15, 17 and 18 under the flame. 15. Rinse the wall of the collection tube with BMES to flush all sorted cells down. 16. Centrifuge the collection tube 1 min at 1250 rpm and RT. 17. Carefully remove the supernatant until about 200 μL liquid remains. 18. Mix liquid by pipetting 3–5 times and plate total volume on a TSA plate with 20 μg/mL Cm. 19. Incubate the plate in a 37  C room overnight. 20. Pick colonies and/or scrape plates for bulk lines. 3.7 VHH Clone DNA Isolation

1. Inoculate a single colony to 5 mL TSB with 20 μg/mL Cm and grow shaking at 200 rpm and 37  C overnight. 2. Centrifuge the culture for 10 min at 7000 rpm and RT, then discard the supernatant. 3. Resuspend the cell pellet in 200 μL TSM. 4. Add 5 μL of lysostaphin (50 μg/mL final concentration). 5. Incubate 10 min at 37  C and vortex occasionally (2–3 times; see Note 36). 6. Centrifuge the lysostaphin-digested culture for 10 min at 7000 rpm and RT, then carefully remove the supernatant (see Note 37). 7. From here on, follow the instructions your plasmid DNA preparation kit (see Note 4).

4

Notes 1. The pSA-VHH-SPAXrc plasmid—and any library—can be prepared in E. coli using 100 μg/mL ampicillin as selection antibiotic. 2. Bio-Rad 1 mm cuvettes (#165-2089) were used; storage at
20  C.

90

Marco Cavallari

3. 10 cm Petri dishes were used for dilution series. 15 cm Petri dishes were used to plate the libraries. 4. Qiagen QIAamp DNA Mini Kits were used. Expect to purify less DNA compared to preparations from E. coli. Start with 10 mL RN4220 culture and use two Mini columns—elute the first column with 50 μL and then, with the recovered volume from the first elution, the second column. Alternatively, perform a MidiPrep of 50–100 mL culture and resuspend the purified DNA in 50–100 μL; this yields a concentration of ~1 μg/mL. 5. Single colonies were obtained by cross-streaking on TSA. Single colonies for inoculation were picked with sterile pipette tips. 6. Usually saturation is reached after more than 8 h. 7. Larger batches of competent cells can be made and stored at
80  C. So, use whatever volume is desired. 8. S. aureus RN4220 reaches an OD660 of 0.5 after 2–2.5 h. If you need more concentrated cells, they can be grown up to an OD660 of 0.8 without significant loss in electrocompetence. 9. The aliquot volume can be altered, as well as the final concentration of the cells, depending on the strain and plasmid size. 10. Optionally, flash freezing can be performed. 11. A pSA-enh-SPAXrc single clone provides high-quality vector material. 12. The pSA-VHH-SPAXrc vector contains a SalI (GTCGAC) site in the amino acids VD before the QVQ start of VHH and a BamHI (GGATCC) site in GS after the VSS end of VHH. BstEII can be used as alternative to BamHI—this might lead to a loss in diversity of the library due to the nature of the BstEII restriction enzyme’s recognition sequence with a variable middle nucleotide. PstI cannot be used to replace SalI because a second site exists in the Xc part of the construct. 13. The primers were designed in a degenerate fashion to maximize the diversity of the library: forward (fwd) 50 CAA CCA GAT CCT AAA gtc gac CAG KTG CAG CTC GTG GAG WCN GGN GG 30 and reverse (rev) 50 CTC TTT TGG TGC TTG gga tcc TGA GGA NAB NDT GAC NHG 30 (restriction sites—SalI in fwd and BamHI in rev—are given in small letters). Nucleotide letters are given according to the IUPAC code. With Q5 polymerase an annealing temperature of 54  C was used for 10 s (check for other polymerases). 14. Digests were performed with the high-fidelity versions of the restriction enzymes. DNA fragments were cut for 16 h at 37  C in a PCR machine to ensure complete creation of sticky ends.

Staphylococcal VHH Display

91

15. The ligation might be purified by a special kit, which might improve electroporation performance but might lead to substantial loss in total library DNA. If not enough DNA can be obtained, perform electroporations into E. coli TG1 electrocompetent cells and purify DNA from TG1 plates (see Note 1). 16. 0.5–1 μg DNA in up to 2 μL is ideal and should give more than 1 million colony forming units. Either the initial source (number of nanobody sequences in the blood sample) or this step are the limitations for the library size; the cloning steps contain at least three logs that number in DNA molecules (there are about 1 billion molecules in 10 ng of DNA at 7 kb). 17. Depending on the electroporator, the cuvette may only fit one direction. If it pops and/or arcs during the pulse, discard the cells. Possible reasons for arcing could include salts remaining in the competent cell preparation (wash additional times during preparation), too much salt in the DNA stock (dialyze DNA preparation), or air bubbles inside the DNA:cell suspension in the cuvette. 18. Reduce concentration of Cm to 10 μg/mL if (very) few cells survive. If the plates cannot absorb the amount of liquid/are too wet, spin the tubes and remove ~150 μL; resuspend and plate the remaining pellet. 19. 96-Well V-bottom plates are preferable for low bacterial numbers. PBS with 2% BSA can be used instead of NET. Different amounts/volumes of cells can be used as well; e.g., for row and/or column pools from 96-well freezing plates. 20. Carefully remove the sealing tape from the plate and re-seal with new tape before freezing the plate anew. 21. Set the shaker to the maximum speed that does not cause splashes on the sealing tape. 22. The elution by more concentrated 2 NuPAGE™ is possible. Also, acidic elution by 100 μL 0.2 M glycine pH 2.2 is suitable but has to be neutralized with 15 μL 1 M Tris pH 9.1. Longer antigen detachment times can be opted for. The cells should still be pelletable after the elution. 23. Reduction by DTT is only desired if denaturing SDS-PAGE is run. For mass spectrometry, digestion might be performed at this step or, preferably, after appropriate gel purification. 24. The dimension of a big gel roughly matches 2 side-by-side A4 pages. For examples of different gels refer to this publication [15]. 25. The tube format protocol is performed if high numbers of bacteria (or small clonal batches) are desired; e.g., use plate sweeps or overnight sub-cultures of clones. Thus, the indicated amount (50 μL) can be adapted.

92

Marco Cavallari

26. Using large amounts of bacteria might increase the background in subsequent detection methods. Thus, include a negative and a positive control. For the precipitation of green fluorescent protein (GFP; or GFP-tagged proteins), use RN4220 without pSA-VHH-SPAXrc and RN4220 with pSAenh-SPAXrc to assay antigen solutions with and without GFP. 27. Adjust for the antigen volume. 2% BSA PBS with protease inhibitors can be used for sensitive antigens. 28. Add the antigen cold if compatible. If working with GFP, include 2.5 μg (for measurements on a Typhoon laser scanner) and 2.5 ng (for detection by Western blot) GFP as controls. 29. The pelleted cell sample with 2.5 μg GFP as input material should appear yellowish to greenish after immunoprecipitation. 30. The elution with 1 sample or loading buffer (
/+ DTT) is possible – omit the 1 M Tris pH 9.1 in step 14, then continue with loading the sample(s). 31. If contaminating cells are left, centrifuge additional 3 min at 9000  g and RT and transfer the supernatant (80–90 μL) to a new tube. 32. Sub-culturing the library is not recommended. 33. For biotinylated antigen repeat steps 7–11 before step 12. 34. The volume might have to be adjusted depending on the sorter; here it is for a BD FACSAria. 35. The settings of the forward and side scatter might not be trivial as bacteria could lie close to the electronic noise of the machine. Log-axes instead of the usual linear axes have to be used. Purity sort mode might not work with bacteria but yield or enrichment mode sorts can be performed. Optionally, staphylococci can be labeled by CMFDA (5-chloromethylfluorescein diacetate), TAMRA (5-carboxytetramethylrhodamine), or similar dyes and the respective channel used as trigger parameter for the machine—this procedure might lead to a change in library diversity. 36. Up to 1 h of exposure to lysostaphin does not reduce the amount of recovered DNA. 37. Be careful not to lose the pellet (some remaining liquid is fine) as it will be mostly translucent and a little gummy. References 1. Hamers-Casterman C, Atarhouch T, Muyldermans S et al (1993) Naturally occurring antibodies devoid of light chains. Nature 363:446–448. https://doi.org/10.1038/363446a0

2. Conrath KE, Wernery U, Muyldermans S, Nguyen VK (2003) Emergence and evolution of functional heavy-chain antibodies in Camelidae. Dev Comp Immunol 27:87–103

Staphylococcal VHH Display 3. Ingram JR, Schmidt FI, Ploegh HL (2018) Exploiting nanobodies’ singular traits. Annu Rev Immunol 36:695–715. https://doi.org/ 10.1146/annurev-immunol-042617-053327 4. Muyldermans S (2013) Nanobodies: natural single-domain antibodies. Annu Rev Biochem 82:775–797. https://doi.org/10.1146/ annurev-biochem-063011-092449 5. Muyldermans S, Cambillau C, Wyns L (2001) Recognition of antigens by single-domain antibody fragments: the superfluous luxury of paired domains. Trends Biochem Sci 26:230–235 6. Finlay WJJ, Almagro JC (2012) Natural and man-made V-gene repertoires for antibody discovery. Front Immunol 3:342. https://doi. org/10.3389/fimmu.2012.00342 7. Govaert J, Pellis M, Deschacht N et al (2012) Dual beneficial effect of interloop disulfide bond for single domain antibody fragments. J Biol Chem 287:1970–1979. https://doi.org/ 10.1074/jbc.M111.242818 8. Marraffini LA, Dedent AC, Schneewind O (2006) Sortases and the art of anchoring proteins to the envelopes of gram-positive bacteria. Microbiol Mol Biol Rev 70:192–221. https://doi.org/10.1128/MMBR.70.1.192221.2006 9. Mazmanian SK, Liu G, Ton-That H, Schneewind O (1999) Staphylococcus aureus sortase, an enzyme that anchors surface proteins to the cell wall. Science 285:760–763. https://doi. org/10.1126/science.285.5428.760 10. Ton-That H, Liu G, Mazmanian SK et al (1999) Purification and characterization of sortase, the transpeptidase that cleaves surface proteins of Staphylococcus aureus at the LPXTG motif. Proc Natl Acad Sci U S A 96:12424–12429 11. Chen I, Dorr BM, Liu DR (2011) A general strategy for the evolution of bond-forming enzymes using yeast display. Proc Natl Acad Sci U S A 108:11399–11404. https://doi. org/10.1073/pnas.1101046108 12. Hirakawa H, Ishikawa S, Nagamune T (2012) Design of Ca2+
independent Staphylococcus aureus sortase A mutants. Biotechnol Bioeng 109:2955–2961. https://doi.org/10.1002/ bit.24585 13. Wu Q, Ploegh HL, Truttmann MC (2017) Hepta-mutant Staphylococcus aureus Sortase A (SrtA7m) as a tool for in vivo protein labeling in Caenorhabditis elegans. ACS Chem Biol 12 (3):664–673. https://doi.org/10.1021/ acschembio.6b00998 14. Dorr BM, Ham HO, An C et al (2014) Reprogramming the specificity of sortase enzymes.

93

Proc Natl Acad Sci U S A 111:13343–13348. https://doi.org/10.1073/pnas.1411179111 15. Cavallari M (2017) Rapid and direct VHH and target identification by staphylococcal surface display libraries. Int J Mol Sci 18:1507. https://doi.org/10.3390/ijms18071507 16. Nair D, Memmi G, Hernandez D et al (2011) Whole-genome sequencing of Staphylococcus aureus strain RN4220, a key laboratory strain used in virulence research, identifies mutations that affect not only virulence factors but also the fitness of the strain. J Bacteriol 193:2332–2335. https://doi.org/10.1128/ JB.00027-11 17. Monk IR, Shah IM, Xu M et al (2012) Transforming the untransformable: application of direct transformation to manipulate genetically Staphylococcus aureus and Staphylococcus epidermidis. MBio 3:e00277. https://doi.org/ 10.1128/mBio.00277-11 18. Moutel S, Bery N, Bernard V et al (2016) NaLi-H1: a universal synthetic library of humanized nanobodies providing highly functional antibodies and intrabodies. Elife 5:e16228. https://doi.org/10.7554/eLife.16228 19. McMahon C, Baier AS, Pascolutti R et al (2018) Yeast surface display platform for rapid discovery of conformationally selective nanobodies. Nat Struct Mol Biol 25:289–296. https://doi.org/10.1038/s41594-018-00286 20. Winter G, Griffiths AD, Hawkins RE, Hoogenboom HR (1994) Making antibodies by phage display technology. Annu Rev Immunol 12:433–455. https://doi.org/10.1146/ annurev.iy.12.040194.002245 21. Benhar I (2007) Design of synthetic antibody libraries. Expert Opin Biol Ther 7:763–779. https://doi.org/10.1517/14712598.7.5.763 22. Chan CEZ, Lim APC, MacAry PA, Hanson BJ (2014) The role of phage display in therapeutic antibody discovery. Int Immunol 26:649–657. https://doi.org/10.1093/intimm/dxu082 23. Pardon E, Laeremans T, Triest S et al (2014) A general protocol for the generation of Nanobodies for structural biology. Nat Protoc 9:674–693. https://doi.org/10.1038/nprot. 2014.039 24. Deschaght P, Vinte´m AP, Logghe M et al (2017) Large diversity of functional Nanobodies from a Camelid immune library revealed by an alternative analysis of next-generation sequencing data. Front Immunol 8:420. https://doi.org/10.3389/fimmu.2017. 00420 25. Daugherty PS (2007) Protein engineering with bacterial display. Curr Opin Struct Biol

94

Marco Cavallari

17:474–480. https://doi.org/10.1016/j.sbi. 2007.07.004 26. Lo¨fblom J (2011) Bacterial display in combinatorial protein engineering. Biotechnol J 6:1115–1129. https://doi.org/10.1002/ biot.201100129 27. van Bloois E, Winter RT, Kolmar H, Fraaije MW (2011) Decorating microbes: surface display of proteins on Escherichia coli. Trends Biotechnol 29:79–86. https://doi.org/10.1016/ j.tibtech.2010.11.003 28. Gautam S, Gniadek TJ, Kim T, Spiegel DA (2013) Exterior design: strategies for redecorating the bacterial surface with small molecules. Trends Biotechnol 31:258–267. https://doi. org/10.1016/j.tibtech.2013.01.012 29. Salema V, Marı´n E, Martı´nez-Arteaga R et al (2013) Selection of single domain antibodies from immune libraries displayed on the surface of E. coli cells with two β-domains of opposite topologies. PLoS One 8:e75126. https://doi. org/10.1371/journal.pone.0075126 30. Nicolay T, Vanderleyden J, Spaepen S (2015) Autotransporter-based cell surface display in Gram-negative bacteria. Crit Rev Microbiol 41:109–123. https://doi.org/10.3109/ 1040841X.2013.804032 ´ (2017) Escherichia 31. Salema V, Ferna´ndez LA coli surface display for the selection of nanobodies. Microb Biotechnol 10:1468–1484. https://doi.org/10.1111/1751-7915.12819 32. Fleetwood F, Devoogdt N, Pellis M et al (2013) Surface display of a single-domain antibody library on Gram-positive bacteria. Cell

Mol Life Sci 70:1081–1093. https://doi.org/ 10.1007/s00018-012-1179-y 33. Kronqvist N, Lo¨fblom J, Jonsson A et al (2008) A novel affinity protein selection system based on staphylococcal cell surface display and flow cytometry. Protein Eng Des Sel 21:247–255. https://doi.org/10.1093/pro tein/gzm090 34. Nelson JW, Chamessian AG, McEnaney PJ et al (2010) A biosynthetic strategy for re-engineering the Staphylococcus aureus cell wall with non-native small molecules. ACS Chem Biol 5:1147–1155. https://doi.org/ 10.1021/cb100195d 35. Schneewind O, Mihaylova-Petkov D, Model P (1993) Cell wall sorting signals in surface proteins of Gram-positive bacteria. EMBO J 12:4803–4811 36. Schneewind O, Model P, Fischetti VA (1992) Sorting of protein a to the staphylococcal cell wall. Cell 70:267–281. https://doi.org/10. 1016/0092-8674(92)90101-H 37. Klock HE, Koesema EJ, Knuth MW, Lesley SA (2008) Combining the polymerase incomplete primer extension method for cloning and mutagenesis with microscreening to accelerate structural genomics efforts. Proteins 71:982–994. https://doi.org/10.1002/prot. 21786 38. Klock HE, Lesley SA (2009) The polymerase incomplete primer extension (PIPE) method applied to high-throughput cloning and sitedirected mutagenesis. Methods Mol Biol 498:91–103. https://doi.org/10.1007/9781-59745-196-3_6

Chapter 6 Restriction-Free Construction of a Phage-Presented Very Short Macrocyclic Peptide Library Valentin Jakob, Saskia Helmsing, Michael Hust, and Martin Empting Abstract Phage display is a commonly used technology for the screening of large clonal libraries of proteins and peptides. The construction of peptide libraries containing very short sequences, however, poses certain problems for conventional restriction-based cloning procedures, which are rooted in the necessity to purify restricted library oligos. Herein, we present an alternative cloning method especially suitable for such very short sequences of about only 21 base pairs resulting in a 60 bp insert. The employed restriction-free hot fusion cloning strategy allows for facile library construction bypassing the need for purification of the small oligo. The library includes one well-defined disulfide bridge rendering the displayed macrocyclic peptide sequences as attractive scaffolds for novel active principles. Key words Library construction, Hot fusion cloning, Restriction-free cloning, Macrocyclic oligopeptide phage display, Panning

1

Introduction Recently, the use of macrocyclic peptides as pharmaceutical agents has regained scientific interest [1]. In general, these circular amino acid sequences are more rigid and stable than their linear counterparts [2]. This usually results in better pharmacokinetics and dynamics and, hence, renders them more suitable for the application in an organism [3]. However, larger peptides and mediumsized ones (>10 amino acids) intrinsically exceed the criteria posed by the Lipinski’s rule of five for oral bioavailability by a huge extend [4]. Hence, it might be worthwhile investigating small macrocyclic peptide scaffolds at the borderline of the Lipinski’s space or just slightly beyond. This requirement directly leads us to a peptide size of about five to seven amino acids resulting in a molecular weight of at least 303 g/mol (five glycines) to a maximum of 1321 g/mol (seven tryptophans). With the aim in mind to include a simple and straightforward macrocyclic motif, this sequence should include two cysteines in order to form a disulfide bond. This alters the

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2070, https://doi.org/10.1007/978-1-4939-9853-1_6, © Springer Science+Business Media, LLC, part of Springer Nature 2020

95

96

Valentin Jakob et al.

minimal and maximal molecular weights of the resulting macrocycles to 393 g/mol and 1153 g/mol, respectively. In our opinion, this provides a suitable range to probe the “near Lipinski’s space” for novel peptide-based active agents. Phage display is a very powerful method to identify targetbinding peptides [5, 6] in general and disulfide-bridged macrocycles [7] in particular. A variety of libraries [8–10] and panning methods [8, 11, 12] has been described in literature. One particular difficulty we encountered in this context was the construction of a phage-presented peptide library based on a very short randomized DNA oligo. Particularly, the cloning of such a peptide library into the appropriate phagemid was challenging. Cloning of short (library) genes with less than 100 bp is rarely described in literature, and in general, it is not common to clone such small DNA fragments into a vector. Conventional cloning procedures make use of restriction and purification steps applied to both phagemid and randomized library oligo prior to ligation [13]. However, gene purification like PCR cleanup for small oligos is very difficult and results at best in low yields. The usage of special kits for the gene extraction out of agarose gels is often advised, but these are usually recommended only down to 40 bp. Another applicable method is PAGE purification, which can also be used for small DNA fragments [14]. In this case, the small oligo has to be extracted out of a polyacrylamide gel after electrophoresis and manual excision. This procedure is rather time-consuming, and large quantities of product will also get lost even if extraction was successful. Notably, a reasonable yield of purified randomized DNA oligo is required to ensure that the desired library is completely incorporated in the phagemid and presented on the phages in the end. An efficient method to circumvent these problems represents restriction enzyme- and ligase-free hot fusion cloning [15]. It also avoids the necessity for cleanup steps as a whole. Given an adequate primer design, it allows the unlabored assembly of very short DNA fragments and cloning into a library phagemid of choice. With this method, we were able to reduce the oligo size of our library to only 60 bp, which contains the 21 bp relevant library sequence. This sequence encodes for a peptide library, which will be presented by the phage and contains five randomized amino acids (except cysteine) and two cysteines at fixed places, which can form a macrocycle under oxidative conditions. Hence, it includes the features of the envisioned very short macrocyclic peptide sequences described above, which we wanted to screen via phage display. The protocol presented herein describes how to design and clone such a small peptide library into a phagemid with the usage of the hot fusion technology and how to optimize the number of positive library-containing clones. Importantly, previously reported methods of library packaging [16] and the phage panning [17] are compatible with this procedure of library construction.

Very Short Macrocyclic Peptide Libraries

2

97

Materials Prepare all solutions using H2O Milli-Q.

2.1 Amplification of the Library Oligomer

1. Lyophilized library oligomer. 2. dNTPs. 3. Forward and reverse primers (in this study HF_PhageLib_01_for and HF_PhageLib_01_rev, Table 1). 4. Q5 HF DNA polymerase and 5
Q5 reaction buffer (NEB). 5. PCR tubes. 6. Thermocycler. 7. Agarose. 8. TAE buffer 50
(instant). 9. Electrophoresis chamber.

2.2 Linearization of the Phagemid Over PCR

1. Phagemid (in this protocol pHAL30 [18]). 2. dNTPs. 3. Forward and reverse primers (in this study: HF_lin_pHAL_01_for and HF_lin_pHAL_01_rev, Table 2). 4. Q5 HF DNA polymerase and 5
Q5 reaction buffer (NEB). 5. PCR tubes. 6. Thermocycler. 7. Agarose. 8. TAE buffer 50
(instant). 9. Electrophoresis chamber.

Table 1 Primers for PCR amplification of the library oligomer: The overlapping part with the oligomer is written in capital letters Oligonucleotide primer

Sequence 50 -30

HF_PhageLib_01_for

ctgctggcagctcagccggcAGCTCAGCCGGCCATG

HF_PhageLib_01_rev

agatcagcttttgttcagaacctgccTGTTCAGAACCTGCGGCCG

Table 2 Primers for PCR linearization of the phagemid pHAL30 Oligonucleotide primer

Sequence 50 -30

HF_lin_pHAL_01_for

gcaggttctgaacaaaagctgatct

HF_lin_pHAL_01_rev

gccggctgagctgccag

98

Valentin Jakob et al.

2.3 Hot Fusion Reaction

1. Amplified library. 2. Linearized phagemid (pHAL30). 3. 5
ISO buffer: 3 mL 1 M Tris–HCl pH 7.5, 150 μL 2 M MgCl2, 240 μL 100 mM dNTP mix, 300 μL 1 M DTT, 1.5 g PEG-8000, 300 μL 100 mM NAD, up to 6 mL H2O Milli-Q. Store in aliquots at 20  C. 4. 2
hot fusion mix: 100 μL 5
ISO buffer, 0.8 μL 1:5 diluted 10 u/μL T5 exonuclease (NEB), 6.25 μL 2 u/μL Q5 HF polymerase (NEB), up to 250 μL H2O Milli-Q. Store in aliquots at 20  C. 5. Thermocycler. 6. PCR tubes. 7. Dialysis plates (Merck). 8. 8.5 cm petri dishes. 9. 10% (v/v) glycerol.

2.4 Transformation and Titration

1. Electrocompetent E. coli ER2738. 2. Desalted hot fusion product. 3. 0.1 cm electroporation cuvette. 4. Electroporator. 5. Recovery medium (Lucigen). 6. Thermomixer. 7. Centrifuge for 1.5 mL tubes. 8. 2
YT medium pH 7.0: 1.6% (w/v) tryptone, 1% (w/v) yeast extract, 0.5% (w/v) NaCl. 9. 2
YT-GAT agar: 2
YT medium, 100 mM glucose, 100 μg/ mL ampicillin, 20 μg/mL tetracycline, 1.2% (w/v) agar-agar. 10. 8.5 cm petri dishes. 11. Polystyrene dish with lid (245 mm
245 mm
25 mm). 12. Single-use Drigalski spatulas. 13. Rocking shaker. 14. 50 mL falcon tubes. 15. Cryotubes. 16. Glycerin. 17. Liquid nitrogen.

2.5 Library Quality Control

1. Forward and reverse primers (in this study MHLacZ-Pro_f and MHgIII_r (Table 3)). 2. 2
DreamTaq™ Green PCR Master Mix (Thermo Fisher Scientific).

Very Short Macrocyclic Peptide Libraries

99

Table 3 Primers for library quality control and sequencing Oligonucleotide primer

Sequence 50 -30

MHLacZ-Pro_f

GGCTCGTATGTTGTGTGG

MHgIII_r

CTAAAGTTTTGTCGTCTTTCC

3. Thermocycler. 4. Agarose. 5. Electrophoresis chamber. 2.6 Library Packaging

1. 2
YT-GA: 2
YT medium, 100 mM glucose, 100 μg/mL ampicillin. 2. Shaker/incubator. 3. 50 mL Falcon tubes. 4. M13K07 helper phage (NEB). 5. Falcon centrifuge. 6. 2
YT: 1.6% (w/v) tryptone, 1% (w/v) yeast extract, 0.5% (w/v) NaCl. 7. 86% (v/v) glycerol. 8. 2 mL cryotubes. 9. Liquid nitrogen. 10. 2
YT-AK: 2
YT, 100 μg/mL ampicillin, 50 μg/mL kanamycin. 11. PEG/NaCl mix: 20% (w/v) PEG 6000, 2.5 M NaCl. 12. Phage dilution buffer: 10 mM Tris–HCl pH 7.5, 20 mM NaCl, 2 mM EDTA. 13. 0.45 μm filter.

2.7 Titration of the Amplified Library

1. 2
YT-T: 2
YT, 20 μg/mL tetracycline. 2. XL1-Blue MRF0 (Agilent Technologies). 3. Phage dilution buffer. 4. 2
YT-GA agar plates: 2
YT medium, 100 mM glucose, 100 μg/mL ampicillin, 1.2% (w/v) agar-agar. 5. Single-use Drigalski spatulas.

2.8 Coating of Microtiter Plate Wells, Panning, and Phage Titration

1. 96-Well ELISA plates. 2. PBS pH 7.4: 8.0 g NaCl, 0.2 g KCl, 1.44 g Na2HPO4
2H2O, 0.24 g KH2PO4 in 1 L. 3. Streptavidin 1 μg/μL in PBS.

100

Valentin Jakob et al.

4. Panning block: 1% (w/v) skim milk powder, 1% (w/v) BSA in PBST, prepare fresh. 5. PBST: PBS, 0.05% (v/v) Tween-20. 6. 2% (w/v) BSA in PBST. 7. Target protein (in this study: CsrA [19]). 8. TG1 cells (Lucigen). 9. 2
YT. 10. Shaker/incubator also for well plates. 11. Rocking shaker. 12. ELISA-plate washer (e.g., Tecan Hydroflex). 13. Trypsin 10 μg/mL in PBS. 14. 96-Well deep-well plate. 15. 10
GA: 1 M glucose, 1 mg/mL ampicillin. 16. M13K07 helper phage. 17. Centrifuge for well plates. 18. 2
YT-AK. 19. XL1-Blue MRF0. 20. 2
YT-GA agar plates. 21. Single-use Drigalski spatulas. 2.9 Gene Extraction and Sequencing

1. 2
YT-GA. 2. 2
YT-GA agar plates. 3. Incubator/shaker. 4. Plasmid miniprep kit. 5. Primer for sequencing, here MHLacZ-Pro_f (Table 3).

3

Methods

3.1 Design of the Peptide Library

1. Design the library oligomer in a way that there will be two cysteines in it (see Table 4). 2. Between these two cysteines, there should be two or three variable amino acids (three in this example). Before the first

Table 4 Sequence of the designed oligomer and the resulting peptide sequence Designed oligonucleotide

AGCTCAGCCGGCCATGGCCXXX TGT XXX XXX XXX TGT XXXGCGGCCGCAGGTTCTGAACA

Resulting peptide

XCXXXCX

Restriction sites NcoI and NotI are underlined; the relevant library sequence is shown in bold

Very Short Macrocyclic Peptide Libraries

101

Table 5 Encoding possibilities for XXX (DNA codon) and resulting amino acid X (single letter code) XXX

X

Lys

AAA

5.26

K

Asn

AAC

5.26

N

Thr

ACT

5.26

T

Ile

ATC

5.26

I

Met

ATG

5.26

M

Gln

CAG

5.26

Q

His

CAC

5.26

H

Pro

CCG

5.26

P

Arg

CGT

5.26

R

Leu

CTG

5.26

L

Glu

GAA

5.26

E

Asp

GAT

5.26

D

Ala

GCA

5.26

A

Gly

GGT

5.26

G

Val

GTT

5.26

V

Tyr

TAC

5.26

Y

Ser

TCT

5.26

S

Cys

TGC

–

C

Trp

TGG

5.26

W

Phe

TTC

5.26

F

Sum

99.94

and after the second cysteine, there should be one variable amino acid. XXX can encode for every amino acid X except cysteine (Table 5) (see Note 1). 3. Before and after this peptide encoding sequence, there should be enough nucleotides for amplification. 4. Include the two restriction sites NcoI before and NotI after the essential library sequence (see Note 2). 5. Every translated amino acid X, there should have the same probability (see Note 3). 6. Ensure that the coding for XXX does not include the restriction sites (here: NcoI and NotI) so that these do not occur in the library coding sequence (see Note 4).

102

Valentin Jakob et al.

Table 6 PCR components for the oligomer amplification Solution or component

Volume

5
Q5 reaction buffer

10 μL

dNTPs (10 mM)

4
1 μL

HF_PhageLib_01_for (10 μM)

2.5 μL

HF_PhageLib_01_rev (10 μM)

2.5 μL

Template (Library Oligo)

1 ng

Q5 HF DNA polymerase

0.5 μL

H2O Milli-Q

Up to 50 μL

Table 7 PCR program for the library oligomer amplification PCR step

Temperature

Duration

Initial denaturation

98  C

30 s

Denaturation

3.2 Amplification of the Library Oligomer



10 s



98 C

Annealing

72 C (NEB calc.)

20 s

Elongation

72  C

25 s/kbp

Final elongation

72  C

2 min

1. Solve the lyophilized oligomer 100 μM in H2O Milli-Q. 2. Design the two primers which allow the amplification of the sequence, and prepare it for the hot fusion reaction (see Note 5). 3. Amplify the library oligomer by means of a PCR using the designed library oligomer: like in Table 6, pipet 10 μL 5
reaction buffer, 1 μL of each dNTP with a concentration of 10 mM, 2.5 μL forward and 2.5 μL reverse primer (10 μM each) and about 1 ng of the solved library oligomer as template. Fill up to 49.5 μL with H2O Milli-Q, add 0.5 μL Q5 HF DNA polymerase, and mix gently by pipetting up and down. 4. Run PCR in a thermocycler based on the program from Table 7. Use about 30 cycles of denaturation, annealing, and elongation. Calculate the annealing temperature with the help of the NEB calculator (see Note 6) and elongation time. For example, oligo: 3 s (25 s/kbp
0.106 kbp). 5. Check the amplified PCR product via agarose gel electrophoresis, and quantify the product (see Note 7).

Very Short Macrocyclic Peptide Libraries

103

Table 8 PCR components for the phagemid linearization Solution or component

Volume

5
Q5 reaction buffer

10 μL

dNTPs (10 mM each)

4
1 μL

HF_lin_pHAL_01_for (10 μM)

2.5 μL

HF_lin_pHAL_01_rev (10 μM)

2.5 μL

Template (pHAL30)

1 ng

Q5 HF DNA polymerase

0.5 μL

H2O Milli-Q

Up to 50 μL

Table 9 PCR program for the phagemid linearization PCR step

Temperature

Duration

Initial denaturation

98  C

30 s

Denaturation

98 C

10 s

Annealing

68  C (NEB calc.)

25 s

Elongation

72  C

25 s/kbp

Final elongation

3.3 Linearization of the Phagemid Over PCR (See Note 8)





72 C

2 min

1. Design primers that create an overlap to the amplified library (Subheading 3.2). 2. Pipette the PCR reaction like in Table 8 (see Note 9). 3. Run PCR in a thermocycler based on the program from Table 9. Use about 30 cycles of denaturation, annealing, and elongation. Calculate the annealing temperature with the help of the NEB calculator and 25 s/kbp for the elongation step. For pHAL30, 2 min 5 s (30 s/kbp
4175 kbp). 4. Check the PCR product via agarose gel electrophoresis (see Note 10). 5. Purify the PCR product over a PCR cleanup kit, elute with 30 μL H2O Milli-Q pH 8–8.5, and determine the concentration with a Nanodrop (see Note 16).

3.4 Hot Fusion Reaction

1. Mix 10 μL 2
hot fusion mix with 0.06 pmol linearized phagemid (pHAL30) with 0.14 pmol amplified library oligomer, and fill up to 20 μL with H2O Milli-Q (see Note 11).

104

Valentin Jakob et al.

2. Hot fusion reaction: Hold a temperature of 50  C for 1 h in a thermocycler and then cool down to 20  C within 5 min (0.1  C/s). 3. Desalt the HF product for about 30 min via dialysis plates against 10% glycerol. 3.5 Transformation and Titration

Transformation and titration are based on Zantow et al. [16] (Subheading 3.5). 1. Thaw 25 μL E. coli ER2738 electrocompetent cells on ice. Add 5 μL of the desalted HF product. Transfer these 30 μL to a precooled 0.1 cm electroporation cuvette, and avoid air bubbles (see Note 12). 2. Perform electroporation for bacteria (1.8 kV; pulse ~3.5–4.5 ms). Immediately add 1 mL prewarmed (37  C) recovery medium, and resuspend three times before transferring the mixture into a prewarmed (37  C) Eppendorf tube. 3. Incubate at 37  C for 1 h and 600 rpm. 4. Centrifuge the tubes for 5 min at 5000
g. 5. Remove about 30 μL of the supernatant, and resuspend the pellet in the remaining medium (see Note 13). 6. Make a 1:100, 1:1000, and a 1:10,000 dilution in recovery medium with 10 μL of the suspension in a volume of 150 μL, and spread them out on 2
YT-GA 8.5 cm agar plates. 7. Plate the remaining 990 μL of the transformation onto a 245
245
25 mm 2
YT-GA agar plate. 8. Incubate all plates at 30  C overnight (see Note 14). 9. Perform the colony counting on the 8.5 cm plates. 10. Add 20 mL of 2
YT to the 245
245
25 mm plate, and incubate on a rocking shaker for 20 min. 11. Carefully scrape the cells from the medium surface with a Drigalski spatula. Collect the liquid-containing cells with a serological pipette in a 50 mL tube, supplement with 20% (v/v) glycerol, and distribute 1 mL in each of 6 cryovials. 12. Flash freeze the cells in liquid nitrogen and store the tubes at 80  C.

3.6 Library Quality Control

The library quality control is based on Zantow et al. [16] (Subheading 3.6). 1. Pick at least 12 colonies from the plate used for colony counting to perform a colony PCR with the help of the DreamTaq™ PCR mastermix polymerase (Table 10). Use the empty phagemid as a negative control (see Note 15). 2. Send positive clones for sequencing, to check whether the peptide library is cloned properly. Additionally, this provides a first hint at the diversity of the library (see Note 16).

Very Short Macrocyclic Peptide Libraries

105

Table 10 Composition of the colony PCR

3.7 Library Packaging

Solution or component

Volume

2
DreamTaq™ mastermix

25 μL

MH_LacZ-Pro_f

1.5 μL

MHgIII_r

1.5 μL

H2O Milli-Q

22 μL

Library packaging is based on Zantow et al. [16] (Subheading 3.7). 1. Inoculate 400 mL 2
YT-GAT with a gently thawed 1 mL glycerol stock (Start OD600 < 0.1). 2. Incubate at 37  C, 250 rpm until an OD600 of 0.4–0.5 is reached. Transfer 25 mL of this culture (1.25
1010 bacteria, OD600 of 1 ¼ 1
109) in a 50 mL falcon tube, and infect with the M13K07 helper phage (MOI 1:20). 3. Incubate at 37  C for 30 min without shaking and 30 min at 250 rpm. 4. Distribute the rest of the culture (375 mL) to eight falcons, centrifuge for 10 min at 3220
g, discard the supernatant, and resuspend each pellet in 800 μL 2
YT. 5. For storage at this point, transfer the suspension into 2 mL cryotubes, add 200 μL 86% glycerol, shock freeze in liquid nitrogen, and store at 80  C. 6. Pellet the bacteria at 3220
g for 10 min, discard the supernatant, carefully resuspend the pellet in a little amount 2
YT, and take up in 400 mL 2
YT-AK. Incubate over night at 30  C, 250 rpm. 7. Centrifuge the bacteria for 20 min at 10,000
g at 4  C in two containers. Transfer the supernatant in two new containers, and precipitate the phage by adding 1/5 volume ice-cold PEG/NaCl, mix it, and incubate on ice on a rocking shaker for 1 h. 8. Resuspend the two pellets in 10 mL phage dilution buffer. 9. Centrifuge for 10 min, 20,000
g at 4  C. 10. Filter the supernatant through 0.45 μm in two fresh centrifugation tubes. 11. Precipitate the phage by adding 1/5 volume ice-cold PEG/NaCl, mix it, and incubate on ice on a shaker for 20 min. 12. Centrifuge for 30 min, 20,000
g at 4  C, and discard the supernatant.

106

Valentin Jakob et al.

13. Place the centrifuge tubes upside down on a paper towel (see Note 17). 14. Resuspend each pellet in 1 mL phage dilution buffer, and centrifuge for 1 min, 16,000
g at 4  C to remove the rest of the bacteria. Transfer the supernatant into an Eppendorf tube. 3.8 Titration of the Amplified Library

Titration of the amplified library is based on Zantow et al. [16] (Subheading 3.7). 1. Inoculate 5 mL 2
YT-T with XL1-Blue MRF0 cells, and incubate over night at 37  C, 200 rpm. 2. On the next day, inoculate 50 mL 2
YT-T with 500 μL overnight, culture, and grow to an OD600 of 0.5 (~5
108 bacteria/mL). 3. Make a 102, 104, 106, 108, 1010, and 1012 dilution of this culture by mixing 990 μL phage dilution buffer with 10 μL phage solution. 4. Add 10 μL of this dilutions to 50 μL ready-grown XL1-Blue MRF0 cells. 5. Incubate for 30 min at 37  C without shaking. 6. Streak out each dilution containing the XL1-Blue MRF0 on 2
TY-GA agar plates, and incubate at 37  C overnight (see Note 18). 7. Determine the titer in pfu/mL by counting the colonies. 8. Pick at least 12 colonies by performing a colony PCR. 9. Save the plates Subheading 3.11.

3.9 Panning Procedure for a Biotinylated Protein

to

perform

a

DNA

extraction

for

The panning procedure is based on Russo et al. [17] (Subheading 3.1). Day 1:

1. Coating: Prepare a solution of 998 μL PBS + 2 μL 1 μg/μL streptavidin in PBS (2 μg), and fill up two ELISA wells with this mixture (see Note 19). Fill up the third ELISA well with panning block (fresh). Incubate for 2 h at 4  C without shaking. 2. Blocking: Empty the first two ELISA wells, wash three times with PBST, and fill up the wells with 2% BSA (fresh) in PBST. Incubate for 2 h at 4  C without shaking. 3. Adding target (protein): Empty the first two ELISA wells and wash three times with PBST. Add 4 μg in 150 μL of the protein in his storage buffer in the first well (see Notes 20 and 21). Fill up the second well with PBS. Incubate over night at 4  C without shaking.

Very Short Macrocyclic Peptide Libraries

107

4. TG1 culture: Inoculate 15 mL 2
YT with TG1 cells. Incubate o/n at 200 rpm and 37  C. Day 2:

1. Adding phage: Empty the third well of the ELISA plate, and add a mixture of 75 μL of the amplified library (normally about 5
1011 pfu/mL) + 70 μL panning block + 5 μL 1 μg/μL (5 μg) streptavidin in PBS to this well. Incubate for 1 h at room temperature on a shaker. 2. Transferring phage I: Empty the second well and wash three times with PBST. Transfer the 150 μL phage solution from the third well into the empty second well. Incubate for 1 h at room temperature on a shaker. 3. Transferring phage II: Empty the first well and wash three times with PBST. Transfer the 150 μL phage solution from the second well into the empty first well. Incubate for 1.5–2 h at room temperature on a shaker. 4. Washing step: Empty the first well and wash ten times with PBST with a plate washer. 5. Phage elution: Add 150 μL (10 μg/mL in PBS) trypsin to the empty first well. Incubate for 30 min at 37  C without shaking. 6. Transfer the 150 μL eluted phage into a well of a 96-well deepwell plate (see Note 22). 7. Grow a TG1 culture (by inoculating with the o/n culture) to an OD600 of 0.5 (can be done earlier and stored on ice at this OD till the next step). 8. Add 150 μL ready-grown TG1 cells (OD600 ¼ 0.5) to the deep well containing the eluted phage. Incubate 30 min at 37  C without shaking and 30 min at 37  C with 450–650 rpm (see Note 23). 9. Coating: Choose three new ELISA wells. Prepare a solution of 998 μL PBS + 2 μL 1 μg/μL streptavidin in PBS (2 μg), and fill up two ELISA wells with this mixture (see Note 19). Fill up a third ELISA well with panning block (fresh). Incubate for 1 h at room temperature without shaking. 10. Blocking: Empty the first two ELISA wells, wash three times with PBST, and fill up the wells with 2% BSA (fresh) in PBST. Incubate for 1 h at room temperature without shaking. 11. Adding target (protein): Empty the first two ELISA wells and wash three times with PBST. Add 4 μg in 150 μL of the protein in his storage buffer in the first well (see Notes 20 and 21). Fill up the second well with PBS. Incubate over night at 4  C without shaking.

108

Valentin Jakob et al.

12. Add 1 mL 2
YT + 150 μL 10
GA to the deep well containing 300 μL eluted phage in 2
YT. Incubate for 1 h at 37  C and 450–650 rpm. 13. Add 1011 pfu M13K07 helper phage to the deep well. Incubate for 30 min at 37  C without shaking and 30 min at 37  C, 450–650 rpm. 14. Centrifuge the deep-well plate for 10 min, 3220
g at room temperature. Discard the supernatant carefully. 15. Resuspend the pellet in 950 μL 2
YT-AK. Incubate o/n at 30  C and 450–650 rpm. 16. TG1-culture: Inoculate 15 mL 2
YT with TG1 cells. Incubate o/n at 200 rpm and 37  C. Day 3:

1. Centrifuge the deep-well plate for 10 min, 3220
g at room temperature. Transfer the supernatant in the neighbored deep well and resuspend for a few times. 2. Adding phage: Empty the third well of the ELISA plate, and add a mixture of 95 μL panning block + 5 μL 1 μg/μL (5 μg) streptavidin in PBS + 50 μL amplified phage from the deep well (step 1) to this well. Incubate for 1 h at room temperature on a shaker. 3. Transferring phage I: Empty the second well and wash three times with PBST. Transfer the 150 μL phage solution from the third well into the empty second well. Incubate for 1 h at room temperature on a shaker. 4. Transferring phage II: Empty the first well and wash three times with PBST. Transfer the 150 μL phage solution from the second well into the empty first well. Incubate for 1.5–2 h at room temperature on a shaker. 5. Washing step: Empty the first well and wash 20 times with PBST with a plate washer. 6. Phage elution: Add 150 μL (10 μg/mL in PBS) trypsin to the empty first well. Incubate for 30 min at 37  C without shaking. 7. Transfer the 150 μL eluted phage into a 96-well deep-well plate (see Note 22). 8. Grow a TG1 culture (by inoculating with the o/n culture) to an OD600 of 0.5 (can be done earlier and stored on ice at this OD till the next step). 9. Add 150 μL ready-grown TG1 cells (OD600 ¼ 0.5) to the deep well containing the eluted phage. Incubate for 30 min at 37  C without shaking and 30 min at 37  C with 450–650 rpm (see Note 23).

Very Short Macrocyclic Peptide Libraries

109

10. Coating: Choose three new ELISA wells. Prepare a solution of 998 μL PBS + 2 μL 1 μg/μL streptavidin in PBS (2 μg), and fill up two ELISA wells with this mixture (see Note 19). Fill up a third ELISA well with panning block (fresh). Incubate for 1 h at room temperature without shaking. 11. Blocking: Empty the first two ELISA wells, wash three times with PBST, and fill up the wells with 2% BSA (fresh) in PBST. Incubate for 1 h at room temperature without shaking. 12. Adding target (protein): Empty the first two ELISA wells and wash three times with PBST. Add 4 μg in 150 μL of the protein in its storage buffer in the first well (see Notes 20 and 21). Fill up the second well with PBS. Incubate over night at 4  C without shaking. 13. Add 1 mL 2
YT + 150 μL 10
GA to the deep well containing 300 μL eluted phage in 2
YT. Incubate for 1 h at 37  C and 450–650 rpm. 14. Add 1011 pfu M13K07 helper phage to the deep well. Incubate for 30 min at 37  C without shaking and 30 min at 37  C, 450–650 rpm. 15. Centrifuge the deep well plate for 10 min, 3220 rpm at room temperature. Discard the supernatant carefully. 16. Resuspend the pellet in 950 μL 2
YT-AK. Incubate o/n at 30  C and 450–650 rpm. Day 4:

1. Centrifuge the deep-well plate for 10 min, 3220
g at room temperature. Transfer the supernatant in the neighbored deep well and resuspend for a few times. 2. Adding phage: Empty the third well of the ELISA plate, and add a mixture of 95 μL panning block + 5 μL 1 μg/μL (5 μg) streptavidin in PBS + 50 μL amplified phage from the deep well (step 1) to this well. Incubate for 1 h at room temperature on a shaker. 3. Transferring phage I: Empty the second well and wash three times with PBST. Transfer the 150 μL phage solution from the third well into the empty second well. Incubate for 1 h at room temperature on a shaker. 4. Transferring phage II: Empty the first well and wash three times with PBST. Transfer the 150 μL phage solution from the second well into the empty first well. Incubate for 1.5–2 h at room temperature on a shaker. 5. Washing step: Empty the first well and wash ten times with PBST with a plate washer.

110

Valentin Jakob et al.

6. Phage elution: Add 150 μL (10 μg/mL in PBS) trypsin to the empty first well. Incubate for 30 min at 37  C without shaking. 7. Save the sample with the eluted phage for titration and further steps. 3.10

Phage Titration

The phage titration is based on Russo et al. [17] (Subheading 3.3). 1. Inoculate 30 mL 2
YT-T in a 100 mL with XL1-Blue MRF0 and grow overnight at 37  C and 250 rpm. 2. Inoculate 50 mL 2
YT-T with 500 μL overnight culture and grow at 250 rpm and 37  C up to OD600 ¼ 0.5. 3. Make serial dilutions of the phage suspension in PBS. 4. Infect 50 μL bacteria with 10 μL phage dilution and incubate for 30 min at 37  C (see Note 24). 5. Plate the 60 μL infected bacteria on 2
YT-GA agar plates (8.5 cm petri dishes). 6. Incubate the plates overnight at 37  C. 7. Count the colonies and calculate the cfu/mL titer according to the dilution.

3.11 Gene Extraction and Sequencing

1. Inoculate up to 50 cultures of 15 mL 2
YT-GA with colonies from the titration plate from the third panning round and up to 50 cultures of 15 mL 2
YT-GA with colonies from Subheading 3.8 (packed library before panning). 2. Inoculate all cultures at 37  C, 200 rpm overnight. 3. Extract the DNA from each culture using a DNA miniprep kit. 4. Use the MHLacZ-Pro_f primer for sequencing each DNA. 5. Compare the peptide encoding sequences before the panning with the sequences after the third panning round. 6. Determine the enrichment of peptide sequences from the sequencing result after the third panning round (see Note 25).

4

Notes 1. 19 canonical amino acids without cysteine at 5 different positions result in about 2.5
106 variants in the library. 2. These two restriction sites are only for the case if you want to use restriction cloning. They will not be used for the hot fusion cloning. When using the restriction sites, only 30 bp will be left. This will probably cause problems for performing a purification step.

Very Short Macrocyclic Peptide Libraries

111

3. The oligomer was synthesized by ELLA Biotech GmbH, which is specialized on the synthesis of randomized oligonucleotides using trinucleotides. 4. This is only necessary if you plan to do a restriction cloning instead of a hot fusion. 5. We recommend using a program like SnapGene to make the design of primers for hot fusion much easier. If the program has no function for hot fusion, it is possible to use the Gibson assembly function for the design. 6. 72  C is the maximum from NEB Tm-calculator, and in our case, due to the primer design, the Tm-difference between the two primers is greater than the recommended limit of 5  C (here: 7  C difference). 7. We recommend using a low-molecular-weight ladder-like N3233S from NEB. The PCR product does not need to be further purified for the hot fusion reaction. The purification of such small genes (in this case about 100 bp) will result in an insufficient recovery. This is also a reason to decide for hot fusion than restriction cloning, because further purification is not necessary. 8. It is also possible to linearize the phagemid via restriction enzymes, for pHAL30 with NcoI and NotI. In addition to these two enzymes, HindIII and MluI can be used to make the restriction-digested DNA fragment smaller. Unfortunately, the efficiency of this procedure was not high in our case, and it seemed that restriction was not quantitative, because in a hot fusion reaction, the library oligomer with the restrictionlinearized pHAL30 often was less than 10% positive clones. 9. It is important to have a nearly quantitative linearization of the phagemid, so less than 1 ng of template will be even better. 10. It is possible that more than one band will occur on the gel (three bands in our case), but if the strongest band is the desired product, it will work nevertheless. 11. Sometimes it is necessary to try different ratios to achieve a successful hot fusion reaction. Consider doing more than one of those reactions in parallel. Combining several reactions after transformation will result in a higher propensity to get librarycontaining clones. 12. Consider doing more than one transformation in parallel. It will be effective to combine the reactions after transformation. Keep also in mind that the transformation efficiency can vary. 13. If using more than one transformation sample, this will be the point to combine these. Be sure to have in total a volume of about 1 mL remaining media in the supernatant, because this will be the ideal volume to streak out on the plates.

112

Valentin Jakob et al.

14. We recommend incubating at 30  C instead of 37  C because the ER2738 is growing very fast. 15. Typical rates are about 9/12 (75%) up to 11/12 (92%) positive clones. 16. Due to the lengths of the pHAL30 (>4000 bp), the linearization via PCR even with the Q5 HF polymerase can cause several point mutations in the backbone of the pHAL30. However, in most cases, this will have no effect on the presented peptide and the functionality of the phage presenting it. 17. The PEG/NaCl solution should be completely removed. 18. It is not necessary to add tetracycline to the culture/plates even though the MRF0 has a tetracycline resistance. 19. For non-biotinylated proteins, see also the instructions from Subheading 3.1 (Russo et al. [17]). 20. When using biotinylated proteins, it does not matter, if there is DTT or glycerol in the buffer. For non-biotinylated proteins, it is necessary to remove such components from the buffer. 21. The needed amount depends on the protein. If you can estimate how well your protein immobilizes, it will be possible to adjust the amount for every panning round. 22. Collect 10 μL of the eluted phage after each panning round for titration and further steps. 23. If you do not have a plate shaker/incubator, use a conventional shaker/incubator with a plate holder at maximum speed, i.e., the Thermo MaxQ 4000, and use 400–500 rpm. 24. Make the titration for every stock of the collected eluted phage from each panning round. Using droplets of just PBS with XL1-Blue MRF0 and droplets of just XL1-Blue MRF0 as a negative control on a separate plate is recommended. This will show if the PBS and the XL1-Blue MRF0 are contamination-free. XL1-Blue MRF0 is used for titration and for following single-clone analysis, because the phagemid is more stable in these bacteria compared to TG1. 25. The more frequently a sequence occurs, the higher is the probability that this is a desired binder for the used target protein.

Acknowledgments This review contains updated and revised parts of former protocols by Zantow et al. [16] and Russo et al. [17]. We thank Rolf W. Hartmann for his continuous support.

Very Short Macrocyclic Peptide Libraries

113

References 1. Craik DJ, Lee M-H, Rehm FBH et al (2018) Ribosomally-synthesised cyclic peptides from plants as drug leads and pharmaceutical scaffolds. Bioorg Med Chem 26(10):2727–2737. https://doi.org/10.1016/j.bmc.2017.08.005 2. Bogdanowich-Knipp SJ, Chakrabarti S, Siahaan TJ et al (1999) Solution stability of linear vs. cyclic RGD peptides. J Pept Res 53 (5):530–541. https://doi.org/10.1034/j. 1399-3011.1999.00052.x 3. Diao L, Meibohm B (2013) Pharmacokinetics and pharmacokinetic-pharmacodynamic correlations of therapeutic peptides. Clin Pharmacokinet 52(10):855–868. https://doi.org/10. 1007/s40262-013-0079-0 4. Lipinski CA (2004) Lead- and drug-like compounds: the rule-of-five revolution. Drug Discov Today Technol 1(4):337–341. https://doi. org/10.1016/j.ddtec.2004.11.007 5. Brown T, Brown N, Stollar EJ (2018) Most yeast SH3 domains bind peptide targets with high intrinsic specificity. PLoS One 13(2): e0193128. https://doi.org/10.1371/journal. pone.0193128 6. Sidhu SS, Lowman HB, Cunningham BC et al (2000) [21] Phage display for selection of novel binding peptides. In: Applications of chimeric genes and hybrid proteins—part C: protein-protein interactions and genomics, vol 328. Elsevier, Amsterdam, pp 333–IN5 7. Sakamoto K, Sogabe S, Kamada Y et al (2017) Discovery of high-affinity BCL6-binding peptide and its structure-activity relationship. Biochem Biophys Res Commun 482(2):310–316. https://doi.org/10.1016/j.bbrc.2016.11.060 8. Rentero Rebollo I, Heinis C (2013) Phage selection of bicyclic peptides. Methods 60 (1):46–54. https://doi.org/10.1016/j. ymeth.2012.12.008 9. Diderich P, Heinis C (2014) Phage selection of bicyclic peptides binding Her2. Tetrahedron 70(42):7733–7739. https://doi.org/10. 1016/j.tet.2014.05.106 10. Ryvkin A, Ashkenazy H, Weiss-Ottolenghi Y et al (2018) Phage display peptide libraries: deviations from randomness and correctives. Nucleic Acids Res 46(9):e52. https://doi. org/10.1093/nar/gky077

11. Watters JM, Telleman P, Junghans RP (1997) An optimized method for cell-based phage display panning. Immunotechnology 3(1):21–29. https://doi.org/10.1016/S1380-2933(96) 00056-5 12. Nguyen X-H, Trinh T-L, Vu T-B-H et al (2018) Isolation of phage-display libraryderived scFv antibody specific to Listeria monocytogenes by a novel immobilized method. J Appl Microbiol 124(2):591–597. https://doi. org/10.1111/jam.13648 13. Hust M, Meyer T, Voedisch B et al (2011) A human scFv antibody generation pipeline for proteome research. J Biotechnol 152 (4):159–170. https://doi.org/10.1016/j. jbiotec.2010.09.945 14. Dretzen G, Bellard M, Sassone-Corsi P et al (1981) A reliable method for the recovery of DNA fragments from agarose and acrylamide gels. Anal Biochem 112(2):295–298. https:// doi.org/10.1016/0003-2697(81)90296-7 15. Fu C, Donovan WP, Shikapwashya-Hasser O et al (2014) Hot Fusion: an efficient method to clone multiple DNA fragments as well as inverted repeats without ligase. PLoS One 9 (12):e115318. https://doi.org/10.1371/jour nal.pone.0115318 16. Zantow J, Moreira GMSG, Du¨bel S et al (2018) ORFeome phage display. Methods Mol Biol 1701:477–495. https://doi.org/10. 1007/978-1-4939-7447-4_27 17. Russo G, Meier D, Helmsing S et al (2018) Parallelized antibody selection in microtiter plates. Methods Mol Biol 1701:273–284. https://doi.org/10.1007/978-1-4939-74474_14 18. Frenzel A, Schirrmann T, Hust M (2016) Phage display-derived human antibodies in clinical development and therapy. MAbs 8 (7):1177–1194. https://doi.org/10.1080/ 19420862.2016.1212149 19. Maurer CK, Fruth M, Empting M et al (2016) Discovery of the first small-molecule CsrARNA interaction inhibitors using biophysical screening technologies. Future Med Chem 8 (9):931–947. https://doi.org/10.4155/fmc2016-0033

Chapter 7 In Vitro Maturation of a Humanized Shark VNAR Domain to Improve Its Biophysical Properties John Steven, Obinna C. Ubah, Magdalena Buschhaus, Marina Kovaleva, Laura Ferguson, Andrew J. Porter, and Caroline J. Barelle Abstract VNAR domains are the binding regions of new antigen receptor proteins (IgNAR) which are unique to sharks, skates, and rays (Elasmobranchii). Individual VNAR domains can bind antigens independently and are the smallest reported adaptive immune recognition entities in the vertebrate kingdom. Sharing limited sequence homology with human immunoglobulin domains, their development and use as biotherapeutic agents require that they be humanized to minimize their potential immunogenicity. Efforts to humanize a human serum albumin (HSA)-specific VNAR, E06, resulted in protein molecules that initially had undesirable biophysical properties or reduced affinity for cognate antigen. Two lead humanized anti-HSA clones, v1.10 and v2.4, were subjected to a process of random mutagenesis using error-prone PCR. The mutated sequences for each humanized VNAR variant were screened for improvements in affinity for HSA and biophysical properties, achieved without a predicted increase in overall immunogenicity. Key words Variable new antigen receptors (VNARs), Humanization, Immunogenicity, Phage display, Error-prone PCR, Human kappa light chain

1

Introduction Phage display technology allows the screening of peptide or protein libraries, displayed on the surface of filamentous phage, for molecules with desired characteristics including specificity and affinity [1–4]. A critical aspect of this technology is the linking of phenotype and genotype within each individual phage, allowing the simultaneous selection of phage-displayed proteins with desired properties and their encoding genetic blueprint. In combination with error-prone PCR [5–8], this display technology facilitates the selective interrogation of a library of mutant proteins derived from a single progenitor molecule and the isolation of those modified proteins with preferred characteristics, while maintaining connection to the mutated DNA fragment genotype. These may be improvements in binding affinity or other biophysical properties

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2070, https://doi.org/10.1007/978-1-4939-9853-1_7, © Springer Science+Business Media, LLC, part of Springer Nature 2020

115

116

John Steven et al.

such as thermal stability, increased solubility, decreased aggregation propensity, or resistance to proteases [9–13]. The development of new biologics for therapeutic use is proceeding at pace, with more than 1600 immunoglobulin and 1300 non-immunoglobulin biologics in development worldwide [14–16]. Some of these protein scaffolds are from nonhuman sources [16–19], and the global biotechnology industry has devoted much time, effort, and resource to making these nonhuman scaffolds more human to reduce the likelihood of an adverse immunological response when they are used clinically. Here, a reduction in potential immunogenicity was the driving force behind a humanization strategy for the therapeutic development of VNAR domains. The initial humanization of VNARs was described by Kovalenko et al. [20] using the anti-hen egg lysozyme (HEL) domain, 5A7 [21], as a model template. The approach adopted was to model 5A7 against numerous human heavy and light chain frameworks to identify the closest structures to that of the parental VNAR. Using this strategy, the human kappa light chain DPK9 was selected as the closest match. This methodology was used to successfully humanize 5A7 and was then transposed onto an antiHSA-binding VNAR domain, E06, resulting in the construction of a clone known as version v1.10. An alternative strategy was used in parallel based on the human kappa germline framework DPK24 giving rise to v2.4. Both of these humanized variants of E06 were studied in greater detail, and this work revealed that the biophysical properties of v1.10 and v2.4 were not ideal for future clinical development. v1.10 had a propensity to dimerize, and v2.4 had a significantly reduced affinity for HSA (Figs. 1 and 2). To overcome these unfavorable characteristics, v1.10 and v2.4 mutagenesis libraries were built and screened to select better preforming humanized versions of these proteins. The mutated sequences were cloned into a phage display vector, and the resultant libraries were screened against HSA. Following this second round of selection, individual clones were expressed, and periplasmic extracts were evaluated by ELISA for binding to HSA and HEL as a nontarget control. Clones with an OD450 of at least 80% of the parental v1.10 or v2.4 signal were selected for further screening by surface plasmon resonance (SPR) (off-rate determination). Analytical size exclusion chromatography (SEC) was used as a secondary screen to ensure those clones selected were not prone to aggregation. Clones with the greatest specificity and affinity were taken forward for further characterization and assessment of biophysical and immunogenic properties. In summary, the use of the molecular techniques described in this text facilitated the selection of matured humanized anti-albumin VNAR domains with improved biophysical properties.

Maturation of a Humanized Shark VNAR Domain

117

Fig. 1 Sequences and ELISA-binding characteristics of VNAR E06 and humanized variants, v1.10 and v2.4. (a) Amino acid sequence of humanized v1.10 and v2.4 clones aligned with wild-type E06 to show framework regions; FW, complementarity determining regions; CDR, and hypervariable regions, HV. (b) Phage-binding ELISA to human serum albumin (HSA) and hen egg lysozyme (HEL)-coated plates. Phage-displayed VNAR domains E06, humanized v1.10 and humanized v2.4, and the negative control clone 5A7 (anti-HEL VNAR) were assessed for binding to both HSA and HEL. (Figure reproduced from Steven et al. (2017) [27] with permission from Frontiers Media Limited 131 Finsbury Pavement, WeWork, office 01-106 London EC2A 1NT United Kingdom)

2 2.1

Materials Error-Prone PCR

1. Humanized VNAR template DNA (diluted in water to 100 ng/μL). 2. 50 and 30 primers (diluted in water to 25 pmol/μL) (see Note 1). 3. Biometra® TProfessional Thermocycler. 4. 0.2 mL thin walled PCR tubes. 5. RNase-free water. 6. GeneMorph II random mutagenesis kit (Agilent Technologies, Santa Clara, CA USA). 7. Agarose gel electrophoresis equipment. 8. 50 TAE buffer: 0.4 M Tris acetate, 50 mM EDTA, pH 8.3 with acetic acid. Dilute to 1 TAE for use in electrophoresis steps. 9. Qiagen PCR clean-up and gel extraction kits (Qiagen Ltd., Manchester, UK). 10. NanoVue UV/visible spectrophotometer (GE Healthcare, Uppsala, Sweden).

118

John Steven et al.

a

b 1

2

M

3

4

c

mAU

mAU

70 80

60 50

60 40 40

30 20

20 10 0

0 10

20 30 Time (min)

40

10

20 30 Time (min)

40

Fig. 2 SDS-PAGE and SEC profiles of native E06 and humanized v1.10. (a) SDS-PAGE and Coomassie blue staining of 2 μg of native E06 and E06 humanized v1.10 proteins. Lane 1, E06; Lane 2, v1.10 samples non-reduced Lane M, molecular weight markers. Lane 3, E06; Lane 4, v1.10 samples reduced. (b) Analytical SEC chromatogram of native E06 protein running as a single peak. (c) Analytical SEC chromatogram of humanized v1.10 protein showing monomer and dimer peaks. (Figure reproduced from Steven et al. (2017) [27] with permission from Frontiers Media Limited 131 Finsbury Pavement, WeWork, office 01-106 London EC2A 1NT United Kingdom)

2.2 Mutant Library Construction

1. Digested amplicon from error-prone PCR. 2. Digested phage display vector DNA (see Note 2). 3. T4 DNA ligase and buffer. 4. Electrocompetent E. coli TG1. 5. 1 mm electroporation cuvette. 6. SOC recovery media. 7. 2 TY: In 1 L deionized water, dissolve 16 g, 10 g, and 5 g of tryptone, yeast extract, and NaCl, respectively, then autoclave. 8. 2 TY-AG: 2 TY, 100 μg/mL ampicillin and 2% (w/v) glucose. 9. 2 TY-AKG: 2 TY, 100 μg/mL ampicillin, 50 μg/mL kanamycin, and 2% (w/v) glucose. 10. TYE-agar: In 400 mL deionized water, dissolve 3.2 g (tryptone), 2 g (yeast extract), 2 g (NaCl), and 6 g (agar), then autoclave.

Maturation of a Humanized Shark VNAR Domain

119

11. TYE-AG agar: TYE-agar, 100 μg/mL ampicillin, 2% (w/v) glucose. 12. TYE-AKG agar: TYE-agar, 100 μg/mL ampicillin, 50 μg/mL kanamycin, and 2% (w/v) glucose. 13. Electroporator. 14. Bioassay plates. 15. 50 mL centrifuge tubes. 16. Cryovials. 2.3 Library Rescue and Selection

1. PEG/NaCl: 20% (w/v) polyethylene glycol 6000/2.5 M NaCl. 2. E. coli TG1 cells. 3. PBS: 1 phosphate-buffered saline; sterile 10 PBS (Severn Biotech) diluted 1/10 in sterile water to give 1 PBS solution. 4. PBS-T: PBS containing 0.05% Tween-20. 5. M-PBS: Marvel phosphate-buffered saline; 1 PBS containing 2% or 4% (w/v) Marvel original dried skimmed milk powder (as indicated). 6. 100 mM triethylamine. 7. 1 M Tris–HCl, pH 7.5. 8. 2 TY. 9. 2 TY-AG. 10. 2 TY-AKG. 11. TYE-AG agar. 12. TYE-AKG agar. 13. Periplasmic fractionation buffer: 50 mM Tris–HCl, 1 mM EDTA, 20% (w/v) sucrose, pH 8.0. 14. ELISA Nunc MaxiSorp flat-bottom 96-well plate (Thermo Fisher Scientific, Loughborough, UK). 15. Breathable seals. 16. 96 deep-well culture plates. 17. IPTG. 18. M13K07 helper phage. 19. Monoclonal anti-polyhistidine peroxidase antibody produced in mouse (Sigma-Aldrich, Missouri, USA). 20. ELISA 3,30 ,5,50 -tetramethylbenzidine (TMB) (Thermo Fisher Scientific, Loughborough, UK).

substrate

21. ELISA stop solution 1 M H2SO4 (27.8 mL concentrated H2SO4 into 472.2 mL distilled water). 22. EnVision 2104 multilabel microplate reader (Perkin Elmer, Buckinghamshire, UK).

120

John Steven et al.

23. 75  12 mm MaxiSorp Immuno™ Tube (Immunotubes). 24. Sigma 6 K 15 centrifuge with appropriate plate carrier, tube carrier bucket adaptors to fit (Sigma Laboratories, Osterode, Germany). 2.4

Periprep ELISA

1. PBS. 2. PBS-T. 3. M-PBS. 4. 2 TY medium. 5. Periplasmic fractionation buffer. 6. Human, rat, and mouse serum albumins (HSA, RSA, and MSA, respectively). 7. Nunc MaxiSorp flat-bottom 96-well plate. 8. Anti-HA-HRP conjugate (Abcam, Cambridge UK). 9. Monoclonal anti-polyhistidine peroxidase antibody produced in mouse (Sigma-Aldrich Missouri, USA). 10. TMB substrate. 11. 1 M H2SO4. 12. EnVision 2104 multilabel microplate reader (Perkin Elmer). 13. Plate washer, Model Zoom HT Washer (Berthold Detection Systems, Germany). 14. Sigma 6 K 15 centrifuge (Sigma Laboratories, Osterode, Germany).

2.5 Reformatting of Selected Clones for Transient Expression in Eukaryotic Cells

1. BioRobot 8000 (Qiagen, Hilden, Germany). 2. Biometra® Tprofessional Thermal Cycler. 3. 0.2 mL thin-walled PCR tubes. 4. 50 and 30 oligonucleotide primers at 25 pmol/μL in water. 5. 96 deep-well culture plates. 6. Electrocompetent E. coli TG1. 7. QIAprep 96 Turbo BioRobot Kit (Qiagen, Hilden, Germany). 8. QPix2 XT colony picker (Genetix, San Jose, CA, USA). 9. Eppendorf 2025 Electroporator. 10. T4 DNA ligase and buffer. 11. Phusion High-Fidelity PCR Master Mix Kit (NEB Ipswich, MA, USA). 12. Perkin Elmer MiniTrak robotic liquid handling system (Waltham, MA, USA). 13. Wheaton standard roll-in carbon dioxide incubator (DWK Life Sciences, Millville, NJ USA).

Maturation of a Humanized Shark VNAR Domain

121

14. New Brunswick Innova 2000 platform shakers (Eppendorf UK Limited, Stevenage UK). 15. 96-Well cell culture plates. 16. Lipofectamine (Fisher Scientific, Loughborough, UK). 17. Freestyle 293 media for HEK293 maintenance growth and expression (Fisher Scientific, Loughborough, UK). 18. Sigma 6 K 15 centrifuge (Sigma Laboratories, Osterode, Germany). 19. NanoVue UV/visible spectrophotometer (GE Healthcare, Uppsala, Sweden). 20. Agarose gel electrophoresis equipment. 21. Qiagen PCR clean-up and gel extraction kits (Qiagen Ltd., Manchester, UK). 2.6 Surface Plasmon Resonance: Off-Rate Determination

1. Biacore® T200 biosensor (GE Healthcare, Amersham, UK). 2. HBS-EP running buffer: 0.01 M HEPES, pH 7.4, 0.15 M NaCl, 3 mM EDTA, 0.005% (v/v) surfactant P20 (GE Healthcare, Amersham, UK). 3. CM5 series S chip (GE Healthcare, Amersham, UK). 4. Primary amine coupling kit type 2 (GE Healthcare, Amersham, UK). 5. HSA. 6. Sartorius Minisart NML syringe filters 0.2 μm filter unit (Fisher Scientific, Loughborough, UK). 7. 10 mM sodium acetate, pH 4.5. 8. 10 mM glycine, pH 1.5. 9. 96-Well polystyrene microplates.

2.7 Sequence Determination and Large-Scale Expression and Purification of Lead Clones

1. Polyethylenimine (PEI): 25 kDa, linear, 1 mg/mL stock in acidified water (a few drops of HCl), then neutralized with NaOH to pH 7, sterilized by filtration (0.22 μm), aliquoted, and stored at 20  C (Polysciences Inc., Warrington, PA, USA). 2. Plasmid mega kit. 3. Freestyle 293 media. 4. Wheaton standard roll-in carbon dioxide incubator. 5. New Brunswick Innova 2000 platform shakers. 6. 2 TY-AG. 7. AKTA prime chromatography system setup for HisTrap purification according to manufacturer’s instructions (GE Healthcare, Uppsala, Sweden).

122

John Steven et al.

8. NanoVue UV/visible spectrophotometer. 9. Sigma 6 K 15 centrifuge (Sigma Laboratories, Osterode, Germany). 10. Amicon® Ultra membrane filtration units with a 10,000 Dalton cut-off (Merck, Darmstadt, Germany). 11. Microscope and hemocytometer and trypan blue stain for cell counting and viability assessment. 12. 20% (w/v) tryptone made up in PBS and 0.2 μm filtered prior to use. 13. Opti-MEM medium (Fisher Scientific, Loughborough, UK). 14. Sterile conical disposable flasks, ranging in size from 250 mL to 2 L for growth of HEK293 cells in suspension culture. 2.8 ELISA and EC50 Measurement

1. HSA. 2. Nunc MaxiSorp flat-bottom 96-well plate. 3. Plate washer, Model Zoom HT Washer. 4. Monoclonal anti-polyhistidine peroxidase antibody produced in mouse. 5. ELISA substrate. 6. 1 M H2SO4. 7. EnVision 2104 multilabel microplate reader. 8. SigmaPlot 9.0 software. 9. PBS. 10. PBS-T.

2.9 Surface Plasmon Resonance Affinity Measurement

1. Biacore® T200 biosensor. 2. HBS-EP running buffer. 3. CM5 series S chip. 4. Primary amine coupling kit type 2. 5. HSA. 6. Sartorius Minisart NML Syringe Filters 0.2 μm filter unit. 7. 10 mM sodium acetate, pH 4.5. 8. 10 mM glycine, pH 1.5. 9. 96-Well polystyrene microplates.

2.10 Analytical Size Exclusion Chromatography

1. Agilent 1200 HPLC system with autosampler and variable wavelength detector. 2. Agilent ZORBAX GF-200 analytical size exclusion column (9.4  250 mm, 4 μm matrix) or Agilent SEC 3 column (7.8  150 mm, 4 μm matrix or analytical size exclusion chromatography column of choice) (see Note 3).

Maturation of a Humanized Shark VNAR Domain

123

3. PBS: 0.14 M NaCl, 10 mM NaH2PO4, 10 mM Na2HPO4, pH 7.4, 0.2 μm filter prior to use. 4. Protein samples for analysis made up in PBS, pH 7.4.

3

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

3.1

Error-Prone PCR

1. Use GeneMorph II random mutagenesis kit in accordance with manufacturer’s directions to introduce the desired number of nucleotide changes into the template DNA sequence (see Note 2). PCR was carried out using 10 pg of template DNA per reaction (30 cycles), and a total of 100 μg of DNA was obtained from 92 reactions. 2. Purify the resultant amplicons, using Qiagen PCR clean-up kit, in accordance with the manufacturer’s instructions. 3. Digest purified amplicons with restriction enzymes, via restriction sites introduced at 50 and 30 ends by PCR primers (see Note 1). 4. Further purify digested mutated amplicon by electrophoresis on a 1% agarose, 1 TAE gel. Excise band and clean up using Qiagen gel extraction kit (see Note 4). 5. Use NanoVue UV/visible spectrophotometer at 260 nm to quantify recovered digested amplicon (see Note 5).

3.2 Mutant Library Construction

1. Ligate vector and insert using a 1:3 ratio (5 μg digested vector:15 μg of purified mutated amplicon). Incubate overnight at 20  C (see Note 6). 2. Ethanol precipitate the ligated DNA and recover by centrifugation at 17,000  g for 20 min. Wash the resultant DNA pellet twice with 70% ethanol, recovering the pellet of DNA by centrifugation after each wash. The final pellet is air-dried at ambient temperature, then resuspended in 5 μL of water. 3. Mix 1–2 μg of ligation solution with 40 μL of electrocompetent E. coli TG1 cells, and incubate on ice for 10 min. 4. Transfer electrocompetent cells/ligation mix to a 1 mm electroporation cuvette, and perform electroporation at 1.8 kV, 4–5 ms pulse. 5. Add 950 μL SOC recovery media to electroporated cells and incubate at 37  C for 1 h. 6. Take 100 μL of recovered E. coli culture and make a tenfold serial dilution to 106. Plate out 100 μL of each dilution onto

124

John Steven et al.

TYE-agar media plates supplemented with ampicillin at 100 μg/mL and glucose at 2% w/v (TYE-AG). Incubate plates overnight at 37  C, and then count colonies to determine library size. 7. After preparation of the serial dilution, plate out the remaining culture on a large bioassay plate containing 2 TY, ampicillin at 100 μg/mL, and glucose at 2% w/v (2 TY-AG), and incubate overnight at 37  C. 8. After determining the library size by counting the serial dilution plates, add 20 mL of 2 TY media to the bioassay plate, used to propagate the library, and recover the library by scraping off the resultant lawn of cells into media. Collect the library into a sterile 50 mL centrifuge tube, and pellet the cells at 4000  g for 10 min (see Note 7). 9. Resuspend the cell pellet in 5 mL of 2 TY-AG media containing 20% (v/v) glycerol. 10. Divide the recovered library into cryovials in 1 mL aliquots. The library can be stored in frozen aliquots prior to rescue, phage preparation, and panning against antigen. 3.3 Library Rescue and Selection 3.3.1 Library Phage Rescue

1. Add the required volume of the library E. coli TG1 stock to obtain an initial OD600 of 0.1 in 150 mL 2 TY-AG media (see Note 8). 2. Grow the culture to mid-log phase (OD600 0.4–0.6) in a 37  C, 250 rpm shaking incubator (see Note 9). 3. Transfer the culture into 50 mL sterile centrifuge tubes, and infect with M13K07 helper phage in a ratio of phage to bacterial cells of 20:1. Incubate without shaking in a 37  C water bath for 30 min, followed by incubation at 37  C for 1 h, with shaking at 150 rpm (see Note 10). 4. To determine the bacterial phage infection efficiency, use 10 μL of the infected culture to prepare serial dilutions (102 to 107) in 2 TY. Thoroughly and evenly spread 100 μL from each dilution onto TYE agar plates containing 100 μg/mL ampicillin, 50 μg/mL kanamycin, 2% (w/v) glucose (TYE-AKG), and TYE-AG agar plates. Incubate TYE agar plates at 30  C overnight in a static incubator (see Note 11). 5. Centrifuge the infected bacterial culture at 2500  g for 15 min at 25  C. 6. Discard the supernatant and resuspend the cell pellet in 150 mL of 2 TY-AK. Incubate the culture at 30  C, 250 rpm, overnight in a shaking platform incubator (see Note 12).

Maturation of a Humanized Shark VNAR Domain 3.3.2 Polyethylene Glycol (PEG) Precipitation

125

1. Centrifuge the overnight culture at 3200  g for 30 min at 4  C. 2. Collect the supernatant into a sterile 50 mL centrifuge tube containing 1/5th final volume ice-cold PEG/NaCl (20% (w/v) PEG, 2.5 M NaCl in distilled H2O). Mix thoroughly but gently and place the tube on ice for 2 h to allow for phage precipitation (see Note 13). 3. Centrifuge the PEG-phage mixture at 3200  g for 30 min at 4  C. Carefully discard the supernatant into Virkon disinfectant. To remove residual cellular debris, resuspend the resulting PEG-precipitated phage pellet in 40 mL sterile PBS, and centrifuge immediately at 3200  g for 30 min at 4  C. 4. Transfer the supernatant to a sterile 50 mL centrifuge tube containing 10 mL of ice-cold PEG/NaCl. Mix and place on ice for 30 min (see Note 14). 5. Centrifuge at 3200  g for 30 min at 4  C and discard supernatant into Virkon disinfectant. Resuspend the pellet in 1–2 mL of PBS. 6. Titrate the rescued phage by preparing a 100-fold dilution in sterile PBS, followed by serial dilutions of 102 to 1012. Transfer 100 μL of each serial dilution into fresh sterile Eppendorf tubes, and add 900 μL of log-phase TG1 cells. Incubate the Eppendorf tubes in a 37  C water bath for 30 min, to allow for infection. Gently mix the infected culture, and transfer 100 μL from each Eppendorf tube onto a corresponding TYE-AG agar plate. Thoroughly and evenly spread the culture on the agar plate. Incubate overnight at 37  C in a static incubator (see Note 15).

3.3.3 Phage Display Selection (See Note 16)

1. Coat a MaxiSorp immunotube with 1 μg/mL of the target antigen in 5 mL PBS. Seal the immunotube with laboratory parafilm, and incubate overnight at 4  C (see Note 17). 2. Inoculate 5 mL 2 TY with a single E. coli TG1 colony, and grow overnight at 37  C on a shaking platform at 250 rpm. 3. The next day, wash the coated immunotube, 3 with 5 mL PBS. Block with 5 mL of 2% (w/v) milk-PBS (M-PBS) for 2 h at 37  C. 4. Wash the immunotube, 3 with 5 mL PBS, before adding 300 μL of the rescued phage (from step 5, Subheading 3.3.2 above). Make up to 5 mL volume with 2% (w/v) M-PBS (see Note 18). 5. Incubate the immunotube at room temperature for 30 min, rotating at 15–20 rpm, followed by static incubation at room temperature for an additional 90 min.

126

John Steven et al.

6. During this time, inoculate 50 mL of 2 TY with the overnight E. coli TG1 culture until an initial OD600 of 0.1 is achieved. Incubate the freshly inoculated culture at 37  C, 250 rpm, and grow until OD600 of 0.4–0.6 is attained (see Note 19). 7. Discard the unbound phage from the immunotube into Virkon disinfectant. Wash the immunotube, 5 with 5 mL PBS containing 0.1% (v/v) tween 20 (PBS-T), followed by another 5 wash with 5 mL PBS (see Note 20). 8. Elute the bound phage by adding 1 mL of 100 mM triethylamine (TEA) into the immunotube (14 μL of 99% TEA in 1 mL dH2O). Rotate the immunotube continuously at 15–20 rpm for 10 min. Transfer the eluted phage into a sterile tube containing 500 μL 1 M Tris–HCl (see Note 21). 9. Add an additional 200 μL of 1 M Tris–HCl to the immunotube to neutralize the remaining phage. This additional eluate is also pooled with the initial eluate from step 8 above. 10. Add 0.75 mL of the neutralized eluted phage to a universal centrifuge tube containing 9.25 mL of exponential phase TG1 cells at OD600 0.4–0.5. Incubate the culture for 30 min in a 37  C water bath without shaking, to allow for TG1 bacteriophage infection. 11. For long-term storage, add glycerol to a final concentration of 15% (v/v) to the remaining 0.75 mL eluted phage, and store at 80  C. 12. Take 100 μL from the infected TG1 culture to prepare serial dilutions of 101 to 107 in a total volume of 1 mL, using 2 TY as the diluent. Dispense and evenly spread 100 μL of the 104 to 107 dilutions on TYE-AG agar plates in duplicate and single TYE agar plates for dilutions 102 and 103. Incubate agar plates at 37  C overnight. 13. Centrifuge the remaining infected TG1 culture at 3200  g for 10 min. Discard the supernatant and resuspend the cell pellet in 2 mL of 2 TY. Carefully spread onto a large bioassay dish of TYE-AG. Incubate overnight at 30  C (see Note 22). 14. The round 1 selection output size can be determined by counting single colonies on each dilution TYE agar plate (see Note 23). These single monoclonal TG1 cells can be picked for binding, protein expression, and DNA sequence characterization. 15. The round 1 selection output library is rescued following the previously described protocol (Subheading 3.3.1). This process is repeated once again for a second selection round. 16. For this project, 96-well monoclonal periplasmic protein expression and soluble protein-binding ELISA were conducted (see Note 24).

Maturation of a Humanized Shark VNAR Domain

3.4

Periprep ELISA

127

1. To identify target-specific binders in the selection outputs, isolated clones were expressed into the bacterial periplasm, and non-purified periplasmic extracts were screened by ELISA. In a 96 deep-well plate, inoculate 1 mL/well of 2 TY, 100 μg/mL ampicillin, 0.1% (w/v) glucose with 5 μL glycerol stock (see Note 25). 2. Cover the plate with a breathable seal, and allow the bacteria to grow for approximately 4 h at 37  C, 250 rpm, on a shaking platform (see Note 26). 3. Induce with 110 μL/well, 1 mM IPTG in 2 TY, and 100 μg/ mL ampicillin (final IPTG concentration of 0.1 mM), and incubate overnight at 25  C, 250 rpm, on a shaking platform. 4. Coat the required number of Nunc MaxiSorp ELISA plates with 100 μL/well of 1 μg/mL antigen, in 1 PBS, overnight at 4  C. To ensure VNAR specificity, prepare negative control plates of a nontarget antigen in parallel (see Note 27). 5. After 18 h of growth, pellet the cells by centrifugation at 2500  g for 10 min at 4  C and discard the supernatant. Tap the plate dry on paper towels to remove any residual liquid. 6. While spinning down cells, wash the ELISA plates 3 with 0.05% (v/v) Tween-20 in 1 PBS (PBS-T) and 200 μL/well. 7. Block plates with 2% (w/v) M-PBS and 200 μL/well for 1 h. 8. Resuspend the pellets in 150 μL/well of ice-cold periplasmic fractionation buffer by vortexing briefly. 9. Add 150 μL/well of 1:5 ice-cold periplasmic fractionation buffer (diluted in H2O). Incubate on ice for 30 min. 10. Following incubation, centrifuge as in step 5 above. 11. Remove 200 μL/well of the periplasmic fraction (supernatant), take care not to disturb the pellet, and add to an equal volume of 4% (w/v) M-PBS in a 96 deep-well plate; incubate for 1 h. 12. Transfer 100 μL/well of the blocked periplasmic fraction to the blocked target and control ELISA plates; incubate for 1 h. 13. Wash the ELISA plates 3 with PBS-T and add 100 μL/well anti-HA-HRP conjugate (1:1000 diluted in PBS-T); incubate for 1 h. 14. Wash ELISA plates 3 with PBS-T. 15. Develop ELISA by adding 100 μL/well of TMB substrate, and incubate until the appearance of signal/onset of saturation (approximately 5–10 min). Halt the reaction with the addition of 50 μL/well of 1 M H2SO4. 16. Measure the absorbance at 450 nm wavelength using EnVision 2104 multilabel microplate reader with Wallace EnVision software, and process the binding data using SigmaPlot 9.0.

128

John Steven et al.

3.5 Reformatting of Selected Clones for Transient Expression in Eukaryotic Cells

1. Clones showing OD450 by periprep ELISA of at least 80% of parental v1.10 or v2.4 were selected for batch conversion into the proprietary mammalian expression vector pSMED2. 2. Inoculate 2 TY-AG with positive clones selected from periprep ELISA screen in 96 deep-well plate, and incubate overnight at 37  C on a shaking platform, 250 rpm. 3. Harvest the cells by centrifugation, and prepare plasmid DNA from the cell pellets using BioRobot 8000 and QIAprep 96 Turbo BioRobot Kit in accordance with the manufacturer’s instructions. 4. Determine the concentration of the plasmid DNA samples by UV A260 spectrophotometry by mixing 1 μg of each, resulting in a mixed pool DNA sample. Dilute the pooled DNA sample to 100 ng/μL (see Note 28). This is the template DNA mix to be used in step 5. 5. In a thin-walled PCR tube, mix the 50 and 30 oligonucleotide primers containing the appropriate restriction enzyme sites for cloning, in frame into the eukaryotic expression vector, and add to 50 ng of the pooled template DNA mix from step 4. Use Phusion High-Fidelity PCR Master Mix and water to make up to the volume according to the manufacturer’s directions. Mix well, then cycle on a PCR block: 98  C for 2 min, then 30 cycles, 94  C for 30 s 55  C for 30 s, 72  C for 1 min. 6. Clean up the resultant amplicon, using Qiagen PCR clean-up kit in accordance with the manufacturer’s instruction. 7. Digest the cleaned up amplicon with the restriction enzymes correspondent to restriction sites introduced at 50 and 30 ends by the PCR primers. In this example, the enzymes used were BssHII and EcoR1. 8. Run the digested mutated amplicon out on an agarose TAE gel, excise the band, and clean up using a Qiagen gel extraction kit. 9. Use UV A260 spectrophotometry to quantify the recovered digested amplicon. 10. Set up the ligation using a vector/insert ratio of 1:3 (a 100 ng of BssHII/EcoR1-digested pSMED2 vector with 300 ng of BssHII/EcoR1-digested cleaned-up amplicon (insert)). Incubate overnight at 20  C. 11. Ethanol precipitate the ligated DNA and recover by centrifugation at 17,000  g for 20 min at 4  C. Wash the resultant DNA pellet twice with 70% ethanol, recovering the pellet of DNA by centrifugation after each wash. The final pellet is air-dried at an ambient temperature and then resuspended in 5 μL of water.

Maturation of a Humanized Shark VNAR Domain

129

12. Mix the ligation solution with 40 μL of electrocompetent E. coli TG1 cells, and incubate on ice for 10 min. 13. Transfer the electrocompetent cells/ligation mix to a 1 mm electroporation cuvette, and perform electroporation at 1.8 kV, 4–5 ms pulse. 14. Add 950 μL SOC recovery media to the electroporated cells, and incubate at 37  C for 1 h. 15. Take 100 μL of the recovered E. coli culture and make a tenfold serial dilution to 106. Plate out each dilution onto TYE-AG agar media plates and grow overnight at 37  C. 16. Select the dilution plates with defined individual colonies, and using a QPix2 XT colony picker, pick a fourfold overrepresentation of the selected batch-converted clones into 2 TY-AG and grow in 96 deep-well culture plates overnight. 17. Harvest the cells by centrifugation, and prepare the plasmid DNA from the cell pellets using a BioRobot 8000 and QIAprep 96 Turbo BioRobot Kit, in accordance with manufacturer instructions. 18. Determine the concentration of the plasmid DNA samples by UVA260 spectrophotometry, and prepare 200 ng of each DNA sample for small-scale HEK293 cell transfection and transient expression (see Note 29). 19. Prepare mid-log-phase HEK 293 cells in fresh Freestyle 293 expression media, at a density of 1  106 cells per mL, and dispense 200 μL aliquots into a 96-well tissue culture plate. 20. Express each clone in 200 μL of HEK293 cells. Transfect the cultures with 200 ng of plasmid DNA using lipofectamine in accordance with the supplier’s instructions, and then grow at 37  C in an air/8% CO2 atmosphere, while shaking at 750 rpm to maintain cells in suspension for 6 days. 21. Remove a 10 μL aliquot of the HEK293 culture media from each 200 μL transfection, and test for binding to HSA and HEL by ELISA as described in Subheading 3.4. Detect binding using an anti-polyhistidine peroxidase antibody. 22. Prepare media samples from clones with positive ELISAbinding signal to HSA for analysis, by off-rate ranking using SPR on a Biacore® T200 instrument. Dilute media samples from small transient transfections, 1 in 4, with HBS-EP buffer and 0.2 μm filter. Store filtered samples at 4  C prior to off-rate selection ranking. 3.6 Surface Plasmon Resonance (SPR) Off-Rate Selection

1. Perform the off-rate ranking on a T200 Biacore® instrument, in a 96-well plate format.

130

John Steven et al.

2. Harvest the crude supernatants from the small-scale (200 μL) HEK293 transient expression; dilute 1:4 in HBS-EP running buffer (10 stock) and 0.2 μm filter. 3. Immobilize the target ligand, HSA onto an activated CM5 series chip in 10 mM sodium acetate, pH 4.5 at 10 μL/min flow rate via covalent primary amine coupling and giving a coating density of approximately 300 response units (RU), using the Biacore® target ligand immobilization program. 4. Allow a 2 min association phase for all samples and a dissociation of 3 min, at a flow rate of 100 μL/min, followed by two 10 μL injections of 10 mM glycine, pH 1.5 at a flow rate of 100 μL/min. Perform all binding at 25  C in HBS-EP buffer. 5. Analyze the resulting sensorgrams with 1:1 global Langmuir binding model, using the Biacore® evaluation software. 6. Select the samples showing the slowest dissociation rates in comparison to the parental variant, and proceed to largerscale protein production for further characterization of purified proteins and sequence validation. 3.7 Sequence Determination, LargeScale Plasmid Preparation, Transient Expression, and Purification of Lead Clones

1. Identify unique lead clones (from Subheadings 3.5 and 3.6) by DNA sequence determination. 2. Select unique sequences for scale-up of plasmid preparation, transient expression, and IMAC purification of protein for further study (see Fig. 3).

Fig. 3 Sequences of v1.10 and v2.4 derived albumin-binding clones selected by surface plasmon resonance off-rate selection. (a) The amino acid sequence of selected clones from the v1.10 mutagenesis library. (b) The amino acid sequence of selected clones from the v2.4 mutagenesis library. Mutated residues differing from the parental sequence are highlighted. (Figure reproduced from Steven et al. (2017) [27] with permission from Frontiers Media Limited 131 Finsbury Pavement, WeWork, office 01-106 London EC2A 1NT United Kingdom)

Maturation of a Humanized Shark VNAR Domain

131

3. Grow 1 L overnight culture of each of the selected clones in 2 TY supplemented with antibiotic as appropriate. 4. Harvest the cells from each culture by centrifugation at 5000  g for 30 min, followed by plasmid DNA preparation using a Qiagen plasmid mega kit. 5. Determine the recovery and concentration of the plasmid DNA samples by UV A260 spectroscopy. 6. Prepare the DNA samples for transfection, into 1 L of log-phase HEK293 cells in Freestyle 293 expression media, by adding 1 mg of each plasmid DNA sample to a 25 mL aliquot of Opti-MEM serum-free media. 7. Add 2.5 mL of PEI solution at 1 mg/mL to fresh 25 mL aliquots of Opti-MEM media. 8. Incubate the DNA and PEI solutions at room temperature for 15 min, and then mix each DNA/Opti-MEM solution with an aliquot of PEI/Opti-MEM solution (so that the PEI and the DNA are at a ratio of 2.5:1). Mix well and allow PEI:DNA complex to form before addition to HEK293 cells. 9. HEK293 cells are grown in Freestyle 293 expression media in conical flasks at 37  C with shaking at 130 rpm (keeping cells in suspension) in 8% CO2/air. Split cells 1:3 when the cell density reaches 1.5–2.0  106 cells/mL, and maintain these growth conditions expanding the cells to give the cell numbers and volume required for transfection and expression. 10. For each transfection, prepare 1 L of HEK293 cells in fresh Freestyle 293 expression media at a density of 1  106 cells/ mL, and equilibrate to 37  C by shaking in 8% CO2/air at 130 rpm for 30 min. 11. To each 1 L of cells, add 1 sample of PEI:DNA complex mixture from step 9. 12. Incubate at 37  C, 130 rpm, 8% CO2/air. 13. 24 h post-transfection, supplement the media with tryptone to 0.5% (w/v) final volume from a 20% (w/v) stock made up in PBS. 14. Check the cell viability over a 3–5-day period, and harvest the media when the cell viability drops to below 50%. 15. For each transfection, remove the cells and cell debris from the harvested media samples by centrifugation at 5000  g for 15 min, followed by filtration of the clarified media through a 0.2 μm membrane filter. 16. Recover the expressed 6 histidine-tagged proteins from each sample by IMAC purification on an AKTA prime chromatography system, set up using the manufacturer’s protocol for HisTrap purification. Pool the fractions containing the eluted

132

John Steven et al.

6 histidine-tagged humanized VNAR protein, concentrate, and buffer exchange to PBS using Amicon® ultramembrane filtration units, with a 10,000 Da molecular weight cutoff. 17. Determine the recovery and the concentration of the protein samples by UV A280 spectroscopy. Store purified proteins at 4  C, prior to further characterization. 3.8 Characterization of Lead Clones 3.8.1 ELISA and EC50 Measurement

1. Coat a Nunc MaxiSorp 96-well ELISA plate with 1 μg/mL antigen (rat, mouse, and human serum albumin) in PBS, 100 μL/well, overnight at 4  C. 2. Wash the ELISA plate, 3 with PBS-T, 200 μL/well. 3. Block the plate with 4% (w/v) M-PBS, 200 μL/well, and incubate for 1 h at room temperature. 4. Wash the plate as described in step 2 above. 5. Dilute the purified VNAR protein variants and a 6 histidinetagged control, 1:3 in PBS, and use this as the top concentration, with further double dilutions across the ELISA plate in PBS, 100 μL/well. 6. Following 1 h incubation at room temperature, wash the ELISA plate, 3 with PBS-T, 200 μL/well. 7. Detect the antigen-bound VNARs with an anti-polyhistidine peroxidase-conjugated monoclonal antibody incubated for 1 h in PBS, 100 μL/well at room temperature. 8. Wash the ELISA plate, 3 with PBS-T, 200 μL/well, and develop by adding TMB substrate, 100 μL/well. When fully developed (approximately 5–10 min incubation at room temperature), halt the reaction with the addition of 50 μL/well 1 M H2SO4. 9. Measure the absorbance at 450 nm wavelength using EnVision 2104 multilabel microplate reader with Wallace EnVision software, and process the binding data using SigmaPlot 9.0. 10. Determine the half maximal effective concentration (EC50) values mathematically by derivation of the sigmoidal doseresponse curve via SigmaPlot 9.0 software (see Fig. 4).

3.9 Affinity Measurement Using Biacore® T200 3.9.1 Immobilization of Human Serum Albumin on CM5 Chip

1. Activate the surface of a research-grade CM5 with EDC-NHS for 7 min at a flow rate of 10 μL/min (see Note 30). 2. Prior to immobilization, dilute HSA in 10 mM sodium acetate buffer (pH 4.5) before loading onto the CM5 chip flow cell 2 at 5 μL/min (aiming for 300 RU). Immobilize rat (600 RU) and mouse (350 RU) serum albumin onto flow cells 3 and 4, respectively. 3. Block the remaining unoccupied activated groups on the CM5 chip with 1.0 M ethanolamine HCl (pH 8.5).

Maturation of a Humanized Shark VNAR Domain

133

Fig. 4 ELISA and EC50 determination of lead v1.10 and v2.4 humanized E06 clones. Purified E06 His-tagged VNAR protein and selected lead clones were added to wells of an HSA-coated plate at a concentration of 1 μg/ mL and serially diluted threefold across the plate. Four parameter logistic curve adjustments and EC50 calculations were performed with SigmaPlot 9.0. (a) Binding curves of selected lead clones derived from mutagenesis of the v1.10 backbone. (b) Binding curves of selected lead clones derived from mutagenesis of the v2.4 backbone. (c) EC50 values of lead v1.10 derived clones. (d) EC50 values of lead v2.4 derived clones. Except for v2.4-2G, clones obtained by mutagenesis of v1.10 performed better with several selected clones showing single-digit nanomolar or lower EC50 values. (Figure reproduced from Steven et al. (2017) [27] with permission from Frontiers Media Limited 131 Finsbury Pavement, WeWork, office 01-106 London EC2A 1NT United Kingdom)

3.9.2 Affinity Measurement of Humanized Anti-HSA VNARs

1. Prepare five concentrations of anti-HSA VNARs, ranging from 1.56 to 100 nM in HBS-EP buffer. Also include a zero concentration (running buffer) in the run for buffer signal correction. 2. Inject samples and record the binding association for 2 min and the dissociation of the complexes for 3 min. 3. Regenerate the surface with two 10 s pulses of 10 mM glycine, pH 1.5. 4. Subtract the reference and buffer signal prior to fitting curves to a 1:1 Langmuir binding model with Biacore® evaluation software (see Fig. 5).

134

John Steven et al.

Fig. 5 Biacore® Affinity Measurements of anti-albumin-humanized VNARs. The affinities of the selected leadhumanized anti-human serum albumin VNARs were determined by surface plasmon resonance using a Biacore® instrument. In addition, measuring affinity for human serum albumin, the affinities for rat and mouse albumin were also determined. Note that BB10 was no longer able to bind rodent albumin. The fold difference in affinities for albumin of the humanized VNAR variants was also calculated

3.10

Analytical SEC

1. Fit ZORBAX GF-200 or Agilent SEC 3 (see Note 3) to the Agilent 1200 HPLC system. 2. Set the detector wavelength to 280 nm. 3. Equilibrate the Agilent 1200 HPLC system and the column with PBS, pH 7.4, flow rate 1 mL per min for 20 min prior to use (flow rate may vary depending on analytical SEC column selected for use). This sets the baseline and allows the UV lamp to warm up.

Maturation of a Humanized Shark VNAR Domain

135

Fig. 6 Analytical SEC analysis of purified lead-humanized E06 proteins. SEC profiles of purified lead monomer proteins overlaid and offset to aid comparisons. Profiles are from left to right: v1.10, 2V, E06, BB10, BA11, BB11, 2G, and E06. Selected clones run as single peak unlike the double peak noted with sample v1.10. Note clone 2V protein was a control and has no affinity for albumin

4. Load the samples into the vials and place them into the autosampler. 5. Commence running samples in a predetermined sequence. 6. Assess chromatograms for elution profiles, selecting those that have a single peak with expected retention times and showing no signs of aggregation or multiple peaks with low retention (see Fig. 6).

4

Notes 1. 50 and 30 primers designed for the error-prone PCR should have restriction endonuclease sites compatible with the phage display cloning vector so that the digested amplicon can be cloned in frame and molecularly fused to a phage coat protein. 2. The phage display vector pWRIL-9 (ampicillin resistance) was used for this work, and the mutated amplicons were cloned in frame as NcoI/NotI fragments. This vector is comparable to pWRIL-1 [22] with the c-Myc tag replaced by an HA tag. Vector DNA should be digested with appropriate restriction endonucleases, purified, and stored frozen, ready for cloning of

136

John Steven et al.

the mutated amplicon repertoire. To ensure full mutational coverage of the humanized VNAR sequences, and therefore benefit from multiple substitutions in the protein sequences, conditions were optimized to deliver a maximum of 9 nucleotide mutations per VNAR sequence (VNAR sequence ~330 bp so mutational frequency desired 30 per kb). 3. Analytical size exclusion chromatography columns and systems from suppliers other than those listed here may also be used to assess the properties of purified proteins. 4. The digested amplicon band should be sharp. If the band appears as a smear, repeat the PCR adjusting conditions to give an amplicon that runs as a clear/sharp band postdigestion. 5. NanoVue UV/visible spectrophotometer requires a minimal sample volume (2 μL) to quantify and determine DNA concentrations, ensuring maximum amplicon availability for library construction. 6. The ratio of vector to the amount of amplicon (insert) used in the ligation can be optimized to ensure maximum library build quality (functional clone diversity) and transformation efficiency. 7. Library size of ~5  107 was achieved using 10 μg of ligated DNA in eight electroporations. For library quality assessment, after counting the serial dilution plates, individual colonies can be selected and grown in 2 TY-AG broth in 96 deep-well plates. Plasmid DNA from the resultant cultures can be prepared and sequenced and the mutational frequency of the library determined by sequence analysis (see Fig. 7). 8. The inoculation volume of library stock used in the initial culture is dependent on the library’s calculated or estimated size. To achieve a good coverage of the library’s diversity, we recommend that at least 10–100 copies of each clone are represented in the “starting” culture. For example, in a 50 mL starting culture with an OD600 of 0.1 (equivalent to 8  107 cells per mL), each clone in a 1  108 library is represented approximately 40. The starting culture volume should be adjusted accordingly for bigger library sizes. 9. Expect to measure culture absorbance at short time intervals using 1 mL culture per disposable cuvette, which is discarded after measurement. One impact of frequent measurements is the loss of culture volume; therefore, factor in an additional 10 mL of culture for these important absorbance measurements. 10. To calculate the volume of helper phage for library coinfection, measure OD600 of the cells (OD600 of 0.5 is equivalent to

Maturation of a Humanized Shark VNAR Domain

137

Fig. 7 Mutant libraries sequencing QC. Clones from both libraries were selected at random and sent for sequencing. The number of nonsilent mutations in each sequence was noted and the results plotted to show the mutation frequency in each library

4  108 cells/mL). The ratio of phage: bacterial cells should be 20:1; therefore take 20  4  108 ¼ 8  109 per mL of culture. For 50 mL culture, add 50  8  109 ¼ 4  1011 helper phage. 11. To ensure that enough cells have been infected to represent the diversity of the inoculum, it is necessary to determine the infection rate. Colonies formed in the absence of kanamycin represent viable bacterial cells, while colonies on ampicillin/ kanamycin agar plates represent phage-infected, bacterial cells. To calculate the infection rate (performed typically after an overnight incubation), initially count the number of colony-

138

John Steven et al.

forming units (cfu) on each of the dilution plates. The number of transformants ¼ the number of cfu/100 μL plated  the dilution factor  the recovered culture volume in mL. 12. It is important to omit glucose from the growth media to enable basal expression. 13. Use 1/5th final volume of ice-cold PEG/NaCl. This is typically a 40 mL phage supernatant added to 10 mL of ice-cold PEG/NaCl. Usually the precipitated phage becomes visible after 2 h incubation on ice. 14. Be careful not to discard the resulting supernatant at this step. Also swirling clouds of phage should be visible at this stage. The phage is now ready to be used in the selection process. Alternatively, the phage solution can be stored at 20  C for a few weeks or 80  C for several months. For long-term storage at 80  C, add 100% glycerol to a final concentration of 15% (v/v) PEG-phage solution, and aliquot into appropriate cryovials. 15. It is important to determine the number of rescued phage. Only transfer 100 μL of the infected culture from dilution tubes 102, 107, to 1012 onto the TYE agar plates. When calculating the rescued phage number, remember to correct for the dilution. We use the formula number of phage (cfu/mL) ¼ no. of colonies  dilution factor  100 per mL. 16. The correct presentation of target antigen during the selection of binders is critical and may affect the outcome of the selection process in terms of the frequency of “hit” isolation, affinity, and even functionality of identified binders. Therefore, it is important to apply a suitable target antigen immobilization strategy. In our group, we have routinely used direct target immobilization on solid support (solid phase) and biotinylated antigen captured on streptavidin-coated beads (solution phase) [23–26]. For this chapter, we have described a solid phaseselection technique. 17. The coating concentration of the target antigen from the first round of selection to the final selection round influences the stringency of the selection process, and this in turn may influence the binding quality (particularly affinity) of the selected clones. Here retention of high affinity was a key deliverable, and so a 1 μg/mL coating concentration was used to bias the isolation of high-affinity matured clones. In contrast, a first round of selection of an unbiased synthetic library may use a coating concentration of 5–10 μg/mL, and then stringency increased by decreasing this coating concentration in subsequent rounds of selection. Stringency may also be increased by performing more wash steps and increasing their duration.

Maturation of a Humanized Shark VNAR Domain

139

18. Ensure that 50% of the rescued phage is retained as backup stock. 19. It is imperative that the TG1 culture does not exceed an OD600 of 0.6. Typically, it should take approximately 90 min to reach OD600 of 0.4–0.6. 20. In this project, only two rounds of selection were performed. While the antigen concentration was maintained at 1 μg/mL in both selection rounds, additional stringency was incorporated into the second round of selection using a wash step of 10 PBS-T and 10 PBS. 21. During phage elution, do not exceed a 10 min incubation with TEA. TEA is very basic (pH 12) at 100 mM and is employed during phage elution to disrupt the interaction between the phage-displayed proteins (VNAR in this case) and the immobilized target antigen. Extended exposure of the phage-displayed proteins to the corrosive TEA might attenuate the phage (loss of infectivity). 22. To recover the output library from the large bioassay dish after an overnight incubation, add 10 mL 2 TY media to the dish and loosen the cells with a sterile spreader. Collect loose cells in a 50 mL centrifuge tube. Repeat this process two more times with fresh 10 mL 2 TY. Centrifuge the collected loose cells at 3200  g for 10 min at 4  C. Resuspend the cell pellet in 5 mL 2 TY containing 5% (w/v) glucose and 20% (v/v) glycerol. Store aliquots at 80  C. 23. In theory, selection round output size is an indication of either the enrichment or depletion of target antigen binders. Ideally, selection round output size is likely to enrich if we kept all conditions the same for a second round of selection or decrease if we increase stringency in the second round. Calculate selection round output size assuming there were 149 and 159 single colonies on the 105 dilution TYE agar plates, 48 and 16 single colonies on the 106 dilution TYE agar plates, and 8 colonies each on the 107 dilution TYE agar plates; For the 105 dilution: (149 cfu + 159 cfu)/2  105  10  10 mL ¼ 1.54  109. For the 106 dilution: (48 cfu + 16 cfu)/2  106  10  10 mL ¼ 3.2  109. For the 107 dilution: (8 cfu + 8 cfu)/2  107  10  10 mL ¼ 8  109. Then calculate the average ¼ 4.2  109 cfu in 10 mL. 24. Instead of a periplasmic protein expression followed by a soluble protein-binding ELISA, a monoclonal or polyclonal phage ELISA can be conducted using the same single colonies from the selection output library. However, on occasion a

140

John Steven et al.

monoclonal phage-binding ELISA might deliver false-positive binders (stickiness of phage) or binders with poor expressibility as soluble protein. Therefore, the more reliable screening strategy (especially for downstream application of the VNAR binders) relies upon soluble protein expressibility and binding to target antigen as a non-pIII-fused binding domain. 25. To enhance throughput selection of clones, use the glycerol stock of single clones (monoclonals) picked into 96-well plates (from the selection output) to inoculate 96 deep-well plates. Alternatively, deep-well plates can be inoculated with the overnight cultures of monoclonals. Overnight cultures can be made from glycerol stocks or freshly picked bacterial colonies. 26. For bacterial strains like TG1 and ER2738, an incubation time of 4 h should be sufficient for most clones; this time can be increased to 5 h. The OD600 (measured in 1 cm cuvettes) of cultures in deep wells, at the end of induction, should be between 1 and 2. 27. Other globular proteins such as thyroglobulin can be used for this purpose. 28. Mini-prep plasmid DNA samples, for reformatting into the proprietary pSMED2 eukaryotic expression vector, were prepared using a Qiagen robotic system in a 96-well plate format. The DNA concentration in each well was estimated by combining 10 μL aliquots from each well. UV spectroscopy was then used to determine the concentration of the pooled sample. This value was then taken as the average concentration of all samples on the plate and used to determine the sample volume that will give ~1 μg sample of each DNA. 29. For HEK293 cells, DNA for transient transfection should be used at 1 μg/mL final concentration in the transfection setup. For DNA prepared in a 96-well plate format, using robotic technology, the concentration of samples can be estimated by combining a 10 μL aliquot from each well. UV spectroscopy can then be used to determine the concentration of the pooled sample. This will give an average concentration of all samples on the plate. This concentration can be used to determine the sample volume that will give ~200 ng sample of each DNA for transfection into 200 μL HEK293 cells in a 96-well tissue culture plate. 30. The immobilization of HSA onto the CM5 chip was achieved using standard amine coupling. The first flow cell was used as a reference surface to correct for bulk refractive index, matrix effects, and non-specific binding.

Maturation of a Humanized Shark VNAR Domain

141

References 1. Smith GP (1985) Filamentous fusion phage: novel expression vectors that display cloned antigens on the virion surface. Science 228 (4705):1315–1317 2. Scott JK, Smith GP (1990) Searching for peptide ligands with an epitope library. Science 249(4967):386–390 3. McCafferty J, Griffiths AD, Winter G et al (1990) Phage antibodies: filamentous phage displaying antibody variable domains. Nature 348(6301):552 4. Lowman HB, Bass SH, Simpson N et al (1991) Selecting high-affinity binding proteins by monovalent phage display. Biochemistry 30 (45):10832–10838 5. Cadwell RC, Joyce GF (1992) Randomization of genes by PCR mutagenesis. PCR Methods Appl 2(1):28–33 6. Cirino PC, Mayer KM, Umeno D (2003) Generating mutant libraries using error-prone PCR. In: Anonymous directed evolution library creation. Springer, New York, pp 3–9 7. Gram H, Marconi LA, Barbas CF 3rd et al (1992) In vitro selection and affinity maturation of antibodies from a naive combinatorial immunoglobulin library. Proc Natl Acad Sci U S A 89(8):3576–3580 8. Deng SJ, MacKenzie CR, Sadowska J et al (1994) Selection of antibody single-chain variable fragments with improved carbohydrate binding by phage display. J Biol Chem 269 (13):9533–9538 9. Stephens DE, Singh S, Permaul K (2009) Error-prone PCR of a fungal xylanase for improvement of its alkaline and thermal stability. FEMS Microbiol Lett 293(1):42–47 10. Malm M, Kronqvist N, Lindberg H et al (2013) Inhibiting HER3-mediated tumor cell growth with affibody molecules engineered to low picomolar affinity by position-directed error-prone PCR-like diversification. PLoS One 8(5):e62791 11. Zhao N, Schmitt MA, Fisk JD (2016) Phage display selection of tight specific binding variants from a Hyperthermostable Sso7d scaffold protein library. FEBS J 283(7):1351–1367 12. Zahnd C, Spinelli S, Luginbuhl B et al (2004) Directed in vitro evolution and crystallographic analysis of a peptide-binding single chain antibody fragment (scFv) with low picomolar affinity. J Biol Chem 279(18):18870–18877. https://doi.org/10.1074/jbc.M309169200

13. Ye J, Wen F, Xu Y et al (2015) Error-prone pcr-based mutagenesis strategy for rapidly generating high-yield influenza vaccine candidates. Virology 482:234–243 14. Reichert JM (2017) Antibodies to watch in 2017. In: Anonymous MAbs, vol 9. Taylor & Francis, Milton Park, Didcot, p 167 15. Ian L (2017) Pharma R&D Annual Review. Informa UK Ltd Mortimer House, 37-41 Mortimer Street, London W1T3JH, UK 16. Hey T, Fiedler E, Rudolph R et al (2005) Artificial, non-antibody binding proteins for pharmaceutical and industrial applications. Trends Biotechnol 23(10):514–522 17. Gebauer M, Skerra A (2009) Engineered protein scaffolds as next-generation antibody therapeutics. Curr Opin Chem Biol 13 (3):245–255 18. Muyldermans S (2013) Nanobodies: natural single-domain antibodies. Annu Rev Biochem 82:775–797 19. Kovaleva M, Ferguson L, Steven J et al (2014) Shark variable new antigen receptor biologics–a novel technology platform for therapeutic drug development. Expert Opin Biol Ther 14 (10):1527–1539 20. Kovalenko OV, Olland A, Piche-Nicholas N et al (2013) Atypical antigen recognition mode of a shark immunoglobulin new antigen receptor (IgNAR) variable domain characterized by humanization and structural analysis. J Biol Chem 288(24):17408–17419. https:// doi.org/10.1074/jbc.M112.435289 21. Dooley H, Flajnik MF, Porter AJ (2003) Selection and characterization of naturally occurring single-domain (IgNAR) antibody fragments from immunized sharks by phage display. Mol Immunol 40(1):25–33 22. Finlay WJ, Cunningham O, Lambert MA et al (2009) Affinity maturation of a humanized rat antibody for anti-RAGE therapy: comprehensive mutagenesis reveals a high level of mutational plasticity both inside and outside the complementarity-determining regions. J Mol Biol 388(3):541–558 23. Winter G, Griffiths AD, Hawkins RE et al (1994) Making antibodies by phage display technology. Annu Rev Immunol 12 (1):433–455 24. Griffiths AD, Williams SC, Hartley O et al (1994) Isolation of high affinity human

142

John Steven et al.

antibodies directly from large synthetic repertoires. EMBO J 13(14):3245–3260 25. Hawkins RE, Russell SJ, Winter G (1992) Selection of phage antibodies by binding affinity: mimicking affinity maturation. J Mol Biol 226(3):889–896 26. Marks JD, Hoogenboom HR, Bonnert TP et al (1991) By-passing immunization: human

antibodies from V-gene libraries displayed on phage. J Mol Biol 222(3):581–597 27. Steven J, Mu¨ller MR, Carvalho MF et al (2017) In vitro maturation of a humanized shark Vnar domain to improve its biophysical properties to facilitate clinical development. Front Immunol 8:1361

Chapter 8 Antibody Phage Display: Antibody Selection in Solution Using Biotinylated Antigens Esther V. Wenzel, Kristian D. R. Roth, Giulio Russo, Viola Fu¨hner, Saskia Helmsing, Andre´ Frenzel, and Michael Hust Abstract Antibody phage display is the most used in vitro technology to generate recombinant, mainly human, antibodies as tools for research, for diagnostic assays, and for therapeutics. Up to now (autumn 2018), eleven FDA/EMA-approved therapeutic antibodies were developed using phage display, including the world best-selling antibody adalimumab. A key to generate successfully human antibodies in vitro is the choice of the most appropriate antibody selection method, for our goal. In this book chapter, we describe the antibody selection process (panning) in solution and its advantages over panning on immobilized antigens. Detailed protocols on the panning procedure and the screening of monoclonal binders are given. Key words Phage display, Panning, Antibody, ScFv

1

Introduction Antibody phage display is a key technology to generate antibodies, mainly human antibodies, in vitro, independent of the restriction of the immune system [1]. This technology is based on the work of G. P. Smith [2]. The antibody technology was developed independently on three sites in parallel and published in 1990/1991. In Cambridge (UK) by John McCafferty [3], by Carlos Barbas III in La Jolla (USA) [4], and in Heidelberg (Germany) by Breitling and Du¨bel who also invented the most common used phagemid system [5]. For phage display, antibody fragments, mainly the single-chain fragment variable (scFv) [6, 7] and the fragment antigen binding (Fab) [8, 9], are used, but also human or camel domain antibodies [10, 11]. The in vitro procedure for isolation of these antibody fragments from antibody gene libraries is called “panning” according to the gold washer procedure [2]. In the panning procedure, the antigen can be immobilized to a solid surface, such as column matrixes [5], nitrocellulose [12], magnetic beads [13], or, most

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2070, https://doi.org/10.1007/978-1-4939-9853-1_8, © Springer Science+Business Media, LLC, part of Springer Nature 2020

143

144

Esther V. Wenzel et al.

widely used, plastic surfaces with high protein-binding capacity as polystyrene tubes, respectively, microtiter wells (MTPs) [14]. A further strategy is to select antibodies in solution using biotinylated antigens followed by a “pull down“ with streptavidin beads [15]. This approach is also a preferred strategy when performing an affinity maturation of antibodies [16]. For the generation of antibodies against cell surface markers, e.g., cancer targets, the panning can be performed directly on cells [17, 18]. For the selection in solution, the first step is to incubate the antibody phage with the soluble biotinylated antigen before magnetic streptavidin beads are used for a pull down. Stringent washing is performed subsequently to remove the vast excess of nonbinding and weak-binding antibody phage. For the next panning round, the bound antibody phage will be eluted and used for infection of E. coli, followed by infection with the helper phage to amplify an enriched antigen-specific antibody phage fraction. This fraction can be used for further panning rounds until a significant enrichment of antigen-specific antibody phage is achieved. The number of antigen-specific antibody phage clones should increase with every panning round. Usually, 2–3 panning rounds are necessary to select specifically binding antibody fragments. For screening of monoclonal binders, they are produced as soluble monoclonal antibody fragments, or in rare cases as monoclonal antibody phage, in microtiter plates. These monoclonal antibodies can be identified by, e.g., ELISA [19], immunoblot [14], flow cytometer [20], or functional assay (Wenzel et al. in preparation). Subsequently, the gene fragments encoding the antibody fragments can be subcloned into any other antibody format, e.g., scFv-Fc or IgG [6, 8, 19, 21]. A schema of the selection procedure is given in Fig. 1. This selection procedure can be performed using patientderived immune libraries [22, 23] or naive libraries like the McCafferty library [7], Pfizer library [24], Tomlinson libraries [25], or HAL 7/8 and 9/10 [6, 26]. Antibody phage display libraries are valuable sources for the generation of antibodies against all kind of target structures including therapeutic targets. The first approved human antibody adalimumab was generated by a phage displaybased approach using the phagemid system [27]. Up to now (autumn 2018), eleven antibodies generated by phage display are FDA/EMA approved, whereas around 3 years ago, only six phage display-derived antibodies were approved. An overview about phage display-derived therapeutic antibodies by phage display is given by Frenzel et al. [28]. The following protocols describe the panning of scFv antibody fragments completely in solution. The antibody selection can be performed in 3 days and the screening and identification of monoclonal antibodies in 3 further days.

Panning in Solution

145

Fig. 1 Schema of antibody (scFv) phage display selection and screening. (Modified figure from former publications [28, 36])

2

Materials

2.1 Preparation of the Magnetic Beads

1. Protein LoBind tubes (Eppendorf). 2. Magnetic streptavidin beads Dynabeads™ M-280 streptavidin (Thermo Fisher Scientific). 3. Magnetic Rack (BioRad).

16-Tube

SureBeads™

Magnetic

Rack

4. Phosphate-buffered saline (PBS): 140 mM NaCl + 2.7 mM KCl + 1.8 mM KH2PO4 + 10 mM Na2HPO4·2H2O. 2.2

Panning

1. BSA-PBST: 2% BSA in PBS with 0.05% Tween 20, prepare fresh. 2. Overhead shaker such as Grant Bio PTR-30 (Grant Instruments) or equivalent instrumentation. 3. 10 μg/mL trypsin in PBS. 4. E. coli TG1 (supE thi-1 Δ(lac-proAB) Δ(mcrB-hsdSM)5 (rK
mK
) [F´ traD36 proAB lacIqZΔM15]). 5. Filter tips Biosphere® (Sarstedt). 6. M13K07 helper phage (Thermo Fisher Scientific).

146

Esther V. Wenzel et al.

7. Round bottom polypropylene (PP) deep-well 96 MTPs (Greiner). 8. Thermoshaker. 9. Eppendorf 5810R, rotor A-4-81 with MTP adapter. 10. 2 TY media pH 7.0: 1.6% (w/v) tryptone, 1% (w/v) yeast extract, 0.5% (w/v) NaCl. 11. 2 TY-T: 2 TY, 50 μg/mL tetracycline. 12. 10 GA: 1 M glucose, 1 mg/mL ampicillin. 13. 2 TY-GA: 2 TY, 100 mM glucose, 100 μg/mL ampicillin. 14. 2 TY-AK: 2 TY, 100 μg/mL ampicillin, 50 μg/mL kanamycin. 15. Glycerol (80%). 2.3

Phage Titration

1. E. coli XL1-Blue MRF0 (recA1 endA1 gyrA96 thi-1 hsdR17 supE44 relA1 lac [F0 proAB lacIqZΔM15 Tn10 (Tetr)]). 2. 2 TY-GA agar plates (2 TY-GA + 1.5% (w/v) agar–agar).

2.4 Production of Soluble Monoclonal Antibody Fragments in Microtiter Plates

1. 96-Well U-bottom polypropylene (PP) microtiter plates (Greiner Bio-One). 2. AeraSeal breathable sealing film (Excel Scientific). 3. Potassium phosphate buffer: 2.31% (w/v) (0.17 KH2PO4 + 12.54% (w/v) (0.72 M) K2HPO4.

M)

4. Buffered 2 TY pH 7.0: 1.6% (w/v) tryptone, 1% (w/v) yeast extract, 0.5% (w/v) NaCl, 10% (v/v) potassium phosphate buffer. 5. Buffered 2 TY-SAI: buffered 2 TY containing 50 mM saccharose + 100 μg/mL ampicillin + 50 μM isopropyl-beta-Dthiogalactopyranoside (IPTG). 2.5 ELISA of Soluble Monoclonal Antibody Fragments

1. Mouse α-myc-tag monoclonal antibody (9E10) (Sigma). 2. Goat α-mouse IgG polyclonal (Fab-specific) HRP conjugated (A0168) (Sigma). 3. TMB solution: mix 20 parts of TMB-A with 1 part of TMB-B directly prior use TMB-A: 50 mM citric acid, 30 mM potassium citrate, pH 4.1 TMB-B: 90% (v/v) ethanol, 10% (v/v) acetone, 10 mM tetramethylbenzidine, 80 mM H2O2 (store in opaque polyethylene bottle at 4  C). 4. 1 N H2SO4. 5. Oligonucleotide primers MHLacZ-Pro_f (50 GGCTCGTATG TTGTGTGG 30 ) and MHgIII_r (50 CTAAAGTTTTGTCG TCTTTCC 30 ).

Panning in Solution

3

147

Methods The time schedule for the complete procedure from antibody selection to identification on monoclonal antibodies is given in Table 1.

3.1 Preparation of the Magnetic Beads

1. Calculate needed amount (see Note 1). 2. Transfer the desired volume of beads to a tube. 3. Add an equal volume of PBS (or at least 1 mL). 4. Mix the tube (vortex for 5 s, or keep on an overhead shaker for at least 5 min). 5. Place the tube on a magnetic rack for 1 min and discard the supernatant. 6. Resuspend in the same volume PBS as the initial volume beads taken from the vial.

3.2

Panning

1. For library preincubation, incubate 1011 to 1012 antibody phage (you should use ~100 more phage particles compared to the library size) from the library in 600 μL BSA-PBST for 30 min at RT with overhead shaking. After 30 min, add specific amount of streptavidin beads from Subheading 3.1 (10 more

Table 1 Time schedule for antibody selection (panning) and screening of monoclonal antibodies Day Procedure steps

Preparation steps

0

–

– Overnight culture of E. coli TG1

1

– – – –

First panning round Infection of E. coli TG1 with eluted phage Infection with helper phage Antibody phage production overnight

– Overnight culture of E. coli TG1

2

– – – –

Second panning round Infection of E. coli TG1 with eluted phage Infection with helper phage Antibody phage production overnight

– Overnight culture of E. coli XL1-Blue MRF0

3

– –

–

–

Third panning round Infection of E. coli XL1-Blue MRF0 with eluted phage Titration on agar plates

4

– –

Picking clones for screening Culture overnight

–

5

–

Production of soluble scFv overnight

– Coating of MTP wells for screening ELISA

6

–

Screening ELISA

–

148

Esther V. Wenzel et al.

binding capacity than needed, see Note 1), and incubate for 60 min at RT with overhead shaking. This step removes unspecific binders which can be present in the antibody gene libraries due to incorrect folding of individual antibodies. Preincubation with uncoupled beads further allows getting rid of unwanted streptavidin-specific binders. 2. Place the tube on a magnetic rack, wait for 1 min to let the beads attach on the tube wall. 3. Carry over the preincubated antibody phage library to a new protein LoBind tube, and add 100–500 ng biotinylated antigen. Incubate at RT for 2 h with overhead shaking for binding of the antibody phage (see Note 2). 4. Add specific amount of streptavidin beads, and incubate for 30 min with overhead shaking at RT to pull down the phageantibody::biotinylated antigen complex with the beads. 5. Place the tube on a magnetic rack, wait for 5 min. 6. Remove the unspecifically bound antibody phage by stringent washing. Therefore, wash the tubes 10 with PBST in the first panning round. In the following panning rounds, increase the number of washing steps (20 in the second panning round, 30 in the third panning round) (see Note 3). 7. Elute bound antibody phage with 150 μL trypsin solution for 30 min at 37  C (see Note 4). 8. Place the tube on a magnetic rack, wait for 1 min, collect supernatant containing eluted phages. 9. After the third panning round, use 10 μL of the eluted phage for titration (see Subheading 3.3). 10. Inoculate 50 mL 2 TY with an overnight culture of E. coli TG1 (see Note 5) in 100 mL Erlenmeyer flasks, and incubate at 250 rpm and 37  C until exponential growth phase is reached O.D.600 0.4–0.5 (see Note 6). 11. Fill 150 μL exponentially growing E. coli TG1 in a polypropylene (PP) deep-well MTP well, and mix with 150 μL of the eluted phage. Incubate the bacteria for 30 min at 37  C without shaking and 30 min at 37  C and 650 rpm (see Note 7). 12. Add 1150 μL of 2 TY-GA (see Note 8) and incubate for 1 h at 37  C and 650 rpm. 13. Infect the bacteria with 2  1011 phage particles/mL (¼1  1010 phage particles, MOI 1:20) M13K07 helper phage (use filter tips to avoid potential cross-contamination!). Incubate for 30 min at 37  C without shaking, followed by 30 min at 37  C and 650 rpm. 14. Centrifuge the MTP plate at 3220  g (e.g., use Eppendorf 5810R, Rotor A-4-81 with MTP carriers). Remove the

Panning in Solution

149

complete supernatant with a pipette. Do not destroy the pellet (see Note 9). 15. Add 950 μL 2 TY-AK and incubate overnight at 30  C and 650 rpm to produce new antibody phage. 16. Centrifuge the MTP plate at 3220  g. Transfer the supernatant (~1  1012 scFv-phage/mL) into a new cryovial. The produced antibody phages can directly be used for the next panning round. 3.3

Phage Titration

1. Inoculate 30 mL 2 TY-T in a 100 mL Erlenmeyer flask with E. coli XL1-Blue MRF0 (see Note 10), and grow overnight at 37  C and 250 rpm. 2. Inoculate 50 mL 2 TY-T with 500 μL overnight culture, and grow at 250 rpm at 37  C up to O.D.600 ~ 0.5 (see Note 6). 3. Make serial dilutions of the phage suspension in PBS (use filter tips). The number of eluted phage depends on several parameters (e.g., antigen, library, panning round, washing stringency, etc.). In case of a successful enrichment, the titer of eluted phage usually is 103 to 105 phages per well after the first panning round and increases two to three orders in magnitude per additional panning round (see Note 11). The phage preparation after reamplification of the eluted phage has a titer of about 1012 to 1013 phage/mL. 4. Infect 50 μL bacteria with 10 μL phage dilution (use filter tips), and incubate for 30 min at 37  C. 5. You can perform titrations in two different ways: (a) Plate the 60 μL infected bacteria on 2 TY-GA agar plates (9 cm Petri dishes). (b) Pipet 10 μL (in triplicate) on 2 TY-GA agar plates. Here, about 20 titering spots can be placed on one 9 cm Petri dish. Dry drops on work bench. 6. Incubate the plates overnight at 37  C. 7. Count the colonies and calculate the cfu or cfu/mL titer according to the dilution.

3.4 Production of Soluble Monoclonal Antibody Fragments in Microtiter Plates

1. Fill each well of a 96-well U-bottom PP MTP with 150 μL 2 TY-GA. 2. Pick 92 clones with sterile tips from the third panning round, and inoculate each well (see Note 12). Seal the plate with a breathable sealing film. 3. Incubate overnight in a microtiter plate shaker at 37  C and 850 rpm.

150

Esther V. Wenzel et al.

4. (a) Fill a new 96-well polypropylene microtiter plate with 180 μL 2 TY-GA, and add 10 μL of the overnight cultures. Incubate for 2 h at 37  C and 850 rpm. (b) Add 30 μL glycerol solution to the remaining 140 μL overnight cultures. Mix by pipetting and store this masterplate at
80  C. 5. Pellet the bacteria in the microtiter plates by centrifugation for 10 min at 3200  g and 4  C. Remove 190 μL glucosecontaining media by carefully pipetting (do not disturb the pellet) (see Note 9). 6. Add 180 μL buffered 2 TY-SAI (containing saccharose, ampicillin, and 50 μM IPTG), and incubate overnight at 30  C and 850 rpm (see Notes 13 and 14). 7. Pellet the bacteria by centrifugation for 10 min at 3200  g in the microtiter plates. Transfer the antibody fragment containing supernatant to a new polypropylene microtiter plate, and use it directly or store at 4  C but not longer than 3 days. 3.5 ELISA of Soluble Monoclonal Antibody Fragments

1. To analyze the antigen specificity of the monoclonal soluble antibody fragments, coat 100–200 ng antigen per well of an MTP plate overnight at 4  C. As control, coat 100–200 ng BSA or streptavidin per well (see Note 15). 2. Wash the coated microtiter plate wells 3 with PBST (washing procedure see Note 16). 3. Block the antigen-coated wells with MPBST for 1 h at RT. The wells must be filled completely. Empty the wells after blocking. 4. Fill 50 μL MPBST in each well and add 50 μL of antibody solution (coming out of Subheading 3.4). Incubate for 1 h at RT (or overnight at 4  C). 5. Wash the microtiter plate wells 3 with PBST (see Note 16). 6. Incubate 100 μL mouse 9E10 α-myc tag antibody solution for 1 h at RT (appropriate dilution in MPBST). 7. Wash the microtiter plate wells 3 with PBST (see Note 16). 8. Incubate 100 μL goat α-mouse HRP conjugate secondary antibody (1:10,000 in MPBST) 1 h at RT. 9. Wash the microtiter plate wells 3 with PBST (see Note 16). 10. Shortly before use, mix 20 parts TMB substrate solution A and 1 part TMB substrate solution B. Add 100 μL of this TMB solution into each well and incubate for 1–15 min at RT. 11. Stop the color reaction by adding 100 μL 1 N sulfuric acid. The color turns from blue to yellow. 12. Measure the extinction at 450 nm using an ELISA reader (see Note 17).

Panning in Solution

151

13. Identify positive candidates with a signal (on antigen) 10 over noise (on control protein, e.g., streptavidin or BSA) (see Note 18). 14. Sequence the DNA of the selected scFv with the oligonucleotide primers MHLacZ-Pro_f and MHgIII_r. We suggest analyzing the antibody sequences using VBASE2 (www.vbase2. org) (Tool: Fab/scFab/scAb/scFv Analysis).

4

Notes 1. Calculate the amount of bead per biotinylated antigen according to the manufacturer protocol, then use 10 the calculated amount. Double this amount of beads so to have one half for library preincubation and the other half for the panning selection process. 2. Instead of using a LoProtein Bind tube, a normal 1.5 mL Eppendorf tube can be used when preincubated 1 h with BSA-PBST and overhead shake. 3. The washing is performed manually. Vortex tube for 5 s, place the tube on a magnetic rack, wait for 1 min, discard supernatant, and finally add fresh PBST to repeat the washing step. 4. Phagemids like pHAL14 [6, 29] or pHAL30 [26] have coding sequences for a trypsin-specific cleavage site between the antibody fragment gene and the gIII. Trypsin also cleaves within antibody fragments but does not degrade the phage particles including the pIII that mediates the binding of the phage to the F pili of E. coli required for the infection. We observed that proteolytic cleavage of the antibody fragments from the antibody::pIII fusion by trypsin increases not only the elution but also enhances the infection rate of eluted phage particles, especially when using hyperphage as helper phage. 5. E. coli TG1 is growing much faster compared to XL1-Blue MRF0. This strain allows to perform one panning round per day. 6. If the bacteria have reached O.D.600 ~ 0.5 before they are needed, store the culture immediately on ice to maintain the F pili on the E. coli cells for up to 1 h. M13K07 helper phage (kan+) or other scFv-phage (amp+) can be used as positive control to check the infectibility of the E. coli cells. 7. After 1 h, a concentration of O.D.600 0.4–0.5 is reached, corresponding to ~4  108 bacteria/mL. 8. Higher concentration of glucose is necessary to efficiently repress the lac promoter controlling the antibody::pIII fusion gene on the phagemid. Low glucose concentrations lead to an

152

Esther V. Wenzel et al.

inefficient repression of the lac promoter and background expression of the antibody::pIII fusion protein. Background antibody expression represents a strong selection pressure frequently causing mutations in the phagemid, especially in the promoter region and the antibody::pIII fusion gene. Bacteria with this kind of mutations in the phagemids proliferate faster than bacteria with nonmutated phagemids. Therefore, 100 mM glucose must be included in every step of E. coli cultivation except during the phage production! 9. To not destroy the pellet, remove the supernatant carefully by having the pipette tip on the side of the well. An alternative is to manually shake out the supernatant (do it with a fast movement of your wrist). 10. Use E. coli XL1-Blue MRF0 for titering and production of soluble antibodies. The plasmid quality and yield in this strain are higher compared to TG1. 11. When the antibody gene library was packaged using hyperphage, the titer of the eluted phage after the second panning may not increase as strongly or even decreases slightly due to the change from oligovalent to monovalent display. 12. We recommend picking 92 clones when using a 96-well microtiter plate. Use the wells H3, H6, H9, and H12 for controls. H3 and H6 are negative controls; these wells will not be inoculated and just contain the bacteria medium. We inoculate the wells H9 and H12 with a clone containing a phagemid encoding a known antibody fragment. In ELISA, the wells H9 and H12 are coated with the antigen corresponding to the control antibody fragment in order to check scFv production and detection by ELISA. 13. The appropriate IPTG concentration for induction of antibody or antibody::pIII expression depends on the vector design. A concentration of 50 μM is well suited for vectors with a Lac promoter like pIT2 [30], pHENIX [31], pHAL14 [6], or pHAL30 [26]. 14. Buffered culture media and the addition of saccharose enhance the production of many but not all scFvs [32]. We observed that antibody::pIII fusion proteins and antibody phage sometimes show differences in antigen binding in comparison to soluble antibody fragments, because some antibodies can bind the corresponding antigen only as pIII fusion [33, 34]. Therefore, we recommend performing the screening procedure only by using soluble antibody fragment, to avoid false-positive binders. On the other hand, some scFv binding as antibody phage, but not as soluble scFv, binds as scFv-Fc after recloning.

Panning in Solution

153

15. While using biotinylated antigens for panning in solution, we recommend using streptavidin as control for the screening ELISA to identify possible biotin binders. 16. Microtiter plate well washing should be performed with an ELISA washer (e.g., TECAN Columbus Plus). To remove antigen or blocking solutions, wash 3 with PBST (“standard washing protocol” for TECAN washer). If no ELISA washer is available, wash manually 3 with PBST. 17. A measurement at 450 nm and 620–650 nm as reference wavelength can improve the readout pattern in ELISA. 18. The background (noise) signals should be about O. D.450 ~ 0.02 after 5–30 min TMB incubation time.

Acknowledgments This review contains updated and revised parts of former protocols [35]. References 1. Winter G, Milstein C (1991) Man-made antibodies. Nature 349:293–299. https://doi. org/10.1038/349293a0 2. Parmley SF, Smith GP (1988) Antibodyselectable filamentous fd phage vectors: affinity purification of target genes. Gene 73:305–318 3. McCafferty J, Griffiths AD, Winter G, Chiswell DJ (1990) Phage antibodies: filamentous phage displaying antibody variable domains. Nature 348:552–554. https://doi.org/10. 1038/348552a0 4. Barbas CF, Kang AS, Lerner RA, Benkovic SJ (1991) Assembly of combinatorial antibody libraries on phage surfaces: the gene III site. Proc Natl Acad Sci U S A 88:7978–7982 5. Breitling F, Du¨bel S, Seehaus T, Klewinghaus I, Little M (1991) A surface expression vector for antibody screening. Gene 104:147–153. https://doi.org/10. 1016/0378-1119(91)90244-6 6. Hust M, Meyer T, Voedisch B, Ru¨lker T, Thie H, El-Ghezal A, Kirsch MI, Schu¨tte M, Helmsing S, Meier D, Schirrmann T, Du¨bel S (2011) A human scFv antibody generation pipeline for proteome research. J Biotechnol 152:159–170. https://doi.org/10.1016/j. jbiotec.2010.09.945 7. Schofield DJ, Pope AR, Clementel V, Buckell J, Chapple SD, Clarke KF, Conquer JS, Crofts AM, Crowther SR, Dyson MR, Flack G, Griffin GJ, Hooks Y, Howat WJ, Kolb-Kokocinski A,

Kunze S, Martin CD, Maslen GL, Mitchell JN, O’Sullivan M, Perera RL, Roake W, Shadbolt SP, Vincent KJ, Warford A, Wilson WE, Xie J, Young JL, McCafferty J (2007) Application of phage display to high throughput antibody generation and characterization. Genome Biol 8:R254. https://doi.org/10.1186/gb-20078-11-r254 8. Hoet RM, Cohen EH, Kent RB, Rookey K, Schoonbroodt S, Hogan S, Rem L, Frans N, Daukandt M, Pieters H, van Hegelsom R, Neer NC, Nastri HG, Rondon IJ, Leeds JA, Hufton SE, Huang L, Kashin I, Devlin M, Kuang G, Steukers M, Viswanathan M, Nixon AE, Sexton DJ, Hoogenboom HR, Ladner RC (2005) Generation of high-affinity human antibodies by combining donor-derived and synthetic complementarity-determining-region diversity. Nat Biotechnol 23:344–348. https://doi. org/10.1038/nbt1067 9. Omar N, Lim TS (2018) Construction of Naive and Immune Human Fab Phage-Display Library. Methods Mol Biol 1701:25–44. https://doi.org/10.1007/978-1-4939-74474_2 10. Holt LJ, Herring C, Jespers LS, Woolven BP, Tomlinson IM (2003) Domain antibodies: proteins for therapy. Trends Biotechnol 21:484–490. https://doi.org/10.1016/j. tibtech.2003.08.007

154

Esther V. Wenzel et al.

11. Roma˜o E, Poignavent V, Vincke C, Ritzenthaler C, Muyldermans S, Monsion B (2018) Construction of high-quality camel immune antibody libraries. Methods Mol Biol 1701:169–187. https://doi.org/10.1007/ 978-1-4939-7447-4_9 12. Hawlisch H, Mu¨ller M, Frank R, Bautsch W, Klos A, Ko¨hl J (2001) Site-specific anti-C3a receptor single-chain antibodies selected by differential panning on cellulose sheets. Anal Biochem 293:142–145 13. Moghaddam A, Borgen T, Stacy J, Kausmally L, Simonsen B, Marvik OJ, Brekke OH, Braunagel M (2003) Identification of scFv antibody fragments that specifically recognise the heroin metabolite 6-monoacetylmorphine but not morphine. J Immunol Methods 280:139–155 14. Hust M, Maiss E, Jacobsen H-J, Reinard T (2002) The production of a genus-specific recombinant antibody (scFv) using a recombinant potyvirus protease. J Virol Methods 106:225–233 15. Schu¨tte M, Thullier P, Pelat T, Wezler X, Rosenstock P, Hinz D, Kirsch MI, Hasenberg M, Frank R, Schirrmann T, Gunzer M, Hust M, Du¨bel S (2009) Identification of a putative Crf splice variant and generation of recombinant antibodies for the specific detection of Aspergillus fumigatus. PLoS One 4:e6625. https://doi.org/10. 1371/journal.pone.0006625 16. Thie H, Toleikis L, Li J, von Wasielewski R, Bastert G, Schirrmann T, Esteves IT, Behrens CK, Fournes B, Fournier N, de Romeuf C, Hust M, Du¨bel S (2011) Rise and fall of an anti-MUC1 specific antibody. PLoS One 6: e15921. https://doi.org/10.1371/journal. pone.0015921 17. Keller T, Kalt R, Raab I, Schachner H, Mayrhofer C, Kerjaschki D, Hantusch B (2015) Selection of scFv antibody fragments binding to human blood versus lymphatic endothelial surface antigens by direct cell phage display. PLoS One 10:e0127169. https://doi.org/10. 1371/journal.pone.0127169 18. Rezaei J, RajabiBazl M, Ebrahimizadeh W, Dehbidi GR, Hosseini H (2016) Selection of single chain antibody fragments for targeting prostate specific membrane antigen: a comparison between cell-based and antigen-based approach. Protein Pept Lett 23:336–342 19. Frenzel A, Ku¨gler J, Wilke S, Schirrmann T, Hust M (2014) Construction of human antibody gene libraries and selection of antibodies by phage display. Methods Mol Biol 1060:215–243. https://doi.org/10.1007/ 978-1-62703-586-6_12

20. Ayriss J, Woods T, Bradbury A, Pavlik P (2007) High-throughput screening of single-chain antibodies using multiplexed flow cytometry. J Proteome Res 6:1072–1082. https://doi.org/ 10.1021/pr0604108 21. J€ager V, Bu¨ssow K, Wagner A, Weber S, Hust M, Frenzel A, Schirrmann T (2013) High level transient production of recombinant antibodies and antibody fusion proteins in HEK293 cells. BMC Biotechnol 13:52. https://doi.org/10.1186/1472-6750-13-52 22. Trott M, Weiβ S, Antoni S, Koch J, von Briesen H, Hust M, Dietrich U (2014) Functional characterization of two scFv-Fc antibodies from an HIV controller selected on soluble HIV-1 Env complexes: a neutralizing V3- and a trimer-specific gp41 antibody. PLoS One 9: e97478. https://doi.org/10.1371/journal. pone.0097478 23. Chan SW, Bye JM, Jackson P, Allain JP (1996) Human recombinant antibodies specific for hepatitis C virus core and envelope E2 peptides from an immune phage display library. J Gen Virol 77(10):2531–2539 24. Glanville J, Zhai W, Berka J, Telman D, Huerta G, Mehta GR, Ni I, Mei L, Sundar PD, Day GMR, Cox D, Rajpal A, Pons J (2009) Precise determination of the diversity of a combinatorial antibody library gives insight into the human immunoglobulin repertoire. Proc Natl Acad Sci U S A 106:20216–20221. https://doi.org/10. 1073/pnas.0909775106 25. de Wildt RM, Mundy CR, Gorick BD, Tomlinson IM (2000) Antibody arrays for highthroughput screening of antibody-antigen interactions. Nat Biotechnol 18:989–994. https://doi.org/10.1038/79494 26. Ku¨gler J, Wilke S, Meier D, Tomszak F, Frenzel A, Schirrmann T, Du¨bel S, Garritsen H, Hock B, Toleikis L, Schu¨tte M, Hust M (2015) Generation and analysis of the improved human HAL9/10 antibody phage display libraries. BMC Biotechnol 15:10. https://doi.org/10.1186/s12896-015-01250 27. Osbourn J, Groves M, Vaughan T (2005) From rodent reagents to human therapeutics using antibody guided selection. Methods 36:61–68. https://doi.org/10.1016/j.ymeth. 2005.01.006 28. Frenzel A, Schirrmann T, Hust M (2016) Phage display-derived human antibodies in clinical development and therapy. MAbs 8:1177–1194. https://doi.org/10.1080/ 19420862.2016.1212149 29. Kirsch M, Hu¨lseweh B, Nacke C, Ru¨lker T, Schirrmann T, Marschall H-J, Hust M, Du¨bel

Panning in Solution S (2008) Development of human antibody fragments using antibody phage display for the detection and diagnosis of Venezuelan equine encephalitis virus (VEEV). BMC Biotechnol 8:66. https://doi.org/10.1186/ 1472-6750-8-66 30. Goletz S, Christensen PA, Kristensen P, Blohm D, Tomlinson I, Winter G, Karsten U (2002) Selection of large diversities of antiidiotypic antibody fragments by phage display. J Mol Biol 315:1087–1097 31. Finnern R, Pedrollo E, Fisch I, Wieslander J, Marks JD, Lockwood CM, Ouwehand WH (1997) Human autoimmune anti-proteinase 3 scFv from a phage display library. Clin Exp Immunol 107:269–281 32. Hust M, Steinwand M, Al-Halabi L, Helmsing S, Schirrmann T, Du¨bel S (2009) Improved microtitre plate production of single chain Fv fragments in Escherichia coli. New Biotechnol 25:424–428. https://doi.org/10. 1016/j.nbt.2009.03.004 33. Goffinet M, Chinestra P, Lajoie-Mazenc I, Medale-Giamarchi C, Favre G, Faye J-C

155

(2008) Identification of a GTP-bound Rho specific scFv molecular sensor by phage display selection. BMC Biotechnol 8:34 34. Lillo AM, Ayriss JE, Shou Y, Graves SW, Bradbury ARM (2011) Development of phagebased single chain Fv antibody reagents for detection of Yersinia pestis. PLoS One 6: e27756. https://doi.org/10.1371/journal. pone.0027756 35. Russo G, Meier D, Helmsing S, Wenzel E, Oberle F, Frenzel A, Hust M (2018) Parallelized antibody selection in microtiter plates. In: Hust M, Lim TS (eds) Phage display: methods and protocols. Springer, New York, NY, pp 273–284 36. Kuhn P, Fu¨hner V, Unkauf T, Moreira GMSG, Frenzel A, Miethe S, Hust M (2016) Recombinant antibodies for diagnostics and therapy against pathogens and toxins generated by phage display. Proteomics Clin Appl 10:922–948. https://doi.org/10.1002/prca. 201600002

Chapter 9 Assessing Antibody Specificity in Human Serum Using Deep Sequence-Coupled Biopanning Kathryn M. Frietze, Susan B. Core, Alexandria Linville, Bryce Chackerian, and David S. Peabody Abstract Affinity selection using phage-display technologies is a powerful tool for identifying the peptide epitopes of monoclonal antibodies. Coupling affinity selection with deep sequencing technologies allows for the broad assessment of selectant populations. Here, we describe a method for using a phage-display platform to assess antibody specificity in human serum. We describe the method with reference to the bacteriophage MS2 virus-like particle (VLP) platform, but it can be adapted to other phage-display technologies as well. Key words Virus-like particle, Affinity selection, Deep sequencing, Serum, Antibodies, Pathogens, Phage display

1

Introduction Affinity selection using phage display is an important molecular biology tool that has helped identify the epitopes of monoclonal antibodies as well as the specificity of antibodies in polyclonal serum [1–6]. Affinity selection relies on several important features: [1] the ability of the phage to display foreign antigens on its surface, [2] the construction of a library of phage displaying diverse foreign antigens, and [3] the link between the phenotype of an individual phage (i.e., the foreign antigen displayed on the surface of the phage) to its genotype (i.e., the nucleic acid encoding the foreign antigen encapsidated within the phage). Mapping the epitopes of monoclonal antibodies is a straightforward application of the technology. However, several phage-based platforms, including T7 [2–4], M13 [7, 8], and MS2 virus-like particles (VLPs) [5, 6], have also been used to more broadly map the targets of antibody responses after infection. Coupling affinity selections with deep sequence technology can provide an in-depth assessment of antibody specificities in polyclonal serum.

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2070, https://doi.org/10.1007/978-1-4939-9853-1_9, © Springer Science+Business Media, LLC, part of Springer Nature 2020

157

158

Kathryn M. Frietze et al.

1. Isolate IgG

IgG

Library

2. Pan library against purified IgG 3. Purify RNA from the selectants Purified coding-RNA

4. Generate cDNA from the selectants

5. PCR amplification of cDNA

Non-binders

Binders

6. Prepare samples for sequence analysis

Amplified cDNA

7. Deep sequencing 8. Data analysis

Deep Sequence and analysis of the population

Fig. 1 Schematic of deep sequence-coupled biopanning. Isolated IgG is mixed with a VLP library of interest and allowed to bind. IgG is then pulled down using protein G magnetic beads, and nonbinding VLPs are washed away. Bound VLPs are then eluted, and associated RNA is recovered by RNA extraction, reverse-transcriptase PCR, and subsequent deep sequencing by Ion Torrent technology to assess the selectant population

In this chapter, we describe a step-by-step protocol for performing a single round of selection with a human serum sample and a pathogen or antigen-specific library (Fig. 1). This protocol was developed using a bacteriophage MS2 virus-like particle (VLP) platform, but can be adapted to utilize other phage-display platforms by altering the primer sequences indicated in this protocol and adapting the data processing scripts. The protocol can be divided into eight steps: [1] isolation of IgG from human serum, [2] library selection using the purified IgG, [3] purification of nucleic acid from selectants, [4] generation of cDNA from selected library members, [5] PCR amplification of cDNA, [6] preparation of samples for sequence analysis, [7] deep sequencing by Ion Torrent technology, and [8] data processing.

2

Materials

2.1 Isolation of IgG from Human Serum

1. Human serum/plasma (see Note 1). 2. Dynabeads™ protein G (ThermoFisher Scientific). 3. Phosphate-buffered saline (PBS): 137 mM NaCl, 8.1 mM Na2HPO4, 3.4 mM KCl, 29.4 mM KH2PO4. 4. Elution buffer: 0.1 M glycine–HCl, pH 2.7.

Deep Sequence-Coupled Biopanning

159

5. Neutralization buffer: 1 M Tris, pH 9.0. 6. DynaMag™ Magnet (ThermoFisher Scientific). 7. Vortex shaker (see Note 2). 8. Microvolume spectrophotometer (i.e., NanoDrop). 2.2 Selection of a Library Against Isolated Human IgG

1. PBS. 2. Library: Bacteriophage MS2 virus-like particle (VLP) library displaying peptides from an antigen or pathogen of interest, generated as described in steps 6 and 9. This library should contain enough transformants to have at least tenfold overrepresentation of the library members (see Note 3). 3. Dynabeads™ protein G (ThermoFisher Scientific). 4. DynaMag™ Magnet (ThermoFisher Scientific). 5. Vortex shaker (see Note 2). 6. PBS/Tween: PBS, 0.05% v/v Tween®20. 7. Mini-centrifuge capable of holding 1.7 mL tubes. 8. Elution buffer: 0.1 M glycine–HCl, pH 2.7. 9. Neutralization buffer: 1 M Tris, pH 9.0.

2.3 Purification and Amplification of the Nucleic Acid from the Selected Library Members (See Note 4)

1. RNeasy Micro Kit (Qiagen) with associated buffers made as indicated in kit instructions (see Note 5). 2. Reverse transcriptase: Superscript II, with buffers and kit components (Invitrogen). 3. 10 mM dNTP mix. 4. E2 Primer: 50 -TCAGCGGTGGCAGCAGCCAA-30 , prepared to a working solution of 10 μM with DNase-free H2O (see Note 6). 5. Platinum Taq HiFi (Invitrogen) with associated buffers and components. 6. 62up primer: 50 -CTATGCAGGGGTTGTTGAAG-30 , prepared to a working solution of 10 μM with DNase-free H2O (see Note 6). 7. Thermocycler. 8. Gel extraction kit with associated buffers made as indicated in kit instructions. 9. Agarose suitable for DNA gel electrophoresis. 10. DNA gel electrophoresis apparatus. 11. Tris-acetate EDTA buffer (TAE): Prepare 1 TAE with 40 mM Tris, 20 mM EDTA disodium salt, 50 mM sodium acetate, and 77.2 mL/L glacial acetic acid.

160

Kathryn M. Frietze et al.

12. Nuclease-free water. 13. UV transilluminator. 14. Ethidium bromide stock: Solution of 10 mg/mL ethidium bromide. 15. Orange G gel-loading dye: Obtained from commercial source or prepared by dissolving 1.5 g Ficoll 400 in 10 mL nucleasefree H2O and adding a very small amount of Orange G powder to achieve a dark orange color. 2.4 Preparation of Samples for Deep Sequencing

1. KpnI restriction enzyme with associated 10 buffer. 2. Nuclease-free water. 3. Gel extraction kit with associated buffers made as indicated in kit instructions. 4. Platinum Taq HiFi, 10 PCR buffer, 50 mM MgSO4 (Invitrogen). 5. ITBC reverse primer: 50 -CCTCTCTATGGGCAGTCGGT GATGTGAACGCGAGTTAGAGC-30 , HPLC purified, prepared to a working solution of 10 μM with DNase-free H2O (see Note 6). 6. Forward barcode primers: 50 -CCATCTCATCCCTGCGTGT CTCCGACTCAG-[BARCODEX]10–12-CGATCTTTACTCA GTTCGTTCTCGTC-30 , HPLC purified, prepared to a working solution of 10 μM with DNase-free H2O (see Note 6). 7. 10 mM dNTP mix. 8. Agarose suitable for DNA gel electrophoresis. 9. DNA gel electrophoresis apparatus. 10. Tris-boric acid EDTA buffer (TBE): 90 mM Tris, 90 mM boric acid, 2 mM EDTA. 11. UV transilluminator. 12. QIAquick Gel Extraction Kit. 13. Scalpel. 14. 1.7 mL tubes. 15. PCR tubes. 16. Thermocycler. 17. Ethidium bromide staining solution: TBE buffer with 1 μg/ mL ethidium bromide prepared fresh each time immediately before use by adding 50 μL of ethidium bromide stock to 500 mL of TBE buffer.

Deep Sequence-Coupled Biopanning

3

161

Methods

3.1 Isolation of IgG from Human Serum

1. Resuspend Dynabeads™ Protein G by vigorous vortexing for 1 min immediately before use. 2. Transfer 80 μL of Dynabeads™ Protein G to a microfuge tube for each sample, pipetting the bead suspension directly into the bottom of the tube. 3. Place tubes on DynaMag™ Magnet with the tube cap hinge nearest the magnet for 1 min to allow beads to move to the magnet side of the tube. 4. With tubes still on the magnet, aspirate the liquid from the bottom of the tube with a pipettor. 5. Remove tubes from the DynaMag™ Magnet. 6. Wash the beads by adding 500 μL PBS to each tube directly to the beads, pipetting up and down gently several times to put beads into suspension. 7. Place tubes on DynaMag™ Magnet with the tube cap hinge nearest the magnet for 1 min to allow beads to move to the magnet side of the tube. 8. With tubes still on the magnet, aspirate the liquid from the bottom of the tube with a pipettor. 9. Remove tubes from the DynaMag™ Magnet. 10. Repeat steps 6–9 for a second time, for two total washes. 11. Add 80 μL of PBS to each tube of washed Dynabeads™ Protein G. 12. Add 20 μL of serum/plasma sample. 13. Pipet up and down gently to mix Dynabeads™ Protein G, PBS, and serum. Parafilm the top of each tube to prevent loss of sample. 14. Secure tubes to a vortex shaker and adjust speed such that Dynabeads™ Protein G is kept in continuous suspension. 15. Shake for 40 min at room temperature. 16. Centrifuge tubes briefly (~5 s) so all liquid and beads are at the bottom of the tubes (see Note 7). 17. Place tubes on DynaMag™ Magnet for 2 min (see Note 8). 18. With tubes still on the magnet, aspirate the liquid from the bottom of the tube with a pipettor—discard as waste. 19. Remove tubes from the DynaMag™ Magnet. 20. Add 500 μL PBS and vortex briefly to wash. 21. Centrifuge tubes briefly (~5 s) so all liquid and beads are at the bottom of the tubes.

162

Kathryn M. Frietze et al.

22. Place tubes on DynaMag™ Magnet for 2 min. 23. With tubes still on the magnet, aspirate the liquid from the bottom of the tube with a pipettor—discard as waste. 24. Remove tubes from the DynaMag™ Magnet. 25. Repeat steps 20–24. 26. Prepare two microfuge tubes per sample with 3 μL neutralization buffer pipetted directly to the bottom of each tube, labeled with the sample identifier. 27. Add 30 μL elution buffer directly to the beads in each tube, and pipette up and down several times to mix beads and elution buffer. 28. Incubate sample for 2 min at room temperature with near constant mixing (see Note 9). 29. Centrifuge tubes briefly (~5 s) so all liquid and beads are at the bottom of the tubes. 30. Place tubes on the DynaMag™ Magnet for 1 min. 31. Aspirate the elution liquid (containing the eluted antibodies), and transfer into one of the prepared tubes from step 26, pipetting the eluted antibodies directly into the neutralization buffer. Mix by pipetting gently a few times. 32. Repeat steps 27–31, this time adding the elution liquid to the other prepared tube from step 29. 33. Consolidate the paired samples into one tube, resulting in 66 μL total volume of isolated IgG. 34. Estimate the antibody concentration by assessing the sample with a microvolume spectrophotometer and measuring protein A280 (see Note 10). 35. This protocol should yield ~300 ng/μL of isolated IgG. 36. If IgG is to be used within the same day, store the IgG on ice until use. Otherwise, store the IgG at 20  C. 3.2 Selection of a Library Against Isolated IgG

Use barrier tips for all procedures in this section. Do not reuse tips for any steps to prevent cross-contamination. All tubes and tips should be certified free of RNA, DNA, RNase, and DNase contamination. 1. Label one microfuge tube per sample. 2. Add 500 ng of IgG to the tube. 3. Add 1.25–40 μg library (see Note 11). 4. Add sufficient PBS to reach 100 μL total volume. 5. Gently mix and centrifuge briefly (~5 s) to bring all liquid to the bottom of the tube. 6. Parafilm the top of each tube to prevent loss of sample.

Deep Sequence-Coupled Biopanning

163

7. Incubate the IgG, library, and PBS mixture overnight at 4  C with a vortex shaker (see Note 12). 8. Wash and prep 10 μL of Dynabeads™ Protein G following steps 1–13 in Subheading 3.1, using 200 μL of PBS for washes instead of 500 μL. 9. Briefly centrifuge the IgG/library tubes (~5 s). 10. Transfer IgG/library to the washed/prepped Dynabeads™ Protein G from step 8. 11. Gently mix and briefly centrifuge. 12. Parafilm the top of each tube to prevent loss of sample. 13. Incubate at room temperature with a vortex shaker (see Note 2). 14. Centrifuge tubes briefly (~5 s) so all liquid and beads are at the bottom of the tubes. 15. Place tubes on the DynaMag™ Magnet for at least 2 min (see Note 8). 16. Aspirate liquid from tube using a pipettor and discard as waste. 17. Remove tubes from magnet. 18. Wash six times with PBS/Tween. Add 200 μL PBS/Tween to each sample, gently pipetting up and down several times to mix. Centrifuge tubes briefly, place tubes on magnet, wait for ~2 min, aspirate liquid, and discard. Repeat for six washes. After washes #1 and #4, transfer to a fresh microfuge tube. 19. Wash two times with PBS, changing the tubes after the first PBS wash (wash #7). Add 200 μL PBS to each sample and follow wash instructions in step 18. Repeat for two washes. 20. Prepare one microfuge tube per sample with 5 μL neutralization buffer pipetted directly to the bottom of the tube, labeled with the sample identifier. 21. Add 50 μL elution buffer directly to the beads in each tube, and pipette up and down gently for several times to mix beads and elution buffer. 22. Incubate for 5 min at room temperature with near constant mixing (see Note 9). 23. Centrifuge tubes briefly (~5 s) so all liquid and beads are at the bottom of the tubes. 24. Place tubes on the DynaMag™ Magnet for 2 min. 25. Aspirate the elution liquid, containing the eluted library members and IgG, and transfer into the prepared tube from step 20, pipetting the eluate directly into the neutralization buffer. Pipette gently a few times to mix. 26. Proceed immediately to the next step; store eluate at 4  C overnight or at 20  C for up to 1 week (see Note 13).

164

Kathryn M. Frietze et al.

3.3 Purification of the Nucleic Acid from the Selected Library Members

Use barrier tips for all procedures in this section. Do not reuse tips for any step to avoid cross-contamination of samples. All tubes and tips should be certified free of RNA, DNA, RNase, and DNase contamination (see Note 14). 1. Use the RNeasy Micro Kit and associated buffers to purify RNA from the eluates obtained in Subheading 3.2, step 26. Follow manufacturer’s instructions, using 295 μL buffer RLT and eluting with 14 μL RNase-free water added directly to the center of the spin column. 2. Store isolated RNA at 20  C for up to 2 weeks (see Note 15).

3.4 Generating cDNA of Selected Library Members

Use barrier tips for all procedures in this section. To avoid crosscontamination of samples, do not reuse tips for any step. All tubes and tips should be certified free of RNA, DNA, RNase, and DNase. Incubation steps can be performed using a thermocycler or a heat block set to appropriate temperatures. 1. Thaw 5 FS buffer (part of Superscript II, Invitrogen), 0.1 M DTT, 10 mM dNTP mix, and E2 primer, and place on ice. 2. Prepare Master Mix 1 (MM1) in a microfuge tube: 4.0 μL E2 primer and 1.0 μL 10 mM dNTP mix per sample. Make enough MM1 to include a negative control tube: (# samples + 1 negative control + 1 for pipetting error). 3. Aliquot 5.0 μL of MM1 into PCR tubes (1 for each sample + 1 for negative control). 4. Add 8 μL RNA from Subheading 3.3, step 2 to the PCR tube. 5. Centrifuge tubes briefly (~5 s) so all liquid is at the bottom of the tubes. 6. Heat to 65  C for 5 min, immediately place on ice to quick chill. 7. Prepare Master Mix 2 (MM2) in a microfuge tube: 4.0 μL 5 FS buffer, 2.0 μL 0.1 M DTT. Make enough MM2 to include a negative control tube (# samples + 1 negative control + 1 for pipetting error). 8. Add 6.0 μL of MM2 to each tube from step 6. 9. Centrifuge tubes briefly (~5 s) so all liquid is at the bottom of the tubes. 10. Place tubes at 37  C for 2 min. 11. Add 1 μL Superscript II to each tube, dispensing directly into the reaction mixture already in the tube and pipetting up and down for several times to mix. 12. Incubate tubes for 50 min at 37  C. 13. Incubate tubes for 15 min at 70  C. 14. Place tubes on ice to chill (see Note 16).

Deep Sequence-Coupled Biopanning

3.5 Amplification of cDNA of Selected Library Members

165

Use barrier tips for all procedures in this section. Do not reuse tips for any step to avoid cross-contamination of samples. All tubes and tips should be certified free of RNA, DNA, RNase, and DNase contamination. 1. Thaw 10 PCR buffer, 62up primer, E3.2 primer, 50 mM MgSO4, and 10 mM dNTP mix and place on ice. 2. Prepare PCR Master Mix (PCR MM) in a microfuge tube: 35.8 μL nuclease-free H2O, 5.0 μL 10 PCR buffer, 2.5 μL 62up primer, 2.5 μL E3.2 primer, 2 μL 50 mM MgSO4, 1 μL 10 mM dNTP mix, and 0.2 μL Platinum Taq HiFi. Make enough PCR MM to include a negative control generated during the cDNA step and a negative control for the PCR (# samples + 1 negative control cDNA + 1 PCR negative control + 1 for pipetting error). 3. Aliquot 49 μL of PCR MM into PCR tubes. Close tubes after adding to avoid contamination. 4. Add 1 μL cDNA (Subheading 3.4, step 14) to PCR tubes. For cDNA-negative control, add 1 μL of the negative control from Subheading 3.4, step 14. For PCR-negative control, add 1 μL nuclease-free H2O. 5. Briefly vortex or flick tubes to mix and centrifuge briefly. 6. Place tubes in the thermocycler and cycle according to the following specifications: 94  C for 2 min (30 cycles of 94  C for 30 s, 60  C for 30 s, 68  C for 30 s), 68  C for 10 min, and 4  C hold (see Note 17). 7. Prepare a 1.0% (w/v) agarose gel in 1 TAE buffer by mixing agarose and 1 TAE buffer in an Erlenmeyer flask and microwaving until all agarose is dissolved. Allow to cool until no longer steaming, add ethidium bromide stock to 0.05 μg/mL, and pour in gel casting system. 8. On a square of parafilm, dispense 2 μL of Orange G gel loading dye and 8 μL of PCR samples. 9. Load prepared samples in gel, and run at 95 V until the dye front is about 1/2 to 2/3 of the length of the gel. Also load a 100 bp DNA ladder. 10. Visualize gel with a UV transilluminator. Successful samples will show a product of ~500 bp (see Notes 18 and 19).

3.6 Preparation of Samples for Deep Sequencing

Use barrier tips for all procedures in this section. Do not reuse tips for any step to avoid cross-contamination of samples. All tubes and tips should be certified free of RNA, DNA, RNase, and DNase contamination. 1. Prepare a KpnI restriction enzyme digest reaction containing the following: 35 μL PCR product from Subheading 3.5, step 6; 50 μL nuclease-free H2O, 10 μL 10 restriction enzyme

166

Kathryn M. Frietze et al.

buffer, 5 μL KpnI restriction enzyme. Incubate for 1 h in a 37  C water bath or heat block (see Note 20). 2. Use the QIAquick PCR Purification Kit to clean up the restriction enzyme digest reaction following the manufacturer’s instructions, eluting at the end with 30 μL of the supplied EB buffer. Store this at 20  C. 3. Assign a unique barcode to each sample and record in a laboratory notebook. 4. Prepare a barcode PCR master mix in a microfuge tube for the barcode PCR. For each sample, add the following: 35.8 μL nuclease-free H2O, 5.0 μL 10 times PCR buffer, 2.0 μL 50 mM MgSO4, 2.5 μL ITBC Reverse primer, 1.0 μL 10 mM dNTP mix, and 0.2 μL Platinum Taq HiFi. Make enough barcode PCR master mix to include a negative control for the PCR (# samples + 1 negative control + 1 for pipetting error). 5. Aliquot 46.5 μL of the barcode PCR master mix into PCR tubes. Close tubes after adding the barcode PCR master mix to avoid contamination. 6. Add 2.5 μL of appropriate barcode primer to each tube, as assigned in step 3. Open tubes one at a time and close immediately after opening to avoid contamination. 7. Add 1 μL of the appropriate KpnI digested PCR product from step 2 to the corresponding PCR tube. 8. Flick tubes or vortex gently to mix and centrifuge briefly to spin all liquid to the bottom of the tubes. 9. Place tubes in the thermocycler and cycle according to the following specifications: 94  C for 2 min; 15 cycles of 94  C for 30 s, 60  C for 30 s, 68  C for 30 s, and 68  C for 10 min; and 4  C hold (see Note 21). Samples can be stored at 20  C. 10. Prepare a 1.5% (w/v) agarose gel in TBE buffer by mixing agarose and TBE buffer in an Erlenmeyer flask and microwaving until all agarose is dissolved. Allow to cool until no longer steaming, and pour in gel casting system (see Note 22). Place gel in the gel electrophoresis apparatus with freshly made TBE buffer. 11. Add 10 μL of Orange G gel loading dye to each PCR tube from step 9. 12. Load prepared samples in the gel (using multiple lanes if necessary), and run at 95 V until the dye front is about 3/4 of the length of the gel. Also load a 100 bp DNA ladder. 13. Remove the gel from the gel electrophoresis apparatus and place in a pyrex (or other suitable) tray. Cover with freshly made ethidium bromide staining solution and stain for 7 min. Transfer ethidium bromide staining solution to an appropriate waste container. Destain the gel for at least 10 min under running dH2O.

Deep Sequence-Coupled Biopanning

167

14. Visualize gel with a UV transilluminator. Successful samples will show products of ~200 bp. 15. Excise products quickly with a scalpel, taking care to avoid any products that are lower in size than the expected product. 16. Transfer gel slices to labeled tubes. Proceed immediately to gel extraction, or store the gel slices for up to 3 days at 4  C. 17. Use the QIAquick Gel Extraction Kit following the manufacturer’s instructions. One QIAquick column should be used for each sample. Centrifuge multiple times to get all the QG buffer/gel solution through the column. Include any optional steps in the protocol. Elute with 30 μL of EB buffer. 18. Submit samples for sequencing using the Ion Torrent sequencing platform (see Note 23). 3.7

4

Data Processing

Data from the Ion Torrent sequencing platform is preprocessed by the associated Ion Torrent software to separate each set of barcoded sequences into a separate FASTQ file format. We have written custom MATLAB scripts to perform data processing that takes FASTQ files, performs quality control steps, identifies the foreign peptide coding sequence, and provides as an output a file that has the peptide sequence, the raw read number (occurrence), and the count per million (used to standardize the population number across various sequencing runs). These files are then processed with additional custom scripts as needed based on the specific questions being addressed.

Notes 1. Dynabeads™ Protein G will isolate primarily IgG from both human and mouse serum/plasma. We predict that antibodies of different classes and from different animal species could be isolated through various methods and used in this protocol with comparable results. 2. Any device that will shake samples fast enough to keep the beads in suspension is suitable for this protocol. We use a vortexer with a plate attachment set to constant shaking, and tape our sample tubes securely to the plate. If no suitable shaker is available, we have had success with rocking samples on an orbital rocker and every 10–15 min during incubation mixing samples by hand to bring beads into suspension. 3. We have developed this protocol using bacteriophage MS2 VLPs that display pathogen-specific peptides. This protocol could be adapted to use another phage-display platform by altering the primers and data analysis scripts. We refer the readers to [6] for detailed information on the construction of

168

Kathryn M. Frietze et al.

MS2 VLP pathogen-specific libraries and [9] for detailed information on the construction of MS2 VLP libraries in general. We have also used this protocol with a bacteriophage MS2 VLP library displaying random peptides [5]. The use of pathogenspecific libraries limits the researcher to identifying linear epitopes, while random libraries allow the possibility of identifying conformational mimics. Although we have successfully used random libraries to assess antibody specificity in human serum, the bioinformatics analysis necessary to identify interesting selectants is challenging. 4. We have carefully optimized the volumes described in this protocol based on the specific PCR and RT reagents we list here. In our experience, reverse transcriptase and Taq polymerase from different companies/sources are not interchangeable. For this reason, we indicate the specific suppliers and catalog numbers for the RT and PCR reagents we use. 5. The protocol here is based on using RNeasy Micro Kit (Qiagen 74004) which, in our experience, is superior to other Qiagen RNeasy Kit formats. This may be due to the small volumes we need to isolate and the size of the columns included in this kit. It may also be due to the purity of the RNA that we obtain from this kit. For this reason, we recommend using this specific kit and not substitute from another company or format. 6. Primers indicated in this protocol are specific for the bacteriophage MS2 VLP platform that we use. The RT and PCR primers used to amplify the selected library members will need to be custom designed for your library platform of choice. Primers should be designed to amplify a sufficient length of nucleic acid (approximately 300–500 bp) to ensure that the resulting product can be easily purified and then used as a template for deep sequence sample preparation. The ITBC reverse primer has two features: an adapter sequence (which is specific for Ion Torrent sequencing platforms) and an 18 nt sequence that is complementary to the PCR product generated. The forward barcode primers have three features: adapter sequence (which is specific for Ion Torrent sequencing platforms), unique 10–12 nt barcode, and 22 nt sequence that is complementary to the PCR product generated. 7. The manufacturer’s instructions for Dynabeads™ Protein G indicate that centrifugation is unnecessary, but in our experience, the magnet works best if the sample is centrifuged briefly prior to each time the magnet is used. Centrifugation is critical; without this step, bead loss can occur and/or the beads may not get washed sufficiently. 8. After 2 min, observe the tubes. Because the beads are loaded with IgG, they tend to stick to the sides of the tube. If it appears that this is the case and not all of the beads have moved to the

Deep Sequence-Coupled Biopanning

169

magnet side of the tube, gently rotate the tube to collect beads on the magnet. 9. When processing many samples at one time, it may be useful to have another laboratory member helping with this step. The elution step should not be allowed to go much longer than 2 min, since the elution buffer is acidic. We usually add elution buffer to all tubes, and then flick each tube several times in sequence until 2 min has passed for the first sample. Then all mixing stops, and the next steps are carried out. Sets of ~8 samples can be processed in this step at one time. If you have more than eight samples, split them into smaller, manageable groups. 10. We use a NanoDrop™ 8000 spectrophotometer and associated software that has a preset Protein A280 setting called “IgG” that estimates the concentration of antibody using a mass extinction coefficient of 13.7. 11. We have successfully used as little as 1.25 μg and as much as 40 μg of our bacteriophage MS2 VLP libraries. 12. Any device that will provide constant mixing of small volumes would be sufficient in this step. 13. We have not systematically assessed the effect of long-term storage of the eluates on subsequent success of the protocol. We usually proceed directly to RNA purification but have, at times, waited over a weekend. We do not recommend waiting longer than a week to continue to the RNA purification step. 14. The protocol described in this section has been extensively adapted and tested. Challenges we have encountered in the past include no amplification in the PCR, low amplification in the PCR, and contamination in the PCR and/or RT. Many of these challenges were encountered when we made minor changes to the source of reagents. For this reason, we do not recommend altering the source of the reagents listed here. Please see the Notes in this section for detailed tips on addressing these challenges if they occur. 15. We have not systematically assessed the long-term storage of purified RNA obtained in this protocol, but we have seen low amplification yields when we use RNA that has been stored longer than several months at 20  C. For this reason, we caution against using RNA stored longer than 2 weeks. 16. At this point, cDNA can be stored at 20  C or can be directly used in the next step. We have not systematically assessed the effect of long-term storage on the cDNA. We have successfully used up cDNA stored for as long as 1 month. 17. PCR product can be stored at 4  C for up to 3 days. If longer storage is needed, the PCR product should be stored at 20  C.

170

Kathryn M. Frietze et al.

18. The presence of product in either of the negative controls indicates contamination in the PCR. There are a number of interventions that can prevent contamination, including changing gloves frequently, never reusing tips, using dedicated tubes that are kept in a closed container and only handled with gloves, having dedicated reagents and tubes for each laboratory member, using barrier tips for every step of the procedure, and promptly throwing out all reagents when any contamination occurs. It is much easier and more cost-effective, in our opinion, to prevent contamination by following these guidelines and then to try to troubleshoot the source of contamination when it occurs. 19. If the agarose gel shows no product, or very faint product, the PCR should be repeated with 40 cycles. In our experience, this almost always results in a product. If there is still no product, the entire procedure should be repeated from the beginning, and troubleshooting efforts should focus on identifying possible defective reagents or user error. 20. Only about 90% of clones in our library carry a peptide insert. Because insertion during library construction knocks out a KpnI site in our display vector, the presence of nonrecombinant sequences can be greatly reduced by KpnI digestion. This step is important, since nonrecombinant sequences are preferentially amplified during preparation for sequence analysis. 21. The cycle number here may need to be adjusted. In our experience, 15 cycles almost always result in a suitable product. In some cases, samples may need more cycles to produce a product. Cycles should be increased in 2 cycle increments if no product is observed with 15 cycles. 22. This is a high-percentage agarose gel and takes considerable boiling to fully dissolve the agarose. We find it best to heat the solution to boiling and swirl vigorously to mix. We then continue this boil/vigorous swirl cycle until all agarose is fully dissolved. Additionally, TBE should be prepared fresh for each use from a 5 stock solution. TBE buffer in the gel electrophoresis apparatus should not be reused, as this results in products that run in a smear. Note that ethidium bromide is not added to the gel. It is important for the product to run in a very tight band on the gel, and ethidium bromide added to the gel makes the product run as more of a smear, putting the excision step at risk of capturing undesirable product. 23. We use our university’s sequencing facility core to sequence our samples. There are a number of companies that could perform this sequencing for you, or your own institution may have a sequencing facility core that could provide this service.

Deep Sequence-Coupled Biopanning

171

References 1. Ryvkin A, Ashkenazy H, Smelyanski L et al (2012) Deep panning: steps towards probing the IgOme. PLoS One 7(8):e41469 2. Larman HB, Zhao Z, Laserson U et al (2011) Autoantigen discovery with a synthetic human peptidome. Nat Biotechnol 29(6):535–541 3. Larman HB, Laserson U, Querol L et al (2013) PhIP-Seq characterization of autoantibodies from patients with multiple sclerosis, type 1 diabetes and rheumatoid arthritis. J Autoimmun 43:1–9 4. Xu GJ, Kula T, Xu Q et al (2015) Comprehensive serological profiling of human populations using a synthetic human virome. Science 348 (6239):aaa0698 5. Frietze KM, Roden RB, Lee JH et al (2016) Identification of anti-CA125 antibody responses in ovarian cancer patients by a novel deep

sequence-coupled biopanning platform. Cancer Immunol Res 4(2):157–164 6. Frietze KM, Pascale JM, Moreno B et al (2017) Pathogen-specific deep sequence-coupled biopanning: a method for surveying human antibody responses. PLoS One 12(2):e0171511 7. de Oliveira-Ju´nior LC, Arau´jo Santos FA, Goulart LR et al (2015) Epitope fingerprinting for recognition of the polyclonal serum autoantibodies of Alzheimer’s disease. Biomed Res Int 2015:267989 8. Somers V, Govarts C, Somers K et al (2008) Autoantibody profiling in multiple sclerosis reveals novel antigenic candidates. J Immunol 180(6):3957–3963 9. Chackerian B, Caldeira Jdo C, Peabody J et al (2011) Peptide epitope identification by affinity selection on bacteriophage MS2 virus-like particles. J Mol Biol 409(2):225–237

Chapter 10 Isolation of Antigen-Specific VHH Single-Domain Antibodies by Combining Animal Immunization with Yeast Surface Display Lukas Roth, Simon Krah, Janina Klemm, Ralf Gu¨nther, Lars Toleikis, Michael Busch, Stefan Becker, and Stefan Zielonka Abstract In addition to conventional hetero-tetrameric antibodies, the adaptive immune repertoire of camelids comprises the so-called heavy chain-only antibodies devoid of light chains. Consequently, antigen binding is mediated solely by the variable domain of the heavy chain, referred to as VHH. In recent years, these single-domain moieties emerged as promising tools for biotechnological and biomedical applications. In this chapter, we describe the generation of VHH antibody yeast surface display libraries from immunized Alpacas and Lamas as well as the facile isolation of antigen-specific molecules in a convenient fluorescenceactivated cell sorting (FACS)-based selection process. Key words Yeast surface display, Antibody engineering, Protein engineering, Camelid antibodies, Single-domain antibodies, VHH (variable domain of the heavy chain of heavy chain-only antibodies)

1

Introduction From a structural aspect, conventional antibodies are homodimers of heavy- and light-chain heterodimers (Fig. 1). The antigen binding site, i.e., paratope is formed by the variable domain of the heavy chain and the variable domain of the light chain. In addition to this, it’s been known for quite some time that camelids and sharks produce heavy chain-only antibodies (HcAbs) [1, 2]. Such molecules are homodimeric by nature since they are composed of two heavy chains and lack light chain counterparts. Moreover, with respect to camelid-derived heavy chain-only antibodies the first constant domain CH1 that typically interacts with the constant region of the light chain is absent in these moieties (Fig. 1). Interestingly, HcAbs specifically recognize an antigen with only one variable domain, referred to as variable domain of the heavy chain of heavy chain-only antibodies (VHH) for camelid-derived

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2070, https://doi.org/10.1007/978-1-4939-9853-1_10, © Springer Science+Business Media, LLC, part of Springer Nature 2020

173

174

Lukas Roth et al.

Fig. 1 Structural illustration of conventional hetero-tetrameric antibodies (left) and camelid-derived heavy chain-only antibodies

molecules [3]. This VHH domain displays several attributes making it interesting for biomedical and biotechnological applications. In this respect, complementarity determining region 3 (CDR3) is typically elongated compared to conventional antibodies [4]. This protruding loop enables engagement of epitopes that are typically not antigenic to canonical antibodies, i.e., cryptic or recessed epitopes such as catalytic clefts [5–7]. Besides, also the hypervariable region, corresponding to CDR1, is extended by four residues at the N-terminal end and is likely involved in antigen binding [3]. Moreover, camelid VHH domains are generally highly water soluble. This is due to replacements of large and hydrophobic residues by small and hydrophilic amino acids in framework region2 that normally builds the interaction surface with the variable domain of the light chain [8, 9]. Furthermore, the surface-exposed loop of CDR3 is often connected with CDR1 or CDR2 via an additional disulfide bond [4, 10]. However, also CDR3 intraloop disulfides have been described for VHHs [11]. Camelid VHH domains share a high sequence similarity with human VH domains [3]. Notwithstanding, to reduce the potential for immunogenicity, a general strategy for humanization has been presented by Vincke and colleagues [12]. Due to their simple architecture, camelid VHH domains afford the benefit of flexible reformatting options, including bi- and multispecific constructs opening new avenues for targeting diseases [13–16]. Consequently, a plethora of different strategies have been described for the generation of antigen-specific VHH fragments, including naı¨ve, semi-synthetic, synthetic, and immunized library approaches [17–22], and several VHH-derived molecules are currently investigated in clinical trials [4, 10, 23].

Isolation of VHH Antibodies by Yeast Surface Display

175

Although the vast majority of engineered VHH domains were obtained using phage display as platform technology [20, 24], it has been shown that such antigen-specific single-domain antibodies can readily be isolated using yeast surface display (YSD) [25]. Pioneered by Boder and Wittrup in 1997 [26], this genotypephenotype coupling technology relies on the expression of the protein of interest as fusion with Aga2p mating adhesion receptor of Saccharomyces cerevisiae. Surface display is enabled through covalent linkage of Aga2p to Aga1p that is anchored to the cell wall. One of the major benefits of YSD is the utilization of a eukaryotic expression system comprising unfolded protein response for the degradation of misfolded or aggregated proteins [27]. Moreover, its compatibility with fluorescence-activated cell sorting (FACS) enables online and real-time analysis of individual library candidates [28]. In this chapter, we provide protocols for facile VHH library construction following Alpaca and Lama immunization as well as for the isolation of antigen-specific molecules using a FACS-based selection strategy. To this end, VHH genes are amplified from PBMC-derived cDNA in a one-step PCR and introduced into a yeast surface display vector in a homologous recombination-based process referred to as gap repair cloning. By application of a two-dimensional sorting strategy for the detection of full-length surface display as well as antigen binding, we show that targetspecific VHHs can readily be obtained within two rounds of FACS selections.

2 2.1

Materials Strains

1. Saccharomyces cerevisiae strain EBY100 (MATa) (URA3-52 trp1 leu2Δ1 his3Δ200 pep4::HIS3 prb1Δ1.6R can1 GAL) (pIU211:URA3). 2. E. coli strain Top10 (F-mcrA Δ(mrr-hsdRMS-mcrBC) Φ80lacZΔM15 ΔlacX74 recA1 araD139 Δ(ara leu) 7697 galU galK rpsL (StrR) endA1 nupG).

2.2

Plasmids

Substantial features of display plasmid are schematically shown in Fig. 2. A pYD-derived backbone is used as destination vector (pDest) containing essential features, i.e., tryptophan auxotrophic marker for selection in EBY100, ampicillin resistance marker (AmpR) for selection in E. coli, GAL1 promoter and replication origins in S. cerevisiae (ARS4/CEN6) and E. coli (ColE1), and terminator sequences (not shown). The coding VHH region is genetically fused in frame to Aga2p by replacement of stuffer sequence after pDest digestion with BsaI and gap repair cloning. Therefore, the VHH gene is flanked by homologous overhangs to

176

Lukas Roth et al.

Fig. 2 Schematic illustration of generated yeast surface display plasmid. Main genetic components of the system are shown. GAL1, GAL1 promoter; Aga2 SP, Aga2 signal peptide; VHH, camelid variable domain of the heavy chain of heavy chain-only antibodies; G/S-linker, glycine-serine linker; Myc, Myc epitope; Aga2p, Aga2p cell adhesion molecule; HA, HA-tag; tryptophan, tryptophan auxotrophic marker; ARS4/CEN6, replication origin for yeast; AmpR, ampicillin resistance marker; ColE1, replication origin for E. coli. Illustrated features from GAL1 to HA were genetically fused in frame on the used plasmid

G/S linker as well as Aga2p signal sequence (see Note 1) facilitating homologous recombination (Fig. 3). A HA-tag is linked to Aga2p allowing for the detection of VHH surface expression. 2.3

Media

1. YPD media: Dissolve 20 g D(+)-glucose, 20 g peptone, and 10 g yeast extract in 1 l deionized H2O. Sterilize by autoclaving. Add 10 ml of Penicillin–Streptomycin (10,000 units/ml) and remove any particles by sterile filtration using a 0.22 μm bottle top filter. 2. SD-Trp media: Dissolve 26.7 g minimal SD-Base (Clontech) in deionized H2O and adjust volume to 890 ml. Sterilize by autoclaving. Dissolve 8.56 g NaH2PO4
H2O, 5.4 g Na2HPO4, and 0.74 g Dropout-mix -Trp (Clontech) in deionized H2O and adjust the volume to 100 ml. Sterilize by autoclaving. Combine both solutions, add 10 ml of Penicillin– Streptomycin (Gibco, 10,000 units/ml), and remove any particles by sterile filtration using a 0.22 μm bottle top filter.

Isolation of VHH Antibodies by Yeast Surface Display

177

Fig. 3 Gap repair cloning scheme for generation of yeast surface display antibody libraries by homologous recombination. Restriction enzyme digested destination plasmid (pDest) and PCR amplicons with homologous overhangs are shown. Homologous regions are embedded in Aga2p signal peptide sequence and G/S-linker with joined myc-tag, respectively, facilitating homologous recombination to generate the final display plasmid

3. SD-Trp plates: Dissolve 23.35 g of minimal SD-Agar Base (Clontech) in deionized H2O and adjust volume to 445 ml. Sterilize by autoclaving. Dissolve 4.28 g of NaH2PO4
H2O, 2.7 g of Na2HPO4, and 0.37 g Dropout-mix -Trp in deionized H2O and adjust the volume to 50 ml. Sterilize by autoclaving. Combine both solutions, add 10 ml of Penicillin–Streptomycin (Gibco, 10,000 units/ml), and prepare plates. 4. SG -Trp media: Dissolve 37 g of minimal SD-Base + Gal/Raf in deionized H2O and adjust volume to 490 ml. Dissolve 8.56 g of NaH2PO4
H2O, 5.4 g of Na2HPO4, and 0.74 g Dropoutmix -Trp in deionized H2O and adjust the volume to 100 ml. Dissolve 110 g of PEG8000 in deionized H2O and adjust the volume to 400 ml. Sterilize by autoclaving and combine all three solutions. Add 10 ml of Penicillin–Streptomycin (10,000 units/ml) and remove any particles by sterile filtration using a 0.22 μm bottle top filter. 5. SD Low -Trp medium: Dissolve 5 g dextrose, 6.7 g yeast nitrogen base (w/o amino acids) in deionized H2O, and adjust volume to 890 ml. Sterilize by autoclaving. Dissolve 8.56 g NaH2PO4
H2O, 5.4 g Na2HPO4 and 0.74 g Dropout-mix -Trp in deionized H2O and adjust the volume to 100 ml. Sterilize by autoclaving. Combine both solutions, add 10 ml of Penicillin–Streptomycin (10,000 units/ml), and remove any particles by sterile filtration using a 0.22 μm bottle top filter.

178

Lukas Roth et al.

6. Yeast library freezing solution: Dissolve 2 g of glycerol and 0.67 g of yeast nitrogen base in a volume of 100 ml deionized H2O. Sterile filter the solution. 7. LB Amp media: Dissolve 10 g NaCl, 10 g peptone, and 5 g yeast extract in 1 l deionized water. Sterilize by autoclaving. Once the medium has chilled (to approximately 50  C), add 1 ml of sterile filtered ampicillin solution (100 mg/ml in deionized H2O). 8. LB Amp plates: Dissolve 10 g NaCl, 10 g peptone, 5 g yeast extract, and 15 g agar to a volume of 1 l in deionized water. Sterilize by autoclaving. When medium has chilled (to approximately 50  C), add 1 ml of sterile filtrated ampicillin solution (100 mg/ml in deionized H2O) and prepare plates. 2.4 Peripheral Blood Mononuclear Cell (PBMC) Isolation by Density Gradient Centrifugation

1. PBMC prep buffer: Dulbecco’s phosphate-buffered saline (DPBS), 2% (v/v) fetal bovine serum (FBS).

2.5 Total RNA Isolation from PBMCs

1. Qiashredder (Qiagen).

2. Biocoll separation solution.

2. RNeasy Mini Kit (Qiagen). 3. Ethanol 70%. 4. RNase-free water.

2.6

cDNA Synthesis

1. Superscript III First-Strand Kit (Invitrogen). 2. RNase-free water.

2.7 Gene-Specific Amplification of VHH Regions

1. Q5 High-Fidelity 2
Master Mix (New England Biolabs). 2. Nuclease-free water. 3. Primer sequences for VHH gap repair cloning are given in Table 1. 4. Wizard SV gel and PCR Clean-Up System (Promega).

Table 1 Oligonucleotide primers used in this study (overhang sequences are shown in italic) Name

Sequence (50 –30 )

VHH_fwd CGCTGTTTTTCAATATTTTCTGTTATTGCTAGCGTTTTAGCAGGTGATGTGCAGC TGCAGGAGTCTGGRGGAGG VHH1_rev CAATTTTTGTTCAGAACCACCACCACCAGAACCACCACCACCGCTGGGGTC TTCGCTGTGGTGCG VHH2_rev CAATTTTTGTTCAGAACCACCACCACCAGAACCACCACCACCTGGTTGTGG TTTTGGTGTCTTGGG

Isolation of VHH Antibodies by Yeast Surface Display

2.8 Destination Vector (pDest) Digestion

179

1. BsaI High-Fidelity (20,000 units/ml, New England Biolabs). 2. Cut Smart Buffer 10
(New England Biolabs). 3. Nuclease-free water. 4. Wizard SV gel and PCR Clean-Up System. 5. pDest according to Subheading 2.2.

2.9 Library Transformation into S. cerevisiae Strain EBY100

1. BsaI digested pDest according to Subheading 3.5. 2. VHH PCR inserts according to Subheading 3.4. 3. Electroporation buffer: 1 M Sorbitol, 1 mM CaCl2
2H2O (autoclaved). 4. LiAc buffer: 100 mM LiAc, 10 mM DTT (sterile filtered). 5. 1 M Sorbitol (autoclaved).

2.10 Yeast Surface Display and Selections Using FluorescenceActivated Cell Sorting (FACS)

1. Dulbecco’s phosphate-buffered saline (DPBS).

2.11

1. Cryogenic vials.

Equipment

2. Penta-His Alexa Fluor 647 Conjugate antibody (Qiagen). 3. Anti-HA tag antibody (FITC) (abcam). 4. Target protein, his-tagged.

2. Freezing container. 3. 0.22 μm Steriflip and Steritop filtration units. 4. Shaking flasks (150 ml–3 l volume). 5. Electroporator Gene Pulser Xcell™ (Bio-Rad). 6. 0.2 cm Electroporation Cuvettes (Bio-Rad). 7. MasterPure Yeast DNA Purification Kit (Lucigen). 8. Shaking incubator (20  C, 30  C, and 37  C). 9. Flow cytometry device (i.e., cell sorter). 10. Thermocycler. 11. Device and reagents for agarose gel electrophoresis. 12. Benchtop centrifuge. 13. 9 cm Petri dishes. 14. Cell density meter. 15. BioSpec Nano or equivalent instrumentation. 16. SepMate tubes.

3

Methods In this section, we describe the construction of large yeast surface display antibody VHH libraries following Lama or Alpaca immunization. In this particular example, two Alpacas (Vicugna pacos) and

180

Lukas Roth et al.

Fig. 4 Example of immune response of two Alpacas and two Lamas after immunization with a protein antigen. Immunizations were performed at Preclinics GmbH and antibody titers were determined by serum ELISA

two Lamas (Lama glama) were immunized at Preclinics GmbH (http://www.preclinics.com/) with a protein antigen. During the immunization procedure, Preclinics GmbH assessed the antigenspecific immune response in a serum ELISA, as exemplarily shown in Fig. 4. Afterwards, total RNA was isolated from 100 ml PBMCs followed by cDNA synthesis which was subsequently used for genespecific amplification. In order to allow for a broader applicability of this herein described protocol, we also include methods for the generation of cDNA starting from whole blood. Gene-specific oligonucleotide sequences for the amplification of camelid VHH repertoires can also be found elsewhere [16, 24]. However, all protocols, components, concentrations and volumes given below can be used without any modification for the introduction of VHH diversities into yeast surface display libraries. Additionally, in Subheading 3.10, we also present a labeling strategy for the detection of surface display and antigen binding using fluorescence-activated cell sorting. 3.1 PBMC Isolation by Density Gradient Centrifugation

1. Dilute blood samples with an equal volume of PBMC prep buffer. 2. Fill SepMate tubes with 16 ml Biocoll Separation Solution. 3. Transfer approx. 30 ml diluted blood carefully into the SepMate tubes. Avoid subsidence of the blood under the separating disc. 4. Centrifuge for 20 min at 1200
g and room temperature with medium acceleration and deceleration profile.

Isolation of VHH Antibodies by Yeast Surface Display

181

5. Remove blood plasma phase by pipetting. Please note that the PBMC fraction is located directly below and should not be affected by pipetting. 6. If PMC cells are aggregated at SepMate inner surface, remove them mechanically by scraping. 7. Decant residual blood plasma and PBMC into clean tube (ensure that no erythrocytes are transferred) and add 50 ml PBMC prep buffer. 8. Centrifuge for 8 min at 300
g and room temperature with maximum acceleration and medium deceleration profile. Discard supernatant. 9. Wash cells twice by resuspending in 50 ml volume with PBMC prep buffer and repeat centrifugation as in step 8, respectively. 10. Resuspend cells in defined volume of PBMC prep buffer and count PBMC number. 3.2 Total RNA Isolation from PBMCs

1. Pellet 1
107 cells by centrifugation for 2 min at room temperature and 1200
g (microfuge). 2. Lyse cells by resuspending in 600 μl of RLT and subsequent vortexing or pipetting. 3. Add the lysate into Qiashredder column in a 2 ml collection tube and centrifuge for 2 min at maximum speed and room temperature. 4. Remove the Qiashredder column and determine the volume of homogenized lysate before adding same amount of 70% ethanol. 5. Transfer the sample into RNeasy spin column in a 2 ml collection tube and centrifuge for 60 s at 8000
g and room temperature. Discard flow-through. If sample volume exceeds 700 μl, centrifuge aliquots stepwise. 6. Add 700 μl of RW1 buffer into RNeasy spin column. Centrifuge for 60 s at 8000
g and room temperature. Discard flowthrough. 7. Wash the sample twice by adding 500 μl of RPE buffer and successive centrifugation for 60 s at maximum speed and room temperature, respectively. 8. Transfer RNeasy spin column containing bound RNA into a clean collection tube and add 50 μl of RNase-free water. 9. Incubate for 15 min at room temperature. 10. Eluate RNA by centrifugation for 60 s at maximum speed and room temperature. To increase RNA yield load flow-through again onto RNeasy, spin column and repeat the latter centrifugation step.

182

3.3

Lukas Roth et al.

cDNA Synthesis

1. Prepare a RNA and primer mix by adding 5 μl RNA solution from RNA isolation as template, 1 μl random hexamer primers (50 ng/μl) as well as 1 μl of 10 mM dNTP mix. Add 10 μl with RNase-free water. 2. Place the mix in thermocycler at 65  C for 5 min, followed by 60 s incubation on ice. 3. Prepare reverse transcriptase Master Mix by combining 2 μl RT buffer, 4 μl 25 mM MgCl2, 2 μl 100 μM DTT as well as 1 μl RNaseOut (40 U/μl), and 1 μl SuperScript III RT (200 U/μl). 4. Add an equal volume of reverse transcriptase Master Mix to the RNA and primer mix. 5. Incubate the reaction solution in a thermocycler with 12 min at 25  C followed by 52 min at 50  C and as final step 5 min at 85  C. 6. Add 20 μl of RNase H solution and incubate at 37  C for 20 min.

3.4 Amplification of Lama and Alpaca HcAbs Variable Domains (VHHs)

1. Place PCR tube on ice and add 25 μl Q5 high fidelity 2
Master Mix. Add 1 μl (100 ng/μl) of DNA encoding for alpaca VHH region as template as well as 2.5 μl corresponding forward primer and 2.5 μl reverse primer (out of a 10 μM stock, primer sequences are listed in Table. 1) and 19 μl nuclease-free water. Alpaca VHHs can be amplified using primer combination VHH_fwd and VHH1_rev. Lama VHH regions can be amplified using the VHH_fwd and both reverse oligonucleotides (see Note 2). 2. Carry out PCR using the following parameters: Initial denaturation at 98  C for 30 s. 30 cycles of 10 s denaturation at 98  C, 30 s primer annealing at 57  C, and 30 s elongation at 72  C, followed by final step at 72  C for 2 min. 3. Analyze PCR products by 1–2% (w/v) agarose gel electrophoresis. Amplified VHH region genes should give a distinct band at approx. 500 bp. Purify PCR products using Wizard® SV Gel and PCR Clean-Up System according to the manufacturer’s instruction (see Note 3). Determine DNA concentration via BioSpec Nano or equivalent instrumentation. PCR products might be stored at 20  C.

3.5 Destination Vector (pDest) Digestion

1. Place PCR tube on ice and add 40 μg of pDest (1 mg/ml), 8 μl Cut Smart 10
. Buffer as well as 2 μl BsaI-HF (20000 U/ml) in a final volume of 80 μl (add nuclease-free water in order to achieve this volume). 2. Carry out reaction at room temperature overnight. 3. Analyze digestion product by 1–2% (w/v) agarose gel electrophoresis. Successful digestion should give a distinct band at

Isolation of VHH Antibodies by Yeast Surface Display

183

approx. 6000 bp. Purify digestion reaction using Wizard® SV Gel and PCR Clean-Up System according to the manufacturer’s instruction. 3.6 Yeast Transformation for Library Establishment

In the following section, we describe library construction steps as modified version of the improved S. cerevisiae yeast transformation protocol of Benatuil and colleagues [29]. Therefore, it is herein only presented in brief. In this protocol all centrifugation steps to pellet yeast cells are performed at 4000
g for 3 min. 1. Incubate EBY100 overnight to stationary phase in YPD media at 120 rpm and 30  C. 2. Inoculate 500 ml fresh YPD media with the overnight culture to an OD600 of about 0.3. 3. Incubate cells at 30  C and 120 rpm until OD600 reaches about 1.6–1.9. 4. Centrifuge cells and remove supernatant. 5. Wash cells twice by resuspension in 250 ml ice-cold water followed by a wash step using 250 ml ice-cold electroporation buffer. 6. Resuspend cells in 100 ml LiAc-buffer and incubate for 30 min at 30  C and 120 rpm. 7. Centrifuge cells and wash afterwards once with 250 ml ice-cold electroporation buffer. 8. Resuspend cell pellet in electroporation buffer to a final volume of approx. 5 ml. This results in ten electroporation reactions with approx. 400 μl electro-competent EBY100 each. 9. Add 4 μg digested pDest and 12 μg VHH PCR product to 400 μl electro-competent cells per reaction. 10. Transfer cell-DNA mix to ice-cold electroporation cuvette (0.2 cm). Electroporate at 2500 V. Time constant should range from 3.0 to 4.5 ms. Transfer cells from each electroporation reaction into 8–10 ml of a mixture of YPD and 1 M sorbitol (1:1 ratio). Incubate for 1 h at 30  C and 120 rpm. 11. Centrifuge cells and resuspend cell pellet in 10 ml SD-Trp media to calculate library size by dilution plating (see Note 4, SD-Trp agar plates; estimate number of transformants after 72 h). 12. Decant cells in 1 l in SD-Trp media and incubate for 2 days at 30  C and 120 rpm. 13. Transfer library to SD Low -Trp medium. To this end, transfer at least a tenfold excess of cells as calculated by dilution plating. Inoculate at an OD600 of ~1.0. An absorbance value of 1 at 600 nm corresponds to approximately 1
107 cells per ml.

184

Lukas Roth et al.

14. Incubate for two more days at 30  C and 120 rpm. 15. The final library is henceforth prepared for long-term storage (see Subheading 3.8) or screening via FACS. To evaluate the quality of the final library, the authors recommend sending out at least 100 clones for sequencing (see Note 5). 3.7 Sequencing of Display Vector from Yeast Cells

1. Pick clones (e.g., from serial dilution) and inoculate approximately 10 ml SD-Trp medium in a baffled flask followed by incubation for 48 h at 30  C and 120 rpm. Subsequently, isolate display plasmid using the MasterPure Yeast DNA Purification Kit following the manufacturer’s protocol. 2. Transform 50 μl electro-competent E. coli Top10 cells with 5 μl isolated plasmid. 3. After incubation for approximately 60 min in LB medium without antibiotics, plate cells on LB Amp agar plates and incubate overnight at 37  C. Afterwards pick single colonies and inoculate 700 μl LB Amp media in a Masterblock microplate. Incubate overnight at 37  C and 700 rpm. 4. Mix 60 μl of E. coli Top10 culture with 40 μl glycerol solution (50% v/v) in an appropriate microplate. Deep-freeze at 80  C until sending out for sequencing.

3.8 Yeast Library Cryopreservation for Long-Term Storage

1. Harvest cells from a freshly grown SD-Trp culture by centrifugation. 2. Inoculate SD Low -Trp medium with cells to an OD600 of 1 and cultivate for 48–72 h at 30  C and 120 rpm. Library diversity should be oversampled by at least the factor of ten. 3. Harvest cells by centrifugation and remove supernatant. 4. Wash cells once with DPBS. Afterwards pellet cells by centrifugation, decant, and discard supernatant. 5. Resuspend cells in yeast library freezing solution with final cell concentrations of approximately 1
1010 cells/ml and transfer suspensions into cryogenic vials. 6. Freeze vials at 80  C.

3.9 Induction of VHH Expression for Antibody Surface Display

1. Thaw an aliquot of the frozen library at room temperature and resuspend cells in SD-Trp. 2. The total number of cells in the starting culture should exceed the calculated library diversity at least ten times. 3. Incubate overnight at 30  C and 120 rpm. 4. Harvest at least the number of cells corresponding to a tenfold excess of the library diversity by centrifugation and resuspend cells in SG -Trp medium to an initiate OD600 of 1. Cultivate cells for at least 24 h at 20  C and 120 rpm for expression of VHHs.

Isolation of VHH Antibodies by Yeast Surface Display

185

Fig. 5 Detection of VHH display and antigen binding of yeast surface display library candidates using FACS. Overlays of VHH display control (blue) and antigen binding yeast cells (red). Blue: yeast cells labeled with Penta-His Alexa Fluor 647 Conjugate antibody and anti-HA tag antibody (FITC) only. Red: staining for VHH surface display (same detection antibodies as for display control) as well as for antigen binding. Left: FACS sorting round 1. Right: FACS sorting round 2. Sorting gates are shown as well as percentage of cells in sorting gate 3.10 Detection of Yeast Surface Display and Antigen Binding by FluorescenceActivated Cell Sorting (FACS)

In the following chapter, we describe a labeling strategy to detect surface expression of VHH molecules via binding of a fluorescently labeled anti-HA tag antibody (FITC) as well as antigen binding via his-tag staining (Penta-His Alexa Fluor 647 Conjugate antibody). An equivalent staining procedure can be carried out for library sorting (see Notes 6 and 7). Please note that the fluorophores should be shielded from light and labeling steps are performed on ice. For all FACS sorting rounds as well as for flow cytometric analysis, we recommend staining an antibody display control (without antigen) in order to be able to adjust sorting gates accordingly (Fig. 5). Furthermore, a positive control, i.e., a single clone displaying a target-positive VHH should be used to ensure functional integrity of labeling reagents as well as target proteins. It is also possible to carry out positive control staining with different antibody formats. Consequently, staining reagents need to be adapted accordingly (e.g., detection of Fab fragment presentation via staining of the constant domain of the light chain).

3.10.1 Antibody Display Control

1. Pellet approximately 2
107 cells by centrifugation and remove supernatant. Wash cells two times with DPBS and resuspend cells in 40 μl DPBS. Incubate on ice for about 30 min.

186

Lukas Roth et al.

2. Wash cells twice with DPBS and resuspend in 40 μl DPBS containing Penta-His Alexa Fluor 647 Conjugate antibody (diluted 1:20) and anti-HA tag antibody (FITC) (diluted 1:20), followed by incubation on ice for 30 min. 3. Wash cells again two times and resuspend the cell pellet finally in 600 μl DPBS. Keep on ice protected from light until FACS analysis. 3.10.2

Library Staining

The following section describes staining protocol of antibody displaying yeast cells regarding VHH surface expression as well as target binding. Essentially the same protocol can be used to label positive controls and libraries in general. Scale up for sorting purposes (see Note 6). 1. Pellet 2
107 yeast cells by centrifugation and remove supernatant. Wash cells twice with DPBS and resuspend in 40 μl DPBS containing antigen with a concentration of 1 μM (see Note 8). Incubate cells on ice for 30 min. 2. Wash cells two times and resuspend in 40 μl DPBS containing Penta-His Alexa Fluor 647 Conjugate antibody (diluted 1:20) and anti-HA tag antibody (FITC) (diluted 1:20), followed by incubation on ice for 30 min (see Note 9). 3. Wash cells again two times and resuspend the cell pellet finally in 600 μl DPBS. Keep on ice protected from light until FACS analysis.

3.10.3 Further Proceeding with Cells After FACS Sorting

This section describes briefly the treatment of the sorted cells after FACS analysis for further cytometric analysis rounds and cell storage. 1. Transfer sorted cells collected in sorter tube containing 500 μl DPBS into corresponding volume of SD media (approximately 20 ml media per 1
106 cells). Incubate at 30  C and 120 rpm for 2 days. 2. Count cell number and induce VHH expression (see Note 6) by media exchange, as described in Subheading 3.9. 3. Deep-freeze remaining cells according to Subheading 3.8 for long-term storage.

4

Notes 1. In general, overhang sequences were part of constant components of the display system. The N-terminal overhang comprised nucleotides that were identical with those of a part of the Aga2p signal peptide, whereas C-terminal overhang was homologous to the glycine-serine linker fused to the myc-tag.

Isolation of VHH Antibodies by Yeast Surface Display

187

The herein described pDest allowed for N-terminal presentation of VHH moieties (i.e., free N-terminus of VHH and C-terminus fused to Aga2p). C-terminal display of VHHs is also a valid strategy for surface expression as already described by our group for the display of shark-derived single-domain antibodies [29–31]. 2. Reverse primers were designed to anneal in the hinge region of camelid-derived IgG2b and IgG3b isotypes. Another alternative is the utilization of oligonucleotides annealing in framework region 4 of the VHH domain. 3. By using the primer sets shown in Table 1, one co-amplifies also the VH as well as CH1 region of conventional IgG antibodies. This results in an unwanted product with a size of approximately 800–900 bp on the agarose gel. We typically do not gel-excise the VHH-specific band since this would require a second-step re-amplification and a potential bias. Still, this crude mixture delivers nearly exclusively VHH domains after library establishment. 4. One transformation electroporation reaction usually yields in approximately 1
107 to 5
107 unique EBY100 clones. Scale up gap repair cloning and electroporation reactions by parallelization to obtain a sufficient library size. This is especially required for the construction of libraries from non-immunized animals. 5. For quality maintenance we recommend sequencing of the constructed library. Thus, the sequencing primer should be designed with respect to VHH region coverage in a single run. To this end, oligonucleotides should anneal approximately 60 bp upstream of the variable region. 6. In general, labeling steps are performed with 2
107 cells in a volume of 40 μl. For library sorting purposes, scale up in order to be able to sample a sufficient number of clones. By increasing the numbers of cells, volumes and amounts of labeling reagents should be increased proportionally. We recommend oversampling the generated diversity by at least the factor of ten. However, when large libraries are being generated, the limiting factor is throughput by FACS. For the first round of FACS, a maximum number of cells should be sorted. For subsequent sorting rounds, we typically oversample the number of isolated cells from the previous round by the factor of ten. 7. A variety of labeling reagents are commercially available. Thus, different combinations of fluorescent dyes and detection reagents are possible. For fluorophore selection, please note that no overlapping of excitation and emission spectra occur. Otherwise, appropriate compensation needs to be applied on the FACS device. Guidance should be given by the respective manufacturer.

188

Lukas Roth et al.

8. To enhance stringency, the antigen concentration can be reduced significantly. However, we recommend starting sorting at an antigen concentration of 1 μM for initial rounds and to successively reduce the concentration as soon as cells are enriched for an antigen-binding population. 9. When working with non-immunized antibody repertoires, different secondary detection reagents should be alternated to avoid enrichment of off-target binders.

Acknowledgments We like to thank Preclinics GmbH for collaborating on this project. References 1. Hamers-Casterman C, Atarhouch T, Muyldermans S et al (1993) Naturally occurring antibodies devoid of light chains. Nature 363:446–448 2. Zielonka S, Empting M, Grzeschik J et al (2015) Structural insights and biomedical potential of IgNAR scaffolds from sharks. MAbs 7:15–25 3. Arezumand R, Alibakhshi A, Ranjbari J et al (2017) Nanobodies as novel agents for targeting angiogenesis in solid cancers. Front Immunol 8:1746 4. Ko¨nning D, Zielonka S, Grzeschik J et al (2017) Camelid and shark single domain antibodies: structural features and therapeutic potential. Curr Opin Struct Biol 45:10–16 5. Wesolowski J, Alzogaray V, Reyelt J et al (2009) Single domain antibodies: promising experimental and therapeutic tools in infection and immunity. Med Microbiol Immunol 198:157–174 6. J€ahnichen S, Blanchetot C, Maussang D et al (2010) CXCR4 nanobodies (VHH-based single variable domains) potently inhibit chemotaxis and HIV-1 replication and mobilize stem cells. Proc Natl Acad Sci U S A 107:20565–20570 7. Maussang D, Mujic´-Delic´ A, Descamps FJ et al (2013) Llama-derived single variable domains (nanobodies) directed against chemokine receptor CXCR7 reduce head and neck cancer cell growth in vivo. J Biol Chem 288:29562–29572 8. Nguyen VK, Hamers R, Wyns L et al (2000) Camel heavy-chain antibodies: diverse germline VHH and specific mechanisms enlarge the antigen-binding repertoire. EMBO J 19:921–930

9. Muyldermans S, Smider VV (2016) Distinct antibody species: structural differences creating therapeutic opportunities. Curr Opin Immunol 40:7–13 10. Krah S, Schro¨ter C, Zielonka S et al (2016) Single-domain antibodies for biomedical applications. Immunopharmacol Immunotoxicol 38:21–28 11. Conrath KE, Wernery U, Muyldermans S et al (2003) Emergence and evolution of functional heavy-chain antibodies in Camelidae. Dev Comp Immunol 27:87–103 12. Vincke C, Loris R, Saerens D et al (2009) General strategy to humanize a camelid single-domain antibody and identification of a universal humanized nanobody scaffold. J Biol Chem 284:3273–3284 13. Tijink BM, Laeremans T, Budde M et al (2008) Improved tumor targeting of anti-epidermal growth factor receptor Nanobodies through albumin binding: taking advantage of modular Nanobody technology. Mol Cancer Ther 7:2288–2297 14. Helma J, Cardoso MC, Muyldermans S et al (2015) Nanobodies and recombinant binders in cell biology. J Cell Biol 209:633–644 ˜ ez-Prado N, 15. Alvarez-Cienfuegos A, Nun Compte M et al (2016) Intramolecular trimerization, a novel strategy for making multispecific antibodies with controlled orientation of the antigen binding domains. Sci Rep 6:28643 16. Desmyter A, Spinelli S, Boutton C et al (2017) Neutralization of human interleukin 23 by multivalent nanobodies explained by the structure of cytokine–nanobody complex. Front Immunol 8:884

Isolation of VHH Antibodies by Yeast Surface Display 17. Goldman E, Liu J, Bernstein R et al (2009) Ricin detection using phage displayed single domain antibodies. Sensors 9:542–555 18. Yan J, Li G, Hu Y et al (2014) Construction of a synthetic phage-displayed Nanobody library with CDR3 regions randomized by trinucleotide cassettes for diagnostic applications. J Transl Med 12:343 19. Bencurova E, Pulzova L, Flachbartova Z et al (2015) A rapid and simple pipeline for synthesis of mRNA–ribosome–V H H complexes used in single-domain antibody ribosome display. Mol BioSyst 11:1515–1524 20. Romao E, Morales-Yanez F, Hu Y et al (2016) Identification of useful nanobodies by phage display of immune single domain libraries derived from camelid heavy chain antibodies. Curr Pharm Des 22:6500–6518 21. Cavallari M (2017) Rapid and direct VHH and target identification by staphylococcal surface display libraries. Int J Mol Sci 18:1507 22. Eden T, Menzel S, Wesolowski J et al (2018) A cDNA immunization strategy to generate nanobodies against membrane proteins in native conformation. Front Immunol 8:1989 23. Wu Y, Jiang S, Ying T (2017) Single-domain antibodies as therapeutics against human viral diseases. Front Immunol 8:1802 24. Pardon E, Laeremans T, Triest S et al (2014) A general protocol for the generation of

189

Nanobodies for structural biology. Nat Protoc 9:674–693 25. McMahon C, Baier AS, Pascolutti R et al (2018) Yeast surface display platform for rapid discovery of conformationally selective nanobodies. Nat Struct Mol Biol 25:289–296 26. Boder ET, Wittrup KD (1997) Yeast surface display for screening combinatorial polypeptide libraries. Nat Biotechnol 15:553–557 27. Lu Z-J (2012) Frontier of therapeutic antibody discovery: The challenges and how to face them. World J Biol Chem 3:187 28. Doerner A, Rhiel L, Zielonka S et al (2014) Therapeutic antibody engineering by high efficiency cell screening. FEBS Lett 588:278–287 29. Zielonka S, Weber N, Becker S et al (2014) Shark Attack: high affinity binding proteins derived from shark vNAR domains by stepwise in vitro affinity maturation. J Biotechnol 191:236–245 30. Zielonka S, Empting M, Ko¨nning D et al (2015) The shark strikes twice: hypervariable loop 2 of shark IgNAR antibody variable domains and its potential to function as an autonomous paratope. Mar Biotechnol (NY) 17:386–392 31. Grzeschik J, Ko¨nning D, Hinz SC et al (2018) Generation of semi-synthetic shark IgNAR single-domain antibody libraries. Methods Mol Biol 1701:147–167

Chapter 11 Selection and Characterization of Anti-idiotypic Shark Antibody Domains Doreen Ko¨nning, Stefan Zielonka, Anna Kaempffe, Sebastian J€ager, Harald Kolmar, and Christian Schro¨ter Abstract The antibody repertoire of cartilaginous fish comprises an additional heavy-chain-only antibody isotype that is referred to as IgNAR (immunoglobulin novel antigen receptor). Its antigen-binding site consists of one single domain (vNAR) that is reportedly able to engage a respective antigen with affinities similar to those achieved by conventional antibodies. While vNAR domains offer a reduced size, which is often favorable for applications in a therapeutic as well as a biotechnological setup, they also exhibit a high physicochemical stability. Together with their ability to target difficult-to-address antigens such as virus particles or toxins, these shark-derived antibody domains seem to be predestined as tools for biotechnological and diagnostic applications. In the following chapter, we will describe the isolation of anti-idiotypic vNAR domains targeting monoclonal antibody paratopes from semi-synthetic, yeast-displayed libraries. Anti-idiotypic vNAR variants could be employed for the characterization of antibody-based therapeutics (such as antibody-drug conjugates) or as positive controls in immunogenicity assays. Peculiarly, when using semisynthetic vNAR libraries, we found that it is not necessary to deplete the libraries using unrelated antibody targets, which enables a fast and facile screening procedure that exclusively delivers anti-idiotypic binders. Key words Shark, IgNAR, vNAR, Yeast surface display, Antibody engineering, Protein engineering, Anti-idiotypic, Anti-ID, Single-domain antibody

1

Introduction Challenging environmental conditions in the marine sea have caused sharks and cartilaginous fish to adapt in a specialized manner [1]. In order to avoid dehydration, these species enrich high concentrations of urea in their blood, however, this puts a strain on the proteins in their organism and also impacts their immune system. Sharks comprise an additional antibody isotype that is formed by only two heavy chains [2]. In contrast to the heterotetrameric architecture of two associated heavy and light chains commonly found in conventional antibodies, the simple structure of shark antibodies reduces the antigen-binding region to a minimum.

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2070, https://doi.org/10.1007/978-1-4939-9853-1_11, © Springer Science+Business Media, LLC, part of Springer Nature 2020

191

192

Doreen Ko¨nning et al.

Precisely, homodimeric shark antibodies of the IgNAR isotype (immunoglobulin novel antigen receptor) possess only a single variable domain that mediates binding to the corresponding antigens [3]. However, the affinities that have been observed for these antibody–antigen interactions range in the same order of magnitude as for conventional mammalian antibodies [4]. When comparing the structures of a human heavy chain variable (V) domain and a shark IgNAR V domain (referred to as vNAR in the following), it becomes apparent that the deletion of a ß-strand that normally contributes to the formation of the CDR2 binding site is missing in the shark equivalent [3]. As such, vNARs comprise only two CDR regions instead of three. Yet, at the CDR2 deletion site, the remaining surface exposed loop wraps around the bottom of the molecule in a “strap-like” manner, and it was shown that, after antigen exposure, somatic hypermutation also occurs in this region. In addition, hypermutation leads to the diversification of an additional loop which corresponds to HV4 in T-cell receptors [5]. Consequently, these regions are referred to as HV2 and HV4, respectively. Interestingly, vNAR domains also exist as T-cell receptor fusion proteins, supporting the perception that IgNAR most likely plays a role in both, T-cell and B-cell immunology [6]. In contrast to the variable domain of a human heavy chain, vNAR domains exhibit drastically elongated CDR3 binding loops (Fig. 1a). Whereas these regions generally comprise between 8 and

Fig. 1 (a) Structural representation of the variable domain of an IgNAR heavy-chain-only antibody (vNAR). The CDR3 binding site is highlighted in magenta. The structure was modified from PDB identifier 2I25 using UCSF Chimera [33]. (b) Schematic depiction of an exemplary protein setup on the yeast surface for high-throughput screening. The orientation of the vNAR domain (C-terminal fusion to Aga2p) can be altered as well as the Cterminal peptide tag that is used for the verification of full-length surface display of the fusion protein. In addition to the Aga1p-Aga2p system, other cell wall anchor proteins have been described in the literature [34]

Selection of Anti-Idiotypic Shark Antibody Domains

193

12 amino acids in the human counterparts, vNAR domains with a CDR3 length of up to 34 amino acids have been reported [6]. This leads to the formation of complex three-dimensional structures, which can additionally be stabilized by intramolecular disulfide bonds, and affords the benefit of engaging hidden and cleft-like epitopes [3]. It has been shown that vNARs are even able to interact with the active site of enzymes [7] as well as virus particles [8, 9] or toxins [10, 11]. VNAR domains can be divided into overall four different subtypes based on the number and pattern of non-canonical disulfide bonds that are usually absent in conventional antibody domains [3]. This results in an unparalleled repertoire of unique loop structures able to engage a wide range of epitopes. Their high thermal stability as well as tolerance to irreversible denaturation are other beneficial attributes of shark vNAR domains [4]. Accordingly, target-specific vNAR molecules have been isolated against a wide range of disease-related antigens [12–16]. In addition, it has been demonstrated by our group that these molecules can also be engineered towards pH-dependent antigen-binding [17, 18], opening up new avenues for the development of robust and tailored affinity ligands. Because of their unique ability to engage cryptic and recessed epitopes, vNAR domains seem to be well suited for various applications ranging from diagnostics to therapy [19]. In the following chapter, we will describe the isolation and characterization of antiidiotypic vNAR domains that can specifically bind to an idiotope located in the variable domain of a therapeutic antibody [20, 21]. In general, anti-idiotypic entities are often employed for the thorough pre-clinical characterization of antibody-based therapeutics (such as an antibody-drug conjugate, for instance) or for the monitoring of these therapeutic drugs in the patient during clinical development (i.e., pharmacokinetic analyses). Whereas these experiments can usually be conducted upon using the drug’s antigen, it becomes a more challenging task as soon as this antigen is not easily available, expensive, or hazardous. The most important prerequisite for anti-idiotypic binders is a very high specificity that enables a clear differentiation between the target molecule and competitor immunoglobulins present in patient serum. Additional applications for anti-idiotypic binders include their utilization as a positive control in the identification of antidrug responses [22] for the assessment of potential immunogenicity issues, the purification of bispecific antibodies [20, 23] and the development of cancer vaccines [24–27].

194

2

Doreen Ko¨nning et al.

Materials

2.1 Cultivation and Immunofluorescence Staining of vNAR-Displaying Yeast Cells

1. vNAR-displaying yeast cell library [18, 20, 28]. 2. SD-CAA medium: 8.6 g/L NaH2PO4
H2O, 5.4 g/L Na2HPO4, 20 g/L D(+)-glucose, 6.7 g/L yeast nitrogen base without amino acids, and 5 g/L Bacto™ Casamino Acids. 3. SD-CAA agar plates: 8.6 g/L NaH2PO4
H2O, 5.4 g/L Na2HPO4, 20 g/L D(+)-galactose, 6.7 g/L yeast nitrogen base without amino acids, 5 g/L Bacto™ Casamino Acids. 4. SG-CAA medium: 8.6 g/L NaH2PO4
H2O, 5.4 g/L Na2HPO4, 20 g/L D(+)-galactose, 6.7 g/L yeast nitrogen base without amino acids, and 5 g/L Bacto™ Casamino Acids. 5. Phosphate-buffered saline: 137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.8 mM KH2PO4, pH 7.4 (can also be purchased). 6. Staining solution: Phosphate-buffered saline with 0.1% (w/v) bovine serum albumin. 7. Monoclonal antibody target of interest (i.e., cetuximab or matuzumab in this case). 8. Unrelated control antibodies (e.g., SILu™ Lite SigmaMAb Universal antibody standard from Sigma-Aldrich, trastuzumab, adalimumab, etc.). 9. Anti-c-myc-biotin antibody (mouse, monoclonal, Miltenyi Biotech). 10. Streptavidin allophycocyanin (Affymetrix eBioscience). 11. Anti-human Fcγ-specific Fab fragment labeled with Alexa Fluor488® (goat, polyclonal, Jackson ImmunoResearch). 12. Spectrophotometer and suitable cuvettes. 13. Sterile cell culture flasks. 14. Lab incubator. 15. Cell sorter and corresponding software.

2.2 Subcloning of Anti-idiotypic vNAR Variants

1. Zymoprep Yeast Plasmid Miniprep (Zymo Research). 2. Suitable expression plasmid (we used the pEXPR-IBA42 plasmid that already encoded a human IgG1 Fc region [29]). 3. Q5 High Fidelity DNA polymerase (New England Biolabs). 4. 10
Q5 reaction buffer (New England Biolabs). 5. dNTP mixture, 10 mM each. 6. Nuclease-free water. 7. Thermocycler. 8. Chamber and reagents for agarose gel electrophoresis.

Selection of Anti-Idiotypic Shark Antibody Domains

195

9. PCR clean-up kits for small- and larger-scale DNA purifications (e.g., the Wizard® SV Gel and PCR Clean-up System from Promega). 10. NheI-HF (New England Biolabs). 11. ApaI-HF (New England Biolabs). 12. 10
CutSmart Buffer (New England Biolabs). 13. T4 DNA ligase (New England Biolabs). 14. 10
T4 ligase buffer (New England Biolabs). 15. Ethanol, absolute. 16. 7 M ammonium acetate. 17. Electrocompetent E. coli cells. 18. Double Yeast Tryptone medium (dYT): 16 g/L tryptone, 10 g/L yeast extract, 5 g/L NaCl. 19. Ampicillin. 20. Electroporator GenePulser Xcell™ (Bio-Rad). 21. Electroporation cuvettes, 0.2 cm (Bio-Rad). 2.3 Recombinant Expression of Antiidiotypic vNARs

1. Gibco™ Expi293F™ cells (ThermoFisher Scientific). 2. Gibco™ ExpiFectamine™ 293 Transfection Kit (including Enhancer 1 and 2; ThermoFisher Scientific). 3. Gibco™ Opti-MEM™ reduced serum medium (ThermoFisher Scientific). 4. Gibco™ Expi293 Scientific).

expression

medium

(ThermoFisher

5. 50 mL conical tubes with filtertops. 6. Sterile cell culture flasks (volume depends on the expression scale). 7. Lab shaker for cell culture flasks. 8. Montage ProSep® A Protein A spin columns (Merck Millipore). 9. Amicon® Centrifugal Filters (Merck Millipore). 2.4 Characterization of Recombinant Antiidiotypic vNAR Variants

1. Phosphate-buffered saline:137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.8 mM KH2PO4, pH 7.4. 2. Tween®-20 (Sigma-Aldrich). 3. Kinetics buffer: PBS, 0.1% (w/v) BSA, 0.02% (v/v) Tween-20. 4. Black, flat-bottom 96-well plates. 5. Rack for sensortips (Forte´Bio®, Pall). 6. Anti-Fab-CH1 (2nd Generation) or Streptavidin Sensortips (Forte´Bio®, Pall).

196

Doreen Ko¨nning et al.

7. Octet®RED96 Bio-Layer Interferometry device with Acquisition and Evaluation Software (Forte´Bio®, Pall). 8. Nunc 96-well MaxiSorp® plates (ThermoFisher Scientific). 9. Blocking solution: PBS, 3% (w/v) BSA. 10. Washing solution: PBS, 0.1% (v/v) Tween-20. 11. Monoclonal antibody target of interest (i.e., cetuximab or matuzumab in this case). 12. Unrelated control antibodies (e.g., SILu™ Lite SigmaMAb Universal antibody standard from Sigma-Aldrich, trastuzumab, adalimumab, etc.). 13. Human serum (Merck Millipore). 14. Anti-human Fab antibody labeled with horseradish peroxidase (Sigma-Aldrich). 15. TMB One solution (Promega). 16. 0.2 M HCl. 17. Plate reader.

3

Methods The starting point for the isolation of anti-idiotypic vNAR domains in this case was a mixture of semi-synthetic and histidine-enriched, yeast-displayed vNAR libraries (Fig. 1b) with varying lengths in the CDR3 binding site (see Notes 1–3). Methods for the establishment of such libraries can be found in refs. 18, 28. For detailed publications on yeast surface display, the reader is referred to refs. 30, 31.

3.1 Induction of vNAR Expression

1. Thaw library cryo stocks (stored at 80  C) and add them to 1 L of sterile SD-CAA medium. 2. After shaking the culture at 180 rpm and 30  C overnight, measure the optical density. 3. Take out a volume that corresponds to an optical density (OD600) of 0.5 in 1 L (corresponds to 0.5
107 cells/mL). 4. Centrifuge the cells at 4000
g for 5 min. 5. Wash the cell pellet once with PBS upon resuspending the cells in 1 mL of buffer, followed by gentle mixing and centrifugation of cells at 4000
g for 5 min and removal of the supernatant. 6. Resuspend the cell pellet in sterile SG-CAA medium and transfer the suspension into a flask containing 1 L of SG-CAA medium. 7. Incubate the cells at 180 rpm and 20  C for 2 days.

Selection of Anti-Idiotypic Shark Antibody Domains

3.2 Immunofluorescence Staining of Yeast Cells

197

Detailed procedures for the immunofluorescence staining of yeast cells can be found in ref. 31. Make sure to prepare a control sample of cells which are not incubated with the target antibody but only with the detection reagents. This will allow you to adequately set a sorting gate later on. 1. Measure the optical density (OD600) of the yeast culture in SG-CAA medium. 2. For all screening rounds, it is recommended to utilize at least ten times the cell number corresponding to the theoretical diversity of your library or the number of cells isolated during the previous sort, respectively. 3. Centrifuge the respective volume of cell suspension at 4000
g for 5 min. 4. Discard the supernatant and wash the pellet twice with PBS. 5. Resuspend the cell pellet in an appropriate volume of the monoclonal antibody target in staining solution for 1 h and at room temperature (when working with a semi-synthetic library, we recommend starting with an antigen concentration of 1 μM). For the preparation of the negative control, simply add the same volume of staining solution only. 6. Centrifuge the cell suspension at 4000
g for 5 min. 7. Wash the cell pellet twice with staining solution. From here, all subsequent steps should be conducted on ice to avoid dissociation of the antigen from the displayed vNAR domains. 8. Resuspend the yeast cells in an appropriate volume of primary detection antibody, i.e., an anti-myc antibody conjugated to biotin, and incubate for 30 min on ice. 9. Wash the cell pellet once with staining solution. 10. Resuspend the cell pellet in an appropriate volume of fluorescently-labeled secondary detection antibodies (streptavidin allophycocyanin diluted at 1:100 in staining solution and an anti-human Fc-specific Fab fragment conjugated to Alexa Fluor® 488 diluted at 1:200), and incubate on ice and in the dark for 30 min. 11. Centrifuge the cell suspension at 4  C for 5 min. 12. Discard the supernatant. 13. Keep the cell pellets on ice. 14. As soon as the cell sorter is set up and the lasers are adjusted, resuspend the cell pellet in an appropriate volume of staining solution and start sorting. Detailed protocols for fluorescence-

198

Doreen Ko¨nning et al.

activated cell sorting (FACS) of yeast libraries can be found elsewhere [31, 32]. 15. Once the sort is finished, transfer the isolated cells into an appropriate volume of SD-CAA medium. 16. Incubate the cell suspension at 30  C for 24 h and repeat the steps listed in Subheadings 3.1 and 3.2. The staining protocol described here (including the antigen concentration) can be applied to all subsequent sorting rounds (see Notes 4–6). If needed, the antigen concentration can be decreased in order to increase the screening stringency and the likelihood of identifying high-affinity binders. In general, when using semi-synthetic vNAR libraries, it is not necessary to deplete the library against unrelated human immunoglobulins or to actively perform counter selections in order to avoid the enrichment of vNAR variants targeting the constant regions of the monoclonal antibody target [20]. 3.3 Verification of Anti-idiotypic Binding on the Surface of Yeast

Once you see an enrichment of target-binding yeast cells, you can start to verify if these binders are anti-idiotypic. The steps described here can also be applied to individual single clones in order to ensure the selection of the best performing vNAR variants for recombinant expression. Single clones can be isolated upon plating the cell suspension after the last sorting round on SD-CAA-W agar plates and upon incubation at 30  C for 2–3 days. Afterwards, single colonies can be picked up with a sterile pipette tip and resuspended in 2–5 mL of SD-CAA-W medium. 1. Induce the yeast cell population or the respective single clone upon transferring the cells from SD-CAA to SG-CAA medium as described in Subheading 3.1. 2. Prepare overall three samples comprising an adequate number of yeast cells (i.e., 5
106 cells for a single clone). 3. Centrifuge the cells and discard the supernatant. 4. Resuspend the cell pellets in an appropriate volume of either staining solution only (negative control), 1 μM of the target monoclonal antibody (sample 1), or a tenfold molar excess of unrelated human antibodies (10 μM; sample 2). 5. Incubate the cells for 1 h at room temperature. 6. Centrifuge the cells at 4000
g for 5 min. 7. Continue with the protocol for immunofluorescence staining (Subheading 3.2) starting from step 4. An example for dot plot diagrams depicting yeast cells that present anti-idiotypic vNAR domains is shown in Fig. 2.

Selection of Anti-Idiotypic Shark Antibody Domains

199

Fig. 2 Dot plot diagrams depicting vNAR-displaying yeast cells that show anti-idiotypic binding towards the respective human monoclonal antibody target (matuzumab in this case) [20]. (a) Cells of the final sorting round were analyzed in terms of anti-idiotypic binding to matuzumab. Towards this end, two samples were prepared. One was solely incubated with the detection antibodies but without matuzumab (left, negative control), whereas the second sample was incubated with 500 nM of matuzumab. (b) Dot plot diagrams depicting an exemplary single clone which was isolated from the final sorting round against matuzumab (depicted in a). Overall, three samples were prepared in this case. The dot plot diagram on the left displays yeast cells which were only incubated with the detection antibodies but without the human antibody target. The plot in the middle represents cells which were incubated with 100 nM of matuzumab, while the dot plot diagram on the right shows cells which were incubated with a tenfold molar excess of cetuximab (an unrelated, EGFR-targeting human antibody)

3.4 Reformatting and Recombinant Expression of Antiidiotypic vNAR Domains

In the following chapter, the identified anti-idiotypic vNAR variants will be reformatted as human Fc fusion proteins and expressed in mammalian cells (see Note 7). Primer sequences utilized for subcloning into a mammalian expression plasmid are depicted in Table 1. In this case, the plasmid pEXPR-IBA42 which already encoded the sequence for a human IgG Fc sequence was utilized for recombinant expression [29]. However, the expression system as well as the format of the recombinant vNAR can be adapted to individual preferences. After isolation of the plasmid DNA from yeast, the vNAR insert is amplified in a two-step PCR reaction that incorporates a hinge region separating the vNAR and the vector-

200

Doreen Ko¨nning et al.

Table 1 Primers utilized for the subcloning of vNAR variants into the pEXPR-IBA42 human Fc expression plasmid [29] Name

Sequence (50 !30 )

pEXPR_NheI_up

GCGCGCGCTAGCCGCTGAGAACCTGTACTTCCAGAGCATGGCCGCAC GGCTTGAACAAACACCGACA

vNAR_lo

GGTGTGGGTCTTGTCGCAGCTCTTGGGCTCGCTTCCGCTCTGGAAG TACAGGTTCTCTTTCACAGTCAGAGGGGTCCCGCCACCTTCa

pEXPR_ApaI_lo

GCGCGCGGGCCCGCCCAGCAGTTCAGGGGCAGGGCAGGGAGGACA GGTGTGGGTCTTGTCGCAGCTCTTGGGCTCGCTTCC

Note that the sequences add recognition sequences for tobacco etch virus protease at the N- and C-terminus of the vNAR domain. In addition, they encode a hinge region that connects the vNAR domain with the human Fc portion, as reformatting into a bivalent Fc fusion protein was desired in this case (see Note 8) a Note that the underlined sequence is vNAR specific. It has to be adapted to the sequence of the individual vNAR domain that is to be subcloned

Table 2 Reaction conditions for polymerase chain reactions 98  C

2 min

98  C 55  C 72  C

10 s 20 s 20 s

72  C

2 min

30 cycles

encoded Fc region, as well as recognition sequences for restriction enzymes. For all PCR reactions, the conditions were set as follows (Table 2): 3.4.1 Isolation of vNAR DNA from Yeast Cells and Subcloning into the pEXPR Plasmid

1. Inoculate 5 mL of SD-CAA medium with the single clone of interest. 2. Grow the culture at 30  C and at 180 rpm overnight. 3. Isolate the plasmid DNA from the yeast cells with a yeast DNA isolation kit according to the manufacturer’s instructions. 4. Prepare a PCR reaction upon mixing the following reagents: approximately 100 ng of yeast DNA (around 1 μL), 2.5 μL of the primer pEXPR_NheI_up, 2.5 μL of the primer vNAR_lo, 0.5 μL of Q5 polymerase, 10 μL of 5
reaction buffer, 1 μL of dNTP mixture. Add nuclease-free water to a final volume of 50 μL.

Selection of Anti-Idiotypic Shark Antibody Domains

201

5. Perform the PCR reaction in a Thermocycler using the protocol given in Table 2. Afterwards, separate the PCR products on a 1% (w/v) agarose gel. The amplified PCR product should be visible as a distinct band at a size of approx. 400 bp. Purify the PCR reaction upon using a PCR clean-up kit according to the manufacturer’s instructions. 6. Carry out a second PCR reaction upon mixing the following reagents: 100 ng of the amplified vNAR PCR product (generated in the previous step), 2.5 μL of the pEXPR_NheI_up, 2.5 μL of the primer pEXPR_ApaI_lo, 0.5 μL of Q5 polymerase, 10 μL of 5
reaction buffer. Add nuclease-free water to a final volume of 50 μL. 7. Perform the PCR reaction in a Thermocycler using the protocol given in Table 2. Afterwards, separate the PCR products on a 1% agarose gel. The amplified PCR product should be visible as a distinct band at a size of approx. 480 bp. Purify the PCR reaction upon using a PCR clean-up kit according to the manufacturer’s instructions. 8. Digest the purified vNAR PCR product with the ApaI restriction enzyme upon pipetting together the following reagents on ice: 44 μL of purified PCR product (approx. 1–2 μg; one may reduce the amount of insert; however, we usually digest as much as possible), 5 μL of 10
CutSmart buffer, and 1 μL of ApaI enzyme. 9. Mix gently and incubate at 25  C for 1 h. 10. Add 1 μL of the NheI-HF restriction enzyme to the reaction and incubate at 37  C for 1 h. 11. Run a preparative 1% (v/v) agarose gel and cut out the band corresponding to the digested PCR product. It is advisable to apply the complete digestion mixture to the gel since the yield after gel extraction tends to be low. 12. Perform a gel extraction of the digested PCR product upon using a PCR clean-up kit according the manufacturer’s instructions. 13. Digest the pEXPR plasmid in parallel upon pipetting the following reagents: a volume of the plasmid that corresponds to around 3–5 μg, 5 μL of 10
CutSmart buffer, and 1 μL of ApaI restriction enzyme. Add nuclease-free water to a final volume of 50 μL. If the concentration of plasmid DNA is low, you can use plasmid solution instead of water. 14. Mix gently and incubate at 25  C for 1 h. 15. Add 1 μL of the NheI-HF restriction enzyme to the reaction and incubate at 37  C for 1 h.

202

Doreen Ko¨nning et al.

16. Run a preparative 1% (w/v) agarose gel and cut out the band corresponding to the digested plasmid. It is advisable to apply the complete digestion mixture to the gel since the yield after gel extraction tends to be low. 17. Perform a gel extraction of the digested PCR product upon using a PCR clean-up kit according the manufacturer’s instructions. 18. Set up the ligation reaction upon pipetting together the following reagents on ice in a total volume of 20 μL: 1 equivalent of digested pEXPR plasmid (we recommend using around 3 μg), 3 equivalents of digested PCR product, 2 μL of 10
T4 ligase buffer, and 1 μL of T4 ligase. If necessary, add nuclease-free water to a final volume of 20 μL. 19. Mix gently and incubate at 16  C for at least 2 h (this can also be performed overnight). 20. Purify the reaction upon precipitating the ligated pEXPRvNAR plasmid with a 7 M solution of ammonium acetate and ethanol. To do so, add 2 μL of ammonium acetate and 60 μL of pure ethanol. 21. Incubate the mixture at 20  C for at least 1 h (this can also be performed overnight). 22. Centrifuge the mixture in a pre-cooled centrifuge at 4  C. 23. Carefully discard the supernatant. 24. Air-dry the DNA pellet (the pellet might not be visible to the eye). 25. Resuspend the pellet in 10 μL of nuclease-free water. 26. Electroporate competent E. coli cells (we use 50 μL aliquots of electrocompetent cells) together with 2–5 μL of the purified ligation product. 27. Incubate E. coli cells at 37  C and at 180 rpm overnight in at least 50 mL of dYT medium with 100 μg/mL ampicillin. 28. Isolate the ligation mixture from E. coli cells upon performing a larger scale DNA preparation (e.g., upon using the Promega Midi Prep Kit according to the manufacturer’s instructions (see Note 9). 29. Sequence-verify the ligated pEXPR-vNAR plasmid. 3.4.2 Transfection of Expi293F™ Cells

For the expression of the vNAR-Fc fusion protein from the pEXPR-vNAR plasmid, we chose the HEK293 cell variant Expi293F™. However, expression in regular HEK cells is also possible. 1. Prepare a sufficient volume of Expi293F™ cells prior to transfection. Ideally, 200–500 mL of a cell suspension with a cell

Selection of Anti-Idiotypic Shark Antibody Domains

203

number of approx. 3
106 cells/mL should be sufficient for several small-scale expressions (25 mL per expression). 2. Dilute 25 μg of pEXPR-vNAR plasmid DNA in 1.25 mL of Opti-MEM™ medium. 3. Dilute 80 μL of ExpiFectamine™ in 1.25 mL of Opti-MEM™ medium. 4. Incubate for 5 min at room temperature. 5. Add the DNA to the ExpiFectamine™ solution and mix. 6. Incubate at room temperature for 30 min. 7. In the meantime, determine the viable cell number of the Expi293F™ cell culture. 8. Prepare 21 mL of cell suspension with a cell number of 2.5
106 cells/mL in a 50 mL conical tube comprising a filtertop. 9. Add the DNA/ExpiFectamine™ mixture slowly to the cells. 10. Incubate the conical tubes at 37  C, 5% CO2, and 180 rpm overnight. 11. After 16–18 h, add 150 μL of Enhancer 1 and 1.5 mL of Enhancer 2. 12. Harvest the cells 5 days after transfection by centrifuging the conical tubes at 1000
g and 4  C for 20 min. Make sure to activate the brake (level ~2). 13. Carefully transfer the supernatant into a new 50 mL conical tube. 14. Sterile-filter the supernatant through a membrane with a pore size of 0.22 μM. 15. Purify the recombinant vNAR-Fc proteins using Protein A chromatography (or any other purification method, depending on the vNAR format decided on in Subheading 3.4). We generally utilize ProSep® Protein A columns according to the manufacturer’s instructions. 16. If necessary and depending on the purification method, perform a buffer exchange using dialysis or an Amicon® Ultrafilter unit with an appropriate cutoff (we typically use a 10 kDa cutoff). 3.4.3 Characterization of Recombinant and Antiidiotypic vNARs via Bio-Layer Interferometry (BLI)

For the initial characterization of vNAR variants, we usually determine the affinity constant KD as well as the binding kinetics via Bio-Layer Interferometry (BLI). Since the vNAR domain is fused to an Fc region, we immobilize the target antibody on anti-Fab sensortips that would not interact with the vNAR-Fc. Alternatively, either the target antibody or the vNAR-Fc protein can be biotinylated and immobilized on streptavidin sensortips.

204

Doreen Ko¨nning et al.

If no BLI device is available, the affinity of the vNAR-Fc variants can also be determined using an ELISA assay (see Subheading 3.4.4). 1. Pipette 200 μL of PBS into 8 wells of a black, flat-bottom 96-well plate (A1 to H1). 2. Place the plate into the Octet® sensortip rack and carefully insert anti-Fab-CH1 sensortips (2nd Generation) into the pinholes. 3. Leave the sensortips to adjust in PBS for at least 10 min at room temperature. 4. Meanwhile, dilute the target antibody to 10 μg/mL in PBS. 5. Pipette 200 μL of the antibody solution into overall 8 wells of a new, black 96-well flat-bottom plate (i.e., wells A1 to H1). 6. Pipette 200 μL of 1
kinetics buffer into wells A2 to H2. 7. Prepare serial dilutions of the vNAR-Fc variant in 1
kinetics buffer. It is recommended to start with a concentration that is around 10–50
higher than the expected affinity binding constant (KD, can be estimated from the binding intensities observed on yeast cells). In our case, we started with a vNARFc concentration of 500 nM and subsequently performed a 1:1 dilution in kinetics buffer (A3 to G3). Well H3 only contained 1
kinetics buffer as a negative control. Make sure to always include appropriate controls during the BLI measurements (such as an unrelated antibody target). 8. Set up the Bio-Layer Interferometry measurement using the following steps and parameters: Loading for 300 s, Baseline for 120 s, Association for 300 s and Dissociation for 600 s (these conditions can be altered individually, however, this is the standard setup we usually employ). 9. Place both, the vNAR sample plate and the rack containing the equilibrated sensortips, into the Octet®RED device. 10. Start the measurement. 11. Determine the binding kinetics by using the respective evaluation software and upon applying a 1:1 Langmuir binding fit (Global Fit). Make sure to subtract the control wells beforehand. 3.4.4 Characterization of Recombinant and Antiidiotypic vNARs via Ligand Binding Assays (ELISA)

With respect to confirming the anti-idiotypic binding behavior of the recombinant vNAR variants, enzyme-linked immunosorbent (ELISA) assays employing unrelated antibodies or the target antibody in the presence of serum can be conducted (Fig. 3). After coating the plate with the vNAR variant, it is recommended to add the target antibody or the unrelated antibody controls in duplicate or triplicate. Make sure to prepare a sufficient number of plates and

Selection of Anti-Idiotypic Shark Antibody Domains

205

Fig. 3 Results of a specificity as well as a serum ELISA assay encompassing two anti-idiotypic vNAR variants targeting matuzumab or cetuximab, respectively (VNAR1 and 2) [20]. (a) Recombinantly expressed antiidiotypic vNAR variants 1 and 2 were coated onto 96-well microtiter plates and incubated with cetuximab, matuzumab, and other unrelated monoclonal antibodies. Detection of bound antibodies was performed upon using a Fab-specific detection antibody conjugated to horseradish peroxidase. (b) Anti-idiotypic vNAR variants vNAR1 and 2 were coated onto 96-well microtiter plates. The vNAR-coated wells were subsequently incubated with serial dilutions of biotinylated matuzumab (vNAR1) or cetuximab (vNAR2) in the presence (light gray) or absence (dark gray) of 10% human serum

scale up the protocol if needed. For the detection of the target antibody from human serum, it is useful to biotinylate the antibody beforehand [20]. 1. Start by preparing a 2 μg/mL solution of vNAR-Fc protein in PBS. 2. Pipette 100 μL of the vNAR solution into each well of one half of a 96-well MaxiSorp® plate. 3. Pipette 100 μL of PBS into the wells of the second half of the plate (this serves the purpose of identifying any unspecific binding of the target and unrelated antibodies as well as secondary reagents to the uncoated plate). 4. Incubate the plate at 4  C overnight. 5. Discard the coating solutions.

206

Doreen Ko¨nning et al.

6. Pipette 100 μL of blocking solution to each well of the plate. 7. Incubate the plate at room temperature for 2 h. 8. Discard the blocking solution and wash the plate thrice with washing solution. 9. Coat both halves of the plate with the target antibody (in duplicate or triplicate) or unrelated antibodies (in duplicate or triplicate) upon adding 100 μL of a 500–1000 nM solution in PBS with 1% (w/v) BSA to each well. Remark: If you aim at performing an affinity titration, simply prepare a serial dilution of the target antibody (i.e., dilute a 1000 nM solution 1:1 with PBS 1% BSA) and add 100 μL to each half of the plate. In addition, you can perform a second set of serial dilutions in 5–10% human serum to further investigate the specificity of the coated, anti-idiotypic vNAR variant. You could also use biotinylated target antibody for this purpose as it might give better signals. 10. Incubate the plate at room temperature for 1 h. 11. Discard the antibody solutions and wash the plate thrice with washing solution. 12. Add 100 μL of an anti-human Fab-specific antibody conjugated to horseradish peroxidase to each treated well. The detection antibody should be diluted in PBS with 1% (w/v) BSA according to the manufacturer’s instructions. If using biotinylated target antibody, the detection would be carried out with a Streptavidin-horseradish peroxidase conjugate. 13. Incubate for 1 h at room temperature. 14. Discard the detection antibody and wash the plate thrice with washing solution. 15. Add 100 μL of TMB solution to each well. 16. Once the blue color has developed to an appropriate extent, stop the reaction upon adding 100 μL of a 0.2 M HCl solution to each well. 17. Read the plate at 450 nm and subtract the value of the uncoated half of the plate from the read values of the coated side.

4

Notes 1. The setup of the proteins on the yeast surface can be adjusted to individual preferences (N- or C-terminal fusion, myc tag or not, etc.). 2. Note that the library does not have to be histidine-enriched.

Selection of Anti-Idiotypic Shark Antibody Domains

207

3. It is possible to elongate the CDR3 even further or add additional lengths. In this particular case, the CDR3 length distribution encompassed 12, 14, 16, and 18 amino acids. 4. Although it is not necessary with semi-synthetic vNAR libraries, a Fab fragment of the antibody target of interest can be employed during the screenings. This way, the likelihood of isolating binders that target the constant rather than the antibody variable region could be decreased. 5. The detection reagents used for immunofluorescence stainings can be alternated throughout the process in order to avoid enrichment of vNAR variants that interact with the detection reagents. 6. Alternatively, the library can be depleted against the detection antibodies prior to screening upon performing magnetic cell sorting. This will also reduce the library size and shortens the sorting time during the first FACS screening round. 7. Reformatting of the vNAR variants as Fc fusion proteins is not necessary per se, they can also be expressed in a solitary form. However, the yields might not be optimal. 8. The primer sequences we used for the reformatting of vNAR variants as Fc fusion proteins attach a hinge region that connects the vNAR domain with the human Fc portion. In addition, the primers add a recognition sequence (amino acid sequence ENLYFQS) for tobacco etch virus protease at the Nand C-terminus of the vNAR variant. 9. It can be appropriate to perform a small-scale DNA isolation first in order to verify the sequence of the ligated plasmid before doing a larger DNA preparation. References 1. Hammerschlag N (2006) Osmoregulation in elasmobranchs: a review for fish biologists, behaviourists and ecologists. Mar Behav Physiol 39:209–228 2. Greenberg AS, Avila D, Hughes M et al (1995) A new antigen receptor gene family that undergoes rearrangement and extensive somatic diversification in sharks. Nature 374:168–173 3. Zielonka S, Empting M, Grzeschik J et al (2015) Structural insights and biomedical potential of IgNAR scaffolds from sharks. MAbs 7:15–25 4. Kovaleva M, Ferguson L, Steven J et al (2014) Shark variable new antigen receptor biologics—a novel technology platform for

therapeutic drug development. Expert Opin Biol Ther 14:1527–1539 5. Stanfield RL, Dooley H, Verdino P et al (2007) Maturation of Shark Single-domain (IgNAR) antibodies: evidence for induced-fit binding. J Mol Biol 367:358–372 6. Dooley H, Flajnik MF (2006) Antibody repertoire development in cartilaginous fish. Dev Comp Immunol 30:43–56 7. Stanfield RL, Dooley H, Flajnik MF et al (2004) Crystal structure of a shark singledomain antibody V region in complex with lysozyme. Science 305:1770–1773 8. Goodchild SA, Dooley H, Schoepp RJ et al (2011) Isolation and characterisation of Ebolavirus-specific recombinant antibody

208

Doreen Ko¨nning et al.

fragments from murine and shark immune libraries. Mol Immunol 48:2027–2037 9. Walsh R, Nuttall S, Revill P et al (2011) Targeting the hepatitis B virus precore antigen with a novel IgNAR single variable domain intrabody. Virology 411:132–141 10. Liu JL, Anderson GP, Delehanty JB et al (2007) Selection of cholera toxin specific IgNAR single-domain antibodies from a naı¨ve shark library. Mol Immunol 44:1775–1783 11. Liu JL, Anderson GP, Goldman ER (2007) Isolation of anti-toxin single domain antibodies from a semi-synthetic spiny dogfish shark display library. BMC Biotechnol 7:78 12. Ubah OC, Steven J, Kovaleva M et al (2017) Novel, Anti-hTNF-α variable new antigen receptor formats with enhanced neutralizing potency and multifunctionality, generated for therapeutic development. Front Immunol 8:1780 13. Kovaleva M, Johnson K, Steven J et al (2017) Therapeutic potential of shark Anti-ICOSL VNAR domains is exemplified in a murine model of autoimmune non-infectious uveitis. Front Immunol 8:1121 14. Zielonka S, Weber N, Becker S et al (2014) Shark attack: high affinity binding proteins derived from shark vNAR domains by stepwise in vitro affinity maturation. J Biotechnol 191:236–245 15. Zielonka S, Empting M, Ko¨nning D et al (2015) The shark strikes twice: hypervariable loop 2 of shark IgNAR antibody variable domains and its potential to function as an autonomous paratope. Mar Biotechnol (NY) 17:386–392 16. Camacho-Villegas T, Mata-Gonza´lez M, Garcı´a-Ubbelohd W et al (2018) Intraocular penetration of a vNAR: in vivo and in vitro VEGF165 neutralization. Mar Drugs 16:113 17. Ko¨nning D, Zielonka S, Sellmann C et al (2016) Isolation of a pH-sensitive IgNAR variable domain from a yeast-displayed, histidinedoped master library. Mar Biotechnol (NY) 18:161–167 18. Ko¨nning D, Hinz SC, Grzeschik J et al (2018) Construction of histidine-enriched shark IgNAR variable domain antibody libraries for the isolation of pH-sensitive vNAR fragments. In: Hust M, Lin T (eds) Phage display. methods in molecular biology. Humana Press, New York, NY, pp 109–127 19. Matz H, Dooley H (2019) Shark IgNARderived binding domains as potential diagnostic and therapeutic agents. Dev Comp Immunol 90:100–107

20. Ko¨nning D, Rhiel L, Empting M et al (2017) Semi-synthetic vNAR libraries screened against therapeutic antibodies primarily deliver antiidiotypic binders. Sci Rep 7:1–13 21. Simmons DP, Streltsov VA, Dolezal O et al (2008) Shark IgNAR antibody mimotopes target a murine immunoglobulin through extended CDR3 loop structures. Proteins 71:119–130 22. Tornetta M, Fisher D, O’Neil K et al (2007) Isolation of human anti-idiotypic antibodies by phage display for clinical immune response assays. J Immunol Methods 328:34–44 23. Godar M, Morello V, Sadi A et al (2016) Dual anti-idiotypic purification of a novel, nativeformat biparatopic anti-MET antibody with improved in vitro and in vivo efficacy. Sci Rep 6:31621 24. Ladjemi MZ (2012) Anti-idiotypic antibodies as cancer vaccines: achievements and future improvements. Front Oncol 2:158 25. Alvarez-Rueda N, Ladjemi MZ, Be´har G et al (2009) A llama single domain anti-idiotypic antibody mimicking HER2 as a vaccine: Immunogenicity and efficacy. Vaccine 27:4826–4833 26. Sanches J de S, de Aguiar RB, Parise CB et al (2016) Anti-bevacizumab idiotype antibody vaccination is effective in inducing vascular endothelial growth factor-binding response, impairing tumor outgrowth. Cancer Sci 107:551–555 27. Hartmann C, Mu¨ller N, Blaukat A et al (2010) Peptide mimotopes recognized by antibodies cetuximab and matuzumab induce a functionally equivalent anti-EGFR immune response. Oncogene 29:4517–4527 28. Grzeschik J, Ko¨nning D, Hinz SC et al (2018) Generation of semi-synthetic shark ignar single-domain antibody libraries. In: Hust H, Lin T (eds) Phage display. Methods in molecular biology. Humana Press, New York, NY, pp 147–167 29. Dickgiesser S, Rasche N, Nasu D et al (2015) Self-assembled hybrid aptamer-Fc conjugates for targeted delivery: a modular chemoenzymatic approach. ACS Chem Biol 10:2158–2165 30. Van Deventer JA, Wittrup KD (2014) Yeast surface display for antibody isolation: library construction, library screening, and affinity maturation. In: Ossipow V, Fischer N (eds) Monoclonal antibodies. Methods in molecular biology (methods and protocols). Springer, Totowa, NJ, pp 151–181

Selection of Anti-Idiotypic Shark Antibody Domains 31. Chao G, Lau WL, Hackel BJ et al (2006) Isolating and engineering human antibodies using yeast surface display. Nat Protoc 1:755–768 32. Gera N, Hussain M, Rao BM (2013) Protein selection using yeast surface display. Methods 60:15–26 33. Pettersen EF, Goddard TD, Huang CC et al (2004) UCSF Chimera—a visualization system

209

for exploratory research and analysis. J Comput Chem 25:1605–1612 34. Ko¨nning D, Kolmar H (2018) Beyond antibody engineering: directed evolution of alternative binding scaffolds and enzymes using yeast surface display. Microb Cell Factories 17:32

Chapter 12 Simplifying the Detection of Surface Presentation Levels in Yeast Surface Display by Intracellular tGFP Expression Steffen C. Hinz, Adrian Elter, Julius Grzeschik, Jan Habermann, Bastian Becker, and Harald Kolmar Abstract Yeast surface display (YSD) is an ultra-high throughput method used in protein engineering. Proteinprotein interactions as well as surface presentation on the yeast cell surface are verified through fluorophoreconjugated labeling agents. In this chapter we describe an improved setup for full-length surface presentation detection. To this end, we used a single open reading frame (ORF) encoding for the protein to be displayed and a 2A sequence and tGFP for an intracellular fluorescence signal. The 2A sequence allows the simultaneous generation of two separate proteins from the same ORF through ribosomal skipping. The entangled expression of the POI on the yeast surface and intracellular tGFP obviates the need for fluorescent staining steps. Key words T2A, 2A peptide, Yeast surface display, GFP, Ribosomal skipping

1

Introduction Yeast surface display is a commonly used technology for ultra-high throughput screenings in protein engineering [1, 2]. It relies on the genotype-phenotype coupling of yeast cells with every yeast cell expressing one defined protein variant on the cell surface [3]. Successful usage of yeast surface display is reported for proteases, antibody fragments, and other proteins [4]. The surface presentation is usually detected by fluorescence staining of peptide tags (FLAG, HA, c-Myc) combined with detection of antigen binding. Since the staining of different peptide-epitopes requires orthogonal labeling agents, the staining can be time-consuming and requires the production or purchase of labeling agents. 2A peptides are virus-derived polypeptide sequences described for different virus strains. They are located between virus proteins

Steffen C. Hinz and Adrian Elter contributed equally to this work. Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2070, https://doi.org/10.1007/978-1-4939-9853-1_12, © Springer Science+Business Media, LLC, part of Springer Nature 2020

211

212

Steffen C. Hinz et al.

and enable the “cleavage” of these proteins during translation utilizing a unique “skipping” mechanism. This mechanism comprises the formation of a helical structure of the nascent polypeptide chain during the ribosomal translation and the nucleophilic attack of a water molecule resulting in an interruption of the polypeptide chain elongation without stopping the translation itself. The 18–22 amino acid-containing 2A peptides derived from different virus strains differ in their skipping efficiency while comprising a preserved C-terminal “GDVEXNPGP” motif [5]. Skipping occurs between the glycine and proline located at the C-terminus of the 2A peptide preventing the glycyl-prolyl peptide bond from forming [6, 7]. Chng et al. identified the 2A peptide derived from the thosea asigna virus (abbreviated T2A) to have nearly 100% cleaving efficiency in CHO cells [8]. In this chapter, we describe the coupling of the expression of a protein of interest (POI) with tGFP via utilization of the T2A peptide. According to the gene sequence (50 -POI-T2A-tGFP-30 ), translation of the T2A gene is necessary for tGFP expression. This allows to identify POI-presenting yeast cells through microscopy or flow cytometry. The bicistronic construct allows tGFP to accumulate intracellular, whereas the POI is presented on the yeast cell surface [9, 10]. This enables the circumvention of labeling for surface presentation, saving time and labeling agents (Fig. 2). In more detail, we describe the construction of a T2A-tGFPcontaining plasmid for yeast surface display which allows the detection of full-length expression and surface presentation through intracellular tGFP fluorescence. Herein, we elaborate on the generation of a single clone construct, but this cloning strategy can be adjusted for the generation of large YSD libraries (Fig. 2).

2

Materials All buffers and solutions were prepared with MilliQ H2O unless declared otherwise.

2.1 Single Clone Construction for Yeast Surface Display

1. BamHI-HF (New England Biolabs). 2. XhoI-HF (New England Biolabs). 3. 10 CutSmart buffer (New England Biolabs). 4. pCT plasmid [3]. 5. Yeast strain: EBY100. 6. YPD media: 20 g/L D(+)-glucose, 20 g/L tryptone, 10 g/L yeast extract.

Detection of Display Levels in YSD using T2A-tGFP

213

7. Buffer A: 1 M Tris–HCl, 2.5 M dithiothreitol (DTT). 8. Buffer E: 10 mM Tris–HCl, 270 mM sucrose 1 mM MgCl2. 9. Electroporator Gene Pulser Xcell™ (Bio-Rad, Dreieich, Germany). 10. 0.2 cm electroporation cuvettes (Bio-Rad). 11. Bacto™ Casamino acids (BD Biosciences, San Jose, USA). 12. SD-CAA: 8.6 g/L NaH2PO4 H2O, 5.4 g/L Na2HPO4, 20 g/ L D(+)-galactose, 6.7 g/L yeast nitrogen base without amino acids, 5 g/L Bacto™ casamino acids. 13. SD-CAA agar plates: 8.6 g/L NaH2PO4·H2O, 5.4 g/L Na2HPO4, 20 g/L D(+)-galactose, 6.7 g/L yeast nitrogen base without amino acids, 5 g/L Bacto™ casamino acids, 5–10 g/L agar–agar. 14. 9 cm Petri dishes. 2.2

Oligonucleotides

The utilized oligonucleotides are listed in Table 1.

Table 1 Oligonucleotides for the construction of the tGFP containing pCT-based plasmid and the randomization of a vNAR fragment Primer

Sequence

c-myc-T2A_up

CACTGTGACTGTGAAAGGATCCGAGCAAAAGCTTATTTCTGAAGAGGACTT GGAGGGCCGCGG

T2A_tGFP_lo

GGTAATACGACATTCAATTTCCATTGGGCC

tGFP_up

GGCCCAATGGAAATTGAATGTCGTATTACC

Term_lo

TTTGTTACATCTACACTGTTGTTATCAGATCTCGAGCTATTATTCTTCACC

CDR1rand_up ACCATCAATTGCGTCCTAAAA(XXX)5TTGGGTAGCACGTACTGGTATTTCAC AAAGAAG H3_Mat_lo

GCTGCCGCGGCCCTCGGATCCWTTCACAGTCASARKGGTSCCSCCNCC

FR1_up

ATGGCCGCACGGCTTGAACAAACACCGACAACGACAACAAAGGAGGCAGG CGAATCACTGACCATCAATTGCGTCCTAA

H3_gr2_up

GTGGTGGTGGTTCTGGTGGTGGTGGTTCTGCTAGCATGGCCGCACGGCT TGAACA

H3_Mat_2A_lo CCTCCACGTCGCCGCAGGTCAGCAGGCTGCCGCGGCCCTCGGATCCWTTC pCT_seq_up

TACGACGTTCCAGACTACGCTCTGCAGGCT

pCT_seq_lo

AGTTGGTAACGGAACGAAAAATAGAAA

214

3

Steffen C. Hinz et al.

Methods Herein we describe the construction and implementation of a T2A-tGFP gene string into a yeast surface display plasmid (pCT) and subsequent flow cytometry analysis.

3.1 Single Clone Construction: Insert

Starting with the T2A genestring and the already amplified tGFP gene, the first PCR is necessary to fuse the T2A with the fluorescent protein (tGFP in this case) which is later used as an indicator for full-length expression. The T2A gene is complemented with the cMyc-tag at 50 for yeast homologous recombination, whereas the tGFP is complemented 30 with the first 30 nucleotides of the terminator (Fig. 1). 1. Amplify the T2A gene using the primers c-myc_T2A_up and T2A_tGFP_lo to incorporate the 50 c-Myc-Tag and a tGFP compatible overhang for fusion PCR. The detailed PCR protocol is listed in Table 2 (PCR 1). 2. Utilize the primers tGFP_up and term_lo for amplification (Table 2, PCR 2) of the tGFP gene including the 30 overhang for homologous recombination and the 50 end for gene fusion with the T2A gene. 3. Prepare a PCR reaction mixture with the purified T2A gene amplicon and the purified tGFP amplicon in an equimolar ratio. After 10 cycles with an annealing temperature of 72  C and an elongation time of 30 s, add 1 μL of 1 μM c-mycT2A_up and 1 μL of 1 μM term_lo primers, and continue the PCR for 20 cycles with given properties. The incubation times and temperatures are listed in Table 2 (PCR 3). Verify the presence of a 792 bp fragment by agarose gel electrophoresis with a 1% (w/v) agarose gel. Purify the amplicons with a PCR purification kit and determine the concentration and the total amount of insert DNA.

3.2 Single Clone Construction: Backbone

1. Digest 5 μg of pCT plasmid with 5 U XhoI-HF® and 5 U BamHI-HF® overnight at 37  C. Perform the digestion in a reaction volume of 50 μL. Supplement the solution with 5 μL of CutSmart (10) buffer and complement nuclease-free water to a total volume of 50 μL. 2. Verify the digestion via agarose gel analysis with a 1% (w/v) agarose gel and purify the digested plasmid. Use a PCR cleanup kit (we use the Wizard® SV Gel and PCR Clean-Up System manufactured by Promega) to purify the plasmid according to the kit’s manual. Determine the plasmid concentration after the cleanup.

Detection of Display Levels in YSD using T2A-tGFP

215

Fig. 1 PCR strategy and plasmid preparation for the implementation of a T2A-tGFP-containing plasmid for yeast surface display. (a) The insert is generated in three consecutive PCRs. The c-Myc-T2A gene fragment is amplified with the 50 overhangs for homologous recombination in yeast. The tGFP gene is amplified in parallel with overhangs for homologous recombination at the 30 site. Both gene fragments are fused by fusion PCR through shared nucleotide sequences between the T2A and the tGFP gene and subsequent amplification with the primers c-myc_T2A_up and term_lo after the first 10 PCR cycles without additional primers. Asterisks indicate the time-displaced additions of both primers. (b) The plasmid for yeast surface display is restricted with BamHI and XhoI and subsequently purified for the gapping procedure. (c) The T2A-tGFP containing plasmid is assembled through homologous recombination in yeast. The necessary nucleotide overhangs are depicted black-striped at the end of the insert. (d) Fully assembled plasmid with the T2A and tGFP genes with the endonuclease restriction sites for further cloning experiments depicted in red with the respective endonuclease

3.3 Single Clone Construction: Transformation

1. It is necessary to start with an overnight culture of EBY100, preferably inoculated from an agar–agar plate ensuring the yeasts are not contaminated. The pre-culture is cultivated at 30  C and medium agitation (140–180 rpm) in YPD medium.

216

Steffen C. Hinz et al.

Table 2 Thermocycler protocol for PCRs performed with Q5® High-Fidelity DNA Polymerase Step

Temperature ( C)

Duration (s)

Initial denaturation

98

30 30 30

30 cycles

98 66 (1. PCR) 58 (2. PCR) 60 (3. PCR) 72

Final elongation

72

300

Hold

4

–

30/kb

2. The pre-culture is utilized to inoculate 50 mL of YPD under the same cultivation conditions as the pre-culture with a starting OD600 of 0.5. 3. The cultivation is continued until an OD600 of 1.3–1.5 which should take approx. 3 h. 4. Treat the yeast cells with 500 μL of Tris-DTT stock solution 10–15 min prior to harvest during the cultivation period (see Note 1). 5. Harvest yeast cells by centrifugation (3 min at 4  C and 3500  g). All the following steps are performed on ice unless declared otherwise. 6. Wash EBY100 cells thrice with ice-cold Buffer E (see Note 2). To this end, resuspend the cell pellet in 25 mL Buffer E and centrifuge the cell suspension afterwards at 4  C and 3500  g for 3 min and discard the supernatant. Repeat this step subsequently with 10 mL and 1 mL, respectively. 7. Resuspend the cell pellet in ice-cold 1 mL Buffer E and transfer 100 μL of the cell suspension into an ice-prechilled electroporation cuvette. The solution is supplemented with 0.5–1 μg of linearized plasmid DNA and threefold molar excess of the DNA insert and incubated for 1–2 min (see Note 3). 8. Perform the electroporation with the following conditions: 2.5 kV, 200 Ω, and 25 μF. Immediately after the transformation, mix the cells with 1 mL YPD and incubate at 30  C for 1 h. 9. Centrifuge the cells after regeneration at 10,000  g for 2 min, aspirate the supernatant, and wash the cells once with 1 mL of PBS and repeat the centrifugation.

Detection of Display Levels in YSD using T2A-tGFP

217

Table 3 Thermocycler protocol for colony PCRs performed with Taq Polymerase Step

Temperature ( C)

Duration (s)

Initial denaturation

98

30

30 cycles

98 55 68

30 30 60/kb

Final elongation

68

300

Hold

4

–

10. Resuspend the yeast cells in 100 μL PBS and plate the cell suspension on SD-CAA plates. Yeast colonies should be visible after 2–3 days of incubation at 30  C. 11. Pick 10 single clones and resuspend each single clone in 20 μL 20 mM NaOH, respectively. Incubate the cell suspension at 98  C for 15 min. Utilize 1 μL of the suspension as a colony PCR template. Perform the colony PCR with the pCT_seq_up and pCT_seq_lo primers and an annealing temperature of 55  C (PCR protocol summarized in Table 3). Analyze the PCR amplicons via a 1% (w/v) agarose gel. Amplicons with a length of ~1300 bp are considered to be correct. Sequence putative positive PCR to confirm the sequence prior to further utilization of the single clones (see Note 4). 3.4 Single Clone Flow Cytometry Analysis

1. Inoculate 2 mL SD-CAA with a yeast single clone comprising the plasmid with the correct sequence and cultivate it overnight at 30  C and 180 rpm. Supplement the cell suspension with 3 mL of SG-CAA medium and continue to cultivate the suspension for at least 8 h at 30  C and 180 rpm. 2. Determine the OD600 of the grown yeast cells (use SD-CAA or PBS as a blank) and harvest 1  107 cells by centrifugation (13,000  g, 1 min). 3. Aspirate the supernatant and resuspend the cell pellet in 20 μL PBSB (PBS buffer supplemented with 0.1 % BSA (w/v)) buffer supplemented with an anti-c-Myc-biotin antibody (diluted 1:30), and incubate the cells for 20 min on ice. As negative control, incubate 1  107 cells in 20 μL PBSB without the antic-Myc antibody in parallel. Use both aliquots in the subsequent staining steps.

218

Steffen C. Hinz et al.

Fig. 2 Schematic depiction of yeast surface display in combination with T2A and intracellular tGFP readout. (a) The different proteins are displayed in different colors according to the respective mRNA section. The T2A residues are located on the yeast cell surface at the C-terminus of the c-Myc-tag (highlighted in red). The C-terminal proline of the T2A protein is N-terminally fused to the intracellular tGFP (also highlighted in red). The ribosomal translation is depicted as green arrows. (b) Correlation between c-Myc-tag labeling and intracellular tGFP fluorescence as reported by flow cytometry. 50,000 yeast cells are displayed. The c-Myctag is labeled in a two-step procedure with an anti-c-Myc-biotin antibody and streptavidin-APC

4. Add 1 mL of ice-cold PBSB to the cell suspension, mix thoroughly, and centrifuge at 13,000  g for 1 min. Remove the supernatant. 5. Resuspend the cells in 20 μL PBSB in the presence of a 1:75 dilution of streptavidin APC and incubate the suspension for 15 min on ice. 6. Wash the cells with 1 mL of ice-cold PBSB and harvest the cells through centrifugation (13,000  g, 4  C). Aspirate the supernatant and resuspend the cells in 300–500 μL of ice-cold PBSB. Depict the tGFP fluorescence and the APC fluorescence in a two-dimensional plot in the cell sorter software. The yeast cells of the negative control should show two distinct populations differing in their tGFP fluorescence intensity; neither population should show APC fluorescence. The anti-c-Myc-biotin-labeled sample will have the entire tGFP-positive population shifted towards APC fluorescence indicating an entangled expression of surfacepresented POI and intracellular accumulation tGFP (see Note 5; Fig. 2B). If the yeast cells do not show sufficient surface presentation levels, increase the incubation time in galactose-containing medium according to your needs (see Note 6).

Detection of Display Levels in YSD using T2A-tGFP

4

219

Notes 1. It is recommended to make a new Tris–DTT solution prior to every transformation. 2. After autoclaving, the sorbitol containing solution tends to have a concentration gradient of sorbitol from top to bottom. Short shaking of the bottle solves this problem. 3. To estimate the number of yeast cells that do not carry the desired insert but the parental plasmid instead, at least one negative control has to be carried along (1 μg of backbone DNA, 0 μg of insert DNA). 4. Even though the labeling of surface presented tags like HAand c-Myc-tag located at the N-/C-terminus of your protein of interest will give you an impression how many full-length constructs are displayed on the yeast cell surface in comparison to truncated variants, particular in case of large YSD libraries, this is not sufficient to determine your library quality. Especially the degree of randomization is an important characteristic which can be determined by deep sequencing. In a quick-and-dirty approach, you can sequence yeast single clones (the more the better, we recommend at least 10 single clone sequences) of your initial library which will indicate if you have some degree of randomization in your POI. 5. Observing the tGFP fluorescence via FACS alone will not give you the possibility to determine if the T2A skipping was successful since the tGFP could also be displayed on the yeast cells. Utilizing a fluorescence microscope—or even better a confocal microscope—will help you to conclude if the tGFP is accumulated in the cytoplasm [10]. 6. In library generation approaches, we utilized the T2A-peptide gene as a homologous overhang for gap repair depleting the C-terminal c-Myc-Tag reducing the C-terminal peptide length. However, you can also generate the library with the c-Myc-Tag at the C-terminus. The utilized primers need to be adjusted accordingly and the plasmid restriction is performed with NheI and BamHI. For the construction of YSD libraries, the transformation protocol should be adjusted according to the protocol of Bernatuil et al. [11]. Libraries can be checked for truncated POIs on the yeast cell surface with HA-tag labeling as described elsewhere [10] (Table 4).

220

Steffen C. Hinz et al.

Table 4 Nucleotide sequences used in this study Construct

DNA sequence

Full-length construct

AGTTACTTCGCTGTTTTTCAATATTTTCTGTTATTGCTTCAG TTTTAGCACAGGAACTGACAACTATATGCGAGCAAATCCCCTCACCAAC TTTAGAATCGACGCCGTACTCTTTGTCAACGACTACTA TTTTGGCCAACGGGAAGGCAATGCAAGGAGTTTTTGAATATTACAAA TCAGTAACGTTTGTCAGTAATTGCGGTTCTCACCCCTCAACAAC TAGCAAAGGCAGCCCCATAAACACACAGTATGTTTTTAAGGACAATAGC TCGACGATTGAAGGTAGATACCCATACGACGTTCCAGACTACGCTC TGCAGGCTAGTGGTGTGGTGGTTCTGGTGGTGGTGGTTCTGGTGGTGG TGGTTCTGCTAGCATGGCCGCACGGC TTGAACAAACACCGACAACGACAACAAAGGAGGCAGGCGAATCAC TGACCATCAATTGCGTCCTAAAACCGGAATGGACTATCTTGGGTAGGACG TACTGGTATTTCACAAAGAAGGGCGCAACAAAGAAGGCGAGGTTATCAAC TGGCGGACGATACTCGGACACAAAGAATACGGCATCAAAGTCCCTTTCC TTGCGAATTAGTGACCTAAGAGTTGAAGACAGTGGTACATATCACTG TGAAGCGTTGATTTATAGCGATATGGGCATGATTATGTGGAAAA TTGAAGGGGGGGGGACCACTGTGACTGTGAAAGGATCCGAGCAAAAGC TTATTTCTGAAGAGGACTTGGAGGGCCGCGGCAGCCTGCTGACC TGCGGCGACGTGGAGGAAAACCCAGGCCCAATGGAAATTGAATGTCGTA TTACCGGCACCCTGAATGGTGTTGAATTTGAACTGGTTGGTGGTGG TGAAGGTACACCGGAACAGGGTCGTATGACCAATAAAA TGAAAAGCACCAAAGGTGCACTGACCTTTAGCCCGTATCTGCTGTCTCA TGTTATGGGCTATGGCTTTTATCATTTTGGCACCTATCCGAGCGGTTA TGAAAATCCGTTTCTGCATGCCATAATAATGGTGGCTATACCAA TACCCGCATTGAAAAATATGAAGATGGTGGTGTTCTGCATGTTAGC TTTAGCTATCGTTATGAAGCCGGTCGTGTGATTGGTGATTTTAAAGTTA TGGGCACCGGTTTTCCGGAAGATAGCGTGATTTTTACCGATAAAATTA TTCGCAGCAATGCCACCGTTGAACATCTGCACCCGATGGGTGATAATGA TCTGGATGGTAGCTTTACCCGTACCTTTAGCCTGCGTGATGGTGGTTA TTATAGCAGCGTTGTGGATAGCCATATGCATTTTAAAAGCGCCATTCA TCCGAGCATTCTGCAGAACGGTGGTCCGATGTTTGCATTTCGTCGTG TGGAAGAAGATCATAGCAATACCGAACTGGGCATTGTTGAATATCAGCA TGCCTTTAAAACACCGGATGCAGATGCCGGTGAAGAATAATAGCTCGAG

Aga2p

ATGCAGTTACTTCGCTGTTTTTCAATATTTTCTGTTATTGCTTCAG TTTTAGCACAGGAACTGACAACTATATGCGAGCAAATCCCCTCACCAAC TTTAGAATCGACGCCGTACTCTTTGTCAACGACTACTA TTTTGGCCAACGGGAAGGCAATGCAAGGAGTTTTTGAATATTACAAA TCAGTAACGTTTGTCAGTAATTGCGGTTCTCACCCCTCAACAAC TAGCAAAGGCAGCCCCATAAACACACAGTATGTTTTTAAGGACAATAGC TCGACGATTGAAGGTAGATACCCA

HA-tag

TACGACGTTCCAGACTACGCTCTGCAGGCT

Gly4Ser linker

AGTGGTGGTGGTGGTTCTGGTGGTGGTGGTTCTGGTGGTGGTGGTTC TGCTAGC

vNAR

ATGGCCGCACGGC TTGAACAAACACCGACAACGACAACAAAGGAGGCAGGCGAATCAC TGACCATCAATTGCGTCCTAAAACCGGAATGGACTATCTTGGGTAGGACG (continued)

Detection of Display Levels in YSD using T2A-tGFP

221

Table 4 (continued) Construct

DNA sequence TACTGGTATTTCACAAAGAAGGGCGCAACAAAGAAGGCGAGGTTATCAAC TGGCGGACGATACTCGGACACAAAGAATACGGCATCAAAGTCCCTTTCC TTGCGAATTAGTGACCTAAGAGTTGAAGACAGTGGTACATATCACTG TGAAGCGTTGATTTATAGCGATATGGGCATGATTATGTGGAAAA TTGAAGGGGGGGGGACCACTGTGACTGTGAAA

BamHI site

GGATCC

c-Myc-tag

GAGCAAAAGCTTATTTCTGAAGAGGACTTG

T2A

GAGGGCCGCGGCAGCCTGCTGACCTGCGGCGACG TGGAGGAAAACCCAGGCCCA

tGFP

ATGGAAATTGAATGTCGTATTACCGGCACCCTGAATGGTGTTGAA TTTGAACTGGTTGGTGGTGGTGAAGGTACACCGGAACAGGGTCGTA TGACCAATAAAATGAAAAGCACCAAAGGTGCACTGACCTTTAGCCCGTA TCTGCTGTCTCATGTTATGGGCTATGGCTTTTATCATTTTGGCACCTA TCCGAGCGGTTATGAAAATCCGTTTCTGCATGCCATTAATAATGGTGGC TATACCAATACCCGCATTGAAAAATATGAAGATGGTGGTGTTCTGCATG TTAGCTTTAGCTATCGTTATGAAGCCGGTCGTGTGATTGGTGA TTTTAAAGTTATGGGCACCGGTTTTCCGGAAGATAGCGTGA TTTTTACCGATAAAATTATTCGCAGCAATGCCACCGTTGAACATC TGCACCCGATGGGTGATAATGATCTGGATGGTAGCTTTACCCGTACC TTTAGCCTGCGTGATGGTGGTTATTATAGCAGCGTTGTGGATAGCCATA TGCATTTTAAAAGCGCCATTCATCCGAGCATTCTGCAGAACGGTGG TCCGATGTTTGCATTTCGTCGTGTGGAAGAAGATCATAGCAATACCGAAC TGGGCATTGTTGAATATCAGCATGCCTTTAAAACACCGGATGCAGA TGCCGGTGAAGAATAATAG

XhoI site

CTCGAG

References 1. Cherf GM, Cochran JR (2015) Applications of yeast surface display for protein engineering. Methods Mol Biol 1319:155–175 2. Angelini A, Chen TF, de Picciotto S et al (2015) Protein engineering and selection using yeast surface display. Methods Mol Biol 1319:3–36 3. Boder ET, Wittrup KD (1997) Yeast surface display for screening combinatorial polypeptide libraries. Nat Biotechnol 15:553–557. https://doi.org/10.1038/nbt0697-553 4. Ko¨nning D, Zielonka S, Grzeschik J et al (2017) Camelid and shark single domain antibodies: structural features and therapeutic potential. Curr Opin Struct Biol 45:10–16. https://doi.org/10.1016/j.sbi.2016.10.019

5. Liu Z, Chen O, Wall JBJ et al (2017) Systematic comparison of 2A peptides for cloning multi-genes in a polycistronic vector. Sci Rep 7:2193. https://doi.org/10.1038/s41598017-02460-2 6. Ryan MD, Mehrotra A, Gani D et al (2001) Analysis of the aphthovirus 2A/2B polyprotein ‘cleavage’ mechanism indicates not a proteolytic reaction, but a novel translational effect: a putative ribosomal ‘skip. J Gen Virol 82:1013–1025. https://doi.org/10.1099/ 0022-1317-82-5-1013 7. Ryan MD, Donnelly M, Lewis A et al (1999) A model for nonstoichiometric, Cotranslational protein scission in eukaryotic ribosomes. Bioorg Chem 27:55–79. https://doi.org/10. 1006/bioo.1998.1119

222

Steffen C. Hinz et al.

8. Chng J, Wang T, Nian R et al (2015) Cleavage efficient 2A peptides for high level monoclonal antibody expression in CHO cells. MAbs 7:403–412. https://doi.org/10.1080/ 19420862.2015.1008351 9. de Felipe P (2004) Skipping the co-expression problem: the new 2A “CHYSEL”technology. Genet Vaccines Ther 2:13. https://doi.org/ 10.1186/1479-0556-2-13 10. Grzeschik J, Hinz SC, Ko¨nning D et al (2017) A simplified procedure for antibody

engineering by yeast surface display: coupling display levels and target binding by ribosomal skipping. Biotechnol J 12:1600454. https:// doi.org/10.1002/biot.201600454 11. Benatuil L, Perez JM, Belk J, Hsieh C-M (2010) An improved yeast transformation method for the generation of very large human antibody libraries. Protein Eng Des Sel 23:155–159. https://doi.org/10.1093/pro tein/gzq002

Chapter 13 Methods for Construction of Yeast Display Libraries of Four-Domain T-Cell Receptors Fla´vio Sa´dio, Gerhard Stadlmayr, Katja Eibensteiner, Katharina Stadlbauer, Florian Ru¨ker, and Gordana Wozniak-Knopp Abstract Since two decades, yeast display methodology is a popular tool for discovery, stability improvement, and affinity maturation of diverse protein scaffolds, intended for antigen recognition. Yeast display is particularly well suited for the selection of heterodimeric proteins, such as antibodies and T-cell receptors (TCRs), as it allows rapid library creation via gap-repair-driven homologous recombination and subsequent construction of a combinatorial library after mating of yeast of opposite mating types. Certain properties of the TCR scaffold, such as its stability, inferior to most antibody fragments, require modifications of traditional antigen selection strategies. Their selection can be monitored and guided upon staining with the soluble versions of their original antigen, peptide-major histocompatibility complex (MHC), or clonotypic antibodies, whose binding is critically dependent on the TCR structural integrity. Overall, this chapter underlines the importance of the versatile yeast display technique for the diversification of the TCR scaffold for antigen recognition and optimization of its stability. Key words TCR, Yeast display, Library construction, Directed evolution, Heterodimer

1

Introduction

1.1 Structure and Function of TCRs

TCRs are heterodimeric transmembrane proteins mostly composed of α- or β-chain, each of which consists of a variable and constant domain. Variable domains are mostly concerned with antigen binding, and typically the cognate antigens are peptide stretches presented in the context of MHCI or MHCII molecules. When expressed on the cell surface, the C-terminal residues of the transmembrane region of the α-chain are in contact with CD3-δ/ε heterodimer [1], and the eight transmembrane units, TCR-α/β, CD3-δ/ε, and CD3-γ/ε and CD3-ζ/ζ, are organized into an evolutionary conserved 8-helix-bundle [2]. Transbilayer signaling is triggered upon antigen binding: the CD3 δ-, γ-, ϵ- and ζ-chains contain immunoreceptor tyrosine-based activation motifs (ITAMs), which are upon ligand recognition phosphorylated by

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2070, https://doi.org/10.1007/978-1-4939-9853-1_13, © Springer Science+Business Media, LLC, part of Springer Nature 2020

223

224

Fla´vio Sa´dio et al.

the Lck (leukocyte-specific tyrosine kinase) [3]. Lck is recruited to the membrane region containing TCR complexes by the co-receptor CD4 when ligated with MHC molecules. Phosphorylated CD3 ITAMs recruit the kinase Zap70 (zeta-activated protein 70 kDa), which then phosphorylates multiple tyrosine residues within a membrane-associated scaffolding protein LAT (linker for the activation of T cells) [4], which can interact with signaling effectors such as phospholipase C-γ (PLCγ). This enzyme hydrolyzes phosphatidylinositol bisphosphate (PIP2) to diacylglycerol and inositol trisphosphate, which recruits a number of downstream proteins to the membrane, among them PKCθ (protein kinase C-θ) and RasGRP (RAS guanyl nucleotide-releasing protein), which activates the small GTPase Ras, an activator of mitogen-activated protein kinase (MAPK) signaling pathway [5]. 1.2 Structure of Soluble TCR

For the purpose of soluble expression and display, TCRs have been reduced to the structural entity composed of the α- and β-chains (Fig. 1a). It has been shown that the variable domains of the TCR alone function as an antigen-recognition unit if they are connected with an amino acid linker [6, 7]. Further, a three-domain construct composed of variable α- and variable and constant domain of the β-chain were successfully expressed [8] and displayed on the surface of mammalian cells, linked to an intracellular CD3-ζ chain, and were shown to be competent of signaling upon their specific antigen recognition [9]. Nevertheless, it has been shown that the structural perturbations relatively remote from the antigen binding site can affect antigen binding [10, 11]. Therefore it would be attractive to display the TCR with variable and constant domains of α- and β-chains, which is the format similar to the one of their end applications [12]. When the sequences of four-domain constructs described here were transferred directly from yeast display format into mammalian expression vectors and transfected into suspension cells, cultivated at standard lab-scale conditions, yields of 40–100 mg/L of purified product were achieved without particular optimization. Such display and selection protocols are hence a promising first step towards a TCR integrative display and screening platform.

1.3 Yeast Display of TCR

TCRs as recognition units differ critically from antibodies in their target repertoire and the type of target interaction. These proteins have evolved to react with intracellular targets with low affinity. Although they share the basic immunoglobulin fold with antibodies [13], TCRs have long been assessed as “difficult to handle” proteins, due to their lower production levels in recombinant expression systems, tendency for aggregation and lower thermal and chemical stability [14]. Their clinical application has therefore been limited to cell-based therapies involving expression of surfaceexposed TCR α- and β-chains coupled to intracellular subunits of

Yeast-Displayed T-Cell Receptors

225

Fig. 1 (a) Cartoon diagram of TCR A6 (PDB: 4grm). Deep purple: variable domain of the α-chain. Magenta: constant domain of the α-chain. Teal: variable domain of the β-chain. Green: constant domain of the β-chain. The figure was prepared using Pymol (Schro¨dinger LLC). (b) Organization of 4-domain TCR displayed on the surface of a yeast cell (color legend as in (a)), with indicated binding site for pMHC

agonistic signaling molecules of the immune system, CD3-ζ, and costimulatory domains such as CD28, 4-1BB, CD27, or OX40 or their combinations in engineered T cells [15]. Only after an improvement in antigen affinity over several orders of magnitude has been achieved using affinity maturation of scTCRs (single-chain TCR) displayed on phage [16, 17], soluble TCRs could enter clinical trials and are since considered valuable therapeutic agents [12]. Their most common mode of action is the engagement of T-cells for cellular killing, which is mediated by the TCR’s fusion partner, a single-chain antibody with CD3-ε subunit targeting activity. Not less crucial for clinical development were the improvements of the TCR stability, which methodologically included rational design [18, 19], but also directed evolution of TCR libraries [20, 21]. Here the pioneering work on TCR scaffold stability was done with use of at the time little known yeast display technology to express the TCR as a single chain-linked α/β heterodimer, create libraries with error-prone mutagenesis, and then select for properly folded mutants by inducing yeast cells at an increased temperature and identifying desired clones by staining with a conformationdependent clonotypic anti-TCR antibodies. Thereby yeast display started to emerge as a valuable system for directed evolution of TCR molecules. Multiparameter analysis supported by FACS-based methods has allowed screening where several properties of the displayed molecule can be analyzed simultaneously. An efficient TCR display on the S. cerevisiae EBY100 surface was achieved over the well-established system employing the

226

Fla´vio Sa´dio et al.

linkage of Aga1p to Aga2p [22], fused with one of the chains of the TCR, and soluble secretion of the second chain (Fig. 1b). This however is not sufficient to explore the formation of functional heterodimer on yeast surface: to test for correct pairing, functional soluble pMHC (peptide-MHC) molecules [23] and their multimeric forms [24] proved invaluable reagents. This readout was important not only to determine the functionality of yeastdisplayed TCR but also for determining the specificity of binding to the cognate peptide in the early stages of discovery of derivate stabilized or affinity matured candidate molecules. Several later studies have confirmed that yeast display is optimally suited for optimization of TCR-derived clones for stability or alteration in antigen-recognition mode [10, 20, 25]. 1.3.1 Preferred Methods for Construction of TCR Yeast Display Libraries

Variable domains of a TCR can be expressed in scTCR-form efficiently over Aga1-Aga2 display system. The display on yeast surface becomes more of a challenge when the desired format consists of two polypeptide chains, each composed of variable and constant domains of heterodimeric TCR chains. Four-domain TCR constructs can be displayed efficiently when cloned into a bicistronic vector, similarly as described for Fab fragments [26] (Fig. 2a). The initial method of identification of assembled TCR on the yeast cell surface is the detection of tagged secreted chain. In the first experiments, it is highly recommended that the influence of the temperature and time of induction for a particular TCR scaffold is determined (Fig. 3a): the α-chain is the less efficient folding unit of the two and requires a folded β-chain as a folding chaperone, similarly as it was observed for the folding of the CH1 domain of antibodies on the CL domain [27]. Nevertheless, in yeast display system, the surface expression of anchored α- and β-chain expressed alone can be detected. There are significant differences in folding and stability between different TCRs, in our case between the anti-HTLV Tax A6 [28] and anti-Melan/MART DMF5 [29] (sequences in Table 1) (Fig. 3a), although a generically stabilizing de novo disulphide bond in both molecules is present to enhance the contact between the constant domains of α- and β-chain [18]. Further, the influence of natively occurring C-terminally positioned cysteine bond connecting α- and β-chain can be easily examined: in both chosen scaffold TCRs, it exerted a negative effect on their level of display on the yeast surface (Fig. 3a). Interestingly, its detrimental effect on the expression level in prokaryotic systems has been described before [30], and several studies sought for an alternative way to increase the affinity between the two chains of a heterodimer, such as introducing a non-native cysteine bond between the constant domains [18], connecting them into a single-chain fragment via a polypeptide linker [6, 9] or via a leucine zipper motive [31, 32].

Yeast-Displayed T-Cell Receptors

227

Fig. 2 Schematic summary of methods for construction of yeast-displayed TCR libraries using: (a) Bicistronic plasmid; (b) yeast mating; (c) plasmid encoding one chain of the TCR integrated into yeast genomic DNA and an episomal plasmid

Alternatively to bicistronic display, α- and β-chain can be transformed separately into haploid yeast of two opposite mating types, and the heterodimeric protein is then displayed on the surface of diploid cells [33, 34] (Fig. 2b). S. cerevisiae exhibits simple sexual differentiation into a-type cells, which produce “a-factor,” a mating pheromone which signals the presence of an a-cell to neighboring α-cells, and conversely α cells produce “α-factor” and respond to a-factor (reviewed in [35]). Yeast cells respond to the mating pheromone by growing a projection (named shmoo, due to its distinctive shape) towards its gradient. The response of haploid cells only to the mating pheromones of the opposite mating type allows mating only between the cells of the distinct mating type, and the fusion proceeds through the interaction of sexual molecules, such as agglutinins, of the opposite mating type [36]. Assuming the soluble version of the α-chain is cloned under a leader peptide allowing soluble secretion from an α-mating type yeast such as BJ5464, and the β-chain as an Aga2p-linked construct into a a-mating type yeast such as EBY100, the display of a four-domain heterodimeric TCR can be achieved by mating of the yeast cells of opposite mating types and selecting the diploid cells on doublenegative nutrient medium (-Leu/-Trp) (Fig. 4). Although the resulting construct is identical when displayed on the surface of yeast cell, the second strategy has important advantages in library construction, as described in Subheading 3.1.2. Importantly, mating allows the construction of significantly larger combinatorial libraries than can be achieved with a single transformation step. On the other hand, a drawback of the method is that the mating procedure requires additional experimental time of at least 3 days after the haploid libraries have been constructed. As a third strategy, integration in the yeast genome of one of the chains of the heterodimer can be attempted (Fig. 2c). This strategy is particularly attractive if library mutagenesis is intended

228

Fla´vio Sa´dio et al.

Fig. 3 (a) The effect on the percentage of displaying yeast cells of a particular TCR (A6 or DMF), native cysteine bond present in constructs labeled with “cys,” time and temperature of induction; (b) the percentage of TCRdisplaying yeast cells when bicistronic plasmid, mating, or combination of integrated and episomal plasmid has been used for expression of the two chains of the heterodimer

Yeast-Displayed T-Cell Receptors

229

Table 1 Amino acid sequences of described TCR receptors described and the cognate peptides A6 TCR α-Chain

QKEVEQNSGPLSVPEGAIASLNCTYSDRGSQSFFWYRQYSGKSPELIMSIYSNGDKEDGR FTAQLNKASQYVSLLIRDSQPSDSATYLCAVTTDSWGKLQFGAGTQVVVTPDIQNPDPAV YQLRDSKSSDKSVCLFTDFDSQTNVSQSKDSDVYITDKCVLDMRSMDFKSNSAVAWSNKS DFACANAFNNSIIPEDTFFPSPESSC

β-Chain

NAGVTQTPKFQVLKTGQSMTLQCAQDMNHEYMSWYRQDPGMGLRLIHYSVGAGITDQGEV PNGYNVSRSTTEDFPLRLLSAAPSQTSVYFCASRPGLMSAQPEQYFGPGTRLTVTEDLKN VFPPEVAVFEPSEAEISHTQKATLVCLATGFYPDHVELSWWVNGKEVHSGVCTDPQPLKE QPALNDSRYALSSRLRVSATFWQDPRNHFRCQVQFYGLSENDEWTQDRAKPVTQIVSAEA WGRADC

Peptide

LLFGYPVYV

DMF5 TCR α-Chain

QKEVEQNSGPLSVPEGAIASLNCTYSYRGSQSFFWYRQYSGKSPELIMFIYSNGDKEDGR FTAQLNKASQYVSLLIRDSQPSDSATYLCAVNFGGGKLIFGQGTELSVKPNIQNPDPAVY QLRDSKSSDKSVCLFTDFDSQTNVSQSKDSDVYITDKCVLDMRSMDFKSNSAVAWSNKSD FACANAFNNSIIPEDTFFPSPESSC

β-Chain

IAGITQAPTSQILAAGRRMTLRCTQDMRHNAMYWYRQDLGLGLRLIHYSNTAGTTGKGEV PDGYSVSRANTDDFPLTLASAVPSQTSVYFCASSWSFGTEAFFGQGTRLTVVEDLNKVFP PEVAVFEPSEAEISHTQKATLVCLATGFYPDHVELSWWVNGKEVHSGVCTDPQPLKEQPA LNDSRYALSSRLRVSATFWQDPRNHFRCQVQFYGLSENDEWTQDRAKPVTQIVSAEAWGR ADC

Peptide

ELAGIGILTV

Color code corresponds to Fig. 1a. Amino acid residues present in constructs with native cysteine bond are in blue for the α-chain and in dark green for the β-chain

to be applied only to a single chain of the heterodimer. The strain with the integrated plasmid can then act as a novel host strain to harbor the episomally encoded library of variant chains. Although the resulting cells phenotypically display the same TCR or analogously mutated library chains, they differ in the level of display of functional TCRs (Fig. 3b). The preferred method of library construction can be chosen according to the desired outcome of TCR mutagenesis and selection, taking into account the particular method of mutagenesis planned. 1.3.2 Gap Repair-Driven Homologous Recombination in TCR Library Construction

One of the properties that has much aided yeast display platform to gain popularity is the simple in vivo library construction in S. cerevisiae, which utilizes gap repair-driven homologous recombination. Here 25-bp homologous ends of the recombination fragment and the recipient vector are sealed into a circular construct in the yeast cell, and the correct recombination of up to 25 such fragments has been described. In this way, libraries in size of 108 members can be achieved by transforming 100 μg of gel-purified PCR fragment encoding the library of mutants and the corresponding quantity of linearized vector with homologous

230

Fla´vio Sa´dio et al.

Fig. 4 Schematic presentation of yeast mating protocol

ends. One potential drawback may be the self-recombination of the bicistronic vector, where promoter and terminator sequences are present multiple times to achieve equimolar expression of both chains of the heterodimer. Here the undesired recombination events occur in 50–60% of the transformants, which largely exceeds the 1% “dart” clones resulting typically from illegitimate recombination with monocistronic vector constructs. This effect cannot be

Yeast-Displayed T-Cell Receptors

231

overcome even if the PCR fragment is added to the transformation reaction in more than tenfold molar excess or if the promoter and TCR-chain-encoded fragment are added as two recombination fragments. On the contrary, the strategies for construction of libraries of TCR heterodimer chains in monocistronic vectors can easily be optimized to deliver nearly 100% correctly assembled clones. 1.3.3 Antigen Selections

After successful library construction, the members of a TCR yeast display library can enter a selection for improved antigen binding, where variants are selected that stain with the antigen at lower concentrations than the threshold for detection of the wild-type TCR. To select for TCRs with improved folding properties and stability, yeast cells can be either induced at a higher temperature (e.g., 37
C), or induced yeast cultures can be incubated at a temperature that exceeds the melting temperature of the wildtype TCR and then selected for members that can still bind the cognate pMHC antigen or react with clonotypic conformationdependent antibodies. Strategies for selections of yeast-displayed TCRs with enhanced antigen affinity have been developed where antigen-presenting cells have been applied as a selection platform to isolate displaying yeast/target cell complexes with density centrifugation [37]. FACS screening allows the very advantageous multiparameter screening, where yeast cells are simultaneously stained with several ligands labeled with different fluorophores. This protocol enables simultaneous evaluation of TCR expression level, for example, with an anti-tag antibody binding to a tagged chain of the heterodimer, and antigen reactivity through the detection of fluorescently labeled streptavidin or neutravidin binding to the biotinylated cognate pMHC ligand.

1.3.4 Detection Agents for Labeling of Yeast Display TCR Libraries

The TCRs differ from antibodies critically in their specific target antigens, composed both of self (MHC) and non-self (peptide) components. The development of multimeric pMHC molecules has primarily been developed for detection of little frequent CD4 + antigen-specific cells [38], and TCR variants with 1000 improvement in affinity have been reported to evolve via yeast display technology [39]. The most common method for labeling of the antigen is biotinylation and the preferred labeling reagent used is a biotin moiety attached to a long linker, which minimizes the steric hindrance of the binding site. A practical method for labeling of pMHC is biotinylation in vivo, an efficient method applicable to the bacterially expressed and refolded MHC molecules equipped with an avi-tag [40]. The detection of antigen binding can be performed using neutravidin or streptavidin as detection reagents labeled with a strong fluorescent conjugate, such as PE, APC, or tandem conjugates, or an adequately labeled anti-biotin-antibody.

232

Fla´vio Sa´dio et al.

Monoclonal antibodies (mAb) directed against the clonotypic structure of the T-cell receptor (TCR) may be useful reagents in the study and therapy of T-cell-mediated diseases. To obtain such antibodies against human TCRs, mice are first immunized with intact human T cells, and then spleen cell populations are depleted of B cells reactive to CD3 and the constant region of the TCR by adsorption to TCR/CD3 complexes derived from an irrelevant T-cell clone and clonotype-specific B cells selected with TCR/CD3 complexes from the T-cell clone of interest [41]. Binding of such antibodies can be conformationally dependent, e.g., can require an oxidized intrachain disulphide bond within the TCR variable domains for recognition [42, 43]. As surface-displayed proteins are transported through the secretory pathway, which retains and degrades misfolded proteins, TCR mutants with enhanced stabilities are expected to be displayed on a yeast cell at a higher density, are less likely degraded by extracellular proteases, and can hence be identified with conformational-dependent binding clonotypic antibody-based probes [20].

2 2.1

Materials Reagents

1. EZ-Link™ Sulfo-NHS-LC-LC-Biotin (21338, Thermo Fisher Scientific). 2. Anti-mouse Fc (Fab)2 fragment-FITC conjugate (F-2653, Sigma-Aldrich). 3. Anti-biotin antibody-APC conjugate (130-090-856, MACS Miltenyi). 4. Anti-pentahis antibody Alexa Fluor-488 conjugate (35310, QIAgen). 5. Neutravidin-PE (A2660, Thermo Fisher Scientific). 6. Soluble pMHC (Immunitrac A/S). 7. Streptavidin-Alexa Fluor-647 conjugate (S21374, Thermo Fisher Scientific). 8. Restriction enzymes. 9. T4 ligase and 10 ligase buffer (New England Biolabs).

2.2 Solutions and Buffers

1. 50% PEG3350. 2. 1 M Li-acetate solution. 3. 2 mg/mL salmon sperm DNA (ssDNA) in TE buffer (see Note 1). 4. Buffer for yeast media: 1 M potassium phosphate buffer, pH 6.0.

Yeast-Displayed T-Cell Receptors

233

5. Buffer for yeast mating: 0.5 M citrate buffer, pH 4.5. 6. 100 leucine solution: 10 g/L leucine. 7. 100 tryptophan solution: 8 g/L tryptophan. 8. Phosphate-buffered saline (PBS): 137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 2 mM KH2PO4. 9. Blocking buffer: 5% bovine serum albumin (BSA) in PBS. 10. MACS wash buffer: 0.25% BSA in PBS with 2 mM EDTA, pH 7.2. 11. FACS staining solution: 2% BSA in PBS. 12. Freezing buffer: 30% glycerol in PBS. 13. TE buffer: 1 M Tris–HCl, 0.1 M EDTA, pH 8.0. 2.3

Media

1. YPD: 20 g/L peptone, 10 g/L yeast extract and 2% glucose. 2. YPD, pH 4.5: YPD with 0.05 M citrate buffer, pH 4.5. 3. SD-CAA: 1% CAA solution with 0.67% YNB with ammonium sulphate, 100 mM KH2PO4/K2HPO4 buffer, pH 6.0, 2% glucose. 4. SG/R-CAA: 1% CAA solution with 0.67% YNB with ammonium sulphate, 100 mM KH2PO4/K2HPO4 buffer, pH 6.0, 2% galactose, 1% raffinose (see Note 2). 5. SD-Leu/-Trp media: 1 Drop-out Supplements solution (DO-Leu/-Trp) (see Note 3), 0.67% YNB with ammonium sulphate, 100 mM KH2PO4/K2HPO4 buffer, pH 6.0, 2% glucose. 6. SG/R-Leu/-Trp media: 1 Drop-out Supplements solution (DO-Leu/-Trp), 0.67% YNB with ammonium sulphate, 100 mM KH2PO4/K2HPO4 buffer, pH 6.0, 2% galactose, 1% raffinose. 7. YPD solid media: 1.5% agar with 20 g/L peptone, 10 g/L yeast extract, and 2% glucose. 8. MDL solid media: 1.5% agar with 0.67% YNB with ammonium sulphate, 100 mM KH2PO4/K2HPO4 buffer, 2% glucose, and 0.1 mg/mL leucine. 9. SD-Leu/-Trp solid media: 1.5% agar with 1 Drop-out Supplements solution (DO-Leu/-Trp), 0.67% YNB with ammonium sulphate, 100 mM KH2PO4/K2HPO4 buffer, 2% glucose. 10. Yeast Mating Agar, pH 4.5: 1.5% agar with 2% peptone, 1% yeast extract, 2% glucose, 0.05 M citrate buffer, pH 4.5. 11. LB medium: 1.5% agar with 10 g/L peptone, 5 g/L yeast extract, and 10 g/L NaCl.

234

Fla´vio Sa´dio et al.

12. 100 pen-strep solution: 100,000 U/mL penicillin and 10 mg/mL streptomycin. 13. 1000 ampicillin solution: 100 mg/mL ampicillin solution. 14. 1000 kanamycin solution: 50 mg/mL kanamycin solution. 2.4

Kits

1. Gel and DNA purification kit. 2. Plasmid isolation kit (for midi-preparation). 3. Zymoprep II kit (D2004, Zymo Research). 4. μMACS streptavidin kit (MACS Miltenyi). 5. MACS LS columns (MACS Miltenyi).

2.5

Equipment

1. 96-well microtiter plates. 2. 96-well compatible stacked micronic tubes (ScreenMate or similar). 3. 1.5 mL microcentrifuge tubes. 4. 1.0 mL kryotubes. 5. 15 and 50 mL conical tubes. 6. 6.125, 500, 1000, and 2000 mL Erlenmeyer flasks. 7. 90 mm Petri dishes. 8. 245  245 mm Petri dishes. 9. Pipettes from P10 to P1000 range with respective tips and multichannel pipettes. 10. Rotating wheel. 11. 80
C refrigerator. 12. Incubator with shaking platform that can be temperated to 20, 30, and 37
C. 13. Heat blocks or water baths. 14. SuperMACS magnetic bead separator. 15. High-speed sorter such as ARIA I (Becton-Dickinson). 16. FACS Analysis apparatus.

2.6

Yeast Strains

1. EBY100 (MATa GAL1-AGA1::URA3 ura3–52 trp1 leu2Δ1 his3Δ200 pep4::HIS2 prb1_1.6R can1 GAL) (ATCC® MYA4941™). 2. BJ5464 (MATα ura3–52 trp1 leu2Δ his3Δ200 pep4::HIS3 prb1-Δ1.6R can1 GAL) (ATCC® 208288™).

Yeast-Displayed T-Cell Receptors

3

235

Methods

3.1 TCR Yeast Display Library Construction 3.1.1 TCR Yeast Library Construction Using a Bicistronic Vector

In this display system, the sequences encoding α- or β-chain are cloned into a vector where their expression is initiated by the same promoter, and in our example also the same terminator sequences are used. Such vectors were described early in the development of yeast display methods for the display of Fab fragments [26]: in the described example, pYD1-based vector is modified with a series of mutagenesis steps to harbor two expression cassettes, and a novel multiple cloning site. One of the chains of the heterodimer is first cloned as a fusion of N-terminally positioned Aga2p-protein and is usually C-terminally tagged. The second chain of the heterodimer is expressed as a soluble protein and can for practical purposes also be C-terminally tagged, which allows its detection on the surface of a yeast cell (Fig. 5a). Although there are specific detection reagents available for detection of each of heterodimer chains of the TCR, such reporter tags can assist normalization during sorting procedure, minimizing the probability of steric hindrance, as well as the effect that a bound molecule could have on antigen binding of the TCR. Another useful feature of the designed construct is the restriction sites that allow a simple variable and constant domain exchange within each chain of the heterodimer (Fig. 5a). This may prove useful where the mutations introduced should be restricted to a certain region of the molecule. For example, in the directed evolution of TCR constant domains towards more stable variants, a greater spatial distance from variable domains might be of advantage, as the modification within the variable domains may affect antigen binding affinity, even when they are as far as 20 A˚ from the antigen binding amino acid residues [44]. 1. Perform a PCR with primers listed in Table 2 to amplify the TCR variable and/or constant regions using a high-fidelity Taq polymerase (see Note 4). 2. Gel-purify the PCR fragments. 70 μg DNA are required for a library of about 108 independent clones. This amount will be produced with about 70 PCR reactions with 100 μL volume. 3. Digest the recipient vector and purify the DNA. 100 μg DNA are required. 4. Start an overnight culture of S. cerevisiae EBY100 in 10 mL YPD, on an orbital shaker at 30
C. 5. Next day, determine the OD600 of the culture and dilute to 500 mL of OD600 of 0.4 in YPD medium. 6. Incubate the yeast culture on an orbital shaker at 30
C for about 5 h, until it reaches an OD600 between 2 and 3. 7. Distribute the culture to 50 mL conical tubes.

236

Fla´vio Sa´dio et al.

Fig. 5 Vector charts allowing (a) bicistronic expression of the TCR, (b) TCR library cloning with mating, (c) expression of a TCR chain from an integrative plasmid. Convenient restriction sites are indicated

ATCAGGCCAGCAAGGCCAAGGTTCTAACGCCGGCGTGACCCAGAC ATTAGCGGCCGCGTCGGCCCTGCCCCAGGCCTC

β-Chain A6Bsfi A6Bnot

ACGACCATGGCGAACGCCGGCGTGACCCAGAC CATCGGCGCGCCTCAGTCGGCCCTGCCCCAGGCCTCGGC

β-Chain A6Bnco A6Basc

TCCGTGCAGTTGGACAATATCAATGCCGTAATC GATTACGGCATTGATATTGTCCAACTGCACGGA

α-Chain A6Afor A6Arev

ATTGCTTCAGTGCTAGCCGCTGGGGCCATGGCGCAGAAGGAGGTGGAGCAGAAC GATTTTGTTACATCGGATCCGGCGCGCCTCAGGGGCTGGGGAAG

Primers for amplification of the library fragments for recombination as α-chain secreted, β-chain Aga2p-linked

Deletion of EcoRV site DelecorV DelecorVa

Primers for modifications of a bicistronic vector

ATCAGGCCAGCAAGGCCAAGGTTCTCAGAAGGAGGTGGAGC ATTGGCGGCCGCGGGGCTGGGGAAGAAGGTG

α-Chain A6Asfi A6Anot

β-Chain secreted, α-chain Aga2p linked

ACGACCATGGCGCAGAAGGAGGTGGAGC CATCGGCGCGCCTCAGGGGCTGGGGAAGAAGGTG

α-Chain A6Anco A6Aasc

α-Chain secreted, β-chain Aga2p-linked

Primers for amplification TCR sequences for cloning into a bicistronic vector

Primer sequence

Table 2 Primers for cloning of TCR sequences into a bicistronic vector and amplification of TCR domain libraries

(continued)

NcoI/AscI

NcoI/AscI

SfiI/NotI

SfiI/NotI

NcoI/AscI

Vector linearized

Yeast-Displayed T-Cell Receptors 237

Primer sequence

ATTGCTTCAGTGCTAGCCGCTGGGGCCATGGCGCAGAAGGAGGTGGAGCAGAAC GGGGTCGGGGTTCTGGATATCGGGGGTCACCACCACCTGGGTACCG

TGGTTCTGCTAGCATGACTGGTGGCCAGCAAGGCCAAGGTTCTAACGCCGGCGTGACCCAG AATCAATGGTGATGGTGATGATGTGCGGCCGCGTCGGCCCTGCCCCAGGC

CGGTACCCAGGTGGTGGTGACCCCCGATATCCAGAACCCCGACCCC GATTTTGTTACATCGGATCCGGCGCGCCTCAGGGGCTGGGGAAG

TGGTTCTGCTAGCATGACTGGTGGCCAGCAAGGCCAAGGTTCTAACGCCGGCGTGACCCAG CGGCCACCTCGGGGGGGAACACGTTCTTAAGGTCCTCGGTCACGGTCAG

GGCACCAGGCTGACCGTGACCGAGGACCTTAAGAACGTGTTCCCCCCCGAGG AATCAATGGTGATGGTGATGATGTGCGGCCGCGTCGGCCCTGCCCCAGGC

Restriction sites for vector linearization are indicated

A6BCfor A6Brev

Constant domain of the β-chain

A6Bfor A6BVrev

Variable domain of the β-chain

A6Bfor A6Brev

β-Chain

A6ACfor A6Arev

Constant domain of the α-chain

A6Afor A6AVrev

Variable domain of the α-chain

Table 2 (continued)

AflII/NotI

SfiI/AflII

SfiI/NotI

EcoRV/AscI

NcoI/EcoRV

Vector linearized

238 Fla´vio Sa´dio et al.

Yeast-Displayed T-Cell Receptors

239

8. Pellet the yeast by centrifugation at 1000  g, 5 min, at room temperature. 9. Wash pellets with 25 mL distilled water per tube. 10. Collect the cells by centrifugation at 1000  g, 5 min, at room temperature. 11. Resuspend each pellet with 3 mL 100 mM Li-acetate solution and let shake at 30
C for 15 min. 12. Collect the cells by centrifugation at 1000  g, 5 min, at room temperature. Remove the Li-acetate solution. 13. Resuspend the cells in the remains of the Li-acetate as the pellets are difficult to resuspend in PEG3350 solution. 14. Add to each tube 2.4 mL 50% PEG3350, 360 μL 1 M Li-acetate, 500 μL heat-shocked (5 min at 95
C) ssDNA (see Note 5), and 7 μg of insert and 10 μg of vector DNA in a volume that should not exceed 340 μL. Let shake at 30
C for 30 min. 15. Incubate at 42
C in a water bath for 45 min. Invert the tubes every 10 min. 16. Pellet the yeast and remove the transformation solution. 17. Add 5 mL YPD medium to each pellet and incubate with shaking at 30
C for 30 min. 18. Collect the cells by centrifugation at 1000  g, 5 min, at room temperature. Inoculate pooled yeast cells into 500 mL SD-CAA medium. 19. At this point, plate an aliquot of the yeast library in SD-CAA to MDL plates to determine the number of independent transformants in the library. Incubate at 30
C for at least 3 days. 20. Incubate the yeast library on an orbital shaker at 30
C for 24 h. 21. Passage 10–490 mL fresh SD-CAA medium and incubate the yeast library on an orbital shaker at 30
C for 24 h. 22. Collect yeast cells by centrifugation at 1000  g, 5 min, at 4
C, discard medium, and mix with an equal volume of freezing buffer. 23. Aliquot to 1 mL kryotubes. About 20 mL yeast suspension will be produced. To determine the number of viable cells, remove an aliquot of the culture and dilute in SD-CAA before plating to MDL plates in tenfold dilutions. A density of 109–1010 cells/mL is expected. 24. Store the library at 80
C (see Note 6). 3.1.2 TCR Yeast Display Library Construction Using Yeast Mating

In this protocol, the pool of α- or β-chains of the TCR heterodimer is transformed separately to the yeast of two opposite mating types. In this case we describe the construction of the library of α-chains in

240

Fla´vio Sa´dio et al.

a vector that guides their soluble expression and transforms S. cerevisiae BJ5464 to leucine prototrophy, and the transformants are selected in minimal medium supplemented with tryptophan. For the library of β-chains in a pYD1-based vector in S. cerevisiae EBY100, the selection proceeds in minimal medium supplemented with leucine (Fig. 5b). The two S. cerevisiae strains are of opposite mating types (α and a, respectively). After the mating procedure, the diploids are selected in double-negative medium (-Leu/-Trp). Optimization of this protocol is oriented towards three main factors: growth phase of the cells at the mating step and cell density as well as the pH of incubation of mating cells. 1. Perform a PCR to amplify the TCR variable and/or constant regions using a high-fidelity Taq polymerase. 2. For transformation of the libraries of the heterodimer chains, follow the steps 5–18 of Subheading 3.1.1. 3. The selection of each of the libraries proceeds in the SD-Leu/Trp medium supplemented with leucine for the yeast transformed to tryptophan prototrophy and supplemented with tryptophan for the yeast transformed to leucine prototrophy. Analogously, the enumeration plates are SD-Leu/-Trp solid medium supplemented with leucine or tryptophan. 4. Follow steps 21–25 as described in Subheading 3.1.1. The expected cell density is about 5  108 cells/mL. 5. For mating, inoculate 110 mL appropriately supplemented SD-Leu/-Trp medium with a representative number of cells (see Note 7) for each library. Increase the volume if the OD600 would exceed 2. Incubate on orbital shaker, at 30
C overnight. 6. The next day, determine the OD600. Dilute the culture to 500 mL with OD600 of 2.0. Let grow on an orbital shaker at 30
C until an OD600 of 4.0 is reached (about 3.5 h). 7. Determine the OD600. Collect 1400 OD600 units of each culture by centrifugation at 1000  g, 5 min at 20
C. 8. Resuspend the cells of each culture in 200 mL YPD, pH 4.5, and incubate on orbital shaker for 1 h at 30
C. 9. Collect the cells with centrifugation at 1000  g, 5 min at 20
C. 10. Resuspend the pellets in 5 mL YPD, pH 4.5. 11. Calculate the amount of cells of the opposite mating types required for mating: mix in total two aliquots of 700 OD600 units (2.5 mL) of the cells of each mating type to be spread onto two 245  245 mm agar plates (about 2 OD600 units/ cm2). 12. Spread the cells onto the mating agar (YPD, pH 4.5), and incubate 6 h at 30
C.

Yeast-Displayed T-Cell Receptors

241

13. Scrape the mated cells off the plates in 70 mL sterile PBS. 14. Collect the yeast cells by centrifugation at 1000  g, 5 min at 20
C, and rinse them three times with 12 mL sterile PBS at 1000  g, 5 min at 20
C. 15. Resuspend the pellet in 50 mL SD-Leu/-Trp medium. 16. Determine the size of the mated library and the number of viable cells of both types by plating dilutions to SD-Leu/-Trp medium and SD-Leu/-Trp medium supplemented with leucine or tryptophan. Incubate the plates at 30
C for 3 days. These figures will be required to calculate the mating efficiency. 17. Inoculate 5 mL of the mated yeast suspension into 10  500 mL SD-Leu/-Trp medium, distributed in 2000 mL flasks. Incubate on a shaking platform at 30
C for 45 h. 18. Passage the cells: Deliver 15 mL from every flask to 500 mL fresh SD-Leu/-Trp medium in a 2000 mL flask. Incubate on a shaking platform at 30
C for 24 h. 19. Harvest the cells at 1000  g, 10 min at 4
C, discard medium, and mix with an equal volume of freezing buffer. 20. Aliquot to 4.5 mL kryotubes. About 35 mL of yeast suspension will be produced. To determine the number of viable cells, remove an aliquot of the culture and dilute in tenfold steps in SD-Leu/-Trp medium before plating to SD-Leu/-Trp plates. About 3  108 cells/mL are expected. Incubate the plates at 30
C for 3 days. 21. Freeze the vials at 80
C. 22. After the 3 days incubation, count the colonies on the plates, and determine the number of haploid cells that participated in the mating procedure from the counts on leucine- and tryptophan-supplemented SD-Leu/-Trp agar plates. The number of diploid cells that can be determined by counting the colonies on SD-Leu/-Trp plates corresponds to the library size. The percent of mating efficiency is determined as the proportion of the number of diploids and the number of the less abundant haploid cells, multiplied by 100. 3.1.3 TCR Yeast Display Libraries Constructed with an Integrative Vector

If only one chain of the heterodimer will be targeted for the mutagenesis, an attractive option of a heterodimer library construction is to integrate the invariant chain in the yeast genome. A suitable vector for such manipulation is, for example, the vector pCM218 [45]. This vector contains a mutated tetR0 moiety under the control of cytomegalovirus promoter. The amplicon with the expression cassette containing the galactose-inducible GAL1,10promoter, a leader peptide sequence to guide soluble expression, and the sequence encoding the desired TCR chain is cloned into

242

Fla´vio Sa´dio et al.

the polylinker site of the integrative vector (Fig. 5c). The integration of the EcoRV-linearized plasmid proceeds at the chromosomal mutated LEU2 locus. 1. Amplify the expression cassette containing the GAL1,10-promotor sequence, leader peptide sequence, and the sequence of the TCR chain with a PCR using primers that allow cloning within the pCM218 polylinker region (e.g., HindIII and EcoRI). 2. Clone the amplicon into pCM218. 3. Linearize the vector with EcoRV: Digest 3 μg of the construct with 5 U EcoRV in 1 SmartCut buffer. Purify the DNA. 4. Prepare competent S. cerevisiae EBY100 and transform 1 μg of digested vector per 5 mL yeast culture as described in Subheading 3.1.1, steps 5–18. Plate the transformation reaction to SD-Leu/-Trp plates supplemented with tryptophan. Incubate at 30
C for 3 days. 5. A positive strain picked as a single clone from the selection plates can be transformed with a plasmid encoding a library of variants of the other chain of the TCR heterodimer, cloned into a vector assigning tryptophan prototrophy such as pYD1. Follow steps 5–18 described in Subheading 3.1.1. 6. Collect the cells by centrifugation at 1000  g, 5 min, at room temperature. Inoculate into 500 mL SD-Leu/-Trp medium. 7. At this point, plate the aliquots to SD-Leu/-Trp plates to determine the number of independent transformants in the library. Incubate at 30
C for at least 3 days. 8. Incubate the yeast library on an orbital shaker at 30
C for 24 h. 9. Passage 10–490 mL fresh SD-Leu/-Trp medium and incubate the yeast library on an orbital shaker at 30
C for 24 h. 10. Collect yeast cells with centrifugation at 1000  g, 10 min at 4
C, discard medium, and mix with an equal volume of freezing buffer. 11. Aliquot to 1 mL kryotubes. About 20 mL of yeast suspension will be produced. To determine the number of viable cells, remove an aliquot of the culture, and dilute in tenfold steps in SD-Leu/-Trp medium before plating to SD-Leu/-Trp plates. About 5  108 cells/mL are expected. 12. Store the library at 80
C. 3.2 Induction of TCR Expression in Yeast Cells

After library construction, the level of their correctness should be confirmed with sequencing a large number of clones. Plasmid DNA is best preserved when isolated with a plasmid isolation kit such as Zymoprep II kit. Sequencing can be then performed from E. coli colonies obtained after transformation and selection on the

Yeast-Displayed T-Cell Receptors

243

appropriate selective medium (LB supplemented with ampicillin or kanamycin). Staining of the induced yeast cultures and comparison with the level of staining displayed by the wild-type TCR is used to confirm the presence of each of the heterodimer chains as well as the formation of functional heterodimer, if the reactivity with the cognate pMHC can be anticipated after the library mutagenesis steps. 1. Inoculate the yeast cells resulting from a library transformation at an OD600 of 0.2 into SD-CAA or SD-Leu/-Trp medium supplemented with penicillin/streptomycin. For quality control purposes, a 10-mL culture can be used; if the library is intended for selection, an amount of yeast cells sufficient for coverage of library diversity should be inoculated (see Note 7). 2. Let grow at 30
C overnight with shaking until the cultures reach an OD600 of at least 5. 3. Collect the cells by centrifugation at 1000  g, 5 min at 20
C. 4. Remove the supernatant. 5. Resuspend the cells in SG/R-CAA or SG/R-Leu/-Trp medium supplemented with penicillin/streptomycin. 6. Incubate the yeast cultures with shaking at 20
C for 48 h. 7. If libraries are to be induced, use for induction an aliquot of the cells that is at least 20-fold the number of the output cells of the previous selection round. The remaining cells can be stored after resuspending in an equal volume of 30% glycerol at 80
C to serve as a backup. 3.3 Selection of TCRDisplaying Yeast Cells with MACS (MagneticActivated Cell Sorting)

MACS-based selection enables processing of a large number of yeast cells, which is usually necessary in the initial steps of antigen selection when the number of cells entering the selection procedure should be 20-fold number of independent library members to achieve proper representation. Oversampling is also recommended for all library pre-cleaning steps. Here we describe a protocol for the selection of pMHC-positive clones, which proved particularly useful for increasing the quality of TCR libraries based on bicistronic vector, which contained initially only about 50% correctly assembled clones as judged by DNA analysis (Fig. 6). 1. Pellet 2  109 induced yeast cells by centrifugation for 5 min at 2500  g, 20
C. 2. Wash by resuspending the pellet in 25 mL MACS wash buffer and centrifuge for 5 min at 2500  g, 20
C. 3. Resuspend the cells in 5 mL blocking buffer with the desired concentration of biotinylated pMHC antigen, and incubate for 1 h at room temperature with gentle agitation.

244

Fla´vio Sa´dio et al.

Fig. 6 Enrichment of pMHC-binding clones in a TCR-display library using MACS

4. Place the yeast cells in antigen solution on ice. Quench the reaction with adding tenfold volume of ice-cold MACS wash buffer. 5. Pellet the cells and antigen solution for 5 min at 2500  g and 4
C. 6. Wash twice with 20 mL MACS wash buffer and pellet after each washing step (see Note 8). 7. Resuspend the cells in 5 mL MACS wash buffer plus 25 μL streptavidin microbeads (μMACS streptavidin kit, MACS Miltenyi), and incubate on ice for 10 min. Mix by inversion every 2 min. 8. Add 15 mL of ice-cold MACS wash buffer. Keep the cell suspension on ice throughout the procedure. 9. Place a MACS LS column, chilled to 4
C, in the separator and apply 3 mL of MACS wash buffer to precondition the column. 10. Load the cell solution in 7 mL batches. Once the column stops dripping, remove it briefly from the magnet and place back in the magnet to reorient the beads in the column to allow trapped cells to flow through. Before loading the next 7 mL of cell suspension, apply 1 mL of MACS wash buffer to the column. Repeat until all cells were loaded. 11. Wash the column with 3 mL MACS wash buffer, briefly removed from the magnet and wash again with 3 mL MACS wash buffer. 12. To elute the binding cells, remove the column from the magnet and add 5 mL wash buffer. 13. Using the supplied plunger, push the cells from the column into a fresh collection tube. 14. Pellet the binding cell fraction for 5 min at 2500  g at 20
C.

Yeast-Displayed T-Cell Receptors

245

15. Resuspend the pellet in 10 mL SD-CAA or SD-Leu/-Trp medium. To estimate the number of eluted cells, dilute 10 μL of eluate in 990 μL SD-CAA or SD-Leu/-Trp, and plate aliquots of 10 and 100 μL were plated onto an MDL or SD-Leu/-Trp plate. Incubate the plates at 30
C for 3 days to estimate the output of MACS procedure. 16. Incubate the eluted cells at 30
C and 180 rpm overnight. 17. Freeze the cells at 80
C after mixing them with an equal volume of freezing medium or induce them for a further step of selection.

4

Notes 1. Salmon sperm DNA should be dissolved in TE buffer overnight at 4
C. As this is a viscous solution, it is considered sterile after being heated to 95
C for 5 min before yeast transformation. This can be repeated three times. 2. Galactose/raffinose solution can be prepared as a 10 concentrate (200 g/L galactose, 100 g/L raffinose) to be added to the medium. The sugars dissolve faster when placed in a water bath at 50
C. 3. 10 DO-Leu-Trp supplement can be sterilized by autoclaving at 121
C, but preferably with sterile filtration as the yeast tend to grow slower when using autoclaved solutions. Storage at 4
C for time periods over 2 months may result in precipitation and is not recommendable. 4. For a PCR protocol with Q5® Hi-Fidelity Polymerase (New England Biolabs), a PCR reaction containing 1 Q5 at Hi-Fidelity Polymerase Mastermix, 1 μM each oligonucleotide, and 100 ng/mL of the template DNA can be used. An amount of 1–2 μg of PCR product will be obtained using a PCR program with initial denaturation at 98
C for 3 min, followed by 30 cycles including 15 sec denaturation at 98
C, 15 s primer annealing at 55
C, and 15 s of extension at 72
C, and a final extension step at 72
C for 5 min. 5. ssDNA works as a carrier for DNA intended for transformation of yeast cells and increases the efficiency of transformation reaction. Single-strand-carrier DNA binds more effectively to the cells, reduces membrane binding to plasmid DNA, and enables a more efficient entry. At the same time, it is a substrate for cytosolic DNases and thus protects the plasmid DNA from degradation. 6. Repetitive freezing and thawing of the yeast kryostocks results in a decrease in the viability of yeast cells and should be avoided, especially in the case of libraries where the inoculum must be of the size representative for the number of independent clones.

246

Fla´vio Sa´dio et al.

7. The representative number of library clones relates to the fact that the frequency of single clones in a population follows the Poisson distribution. A total of 95% library coverage is assured if 20-fold number of its independent members, i.e., the 20-fold of the library size, participates in the selection. 8. If MACS is performed in the first steps of the selection of the naı¨ve library for antigen binding, where low affinities are expected, the number of wash steps can be reduced without an impact on the reproducibility of the selection outcome, to avoid the probability of losing weak binders of potential biological interest. For the same reason, it is recommended that all solutions applied after the antigen-incubation step are used ice-cold.

Acknowledgments This work was supported by the company F-star Biotechnology Ltd. and Christian Doppler Research Association (Christian Doppler Laboratory for Innovative Immunotherapeutics). The financial support by the Austrian Federal Ministry for Digital and Economic Affairs and the National Foundation for Research, Technology and Development is gratefully acknowledged. F.S. was also supported by the Ph.D. program BioToP (Biomolecular Technology of Proteins) funded by the Austrian Science Fund (FWF W1224). References 1. Shelton JG, Gu¨lland S, Nicolson K et al (2001) Importance of the T cell receptor α-chain transmembrane distal region for assembly with cognate subunits. Mol Immunol 38:259–265. https://doi.org/10.1016/S0161-5890(01) 00062-1 2. Chervin AS, Aggen DH, Raseman JM, Kranz DM (2008) Engineering higher affinity T cell receptors using a T cell display system. J Immunol Methods 339:175–184. https://doi.org/ 10.1016/j.jim.2008.09.016 3. Kane LP, Lin J, Weiss A (2000) Signal transduction by the TCR for antigen. Curr Opin Immunol 12:242–249 4. Sommers CL, Samelson LE, Love PE (2004) LAT: a T lymphocyte adapter protein that couples the antigen receptor to downstream signaling pathways. BioEssays 26:61–67 5. Huse M, Klein LO, Girvin AT et al (2007) Spatial and temporal dynamics of T cell receptor signaling with a photoactivatable agonist. Immunity 27:76–88. https://doi.org/10. 1016/j.immuni.2007.05.017

6. Novotny J, Ganju RK, Smiley ST et al (1991) A soluble, single-chain T-cell receptor fragment endowed with antigen-combining properties. Proc Natl Acad Sci U S A 88:8646–8650 7. Hoo WF, Lacy MJ, Denzin LK et al (1992) Characterization of a single-chain T-cell receptor expressed in Escherichia coli. Proc Natl Acad Sci U S A 89:4759–4763 8. Plaksin D, Polakova K, McPhie P, Margulies DH (1997) A three-domain T cell receptor is biologically active and specifically stains cell surface MHC/peptide complexes. J Immunol 158:2218–2227 9. Chung S, Wucherpfennig KW, Friedman SM et al (1994) Functional three-domain singlechain T-cell receptors. Proc Natl Acad Sci U S A 91:12654–12658 10. Richman SA, Aggen DH, Dossett ML et al (2009) Structural features of T cell receptor variable regions that enhance domain stability and enable expression as single-chain ValphaVbeta fragments. Mol Immunol 46:902–916.

Yeast-Displayed T-Cell Receptors https://doi.org/10.1016/j.molimm.2008. 09.021 11. Rangarajan S, He Y, Chen Y et al (2018) Peptide-MHC (pMHC) binding to a human antiviral T cell receptor induces long-range allosteric communication between pMHCand CD3-binding sites. J Biol Chem 293 (41):15991–16005. https://doi.org/10. 1074/jbc.RA118.003832 12. Bossi G, Buisson S, Oates J et al (2014) ImmTAC-redirected tumour cell killing induces and potentiates antigen crosspresentation by dendritic cells. Cancer Immunol Immunother 63:437–448. https://doi. org/10.1007/s00262-014-1525-z 13. Wilson IA, Christopher Garcia K (1997) T-cell receptor structure and TCR complexes. Curr Opin Struct Biol 7:839–848 14. Maynard J, Adams EJ, Krogsgaard M et al (2005) High-level bacterial secretion of single-chain αβ T-cell receptors. J Immunol Methods 306:51–67. https://doi.org/10. 1016/j.jim.2005.07.022 15. Mirzaei HR, Rodriguez A, Shepphird J et al (2017) Chimeric antigen receptors T cell therapy in solid tumor: challenges and clinical applications. Front Immunol 8. https://doi. org/10.3389/fimmu.2017.01850 16. Dunn SM, Rizkallah PJ, Baston E et al (2006) Directed evolution of human T cell receptor CDR2 residues by phage display dramatically enhances affinity for cognate peptide-MHC without increasing apparent cross-reactivity. Protein Sci 15:710–721. https://doi.org/10. 1110/ps.051936406 17. Li Y, Moysey R, Molloy PE et al (2005) Directed evolution of human T-cell receptors with picomolar affinities by phage display. Nat Biotechnol 23:349–354. https://doi.org/10. 1038/nbt1070 18. Boulter JM, Glick M, Todorov PT et al (2003) Stable, soluble T-cell receptor molecules for crystallization and therapeutics. Protein Eng 16:707–711 19. Lunde E, Loset GA, Bogen B, Sandlie I (2010) Stabilizing mutations increase secretion of functional soluble TCR-Ig fusion proteins. BMC Biotechnol 10:61. https://doi.org/10. 1186/1472-6750-10-61 20. Kieke MC, Shusta EV, Boder ET et al (1999) Selection of functional T cell receptor mutants from a yeast surface-display library. Proc Natl Acad Sci U S A 96:5651–5656 21. Shusta EV, Holler PD, Kieke MC et al (2000) Directed evolution of a stable scaffold for T-cell receptor engineering. Nat Biotechnol

247

18:754–759. https://doi.org/10.1038/ 77325 22. Boder ET, Wittrup KD (1997) Yeast surface display for screening combinatorial polypeptide libraries. Nat Biotechnol 15:553–557. https:// doi.org/10.1038/nbt0697-553 23. Garboczi DN, Hung DT, Wiley DC (1992) HLA-A2-peptide complexes: refolding and crystallization of molecules expressed in Escherichia coli and complexed with single antigenic peptides. Proc Natl Acad Sci U S A 89:3429–3433. https://doi.org/10.1073/ pnas.89.8.3429 24. Wooldridge L, Clement M, Lissina A et al (2010) MHC class I molecules with superenhanced CD8 binding properties bypass the requirement for cognate TCR recognition and nonspecifically activate CTLs. J Immunol 184:3357–3366. https://doi.org/10.4049/ jimmunol.0902398 25. Weber KS, Donermeyer DL, Allen PM, Kranz DM (2005) Class II-restricted T cell receptor engineered in vitro for higher affinity retains peptide specificity and function. Proc Natl Acad Sci U S A 102:19033–19038. https:// doi.org/10.1073/pnas.0507554102 26. Van Den Beucken T, Pieters H, Steukers M et al (2003) Affinity maturation of Fab antibody fragments by fluorescent-activated cell sorting of yeast-displayed libraries. FEBS Lett 546:288–294. https://doi.org/10.1016/ S0014-5793(03)00602-1 27. Feige MJ, Behnke J, Mittag T, Hendershot LM (2015) Dimerization-dependent folding underlies assembly control of the clonotypic alphabetaT cell receptor chains. J Biol Chem 290:26821–26831. https://doi.org/10. 1074/jbc.M115.689471 28. Cole DK, Sami M, Scott DR et al (2013) Increased peptide contacts govern high affinity binding of a modified TCR whilst maintaining a native pMHC docking mode. Front Immunol 4(168). https://doi.org/10.3389/fimmu. 2013.00168 29. Borbulevych OY, Santhanagopolan SM, Hossain M, Baker BM (2011) TCRs used in cancer gene therapy cross-react with MART-1/ Melan-A tumor antigens via distinct mechanisms. J Immunol 187:2453–2463. https://doi. org/10.4049/jimmunol.1101268 30. Pecorari F, Tissot AC, Plu¨ckthun A (1999) Folding, heterodimeric association and specific peptide recognition of a murine alphabeta T-cell receptor expressed in Escherichia coli. J Mol Biol 285:1831–1843. https://doi.org/ 10.1006/jmbi.1998.2422

248

Fla´vio Sa´dio et al.

31. Garcia K, Degano M, Stanfield R, Brunmark A (1996) An alpha-beta T cell receptor structure at 2.5 microns and its orientation in the TCRMHC complex. Science (80) 274:209–219 32. Chang HC, Bao Z, Yao Y et al (1994) A general method for facilitating heterodimeric pairing between two proteins: application to expression of alpha and beta T-cell receptor extracellular segments. Proc Natl Acad Sci U S A 91:11408–11412 33. Weaver-Feldhaus JM, Lou J, Coleman JR et al (2004) Yeast mating for combinatorial Fab library generation and surface display. FEBS Lett 564:24–34. https://doi.org/10.1016/ S0014-5793(04)00309-6 34. Baek DS, Kim YS (2014) Construction of a large synthetic human Fab antibody library on yeast cell surface by optimized yeast mating. J Microbiol Biotechnol 24:408–420. https:// doi.org/10.4014/jmb.1401.01002 35. Bardwell L (2004) A walk-through of the yeast mating pheromone response pathway. Peptides 25:1465–1476 36. Lipke PN, Kurjan J (1992) Sexual agglutination in budding yeasts: structure, function, and regulation of adhesion glycoproteins. Microbiol Rev 56:180–194 37. Richman SA, Healan SJ, Weber KS et al (2006) Development of a novel strategy for engineering high-affinity proteins by yeast display. Protein Eng Des Sel 19:255–264. https://doi. org/10.1093/protein/gzl008 38. Bill JR, Kotzin BL (2002) Use of soluble MHC class II/peptide multimers to detect antigen-

specific T cells in human disease. Arthritis Res 4:261–265 39. Holler PD, Chlewicki LK, Kranz DM (2003) TCRs with high affinity for foreign pMHC show self-reactivity. Nat Immunol 4:55–62 40. Cull MG, Schatz PJ (2000) Biotinylation of proteins in vivo and in vitro using small peptide tags. Methods Enzymol 326:430–440. https://doi.org/10.1016/S0076-6879(00) 26068-0 41. Steenbakkers PGA, Boots AMH, Rijnders AWM (1997) Generation and functional characterization of anti-clonotype antibodies to human T-cell receptors. J Immunol Methods 210(1):51–64 42. Brodnicki TC, Holman PO, Kranz DM (1996) Reactivity and epitope mapping of single-chain T cell receptors with monoclonal antibodies. Mol Immunol 33:253–263. https://doi.org/ 10.1016/0161-5890(95)00142-5 43. Manning TC, Schlueter CJ, Brodnicki TC et al (1998) Alanine scanning mutagenesis of an alphabeta T cell receptor: mapping the energy of antigen recognition. Immunity 8:413–425. S1074-7613(00)80547-6 [pii] 44. Sharma P, Kranz DM (2018) Subtle changes at the variable domain interface of the T-cell receptor can strongly increase affinity. J Biol Chem 293:1820–1834. https://doi.org/10. 1074/jbc.M117.814152 45. Bellı´ G, Garı´ E, Piedrafita L et al (1998) An activator/repressor dual system allows tight tetracycline-regulated gene expression in budding yeast. Nucleic Acids Res 26:942–947. https://doi.org/10.1093/nar/26.4.942

Chapter 14 Isolation of Tailor-Made Antibody Fragments from YeastDisplayed B-Cell Receptor Repertoires by Multiparameter Fluorescence-Activated Cell Sorting Anna Kaempffe, Sebastian J€ager, Doreen Ko¨nning, Harald Kolmar, and Christian Schro¨ter Abstract In the past decades, monoclonal antibodies have made an unprecedented transformation from research tools to a rapidly growing class of therapeutics. Advancements in the yeast surface display platform enable the selection of favorable mouse or human antibody variants from large B-cell receptor (BCR) gene repertoires that are derived from immunized normal or transgenic animals. Application of high-throughput fluorescence-activated cell sorting (FACS) screening along with well-chosen selection settings can be utilized to identify variants with desired affinities and predefined epitope binding properties. In the following chapter, we describe in detail a multiparameter screening protocol for the selection of antibody variants from yeast libraries generated from BCR gene repertoires from immunized transgenic rats. The procedure provides guidance for the selection of antigen-specific, high-affinity binding, and species crossreactive human antibodies with a broad epitope coverage. Essentially, this can accelerate target-specific antibody characterization as multiple desirable antibody features can be easily integrated into the selection procedure. In addition, we provide information on how to validate binding behavior of selected candidates after expression as soluble, full-length IgG molecules. Key words Fluorescence-activated cell sorting (FACS), Yeast surface display, Fab display, Species cross-reactivity, Epitope binning, Affinity, Specificity, B-cell receptor gene repertoires

1

Introduction Since the first approval of a therapeutic monoclonal antibody (mAb) in 1986, this class of therapeutics evolved rapidly to become the most successful product within the biopharmaceutical market [1]. Currently, more than 70 antibody-based molecules, e.g., IgGs, antibody-drug conjugates, and bispecific antibodies, have been approved and more than 700 are in clinical development [2].

Anna Kaempffe and Sebastian J€ager contributed equally to this work. Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2070, https://doi.org/10.1007/978-1-4939-9853-1_14, © Springer Science+Business Media, LLC, part of Springer Nature 2020

249

250

Anna Kaempffe et al.

Although the technologies for the generation of therapeutic antibodies showed rapid advancement in the last decades, e.g., by further refining the hybridoma technology [3] or the development of display technologies [4, 5], it is still challenging to obtain candidates with desired stability, affinity, and specificity. The hybridoma technology generates antibody-secreting immortalized cells upon fusion of myeloma cells with B cells and still represents the gold standard [6]. However, one imminent problem associated with this technology is related to the poor fusion efficiency of myeloma and B cells. This can result in a rather limited number of antibodysecreting cells and a significant decrease of antibody diversity [3, 5]. In contrast, display technology platforms like yeast surface display (YSD) in combination with high-throughput FACS enable the selection of advantageous variants from large yeast-displayed antibody repertoires [4]. Important advantages of YSD and FACS are the inherent control mechanisms of the advanced eukaryotic expression machinery, which enhances functional display of library variants, and FACS-based quantitative high-throughput analysis of large yeast libraries with up to 109 clones [7, 8]. Nowadays, there are several ways to generate yeast libraries, and antibody repertoires can be easily obtained from various sources such as synthetic gene repertoires or BCR gene repertoires from naı¨ve or immunized animals [9–12]. One beneficial attribute of BCR gene repertoires from immunized animals is that they typically comprise antibodies with high target specificity and binding affinity [9, 13]. If the BCR gene repertoires are obtained from transgenic animals bearing human immunoglobulin loci, laborious humanization approaches can be omitted [14, 15]. In addition, in vitro selection allows comprehensive sampling of BCR gene repertoires thereby enhancing the chance to identify and enrich infrequent antibody specificities [9, 16]. The diversities can be further enhanced by application of yeast mating which allows shuffling of heavy- and light- chain diversities, yielding novel chain pairings and, subsequently, antibodies with novel attributes [7]. The abovementioned features of YSD have been acknowledged with several successful FACS-based approaches, where protein- and hapten-targeting antibodies, e.g., against gp120 or ricin, were isolated from BCR gene repertoires [12, 13, 17]. Altogether, recent advancements show that YSD has become one of the most powerful tools for the rapid selection of hit candidates with defined characteristics from a variety of different combinatorial libraries [4, 8]. Our group has recently published a novel YSD multiparameter high-throughput screening procedure where applied stringencies finally translated into binding properties of isolated candidates [9]. This procedure enabled the efficient screening of antibody repertoires from immunized transgenic rats, which were challenged with a tumor-associated antigen, and resulted in the isolation of high-affinity, species cross-reactive mAbs, binding to distinct epitopes. Essentially, this procedure can

Multiparameter FACS-Sorting for Isolation of Tailor-Made Antibodies

251

Immunization transgenic rats

Yeast library generation

Antigen preparation

Sorting stringency

FACS Screening

1

Enrichment

2

Species crossreactivity

Selection parameter

3 Domain Specificity

IgG Expression of isolated single clones

Characterization

Epitope binning

Biolayer interferometry

Cellular binding

Fig. 1 General workflow for the generation and characterization of high-affinity, species cross-reactive, and domain-specific human antibodies. The Fab-displaying yeast library is generated from RNA of lymph node cells from immunized transgenic rats harboring human antibody variable regions. The immune library is subjected to the FACS screening process together with differentially labeled antigens. The FACS screening includes the selection parameters for the isolation of high-affinity, species cross-reactive, and domain-specific antibodies. The isolated clones are analyzed, sub-cloned, and expressed as full-length IgG molecules in order to verify whether they comprise the prescribed and desirable features

accelerate antibody discovery and target-specific mAb characterization, as multiple desirable antibody features can be easily integrated in the selection procedure. A general outline of the procedure to identify high-affinity, species cross-reactive mAbs against distinct epitopes is depicted in Fig. 1.

252

2

Anna Kaempffe et al.

Materials

2.1 Cells and Plasmids

1. S. cerevisiae strain EBY100 (trp-, leu-): MATa URA3–52 trp1 leu2Δ1 his3Δ200 pep4::HIS3 prb1Δ1.6R can1 GAL (pIU211: URA3) (Thermo Fisher Scientific). 2. S. cerevisiae strain BJ5464 (trp-, leu-): MATα URA3–52 trp1 leu2Δ1 his3Δ200 pep4::HIS3 prb1Δ1.6R can1 GAL (American Type Culture Collection, ATCC). 3. E. coli strain Top10: F-mcrA Δ(mrr-hsdRMS-mcrBC) Φ80lacZΔM15 ΔlacX74 recA1 araD139 Δ(ara-leu)7697 galU galK λ
rpsL(StrR) endA1 nupG (Thermo Fisher Scientific). 4. Vectors: Two modified pYD1 vectors (Thermo Fisher Scientific) which are under control of a galactose-inducible GAL1 promoter are used for cell surface display of heterodimeric antibody Fab fragments. The heavy chain (VH-CH1) is linked to the cell wall due to the genetic fusion to Aga2p (Xpress-tag(G4S)-VH-CH1-(G4S)-Aga2p). The light chain (VL-Cλ or VK-Cκ) is expressed in a soluble form due to the use of an aMFpp8 leader sequence for secretion [18]. To avoid loss of plasmids in yeast cells, both plasmids contain different auxotrophic markers, which allow cultivation of the yeast cells in tryptophan- or leucine-deficient media. In addition, plasmids carry resistance genes for ampicillin or kanamycin for the selection in E. coli. 5. Library: Detailed protocols for the generation of YSD libraries can be found elsewhere [10, 17, 19]. Nevertheless, we provide a brief outline of the generation of human Fab fragment yeast libraries in the following. The generation of human antibody repertoires against an antigen can be achieved by immunization of OmniRats® which carry human genes for variable regions of antibodies [9, 15, 20] (see Note 1). Animals showing a high serum titer for binding to the recombinantly produced target protein are euthanized, and secondary lymphoid tissues are isolated and subjected to total RNA extraction for further cDNA synthesis [9]. The cDNA is used to generate VH and VL gene libraries by amplifying the variable regions in a nested PCR reaction. VH and VL gene libraries are separately cloned into corresponding yeast shuttle vectors by gap repair cloning to generate two separate libraries in EBY100 and BJ5464 cells, respectively [19]. Mating of the haploid cells with opposite mating types is then applied to combine heavy- and lightchain repertoires [7, 17]. The generated human Fab fragment yeast library cells (Fig. 2) are cultured in Leu- and Trp-deficient media and serve as starting material for the flow cytometric screening (see Note 2).

Multiparameter FACS-Sorting for Isolation of Tailor-Made Antibodies

253

Antigen Alexa FluorTM 647 conjugate

Anti Kappa F(ab’)2 PE conjugate

Fab fragment

Aga2p S S

S S

Aga1p HC plasmid: Gal1

HC-Fab

Aga2p

Trp

AmpR

LC plasmid: Gal1

LC-Fab

Leu

KanR

yeast genome: Gal1

Aga1p

Fig. 2 Schematic illustration of an antibody Fab fragment-displaying Saccharomyces cerevisiae cell. The AntiHuman Kappa F(ab0 )2 PE conjugate binds to the constant region of the kappa light chain. The variable domains of light and heavy chain form the paratope that can bind the different antigens. Heavy and light chain of the Fab fragment are encoded on two shuttle plasmids which carry the LEU and TRP auxotrophic marker for yeast selection and the kanamycin and ampicillin resistance genes for selection in E. coli. The heavy-chain yeast plasmid encodes a fusion protein composed of the variable and CH1 domain of the heavy chain and the aagglutinin yeast adhesion receptor protein Aga2p. Aga2p assembles with the chromosomally encoded and β-glycan-bound Aga1p by disulfide bond formation, thereby anchoring the Fab fragment to the extracellular yeast cell wall. Aga1p, the light chain, and the Aga2p- heavy- chain fusion protein are controlled by galactoseinducible Gal1 promoters 2.2 Yeast Media and Reagents

1. SD -Trp/-Leu medium: Dissolve 26.7 g minimal SD-Base (Clontech) in overall 890 mL deionized H2O. Dissolve 8.56 g NaH2PO4  H2O, 5.4 g Na2HPO4 and 0.64 g Dropout-mix -Trp/-Leu (Clontech) in overall 100 mL deionized H2O. Sterilize by autoclaving. Cool below 50  C, combine both solutions, add 10 mL of Penicillin-Streptomycin

254

Anna Kaempffe et al.

(10,000 units/mL), and filtrate using a 0.22 μm bottle top filter to remove any particles. 2. SD -Trp/-Leu plates: Dissolve 26.7 g of minimal SD-Agar (Clontech) in overall 900 mL deionized H2O. Dissolve 8.56 g of NaH2PO4  H2O, 5.4 g of Na2HPO4 and 0.64 g Dropout-mix -Trp/-Leu in deionized H2O, and adjust the volume to 100 mL. Sterilize by autoclaving, combine both solutions and prepare plates. 3. SG -Trp/-Leu medium: Dissolve 37 g of minimal SD-Base +Gal/Raf (Clontech) in overall 490 mL deionized H2O. Dissolve 8.56 g of NaH2PO4  H2O, 5.4 g of Na2HPO4 and 0.64 g Dropout-mix -Trp/-Leu in final 100 mL deionized H2O. Dissolve 110 g of PEG 8000 in overall 400 mL deionized H2O. Sterilize all three solutions by autoclaving. Cool down solutions below 50  C, combine solutions, add 10 mL of Penicillin-Streptomycin (10,000 units/mL), and remove any particles by sterile filtration with a 0.22 μm bottle top filter. 4. Yeast library freezing solution: Dissolve 2 g glycerol and 0.67 g yeast nitrogen base (Difco) in 100 mL Dulbecco’s phosphatebuffered saline (DPBS) and sterilize by filtration with a 0.22 μm bottle top filter. 2.3 Reagents and Equipment for FACS

1. Goat F(ab0 )2 Anti-Human (SouthernBiotech).

Kappa-PE,

0.25

mg/mL

2. Full-length target proteins carrying a His-tag for detection (see Note 3). 3. Target protein fragments devoid of at least one protein domain compared to the full-length proteins (e.g., human antigenDomain-XY, human antigen-Domain-YZ). 4. Alexa Fluor™ 647 NHS Ester (Thermo Fisher Scientific). 5. EZ-Link™ Sulfo-NHS-Biotin (Thermo Fisher Scientific). 6. Pierce™ Biotin Quantitation Kit (Life technologies). 7. Streptavidin, Alexa Fluor™ 647 conjugate (SA-Alexa Fluor™ 647), 2 mg/mL (Invitrogen). 8. Streptavidin, R-phycoerythrin conjugate (SA-PE), 1 mg/mL (Invitrogen). 9. DPBS: 200 mg/L KCl, 200 mg/L KH2PO4, 8000 mg/L NaCl, 2160 mg/L Na2HPO4  7H2O. 10. Shaking incubator (20 and 30  C). 11. Microcentrifuge (at least 3000  g). 12. 0.22 μm Steriflip® and Steritop™ filtration units. 13. Pre-separation filters (70 μm). 14. Petri dishes.

Multiparameter FACS-Sorting for Isolation of Tailor-Made Antibodies

255

15. Cryogenic vials. 16. Water bath. 17. Fluorescence-activated cell sorter. 2.4 Reagents and Equipment for Epitope Binning Enzyme-Linked Immunosorbent Assay (ELISA)

1. PBS-BSA: 1% (w/v) BSA in DPBS. 2. PBS-T: 0.05% (v/v) Tween-20 in DPBS. 3. Target proteins with His-tag (human antigen). 4. Mouse Anti-His tag antibody, HRP-conjugated (Roche). 5. 1-Step™ Ultra TMB-Blotting Solution (Thermo Fisher Scientific). 6. H2SO4, 1 M. 7. Nunc MaxiSorp™ flat-bottom, 96-well (Thermo Fisher Scientific). 8. Plate Washer for 96-well plates. 9. Microplate Reader able to measure absorbance at 450 nm.

3

Methods For the following sections, all incubation steps with yeast cells should be performed on ice and after incubation with fluorescent proteins cells should be shielded from light. Yeast cell pellets are formed by centrifugation for 3 min at 3000  g and removal of the supernatant.

3.1 Antigen Preparation

3.2 Induction of Expression for the Display of Antibody Fab Fragments on Yeast Cells

Many different tags and fluorophores are commercially available for the labeling of antigens (see Note 4). We recommend using N-hydroxysuccinimide (NHS)-ester conjugation according to the manufacturer’s protocol. The fluorophores, e.g., Alexa Fluor™ 647- or biotin-linked NHS-ester, are conjugated N-terminally to accessible primary amines or lysine residues of the antigen. However, labeling of the antigen may affect its functionality. Therefore, the quality of the labeled antigen should be assessed by YSD (see Subheading 3.3.1) (see Note 5). Biotinylated antigens require a secondary labeling step with fluorophore-conjugated streptavidin to be detected by flow cytometry devices. 1. Thaw one aliquot of frozen library at 30  C for 1 min in a water bath. 2. Resuspend cell solution in SD -Trp/-Leu to a final optical density at 600 nm (OD600) of 0.1–0.5. The total number of cells in the prepared culture should be at least ten times higher than the calculated library diversity (OD600 ¼ 1 corresponds to 107 cells/mL).

256

Anna Kaempffe et al.

3. Incubate the culture overnight at 30  C and 130 rpm until mid-log up to stationary phase is reached (OD600 of approximately 4–7). 4. Harvest yeast cells by centrifugation. The number of harvested cells should also exceed the calculated library diversity by a factor of 10. 5. Resuspend cells in SG -Trp/-Leu medium to an OD600 of 1. 6. Incubate cells for 24–48 h at 20  C and 130 rpm for expression and display of antibody Fab fragments (see Note 6). 3.3 Labeling of Yeast Cells for FluorescenceActivated Cell Sorting

The following protocols refer to the labeling of 1  107 yeast cells. Reagent volumes can be upscaled proportionally to the required cell number. In general, sort samples and control samples should be processed in parallel to ensure comparability of obtained signals.

3.3.1 Determination of Saturating Antigen Concentrations

To evaluate the quality of labeled antigen and to determine appropriate antigen concentrations for screening, binding signals of different antigen concentrations on Fab fragment expressing yeast cells are analyzed by FACS. 1. Measure OD600 of induced yeast cells (library cells and/or positive control yeast cells displaying antigen-specific-antibody fragments, Subheading 3.2) (see Note 7). 2. Transfer three times 1  107 cells into a 1.5 mL tube each for testing three different antigen concentrations. 3. Pellet cells, resuspend cells in 500 μL DPBS, and pellet cells again. 4. Primary labeling: Resuspend pellet in 20 μL DPBS containing different concentrations of the biotinylated antigen (e.g., 250 nM, 500 nM, and 1 μM) (see Note 8). 5. Incubate cells for 30 min (see Note 9). 6. Pellet cells, resuspend cells in 500 μL DPBS, and pellet cells again. 7. Secondary labeling: Resuspend pellet in 20 μL DPBS containing 1:20 (v/v) diluted SA-PE or SA-Alexa Fluor™ 647. 8. Pellet cells, resuspend cells in 500 μL DPBS, and pellet cells again. 9. Resuspend cells in 100 μL DPBS and subject samples to FACS analysis. 10. The antigen binding signals can be quantified in the FACS histograms. Saturation of the antigen can be assumed if the relative fluorescence intensities of the antigen concentrations converge. If saturation is not reached with the concentrations mentioned above, a new experiment using higher antigen concentrations should be performed.

Multiparameter FACS-Sorting for Isolation of Tailor-Made Antibodies 3.3.2 Labeling Strategy

257

The following list provides details about the labeling setup for each sorting round. We want to emphasize that this is an exemplary labeling strategy which can and should be modified according to the specific purpose of the respective screening campaign. In the first two screening rounds, human antigen-specific clones are enriched. Cross-species reactivity screening is incorporated in round 3, whereas screening for domain specificity is included in round 4 (see Note 3, 10, and 11). In addition to the Sort Sample, different control samples to evaluate specific Fab display signals and antigen binding signals are prepared. Overview of Sort Samples to be prepared: 1. Round 1: Primary labeling with biotin-labeled human antigen (1.5 μM in DPBS). Secondary labeling with SA-Alexa Fluor™ 647 and Goat F(ab0 )2 Anti-Human Kappa-PE (both 1:20 (v/v) diluted in DPBS). 2. Round 2: Primary labeling with biotin-labeled human antigen (0.75 μM in DPBS). Secondary labeling with SA-Alexa Fluor™ 647 and Goat F(ab0 )2 Anti-Human Kappa-PE (both 1:20 (v/v) diluted in DPBS). 3. Round 3: Primary labeling with biotin-labeled human antigen (0.1 μM in PBS). Secondary labeling with SA-PE (1:20 (v/v) diluted in DPBS) and Alexa Fluor™ 647 labeled murine antigen (0.75 μM in DPBS) (Fig. 3).

Antigen species B (murine)

Antigen species A (human) X

Y

Z

X

hu-antigen

X

Y

hu-antigen-Domain-XY

Y

Z

mu-antigen

Y

Z

hu-antigen-Domain-YZ

Fig. 3 Schematic illustration of differentially labeled antigens used in this multiparameter screening procedure. The human and the murine antigen share a certain sequence and structural similarity (herein represented by the domains X, Y, and Z). Species cross-reactive antibodies would recognize such conserved regions in both antigens. In order to isolate domain-specific antibodies, two different proteins comprising either the human X and Y domain or the human Y and Z domain were recombinantly produced and labeled differently

258

Anna Kaempffe et al.

4. Round 4: Primary labeling with biotin-labeled human antigenDomain-XY (0.3 μM in DPBS) (Fig. 3). Secondary labeling with SA-PE (1:20 (v/v) diluted in DPBS) and Alexa Fluor™ 647 labeled human antigen-Domain-YZ (0.75 μM in DPBS). Preparing the Sort Samples

1. Measure the OD600 of the induced yeast cell culture and calculate the volume of cell suspension you need to transfer to a new vial for labeling (see Note 12). 2. Pellet cells, resuspend cells in 200 μL DPBS, and pellet cells again. Repeat this washing step. 3. Resuspend the cell pellet in 20 μL DPBS containing the primary labeling reagent of the corresponding sorting round and incubate for 30 min. 4. Pellet cells, resuspend cells in 200 μL DPBS, and pellet cells again. Repeat this washing step. 5. Resuspend the cell pellet in 20 μL DPBS containing the secondary labeling reagents of the corresponding sorting round, and incubate for 30 min. 6. Pellet cells, resuspend cells in 200 μL DPBS, and pellet cells again. Repeat this washing step. 7. Resuspend the cell pellet in 300 μL DPBS. Keep the cell suspension on ice in the absence of light until sorting.

Preparing the Fab Display Control Samples

Although Fab display labeling for sorting is only used in round 1 and 2, it is recommended to prepare a Fab Display Control Sample for all sorting rounds to ensure that cells display Fab fragments properly (see Note 7). 1. Measure the OD600 of the induced yeast cell culture, and transfer 1  107 cells to a new vial for labeling. 2. Pellet cells, resuspend cells in 200 μL DPBS, and pellet cells again. Repeat this washing step. 3. Resuspend the cell pellet in 20 μL DPBS containing Goat F (ab0 )2 Anti-Human Kappa-PE (1:20 (v/v) diluted) and incubate for 30 min. 4. Pellet cells, resuspend cells in 200 μL DPBS, and pellet cells again. Repeat this washing step. 5. Resuspend the cell pellet in 300 μL DPBS. Keep the cell suspension on ice in the absence of light until sorting.

Preparing the Antigen Binding Control Sample

The following procedure describes the preparation of the Antigen Binding Control Sample for round 1 and 2. For the preparation of Antigen Binding Control Samples in round 3 and 4, please adapt the protocol by using the corresponding antigen and secondary labeling reagent (see Note 7).

Multiparameter FACS-Sorting for Isolation of Tailor-Made Antibodies

259

1. Measure the OD600 of the induced yeast cell culture, and transfer 1  107 cells to a new vial for labeling. 2. Pellet cells, resuspend cells in 200 μL DPBS, and pellet cells again. Repeat this washing step. 3. Resuspend the cell pellet in 20 μL DPBS containing the biotinlabeled human antigen and incubate for 30 min. 4. Pellet cells, resuspend cells in 200 μL DPBS, and pellet cells again. Repeat this washing step. 5. Resuspend the cell pellet in 20 μL DPBS containing SA-Alexa Fluor™ 647 (1:20 (v/v) diluted) and incubate for 30 min. 6. Pellet cells, resuspend cells in 200 μL DPBS, and pellet cells again. Repeat this washing step. 7. Resuspend the cell pellet in 300 μL DPBS. Keep the cell suspension on ice in the absence of light until sorting. Once the labeling procedure is completed, cells should be directly subjected to sorting. 3.4 FluorescenceActivated Cell Sorting

In the sorting rounds 1 and 2, human antigen-specific Fab clones are enriched, and variants that do not bind the antigen are discarded. In round 3, human and murine species cross-reactive binders are selected, and in round 4, domain-specific binders are isolated (see Note 13). If using a MoFlo cell sorter device, the parameters in the Summit software can be set as follows (see Note 14): side scatter-LOG mode, 650; forward scatter-LIN mode, 570; FL8-LOG mode (Alexa Fluor™ 647), 600; FL2-LOG mode (PE), 400; trigger parameter, side scatter. Adjust the sample flow rate to an event rate of approximately 10,000–30,000 s
1. Other cell sorters require different but similar settings. 1. Just before FACS measurement, pass the prepared samples through pre-separation filters to remove yeast cell aggregates. 2. Measure the prepared control samples and the sort sample on a suitable FACS device. 3. Round 1 and 2: Define the negative gating borders by aligning them according to the Fab Display and Antigen Binding Control Samples. The sorting gate should include all events in the upper right quadrant to isolate all Fab-display and antigenbinding positive cells (Fig. 4) (see Note 15). Round 3: Draw a gate around the double positive clones. The number of cells in this gate depends on sort stringency. For example, see Fig. 4, where a stringent gating was applied to collect 0.73% of the cells positive for human- and murineantigen binding. Round 4: Select subdomain-specific cells by applying three different gates (R1, hu-antigen-Domain-Y-specific; R2,

260

Anna Kaempffe et al.

Round 2

Enrichment: Enrichment of binders for hu-antigen

Display

Round 1

hu-antigen binding

Species cross-reactivity: Selection of hu + mu species cross-reactive binders

Z Y X

hu-antigen binding

Round 3

mu-antigen binding

X

Y

Y

Z

Z

Y

Domain-specificity: Selection of domainspecific binders

X

hu-antigenDomain-XY binding

Round 4

hu-antigen-Domain-YZ binding

Fig. 4 Multiparameter FACS selection process for the isolation of high-affinity, species cross-reactive, and domain-specific binders from yeast surface immune libraries. In round 1 and 2, the selection process encompasses Fab surface display and antigen binding. Binders for the hu-antigen are enriched by applying relatively high antigen concentrations to avoid out-competition of low-affinity binders by high-affinity binders (round 1, 1.5 μM hu-antigen; round 2, 0.75 μM hu-antigen). Antigen concentrations are further reduced during subsequent rounds of sorting to enable the selection of high-affinity binding clones. Cells in the upper right gate are selected. In round 3, yeast cells are labeled with hu-antigen (0.1 μM) and mu-antigen (0.75 μM) to select binders that simultaneously recognize the human and murine antigen. Once the cross-species reactivity is included in the screening process, in round 4, yeast cells are co-incubated with hu-antigen-Domain-XY and hu-antigen-Domain-YZ proteins to simultaneously select human subdomain-specific clones by applying three different gates (R1, Domain-Y; R2, Domain-X; R3, Domain-Z)

hu-antigen-Domain-X-specific; R3, hu-antigen-Domain-Zspecific) (Fig. 4). 4. Start the sort. Ideally, the number of sorted cells should exceed the theoretical diversity by a factor of 10 (see Note 2).

Multiparameter FACS-Sorting for Isolation of Tailor-Made Antibodies

261

5. Transfer the isolated yeast cells to 25–100 mL SD -Trp/-Leu medium and regrow at 30  C and 130 rpm until OD600 is about 4–6. 6. For preparation of a successive round of sorting, induce expression of surface display according to Subheading 3.2. 7. Subsequently, label cells as described in Subheading 3.3.2. 8. The remaining cells can be used for the preparation of cryostocks. Therefore, harvest and wash the cells as described before, resuspend the cells in yeast library freezing solution considering a tenfold over-sampling, transfer them into cryogenic vials, and freeze the cells at
80  C. 3.5 Characterization of Isolated Clones 3.5.1 Sequencing of Plasmids from Yeast Cells

After sorting round 4, the complementarity-determining region (CDR) diversity can be analyzed by sequencing (see Note 16). 1. Plate serial dilutions (in DPBS) from regrown sorted cells onto selective SD -Trp/-Leu agar plates and incubate at 30  C for 48–72 h. 2. Pick single yeast clones and sequence the variable domains on light-chain and heavy-chain plasmids (see Note 17). 3. Select most abundant clones for sub-cloning and expression as full-length IgG antibodies as described elsewhere [21].

3.5.2 Epitope Binning ELISA

The epitope binning ELISA can be used to test if antibodies compete for antigen binding or if they bind to different epitopes. For this, antibodies should be expressed and purified as full-length IgG before all antibodies are tested against each other and against oneself (negative control). Carry out all incubation steps, except for the coating, at room temperature with 100 μL sample volume per well. Using the plate washer, each washing step is performed by three successive cycles where 300 μL/well PBS-T are applied, followed by removal of solution (see Note 18). 1. Coat a Nunc MaxiSorp™ flat-bottom 96-well plate with a 2.5 μg/mL DPBS antibody solution overnight at 4  C. 2. Wash plate. 3. Block plate with 300 μL/well PBS-BSA for 1 h. 4. In the meantime, mix 90 nM of every isolated antibody separately with 20 nM full length hu-antigen-His in PBS-BSA (see Note 19), vortex, and pre-incubate for 1 h. 5. Wash plate. 6. Incubate with pre-incubated antibody-antigen mixtures for 1 h.

262

Anna Kaempffe et al.

7. Wash plate and pre-warm TMB Ultra-Step solution to room temperature. Make sure the solution stays in the dark and is not exposed to light. 8. Incubate for 1 h with HRP-conjugated anti-His6 previously diluted 1:500 in PBS-BSA. 9. Wash plate. 10. Add TMB Ultra-Step solution and incubate for about 2 min. 11. Stop reaction by adding 1 M of aqueous H2SO4. 12. Measure the optical density at 450 nm using a microplate reader. High binding signals (OD450 > 0.5) are an indication for noncompetitive binding, i.e., the two antibodies bind different epitopes (Fig. 5). Weak binding signals (OD450 0.01–0.5) are measured if both antibodies compete for an epitope. The competition should be maximal when the same antibody was used for coating and for the applied antigenantibody mix (negative control). This should result in no binding (OD450 < 0.01) (see Note 20).

Pre-incubation

His6

His6

His6

substrate

Fig. 5 Schematic illustration of the epitope binning analysis in a pre-mix ELISA format. After antibodies (first analyte) are bound to the well surfaces, pre-incubated antibody- (second analyte) hu-antigen-His-samples are added, followed by the detection of bound hu-antigen-His. A high signal can be detected if the antibody bound to the surface (gray) and the soluble pre-incubated antibody (light blue) bind to different epitopes of the target (left side). If the pre-incubated antibody and the surface-bound antibody bind the same epitope (right side), both antibodies compete for binding and a weak or no signal can be detected

Multiparameter FACS-Sorting for Isolation of Tailor-Made Antibodies

4

263

Notes 1. Gene repertoires from immunized transgenic animals bear the advantage that they usually yield antibodies with high affinity and target specificity. Besides this, other gene repertoires of naı¨ve or synthetic libraries can be used. The advantage of these repertoires is that antibodies against all kinds of antigens, e.g., toxic or non-immunogenic substances, can be generated [22]. Yeast libraries derived from naı¨ve or synthetic gene repertoires were not yet tested with the multiparameter screening procedure. However, the screening procedure should be applicable to other library types, e.g., other sources of antibody repertoires. 2. If the maximum capacity of FACS is exceeded by the library diversity, it is recommended to use magnetic activated cell sorting (MACS) before the first flow cytometric sorting round of the yeast library. MACS is a simple and fast method to isolate a large number of cells [23]. 3. For cross-reactivity screenings, we exemplarily refer to human antigen and murine antigen. Antigen variants from distinct species can be screened as well. However, cross-reactivity cannot be observed with every ortholog antigen. The antigens of the distinct species should be homologous and should exhibit a similar conformation in the regions of interest. The maximum likelihood to isolate cross-reactive antibodies is given when linear epitopes are present. 4. Different combination of detection reagents and fluorophores can be used, but it should be considered that excitation and emission spectra of the fluorophores do not overlap (for guidance, see www.bdbiosciences.com). 5. If the ratio of biotin/protein is too high, it can reduce the portion of functional protein. Labeling of the antigen with 2–3 biotin molecules per protein molecule is sufficient for detection using streptavidin-fluorophore conjugates. 6. The ideal incubation time for the induction and generation of a good surface display signal should be determined in a single experiment using the starting library. 7. It is recommended to prepare additional controls for all FACS analysis and screening rounds to estimate unspecific binding of labeling reagents to wild-type Fab-displaying yeast cells. Additionally, the library cells can be labeled only with the streptavidin-fluorophore conjugates to assess unspecific binding of these reagents to yeast cells in the absence of biotinlabeled antigens.

264

Anna Kaempffe et al.

8. Instead of using biotinylated antigen, it is also possible to use an antigen that is directly labeled with a fluorophore. In this case, the second labeling step with fluorophore-labeled streptavidin is not required. 9. Within the chosen incubation time, an equilibrium should be reached (for details, see [24]). Nevertheless, usually 20–30 min of incubation is sufficient to gain adequate detection signals. 10. The antigen domain combinations that are used for the selection of different antibodies specific for different epitopes of the antigen should be selected carefully with consideration of accessibility of the domain in the complete antigen and conformational stability. 11. Different labeling reagents should be alternated in successive screening rounds to avoid the enrichment of antibody fragments that bind to labeling reagents. 12. Make sure that enough cells are labeled, as the number of analyzed cells during the sort should exceed the theoretical library diversity (round 1) or the isolated cell number of the previously sorted round by at least ten times to avoid losing unique clones. 13. Throughout subsequent sorting rounds, the concentration for antigen labeling can be reduced to enhance the selection stringency which favors selection of high-affinity, target binding variants. However, antigen concentrations should always allow sufficient signal detection during FACS. Details about the optimal antigen concentration for screening are given in [25]. 14. Fluorophores with overlapping emission spectra such as fluorescein (FITC) and phycoerythrin (PE) require proper compensation of the flow cytometer for adequate data interpretation. 15. During round 1 and 2, it is important to isolate a multitude of different binders to make sure that in round 3 most of the functional diversity (including lower affinity variants) can be screened for human- and murine-antigen cross-reactive clones. 16. Usually, the CDR diversity would drastically converge after various rounds of sorting, leaving only a few clusters of clones bearing CDR sequence homology >90% in each cluster. Typically, one clone of each cluster is selected for IgG full-length expression. 17. Single yeast colony sequencing service is available at Quintara Biosciences. Alternatively, the variable regions on heavy- and light-chain plasmids of each yeast clone can be amplified by yeast colony PCR after zymolyase treatment of yeast cells as

Multiparameter FACS-Sorting for Isolation of Tailor-Made Antibodies

265

described in [26] followed by sequencing of the amplification products (e.g., at Microsynth). 18. A plate washer is not necessarily required. Washing steps can also be performed manually. 19. The concentrations of the antibody (90 nM) and antigen (20 nM) are exemplary for antibodies with affinities expected in a single digit nanomolar range and should be adjusted accordingly. 20. The epitope binning ELISA can be performed in different settings. It can be conducted in a classical sandwich ELISA format, where the antigen and soluble antibody are added successively, or in a tandem format, where antigens are bound to the surface and both antibodies are added in parallel. Nevertheless, we recommend using the pre-mixed format. Additionally, epitope binning assays can be performed using bio-layer interferometry (BLI) or surface plasmon resonance (SPR). For advantages and disadvantages of these different possibilities, see ref. 27. References 1. Ecker DM, Jones SD, Levine HL (2015) The therapeutic monoclonal antibody market. MAbs 7:9–14 2. Strohl WR (2018) Current progress in innovative engineered antibodies. Protein Cell 9:86–120 3. Lo MMS, Tsong TY, Conrad MK et al (1984) Monoclonal antibody production by receptormediated electrically induced cell fusion. Nature 310:792 4. Doerner A, Rhiel L, Zielonka S et al (2014) Therapeutic antibody engineering by high efficiency cell screening. FEBS Lett 588:278–287 5. Bradbury ARM, Sidhu S, Du¨bel S et al (2011) Beyond natural antibodies: the power of in vitro display technologies. Nat Biotechnol 29:245–254 6. Ko¨hler G, Mistein C (1975) Continuous cultures of fused cells secreting antibody of predefined specificity. Nature 256:495–497 7. Blaise L, Wehnert A, Steukers MPG et al (2004) Construction and diversification of yeast cell surface displayed libraries by yeast mating: application to the affinity maturation of Fab antibody fragments. Gene 342:211–218 8. Gai SA, Wittrup KD (2007) Yeast surface display for protein engineering and characterization. Curr Opin Struct Biol 17:467–473 9. Schro¨ter C, Beck J, Krah S et al (2018) Selection of antibodies with tailored properties by application of high-throughput multiparameter fluorescence-activated cell sorting of yeast-

displayed immune libraries. Mol Biotechnol 60:727–735 10. Boder ET, Wittrup KD (1997) Yeast surface display for screening combinatorial polypeptide libraries. Nat Biotechnol 15:553–557 11. Benhar I (2007) Design of synthetic antibody libraries. Expert Opin Biol Ther 7:763–779 12. Bowley DR, Labrijn AF, Zwick MB et al (2007) Antigen selection from an HIV-1 immune antibody library displayed on yeast yields many novel antibodies compared to selection from the same library displayed on phage. Protein Eng Des Sel 20:81–90 13. Wang B, Lee CH, Johnson EL et al (2016) Discovery of high affinity anti-ricin antibodies by B cell receptor sequencing and by yeast display of combinatorial VH:VL libraries from immunized animals. MAbs 8:1035–1044 14. Bru¨ggemann M, Osborn MJ, Ma B et al (2015) Human antibody production in transgenic animals. Arch Immunol Ther Exp 63:101–108 15. Osborn MJ, Ma B, Avis S et al (2013) Highaffinity IgG antibodies develop naturally in Ig-knockout rats carrying germline human IgH/Igκ/Igλ loci bearing the rat CH region. J Immunol 190:1481–1490 16. Hoogenboom HR (2005) Selecting and screening recombinant antibody libraries. Nat Biotechnol 23:1105–1116 17. Weaver-Feldhaus JM, Lou J, Coleman JR et al (2004) Yeast mating for combinatorial Fab

266

Anna Kaempffe et al.

library generation and surface display. FEBS Lett 564:24–34 18. Rakestraw JA, Sazinsky SL, Piatesi A et al (2009) Directed evolution of a secretory leader for the improved expression of heterologous proteins and full-length antibodies in Saccharomyces cerevisiae. Biotechnol Bioeng 103:1192–1201 19. Benatuil L, Perez JM, Belk J et al (2010) An improved yeast transformation method for the generation of very large human antibody libraries. Protein Eng Des Sel 23:155–159 20. Pertmer TM, Eisenbraun MD, McCabe D et al (1995) Gene gun-based nucleic acid immunization: elicitation of humoral and cytotoxic T lymphocyte responses following epidermal delivery of nanogram quantities of DNA. Vaccine 13:1427–1430 21. Krah S, Schro¨ter C, Eller C et al (2017) Generation of human bispecific common light chain antibodies by combining animal immunization and yeast display. Protein Eng Des Sel 30:291–301 22. Hoogenboom HR (1997) Designing and optimizing library selection strategies for

generating high-affinity antibodies. Trends Biotechnol 15:62–70 23. Orcutt KD, Wittrup KD (2010) Yeast display and selection. In: Kontermann R, Du¨bel S (eds) Antibody engineering, vol 1. Springer, Heidelberg, pp 207–233 24. Gera N, Hussain M, Rao BM (2013) Protein selection using yeast surface display. Methods 60:15–26 25. Boder ET, Wittrup KD (1998) Optimal screening of surface-displayed polypeptide libraries. Biotechnol Prog 14:55–62 26. Fuxman Bass JI, Reece-Hoyes JS, Walhout AJM (2016) Zymolyase-treatment and polymerase chain reaction amplification from genomic and plasmid templates from yeast. Cold Spring Harb Protoc 2016. https://doi.org/ 10.1101/pdb.prot088971 27. Abdiche YN, Lindquist KC, Stone DM et al (2012) Label-free epitope binning assays of monoclonal antibodies enable the identification of antigen heterogeneity. J Immunol Methods 382:101–116

Chapter 15 Isolation of Anti-Hapten Antibodies by FluorescenceActivated Cell Sorting of Yeast-Displayed B-Cell Receptor Gene Repertoires Sebastian J€ager, Simon Krah, Doreen Ko¨nning, Janis Rosskopf, Stephan Dickgiesser, Nicolas Rasche, Harald Kolmar, Stefan Hecht, and Christian Schro¨ter Abstract Anti-hapten antibodies are widely used as specific immunochemical detection tools in a variety of clinical and environmental analyses. The sensitivity, however, is limited due to the resulting antibody affinities to the haptens which, in turn, leads to a high demand for specific affinity reagents. A well-established path for the generation of high-affinity antibodies is the immunization of animals with the target antigen. However, the generation of anti-hapten antibodies via immunization remains challenging as small molecule haptens usually possess low immunogenicity and, therefore, must be coupled to an immunogenic and high molecular weight carrier to provoke an immune response. Consequently, antibodies are primarily raised against the carrier molecule or structural features of the hapten-linker fused to the carrier protein. This turns the generation of antibodies which bind exclusively to the hapten structure into a search for the needle in a haystack. In the following chapter, we describe how yeast surface display and high-throughput fluorescence-activated cell sorting can be used to isolate antihapten antibodies from a large, yeast-displayed B-cell receptor gene library derived from immunized animals. For this, we describe in detail the preparation of protein-hapten conjugates, the immunization procedure, and the subsequent screening process. Moreover, we provide a simple flow cytometry protocol that allows for a rapid analysis of the enriched clones toward free hapten binding. Key words Anti-hapten antibodies, Carrier protein, Hapten, Protein-hapten conjugate, Immunization, Yeast surface display, Fab display, B-cell receptor gene repertoires, Fluorescence-activated cell sorting

1

Introduction The remarkable target specificity of monoclonal antibodies (mAbs) represents a major driving force for the clinical success of more than 70 mAbs approved by the FDA [1]. Their great specificity has not only been acknowleged in therapeutics but also in diagnostic and technical applications where specific and sensitive detection of an

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2070, https://doi.org/10.1007/978-1-4939-9853-1_15, © Springer Science+Business Media, LLC, part of Springer Nature 2020

267

268

€ger et al. Sebastian Ja

analyte is required (such as ELISA or immunohistochemistry [2]). The vast majority of diagnostic antibodies are generated against high molecular weight protein structures, which also holds true for their therapeutic counterparts. Nevertheless, it has been successfully shown that antibodies against small molecular weight substances (typically 3 for the 1:200 diluted sample when applying the settings of the serum titer ELISA described in Subheading 3.2.2. 13. When preparing the immunization-derived BCR gene yeast library for the first-round sort, the total number of cells should exceed the library diversity at least by the factor 10. 14. The length of incubation in SG medium depends on the time at which a maximum of surface display signal is reached. Induce cells, for example, for 24, 48, and 72 h and perform display labeling as described in Subheading 3.4.2, followed by FACS analysis to determine the time for maximum surface display signal. 15. For the first round of sorting, the appropriate antigen (human Fc-hapten conjugate) concentration should be determined in a separate titration experiment. Usually, one would choose the lowest concentration by which antigen saturation on yeast cells is reached. Therefore, incubate yeast cells with dilutions of varying antigen concentrations (e.g., 1 μM, 500 nM, and

Generation & Isolation of Anti-Hapten Antibodies

285

250 nM) according to the procedure in Subheading 3.4.3. The lowest antigen concentration at which the antigen-binding signal reaches a constant value should be used as it is amenable to saturate the binding signals. Human Fc-hapten conjugate dilutions can be prepared in DPBS. During the screening process, the antigen concentration can be lowered to increase sorting stringency which favors selection of high-affinity target binding variants. 16. A variety of different fluorescently labeled antibodies is commercially available and can be used. When selecting the fluorophore combination, consider that excitation and emission spectra should not overlap. Here, we used either a combination of goat F(ab0 )2 anti-human lambda-PE conjugate and goat IgG anti-human Fcγ fragment-specific Alexa Fluor™ 647 conjugate or a combination of goat F(ab0 )2 anti-human lambda-Alexa Fluor™ 647 conjugate and goat F(ab0 )2 anti-human Fcγ fragment-specific PE conjugate. To avoid the enrichment of antibody fragments that bind to labeling reagents, both combinations of labeling antibodies can be alternated throughout successive screening rounds. 17. It is recommended to prepare at least once a single clone negative control sample to assess unspecific binding signal of human Fc-hapten conjugate to a non-related yeast cell single clone. If there is no suitable yeast cell single clone at hand, we suggest preparing an OKT3 Fab-presenting yeast clone. The preparation of heavy and light chain plasmids can be found elsewhere [29, 30]. Light and heavy chain sequences for OKT3 can be obtained from www.drugbank.ca. 18. When measuring the sort sample, the events in the defined gate should exceed the gated events of the single clone negative control sample to ensure that the double-positive signal in the gate does not result from unspecific binding of the Fc-hapten conjugate or the labeling antibodies. 19. The number of sorted cells should exceed the library size, or the number of isolated cells in a previous sort, by a factor of 10. If the maximum capacity of FACS is exceeded in the first round by the library diversity, it is recommended to use magnetic activated cell sorting (MACS) before the first flow cytometric sorting round of the yeast library. MACS is a simple and fast method to isolate a large number of cells [31]. 20. It is necessary to include one reference control sample skipping the free hapten incubation step (see Fig. 5). We recommend also to prepare various samples with different concentrations of free hapten (e.g., 1, 10, 100 μM). When choosing the concentrations, consider that an excess of hapten is needed for optimal signal suppression as each yeast cell displays ~104–105 Fab fragments on its surface [18].

286

€ger et al. Sebastian Ja

References 1. Grilo AL, Mantalaris A (2019) The increasingly human and profitable monoclonal antibody market. Trends Biotechnol 37:9–16 2. Gao Y, Huang X, Zhu Y, Lv Z (2018) A brief review of monoclonal antibody technology and its representative applications in immunoassays. J Immunoass Immunochem 39:351–364 3. Sheedy C, Roger MacKenzie C, Hall JC (2007) Isolation and affinity maturation of hapten-specific antibodies. Biotechnol Adv 25:333–352 4. Kavanagh O, Elliott CT, Campbell K (2015) Progress in the development of immunoanalytical methods incorporating recombinant antibodies to small molecular weight biotoxins. Anal Bioanal Chem 407:2749–2770 5. Oyama H, Yamaguchi S, Nakata S et al (2013) “Breeding” diagnostic antibodies for higher assay performance: a 250-fold affinity-matured antibody mutant targeting a small biomarker. Anal Chem 85:4930–4937 6. Salomon PL, Singh R (2015) Sensitive ELISA method for the measurement of catabolites of antibody-drug conjugates (ADCs) in target cancer cells. Mol Pharm 12:1752–1761 7. Kovtun YV, Audette CA, Ye Y et al (2006) Antibody-drug conjugates designed to eradicate tumors with homogeneous and heterogeneous expression of the target antigen. Cancer Res 66:3214–3221 8. Erlanger BF (1980) The preparation of antigenic hapten-carrier conjugates: a survey. In: Methods in enzymology, vol 70. Academic, London, pp 85–104 9. Sun Y, Ban B, Bradbury A et al (2016) Combining yeast display and competitive FACS to select rare hapten-specific clones from recombinant antibody libraries. Anal Chem 88:9181–9189 10. Moghaddam A, Løbersli I, Gebhardt K et al (2001) Selection and characterisation of recombinant single-chain antibodies to the hapten aflatoxin-B1 from naive recombinant antibody libraries. J Immunol Methods 254:169–181 11. Kobayashi N, Oyama H (2011) Antibody engineering toward high-sensitivity high-throughput immunosensing of small molecules. Analyst 136:642–651 12. Benatuil L, Perez JM, Belk J, Hsieh CM (2010) An improved yeast transformation method for the generation of very large human antibody libraries. Protein Eng Des Sel 23:155–159

13. Sidhu SS, Lowman HB, Cunningham BC, Wells JA (2000) Phage display for selection of novel binding peptides. In: Methods in enzymology, vol 328. Academic, London, pp 333–363 14. Li Y, Cockburn W, Kilpatrick JB, Whitelam GC (2000) High affinity ScFvs from a single rabbit immunized with multiple haptens. Biochem Biophys Res Commun 268:398–404 15. Chames P, Coulon S, Baty D (1998) Improving the affinity and the fine specificity of an anti-cortisol antibody by parsimonious mutagenesis and phage display. J Immunol 161:5421–5429 16. Orcutt KD, Slusarczyk AL, Cieslewicz M et al (2011) Engineering an antibody with picomolar affinity to DOTA chelates of multiple radionuclides for pretargeted radioimmunotherapy and imaging. Nucl Med Biol 38:223–233 17. Boder ET, Midelfort KS, Wittrup KD (2000) Directed evolution of antibody fragments with monovalent femtomolar antigen-binding affinity. Proc Natl Acad Sci 97:10701–10705 18. Boder ET, Raeeszadeh-Sarmazdeh M, Price JV (2012) Engineering antibodies by yeast display. Arch Biochem Biophys 526:99–106 19. Charlton KA, Moyle S, Porter AJR, Harris WJ (2000) Analysis of the diversity of a sheep antibody repertoire as revealed from a bacteriophage display library. J Immunol 164:6221–6229 20. Grzeschik J, Yanakieva D, Roth L et al (2018) Yeast surface display in combination with fluorescence-activated cell sorting enables the rapid isolation of antibody fragments derived from immunized chickens. Biotechnol J. https://doi.org/10.1002/biot.201800466 21. Spinelli S, Frenken LGJ, Hermans P et al (2000) Camelid heavy-chain variable domains provide efficient combining sites to haptens. Biochemistry 39:1217–1222 22. Krah S, Schro¨ter C, Zielonka S et al (2016) Single-domain antibodies for biomedical applications. Immunopharmacol Immunotoxicol 38:21–28 23. Weaver-Feldhaus JM, Lou J, Coleman JR et al (2004) Yeast mating for combinatorial Fab library generation and surface display. FEBS Lett 564:24–34 24. Boder ET, Wittrup KD (1997) Yeast surface display for screening combinatorial polypeptide libraries. Nat Biotechnol 15:553–557 25. Hermanson G (2008) Preparation of haptencarrier immunogen conjugates. In:

Generation & Isolation of Anti-Hapten Antibodies Bioconjugate techniques, 2nd edn. Academic, London, pp 745–782 26. Pedersen MK, Sorensen NS, Heegaard PMH et al (2006) Effect of different hapten-carrier conjugation ratios and molecular orientations on antibody affinity against a peptide antigen. J Immunol Methods 311:198–206 27. Li Q, Rodriguez LG, Farnsworth DF, Gildersleeve JC (2010) Effects of hapten density on the induced antibody repertoire. Chembiochem 11:1686–1691 28. Hermanson G (2008) Antibody Modification and Conjugation. In: Bioconjugate techniques, 2nd edn. Academic, London, pp 783–823

287

29. Krah S, Schro¨ter C, Eller C et al (2017) Generation of human bispecific common light chain antibodies by combining animal immunization and yeast display. Protein Eng Des Sel 30:291–301 30. Schro¨ter C, Gu¨nther R, Rhiel L et al (2015) A generic approach to engineer antibody pH-switches using combinatorial histidine scanning libraries and yeast display. MAbs 7:138–151 31. Orcutt KD, Wittrup KD (2010) Yeast display and selection. In: Kontermann R, Du¨bel S (eds) Antibody engineering, vol 1. Springer, Heidelberg, pp 207–233

Chapter 16 Rapid Generation of Chicken Immune Libraries for Yeast Surface Display Jan P. Bogen, Julius Grzeschik, Simon Krah, Stefan Zielonka, and Harald Kolmar Abstract Fluorescence-activated cell sorting (FACS) in combination with yeast surface display has emerged as a vital tool for the isolation and engineering of antibodies and antibody-derived fragments from synthetic, naı¨ve, and immune libraries. However, the generation of antibodies against certain human antigens from immunized animals, e.g., mice, can remain challenging due to the homology to the murine counterpart. Due to the phylogenetic distance from humans, avian immunization can be a powerful technique for the generation of antibodies with high specificity against human antigens. Additionally, the peculiar Ig gene diversification in chickens enables the amplification of heavy and light chain genes utilizing single primer pairs, resulting in a convenient library generation. Herein, we describe the protocol for the construction of a single chain fragment variable (scFv) library derived from chickens after immunization with epidermal growth factor receptor (EGFR) for subsequent yeast surface display as well as the screening process utilizing FACS for the isolation of high-affinity antibodies. Key words Chicken antibody, scFv, Yeast surface display, Fluorescence-activated cell sorting, Library generation, Immune antibody library

1

Introduction Besides IgA and IgM isotypes, the major antibody isotype in birds is the immunoglobulin Y (IgY) [1]. Despite the similarities between mammalian IgG and avian IgY, e.g., their biological role, there are profound differences in structure, molecular size, and biochemical properties. IgY antibodies comprise an additional constant domain in their heavy chain resulting in a higher molecular mass. Furthermore, they lack of a hinge region leading to a less flexible binding scaffold in comparison with mammalian IgGs [2, 3]. One major difference is the generation of diversity in the antibody repertoire. While mammalians comprise different V-, D-, and J-gene segments that are recombined [4], chickens possess theoretically uniform V-gene sequences that are diversified based on somatic gene

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2070, https://doi.org/10.1007/978-1-4939-9853-1_16, © Springer Science+Business Media, LLC, part of Springer Nature 2020

289

290

Jan P. Bogen et al.

conversion [5–9]. Additionally, the CDR3 loops in IgYs tend to be longer and have a higher cysteine content in comparison with human or murine IgGs enabling the potential generation of non-canonical disulfide bonds that increase stability and complexity of CDR loops [10]. Furthermore, despite the mentioned differences, a humanization process of chicken antibodies is reported to be feasible [11, 12]. Due to the phylogenetic distance to humans, avian immunization enables the generation of antibodies against epitopes on human antigens, which might not be able to be targeted with antibodies from mice or rats [13, 14]. Using the scFv format, several chicken antibodies have been generated and isolated utilizing phage display [13, 15–17]. Alternatively, yeast surface display can be utilized as a screening platform to isolate antibodies and antibody-derived fragments [18]. This process allows the real-time analysis of library candidates and their enrichment by fluorescence-activated cell sorting [19] and ensures a correct protein folding due to the usage of an eukaryotic expression apparatus [20, 21]. In yeast libraries, the gene of interest is genetically fused with either C- or N-terminally to the a-agglutinin II gene, encoding for the subunit of cell wall-anchored Aga2p protein [22]. Upon expression, yeast cells secrete the fusion protein being covalently attached to the membrane-located a-agglutinin I subunit (Aga1p) via two disulfide bonds. In this chapter, we describe protocols for the cDNA synthesis from splenic RNA after chicken immunization, the establishment of a yeast surface-displayed library of chicken-derived scFv molecules, and the subsequent FACS-based isolation of high-affinity binders (Fig. 1).

2

Materials

2.1 Chicken Immunization and Isolation of Total Spleen RNA

1. AddaVax (InvivoGen).

2.2 cDNA Synthesis and Amplification of VH and VL Genes

1. SuperScript III First-Strand Kit (Thermo Fisher Scientific).

2. TriFast reagent (VWR).

2. Taq DNA polymerase (New England Biolabs). 3. 10
Taq buffer (New England Biolabs). 4. dNTPs. 5. Nuclease-free water. 6. Thermocycler. 7. Device and reagents for agarose gel electrophoresis. 8. PCR Clean Up System. 9. BioSpec Nano or equivalent instrumentation.

Chicken Immune Libraries for Yeast Surface Display

291

Fig. 1 Flow diagram for the generation of chicken-derived scFv molecules from immune evolved repertoires. Isolation of spleen cells was followed by extraction of total RNA and consecutive cDNA synthesis. cDNA was further used as template for separate amplification of VH and VL genes. Resulting scFv molecules were displayed on yeast surface as Aga2p fusions including a C-terminal cMyc-tag to confirm correct antibody surface expression

2.3 Construction of Yeast Surface Display Library

1. BamHI-HF® (New England Biolabs). 2. NheI-HF® (New England Biolabs). 3. 10
CutSmart Buffer (New England Biolabs). 4. pCT plasmid [22]. 5. Yeast strain: Saccharomyces cerevisiae strain EBY100 [22]. 6. YPD media: 20 g/l D(+)-glucose, 20 g/l tryptone, 10 g/l yeast extract, and 100 μg/ml ampicillin. 7. Electroporation buffer: 1 M sorbitol, 1 mM CaCl2. 8. LiAc buffer: 0.1 M LiAc, 10 mM DTT. 9. 1 M sorbitol. 10. PBS: 8.1 g/l NaCl, 1.13 g/l Na2HPO4, 0.75 g/l KCl, and 0.27 g/l KH2PO4. 11. Electroporator Gene Pulser Xcell™ (Bio-Rad). 12. 0.2 cm Electroporation cuvettes (Bio-Rad). 13. Bacto™ Casamino acids (BD Biosciences, San Jose, USA). 14. SD-CAA: 8.6 g/l NaH2PO4
H2O, 5.4 g/l Na2HPO4, 1.7 g/l yeast nitrogen base without amino acids, 5 g/l

292

Jan P. Bogen et al.

ammonium sulfate, 5 g/l Bacto Casamino Acids, 20 g/l glucose, and 100 μg/ml ampicillin (+12 g/l agar agar for agar plates). 15. SG-CAA: 8.6 g/l NaH2PO4
H2O, 5.4 g/l Na2HPO4, 1.7 g/l yeast nitrogen base without amino acids, 5 g/l ammonium sulfate, 5 g/l Bacto Casamino Acids, 20 g/l galactose, and 100 μg/ml ampicillin. 16. 9 cm Petri dishes. 2.4 Surface Expression of Chicken scFvs and Library Screening

1. BPBS: PBS, 0.1% (w/v) bovine serum albumin. 2. His-tagged EGFR (produced in-house). 3. Fc-tagged EGFR (R&D Systems). 4. Anti-Penta-His Alexa Fluor 647 Conjugate antibody (Qiagen). 5. Anti-Human IgG Fc-PE conjugate (Affymetrix eBioscience). 6. Anti-c-myc-biotin antibody (Miltenyi Biotec). 7. BD Influx or comparable cell sorter.

2.5 Sequencing of High-Affinity scFvs

3

1. Zymoprep Yeast Plasmid Miniprep (Zymo Research). 2. E. coli Top10.

Methods In the following section, we describe the two-step PCR protocol for amplification of VH and VL genes based on synthesized cDNA and the subsequent fusion PCR for the construction of full-length scFvs. With the aid of homologous recombination, the scFv genes are inserted into a linearized yeast shuttle vector, enabling the display of the antibody fragment on the surface of yeast cells. We also present the exemplary labeling strategy utilized for the screening of the yeast display library for EGFR binders, as well as the sorting procedure until sequencing of high-affinity binders.

3.1 Chicken Immunization and Isolation of Total Spleen RNA

The immunization of chickens (Gallus gallus domesticus) should be performed in accordance with local animal welfare protection laws and regulations. In the herein presented example, immunization of chickens was executed at Davids Biotechnologie (Regensburg, Germany). Pathogen-free laying hens were immunized with 150 μg human EGFR (extracellular domain, produced and purified in-house) using AddaVax (InvivoGen) as vaccine adjuvant. Following the first immunization, booster immunizations were performed after 2, 4, 5, and 8 weeks. After 5 weeks, the immune response to the antigen was determined by monitoring their antigen-specific titer utilizing ELISA. Chickens were sacrificed after 9 weeks, and the total splenic RNA was isolated using TriFast reagent (VWR).

Chicken Immune Libraries for Yeast Surface Display

3.2 cDNA Synthesis and Amplification of VH and VL Genes

293

cDNA synthesis is performed utilizing the SuperScript™ III FirstStrand Synthesis System Kit by Invitrogen according to the manufacturer’s instruction. Subsequently, VH and VL genes are amplified from the resulting cDNA in separate PCR reactions. For the construction of scFv-encoding fragments, VH and VL genes are fused in a second PCR. 1. cDNA synthesis is performed following the protocol of the SuperScript™ III First-Strand Synthesis System Kit by Invitrogen utilizing the provided random hexamer primer and SuperScript III reverse transcriptase. The mixture is incubated for 5 min at 25  C followed by an incubation step for 60 min at 50  C. Inactivation is performed at 85  C for 5 min. The RNA templates are subsequently digested using 1 μl RNase H at 37  C for 20 min. 2. For the amplification of VH genes, the primer pair VH_gr_up/ VH_SOE_lo is used and for the VL genes the primer pair VL_SOE_up/VL_gr_lo (listed in Table 1). These primers incorporate an overhang encoding for the (Gly4Ser)3 linker for the following fusion PCR. Amplification is performed using Taq polymerase with 3 μl cDNA for each 100 μl PCR. Add 2 μl of each primer (10 μM stock) together with 2 μl dNTPs (10 mM each), 10 μl 10
Standard Taq Buffer, and finally 0.5 μl Taq polymerase. Add nuclease-free water to a final volume of 100 μl. At least six reactions are prepared for each chain. The PCR conditions are listed in Table 2. 3. Analyze the PCR products on a 1% (w/v) agarose gel to confirm the successful amplification of VH (~450 bp) and VL genes (~380 bp) (Fig. 2) (see Note 1). Purify the resulting PCR products utilizing the Promega Wizard® SV Gel and PCR Clean-up System or a comparable kit. Determine the DNA concentration. 4. In the second PCR, the scFvs are assembled by fusing 200 ng PCR product of each chain (VH and VL) by PCR utilizing Taq polymerase. Carry out approx. 40 reactions in parallel with a total reaction volume of 100 μl. Mix DNA templates with 2 μl dNTPs (10 mM each), 10 μl 10
Standard Taq Buffer, and finally 0.5 μl Taq polymerase. The first 15 cycles are performed without addition of primers to allow the annealing and elongation of both fragments. After addition of 2 μl of each primer VH_gr_up and VL_gr_lo (10 μM stock), 15 additional amplification cycles are performed. The resulting PCR product encodes for the scFv and possesses overhangs for homologous recombination in yeast. The conditions for the second PCR are listed in Table 2.

294

Jan P. Bogen et al.

Table 1 Primer sequences for the amplification of avian VH and VL genes and the subsequent fusion PCR Name

Sequence (50 -30 )

Variable heavy chain VH_gr_up

GGTGGTGGTGGTTCTGGTGGTGGTGGTTCTGCTAGCGCCGTGACG TTGGACGAG

VH_SOE_lo TCCGCCCCCCGACCCGCCGCCGCCTGAGCCGCCTCCCCCGGAGGAGACGA TGACTTCGGT Variable light chain VL_SOE_up GGCGGCTCAGGCGGCGGCGGGTCGGGGGGCGGAGGGAGCGCGCTGAC TCAGCCGTCCTCG VL_gr_lo

CAAGTCCTCTTCAGAAATAAGCTTTTGTTCGGATCCTAGGACGGTCAGGG TTGTCCC

Sequencing pCT_seq_up TACCCATACGACGTTCCAGACTAC pCT_seq_lo CAGTGGGAACAAAGTCGATTTTGTTAC Restriction enzyme sites and extra bases to enable homologous recombination or fusion PCR are underlined

Table 2 PCR program for the amplification of VH and VL genes (1. PCR) and the following fusion PCR for the construction of scFvs (2. PCR) Initial denaturation 95  C 30 s Denaturation Annealing Extension Final extension

95  C 20 s 55  C 30 s (1. PCR) 50 s (2. PCR) 68  C 30 s (1. PCR) 50 s (2. PCR)

30 cycles (1. PCR)/15 cycles without primers + 15 cycles with primers (2. PCR)

68  C 5 min

5. Analyze the PCR products on a 1% (w/v) agarose gel (~800 bp) (Fig. 2), and purify the resulting PCR products utilizing the Promega Wizard® SV Gel and PCR Clean-up System or a comparable kit. Determine the DNA concentration and store the DNA at 20  C until further use. 3.3 Generation of Chicken scFv Library in Yeast

The following protocol for the generation of yeast surface display library involves a modified version of the improved yeast transformation protocol of Benatuil and colleagues [23].

Chicken Immune Libraries for Yeast Surface Display

295

Fig. 2 Agarose gel electrophoresis showing the amplicons for VH and VL chains and the scFv gene construct. VH ~ 450 bp; VL ~ 380 bp; scFv ~ 800 bp 3.3.1 Digestion of pCT Plasmid

Libraries for yeast surface display are typically generated by a homologous recombination-based process, referred to as gap repair. This is performed by transforming a linear vector with the respective PCR product, both harboring homologous overhangs. To perform 14 transformations per library, digest approx. 100 μg pCT vector. 1. Perform the double digestion in a 300 μl reaction volume comprising 100 μg pCT plasmid, 120 U of BamHIHF®,120 U NheI-HF®, and 30 μl CutSmart Buffer. Add nuclease-free water to a final volume of 300 μl, and incubate the mixture overnight at 37  C. 2. Analyze an aliquot of the reaction on a 1% (w/v) agarose gel to confirm complete plasmid digestion. Purify the digested plasmid utilizing a PCR clean-up kit according to the supplier’s manual. Determine the DNA concentration and store the DNA at 20  C until needed.

3.3.2 Yeast Transformation

The protocol is designed for two separate electroporation reactions. For larger libraries, scale up the protocol. The detailed version of the used protocol for the improved yeast electroporation can be found elsewhere [23]. To ensure a sufficient transformation efficiency, keep cells on ice whenever possible (see Note 2). In brief: 1. Grow EBY100 overnight in YPD media at 30  C at 180 rpm. 2. Inoculate 100 ml fresh YPD media with the overnight culture to an OD600 of about 0.3. 3. Incubate cells at 30  C and 180 rpm until OD600 reaches about 1.6. 4. Centrifuge cells at 4000
g for 3 min and discard the supernatant.

296

Jan P. Bogen et al.

5. Wash cells twice (by resuspending) with 50 ml ice-cold water, followed by a washing step with 50 ml ice-cold electroporation buffer. 6. Resuspend cells in 20 ml LiAc buffer and incubate for 30 min at 30  C and 180 rpm. 7. Centrifuge cells and wash once with 50 ml ice-cold electroporation buffer. 8. Resuspend the cells in approx. 200 μl electroporation buffer, resulting in about 1 ml cell suspension. This is a sufficient volume for performing two electroporation steps using 400 μl each. Keep cells on ice until electroporation. 9. Mix 4 μg linear pCT and 12 μg scFv-encoding PCR product with 400 μl EBY100 cells. The volume of the DNA mixture should not exceed 50 μl. 10. Transfer the cell-DNA mixture into an ice-cold electroporation cuvette (0.2 cm). Electroporate at 2.500 V. The time constants should range from 3.0 to 4.5 ms. Transfer 400 μl of electroporated cells in 8–10 ml of a 1:1 mixture of YPD and 1 M sorbitol, and incubate for 1 h at 30  C and 180 rpm. 11. Centrifuge cells and resuspend in 10 ml PBS. To enable subsequent estimation of the library complexity, perform dilution plating on SD-CAA plates, and incubate cells at 30  C for approx. 72 h. Transfer remaining cells in 1 l SD-CAA media, and incubate for 2 days at 30  C and 180 rpm. 12. For long-time storage, centrifuge 1
1010 (see Note 3) cells, and resuspend in 2 ml of 5% (v/v) glycerol and 0.67% (w/v) yeast nitrogen base, and store vial at 80  C (see Note 4). To evaluate the constructed library, the authors recommend picking at least ten single clones from the dilution plates and performing a colony PCR utilizing the primer pair pCT_Seq_up and pCT_Seq_lo to confirm the presence of the correct inserts at ~900 bp. Sequence PCR products of positive clones to confirm diversity within those ten clones. 3.4 Screening for High-Affinity Chicken-Derived scFvs 3.4.1 Induction of scFv Display

1. Thaw an aliquot of the frozen library, and transfer the cells into fresh SD-CAA media with a final OD600 of approx. 0.5. Note that the total number of cells used for inoculation should exceed the calculated library complexity by at least the factor of 10. 2. Incubate cells overnight at 30  C and 180 rpm. 3. Harvest at least a cell number exceeding the library complexity by the factor of 10, and resuspend cells in SG-CAA media to an initial OD600 of 1. Incubate cells for 24–48 h at 30  C and 180 rpm for induction of scFv-surface presentation.

Chicken Immune Libraries for Yeast Surface Display 3.4.2 Analysis of Surface Presentation

297

To confirm the surface presentation of scFv fragments, the c-terminally located c-myc tag is used for labeling of full-length scFv variants. 1. To verify successful induction of the surface display of chickenderived scFvs, approx. 1
107 cells of the induced library are centrifuged (see Note 5). 2. Resuspend cells in 20 μl ice-cold BPBS in the presence of a 1:50 dilution of an anti-c-myc-biotin antibody, and incubate the mixture on ice for 15 min. As negative control, use an aliquot of yeast cells and skip this staining step and continue with step 4 (see Note 6). 3. Wash cells twice with 1 ml of ice-cold BPBS. 4. Resuspend cells in 20 μl BPBS, and add Streptavidin-PE or Streptavidin-APC conjugate in a final dilution of 1:75, and incubate the mixture for additional 15 min on ice. 5. After washing twice with ice-cold BPBS, resuspend cells in 500 μl BPBS (see Note 7). To determine surface presentation, analyze at first the negative control by flow cytometry. Apply a gate that allows max. 1% of cells being located within the gate. Now examine the presentation control utilizing the same gate. The percentage of cells within this gate corresponds to the surface presentation. If the library shows a sufficient surface presentation (see Note 8), additional cells can be stained with the antigen to enable fluorescence-activated cell sorting (Fig. 3a).

Fig. 3 FACS staining of chicken-derived scFv library derived from immunized chickens. (a) Histogram of cMyc surface expression of the constructed library assessed by indirect immunofluorescence labeling and flow cytometry. Red, negative control (cells after 2 days induction of gene expression) labeled with detection reagents only; green, immunofluorescence staining after 2 days induction of gene expression. (b) Library screening against EGFR. Target concentrations, sorting gates, and percentages of cells in the sorting gates are shown. Yeast cells were labeled for simultaneous detection of antigen binding and surface presentation. Cells in the sorting gate were isolated, grown, and induced for the next round of selection. The output after second sorting round was only analyzed via FACS but not sorted again

298

Jan P. Bogen et al.

3.4.3 Library Staining

The following protocol describes the staining of a scFv library, derived from chickens after immunization with the target protein EGFR. For the screening procedure, a his-tagged and an Fc-fused variant of EGFR were utilized, but all staining steps could be adjusted for other antigens and varying staining strategies. Herein we describe the sorting process for two consecutive rounds, resulting in the isolation of high-affinity binders.

First Sorting Round

The protocol describes the staining of 1
107 clones for one sorting round. Scale up depending on library size (see Note 9). 1. Harvest 1
107 clone from the induced library by centrifugation, and wash cells once with 1 ml ice-cold BPBS. 2. In a first staining step, add a 1:50 dilution of an anti-c-mycbiotin antibody in the presence of 1 μM his-tagged EGFR, and incubate the mixture on ice for 30 min. For a negative control, perform a separate reaction in the absence of antigen. 3. Wash cells twice with 1 ml of ice-cold BPBS. 4. Resuspend cells in a mixture of a 1:75 dilution of StreptavidinPE conjugate and a 1:20 dilution of anti-Penta-His Alexa Fluor 647 conjugate antibody, and incubate the reaction for 15 min on ice. 5. After washing twice with ice-cold BPBS, resuspend cells in 1 ml BPBS for the subsequent FACS-assisted sorting process (Subheading 3.5).

Second Sorting Round

The protocol describes the staining of 1
107 clones for one sorting round. Scale up depending on library size (see Note 9). For the second round, the staining procedure is altered to avoid the enrichment of scFvs that bind to the detection reagents. 1. Harvest 1
107 clones from the induced library by centrifugation, and resuspend cells in 20 μl ice-cold BPBS. 2. In the first staining step, add a 1:50 dilution of an anti-c-mycbiotin antibody in the presence of 1 nM Fc-fused EGFR, and incubate the mixture on ice for 30 min. As a negative control, perform this step in the absence of EGFR in a separate reaction. 3. Wash cells twice with ice-cold BPBS, and resuspend cells in 20 μl BPBS. 4. Add the Streptavidin-APC conjugate in a 1:75 dilution and the anti-Human IgG Fc-PE conjugate in a 1:75 dilution in BPBS. Incubate mixture on ice for 15 min. 5. After washing twice with ice-cold BPBS, resuspend cells in 1 ml BPBS for the subsequent FACS-assisted sorting process (Subheading 3.5).

Chicken Immune Libraries for Yeast Surface Display

3.5 FluorescenceActivated Cell Sorting

299

In this section, we describe the cell sorting process after all staining steps are completed. For sorting of the anti-EGFR chicken-derived scFv library, the authors used a BD Influx device, but in general, all cell sorters with at least two different lasers might be used. The first round of sorting is performed using the “sort enrich mode,” following rounds in the “sort pure mode.” If other cell sorters are used, settings need to be adjusted accordingly. 1. Define a suitable sorting gate that includes cells that are double positive for surface presentation and target binding. Decide for gate settings that include max. 0.3% of cells of the respective negative control in the absence of antigen. An example can be seen in Fig. 3b. The number of sorted cells should at least exceed the library complexity by the factor 10. 2. After sorting, transfer cells to SD-CAA media and incubate at 30  C for 48 h. For later use, freeze sorted cells in cryo-vials as described above (Subheading 3.3.2; see Note 10). The number of frozen cells should exceed the number of isolated cells at least ten times. 3. To perform further screening rounds, inoculate SG-CAA media using the previously isolated cells to an OD600 of ~0.5, and incubate cells for 24–48 h at 30  C and 180 rpm for induction of scFv-surface presentation, and repeat library staining (Subheading 3.4.3) and FACS sorting (Subheading 3.5). Repeat library staining and cell sorting (see Note 11) until an antigen-binding population is enriched (see Note 12). In case of the anti-EGFR library, two rounds of sorting were sufficient to enrich exclusively clones showing surface presentation and EGFR binding (Fig. 3b). However, the number of rounds might depend on the respective antigen, the animals’ immune response, and the library quality. Prior to analysis of single clones or sequencing of scFvs, the sorted cells resulting from the second screening round were stained again and analyzed by FACS (Fig. 3b). Flow cytometry confirmed nearly all scFv-presenting cells being able to bind the antigen with high affinity. Consequently, scFvs were further analyzed for their sequences in order to analyze the diversity of the enriched population.

3.6 Sequencing of Antigen-Binding scFvs

After successful enrichment of antigen-binding scFvs, the population of yeast cells is lysed utilizing the Zymoprep kit according to the manufacturer’s instructions. The isolated plasmids are transformed into E. coli Top10, and cells are grown overnight on dYT amp+agar plates. Subsequently, single clones are picked, and either they are sent to a sequencing service, or plasmids are isolated individually and sequenced using the primers pCT_Seq_up or pCT_Seq_lo (see Note 13). Based on resulting CDR sequences, a clustering is possible enabling the identification of high-affinity binders.

300

4

Jan P. Bogen et al.

Notes 1. After initial PCR, when multiple bands (i.e., unwanted side products) are visible on the gel, the band corresponding to the correct size can be selectively purified via agarose gel electrophoresis followed by gel excision. 2. In order to ensure a high efficiency in yeast transformation, keep cells on ice at every step, and only use ice-cold buffers and solutions. 3. An OD600 of 1.0 corresponds to about 1
107 yeast cells. 4. To improve the freezing condition for yeast aliquots, place cryo-vials in a bath of isopropanol and slow-freeze cells to 80  C. For the generation of cryo-vials, always freeze a cell number exceeding your library size or your number of sorted cells by the factor of 10, respectively. 5. A cell number of 1
107 cells should be stained in a volume of 20 μl. Scale the volume up for larger staining reactions, e.g., use 200 μl staining volume for 1
108 cells. 6. As a negative control, the staining for verification of the surface presentation should be done without the primary antibody (anti-myc-biotin). These negative cells are used later in comparison with the presentation control to evaluate the surface presentation. 7. Depending on the utilized fluorophores, cells should be shielded from light until FACS analysis to prevent fluorophore bleaching. 8. The authors recommend that the surface presentation should be at least 20% in order to stain yeast cells for sorting experiments. 9. If possible, screen at least three times more cells than determined via dilution plating to ensure sufficient coverage of library candidates. 10. Freeze yeast cells in cryo-vials after each sorting round in case screening rounds have to be repeated utilizing different staining strategies. 11. The screening reagents should be alternated within consecutive rounds. This is not always mandatory but should be done to avoid the enrichment of scFvs binding to secondary detection reagents. 12. To isolate scFv variants with higher affinities, the authors recommend utilizing decreasing antigen concentrations over consecutive screening rounds.

Chicken Immune Libraries for Yeast Surface Display

301

13. Sequence of Aga2p-scFv fusion protein for yeast surface display: ATGCAGTTACTTCGCTGTTTTTCAATATTTTCTGTTATTGCTTCAGTTTTAGCACAGGAA CTGACAACTATATGCGAGCAAATCCCCTCACCAACTTTAGAATCGACGCCGTACTCTTT GTCAACGACTACTATTTTGGCCAACGGGAAGGCAATGCAAGGAGTTTTTGAATATTAC AAATCAGTAACGTTTGTCAGTAATTGCGGTTCTCACCCCTCAACAACTAGCAAAGGCA GCCCCATAAACACACAGTATGTTTTTAAGGACAATAGCTCGACGATTGAAGGTAGATA CCCATACGACGTTCCAGACTACGCTCTGCAGGCTAGTGGTGGTGGTGGTTCTGGTGGT GGTGGTTCTGGTGGTGGTGGTTCTGCTAGCGCCGTGACGTTGGACGAGTCCGGGGGCG GCCTCCAGACGCCCGGAGGAGGGCTCAGCCTCATCTGCAAGGGCTCCGGGTTCACCTT CAGCAGCTTCAACATGCTCTGGGTGCGACAGGCGCCCGGCAAGGGGCTGGAGTTCGTC GCTGACATTTACAGCACTGGTAGTTACACGAGATACGCGCCGGCGGTGGATGGCCGTG CCACCATCTCGAGGGACAACGGGCAGAGCACAGTGAGACTGCAGCTGAACAACCTCA GGGCTGAGGACACCGCCACCTACTACTGCGCCAAAAGTTCTACTAGTGGTTTTTGTGG TGGTGTTAGTTGTGGCGGACTCATCGACGCATGGGGCCACGGGACCGAAGTCATCGTC TCCTCCGGGGGAGGCGGCTCAGGCGGCGGCGGGTCGGGGGGCGGAGGGAGCGCGCTG ACTCAGCCGTCCTCGGTGTCAGCAAACCTGGGAGGAACCGTCGAGATCACCTGCTCCG GGGGTGTTAACAGCAACCACTATGGCTGGTACCAGCAGAAGTCTCCTGGCAGTGCCCC TGTCACTCTGATCTATGCTAACACCAACAGGCCCTCGGACATCCCTTCACGATTCTCCG GTTCCAAATCCGGCTCCACAGCCACATTGACCATCACTGGGGTCCAAGCCGAGGACGA GGCTGTCTATTTCTGTGGGAGTGGAGACAGCAGTGGTGCTGCATTTGGGGCCGGGACA ACCCTGACCGTCCTAGGATCCGAGCAAAAGCTTATTTCTGAAGAGGACTTGTAA Red: Aga2p; dark green: HA tag; grey: (Gly4Ser)3 linker; blue: chicken VH; yellow: chicken VL; light green: cMyc tag; NheI and BamHI recognition sites shown in bold.

References 1. Tizard I (2002) The avian antibody response. Semin Avian Exotic Pet Med 11(1):2–14. https://doi.org/10.1053/saep.2002.28216 2. Sun S, Mo W, Ji Y, Liu S (2001) Preparation and mass spectrometric study of egg yolk antibody (IgY) against rabies virus. Rapid Commun Mass Spectr 15(9):708–712. https:// doi.org/10.1002/rcm.271 3. Cova L (2005) DNA-designed avian IgY antibodies: novel tools for research, diagnostics and therapy. J Clin Virol 34(Suppl 1):S70–S74 4. Market E, Papavasiliou FN (2003) V(D)J recombination and the evolution of the adaptive immune system. PLoS Biol 1(1):E16. https://doi.org/10.1371/journal.pbio. 0000016

5. McCormack WT, Tjoelker LW, Thompson CB (1993) Immunoglobulin gene diversification by gene conversion. In: Cohn WE, Moldave K (eds) Progress in nucleic acid research and molecular biology, vol 45. Academic, New York, NY, pp 27–45. https://doi.org/ 10.1016/S0079-6603(08)60865-X 6. Reynaud CA, Anquez V, Grimal H, Weill JC (1987) A hyperconversion mechanism generates the chicken light chain preimmune repertoire. Cell 48(3):379–388 7. Reynaud C-A, Dahan A, Anquez V, Weill J-C (1989) Somatic hyperconversion diversifies the single VH gene of the chicken with a high incidence in the D region. Cell 59 (1):171–183. https://doi.org/10.1016/ 0092-8674(89)90879-9

302

Jan P. Bogen et al.

8. Reynaud CA, Anquez V, Weill JC (1991) The chicken D locus and its contribution to the immunoglobulin heavy chain repertoire. Eur J Immunol 21(11):2661–2670. https://doi. org/10.1002/eji.1830211104 9. Thompson CB, Neiman PE (1987) Somatic diversification of the chicken immunoglobulin light chain gene is limited to the rearranged variable gene segment. Cell 48(3):369–378. https://doi.org/10.1016/0092-8674(87) 90188-7 10. Wu L, Oficjalska K, Lambert M, Fennell BJ, Darmanin-Sheehan A, Ni Shuilleabhain D, Autin B, Cummins E, Tchistiakova L, Bloom L, Paulsen J, Gill D, Cunningham O, Finlay WJ (2012) Fundamental characteristics of the immunoglobulin VH repertoire of chickens in comparison with those of humans, mice, and camelids. J Immunol 188 (1):322–333. https://doi.org/10.4049/ jimmunol.1102466 11. Tsurushita N, Park M, Pakabunto K, Ong K, Avdalovic A, Fu H, Jia A, Vasquez M, Kumar S (2004) Humanization of a chicken anti-IL-12 monoclonal antibody. J Immunol Methods 295(1–2):9–19. https://doi.org/10.1016/j. jim.2004.08.018 12. Nishibori N, Horiuchi H, Furusawa S, Matsuda H (2006) Humanization of chicken monoclonal antibody using phage-display system. Mol Immunol 43(6):634–642. https:// doi.org/10.1016/j.molimm.2005.04.002 13. Davies EL, Smith JS, Birkett CR, Manser JM, Anderson-Dear DV, Young JR (1995) Selection of specific phage-display antibodies using libraries derived from chicken immunoglobulin genes. J Immunol Methods 186(1):125–135 14. Carlander D, Sta˚lberg J, Larsson A (1999) Chicken antibodies. Ups J Med Sci 104 (3):179–189. https://doi.org/10.3109/ 03009739909178961 15. Yamanaka HI, Inoue T, Ikeda-Tanaka O (1996) Chicken monoclonal antibody isolated by a phage display system. J Immunol 157 (3):1156–1162 16. Li J, Xu Y, Wang X, Li Y, Wang L, Li X (2016) Construction and characterization of a highly

reactive chicken-derived single-chain variable fragment (scFv) antibody against Staphylococcus aureus developed with the T7 phage display system. Int Immunopharmacol 35:149–154. https://doi.org/10.1016/j.intimp.2016.02. 024 17. Hu ZQ, Li HP, Zhang JB, Huang T, Liu JL, Xue S, Wu AB, Liao YC (2013) A phagedisplayed chicken single-chain antibody fused to alkaline phosphatase detects Fusarium pathogens and their presence in cereal grains. Anal Chim Acta 764:84–92. https://doi.org/ 10.1016/j.aca.2012.12.022 18. Grzeschik J, Yanakieva D, Roth L, Krah S, Hinz SC, Elter A, Zollmann T, Schwall G, Zielonka S, Kolmar (2019) Yeast Surface Display in Combination with Fluorescence‐ activated Cell Sorting Enables the Rapid Isolation of Antibody Fragments Derived from Immunized Chickens. Biotechnol J 14 (4):1800466. https://doi.org/10.1002/biot. 201800466 19. Doerner A, Rhiel L, Zielonka S, Kolmar H (2014) Therapeutic antibody engineering by high efficiency cell screening. FEBS Lett 588 (2):278–287. https://doi.org/10.1016/j. febslet.2013.11.025 20. Welihinda AA, Kaufman RJ (1996) The unfolded protein response pathway in Saccharomyces cerevisiae. Oligomerization and transphosphorylation of Ire1p (Ern1p) are required for kinase activation. J Biol Chem 271 (30):18181–18187 21. Lu Z-J, Deng S-J, Huang D-G, He Y, Lei M, Zhou L, Jin P (2012) Frontier of therapeutic antibody discovery: the challenges and how to face them. World J Biol Chem 3(12):187 22. Boder ET, Wittrup KD (1997) Yeast surface display for screening combinatorial polypeptide libraries. Nat Biotechnol 15(6):553–557. https://doi.org/10.1038/nbt0697-553 23. Benatuil L, Perez JM, Belk J, Hsieh CM (2010) An improved yeast transformation method for the generation of very large human antibody libraries. Protein Eng Design Select 23(4):155–159. https://doi.org/10. 1093/protein/gzq002

Chapter 17 Ligand Engineering via Yeast Surface Display and Adherent Cell Panning Lawrence A. Stern, Patrick S. Lown, and Benjamin J. Hackel Abstract High-throughput ligand discovery and evolution—via genotype-phenotype linkage strategies—empower molecularly targeted therapy, diagnostics, and fundamental science. Maintaining high-quality target antigen in these selections, particularly for membrane targets, is often a technical challenge. Panning yeastdisplayed ligand libraries on intact mammalian cells expressing the molecular target has emerged as an effective strategy. Herein we describe the techniques used to select target-binding ligands via this approach including the use of target-negative cells to deplete non-specific binders and avidity reduction to preferentially select high-affinity ligands. Key words Avidity, Cell panning, Depletion, Ligand, Protein engineering, Specificity, Yeast surface display

1

Introduction Protein engineering uses rational design and directed evolution to modulate the function or physical characteristics of proteins. Protein ligands that selectively bind to molecular targets have been successfully applied for clinical diagnostics, such as targeted molecular imaging and the analysis of blood and urine, as well as targeted therapy through inhibition, drug delivery, radioisotope delivery, and immune system engagement [1–5]. Recent advances in genomics, proteomics, and chemical biology have drastically increased our knowledge of disease states on the molecular level, increasing the number of characterized, clinically relevant biomarkers [6]. With this increase, a new demand has emerged for engineered proteins that target these biomarkers. To meet this demand, a variety of high-throughput screening methods for binding activity have been employed. A key principle

Lawrence A. Stern and Patrick S. Lown contributed equally to this work. Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2070, https://doi.org/10.1007/978-1-4939-9853-1_17, © Springer Science+Business Media, LLC, part of Springer Nature 2020

303

304

Lawrence A. Stern et al.

of high-throughput protein screening is genotype-phenotype linkage—a method by which the function or biophysical form of a protein can be linked to its corresponding DNA. This facile approach allows million to billions of proteins to be screened for binding activity with simple identification of successful candidates by sequencing of the attached gene. Several common genotypephenotype linkage strategies—including ribosome display [7–9], mRNA display [10–12], phage display [13–15], yeast display [16, 17], and mammalian cell display [18, 19]—have been successfully applied for the selection and evolution of engineered binding proteins. While the library sizes of mammalian cell (103–106 transformants) [18] and yeast cell (109 transformants) [16] surface display are smaller than those of mRNA, ribosome, or phage display, both yeast and mammalian cell surface displays are of particular interest because of multiple benefits including eukaryotic protein processing [20, 21], the efficiency of homologous recombination for library assembly (in yeast) [22], and their ability to express 103–105 ligand copies per cell [23]. This high multivalency of ligands, coupled with multivalent target expression on the antigen-expressing cells, can provide avidity—the strength of multiple binding interactions—allowing ligands with very weak affinity to be recovered in a way not possible by monovalent selection techniques (Fig. 1). In part due to these advantages, yeast surface display has been extensively applied to the selection and evolution

Fig. 1 Schematic of cell panning with yeast surface display. A candidate binding ligand (blue) is produced as a fusion to yeast mating protein agglutinin 2 (Aga2p, green) linked by a flexible polypeptide (orange). This fusion is secreted and anchored to the yeast cell wall by disulfide linkage to yeast mating protein agglutinin 1 (Aga1p, pink). When brought into contact with target-expressing cells, the displayed ligands can interact with their binding partners (black) in an avid fashion. Additional cellular macromolecules (red) are also present, which can also interact with displayed ligands

Cell Panning Ligand Selections with Yeast Display

305

of multiple engineered binding protein scaffolds against a variety of target molecules [16, 24, 25]. Experimental protein engineering selection strategies require the use of the biomarker of interest to isolate binding proteins. However, many clinically relevant biomarkers possess hydrophobic transmembrane domains, making them difficult to work with in an aqueous system. In lieu of full-length proteins, recombinant extracellular domains are often used as analogs during ligand selections [26–32]. A limitation of this method is that ligands isolated by recombinant selections may not bind to full-length protein expressed in intact cells [33, 34]. These failures indicate that recombinant analogs may not fully recapitulate the true cellular target. While the issue remains largely unresolved, factors such as protein stability, folding, and modification are likely to blame for these failures. Loss of protein stability can induce misfolding of the protein, whereby the polypeptide is of the proper sequence but adopts an alternate conformation relative to the native protein. The use of such misfolded proteins can induce experimental failure due to a loss of proper biological activity and structure [33, 34]. Additionally, non-natural epitopes are often introduced in the form of additional chemical tags required for purification or immobilization or truncations of the transmembrane domain. In these cases, the isolation of tag-binding ligands can be problematic, even when depletion strategies are employed [28, 35]. These factors motivate the need for an alternate selection approach where full-length protein is expressed in its native conformation, which would enhance selection of genuine cell-binding ligands. To overcome these issues and advance ligand discovery by providing a source of full-length protein in its natural conformation, selection campaigns against intact mammalian cells have been investigated (Fig. 2). These have been employed successfully using both phage and yeast surface display to generate antibody fragments that bind to cancer and blood-brain barrier targets [36–46]. Along with these noted successes, limitations on the fractional yield of ligand-displaying yeast [47] and the generation of ligands not specific to the desired cell line or target remain as challenges for the broader use of these methods [35]. Yield, which is strongly correlated to the target expression on the mammalian cell, is severely hindered in these systems, often to the point of ineffectiveness, when using cells with low-to-medium (105 targets/cell) expression [47]. Previous efforts to optimize selections against adherent mammalian cells and redesign the yeast surface display construct have improved ligand recovery, but in one thorough study ligands of micromolar affinity were robustly recovered only on cells that expressed on the order of one million targets per cell [47]. This limits the applicability of selections against mammalian cells, as the affinity of a ligand in a naı¨ve library may

306

Lawrence A. Stern et al.

Fig. 2 Schematic of the cell panning selection process. A yeast display library is introduced to a mammalian cell monolayer. After sufficient incubation, nonbinding yeast are washed away, leaving only bound yeast on the mammalian cell monolayer. Sequential application of this approach will further enrich binding clones

be weak and most mammalian cell lines do not express sufficient biomarkers to provide effective recovery. Compounding this problem, yeast display selections against mammalian cells often lead to isolation of ligands with activity to an undesired target on the presented cell line, allowing biomarkerspecific or non-target-specific (including off-target specific and non-specific) ligands to simultaneously enrich. While target-specific ligands have been isolated in several cases, isolated binders are often accompanied by a high frequency of non-target-specific binding ligands [35]. Depletions against target-negative mammalian cell monolayers, such as those used to reduce nonspecific binding in phage display cell panning [48], have shown ambiguous efficacy in mitigating this problem. However, preliminary work using disadhered, target-negative mammalian cells has shown promise in providing effective depletion on non-target-specific ligands [35, 42]. Several techniques presented herein are used to effectively deplete non-target-specific ligands and provide the ability to reliably isolate target-specific ligands from cell-based selections. Cell panning with mammalian cell pre-blocking is recommended when screening for specific ligands to a tissue, cell line, or cell surface target of interest from a naı¨ve library to limit nonspecificity; however, solely enrichment-based cell panning can be used as an alternative in cases where an appropriate negative cell line is not readily available. Sequential pannings can be conducted until a suitable endpoint of diversity and enrichment are reached for the particular application. The use of titratable avidity reduction allows for finer discrimination of recovered ligand affinities and is suited for use after the initial screening of a naı¨ve library to provide a further

Cell Panning Ligand Selections with Yeast Display

307

enrichment advantage to higher affinity clones or in the case of more focused libraries where the expected average affinity of a clone is substantially higher than that of a naı¨ve library [49]. Collectively, these techniques comprise an effective strategy for discovery and evolution of ligands against biomedically important, yet challenging, cell membrane targets.

2

Materials 1. pCT-40 (yeast surface display vector designed for cell panning; available upon request; Fig. 3).

2.1 Cell Panning of Yeast SurfaceDisplayed Libraries

2. Growth medium for target-expressing cell line of interest (see Note 1).

2.1.1 Media, Buffers, and Reagents

3. PBSACM buffer: 1 phosphate-buffered saline (PBS), pH 7.4 supplemented with 1% w/w bovine serum albumin, 0.1 mM CaCl2, and 0.1 mM MgCl2∙6H2O; sterile filter (see Note 2). 4. SD-CAA medium: 16.8 g/L sodium citrate dihydrate, 3.9 g/L citric acid, 20.0 g/L dextrose, 6.7 g/L yeast nitrogen base, and 5.0 g/L casamino acids with deionized water; sterile filter.

6500bp

CEN6_ARS4 other 5500bp

f1 origin

5000bp

T7 prom

ampR

500bp

6000bp

1000bp

Trp1

pCT-40 10Fn3 6688bp

1500bp

Origin of replication Other gene

2000bp

Promoter Selectable marker

4500bp

pBR322 origin

2500bp

4000bp 3500bp 3000bp

10Fn3 PAS Linker Aga2p GAL1 prom

GAL1

Aga2p Xa HA E

PAS P

myc

10Fn3 N

B

ZZ

Term X

Fig. 3 pCT-40 vector. CEN6-ARS4 centromeric sequence 6 and autonomously replicating sequence H4, ampR ampicillin resistance selection marker, pBR332 origin pBR332 origin of replication, GAL1 GAL1 promoter, E EcoRI site, Aga2p agglutinin 2p protein subunit, Xa factor Xa cleavage site, HA hemagglutinin epitope, P PstI site, PAS linker 40 subunit PAS linker, N NheI site, 10Fn3 wild-type tenth human fibronectin domain 3 gene, B BamHI site, myc c-Myc epitope, ZZ TAATAG stop codons, X XhoI site, Term alpha mating factor terminator, f1 origin f1 origin of replication, Trp1 Trp1 gene. The lower representation is not to scale

308

Lawrence A. Stern et al.

5. SG-CAA medium: 10.2 g/L sodium phosphate dibasic heptahydrate, 8.6 g/L sodium phosphate monobasic monohydrate, 19.0 g/L galactose, 1.0 g/L dextrose, 6.7 g/L yeast nitrogen base, and 5.0 g/L casamino acids with deionized water; sterile filter. 6. SD-CAA plates: Combine 16.8 g/L sodium citrate dihydrate, 3.9 g/L citric acid, 15 g/L agar with 900 mL/L of deionized water; autoclave; separately combine 20.0 g/L dextrose, 6.7 g/L yeast nitrogen base, and 5.0 g/L casamino acids with 100 mL/L of deionized water, and filter sterilize into autoclaved contents once they have cooled to 50  C; and pour 25 mL into each Petri dish. 7. PBS, pH 7.4, sterile filtered. 8. Trypsin-EDTA. 2.1.2 Equipment and Consumables

1. Stationary incubator with CO2 control at 37  C (for mammalian cell growth). 2. Biosafety cabinet for mammalian cell culture. 3. Shaking incubator at 30  C, 220 rpm (for yeast liquid culture growth). 4. Stationary incubator at 30  C (for yeast plate culture growth). 5. Microcentrifuge with rotor to accommodate 1.7 mL tubes (for preparing yeast for selections). 6. Centrifuge with bucket rotor to accommodate 15 or 50 mL conicals (for preparing yeast for selections). 7. Spectrophotometer for measuring optical density at 600 nm (OD600) (see Note 3). 8. Hemocytometer or other instrument for counting mammalian cells. 9. Mammalian cell culture vessel, e.g., T-75 cell culture flask. 10. 6-Well polystyrene cell culture treated plates. 11. 14 mL polystyrene round-bottom tubes (for yeast liquid culture growth). 12. 250 mL baffled flasks (for yeast liquid culture growth). 13. Pipette controller. 14. Serological pipettes. 15. Pipette tips. 16. Cell scrapers. 17. 1.7 mL tubes. 18. Petri dishes. (All plates, tubes, flasks, pipettes, tips, scrapers, and dishes must be sterile.)

Cell Panning Ligand Selections with Yeast Display 2.1.3 Cells

309

1. Saccharomyces cerevisiae yeast surface display strain EBY100 (available from ATCC) [16]. 2. Target-expressing adherent cell line.

2.2 Cell Panning with Depletion by Mammalian Cell Pre-blocking 2.2.1 Cells

2.3 Cell Panning with Titratable Avidity Reduction 2.3.1 Media, Buffers, and Reagents

Same as Subheading 2.1, with the following addition:

1. Target-negative cell line. Cell line(s) with similar expression profiles but lacking target are preferable. Same as Subheading 2.1, with the following additions:

1. 10 mM Tris buffer, pH 7.5: 1.24 g/L Tris–HCl and 0.26 g/L Tris base with deionized water. Validate pH. Adjust with 10 mM Tris–HCl or 10 mM Tris base as necessary. Sterile filter. 2. Dithiothreitol (DTT) powder. 3. Mouse anti-c-Myc antibody 9E10 (BioLegend). 4. Goat anti-mouse Alexa Fluor 647 antibody (Thermo Scientific). 5. Quantum Simply Cellular anti-mouse IgG kit (Bangs Laboratories).

2.3.2 Equipment and Consumables

3

1. Flow cytometer. 2. Flow cytometry tubes compatible with available flow cytometer.

Methods

3.1 Cell Panning of Yeast SurfaceDisplayed Libraries

This technique is appropriate for selections of ligands—from both naı¨ve and enriched libraries—that bind to adherent mammalian cells. If available, it is recommended that this protocol be paired with a strategy to deplete non-target-specific binding ligands, including mammalian cell pre-blocking (Subheading 3.2) or ligand selection with purified recombinant antigen. In instances where the antigen is not known (e.g., panning against patient-derived tumors) or no appropriate target-negative cell line exists, this protocol can be used with the understanding that an abundant background of non-target-specific ligands will be recovered. Enrichment and yield may vary based on ligand affinity, target expression, and other factors (Fig. 4) [47].

310

Lawrence A. Stern et al.

Fig. 4 Cell panning enrichment of yeast surface-displayed binding ligands. Yeast expressing a ligand library are introduced to a target-expressing mammalian cell monolayer (a). Previous experience with an affinity series of EGFR-binding fibronectin domains panned against EGFR-high MDA-MB-468 breast cancer informs experimental expectations. Binding clones with affinities of low nM to μM were effectively enriched from a dilute pool (1:1000 ligand-displaying yeast: non-displaying yeast) (b). High- and mid-affinity binding resulted in significantly higher yield ( p < 0.05) than low-affinity binding (c) [47] 3.1.1 Target-Expressing Mammalian Cell Preparation

1. Seed target-expressing mammalian cell line in 6-well plates according to their recommended seeding density (see Notes 4 and 5). One well of mammalian cells should be grown per 108 yeast to be sorted. 2. Culture cells at 37  C in a humidified atmosphere with 5% CO2 in the appropriate growth medium. Grow cells to 70% confluence, as quantified by light microscopy.

3.1.2 Yeast Cell Preparation

1. Grow the library of plasmid-harboring EBY100 in SD-CAA medium at 30  C with shaking at 220 rpm for at least two doublings (see Note 3). 2. While the yeast are in logarithmic phase (OD600nm < 6), pellet yeast (8000  g for 1 min), and aspirate medium. Resuspend yeast in fresh SG-CAA to an initial OD600nm  1. Grow at 30  C, 220 rpm for at least 8 h to induce ligand display. Use yeast immediately or store yeast at 4  C (see Note 6).

Cell Panning Ligand Selections with Yeast Display 3.1.3 Cell Panning Selection

311

1. Measure an appropriate amount of yeast for selection (see Note 3). Wash yeast once with ice-cold PBSACM by pelleting at 8000  g for 1 min followed by aspirating the medium. Resuspend in 1 mL of PBSACM before again pelleting and aspirating the supernatant. Leave yeast as a wet pellet on ice. 2. To prepare the mammalian cells for selection, aspirate the cell culture medium from each well using a serological pipette (see Note 7). Wash the cells by gently adding 1 mL ice-cold PBSACM down the wall of the well using a serological pipette. Gently tilt plate five times (see Note 8). Aspirate the buffer using a serological pipette. Visually check the plate surface to ensure mammalian cells remain adhered to the plate surface (see Note 9). 3. Resuspend pelleted yeast using ice-cold PBSACM to 108 yeast/mL (see Note 10). 4. Apply 1 mL of yeast to each well of target-expressing mammalian cells using a 1 mL pipette (see Note 11). 5. Incubate this mixture statically for 15 min at 4  C. 6. Aspirate unbound yeast. Gently add 1 mL of ice-cold PBSACM to each well of selection using a serological pipette. Gently tilt the plate 25 times and nutate it 5 times by hand (see Note 12). Repeat this process three times for a total of four washes this way. For the fifth wash, add 1 mL of ice-cold PBSACM using a serological pipette, and nutate the plate ten times. Aspirate the supernatant. 7. Add 1 mL of SD-CAA using a 1 mL pipette. Remove adhered mammalian cells, and bound yeast by thoroughly scraping the plate surface using a cell scraper. Pipette all liquid and place into a 1.7 mL tube. 8. Plate dilutions of the final yeast population on SD-CAA plates to quantify yield (see Note 13). Culture final population in 5 mL SD-CAA in a 14 mL polystyrene round-bottom tube overnight at 30  C with shaking.

3.2 Cell Panning with Depletion by Mammalian Cell Pre-blocking

3.2.1 Target-Negative Mammalian Cell Preparation

This technique is appropriate for selection from both naı¨ve and enriched libraries. Previous experience with this protocol shows that it is effective in lessening the background of non-target-specific ligands while not significantly impeding the enrichment of targetspecific ligands (Fig. 5). For preparation of target-expressing mammalian cells and yeast, follow Subheadings 3.1.1 and 3.1.2, respectively. 1. Seed target-negative mammalian cell line in T-75 flask according to their recommended seeding density (see Notes 4 and 5). Culture mammalian cells at 37  C in a humidified atmosphere

312

Lawrence A. Stern et al.

Fig. 5 Cell panning with depletion by mammalian cell pre-blocking. Yeast expressing a ligand library are incubated with disadhered target-negative mammalian cells. This mixture is then panned against targetpositive mammalian cells, resulting in depletion of non-target-specific binding yeast and enrichment of target specific interactions (a). Previous experiments panning a panel of yeast-displayed affibody domains on targetexpressing cells as per Subheading 3.1 (nondepleted) or Subheading 3.2 (depleted) using target-negative cells as a depletion agent inform expectations. The yields of specific ligands (black) are not significantly different between the two procedures. In contrast, the yield of nonspecific ligands (gray) is significantly decreased ( p < 0.05) in six of eight clones using target-negative cells as a depletion agent (b)

with recommended CO2 in the appropriate growth medium. Grow to 70–90% confluence, as quantified by light microscopy. 2. Aspirate the cell culture medium from the flask using a serological pipette. Wash the cells by gently adding 5 mL PBS using a serological pipette. Disadhere cells by trypsin-EDTA treatment. Neutralize trypsin using serum-containing cell culture medium. Centrifuge cells at 500  g for 3 min. Aspirate medium/trypsin mixture. Resuspend cells in serum-containing cell culture medium. 3. Count the disadhered cells using a hemocytometer. 4. Harvest 106 cells per 108 yeast required. Centrifuge mammalian cells at 500  g for 3 min at 4  C. Remove supernatant. Resuspend cells in 5 mL of ice-cold PBSACM. Repeat twice. 5. Resuspend mammalian cells in ice-cold PBSACM to a concentration of 106 cells/mL, with a total volume equal to 1 mL/ well.

Cell Panning Ligand Selections with Yeast Display 3.2.2 Cell Selection

313

1. Resuspend pelleted yeast using the target-negative cell solution to a concentration of 108 yeast/mL in PBSACM with 106 target-negative mammalian cells per mL. Split the yeast and mammalian cell mixture into 1 mL aliquots in 1.7 mL tubes. 2. Allow the yeast and mammalian cells to rotate for 2 h at 4  C. 3. Gently apply 1 mL of yeast and mammalian cell mixture to each well of target-expressing mammalian cells using a 1 mL pipette (see Note 11). 4. Incubate selection mixture statically for 15 min at 4  C. 5. Aspirate unbound yeast and target-negative mammalian cells. 6. Gently add 1 mL ice-cold PBSACM dropwise to each well of selection using a serological pipette. Gently tilt the plate 25 times and nutate it 5 times by hand (see Note 12). Repeat this three times for a total of four washes this way. For the fifth wash, add 1 mL of ice-cold PBSACM using a serological pipette, and nutate the plate ten times. Aspirate the supernatant. 7. Add 1 mL SD-CAA using a 1 mL pipette. Remove adhered mammalian cells, and bound yeast by thoroughly scraping the plate surface using a cell scraper. Pipette all liquid and place into a 1.7 mL tube. 8. Plate dilutions of the final yeast population on SD-CAA plates to quantify yield (see Note 13). Culture final population in SD-CAA overnight at 30  C with shaking.

3.3 Cell Panning with Titratable Avidity Reduction

3.3.1 Titratable Avidity Reduction

This technique is appropriate for selections from enriched pools of target specific ligands with lower binding affinity than desired. Enrichment of ligands with low nanomolar affinity, if present in the unenriched pool, can be expected after multiple rounds of cell panning with titratable avidity reduction. This protocol may not be appropriate for disulfide bonded ligands (e.g., antibody fragments) as it relies on reduction with DTT. For these ligands, avidity can be reduced enzymatically through digest with factor Xa protease in place of the DTT reaction. For preparation of target-expressing mammalian cells and yeast, follow Subheadings 3.1.1 and 3.1.2, respectively. 1. Prepare the appropriate number of yeast in five separate tubes such that each tube contains the full desired library oversampling. 2. Pellet yeast (8000  g for 1 min) and wash twice with 10 mM Tris buffer pH 7.5. Leave yeast as a wet pellet. 3. Freshly prepare DTT solutions by dissolving DTT powder in 10 mM Tris buffer pH 7.5 and serial diluting in 10 mM Tris buffer pH 7.5, yielding 2.5, 5, 7, and 9 mM DTT (see Note 14).

314

Lawrence A. Stern et al.

4. Resuspend four yeast library samples in each of the four DTT solutions to a concentration of 5  106 yeast/20 μL DTT solution. Resuspend the fifth yeast library sample in 10 mM Tris buffer pH 7.5 without DTT at the same cell concentration as the DTT samples. 5. Incubate yeast for 20 min in a 30  C static incubator. 6. Pellet yeast (8000  g for 1 min) and wash twice with PBSACM. 7. Resuspend yeast in 1 mL PBSACM. Remove 20 μL yeast from each sample for assessment of ligand expression by flow cytometry. 3.3.2 Assessment of Ligand Expression by Flow Cytometry

Labeling of yeast to assess ligand expression should be completed concurrently with step 3.3.3. 1. In addition to the yeast recovered from Subheading 3.1.2, prepare a sample of 1  105 yeast that are not treated with DTT but are also not labeled with mouse anti-c-Myc antibody (secondary only control) by pelleting yeast (8000  g for 1 min) and washing once with 1 mL PBSACM. 2. Pellet yeast and aspirate the supernatant. 3. Resuspend yeast with 20 μL of mouse anti-c-Myc antibody (diluted 1:100, 5 μg/mL final concentration). 4. Incubate for 20 min at room temperature. 5. Pellet yeast, aspirate supernatant, and wash with 1 mL PBSACM. 6. Resuspend yeast with 20 μL of goat anti-mouse Alexa Fluor 647 antibody (diluted 1:1000, 10 μg/mL final concentration). 7. Incubate for 15 min at room temperature in the dark. 8. Pellet yeast, aspirate supernatant, and wash with 1 mL PBSACM. 9. Resuspend yeast in 200 μL of PBSACM, and analyze fluorescence of at least 10,000 events using an appropriate flow cytometer.

3.3.3 Generating a Fluorescence Calibration Curve to Assess Ligand Expression

Labeling of the Quantum Simply Cellular anti-mouse IgG beads should be performed concurrently with labeling of yeast in Subheading 3.3.2. 1. In a single tube, pool 5 μL each of Beads B, Beads 1, Beads 2, Beads 3, and Beads 4. Add 1 mL PBSACM to the bead pool. Pellet the beads (2500  g for 2.5 min) and carefully aspirate the supernatant. 2. Resuspend beads in 20 μL mouse anti-c-Myc antibody (diluted 1:100, 5 μg/mL final concentration).

Cell Panning Ligand Selections with Yeast Display

315

3. Incubate for 20 min at room temperature. 4. Pellet the beads, aspirate the supernatant, and wash with 1 mL PBSACM. 5. Resuspend beads in 20 μL of goat anti-mouse Alexa Fluor 647 (diluted 1:1000, 10 μg/mL final concentration). 6. Incubate for 15 min at room temperature in the dark. 7. Pellet the beads, aspirate the supernatant, and wash with 1 mL PBSACM. 8. Resuspend the beads in 100 μL of PBSACM, and analyze fluorescence of at least 5000 events using the appropriate flow cytometer. 3.3.4 Analysis of Ligand Expression Data

The fluorescence standard curve and absolute quantification of ligand expression are generated using a Microsoft Excel spreadsheet specific to each individual bead lot provided by the manufacturer of the Quantum Simply Cellular anti-mouse IgG kit. 1. With flow cytometry analysis software, generate a dot plot for the Quantum Simply Cellular beads of forward scatter (FSC) vs. side scatter (SSC). Draw a gate around the singlet beads (the most abundant population) as in Fig. 6b. 2. Apply this gate to a histogram of Alexa Fluor 647 fluorescence. 3. Using a histogram gate, assess the median fluorescence of each individual peak as in Fig. 6c. 4. Insert each individual fluorescence value into the spreadsheet corresponding to each individual bead population. This will generate the standard curve. The R2 value should be at least 0.99 to continue. 5. Generate a density plot for the secondary-only control yeast sample of FSC vs. SSC. Draw a gate around the most abundant population (corresponding to singlet yeast) as in Fig. 6d. 6. Apply this gate to a histogram of Alexa Fluor 647 fluorescence as in Fig. 6e. 7. Assess the median fluorescence of all yeast. 8. Repeat this process for each of the treated yeast samples. 9. In an individual row of the spreadsheet, enter the median fluorescence of a single yeast sample, subtracting the median fluorescence of the secondary-only control. The spreadsheet will automatically generate an absolute ligand expression level for each treated sample. 10. Continue to cell selection with a yeast sample expressing 3000–6000 ligands/cell following the protocol from Subheading 3.1.3.

316

Lawrence A. Stern et al.

Fig. 6 Titratable avidity reduction. Yeast expressing a ligand library are treated with DTT to decrease the number of ligands per cell. The avidity-reduced library is sorted by cell panning on target-expressing mammalian cell monolayers (a). In order to quantify ligand expression on yeast cells, a fluorescence calibration curve is generated by Quantum Simply Cellular beads. Gating on the most abundant population in FSC vs. SSC yields singlet beads (b). The beads show five fluorescence peaks, corresponding to four different quantities of antibody binding sites and negative control beads (c). Entering the median fluorescences into the manufacturer’s spreadsheet yields a calibration curve for converting sample fluorescence to the number of antibody binding sites (d). Yeast fluorescence is quantified first by gating on the most abundant population in FSC vs. SSC, yielding singlets (e). Examples of fluorescence displayed by secondary-only control yeast (black), untreated yeast (red), and yeast treated with increasing levels of DTT (green and blue) are shown (f)

Cell Panning Ligand Selections with Yeast Display

4

317

Notes 1. Appropriate growth medium will depend upon the cell line. Please refer to the medium formula recommended by your cell line provider for optimal results. 2. All reagents must be sterile filtered (0.2 μm) into autoclaved glassware prior to use. 3. The number of yeast used in a selection should be at least tenfold of the library diversity for all steps. Reference [50] gives equations to calculate the amount of oversampling needed to expect to screen a percentage of your library as well as the amount of oversampling required to have a degree of certainty that you have sampled all possible variants. Briefly, threefold oversampling is required to expect to sample 95% of your initial library, while roughly fivefold oversampling is required to expect to sample 99% of your initial library. Small libraries (107 variants) can be grown in 50–100 mL SD-CAA in 250 mL baffled flasks. 4. Culture plates can be treated with poly-L-lysine prior to seeding to encourage mammalian cell adhesion without significantly increasing the background binding of yeast cells. 5. The yield of bound yeast is strongly correlated to target expression on the cell surface in this assay. Expression levels greater than 104 targets per cell are recommended to see positive enrichment of binding yeast. 6. Depending on ligand stability, induced yeast can be stored for multiple weeks. 7. Aspiration is conducted by tilting the plate until liquid pools on one side of the well. The liquid is slowly drawn into the serological pipette to minimize shear on the mammalian cell monolayer and prevent the removal of adhered mammalian cells. 8. A single tilt is conducted by rotating the plate around its middle axis approximately 30 from level in either direction. 9. In cases where mammalian cell lines fail to remain adhered, gentler pipetting and tilting may reduce the shear experienced by the cell surface and lower the probability of cell detachment. Poly-L-Lysine treatment of culture plates may enhance cell attachment while not significantly affecting the background retention of nonbinding yeast. 10. Fresh tips must be used when drawing clean buffer, and tips must be changed between cell lines or yeast libraries when conducting parallel selections to minimize crosscontamination.

318

Lawrence A. Stern et al.

11. When adding liquid to the wells, place plate on a level surface, and put the pipette tip against the well wall. Slowly inject liquid to minimize shear on the mammalian cell surface. 12. A single nutation is conducted similar to the motion of a laboratory rocker, causing the liquid to gently swirl along the wall of the plate. 13. Calculate dilutions based upon the expected yield of your sort (on the order of 0.01–0.1% for initial sorting of naı¨ve libraries and up to the order of 1–10% depending upon the expression level and average affinity of your library) so that colonies are countable (1011 (~0.4 fmol) of any one compound for robust detection. Genetic methods allow amplification of as little as a single DNA molecule, and next-generation sequencing (NGS) facilitates differentiation of every molecule present in the solution. Cell surface display uses cells modified to express proteins of interest tethered to the cell surface (“displayed”). Displayed peptides are screened for target binding, and their cells are typically sorted by fluorescence (FACS) or magnetic (MACS) characteristics. As one collects cells displaying the peptide, the coding DNA is also collected, allowing straightforward candidate identification. Binder screens by cell sorting are simple, cost-effective, and can be coupled with natural or artificial evolutionary processes to further refine the compound’s activity [30–32]. Yeast (Saccharomyces cerevisiae) was the first eukaryote used for surface display [33], and it has been a workhorse, producing peptides (including CDPs) of clinical development interest. To perform a cell surface display screen of a CDP library, we explored an alternative approach (Fig. 1). Yeast can indeed display folded CDPs with proper disulfide topology (cysteine–cysteine interaction pattern), but only a handful of individual CDPs have been successfully used as a scaffold for diversity screening [34]. The yeast secretory pathway does not natively process many proteins with cysteine-rich regions (10 mL volume) is at 125 rpm. Suspension 24-well plate growth (0.5–8 mL volume) is at 300 rpm (see Note 1). 3. Static tissue culture incubator. 4. Biosafety cabinet for tissue culture work. 5. Electroporator and 1 mm gap-width cuvettes. We use a BTX Gemini X2 Electroporation System (BTX 45-2040). 6. Thermocycler for PCR and cloning reactions.

368

Zachary R. Crook et al.

7. Access to a Miltenyi autoMACS® Pro Separator with two autoMACS® Columns equipped (for high-diversity magnetic sorting). 8. Access to flow cytometry equipment (sorter and analyzer) capable of visualizing DAPI, GFP, and your chosen co-stain fluorophore (we recommend one visible in the APC channel). We use a BD FACSAria™ II System for sorting and an Acea NovoCyte for flow analysis. 9. Agarose gel electrophoresis setup with blue light transilluminator and SYBR™ Safe DNA Gel Stain (Thermo Fisher S33102) (see Note 2). 10. DNA quantitation spectrophotometer or fluorometer, for example, NanoDrop™ (Thermo Fisher). Optional for high sensitivity/accuracy: Qubit™ Fluorometer (Invitrogen). 11. Access to an NGS Sequencer; we use an Illumina HiSeq 2500 instrument set on Rapid Run mode and using conventional Illumina sequencing primers. 12. Benchtop centrifuge compatible with Eppendorf tubes. 13. Centrifuge (benchtop or floor) compatible with 15 and 50 mL conical tubes, as well as cultures of up to 1 L. 14. Cell strainers (35 μm Corning™ 352235 for flow sorting, 100 μm Corning™ 431752 for MACS). 2.2

Software

These programs are optional and alternatives are available. For DNAWorks, Bowtie 2, and Barcode Splitter, familiarity with basic command line scripting will be required; consult your institution’s scientific computing staff for assistance if necessary. Because their use is not essential to a screening campaign, we have not included detailed code or command line tools for their use in this protocol, but such tools and/or advice for use are available from the authors upon request. 1. DNAWorks (NIH, https://hpcwebapps.cit.nih.gov/ dnaworks/): used for reverse translation of peptide sequences and for Assembly PCR primer design. 2. Bowtie 2 ([36], http://bowtie-bio.sourceforge.net/bowtie2/ index.shtml): used for NGS data processing from library sorting experiments. 3. Barcode Splitter, a component of the FASTX-Toolkit (Hannon Lab at CSHL, http://hannonlab.cshl.edu/fastx_toolkit/down load.html): used to de-multiplex pooled NGS data. 4. Microsoft Excel: used for routine arithmetic calculation of enrichment scores. 5. Flow cytometry data processing software (e.g., FlowJo or BD FACSDiva™. We use FlowJo v. 10).

Mammalian Display Screening of Cystine-Dense Peptides

2.3 Plasmids and Cell Lines

369

1. 293T cells. 2. FreeStyle™ 293F cells (Thermo Fisher R79007), a suspensionadapted variant of 293T. 3. FreeStyle™ 293 Expression Medium (Invitrogen 12338018). No additives are necessary. 4. Mammalian surface display vector SDGF: This mammalian expression vector was derived from the Daedalus [37] vector for membrane-tethered protein expression and optimized for CDP display. Displayed proteins are introduced through BamHI/NotI cloning sites. It contains an ampR resistance marker for bacterial growth. Available from the authors upon request. Expression cassette sequence was deposited to GenBank under accession number MF958494. 5. Lentiviral packaging and VSV-G envelope vectors: we use psPAX2 (Addgene 12260) and pMD2.G (Addgene 12259). 6. Chemically competent E. coli; we use Stellar Competent Cells (Clontech). 7. Electrocompetent E. coli; we use NEB® 10-beta Electrocompetent E. coli (NEB C3020K).

2.4 Molecular Tools for Working with Peptide Libraries

If available, we recommend using a PCR hood or a dedicated room for preparing reaction buffers that is not contaminated with purified SDGF insert DNA. Always include a negative control reaction. 1. Pooled oligonucleotide libraries: we use Twist Bioscience for our vendor, allowing library sizes of up to 10,000 peptides per pool at oligo lengths of up to 200 bases. 2. PCR primers for SDGF insert amplification, cloning, and NGS sequencing (see Table 1): we use Integrated DNA Technologies for our vendor. 3. Restriction enzymes BamHI and NotI. 4. Reagents for cloning PCR-amplified peptide sequences into linearized vectors. We use NEB Gibson Assembly® Master Mix (NEB E2611L). 5. Phusion® High-Fidelity PCR Master Mix with HF Buffer (NEB M0531L). 6. Terra™ PCR Direct Polymerase Mix (Takara 639271). 7. GeneMorph II Random Mutagenesis Kit (Agilent 200550). 8. QIAquick Gel Extraction Kit (QIAgen 28706). 9. QIAquick PCR Purification Kit (QIAgen 28106). 10. QIAprep Spin Miniprep Kit (QIAgen 27106). 11. HiSpeed Plasmid Maxi Kit (QIAgen 12662).

AACCACTACCTGAGCACCCAG

TAA[NNN]xTGCGGCCGCTCATCACCATTA (see Note 3)

Downstream SDGF flanking

AATGGTGATGATTAATGGTGATGAGCGGCC

SDGF Gibson reverse

CAAGCAGAAGACGGCATACGAGATGTGACTGGAGTTCAGACGTGTGCTCTTCCGATCTAA TGGTGATGAGCGGCC

SDGF Illumina reverse

GCCAAT TAGCTT GGCTAC TTAGGC GATCAG GAGTGG AGTCAA AGTTCC

barcode_01

barcode_02

barcode_03

barcode_04

barcode_05

barcode_06

barcode_07

barcode_08

6 bp barcodes for SDGF Illumina forward primer

AATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACACGACGCTCTTCCGATCT-[6 bp barcode]-TGTACTTCCAGGGAGGATCC

SDGF Illumina forward

PCR primers for library amplification for Illumina NGS

GCCGAAAACCTGTACTTCCAGGGAGGATCC

SDGF Gibson forward

PCR primers for SDGF insert cloning by Gibson assembly and general amplification

TGTACTTCCAGGGAGGATCC

Upstream SDGF flanking

Flanking DNA to append to SDGF insert constructs

SDGF GFP seqF

SDGF insert sequencing primers

Table 1 Relevant DNA sequences (flanking and primer)

370 Zachary R. Crook et al.

ACAGTG CTTGTA CAGATC CGTACG GTCCGC TGACCA ACTTGA ATCACG CGATGT ATGTCA CCGTCC GTGAAA

barcode_09

barcode_10

barcode_11

barcode_12

barcode_13

barcode_14

barcode_15

barcode_16

barcode_17

barcode_18

barcode_19

barcode_20

Mammalian Display Screening of Cystine-Dense Peptides 371

372

Zachary R. Crook et al.

12. LB agar plates with carbenicillin (100 μg/mL): one or two 25
25 cm square plates per library or sublibrary, plus many 10 cm petri dishes for singleton cloning. 13. Liquid LB media with carbenicillin (100 μg/mL). 14. Cell scrapers, spreaders, or lifters for scraping large LB agar plates. 15. Optional: Qubit™ dsDNA HS Assay Kit (Thermo Fisher Q32854). 16. Transfection reagents for 293T cells; we use polyethyleneimine (PEI), linear 25 kDa (Polysciences 23966-1). 17. Optional: sheared salmon sperm DNA for SDGF dilution, if desired for lentivector production. 18. Molecular biology-grade water or Tris–EDTA buffer. 2.5 Target Protein Biotinylation and Cell Sorting

1. BirA biotin–protein ligase standard reaction kit (Avidity, BirA500). 2. Amicon Centrifugal Filter Units; volume and kilodalton cutoff specifications will depend on your target protein and the desired scale of your biotinylation reaction. 3. Glycerol.

2.6 Cell Sorting by MACS and Flow Cytometry

Anti-Biotin MicroBeads UltraPure (Miltenyi, 130-105-637), one bottle (2 mL) per MACS screen. 1. Dye-labeled streptavidin: we use Streptavidin Alexa Fluor™ 647 conjugate (Thermo Fisher S21374) (see Note 4). 2. Dye-labeled anti-6
His antibody: we use THE™ His Tag Antibody [iFluor 647] (GenScript A01802) (see Note 4). 3. Phosphate buffered saline (PBS), pH 7.4. 4. Flow buffer: PBS containing 0.5% bovine serum albumin (BSA) and 2 mM EDTA. 5. High BSA flow buffer: PBS containing 3% BSA and 2 mM EDTA. 6. MACS wash buffer: PBS containing 2 mM EDTA. 7. DAPI; 1000
solution is 1 mg/mL in water. 8. Your target protein, preferably 6
His tagged and biotinylated. The amount will depend on chosen sorting concentration, but no more than 10 nmol is typically required for a full campaign. For a 50 kDa target protein, this is 0.5 mg. 9. Optional: a control protein, labeled identically as your target protein but not used for enrichment screening. 10. Dithiothreitol (DTT). 11. Reduced glutathione (GSH).

Mammalian Display Screening of Cystine-Dense Peptides

3

373

Methods

3.1 293F Cell Culture Conditions

293F cells are extremely robust and double rapidly (approximately every 24 h). The standard growth conditions use FreeStyle™ serum-free medium without supplementation, in filter-top, baffled flasks at a volume between one-tenth and one-third of the maximum size of the flask. Cells are often passaged at densities between 0.25 and two million cells/mL and typically begin dying once they reach ~5 million cells/mL.

3.2 Library Design for Optimized Cloning into SDGF

The success of screening approaches depends on the sensitivity of the selection and the diversity of the library. For the latter, the use of both magnetic and flow sorting allows for ~109 cells to be processed for routine screening efforts. This necessitates transduction of one-eighth of that amount, or >1.2
108 cells, to account for the necessary 3 days required for optimal construct expression and a ~24 h doubling time. If your primary interest is in a limited number of peptide scaffolds, library diversity can be achieved through saturation mutagenesis (e.g., NNK). However, if a more diverse scaffold library is to be studied, an economical approach is to winnow your library to 10,000 members, order as an oligonucleotide pool, and further diversify it by error-prone PCR mutagenesis [38]. For libraries diversified by error-prone PCR mutagenesis, the total library size (i.e., the number of library variants) will be heavily dependent on the degree of mutagenesis imposed. CDP libraries have the additional complication that both stop codons and alteration of the cysteine topology can render a peptide nonfunctional. We favor lower mutagenesis rates (1–1.5 per peptide), as our library CDPs typically have 6–8 cysteines and are more vulnerable to cysteine topology disruption, but this should be optimized to your library. 1. Determine your desired peptide library. CDPs can come from native sources by identifying cysteine-rich sequences in existing databases (e.g., Uniprot) or can be derived from in silico docking simulations using protein design software (e.g., Rosetta) [39]. Take care not to include library members whose complete oligonucleotide sequence would exceed the maximum length available from your vendor (see Note 5). 2. Convert the peptide sequences to cloning-ready oligonucleotides (Fig. 3). (a) Reverse-translate the peptide sequences, using human codon-optimization (see Note 6). (b) To each sequence, incorporate uniform flanking sequences (Table 1) to facilitate PCR amplification and cloning (see Note 7).

374

Zachary R. Crook et al.

Fig. 3 Oligonucleotide design strategy for cloning into SDGF. An example peptide (chlorotoxin) sequence is reverse-translated using human codon optimization. Subsequently, adapter sequences for compatibility with SDGF cloning (bold font) are appended to the 50 and 30 ends. The 30 adapter contains a site for stuffer addition (NNN∗), indicating that random 3-mers of bases can be added here to bring oligonucleotides encoding shorter peptides up to a uniform size

(c) If NGS is to be employed for library characterization, verify that each library member contains a unique sequence within the entirety of the NGS read. Introduce silent mutations as necessary to ensure unique mapping is possible (see Note 8). (d) Order the finalized library from a pooled oligonucleotide vendor. 3.3 Cloning Pooled Oligonucleotide Libraries into SDGF

1. Resuspend the pooled oligonucleotide library in 50 μL molecular biology-grade water or Tris–EDTA buffer (see Note 9). 2. Prepare four reactions to PCR amplify the library using Phusion® High-Fidelity PCR Master Mix with HF Buffer and cloning primers (Table 1). 2
Phusion® Master Mix

30 μL/reaction

Primers SDGF Gibson F + R

500 nM final each; 2.4 μL/reaction of 12.5 μM stock

Oligonucleotide library DNA

1 μL/reaction (from 50 μL library resuspension)

Molecular biology-grade To 60 μL/reaction H 2O

Mammalian Display Screening of Cystine-Dense Peptides

375

Fig. 4 Agarose gel with PCR products for cloning of pooled oligonucleotide libraries into SDGF as per Subheading 3.3. Lane 1 (L) contains ladder. Lanes 2–5 contain the reaction run at 14, 16, 18, or 20 cycles. Lane 6 (Neg) contains a reaction run for 20 cycles without a template. The black arrow indicates the mobility of the desired cloning product, of ~210 bp. The white arrow shows the mobility of primers, while the red arrow shows the undesired byproduct of amplicon cross-priming. Notice the appearance of this byproduct by 18 cycles, as the primer band has disappeared. After two more (20) cycles, the byproduct is more abundant, while the band containing the properly amplified product is beginning to disappear. The band indicated by the cyan arrowhead is what was cut out of the gel and purified for the cloning reaction

Perform amplification, running the four reactions at different cycle numbers (e.g., 14, 16, 18, and 20) (see Notes 10 and 11). 1 cycle:

10 min at 98  C

20 cycles:

15 s at 95  C 20 s at 60  C 1 min at 72  C

1 cycle:

10 min at 72  C

3. Run all PCR products on an agarose gel (Fig. 4) and extract the expected band from a lane lacking the inappropriate, high molecular weight amplicon smear above the desired product (a result of PCR product cross-priming when primers are depleted). Purify using a QIAquick Gel Extraction Kit (see Note 12). 4. If your library would benefit from further diversification, you can perform mutagenesis by error-prone PCR using a GeneMorph II Random Mutagenesis Kit and the SDGF Gibson primers. Use manufacturer-recommended reaction conditions, with the PCR product from the previous step as the template.

376

Zachary R. Crook et al.

Run on a gel and purify the desired band from a lane without a substantial cross-priming smear, as above (see Note 13). 5. Prepare a sufficient amount of BamHI/NotI-digested SDGF backbone. We recommend 2 μg per library and 100 ng per singleton to be cloned. 6. Clone your amplicon into SDGF using your methodology of choice. If performing Gibson assembly using NEB Gibson Assembly® Master Mix, prepare your reaction as follows: 2
Gibson Assembly® Master Mix

40 μL

BamHI/NotI digested SDGF backbone

2 μg

Library PCR product

200 ng

Molecular biology-grade water

To 80 μL

Incubate the Gibson reaction for 2 h at 50  C instead of the recommended 1 h. 7. De-salt the reaction using a QIAquick PCR Purification Kit, eluting in a reduced volume of 20 μL. 8. Electroporate your library into competent E. coli, and incubate at 37  C overnight. Use specialized electrocompetent cells such as NEB® 10-beta Electrocompetent E. coli. The conditions are as directed by your instrumentation for E. coli using 1 μL purified assembly product and 50 μL cells in a 1 mm cuvette. Plate electroporated cells onto a large plate (25
25 cm) of LB agar + carbenicillin, taking care to spread completely and letting it dry completely before inversion and overnight 37  C incubation. Be sure to include control petri dishes containing serial dilutions of the library to establish library CFUs (see Note 14). 9. Following morning, count your control plates to establish CFUs. If sufficient, scrape the plate(s) by adding 25–50 mL LB media plus carbenicillin (100 μg/mL) and scraping with a cell scraper, lifter, or spreader. Repeat with more media and scraping until transfer is complete. 10. Shake bacteria in a final volume of 200 mL for 1 h at 37  C, and then pellet 50 mL aliquots (4000
g for 20 min). 11. Freeze all but one pellet at 20  C, and maxiprep the remaining pellet using the HiSpeed Plasmid Maxi Kit. We recommend sterile filtering your final DNA product prior to lentiviral production. 3.4 Target Protein Design and Labeling

Target protein, labeled with an appropriate affinity tag, can be purchased from a commercial vendor or produced in-house. If making a custom target protein, we recommend using a

Mammalian Display Screening of Cystine-Dense Peptides

377

“His-Avi” tag (HHHHHHGGGSLNDIFEAQKIEWHE), which combines a 6
His affinity tag with an AviTag recognition sequence (LNDIFEAQKIEWHE) separated by a short linker. BirA enzyme, found in the BirA biotin–protein ligase standard reaction kit, specifically biotinylates proteins at the AviTag sequence. Such proteins should be biotinylated as per manufacturer’s recommendations (see Note 15). Afterwards, free biotin must be removed, and hence, we use spin-column dialysis using Amicon Centrifugal Filter Units, but a number of alternative methods can be applied. 3.5 Lentivector Generation and Transduction for Mammalian Display SDGF Screening

Lentivirus for transduction of SDGF into suspension 293F cells is prepared using conventional methodology in 293T cells [40] or your chosen protocols; a sample protocol is described below. We use psPAX2 and pMD2.G vectors for packaging and envelope, respectively, but any VSV-G pseudotyped lentiviral system will work. We also recommend PEI as an economical transfection reagent, although others of your choice can be substituted. However, be advised that the SDGF vector has shown substantial toxicity in the context of lentiviral production. For this reason, it is necessary to either use 20% of the typical amount of vector DNA or dilute the SDGF plasmid pool fivefold with carrier DNA (e.g., sheared salmon sperm DNA or similar) prior to inclusion in the transfection mix (see Note 16). For example, a 10-cm dish containing 60% confluent 293T cells can be transfected with 3.2 μg carrier DNA, 0.8 μg SDGF, 2 μg psPAX2, and 1 μg pMD2.G using 10 μg PEI. After changing media 24 h posttransfection, viral supernatant can be harvested 48 h posttransfection. For executing the screens, the library must be transduced into suspension 293 Freestyle™ (293F) cells. The multiplicity of infection (MOI) of this transduction can be varied, and there are advantages and disadvantages for either direction. We typically use an MOI of 1 because it ensures that >60% of our cells are useful for screening, but the risk of vector toxicity and need for deconvolution of hits from cells infected by multiple vectors is manageable. That said, this can be optimized for your own uses. Our one caution for a high MOI screen is that you should establish beforehand, using positive control constructs and increasing amounts of negative constructs, that a hit will be detectable in the context of a high MOI transduction.

3.6 Cell Staining for a Large-Scale, High-Diversity Screen Using an autoMACS Pro Sorter

For high-diversity library screening, magnetic cell sorting (MACS) is employed. Flow cytometry is more sensitive, but it has throughput limitations (100,000 variants, or anything diversified by error prone PCR, so we can screen a billion cells or more. The below protocol requires biotinylated target protein. If your target protein carries an alternative affinity

378

Zachary R. Crook et al.

tag, consult with the Miltenyi catalog of magnetic microbeads for one capable of coupling with your target protein. 1. Transduce high-diversity SDGF library into 200 million 293F cells at your desired MOI; we use an MOI of 1 (see Note 17). 2. On the day of the sort (3–4 days post-transduction), prepare MACS stain solution as follows: Flow buffer

To 23 mL

Antibiotin microbeads

2 mL

Biotinylated target protein

Up to 5 nmol (200 nM)

3. Collect and pellet 109 transduced cells (500
g for 10 min). 4. Resuspend the pelleted cells in the MACS stain solution, and incubate on ice for 30 min, agitating often to prevent settling of cells (see Note 18). 5. After incubation, dilute the cells tenfold (225 mL flow buffer), and pellet again. 6. Remove the diluted stain solution and resuspend stained cells in 38 mL High BSA Flow Buffer and filter through a cell strainer (100 μm). The final volume with pellet will be ~40 mL. 7. Proceed with MACS on a Miltenyi autoMACS® Pro Separator. Make sure two autoMACS® Columns are equipped. Perform four separate 10 mL sorts in the “Chill 15” rack. Use the “posseld” protocol and “quick rinses” after each sort. The running and wash buffers should be High BSA flow buffer, and PBS with 2 mM EDTA, respectively (see Note 19). 8. Sorted cells are pooled and pelleted (500
g for 5 min). (a) If planning to grow the cells out for further screening at medium scale (Subheading 3.7), resuspend the pellet in 0.5 mL FreeStyle® media and grow in a 24-well plate, expanding as cells grow. (b) For sub-library production (Direct PCR Subheadings 3.8, steps 1–5, followed by library cloning Subheading 3.3, steps 6–11), remove as much supernatant as possible from the pellet prior to storage at 20  C (see Note 20). 3.7 Medium-Scale FACS Staining for Low Diversity (2 million cells/ mL) and then eventually move to a baffled flask. 13. When sorted cells reach a population of tens of millions, they can be banked and/or screened against the target again. If pelleting for Direct PCR (Subheading 3.8), aliquot 1–2 million cells and pellet (500
g for 5 min). Remove as much supernatant as possible prior to storage at 20  C. 14. For routine screening campaigns, multiple rounds of flow sorting (beginning at step 2 of this section) are necessary to identify rare target-binding variants in the population (Fig. 6). When our campaigns are successful at identifying specific target binders, they generally appear as a distinct clonal population in the GFP vs. APC flow profile no later than the fourth round of sorting. However, this may vary depending on the diversity of your library and the stringency of your sorting criteria. If your target protein contains multiple affinity tags (e.g., a His-Avi tag), we recommend alternating the co-stain every round of sorting. This reduces the impact of nonspecific binders becoming enriched because they bind to the co-stain (streptavidin or anti-6
His). 3.8 Hit Identification by Conventional Sanger and NextGeneration Sequencing

Cell pellets are used as input for a two-step direct PCR protocol, resulting in amplified library for use in conventional Sanger sequencing or NGS sequencing (e.g., Illumina HiSeq) (see Note 27). 1. Prepare direct PCR and secondary PCR buffers. (a) Direct PCRs are as follows: 2
Terra™ reaction buffer

25 μL/reaction

Primers SDGF Gibson F + R

500 nM final each; 2 μL/reaction of 12.5 μM stock

Terra™ polymerase

1 μL/reaction

Molecular biology-grade To 50 μL/reaction water

One 50-μL reaction is required per cell pellet, plus an additional 50-μL reaction to be processed in parallel as a

382

Zachary R. Crook et al.

Fig. 6 Typical GFP vs. co-stain flow profiles through a screening campaign. (a) First flow plot shows cells that have already been sorted by one round of MACS (anti-biotin microbeads) and one round of flow sorting (streptavidin co-stain). Two more sorts, one with a streptavidin co-stain and the other with an anti-6
His co-stain, are then applied, at which point a population of likely target binders (black arrow) has clearly emerged. (b) Control stains of the material after MACS and three flow sorts (a, right) using a control protein and either streptavidin or anti-6
His co-stains are shown. Indicated with a magenta arrow, some nonspecific streptavidin-binding cells may be appearing (b, left), while a nonspecific anti-6
His binder is almost certainly present (b, right). This highlights the utility of alternating co-stains throughout the campaign, as it blunts the enrichment of peptides that bind to co-stain instead of the target. In this example, campaign, the apparent antibody binder, might have overtaken the population if anti-6
His were the co-stain used in every sort

negative control. We use our standard SDGF Gibson forward and reverse primers in the Direct PCR (Table 1). (b) Secondary PCRs are as follows: 2
Phusion® Master Mix

30 μL/reaction

PrimersSDGF Gibson F + R– or 500 nM final each; 2.4 μL/ –SDGF Illumina F + R reaction of 12.5 μM stock Molecular biology-grade water

To 56.2 μL/reaction

Primers used for the secondary PCR are the same as those used in direct PCR unless Illumina NGS is to be performed, in which case the SDGF Illumina forward and reverse primers are used (Table 1). Prepare four secondary PCRs per direct PCR (including negative control); because over-amplifying the product can lead to self-

Mammalian Display Screening of Cystine-Dense Peptides

383

priming of amplicons and depletion of the proper product, the reactions will be subjected to varying thermocycling rounds to generate the correct amplicon for cloning. 2. Retrieve cell pellets, and to each pellet, add 50 μL direct PCR mix. Ensure the cell pellet is completely lysed by pipetting repeatedly, and then transfer the mixture into a PCR tube. Your negative control can either be reaction mix only or can use a pellet of nontransduced cells. 3. Perform the direct PCR using touchdown conditions, 16 cycles in total: 1 cycle:

10 min at 98  C

8 cycles:

15 s at 95  C 20 s at 68–61  C (reducing 1  C/cycle) 1 min at 72  C

8 cycles:

15 s at 98  C 20 s at 60  C 1 min at 72  C

4. When the direct PCR is complete, use the 3.8 μL direct PCR as the template for secondary PCRs; the final volume is 60 μL each (see Note 28). The secondary amplification conditions are as follows: 20 cycles:

15 s at 95  C 20 s at 60  C 1 min at 72  C

1 cycle:

10 min at 72  C

Using four reactions per sample, we will typically amplify for 20 cycles, pulling one reaction from the thermocycler each after 14, 16, and 18 cycles. 5. Run the secondary PCR on a 2% agarose gel and cut out the desired band from a lane with minimal cross-priming. Purify using a QIAquick Gel Extraction Kit (see Note 29). If generating a sublibrary for further screening, use this material for library production as per Subheading 3.3, steps 6–11. 6. For hit identification by conventional Sanger sequencing: (a) Re-clone this purified amplicon into BamHI/NotIdigested SDGF vector, using the same strategy as that used for initial library cloning (Subheading 3.3) at a smaller scale (100 ng SDGF backbone, 10 ng insert, 10 μL final volume Gibson reactions) and conventional heat shock transformation.

384

Zachary R. Crook et al.

(b) From plated bacterial transformants, pick 24–96 colonies, depending on how clonal your flow cytometry data appeared (and therefore how many independent entities you expect to see in your library), and grow for minipreps. (c) Submit plasmids for Sanger sequencing; hits can be identified as enriched sequences (Subheading 3.9), or you can simply test every candidate in singleton validation (see Note 30). 7. For hit identification by multiplex NGS (here, Illumina HiSeq): (a) Carefully quantitate your gel-extracted PCR products prior to pooling for multiplex sample submission, using Qubit™ or similar (see Note 31). (b) Pool reactions as desired, and submit for NGS. (c) Process the data using conventional NGS data analysis methods. If using our recommended primer design for multiplexing, de-multiplex prior to Bowtie 2 mapping using Barcode Splitter software. Adjust scoring thresholds as necessary (see Note 32). 3.9 Hit Identification and Prioritization by Sequence Characteristics

After sequencing efforts, we are left with a list of candidate peptide sequences that demonstrate enrichment in pooled surface display binding assays, which will be referred to as hits. Not all hits represent specific target binders or viable candidates for drugs or diagnostic tools, so separating relevant from irrelevant hits is as important as executing the screen itself. 1. If colony picking and conventional Sanger sequencing was performed, all sequences are already cloned into SDGF and available for validation. If time or resources are limited, we recommend focusing efforts on sequences that appear more than once among your picked colonies. 2. If NGS sequencing was performed, there are two routine methods for identifying hits of interest (both of which can be performed in Microsoft Excel or other array manipulation software); other analytical methods can be applied as desired. (a) Simple read abundance can identify peptide sequences that made up a large proportion of the enriched cell population. This is of particular utility if you have a clear clonal population appearing in your flow profiles. (b) Enrichment analysis can identify high-performing peptide sequences by normalizing to input abundance. l

Within a given sample, normalize each gene’s read count to the total read count of that sample. This gives you the relative abundance of each gene.

Mammalian Display Screening of Cystine-Dense Peptides

385

l

For each gene, calculate the fold change in relative abundance between your sorted population of interest and the input population.

l

If you have a reference gene in the set (e.g., the nonmutated allele in an affinity maturation screen), you can normalize each gene’s fold change to that of the reference gene; this allows you to discriminate between sequences that are enriched-but-not-improvements and sequences that are enriched-and-improvements.

l

These relative fold change enrichment values can be log-transformed for easier viewing, and sorted. The user can apply whatever significance thresholds they desire.

3. Identify problematic sequences and either discard or note for increased scrutiny. (a) Sequences containing indels or nonsense mutations can generally be discarded, unless the truncation is expected to be of minimal impact (e.g., at the C-terminus, and not encompassing any predicted structural elements). (b) For CDPs, alterations to cysteines (addition or removal) may alter disulfide topology, which could compromise stability; these should be viewed with particular scrutiny. (c) If the scaffold’s structure is known or can be confidently predicted, introduction of prolines in beta sheet elements can compromise the inter-chain hydrogen bond network and may present stability issues compared to the original scaffold. (d) Potential cryptic N-linked glycosites (NXS/T) should be identified. If possible, you should clone and surface express the N!Q variant of the candidate, testing for target engagement (see Note 33). 3.10 Small-Scale Surface Display Hit Validation

1. For hits identified by conventional Sanger sequencing in SDGF, skip to step 5 of this protocol. For hits identified by NGS, reverse translate singleton candidates (DNAWorks or similar software) and append upstream and downstream SDGF flanking DNA, as was done for library design (Table 1). 2. Order singleton candidates of interest from a DNA vendor for cloning into SDGF. The most straightforward method is to order the dsDNA; we use Integrated DNA Technologies gBlocks® for this purpose. Alternatively, primers for Assembly PCR [41] can be ordered. This is often more economical but includes additional molecular manipulations in the form of the assembly reaction and a second PCR step. DNAWorks can

386

Zachary R. Crook et al.

convert input DNA into primer sequences compatible with Assembly PCR. 3. Clone the constructs into BamHI/NotI digested SDGF using the same strategy as for initial library cloning (Subheading 3.3) at a smaller scale (100 ng SDGF backbone, 10 ng insert, 10 μL final volume Gibson reactions) and conventional heat shock transformation. No purification of the Gibson reaction product is necessary, and conventional heat shock transformation should be employed. 4. Pick colonies, grow in LB carbenicillin media as usual, and submit miniprepped plasmid for Sanger sequencing. 5. Transfect sequence-validated SDGF clones of interest into two million 293F cells in 1 mL FreeStyle™ media by adding 2.5 μg plasmid DNA and 4.5 μg PEI. 6. Grow transfected 293F cells for 2–4 days in suspension in 24-well plates, splitting 1:1 or expanding daily. On the day of staining (2–4 days post-transfection), prepare stain solutions as follows: Target stain solution (if testing under other conditions [e.g., reducing agents], increase the amount of target protein stain solution appropriately): Flow buffer

250 μL/sample

DAPI

0.25 μL/sample (1 mg/ mL stock)

Target protein

Varies (see Note 23)

Fluorescent co-stain (e.g., streptavidin or Varies; equimolar to target anti-6
His) protein

Control stain solution: Flow buffer

250 μL/sample

DAPI

0.25 μL/sample (1 mg/mL stock)

Control protein

Equimolar to target protein (see Note 24)

Fluorescent co-stain (e.g., streptavidin Equimolar to target protein or anti-6
His)

7. Collect your samples. Each sample should be divided for use in both stain conditions; typically, this is 0.4 mL cell suspension per condition from your transfection plate. Pellet your cells (500
g, 5 min).

Mammalian Display Screening of Cystine-Dense Peptides

387

8. Optional for CDP candidates: resuspend the cells in 500 μL PBS, PBS with 10 mM DTT, or PBS with 10 mM GSH. Incubate for 5 min at room temperature, and pellet again. This requires additional cell replicates and preparing more target stain solution (no need to test negative control protein stains against reduction-treated samples). However, it serves as validation that any binding seen is dependent on disulfides, suggesting a stable and functional CDP fold (see Note 34). 9. Resuspend each cell pellet in 225 μL stain solution, either target or control. Incubate on ice for 30 min, gently vortexing or inverting to mix every 5–10 min. 10. After incubation, a mild dilution is performed (add 675 μL of flow buffer) and the cells are pelleted. 11. Optional: if you wish to add an additional degree of binding stringency, a wash with 1 mL flow buffer can be performed at this step. 12. Resuspend each pellet in 500 μL flow buffer and proceed to flow cytometry analysis. Gating parameters should be similar or identical to that used for flow sorting (Subheading 3.7). 13. For transfected cells, there should be two populations apparent on the GFP profile: positive (GFP+) and negative (GFP). A validated hit is defined here as one whose cell staining (here, the APC signal) is higher in the GFP+ cells than in the GFP cells when incubated with target protein but not with a control protein (Fig. 7). If the vector confers no specific fluorescence under either stain condition, it was a false positive; if it stains positive in both conditions, it is likely a binder that recognizes not your target protein but instead the co-stain reagent. 14. Positive hits at this stage can be further validated in targetspecific ways (see Note 35). 3.11 Site Saturation Mutagenesis for Affinity Maturation

A site saturation mutagenesis (SSM) library consists of the original hit, plus every possible noncysteine single mutant variant. If your peptide had 40 noncysteine residues, and as there are 18 possible noncysteine substitutions, this means your library would contain your original hit and 720 (40
18) variants, or 721 unique peptides in all. Affinity maturation by SSM is nearly identical to the methodology used for the primary screen. Subheadings 3.2, 3.3, 3.5, 3.7–3.10 are employed, with the following modifications and considerations. 1. Your screening can be done via flow cytometry (Subheading 3.7), as it is a low-diversity (typically 30 ng/μl (see Note 8). 7. Ligate the hu4D5 Fab fragment insert into the pPyEBV vector: The molar insert-to-vector ratio should be 5:1. Calculate the volume of digested insert and vector required for a total DNA mass of 200 ng at this ratio. For example, for digested vector (10,673 bp) and insert (1644 bp) each at 50 ng/μl, a volume of 1.8 μl insert (87 ng) and 2.2 μl vector (113 ng) should be used. Combine the insert and vector volumes calculated, and add 2 μl 10
T4 DNA Ligase buffer, 1 μl T4 DNA Ligase, and water to a final volume of 20 μl in a PCR tube. Prepare a second tube with vector DNA only (replace the insert with water) as a control to monitor background ligation of undigested and singly digested vector if desired. Incubate the tube on a heat block or thermocycler at 16  C for 12 h, and then heat inactivate the reaction for 10 min at 65  C.

404

Annalee W. Nguyen et al.

8. Desalt the ligation by dialysis: incubate 10 μl of each ligation on a MF™ 0.025 mm VSWP filter floating on water in a petri dish for 1 h (see Note 9). 9. Transform 3 μl of each desalted ligation to 50 μl electrocompetent DH5α cells in a 1 mm electroporation cuvette at the settings recommended for bacteria with the electroporation apparatus. Immediately add 1 ml of SOC media at 37  C and grow for 1 h with shaking at 225 rpm and 37  C. 10. For each transformation, place 20 μl of recovered cells with 80 μl of SOC media on an ampicillin agar plate and spread evenly with a sterile spreader. Place the plates upside down in a 37  C incubator for overnight growth. 11. The following day, compare the vector plus insert ligation colony count to the vector alone colony count. If the ligation appears successful (at least twice as many colonies on the vector plus insert versus vector alone plates), start 12 ml cultures of several colonies in TB media with 12 μl of 100 mg/ml ampicillin (final concentration 100 μg/ml ampicillin) in 125 ml sterile, baffled flasks. Grow overnight with 225 rpm shaking at 37  C. 12. Miniprep ~2–3 ml of the overnight cultures and confirm proper cloning by sequencing. Prepare frozen stocks (0.5 ml culture and 0.5 ml 50% glycerol in cryogenic tubes placed in a 80  C freezer) and freshly streaked agar plates from the remaining culture for later cultures of DH5α containing the pPy4D5Disp vector. 13. Grow a 35-ml culture of DH5α cells containing pPy4D5Disp in TB media with 100 μg/ml ampicillin overnight with 225 rpm shaking at 37  C. 14. Midi-prep each culture according to the instructions provided in the ZymoPURE II Plasmid Midi-prep kit. The 200-μl plasmid DNA preparations should be at a concentration of ~0.5–2 μg/μl. 3.1.2 Cloning GFP Genes to Create the Reporter Plasmid pPyEGFP with Episomal Replication

1. Digest pPyEBV vector with NheI and XhoI according to the NEB double digest instructions for these enzymes: Mix 2 μg pPyEBV with 5 μl 10
CutSmart Buffer, 1 μl NheI-HF, 1 μl XhoI, and nuclease-free water to bring the reaction volume to 50 μl in a PCR tube. Incubate at 37  C for 3 h on a heat block or thermocycler block. Heat-inactivate the digestion at 65  C for 20 min. 2. Simultaneously digest 2 μg of pEGFP-C1 using the same protocol described in the previous step.

Engineering Antibodies on the Surface of CHO Cells

405

3. Continue with the cloning process described in Subheading 3.1.1 to introduce the EGFP fragment (752 bp) into the digested pPyEBV vector (10,697 bp). The product will be purified, midi-prepped plasmid pPyEGFP. 3.2 Antibody Display Optimization

For screening of surface displayed antibody libraries on CHO-T cells, there are two key concerns: restricting plasmid copy number to no more than one plasmid per cell to avoid complicated deconvolution of heavy/light chain pairs and maximizing cell health during display to ensure unbiased growth during the selection process. To address these issues, the approximate plasmid copy number per cell after transfection was calculated, and a doping plasmid ratio was determined to bias the pPy4D5Disp or pPyEGFP copy number to one plasmid per cell (Subheading 3.2.3). Then, the expression level of pPy4D5Disp was modulated with the introduction of μORFs and suboptimal Kozak sequences (Subheadings 3.2.1 and 3.2.2) to maximize display while minimizing expression stress.

3.2.1 Optimize Plasmid Copy Number in CHO-T Cells

1. Start a culture of CHO-T cells in growth media (see Note 10) from a frozen stock. Grow the cells in filter-capped flasks at 37  C and 5% CO2 in a jacketed incubator. 2. Expand the cells until there are at least >1
106 cells/ml in a T-150 flask (~30 ml) for a total of >3
107 CHO-T cells. The day before transfection, ensure the cells are healthy (>95% alive as measured by Coomassie blue staining and counting on a hemocytometer or your favorite method) and dilute to a concentration of 1
106 cells/ml for overnight growth. 3. Prepare DNA for transfection by mixing midi-prepped pPy4D5Disp (Subheading 3.1.1) and pPyEGFP (Subheading 3.1.2) at the ratios shown in Table 1 in sterile microcentrifuge tubes. Table 1 pPy4D5Disp and pPyEGFP ratios for copy number testing pPy4D5Disp Ratio

%

pPyEGFP μg

%

μg

0:1

0

0.0

100

13.8

1:1

50

6.9

50

6.9

4:1

80

11.0

20

2.8

9:1

90

12.4

10

1.4

19:1

95

13.1

5

0.7

blank

0

0.0

0

0.0

406

Annalee W. Nguyen et al.

4. Add 550 μl growth media to each of the six tubes of DNA. 5. Mix 143 μl Lipofectamine™ 2000 with 3.575 ml growth media in a sterile 15 ml conical vial. 6. Let both the DNA/media mixture and the Lipofectamine™ 2000/media mixture sit at room temperature for 5 min. 7. Add 550 μl of the Lipofectamine™ 2000/media mixture to each of the six tubes containing DNA and allow to sit at room temperature for 20 min. 8. While allowing the DNA and Lipofectamine™ 2000 to incubate, count the overnight CHO-T culture and spin down 3
107 cells in 20 ml media at 200
g for 2 min. 9. Resuspend the CHO-T cells in 19.5 ml growth media for a cell density of 1.5
106 cells/ml, and add 1.5 ml of suspended CHO-T cells to each well of two 6-well tissue culture treated plates. 10. Add 500 μl of Lipofectamine™ 2000/DNA mixture to each well such that each ratio is transfected to duplicate wells. 11. Allow the cells to grow overnight at 37  C and 5% CO2 in a jacketed incubator. 12. Spin the cells (in the wells or transfer to tubes) at 200
g for 2 min. 13. Vacuum aspirate the transfection media to waste. 14. Add 2 ml growth media per well. 15. Allow the cells to grow overnight at 37  C and 5% CO2 in a jacketed incubator. 16. Two days after transfection, transfer the contents of each well to a microcentrifuge tube. 17. Wash: Centrifuge tubes containing cells at 200
g for 2 min. Vacuum aspirate the media to waste being careful not to disturb the pellet. Resuspend each cell pellet in 1 ml PBS. Centrifuge at 200
g for 2 min. 18. Vacuum aspirate the PBS to waste being careful not to disturb the pellet. 19. Resuspend each pellet in 250 μl of Stain A. 20. Incubate at room temperature for 20 min. 21. Wash as in step 17. 22. Vacuum aspirate the PBS to waste being careful not to disturb the pellet. 23. Resuspend each pellet in 250 μl of Stain B. 24. Incubate at room temperature for 20 min. 25. Wash as in step 17.

Engineering Antibodies on the Surface of CHO Cells

407

26. Vacuum aspirate the PBS to waste being careful not to disturb the pellet. 27. Resuspend in 500 μl PBS and transfer to a tube for flow cytometry. 28. Scan at least 10,000 cells by flow cytometry with blue (488 nm) laser excitation of EGFP and a 530/30 bandpass filter for emission detection and red (640 nm) laser excitation of Alexa Fluor® 647 and a 670/30 bandpass filter for emission detection. 29. Based on the completely negative cells (no transfected DNA), draw a quadrant gate (Fig. 2) separating each transfected

Fig. 2 Determining the number of plasmids per cell after transfection. The plasmids pPyEGFP (EGFP) and pPy4D5Disp (4D5) were transfected into CHO-T cells at various EGFP:4D5 mass ratios in duplicate: a, b at 0:0; c, d at 1:0; e, f at 1:1; g, h at 1:4; i, j at 1:9, and k, l at 1:19. Cells gated for appropriate forward and side scatter were analyzed for EGFP expression (x-axis) and 4D5 Fab expression (y-axis). As shown in panels a and m, each population was evaluated for the percentage of cells appearing in quadrant I (% QI) with both EGFP and 4D5 expression, quadrant II (% QII) with only 4D5 expression, quadrant III (% QIII) without expression of either protein, or quadrant IV (% QIV) with only EGFP expression. The percent double positive (PDP) was determined for each of the populations as described

408

Annalee W. Nguyen et al.

population into those that are positive for both EGFP and Alexa Fluor® 647 (quadrant I), those that are positive for only Alexa Fluor® 647 (quadrant II), those that are not fluorescent (quadrant III), and those that are positive for only EGFP (quadrant IV). 30. The value required for statistical determination of plasmids transfected per cell (percent double positive or PDP) is the percentage of transfected cells positive for both EGFP and Alexa Fluor® 647. This is determined by dividing the percentage of cells in quadrant I by the percentage of positive cells total (100% minus percentage in quadrant III). Determine this value for each transfection. The data and PDP values we obtained are shown in Fig. 2. 31. Determine the average number of plasmids per cell by fitting the PDP to that predicted by a Poisson distribution [22] and ratiometric incorporation of pPyEGFP and pPy4D5Disp into each transfection according to the mixtures used (Table 1). With our data set, the fit gave us an average of 2.07 plasmids per cell, predicting the 1:1 mixture would be 41.7% double positive, the 1:4 mixture would be 27.5% double positive, the 1:9 mixture would be 15.8% double positive, and the 1:19 mixture would be 8.5% double positive. These values matched the data (Fig. 2) closely. 32. The Poisson distribution with an average of 2.07 plasmids per cell is then used to choose a ratio of antibody encoding plasmid to blank carrier plasmid for use with the library. This determination is at the discretion of the researcher as a very high amount of blank plasmid will reduce the number of cells with more than one antibody sequence being produced but will also greatly increase the number of cells not producing antibody at all. We chose to use 74% blank carrier plasmid and 26% antibody encoding plasmid for our library transfections, which was predicted to result in ~17% of all cells producing displayed antibody and about 75% (13% of all cells) of those containing only a single antibody encoding plasmid. 3.2.2 Clone Expression Level Variants of pPy4D5Disp

1. PCR amplify expression level variants of pPy4D5Disp with the primers listed in Table 2 and Phusion HF polymerase: On ice, mix 10 μl 5
Phusion HF Buffer, 1 μl 10 mM dNTPs, 2.5 μl 10 μM Forward Primer (4D5dispF0, 4D5dispF1, 4D5dispF2, 4D5dispF3, or 4D5dispF4), 2.5 μl 10 μM Reverse Primer (the same reverse primer, 4D5dispR, will be used for all five reactions), 0.3 μl pPy4D5Disp plasmid DNA, 0.5 μl Phusion HF Polymerase, 1.5 μl DMSO, and 31.7 μl water for each variant. The PCR cycle is as follows: Denature at 98  C for 30 s; Cycle 30 times at 98  C for 5 s, 68  C for 30 s, 72  C for 30 s; Extend

Engineering Antibodies on the Surface of CHO Cells

409

Table 2 Primers required to generate pPy4D5disp expression level variants Name

Oligonucleotide Sequence (5’ to 3’)

4D5dispF0

Aagctgggtaccgcgcgccttttttggctggtcctgcatcatcctg

4D5dispF1

aagctgggtaccgcgcgccgggatgggttgatttatgggctggtcctgcatcat

4D5dispF2

aagctgggtaccgcgcgcctaaatgggttgaaccatgggctggtcctgcatcat

4D5dispF3

Aagctgggtaccgcgcgccttgatgggctggtcctgcatcat

4D5dispF4

Aagctgggtaccgcgcgcctaaatgggctggtcctgcatcat

4D5dispR

Gacttcgcatgcgtagactttgtg

Note: Restriction sites are highlighted in gray (ggtacc ¼ KpnI; gcatgc ¼ SphI). Bold text refers to the coding regions, underlined text refers to the introduced alterations to the kozak region. F0 is predicted to result in blocked expression of the coding sequence, while F1–F4 are predicted to result in increasing expression levels

at 72  C for 5 min. Gel purify the reaction as described in step 6 of Subheading 3.1.1. 2. Digest both the pPy4D5Disp vector and the PCR products with KpnI-HF and SphI-HF, following the steps described in step 1 of Subheading 3.1.1 and continue cloning and plasmid purification as described in that section. The cut vector will be ~11,654 bp and the inserts will be ~660 bp. The resulting plasmids after cloning will be the unmodified pPy4D5Disp from Subheading 3.1.1, pPy4D5Disp0, pPy4D5Disp1, pPy4D5Disp2, pPy4D5Disp3, and pPy4D5Disp4. 3.2.3 Test Expression Level Variants of pPy4D5Disp

1. Start a culture of CHO-T cells in growth media from a frozen stock. Grow the cells in filter-capped flasks at 37  C and 5% CO2 in a jacketed incubator. 2. Expand the cells until there are >1
106 cells/ml in a T-150 flask (~30 ml) for a total of >3
107 CHO-T cells. The day before transfection, ensure the cells are >95% viable (by Coomassie blue staining and counting on a hemocytometer or your favorite method) and dilute to a concentration of 1
106 cells/ml for overnight growth. 3. Mix 10.2 μg of pPyEBV plasmid with 3.6 μg of each pPy4D5Disp variant (for a total of six mixtures) in 1.5 ml sterile microcentrifuge tubes. 4. To transfect and stain cells expressing these variants, follow steps 4–32 in Subheading 3.2.1 using these tubes of DNA. 5. Determine the percent of cells displaying hu4D5 and the mean fluorescence intensity (MFI) of the displaying population

410

Annalee W. Nguyen et al.

Fig. 3 Effect of Kozak sequence modification on Fab display. Each mORF/Kozak sequence variant was transfected into CHO-T cells in duplicate: a, b, pPy4D5Disp0; c, d, pPy4D5Disp1; e, f, pPy4D5Disp2; g, h, pPy4D5Disp3; i, j, pPy4D5Disp4; k, l, pPy4D5Disp. Two days after transfection, each cell population was stained for 4D5 Fab binding to HER2 with HER2-Fc, followed by goat-anti-human Fc-AF647, and each population was analyzed by flow cytometry. Only pPy4D5Disp0 (a, b) and 1 (c, d) had substantially decreased expression levels. Of those with high expression, pPy4D5Disp2 (e, f) had the lowest predicted transcriptional load and highest detectable display compared to unmodified pPy4D5Disp (k, l)

(Fig. 3). In this case, the pPy4D5Disp2 variant has equivalent expression to unmodified pPy4D5Disp and was noted to qualitatively have more consistent and healthy growth, presumably due to lower antibody expression and display load. The μORF and Kozak sequenced included in the pPy4D5Disp2 construct was used for all subsequent experiments. 3.3 bD1 Library Generation

This method closely follows that outlined in Bessette et al. [23] to generate a bD1 library. The bD1 variant has reduced ligand affinity relative to hu4D5 by virtue of amino acid changes in the VL region only. To avoid recovery of the hu4D5 sequence, the library was created with mutations in VH only.

3.3.1 Clone pPybD1Disp

1. Repeat the cloning process described in Subheading 3.1.1 with the bD1 Fab gene fragment (Supplementary Fig. 1) and pPyEBV. The product will be the plasmid pPybD1Disp. This plasmid should be midi-prepped to a concentration of ~1 μg/μl or greater.

3.3.2 Make bD1 Variant Library

1. Add 2 μl of each 100 μM VH library oligos (Table 3) to a microcentrifuge tube and mix by pipetting up and down. 2. On ice, mix 20 μl 5
Phusion HF Buffer, 2 μl 10 mM dNTPs, 2 μl oligo mixture (from step 1), 1 μl Phusion HF Polymerase, 3 μl DMSO, and 72 μl water. Assemble the VH library by PCR cycling with the following conditions: Denature at 98  C for 30 s; Cycle 30 times at 98  C for 5 s, 52  C for 30 s, 72  C for 30 s; Extend at 72  C for 5 min.

Engineering Antibodies on the Surface of CHO Cells

411

Table 3 Oligonucleotides used for the assembly of the bD1 VH library Name

DNA sequence

HCLibF1

CTCAACTTCGACCTGCTGAAGCTTGCCGGCGACGTGGAGT

HCLibF2

CTAACCCTGGCCCTGAAGTGCAGCTGGTGGAATCTGGCGG

HCLibF3

CGGACTGGTGCAGCCTGGCGGATCTCTGAGACTGTCTTGT

HCLibF4

GCCGCCTCCGGCTTCAACATCAAGGACACCTACATCCACT

HCLibF5

GGGTGCGACAGGCCCCTGGCAAGGGACTGGAATGGGTGGC

HCLibF6

CAG M ATCT M TCCC WM TA R TGGCTACACCAGATACGCCGAC

HCLibF7

TCTGTGAAGGGCCGGTTCACCATCTCCGCCGACACCTCCA

HCLibF8

AGAATACCGCCTACCTGCAGATGAACTCCCTGAGAGCCGA

HCLibF9

GGATACCGCCGTGTACTACTGCTCCAGA BDK GGAGGC RV T

HCLibF10

GGCT NK TAC KV T NBK GACTATTGGGGCCAGGGAACCCTCG

HCLibF11

TGACCGTGTCCTCTGCGTCGACCAAGGGTCCATCCGTCTT

HCLibR1

TACTCTTGGAGGAGGGAGCTAGCGGAAAGACGGATGGACCCTTGGT

HCLibR2

CGACGCAGAGGACACGGTCACGAGGGTTCCCTGGCCCCAA

HCLibR3

TAGTC MVN A BM GTA MN AGCCA BY GCCTCC MHV TCTGGAGC

HCLibR4

AGTAGTACACGGCGGTATCCTCGGCTCTCAGGGAGTTCAT

HCLibR5

CTGCAGGTAGGCGGTATTCTTGGAGGTGTCGGCGGAGATG

HCLibR6

GTGAACCGGCCCTTCACAGAGTCGGCGTATCTGGTGTAGC

HCLibR7

CA Y TA KW GGGA K AGAT K CTGGCCACCCATTCCAGTCCCTT

HCLibR8

GCCAGGGGCCTGTCGCACCCAGTGGATGTAGGTGTCCTTG

HCLibR9

ATGTTGAAGCCGGAGGCGGCACAAGACAGTCTCAGAGATC

HCLibR10

CGCCAGGCTGCACCAGTCCGCCGCCAGATTCCACCAGCTG

HCLibR11

CACTTCAGGGCCAGGGTTAGACTCCACGTCGCCGGCAAGC

Note: Introduced degenerated nucleotides are indicated in bold: B ¼ C/G/T; D ¼ A/G/T; H ¼ A/C/T; K ¼ G/T; M ¼ A/C; N ¼ A/C/G/T; R ¼ A/G; V ¼ A/C/G; W ¼ A/T; Y ¼ C/T

3. Amplify the assembled VH library in large scale. First mix, on ice, 50 μl 5
Phusion HF Buffer, 5 μl 10 mM dNTPs, 5 μl assembly reaction (from step 2), 2.5 μl Phusion HF Polymerase, 7.5 μl DMSO, 12.5 μl 10 μM HCLibF1 primer, 12.5 μl 10 μM HCLibR1 primer, and 155 μl water. Split the reaction into five PCR tubes and cycle with the following conditions: Denature at 98  C for 30 s; Cycle 30 times at 98  C for 5 s, 72  C for 60 s; Extend at 72  C for 5 min.

412

Annalee W. Nguyen et al.

4. Run a 2% agarose gel of the amplified VH library (as in Subheading 3.1.1) and gel purify the band at 466 bp. Elute to 30 μl elution buffer provided in the gel purification kit. 5. Digest the purified VH library with HindIII-HF and NheI-HF. Mix all 30 μl of purified VH library insert with 10 μl CutSmart Buffer, 2 μl HindIII-HF, 2 μl NheI-HF, and water to bring the total volume to 100 μl. Incubate at 37  C for 2 h. 6. Simultaneously digest the pPybD1Disp vector (20 μg) with HindIII-HF and NheI-HF. Mix 20 μg pPybD1Disp with 20 μl CutSmart Buffer, 2.5 μl HindIII-HF, 2.5 μl NheI-HF, and water to bring the total volume to 200 μl. Incubate at 37  C for 2 h. 7. Gel purify the digested vector and insert fragments as described in Subheading 3.1.1. 8. Determine the concentration and purity of DNA in each digestion by analysis on a NanoDrop or similar. The 260/280 absorbance ratio should be >1.8 and the 260/230 ratio should be >2.0 for efficient ligation. The digestions may be desalted as described in Subheading 3.1.1, step 8 if these values are not obtained. 9. Test the digested products by ligating the VH library insert into the cut pPybD1Disp vector at an insert to vector molar ratio of 5:1 and total DNA mass of 200 ng. Mix 32 ng digested VH library insert with 168 ng digested pPybD1Disp vector with 2 μl T4 DNA Ligase Buffer, 1 μl T4 DNA Ligase, and water to make 20 μl total in a PCR tube. As a control, mix 168 ng digested pPybD1Disp vector with 2 μl T4 DNA Ligase Buffer, 1 μl T4 DNA ligase, and water to make 20 μl. Incubate both reactions at 16  C for 12 h, and then heat inactivate at 65  C for 20 min. Desalt these reactions, transform them to DH5α cells, and plate the transformants as described in Subheading 3.1.1. Count the resulting colonies and determine the ratio of vectoronly control colonies versus library colonies. This value is the percentage background for your library and is ideally under ~5%. Calculate the total library size of the small-scale transfection. A total of greater than 104 library transformants is indicative of digested insert and vector of high enough quality to proceed to scale up (see Note 11). 10. Scale up the ligations of the VH library insert into the cut pPybD1Disp vector at an insert to vector molar ratio of 5:1 and total DNA mass of 1 μg. Mix 160 ng digested VH library insert with 840 ng digested pPybD1Disp vector with 20 μl T4 DNA Ligase Buffer, 10 μl T4 DNA Ligase, and water to make 200 μl total. Split into two PCR tubes and incubate both at 16  C for 12 h, and then heat inactivate at 65  C for 20 min.

Engineering Antibodies on the Surface of CHO Cells

413

11. Precipitate the large-scale ligations by transferring them to separate 1.5 ml microcentrifuge tubes and adding 1 ml room temperature butanol to each tube. Spin at the highest speed available (>15500
g) for 5 min and carefully pipette off the butanol. Wash the delicate pellet with 500 μl room temperature 70% ethanol, spin again for 5 min, and allow the pellets to dry on a heat block just until no ethanol is visible. Resuspend each pellet in 15 μl water and combine for a total of 30 μl of ligated library. 12. Desalt the ligation as described in Subheading 3.1.1, step 8 for 2 h. Recover the desalted product and determine the concentration and purity of the DNA by Nanodrop. 13. Chill 1 mm gap electroporation cuvettes on ice while preparing the electroporation mixtures. Mix 50 μl DH5α electrocompetent cells with 5 μl desalted ligation in six different microcentrifuge tubes on ice by gently pipetting. Immediately transfer each mixture to one cold 1 mm gap electroporation cuvette. Electroporate and recover cells as described in Subheading 3.1.1, steps 9 and 10. The library transfections in SOC may be pooled for recovery in a single 50 ml flask. Be sure to plate multiple dilutions of the library as the concentration of cells may be >105 cells/ml. Library size is determined by the cell counts corrected for dilution factor. 14. Transfer the remaining ~6 ml of recovered library culture to 250 ml of LB media with 100 μg/ml ampicillin for overnight growth at 37  C and 250 rpm. 15. The following day, count colonies on plates to determine library size (ideally ~106 cfu), make five to ten frozen stocks of the library culture for storage at 80  C and Midi-prep the plasmid from the library. The resulting purified plasmid is called pPybD1HCLib. 3.4 bD1 Library Screening on CHO-T Cells

Throughout the library transformation and screening process, special care is taken to ensure that few clones as possible are lost from the library. During the initial transfection into CHO-T cells, the transformed plasmid is 74% irrelevant blank background plasmid and 26% library plasmid, resulting in 17% of transfected cells expressing Fab, and the transfection efficiency overall is expected to be about 40–50%. Therefore, the initial number of CHO-T cells transfected must be approximately eight times the actual library size determined in Subheading 3.3.2, step 15. We recommend targeting ten times the library size or 107 cells for the library transfection. The entire sorting process takes approximately 5 weeks.

414

Annalee W. Nguyen et al.

3.4.1 Transform the Library to CHO-T Cells

1. Start a culture of CHO-T cells in growth media (see Note 10) from a frozen stock. Grow the cells in filter-capped flasks at 37  C and 5% CO2 in a jacketed incubator. 2. Expand the cells until there are at least >1
106 cells/ml in three T-150 flasks (~30 ml each) for a total of >9
107 CHO-T cells. The day before transfection, ensure the cells are >95% viable (by Coomassie blue staining and counting on a hemocytometer or your favorite method) and at a concentration of 1
106 cells/ml. 3. Prepare DNA for transfection by mixing 111 μg of midiprepped, nonexpressing, and irrelevant pCTCON plasmid (see Note 4) with 39 μg pPybD1HCLib plasmid (Subheading 3.3.2) in a 15 ml conical tube. In separate microcentrifuge tubes, prepare control transfections by mixing 9.25 μg pCTCON with 3.25 μg pPyEBV, pPybD1Disp (Subheading 3.3.1), and pPy4D5Disp2 (Subheading 3.2). 4. Add 6 ml growth media to the pPybD1HCLib tube and 0.5 ml media to each control tube. 5. Mix 320 μl Lipofectamine™ 2000 with 8 ml growth media in a 15 ml conical tube. 6. Let the DNA/media mixtures and the Lipofectamine™ 2000/ media mixture sit at room temperature for 5 min. 7. Add 6 ml of the Lipofectamine™ 2000/media mixture to the tube containing pPybD1HCLib. Add 0.5 ml of the Lipofectamine™ 2000/media mixture to each of the three control tubes. Allow to the transfection mixtures to sit at room temperature for 20 min. 8. While allowing the DNA and Lipofectamine™ 2000 to incubate, count the overnight CHO-T culture and spin down 7.5
107 cells at 200
g for 2 min. 9. Resuspend the CHO-T cells in 50 ml growth media for a density of 1.5
106 cells/ml. Transfer 18 ml of CHO-T cells to two T-150 tissue culture flasks and add 3 ml of CHO-T cells to each of three T-25 flasks. 10. Add 6 ml of Lipofectamine™ 2000/pPybD1HCLib to each of the T-150 flasks, and 1 ml of each control Lipofectamine™ 2000/DNA mixture to separate T-25 flasks. 11. Allow the cells to grow overnight at 37  C and 5% CO2 in a jacketed incubator. 12. Transfer the contents of each flask to 50 or 15 ml conical tubes and centrifuge at 200
g for 2 min. 13. Vacuum aspirate the transfection media to waste. 14. Add 24 ml of fresh growth media to each tube of pPybD1HCLib transfected cells and 4 ml of fresh growth media to each

Engineering Antibodies on the Surface of CHO Cells

415

control tube. Transfer the cells back to their original flask for continued growth. 15. Two days after transfection, centrifuge the cells again as in steps 12 and 13 and resuspend in growth media containing 150 μg/ml Hygromycin B for selection of cells containing pPyEBV-based plasmids, which encode a Hygromycin B resistance gene (see Notes 5 and 12). 16. Five days after transfection, centrifuge the cells again as in steps 12 and 13, and resuspend in growth media containing 300 μg/ml Hygromycin B. Split the cultures 1:2 if needed due to high viable cell density (>1.5
106 cells/ml). 17. Continue monitoring the cultures every 2–3 days. The Hygromycin B will not kill the untransfected cells immediately, but will take several days. During this time, continue splitting the cells as needed into fresh 300 μg/ml Hygromycin B growth media, ensuring that at least 1
107 live cells total are present in the pPybD1HC Lib transfected cultures at all times to avoid loss of diversity. This may initially require expanding to four or more T-150 flasks but should scale down to two flasks after about 7–9 days post-transfection. Be sure to mix all of the cells displaying the library together before splitting and discarding any cells. 3.4.2 Sort for Improved Antigen Binding

1. Two weeks after transfection, the majority of cells lacking a pPyEBV-based plasmid have been killed by Hygromycin B, and the library is ready for sorting. All steps should be performed in a laminar flow biosafety cabinet to prevent contamination. 2. Transfer 15 ml of confluent pPybD1HCLib-transfected cells (Lib R0) into two separate 15 ml conical flasks (30 ml total). Transfer 2 ml of each control, pPyEBV-transfected cells (neg), pPy4D5Disp2-transfected cells (4D5), and pPybD1Disptransfected cells (bD1) into separate 15 ml conical tubes. 3. Centrifuge the cells at 200
g for 2 min. Vacuum aspirate the supernatant to waste. 4. Wash the Lib R0 pellets with 3 ml Buffer C and wash the neg, 4D5, and bD1 cell pellets with 1 ml Buffer C. Be sure the pellets are fully resuspended by gentle vortexing. 5. Centrifuge the cells at 200
g for 2 min. Vacuum aspirate the supernatant to waste. 6. Add 0.75 ml Stain C to each Lib R0 pellet and 0.25 ml Stain C to the neg, 4D5, and bD1 cell pellets. Fully resuspend the pellets in the stain by gentle vortexing. Incubate for 20 min at room temperature in the dark. 7. Centrifuge the cells at 200
g for 2 min. Vacuum aspirate the supernatant to waste.

416

Annalee W. Nguyen et al.

8. Wash the Lib R0 pellets with 3 ml Buffer C and wash the neg, 4D5, and bD1 cell pellets with 1 ml Buffer C. Be sure the pellets are fully resuspended by gentle vortexing. 9. Add 0.75 ml Stain D to each Lib R0 pellet and 0.25 ml Stain C to the neg, 4D5, and bD1 cell pellets. Fully resuspend the pellets in the stain by gentle vortexing. Incubate for 20 min at room temperature in the dark. 10. Centrifuge the cells at 200
g for 2 min. Vacuum aspirate the supernatant to waste. 11. Wash the Lib R0 pellets with 3 ml Buffer C and wash the neg, 4D5, and bD1 cell pellets with 1 ml Buffer C. Be sure the pellets are fully resuspended by gentle vortexing. 12. Determine the concentration of the Lib R0 cells in Buffer C by counting on a hemocytometer. Determine the total volume required to bring the cell concentration to ~3
106 cells/ml. 13. Centrifuge the cells at 200
g for 2 min. Vacuum aspirate the supernatant to waste. 14. Resuspend the control cell pellets (neg, 4D5, and bD1) in 0.5 ml Buffer C. Resuspend the Lib R0 cells in the volume calculated in step 12. 15. Scan each sample as described in section on a BD FACSAria instrument. Use the filter described for detection of EGFP for FITC and excite Cy3 with a 561 nm laser and detect emission through a 582/15 nm bandpass filter. Gate each population on forward and side scatter, and then use gating to exclude doublets and to exclude those cells with nonspecific binding (binding to the goat-antirabbit-Cy3 antibody). For the first sort, collect the 1% of cells with the highest HER2 ligand-binding (HER2-Fc/anti-Fc-AF647) signal versus display (anti-human kappa-FITC) signal as shown in Fig. 4. Use the 100-μm nozzle at 20 psi and collect ~100,000 cells into 4 ml growth media with 300 μg/ml Hygromycin B. This population is the round 1 sort of the bD1 heavy chain library (Lib R1). 16. Transfer the sorted cells to a T-25 flask and allow to grow at 37  C and 5% CO2 in a jacketed incubator for 1 week, expanding into a T-75 (12 ml) and T-150 (24 ml) as needed. Maintain the cultures in growth media with 300 μg/ml Hygromycin B. Continue to maintain the control samples (neg, 4D5, and bD1) as well as the Lib R0 sample in T-25 flasks by splitting into growth media with 300 μg/ml Hygromycin B as needed. 17. One week after the first sort round, count the Lib R1 cells using a hemocytometer, and centrifuge 2–3 batches of ~3
106 Lib R1 cells. Vacuum aspirate the media to waste, and place the tubes containing the Lib R1 pellets at 20  C for later DNA isolation.

Engineering Antibodies on the Surface of CHO Cells

417

Fig. 4 CHO-T surface display sorting strategy. To sort mammalian cells, it is important to gate on live, averagesized cells by (a) side scatter (SSC) versus forward scatter (FSC), then singlets by sequentially gating on (b) SSC width versus height and (c) FSC width versus height. Single cells are then gated to exclude nonspecific binders and nondisplaying cells by gating for those like bD1 (d) in a plot of nonspecific binding of an anti-rabbit Cy3 secondary versus HER2-Fc binding. Finally, a diagonal gate is drawn to collect ~1–2% of cells in each sort round based on the (e) bD1 display (anti-human kappa—FITC) versus binding (HER2-Fc/anti-Fc-AF647) features. Enrichment is apparent in this gate drawn to compare the (f) unsorted Lib R0 population to the (g) Lib R1 population and (h) Lib R2 population

18. Sort Lib R1 following the same process outlined in steps 1–15. Again, 1% of the cells exhibiting the highest HER2 binding (HER2-Fc/anti-Fc-AF647) signal versus display (anti-human kappa-FITC) signal should be sorted. This population is the round 2 sort of the bD1 heavy chain library (Lib R2). 19. When Lib R2 has grown to confluence in a T-150 flask, scan all of the populations again by staining 1–2 ml of each culture as described above to determine the enrichment achieved in the final sort (Fig. 4). 20. Count the Lib R2 cells using a hemocytometer and centrifuge 2–3 batches of ~3
106 Lib R2 cells. Vacuum aspirate the media to waste and place the tubes containing the Lib R2 pellets at 20  C for later DNA isolation. 3.5

Variant Recovery

After screening, the isolated cells still contain a pool of variant Fab sequences. The episomal DNA is easily recovered alongside genomic DNA, from which the antibody heavy and light chain genes can be amplified and subcloned by PCR, digestion, and ligation to isolate individual clones. Sequencing and flow cytometry analysis

418

Annalee W. Nguyen et al.

of individual clones guides selection of variants for further analysis as desired. 3.5.1 Recover Fab DNA Sequences

1. Purify the total DNA from frozen Lib R1 and Lib R2 cell pellets (Subheading 3.4.2, steps 17 and 20) using the PureLink Genomic DNA Mini Kit according to the instructions. 2. Clone the VH sequences from the Lib R1 and Lib R2 total DNA into E. coli as described in Subheading 3.3.2, steps 3–9 (see Note 13). 3. After plating, inoculate 5–20 or more bacterial transformants into 12 ml TB with 100 μg/ml ampicillin for overnight culture at 37  C. 4. Make frozen stocks of each overnight culture, miniprep the plasmid, and submit purified plasmid for DNA sequencing. 5. After sequencing, analyze variants for diversity and consensus. Decide on a panel of variants to evaluate individually. 6. Prepare midi-prepped plasmid DNA for CHO-T cell transfection of individual variants.

3.5.2 Analysis of Individual Fab Variants

1. Transfect 6.25 μg of each variant plasmid DNA preparation and each control (neg, 4D5, and bD1) in duplicate wells of six-well plates as described in Subheading 3.2.1, steps 4–16. 2. Two days after transfection, stain and scan the variants and controls by flow cytometry as described for sorting (Subheading 3.4.2), scaling the staining buffers appropriately for the required volumes depending on the number of variants tested. 3. Compare each variant to bD1 and 4D5, evaluating expression level, HER2 ligand binding, and non-specific binding to irrelevant anti-rabbit-Cy3 antibody. Due to the large CHO-T cell surface area and avidity effects, even small differences by flow cytometry can reflect large binding differences (Fig. 5). 4. Based on this analysis, choose the most improved clones (~5–10 total) to express as full length antibodies or in the antibody format of choice along with a negative, bD1, and 4D5 control. As full-length human antibodies, the improved variants isolated through this method bound HER2 expressed on SK-OV-3 ovarian cancer cells better than bD1 with some approaching levels observed for the positive control 4D5 (Fig. 6). Moreover, the CHO-T expression levels, stabilities, and apparent affinities of the CHO-T cell-selected Fab variants were similar to 4D5 [18].

Engineering Antibodies on the Surface of CHO Cells

419

Fig. 5 Analysis of representative individual clones after sorting. CHO-T cells displaying the parental bD1 Fab or individual variant Fabs were analyzed for (a) nonspecific binding by comparing irrelevant antibody binding versus HER2 binding and analyzed for (b) affinity maturation with HER2 binding relative to display level (antihuman kappa—FITC). These overlays compare bD1 (black), and representative clones from each sorting round including R1G (green), R2D (blue), and R1J (pink). The arrow indicates the increasing affinity relative to display of the clones, while nonspecific binding levels were relatively constant for R1G and R2J but slightly increased for R2D

Fig. 6 Staining of HER2-expressing SK-OV-3 cells with HER2-binding variants. HER2-expressing SK-OV-3 cells were incubated with full-length human HER2-binding antibodies at 0.8 nM buffer and then washed and stained with goat-anti-human Fc-AF647 for detection by flow cytometry. No binding was present with (a) irrelevant antibody, and increasing binding was apparent with incubation with (b) bD1, (c) R2D, (d) R2J, and (e) 4D5

420

4

Annalee W. Nguyen et al.

Notes 1. Multiple commercial options are available for the purchase of gene fragments. We prefer Integrated DNA Technologies gBlocks. To ensure adequate DNA for cloning, we use PCR with outside primers to amplify the gene fragment and purify it before use. Alternatively, the gene insert may be generated through oligo assembly [24] or other PCR-based methods. 2. Ethidium bromide is a mutagen and potentially a carcinogen. Care should be taken when handling this reagent. We recommend purchasing ethidium bromide in solution to avoid inhalation hazards associated with the powdered reagent. Wear gloves at all times and follow hazardous waste disposal recommendations for your institution. 3. We have found that use of freshly prepared DNA from freshly streaked frozen stocks and fresh cultures results in larger library sizes with this large plasmid. We use the ZymoPURE II Plasmid Midiprep Kit (Zymogen # D4201). 4. A relatively large blank plasmid comparable in size to pPyEBV without a Hygromycin B resistance gene is optimal for doping in place of pPyEGFP used in Subheading 3.2.1. Plasmid pCTCON [21] is >6000 kb and works well for this, but many other options would likely work equally well. 5. Hygromycin B is a highly toxic chemical. Be sure to handle it carefully following appropriate safety practices. 6. Cloning into pPyEBV as described in this step may be performed using any method preferred by the researcher. The pPyEBV vector is quite large (10,720 bp), so any protocol using PCR amplification of the plasmid (e.g., Gibson assembly) must be carefully performed to avoid unwanted errors in plasmid replication. We prefer the traditional digestion and ligation protocol described here for this large plasmid. 7. Eye and skin protection must be utilized when using a UV light Table. UV light also damages sample DNA, and the time DNA samples are exposed to UV should be minimized as much as possible. To further minimize UV exposure, we do not recommend taking photos of the gel under UV. Non-UV excitable dyes and other gel purification methods (e.g., electroelution) may be substituted to minimize DNA damage for all gel purification steps described. 8. The large size of the pPyEBV fragment results in reduced yield relative to smaller digested plasmids in many gel purification kits. We have found that the Zymoclean™ Gel DNA Recovery Kit works well for this large plasmid, but other kits may be

Engineering Antibodies on the Surface of CHO Cells

421

substituted if required. The small elution volume possible with this kit is an added convenience. 9. Column desalting is not recommended for this large plasmid size. 10. Although an anticlumping agent is recommended by Acyte Biotech for growth of CHO-T cells, it must be completely removed prior to transfection. Moreover, the impact of this reagent on surface displayed antibodies is unclear. For these reasons, we do not recommend using anticlumping agent during growth of CHO-T cells in this protocol. 11. Key components to optimize if the number of transformants is low: (a) complete digestion (but not over digestion) of the vector and insert and (b) removal of all agarose after gel purification. Each DNA fragment should be made fresh and the library should be made over contiguous days, minimizing storage of PCR products and digested DNA during the process as much as possible. For the large-scale transformation, it is very important to use freshly made, high quality electrocompetent cells with an efficiency of >109 cfu per μg pUC19 DNA to ensure adequate library size. 12. It may be helpful to use 1 ml of each culture to evaluate transfection efficiency on day 2 after transfection. This may be performed by flow cytometry as described in Subheading 3.2.1, steps 16–31. 13. After cloning the sequences isolated by PCR from Lib R1 and Lib R2, the E. coli transformants left over after plating may be propagated overnight and midi-prepped the following day to recover the Lib R1 pool and the Lib R2 pool. These may be transfected back into CHO-T cells for further sorting or analysis of multiple sort rounds simultaneously. References 1. Hoogenboom HR (2005) Selecting and screening recombinant antibody libraries. Nat Biotechnol 23(9):1105–1116. https://doi. org/10.1038/nbt1126 2. Nieba L, Honegger A, Krebber C, Pluckthun A (1997) Disrupting the hydrophobic patches at the antibody variable/constant domain interface: improved in vivo folding and physical characterization of an engineered scFv fragment. Protein Eng 10(4):435–444 3. Ni M, Lee AS (2007) ER chaperones in mammalian development and human diseases. FEBS Lett 581(19):3641–3651. https://doi. org/10.1016/j.febslet.2007.04.045 4. Wildt S, Gerngross TU (2005) The humanization of N-glycosylation pathways in yeast. Nat

Rev Microbiol 3(2):119–128. https://doi. org/10.1038/nrmicro1087 5. Bradbury AR, Sidhu S, Dubel S, McCafferty J (2011) Beyond natural antibodies: the power of in vitro display technologies. Nat Biotechnol 29(3):245–254. https://doi.org/10.1038/ nbt.1791 6. Beerli RR, Bauer M, Buser RB, Gwerder M, Muntwiler S, Maurer P, Saudan P, Bachmann MF (2008) Isolation of human monoclonal antibodies by mammalian cell display. Proc Natl Acad Sci U S A 105(38):14336–14341. https://doi.org/10.1073/pnas.0805942105 7. Bowers PM, Horlick RA, Kehry MR, Neben TY, Tomlinson GL, Altobell L, Zhang X, Macomber JL, Krapf IP, Wu BF, McConnell

422

Annalee W. Nguyen et al.

AD, Chau B, Berkebile AD, Hare E, Verdino P, King DJ (2014) Mammalian cell display for the discovery and optimization of antibody therapeutics. Methods 65(1):44–56. https://doi. org/10.1016/j.ymeth.2013.06.010 8. Bowers PM, Horlick RA, Neben TY, Toobian RM, Tomlinson GL, Dalton JL, Jones HA, Chen A, Altobell L 3rd, Zhang X, Macomber JL, Krapf IP, Wu BF, McConnell A, Chau B, Holland T, Berkebile AD, Neben SS, Boyle WJ, King DJ (2011) Coupling mammalian cell surface display with somatic hypermutation for the discovery and maturation of human antibodies. Proc Natl Acad Sci U S A 108 (51):20455–20460. https://doi.org/10. 1073/pnas.1114010108 9. Bowers PM, Neben TY, Tomlinson GL, Dalton JL, Altobell L, Zhang X, Macomber JL, Wu BF, Toobian RM, McConnell AD, Verdino P, Chau B, Horlick RA, King DJ (2013) Humanization of antibodies using heavy chain complementarity-determining region 3 grafting coupled with in vitro somatic hypermutation. J Biol Chem 288 (11):7688–7696. https://doi.org/10.1074/ jbc.M112.445502 10. Ho M, Nagata S, Pastan I (2006) Isolation of anti-CD22 Fv with high affinity by Fv display on human cells. Proc Natl Acad Sci U S A 103 (25):9637–9642. https://doi.org/10.1073/ pnas.0603653103 11. Horlick RA, Macomber JL, Bowers PM, Neben TY, Tomlinson GL, Krapf IP, Dalton JL, Verdino P, King DJ (2013) Simultaneous surface display and secretion of proteins from mammalian cells facilitate efficient in vitro selection and maturation of antibodies. J Biol Chem 288(27):19861–19869. https://doi. org/10.1074/jbc.M113.452482 12. King DJ, Bowers PM, Kehry MR, Horlick RA (2014) Mammalian cell display and somatic hypermutation in vitro for human antibody discovery. Curr Drug Discov Technol 11 (1):56–64 13. Li CZ, Liang ZK, Chen ZR, Lou HB, Zhou Y, Zhang ZH, Yu F, Liu S, Zhou Y, Wu S, Zheng W, Tan W, Jiang S, Zhou C (2012) Identification of HBsAg-specific antibodies from a mammalian cell displayed full-length human antibody library of healthy immunized donor. Cell Mol Immunol 9(2):184–190. https://doi.org/10.1038/cmi.2011.55 14. Mason DM, Weber CR, Parola C, Meng SM, Greiff V, Kelton WJ, Reddy ST (2018) Highthroughput antibody engineering in mammalian cells by CRISPR/Cas9-mediated homology-directed mutagenesis. Nucleic Acids Res

46(14):7436–7449. https://doi.org/10. 1093/nar/gky550 15. McConnell AD, Do M, Neben TY, Spasojevic V, MacLaren J, Chen AP, Altobell L 3rd, Macomber JL, Berkebile AD, Horlick RA, Bowers PM, King DJ (2012) High affinity humanized antibodies without making hybridomas; immunization paired with mammalian cell display and in vitro somatic hypermutation. PLoS One 7(11):e49458. https://doi.org/10. 1371/journal.pone.0049458 16. Schenk JA, Sellrie F, Bottger V, Menning A, Stocklein WF, Micheel B (2007) Generation and application of a fluorescein-specific single chain antibody. Biochimie 89(11):1304–1311. https://doi.org/10.1016/j.biochi.2007.06. 008 17. Walsh G (2014) Biopharmaceutical benchmarks 2014. Nat Biotechnol 32 (10):992–1000. https://doi.org/10.1038/ nbt.3040 18. Nguyen AW, Le KC, Maynard JA (2018) Identification of high affinity HER2 binding antibodies using CHO Fab surface display. Protein Eng Des Sel 31(3):91–101. https://doi.org/ 10.1093/protein/gzy004 19. Kunaparaju R, Liao M, Sunstrom NA (2005) Epi-CHO, an episomal expression system for recombinant protein production in CHO cells. Biotechnol Bioeng 91(6):670–677. https:// doi.org/10.1002/bit.20534 20. Bostrom J, Yu SF, Kan D, Appleton BA, Lee CV, Billeci K, Man W, Peale F, Ross S, Wiesmann C, Fuh G (2009) Variants of the antibody herceptin that interact with HER2 and VEGF at the antigen binding site. Science 323(5921):1610–1614. https://doi.org/10. 1126/science.1165480 21. Colby DW, Kellogg BA, Graff CP, Yeung YA, Swers JS, Wittrup KD (2004) Engineering antibody affinity by yeast surface display. Methods Enzymol 388:348–358. https://doi.org/ 10.1016/S0076-6879(04)88027-3 22. https://en.wikipedia.org/wiki/Poisson_ distribution 23. Bessette PH, Mena MA, Nguyen AW, Daugherty PS (2003) Construction of designed protein libraries using gene assembly mutagenesis. Methods Mol Biol 231:29–37. https://doi. org/10.1385/1-59259-395-X:29 24. Stemmer WP, Crameri A, Ha KD, Brennan TM, Heyneker HL (1995) Single-step assembly of a gene and entire plasmid from large numbers of oligodeoxyribonucleotides. Gene 164(1):49–53

Chapter 23 Single B Cell Cloning and Production of Rabbit Monoclonal Antibodies Juliet Rashidian and Joshua Lloyd Abstract Monoclonal antibodies are among the most significant biological tools used in medicine and biology that have revolutionized the field of diagnostics, therapeutics, and targeted drug delivery systems for many diseases. Among them, rabbit monoclonal antibodies have attracted significant attention for having high affinity and specificity. During the past few decades, different techniques have been developed to produce monoclonal antibodies. Single B cell cloning technology offers many advantages compared to other methods and has been used to generate monoclonal antibodies from different species including rabbits. This review briefly describes some of these methods, with main focus on single B cell cloning and production of rabbit monoclonal antibodies. Key words Single B cell cloning, Rabbit monoclonal antibodies, B cell sorting

1

Introduction In August 1975, Ce´sar Milstein and Georges Ko¨hler published a three-page report and described a method for generating mouse monoclonal antibodies with predefined specificity [1]. This method revolutionized biomedical research and diagnostics and opened a door of therapies for many diseases. Since then, the technology of developing monoclonal antibodies has been expanded to other animal sources, and molecular biology has extremely improved the performance of these antibodies. Monoclonal antibodies are generated by a homogeneous population of B cells derived from one parental cell, and they detect a single epitope on any antigen. Polyclonal antibodies, on the other hand, originate from multiple lineages of B cells and recognize multiple epitopes on one antigen. Monoclonal antibodies are highly specific and show low cross-reactivity with nonspecific antigens compared to the polyclonal antibodies [2]. Table 1 compares some features of monoclonal and polyclonal antibodies.

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2070, https://doi.org/10.1007/978-1-4939-9853-1_23, © Springer Science+Business Media, LLC, part of Springer Nature 2020

423

424

Juliet Rashidian and Joshua Lloyd

Table 1 Comparison between monoclonal and polyclonal antibodies Monoclonal

Polyclonal

Produced by a single B cell clone

Produced by multiple lineages of B cells

Detects a single epitope on an antigen

Detects multiple epitopes on an antigen

A homologous antibody population

A heterogeneous antibody population

High specificity

High affinity and robust detection

Higher time to produce

Lesser time to produce

Lower cross-reactivity

Higher cross-reactivity

There is a consistent source of antibody with the same performance

There is potential inconsistency between batches

Higher cost to develop the antibody

Lower cost to develop the antibody

Monoclonal antibodies are among the most significant biological tools used in the field of molecular biology, biochemistry, and medicine. Due to their high specificity, monoclonal antibodies have been successfully used in the field of diagnostics, therapeutics, and targeted drug delivery systems for infectious diseases, cancers, autoimmune diseases, and metabolic and hormonal disorders [3–7]. Numerous monoclonal antibodies undergo evaluation in therapeutic clinical studies every year, and Food and Drug Administration (FDA) has already approved a number of these antibodies for cancer and noncancer indications [8, 9]. More than 230 monoclonal antibodies have been evaluated in the clinical phase in 2017 [9]. In this chapter, we will review the widely used methods for generating monoclonal antibodies and compare the advantages and disadvantages of every method. The main focus of the chapter, however, is on single B cell cloning, and a relatively detailed description is provided for generating rabbit monoclonal antibodies by this technology at the end of the chapter.

2

Mouse Monoclonal Antibodies Versus Rabbit Monoclonal Antibodies The majority of the monoclonal antibodies have been generated in mice; however, there has been an increasing tendency of using rabbit antibodies compared to mouse antibodies. A rabbit’s immune system is evolutionarily distinct from that of a rodent and uses different mechanisms to generate, diversify, and optimize the affinity of the antibodies. In general, rabbit monoclonal antibodies have several advantages over traditional mouse monoclonal antibodies in diagnostics [10–14], which are as follows:

Review: Single B Cell Cloning Technology

3

425

l

Rabbit antibodies have higher affinity and specificity than mouse antibodies.

l

Rabbit antibodies have more diverse epitope recognition than mouse antibodies.

l

Rabbit immune system can recognize small-sized epitopes that are not immunogenic in mice and can produce strong immune responses.

l

As immunotherapeutic agents, mouse antibodies have a short half-life in serum and induce human antimouse antibodies. Engineered rabbit monoclonal antibodies show reduced immunogenicity, while they retain their high specificity and affinity to human antigens.

Technologies and Methods for Monoclonal Antibody Production The mouse hybridoma method is the first and most commonly used approach for generating mouse monoclonal antibodies [1]. In the past few years, several other techniques including display methods [15–18], rabbit hybridoma technique [19, 20], and isolation and cloning antibody encoding genes from B cells [21–23] have been used for producing monoclonal antibodies. In this review, we briefly review these techniques with focus on B cell cloning.

3.1 Hybridoma Technology

Generating mouse monoclonal antibodies by the hybridoma technique was described for the first time by Milstein and Ko¨hler [1]. This approach is based on the cell fusion between B cells isolated from an immunized animal and myeloma partner cells. To generate the hybridomas, a mouse is immunized with an antigen of interest, and spleen cells are used as a source of antibodyproducing B cells [24]. The freshly isolated spleen cells are fused with immortal myeloma cells derived from the BALB/c mouse (such as SP2/0) using polyethylene glycol (PEG) compound to form hybridoma cells. The hybridoma cells inherit the qualities of their parental cells; they are immortal and able to grow continually like their parental tumor cells and produce antibodies like their parental B cells. The cells are re-suspended in a medium containing hypoxanthine, aminopterin, and thymidine (HAT) as a selective agent following fusion. Aminopterin blocks the de novo synthesis of DNA. Cells can survive in the HAT medium only if they synthesize DNA through a pathway called the salvage pathway, in which hypoxanthine and thymidine are incorporated into the DNA by phosphoribosyl transferase (HGPRT). The parental myeloma cells lack HGPRT, and hence, they will die. Unfused B cells will not survive, as they have a short lifespan. Thus, only fused cells will grow in the HAT medium. The cells from the fusion step are then distributed to 96-well plates containing the HAT medium, and the

426

Juliet Rashidian and Joshua Lloyd

plates are incubated for 10–14 days until colonies grow as small clusters in wells. When colonies grow, they are screened for specific polyclonal antibodies by enzyme-linked immunosorbent assay (ELISA). The positive hybridomas are cloned twice by limiting dilution in a microtiter plate to have at most a single clone in every well. Once again, the production of antibody by single clones is tested by ELISA, and the hybridomas are expanded in a medium containing HT. The newly formed hybridomas are co-cultured with feeder cells (such as macrophages from mice or rats, nonimmunized spleen cells, human fibroblasts, and thymus cells) to provide optimal growth conditions and antibody production. Developing rabbit hybridomas was rather complicated in the absence of rabbit myeloma cells. Therefore, efforts were originally made to generate rabbit–mouse hybridomas [25]. These hybridomas were produced by fusing immunized rabbit spleen cells with mouse SP2/0 myeloma cells. However, this strategy was not successful due to the poor fusion efficiency, genetic instability, poor heavy and light-chain pairing, and decrease in antibody secretion [26, 27]. The first rabbit homo-hybridomas were developed by Spieker-Polet et al. in 1995 [19]. This group generated v-abl/cmyc double transgenic rabbits and obtained the first stable rabbit plasmacytoma cell line 240E-1 [19, 20]. The rabbit homohybridomas, however, were not stable, and hence, antibody secretion decreased over time. To overcome these issues, the initial 240E-1 cell line was further improved by sub-cloning and screening for better clones. The improved fusion cell line 240E-W and its descendants 240E-W2 and 240-W3 clones have higher stability, better fusion efficiency, and more stable antibody secretion and have been successfully used to generate rabbit hybridomas and rabbit monoclonal antibodies (US Patent 7,429,487) [28]. Human hybridoma technology has been also developed using different strategies including fusion between a murine cell line and human B-lymphocytes [29, 30]. 3.2 Display Technology: Phage Display

Several display platforms have been introduced to produce monoclonal antibodies during the past few years, such as yeast surface display [17], ribosome display [18], and phage display [15]. Since phage display is used more widely among the other display platforms, we will review this technology. Phage display was introduced by George Smith for the first time in 1985 [15] and further developed by other groups [31–33]. Phage display is based on genetic engineering of bacteriophages (viruses that infect bacteria like Escherichia coli (E. coli)) so that they express the desired peptides as a component of their surface capsid proteins. In antibody phage display, the engineered phages that express functional fragments of antibodies are used in repeated rounds of selection using specific antigen. In this platform, animals are immunized, and B cells are isolated from peripheral

Review: Single B Cell Cloning Technology

427

blood mononuclear cells (PBMCs) [34, 35]. Next, an antibody library is constructed by extracting mRNA from pooled B cells, followed by complementary DNA (cDNA) synthesis. The cDNA is used for amplification of variable domains of heavy chain and light chain (VH and VL) in polymerase chain reaction (PCR) [36]. In this technology, the synthesized antibody fragments are cloned in phagemid vectors. Phagemids have both bacteriophage and plasmid properties. These vectors are engineered to express the VH and VL as a single-chain variable fragment (scFv; VH and VL are joined by a linker) or fragment antigen binding (Fab; composed of one constant and one variable domain of heavy chain and light chain) fused to a capsid protein [37]. The phagemid vectors do not have all the genes required for coding a complete bacteriophage in bacteria. Therefore, a helper phage is transformed into E. coli together with the phagemid vector. In this case, phage particles are packaged and released into the medium, each expressing functional fragments of antibody on their surfaces [38]. The phages carrying the specific antibody are isolated from nonspecific phages based on their ability to bind to the target antigen [39]. This step is called biopanning and is usually repeated three to five times to isolate specific antibodies with high affinity [34]. The phage pools (polyclonal antibody) are tested by ELISA. If ELISA results show binding to the target antigen, then the bacteria infected with polyclonal phages are plated, and individual colonies are picked and grown for monoclonal phage production. Phage display can be successfully used to generate antibodies from any species if the antibodies genes are identified for that species. This technology allows the generation of entirely human antibodies and is a well-established and reliable platform for the generation of antibodies for therapeutic purposes [40]. One advantage of phage display is that it allows to design and select antibodies with desired specificity through different panning strategies and controlling biochemical parameters [41, 42]. 3.3 Downsides of Hybridoma and Phage Display Technologies

Although hybridoma screening and phage display have been widely used to produce monoclonal antibodies, they both have some downsides [43, 44]: Hybridoma l

The efficiency of cell fusion is very low (5
106 efficiency with PEG), and therefore, the vast majority of B cells are not sampled, and repertoire for rare antibodies is lost.

l

It is time consuming and involves a labor-intensive multistep process.

l

Hybridomas are grown in culture media. Therefore, there is a constant risk of contamination.

428

Juliet Rashidian and Joshua Lloyd l

Hybridomas might be genetically unstable and hence stop producing the antibody due to chromosomal re-arrangement after a few rounds of culture.

Phage Display l

During construction of antibody library, the variable domain genes are randomly combined. Therefore, the natural cognate pairing of heavy chain and light chain is lost, and as a result, the antibody might be unstable.

To overcome these issues, huge efforts have been made to develop some alternative methods discussed in the following sections. 3.4 B Cell Clonal Expansion Technology

Immortalization of peripheral B cells with Epstein-Barr virus (EBV) is an effective procedure for inducing long-term growth of B cells [45]. In this procedure, the marmoset cell line culture B95-8 is used as a source of virus since the supernatant contains a high titer of EBV [46]. The lymphocytes are isolated from an immunized animal using a density gradient medium and re-suspended in B95-8 supernatant containing EBV. These cells are next incubated at 37  C and 5% CO2. After 24 h of incubation, a fresh transformation medium is added to the cells and cultures are incubated for 7–10 days. The medium should be allowed to become acidic and the cells allowed clumping. The transformed lymphocyte cell line can be expanded and frozen at this time. This method was further advanced by fusing immortalized B cells with myeloma cells, which resulted in B cells with higher stability and productivity [47]. To improve the efficiency of the method, EBV-transformed cells that express a human monoclonal antibody have been infected with a retrovirus encoding an activated form of ras oncogene (v-ras) [48].

3.5 Single B Cell Cloning

The most straightforward, yet technically challenging, method for generating monoclonal antibodies is single B cell cloning. This technology allows direct sampling of the immune repertoire from a single B cell. The platform consists of detection and screening of specific B cells from a large population of primary B cells, amplification of antibody genes, cloning the genes into expression vectors, and expressing the monoclonal antibody in mammalian cells (such as HEK293 and CHO cells) or bacterial systems (E. coli). The single B cell cloning method has been used to produce human, mouse, rat, and rabbit monoclonal antibodies [21, 49, 50] and has yielded some therapeutic reagents for several diseases including cancer, autoimmune disorders, and infectious diseases [21]. In general, either antigen-specific memory B cells expressing surface IgG [44] or IgG-secreting plasma cells [23] can be used as sources of immune repertoire.

Review: Single B Cell Cloning Technology 3.5.1 Advantages of Single B Cell Cloning

3.5.2 Screening Single B Cells by Flow Cytometry

429

Single B cell cloning has several advantages compared to other technologies for the production of monoclonal antibodies [21, 43, 44]: l

Unlike the inefficient fusion step in hybridoma technology, single B cell cloning is highly efficient and a large part of the antibody repertoire is preserved. This favors generation of antibodies with higher diversity.

l

Both VH and VL are amplified from single cells. Therefore, unlike the display methods, the natural cognate pairing and development of antibodies is maintained, and antibodies with higher affinity, stability, and specificity are formed.

l

In this platform, the DNA sequences coding for VH and VL are cloned. Therefore, the procedure does not rely on growing cells, and there is no risk of contamination and losing antibodyproducing cells.

l

Preserving the B cells is not required, and therefore, the procedure is not complicated by immortalizing cells by viral transformation or any other method.

Several approaches have been applied for screening antibodysecreting cells either in a random manner or in an antigen-specific manner. Micromanipulation and picking labeled B cells from tissue [51] and fluorescence activated cell sorting (FACS; flow cytometry) based on the surface markers [22, 52] are some of the methods used to randomly screen B cells. Alternatively, flow cytometry has been also used to screen antigen-specific B cells [49]. Other methods are screening by antigen-coated magnetic beads [53] or microwell and chip-based arrays [54, 55], and developing fluorescent foci to isolate antigen-specific plasma cells [23]. Although most of the techniques stated above are attractive for single cell isolation, they rely on manual manipulation and therefore have low output. Among them, however, flow cytometry has been recently shown to be an efficient technique to isolate and sort antigen-specific B cells [44, 52, 56–59]. This technology has been easily applied for sorting human B cells. Several markers have been identified on the surface of these cells and a variety of highly characterized antibodies is available to detect the markers and differentiate subsets of human immune cells. For example, Di Niro et al. generated rotavirus-specific human monoclonal antibodies from small-intestinal mucosa. They applied GFP-labeled viruslike particles and used fluorophore-conjugated anti-CD19 (a marker for human B cells) and anti-CD138 (a marker for human plasma cells) antibodies to harness specific cells [60]. Tiller et al. isolated human B cells at different stages of development based on surface markers [52]. This group stained isolated mononuclear cells with antibodies conjugated to different fluorophores

430

Juliet Rashidian and Joshua Lloyd

such as fluorescein-isothiocyanate (FITC), phycoerythrin (PE), allophycocyanin (APC), or biotin to distinguish subpopulations of B cells at different stages [52]. Unlike the human B cells, there are fewer examples of using flow cytometry to sort B cells from immunized animals [61]. Starkie et al. have described a multiparameter flow cytometry approach to sort single IgG+ memory B cells derived from mouse spleen [44]. They applied two-color antigen staining in combination with several mouse B cell-surface markers to efficiently detect and sort specific B cells. The B cells were enriched using an anti-CD45R (B220) (a marker for mouse B cell) reagent and then stained with anti-CD19 (a marker for mouse B cells) and anti-IgG antibodies. Finally, they included fluorophore-conjugated antigen for accurate detection of specific memory B cells in the flow cytometry [44]. Regarding rabbit system, identifying and sorting B cells by flow cytometry is even more challenging due to lack of general rabbit B cell markers. In this case, staining of rabbit B cells is restricted to staining with fluorophore-conjugated antigen. Incorporation of an antigen-binding step increases the recovery of specific B cells; including markers for staining unwanted cells such as T cells, and IgM+ cells will be helpful to sort desired IgG+ cells [44]. To overcome the limitations of surface markers, Kurosawa et al. have applied a different strategy without using these markers [62]. They stained lymph node cells with an antibody for IgG and a fluorescent dye specific for the endoplasmic reticulum (ER-tracker) assuming that these cells contain plenty of ER to produce large amounts of antibodies. They called this method endoplasmic reticulum (ER)-based identification of antigenspecific antibody-producing cells (ERIAA) and applied this strategy to identify plasma cells from human, mouse, rabbit, rat, and guinea pig in flow cytometry [62]. In another attempt to produce rabbit monoclonal antibodies by B cell cloning, Seeber et al. developed a platform for panning lymphocytes [63]. In this platform, they isolated PBMCs from an immunized rabbit and enriched the specific B cells with an antigencoated plate (panning). The attached cells were then removed and stained with antirabbit IgG FITC antibody and subjected to single cell sorting by flow cytometry. Single B cells were next co-cultivated with murine EL-4 B5 cells as feeders in a medium containing rabbit thymocyte supernatant (as a source of interleukins) and allowed to proliferate over time. The positive clones for IgG were then lysed and used for mRNA extraction and cDNA synthesis [63]. This platform is relatively rapid and generates a diverse set of antibodies. Importantly, natural cognate pairing of the light and heavy chains is maintained. However, one of the shortcomings of this method is the need for co-cultivating B cells with feeder cells and supplementing medium with interleukins. Moreover, it will take more than a

Review: Single B Cell Cloning Technology

431

week for B cells to proliferate and make colonies and produce enough antibody in the supernatant. Also, contamination and losing the cells is always a risk for growing any type of cells. 3.5.3 Single-Cell Antibody Gene Amplification, Cloning, and Expression

Construction of cDNA is the next step following sorting of B cells. If the B cells are cultivated, then they are lysed and total RNA and then mRNA is extracted from the lysates [63]. If sorted cells are not cultivated, then cDNA will be synthesized from a single cell. The cDNA is synthesized by reverse transcription of the mRNA during reverse transcription polymerase chain reaction (RT-PCR) using random hexamers, oligo-dT, or gene-specific primers [49, 52, 63, 64]. Usually, single cells are sorted in a 96-well plate, and cell lysis and cDNA synthesis are performed in the original well that the cell was deposited in [49]. This ensures an easier handling of large numbers of samples; however, care must be taken to minimize the risk of cross-contamination between adjacent wells during pipetting. The full-length gene transcripts for VH and VL are usually amplified from the synthesized cDNA by nested or semi-nested RT-PCR during two rounds of PCR [21, 49]. The forward primer could be a sequence or a combination of sequences complementary to the corresponding heavy- and light-chain leader sequences. The reverse primer would be a single sequence or a mixture of sequences within the constant region of heavy and light chains [14, 63]. In case of rabbit IgG, for instance, there is only one subclass of IgG, and there is only one gene for the constant region of heavy chain. Regarding light chain, more than 70% of rabbit IgG light chains are kappa 1 and the remaining are kappa 2 and lambda [63]. Therefore, these details need to be considered during designing primers for PCR. The second round of amplification with nested primers is to increase specificity and to include restriction sites for the subsequent cloning of the variable genes [65]. There are different strategies to clone amplified genes coding for variable domains and to express the antibody: One strategy is to clone the PCR fragments into a plasmid vector with an expression cassette containing the leader sequence and constant region for either heavy chain or light chain. The vector also needs to have all the essential elements required for transcription and translation such as a promoter sequence and a poly (A) sequence. Next, the vectors are transformed into E. coli to prepare enough DNA for transfection. Then, plasmids harboring heavy-chain and light-chain expression genes are co-transfected into mammalian expression cells. The heavy and light chains are expressed by the cells and released into culture media as fully assembled antibody structures [49]. Sequence- and ligation-independent cloning (SLIC) is a different method to clone fragments encoding variable domains [63]. A

432

Juliet Rashidian and Joshua Lloyd

plasmid that carries heavy- or light-chain constant domain sequence, and all the elements needed for transcription and translation are digested with restriction enzymes. SLIC uses an exonuclease such as T4 DNA polymerase to generate complementary single-stranded DNA overhangs in the linear plasmid and the insert DNA fragment (VH or VL). These are then combined in vitro to generate HC- and LC-expressing plasmids. The plasmids are next transfected into mammalian cells such as HEK293 [63]. There is an alternative strategy to amplify VH and VL fragments and to express recombinant antibody. This strategy is PCR-amplifying linear immunoglobulin expression cassettes and constructing transcriptionally active fragments (transcriptionally active PCR; TAP). The TAP strategy does not require the cloning step, and the constructed fragments would be directly transfected into mammalian cells for expression [44]. Using TAP, Starkie et al. successfully generated antigen-specific recombinant antibodies from both rabbit and mouse IgG+ memory B cells. Following cDNA synthesis and in the second round of PCR for amplifying variable domains, this group used a set of primers that not only introduced restriction sites for downstream cloning but also provided a 25-bp overlap region at both 50 and 30 ends. Then, in a tertiary PCR, variable domain DNA, a promoter fragment, and a constant region fragment containing a poly(A) sequence were combined and amplified to produce two separate linear transcriptionally active PCR products, one encoding the heavy chain and the other encoding the light chain. Next, heavy- and light-chain transcriptionally active fragments were co-transfected into mammalian cells for transient expression of the antibodies [44]. Liao et al. [65] also used linear heavy-chain and light-chain gene expression cassettes to produce recombinant influenza monoclonal antibodies from sorted single human plasmablasts without a cloning step. The cassettes contained promoter, antibody leader sequences, constant domains for IgG1 heavy chain or light chain, poly (A) sequence, and VH or VL genes [65]. A 50 -rapid amplification of cDNA ends (50 -RACE) technique has been reported to synthesize cDNA from a single cell to amplify variable domain genes [64]. The human B cells are isolated form blood and first-strand cDNA is synthesized from single cells using a gene-specific primer. A 16-fold excess of dGTP is added to the firststrand cDNA synthesis to tail them with G bases. The secondstrand cDNA is synthesized with an oligo(dC) adaptor. The cDNA is then amplified by PCR using a primer annealing within the adaptor sequence and a gene-specific primer. The second PCR is run using nested primers. The amplified cDNA fragment contained a full-length variable region including a 50 -untranslated region, a leader sequence, and an initiation codon [64].

Review: Single B Cell Cloning Technology

433

Another novel platform is called “single-cell RT-PCR-linked in vitro expression” (SICREX). SICREX has been used to produce rabbit monoclonal antibodies [66]. In this platform, single specific B cells are isolated, and variable domains coding sequences are amplified by RT-PCR and PCR. Next, a linear expression cassette contains all essential elements for transcriptional and translational regulation, including promoter and terminator, and fragment of VH or VL is made. Monoclonal antibodies are next expressed in a solution containing biomolecular translation machinery extracted from cells (cell-free protein synthesis; CFPS). In this system, protein synthesis occurs in cell lysates rather than within cultured cells and reduces the antibody production time to a few days [66]. 3.5.4 Quantification and Characterization of Generated Monoclonal Antibodies

If the antibody is generated by transfected cells, the supernatants are harvested following appropriate time of incubation to determine the quantity and reactivity profile of the antibodies. In some cases, purification of antibodies is required. The concentration of the secreted recombinant antibodies is measured by quantitative ELISA. Additionally, an antigen-binding ELISA is conducted to determine the specificity of the antibody. The immunogen (synthetic peptide, recombinant polypeptide, or full protein) is immobilized on a solid matrix, and it is determined whether the monoclonal antibody can bind to the specific immunogen [49]. The antibodies are also tested in immunoblotting assays such as western blotting.

3.6 Overview for Generating Rabbit Monoclonal Antibodies by Single B Cell Cloning

The following sections are a detailed overview for generating rabbit monoclonal antibodies through single B cell cloning. Figure 1 summarizes the entire process.

3.6.1 Immunization

New Zealand White Rabbits are immunized with a synthetic peptide coupled to Keyhole limpet hemocyanin (KLH). Usually, more than one sequence is used as the immunogen for a target antibody and at least two rabbits are immunized by the same peptide. The sequence of the peptide must be unique to the target immunogen and has minimum homology with other proteins. There are different immunization protocols, and the entire process may take 70–120 days. Immunization starts with subcutaneous injection of the immunogen prepared as emulsions in complete Freund’s adjuvant (CFA) to induce the immune system. The immunogen is prepared in incomplete Freund’s adjuvant (IFA) for the next boosts. Rabbits are usually bled before the first injection to provide control sera for evaluating the immune system response following injections. A sample of serum is collected 1 week after every boost to test the level of polyclonal antibodies and their function by

434

Juliet Rashidian and Joshua Lloyd

Peptide KLH

1 Immunization

Peripheral Blood

PBMCs Isolation

Rabbit IgG+ B Cells Staining

2

FACS Analysis for Screening Antigen-Specific Single Antigen-Specific B Cells B Cells Sorting

VH LS and HCD Expressing Plasmid

Single B Cell RT-PCR and PCRs

Amplification of VH and VL

Quantitative ELISA

HC Expressing Plasmid

3

Incubation VL LS and LCD Expressing Plasmid

Antigen-Binding ELISA Western Blotting

LC Expressing Plasmid

Cloning VH and VL in Expression Vectors

Immunohistochemistry

Co-Transfection of HC and LC in Mammalian Cells and Antibody Production

Antibody Evaluation Tests

Fig. 1 Diagram summarizing the generation of antigen-specific rabbit monoclonal antibodies by single B cell cloning. PBMCs peripheral blood mononuclear cells, FACS fluorescence-activated cell sorting, RT-PCR reverse transcription polymerase chain reaction, VH variable domain of heavy chain, VL variable domain of light chain, LS leader sequence, HCD heavy-chain constant domain, LCD light-chain constant domain, HC heavy chain, LC light chain, ELISA enzyme-linked immunosorbent assay

ELISA, immunoblotting, or immunohistochemistry (IHC) assays. The rabbit with the best results from these assays (strong specific signal and low background signal) is selected for B cell cloning. 3.6.2 Isolation of PBMCs

A volume of 20 ml of peripheral blood is collected from the rabbit in tubes containing an anticoagulant. The blood should be maintained at room temperature and processed as soon as possible. PBMCs are isolated from the blood by density-gradient centrifugation and frozen in liquid nitrogen.

3.6.3 Staining and Sorting Rabbit-Specific B Cells

The IgG+ B cells are enriched before staining. This is done using micro beads for magnetic labeling and separation of IgG expressing B cells from the PBMC pool. Next, these cells are stained with a viability dye to exclude the dead cells during the flow cytometry process. There are several Live/Dead staining kits from different sources. The kits are based on the reaction of a fluorescent reactive

Review: Single B Cell Cloning Technology

435

dye with cellular proteins. These dyes cannot penetrate live cell membranes, so only cell surface proteins are available to react with the dye, resulting in a dim or even no staining (live cells). On the contrary, the fluorescent dye can permeate the damaged membranes of dead cells and stain both the intracellular and membranous proteins, resulting in a more intense staining. Eventually, these cells are stained with antirabbit IgG antibody and immunogen and are sorted in a 96-well plate. 3.6.4 Cloning VH and VL

The single specific B cells are lysed, and total RNA is released in the wells. The first-strand cDNA is synthesized in each well and used to amplify VH and VL in PCRs. Next, the PCR products are analyzed to determine frameworks (FR) and complementarity-determining regions (CDRs) sequences. The last step is a simple cloning procedure, and VH and VL are cloned separately in a plasmid vector with built-in leader sequence and constant region for heavy chain or light chain.

3.6.5 Transfection

The plasmids with cloned VH and VL, originated from the same well, are co-transfected into mammalian expression cells and incubated for 7 days.

3.6.6 Rabbit IgG Quantification and Characterization

The supernatants of transfected cells are collected 7 days after transfection, and the concentration of monoclonal IgG and its specificity is evaluated by ELISA. Either crude supernatant or purified antibody can be used for both assays. There are several commercial kits available to measure rabbit IgG. The specificity of the antibody is determined by an antigen-binding ELISA. In this assay, a 96-well plate is coated with the peptide used for immunization and is probed with serial dilutions of the supernatants. To analyze the structure of the antibody, both reduced and nonreduced forms of the purified antibody are subjected to protein gel electrophoresis. Two sets of samples are prepared, and a reducing reagent is added to one set and run onto the gel. The size of the bands will be evaluated on Coomassie blue-stained gel. IgG molecular weight is 150 kDa. So, there should be a single band at the 150 kDa position, representing a full and intact IgG molecule for the nonreduced sample. On the other hand, there should be two bands for the reduced sample, one at 50 kDa (heavy chain) and the other at 25 kDa (light chain) representing dissociated chains.

IHC

If the generated monoclonal antibody is intended to be used for IHC, it should be tested on different tissues depending on the antibody.

436

Juliet Rashidian and Joshua Lloyd

3.7 Nonsingle B Cell Cloning

Some alternative platforms have been recently reported to generate monoclonal antibodies from a pool of specific B cells. In the next sessions, we will briefly discuss two of these methods.

3.7.1 HybriFree

The HybriFree platform has been applied to generate monoclonal antibodies from different species. This method does not require single cell sorting and flow cytometry. The antigen-specific B cells are captured on an immunogen-coated solid matrix like a 96-well plastic plate. Next, the plastic-bound cells are lysed, and total RNA is extracted and subjected to cDNA synthesis. The VH and VL are amplified by PCR using a cocktail of primers. These primers contain a linker at the end for constructing a library in the next step. The VH–VL combinatorial libraries are then constructed in a mammalian expression vector by the circular polymerase extension cloning (CPEC) technique. In CPEC, VH and VL are cloned between a leader sequence and a built-in FC sequence (fragment crystallizable region) in a PCR using the linker added to the variable domains from the previous step. This cloning does not need ligation, and it is done in one PCR. The VH and VL are randomly cloned in one vector and the antibody is expressed as scFV-FC structures. In the end, the library pool is transfected into mammalian expression cells [67]. The HybriFree technique is rapid and potentially applicable to any species. The main disadvantage of this method is the random pairing of the VH and VL. Similar to the phage display technique, natural pairing of heavy and light chains is lost, and the antibodies may not be stable.

3.7.2 Next-Generation Sequencing

Another method for generating monoclonal antibodies that has recently emerged utilizes next-generation sequencing (NGS) to analyze the antibody repertoire in response to immunization with a target antigen [68–71]. Different variations of this methodology have been published. However, they all share commonalities of identifying the highest frequency heavy-chain and/or light-chain variable domain sequences following immunization, constructing synthetic candidate antibody expression constructs, and screening candidate monoclonal antibodies for antigen reactivity. Reddy et al. [69] used NGS to analyze the antibody repertoires of immunized mice. Their methods identified the highest frequency heavy-chain and light-chain variable region sequences from a pool of bone marrow plasma cells. Synthetic constructs were created for expressing the highest frequency heavy-chain sequences paired with the highest-frequency light-chain sequences. A total of three separate antigens, six mice (two per antigen), and

Review: Single B Cell Cloning Technology

437

27 recombinantly expressed antibodies were included in their study. Of the 27 candidate antibodies identified through NGS, 21 (78%) were verified to bind to their respective targets by antigen-binding ELISA. Wang et al. [71] used NGS in combination with single B cell cloning to identify the highest-frequency heavy chain and light chain naturally paired combinations from approximately 2.5 3.5
104 B cells expressing antigen-specific antibodies. A total of 14 of the highest-frequency heavy chain and cognate light chain sequence combinations identified in two separate mice (seven per animal) were synthesized, recombinantly expressed and screened for antigen reactivity. All 14 of the recombinantly expressed antibodies displayed antigen binding by ELISA. In a different study, Gray et al. [68] performed NGS of heavychain sequences derived from a pool of antibody-expressing B cells. They selected 35 of the highest-frequency heavy-chain sequences (each having a unique CDR3 sequence) and recombinantly expressed the 35-heavy-chain candidate constructs in combination with a panel of 20 unique germline kappa light-chain sequence constructs. Antigen-binding ELISA screening of the recombinantly expressed antibodies indicated that 17 (48.5%) of the candidate heavy-chain constructs bound the target antigen. Antigen binding was observed when the candidate heavy-chain constructs were expressed in combination with six of the 20 unique germline kappa light-chain constructs. Potential advantages to using NGS for monoclonal antibody development include reduced cost, time saving, and increased throughput. These advantages are most applicable to the approach described by Reddy et al. [69]. Potential disadvantages of this method include the possible loss of natural pairing of heavy- and light-chain components, unless a similar approach to that described by Wang et al. [71] is utilized, in which case the aforementioned advantages would be offset. 3.8

Summary

Over the last few years, single B cell cloning has become an attractive and useful technology to generate monoclonal antibodies for a variety of applications. This technique offers several advantages over the conventional methods of producing antibodies, summarized in Table 2, and is a valuable platform to provide novel antibodies to use in diagnostic, therapeutic, and preventive fields.

438

Juliet Rashidian and Joshua Lloyd

Table 2 Advantages and disadvantages of several platforms used to generate monoclonal antibodies Technique

Advantages

Disadvantages

Hybridoma

1. Natural cognate pairing of heavy chain, and light chain is maintained

1. Lack of a suitable myeloma partner for many species. 2. Low efficiency of cell fusion. 3. Hybridomas are genetically unstable and may stop producing antibody. 3. Time consuming. 4. A continuous risk of contamination

Phage display

1. Can be used to generate antibody from any 1. Natural cognate pairing of heavy chain, and light chain is lost species with known genes of antibodies 2. Allows selection of antibodies with desired 2. It might be challenging to create a good library specificity by controlling panning strategies

B cell immortalization and clonal expansion

1. Natural cognate pairing of heavy chain, and light chain is maintained

1. Requires virus for immortalization 2. Risk of contamination

Single B cell cloning 1. Higher efficiency than hybridoma since a 1. RT-PCR might be challenging with single cells large part of the antibody repertoire is 2. Requires special equipment preserved 2. Natural cognate pairing of heavy chain, and light chain is maintained 3. Antibody gens are cloned and no need to grow B cells 4. A continuous source of antibody with consistent performance 5. Time efficient compared to hybridoma and phage display HybriFree, NGS

1. Natural cognate pairing of heavy 1. Potentially applicable to any species chain, and light chain is lost 2. Antibody genes are cloned, and no need to 2. Requires special equipment grow B cells 3. Time efficient compared to hybridoma 4. A continuous source of antibody with consistent performance

References 1. Kohler G, Milstein C (1975) Continuous cultures of fused cells secreting antibody of predefined specificity. Nature 256:495–497 2. Tansey EM, Catterall PP (1994) Monoclonal antibodies: a witness seminar in contemporary medical history. Med Hist 38:322–327

3. Zhang H, Chen J (2018) Current status and future directions of cancer immunotherapy. J Cancer 9:1773–1781 4. Henricks LM, Schellens JH, Huitema AD et al (2015) The use of combinations of monoclonal antibodies in clinical oncology. Cancer Treat Rev 41:859–867

Review: Single B Cell Cloning Technology 5. Siddiqui MZ (2010) Monoclonal antibodies as diagnostics; an appraisal. Indian J Pharm Sci 72:12–17 6. Chan AC, Carter PJ (2010) Therapeutic antibodies for autoimmunity and inflammation. Nat Rev Immunol 10:301–316 7. Scott AM, Wolchok JD, Old LJ (2012) Antibody therapy of cancer. Nat Rev Cancer 12:278–287 8. Reichert JM (2017) Antibodies to watch in 2017. MAbs 9:167–181 9. Kaplon H, Reichert JM (2018) Antibodies to watch in 2018. MAbs 10:183–203 10. Rossi S, Laurino L, Furlanetto A et al (2005) Rabbit monoclonal antibodies: a comparative study between a novel category of immunoreagents and the corresponding mouse monoclonal antibodies. Am J Clin Pathol 124:295–302 11. Weber J, Peng H, Rader C (2017) From rabbit antibody repertoires to rabbit monoclonal antibodies. Exp Mol Med 49:e305. https://doi. org/10.1038/emm.2017.23 12. Landry JP, Ke Y, Yu GL et al (2015) Measuring affinity constants of 1450 monoclonal antibodies to peptide targets with a microarray-based label-free assay platform. J Immunol Methods 417:86–96 13. Shawler DL, Bartholomew RM, Smith LM et al (1985) Human immune response to multiple injections of murine monoclonal IgG. J Immunol 135:1530–1535 14. Rader C, Ritter G, Nathan S et al (2000) The rabbit antibody repertoire as a novel source for the generation of therapeutic human antibodies. J Biol Chem 275:13668–13676 15. Smith GP (1985) Filamentous fusion phage: novel expression vectors that display cloned antigens on the virion surface. Science 228:1315–1317 16. Murai T, Ueda M, Yamamura M et al (1997) Construction of a starch-utilizing yeast by cell surface engineering. Appl Environ Microbiol 63:1362–1366 17. Rakestraw JA, Aird D, Aha PM et al (2011) Secretion-and-capture cell-surface display for selection of target-binding proteins. Protein Eng Des Sel 24:525–530 18. Roberts RW, Szostak JW (1997) RNA-peptide fusions for the in vitro selection of peptides and proteins. Proc Natl Acad Sci U S A 94:12297–12302 19. Spieker-Polet H, Sethupathi P, Yam PC et al (1995) Rabbit monoclonal antibodies: generating a fusion partner to produce rabbit-rabbit hybridomas. Proc Natl Acad Sci U S A 92:9348–9352

439

20. Yam PC, Knight KL (2014) Generation of rabbit monoclonal antibodies. Methods Mol Biol 1131:71–79 21. Tiller T (2011) Single B cell antibody technologies. New Biotechnol 28:453–457 22. Smith K, Garman L, Wrammert J et al (2009) Rapid generation of fully human monoclonal antibodies specific to a vaccinating antigen. Nat Protoc 4:372–384 23. Clargo AM, Hudson AR, Ndlovu W et al (2014) The rapid generation of recombinant functional monoclonal antibodies from individual, antigen-specific bone marrow-derived plasma cells isolated using a novel fluorescence-based method. MAbs 6:143–159 24. Holzlohner P, Hanack K (2017) Generation of murine monoclonal antibodies by hybridoma technology. J Vis Exp 119. https://doi.org/ 10.3791/54832 25. Raybould TJ, Takahashi M (1988) Production of stable rabbit-mouse hybridomas that secrete rabbit mAb of defined specificity. Science 240:1788–1790 26. Yarmush ML, Gates FT 3rd, Weisfogel DR et al (1980) Identification and characterization of rabbit-mouse hybridomas secreting rabbit immunoglobulin chains. Proc Natl Acad Sci U S A 77:2899–2903 27. Dreher K, Sogn JA, Gates FT 3rd et al (1983) Allotype-defined mRNA for rabbit immunoglobulin H and L chains isolated from rabbitmouse hybridomas. J Immunol 130:442–448 28. Pytela R, Zhu W, Ke Y et al (2005) Fusion partner for production of monoclonal rabbit antibodies US Patent US7429487B2 29. Dessain SK, Adekar SP, Stevens JB et al (2004) High efficiency creation of human monoclonal antibody-producing hybridomas. J Immunol Methods 291:109–122 30. Li J, Sai T, Berger M et al (2006) Human antibodies for immunotherapy development generated via a human B cell hybridoma technology. Proc Natl Acad Sci U S A 103:3557–3562 31. Breitling F, Dubel S, Seehaus T et al (1991) A surface expression vector for antibody screening. Gene 104:147–153 32. McCafferty J, Griffiths AD, Winter G et al (1990) Phage antibodies: filamentous phage displaying antibody variable domains. Nature 348:552–554 33. Barbas CF 3rd, Kang AS, Lerner RA et al (1991) Assembly of combinatorial antibody libraries on phage surfaces: the gene III site. Proc Natl Acad Sci U S A 88:7978–7982 34. Hammers CM, Stanley JR (2014) Antibody phage display: technique and applications. J Invest Dermatol 134:1–5

440

Juliet Rashidian and Joshua Lloyd

35. Pande J, Szewczyk MM, Grover AK (2010) Phage display: concept, innovations, applications and future. Biotechnol Adv 28:849–858 36. Sidhu SS, Li B, Chen Y et al (2004) Phagedisplayed antibody libraries of synthetic heavy chain complementarity determining regions. J Mol Biol 338:299–310 37. Rakonjac J, Russel M, Khanum S et al (2017) Filamentous phage: structure and biology. Adv Exp Med Biol 1053:1–20 38. Ledsgaard L, Kilstrup M, Karatt-Vellatt A et al (2018) Basics of antibody phage display technology. Toxins 10:236. https://doi.org/10. 3390/toxins10060236 39. Carmen S, Jermutus L (2002) Concepts in antibody phage display. Brief Funct Genomic Proteomic 1:189–203 40. Frenzel A, Kugler J, Helmsing S et al (2017) Designing human antibodies by phage display. Transfus Med Hemother 44:312–318 41. Moghaddam A, Borgen T, Stacy J et al (2003) Identification of scFv antibody fragments that specifically recognise the heroin metabolite 6-monoacetylmorphine but not morphine. J Immunol Methods 280:139–155 42. Blokzijl A, Zieba A, Hust M et al (2016) Single chain antibodies as tools to study transforming growth factor-beta-regulated SMAD proteins in proximity ligation-based pharmacological screens. Mol Cell Proteomics 15:1848–1856 43. Zhang Z, Liu H, Guan Q et al (2017) Advances in the isolation of specific monoclonal rabbit antibodies. Front Immunol 8:494. https://doi.org/10.3389/fimmu 44. Starkie DO, Compson JE, Rapecki S et al (2016) Generation of recombinant monoclonal antibodies from immunised mice and rabbits via flow cytometry and sorting of antigen-specific IgG+ memory B cells. PLoS One 11:e0152282. https://doi.org/10. 1371/journal.pone.0152282 45. Pattengale PK, Smith RW, Gerber P (1973) Selective transformation of B lymphocytes by E.B. virus. Lancet 2:93–94 46. Kozbor D, Roder JC (1981) Requirements for the establishment of high-titered human monoclonal antibodies against tetanus toxoid using the Epstein-Barr virus technique. J Immunol 127:1275–1280 47. Beerli RR, Rader C (2010) Mining human antibody repertoires. MAbs 2:365–378 48. Shammah S, Mantovani TL, Dalla-Favera R et al (1993) Generation of human monoclonal antibodies by transformation of lymphoblastoid B cells with ras oncogene. J Immunol Methods 160:19–25

49. Rashidian J, Copaciu R, Su Q et al (2017) Generation and performance of R132H mutant IDH1 rabbit monoclonal antibody. Antibodies 6. https://doi.org/10.3390/ antib6040022 50. Ouisse LH, Gautreau-Rolland L, Devilder MC et al (2017) Antigen-specific single B cell sorting and expression-cloning from immunoglobulin humanized rats: a rapid and versatile method for the generation of high affinity and discriminative human monoclonal antibodies. BMC Biotechnol 17:3. https://doi.org/10. 1186/s12896-016-0322-5 51. Kuppers R, Zhao M, Hansmann ML et al (1993) Tracing B cell development in human germinal centres by molecular analysis of single cells picked from histological sections. EMBO J 12:4955–4967 52. Tiller T, Meffre E, Yurasov S et al (2008) Efficient generation of monoclonal antibodies from single human B cells by single cell RT-PCR and expression vector cloning. J Immunol Methods 329:112–124 53. Lagerkvist AC, Furebring C, Borrebaeck CA (1995) Single, antigen-specific B cells used to generate Fab fragments using CD40-mediated amplification or direct PCR cloning. BioTechniques 18:862–869 54. Jin A, Ozawa T, Tajiri K et al (2011) A Rapid isolation of antigen-specific antibody-secreting cells using a chip-based immunospot array. Nat Protoc 6:668–676 55. Park S, Han J, Kim W et al (2011) Rapid selection of single cells with high antibody production rates by microwell array. J Biotechnol 156:197–202 56. Meijer PJ, Andersen PS, Haahr Hansen M et al (2006) Isolation of human antibody repertoires with preservation of the natural heavy and light chain pairing. J Mol Biol 358:764–772 57. Franz B, May KF Jr, Dranoff G et al (2011) Ex vivo characterization and isolation of rare memory B cells with antigen tetramers. Blood 118:348–357 58. Amanna IJ, Slifka MK (2006) Quantitation of rare memory B cell populations by two independent and complementary approaches. J Immunol Methods 317:175–185 59. Doucett VP, Gerhard W, Owler K et al (2005) Enumeration and characterization of virusspecific B cells by multicolor flow cytometry. J Immunol Methods 303:40–52 60. Di Niro R, Mesin L, Raki M et al (2010) Rapid generation of rotavirus-specific human monoclonal antibodies from small-intestinal mucosa. J Immunol 185:5377–5383

Review: Single B Cell Cloning Technology 61. Tiller T, Busse CE, Wardemann H (2009) Cloning and expression of murine Ig genes from single B cells. J Immunol Methods 350:183–193 62. Kurosawa N, Yoshioka M, Fujimoto R et al (2012) Rapid production of antigen-specific monoclonal antibodies from a variety of animals. BMC Biol 10:80. https://doi.org/10. 1186/1741-7007-10-80 63. Seeber S, Ros F, Thorey I et al (2014) A robust high throughput platform to generate functional recombinant monoclonal antibodies using rabbit B cells from peripheral blood. PLoS One 9:e86184. https://doi.org/10. 1371/journal.pone.0086184 64. Ozawa T, Kishi H, Muraguchi A (2006) Amplification and analysis of cDNA generated from a single cell by 50 -RACE: application to isolation of antibody heavy and light chain variable gene sequences from single B cells. BioTechniques 40:469–470 65. Liao HX, Levesque MC, Nagel A et al (2009) High-throughput isolation of immunoglobulin genes from single human B cells and expression as monoclonal antibodies. J Virol Methods 158:171–179 66. Ojima-Kato T, Hashimura D, Kojima T et al (2015) In vitro generation of rabbit antiListeria monocytogenes monoclonal antibody

441

using single cell based RT-PCR linked cell-free expression systems. J Immunol Methods 427:58–65 67. Kivi G, Teesalu K, Parik J et al (2016) HybriFree: a robust and rapid method for the development of monoclonal antibodies from different host species. BMC Biotechnol 16:2. https://doi.org/10.1186/s12896-016-02326 68. Gray SA, Moore M, VandenEkart EJ et al (2016) Selection of therapeutic H5N1 monoclonal antibodies following IgVH repertoire analysis in mice. Antivir Res 131:100–108 69. Reddy ST, Ge X, Miklos AE et al (2010) Monoclonal antibodies isolated without screening by analyzing the variable-gene repertoire of plasma cells. Nat Biotechnol 28:965–969 70. Saggy I, Wine Y, Shefet-Carasso L et al (2012) Antibody isolation from immunized animals: comparison of phage display and antibody discovery via V gene repertoire mining. Protein Eng Des Sel 25:539–549 71. Wang B, Kluwe CA, Lungu OI et al (2015) Facile discovery of a diverse panel of antiEbola virus antibodies by immune repertoire mining. Sci Rep 5:13926. https://doi.org/ 10.1038/srep13926

INDEX A Affitins........................................................................19–40 Aho7c.........................................................................19–40 Antibody display..................................185, 186, 398–410 Anti-hapten antibodies ........................................ 267–285 Anti-idiotypic antibody........................................ 191–207 2A peptides.................................................. 211, 212, 322

B B cell clonal expansion......................................... 428, 438 Biotinylation .......................................................... 39, 230, 339, 344, 347, 372 BSA-hapten conjugate .................................270, 272–275 Buffer exchange...................................44, 50, 55, 63, 203

C Camelid antibodies ....................................................... 173 cDNA display ............................................................43–75 cDNA synthesis ................................................... 177, 180, 182, 252, 290, 291, 293, 427, 430, 432, 436 Cell panning ......................................................... 303–318 Cell selection .......................................312, 315, 359, 418 Chemical modification...................................45, 335–348 Chicken immune libraries.................................... 289–301 Chinese hamster ovary (CHO) display.............. 398–400, 402, 405, 417, 419

D Deep sequence-coupled biopanning................... 157–170 Deep sequencing ........................ 158, 160, 165–167, 219 Depletion ............................................................. 139, 305, 306, 309, 311–312, 383 Directed evolution ......................v, 43, 47, 225, 235, 303 DNA aptamer ....................................................... 1–17, 75

E Electroporation .................................................80, 83, 85, 86, 91, 98, 104, 118, 123, 129, 136, 179, 183, 187, 195, 213, 216, 291, 295, 296, 339, 341, 342, 367, 391, 393, 400, 404, 413 Epitope binning ..................................255, 260, 262, 265 Error-prone PCR .................................................. 56, 115, 117, 118, 123, 135, 339, 373, 375, 377

Escherichia coli ................................................. v, 8, 20, 25, 28, 30, 36, 37, 39, 73, 74, 80, 81, 83, 89–91, 98, 104, 118–120, 123–125, 129, 144–148, 151, 152, 175, 176, 184, 195, 202, 242, 252, 253, 292, 299, 324, 330, 347, 354, 355, 369, 376, 418, 421, 426, 428, 430

F 293F cells............................................................. 355–357, 359, 360, 367, 369, 373, 377, 378, 386, 392, 394 Fc-hapten conjugate ........................................... 270, 273, 275, 277–279, 281, 284, 285 Flow cytometry .................................................82, 83, 88, 89, 179, 212, 255, 269, 297, 309, 314, 315, 322, 323, 326, 329, 330, 345, 355, 357, 368, 372, 377, 384, 387, 407, 410, 418, 419, 421, 429–431, 434, 436 Fluorescence activated cell sorting (FACS) ............ v, 175, 179, 180, 184, 185, 187, 198, 207, 216, 219, 225, 230, 233, 234, 249–265, 268–270, 272, 273, 276–280, 282–285, 290, 297–300, 335, 337, 343, 345–348, 355, 357, 365, 378, 379, 429, 434 Fragment antigen binding (Fab).................................143, 146, 151, 185, 194, 196, 197, 205–207, 225, 235, 251–253, 255–260, 268, 269, 271, 276–279, 281, 285, 347, 398–403, 407, 410, 411, 416, 418, 419, 427 Full-length antibody display ...............284, 335–348, 418

G Green fluorescent protein (GFP) .................................. 92, 211–219, 353–354, 357, 359, 360, 366, 368, 370, 380–382, 387, 389, 400

H Haptens.............................. 250, 268–275, 279–283, 285 High performance liquid chromatography (HPLC).................................................... 2, 37, 48, 52, 53, 134, 160, 337 His-tag affinity purification ......................................50, 54 Hot fusion cloning................................................. 96, 107 Human serum ..................................................... 117, 131, 134, 157–170, 196, 205, 206

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2070, https://doi.org/10.1007/978-1-4939-9853-1, © Springer Science+Business Media, LLC, part of Springer Nature 2020

443

GENOTYPE PHENOTYPE COUPLING: METHODS

444 Index

AND

Hybridoma technology.......................250, 425, 426, 429 HybriFree ............................................................. 436, 438

I Immunization...................................................v, 173–188, 252, 269, 270, 272–277, 284, 290, 292, 298, 432, 434–436 Immunoglobulin G (IgG) ...........................................144, 146, 158, 159, 161–163, 166, 168, 169, 187, 199, 249, 251, 260, 264, 271, 276, 282, 285, 289, 292, 298, 314, 315, 337, 339, 344–347, 366, 401, 402, 428, 430, 432, 434, 435 Immunoglobulin Y (IgY) .................................... 289, 290 Internal ribosomal entry site (IRES) .................. 352–354 In vitro maturation .............................................. 115–140 In vitro transcription............................................... 23, 25, 27, 30, 34, 38, 47, 51 In vitro translation .................................................. 23, 25, 37, 39, 44, 49, 53, 54

L Labeling .............................................................19, 63, 71, 180, 185, 187, 211, 212, 218, 219, 230, 255, 256, 258, 259, 263, 264, 271, 276–279, 281–285, 292, 297, 314, 332, 339, 376 Lentivirus.............................................................. 367, 377 Library ................................................................. 2, 20, 43, 57, 80, 96, 115, 143, 157, 174, 194, 212, 225, 250, 269, 290, 304, 321, 335, 364, 398, 427 Library packaging.................................... 96, 99, 105, 106 Ligands ............................................................. vi, 1, 3, 20, 58, 70, 72, 130, 193, 204–206, 223, 230, 303–318, 366, 394, 410, 416, 418

M Macrocyclic peptide ................................................95–112 Mammalian surface display.................................. 363–394 Mouse monoclonal antibodies ............................ 424–437 mRNA display ...................................43–46, 50, 304, 321 Multiparameter FACS.......................................... 249–265

N Next-generation sequencing (NGS) ............................. 38, 75, 80, 365, 368–370, 374, 381, 382, 384, 385, 389, 390, 394, 436–438

P Panning.................................................................v, 96, 99, 106–110, 112, 124, 143–149, 151–153, 303–318, 427, 430, 438 Panning in solution.............................................. 143–153 Peptide libraries..........................................57–75, 95–112

PROTOCOLS Phage display ........................................................v, 21, 43, 80, 96, 115–118, 125, 135, 139, 143–153, 157, 158, 166, 175, 268, 269, 290, 304, 306, 321, 426–428, 438 Phagemid .........................................................96–98, 103, 104, 111, 112, 143, 144, 151, 152, 427 Phage titration.......................................99, 107, 146, 148 Photocrosslinking..................................44, 45, 49, 53, 54 Polymerase chain reaction (PCR) ................................... 3, 22, 47, 61, 84, 96, 115, 158, 175, 195, 214, 227, 252, 270, 290, 324, 339, 367, 401, 427 Primer .................................................................. 3, 21, 22, 45, 65, 84, 96, 117, 146, 158, 177, 199, 213, 235, 270, 293, 324, 337, 368, 401, 430 Protein disulfide isomerase (PDI) ..................... 59, 61, 67 Protein-hapten conjugate ................................... 268–270, 272–275, 283 Puromycin linker ............................ 43–55, 59, 61, 67, 72

R Rabbit monocloncal antibodies........................... 423–437 Recombinant expression............ 195, 198–200, 224, 321 Reverse transcription...................................................... 32, 43–45, 49, 54, 58, 63, 69, 430 Ribosomal skipping......................................321–333, 336 Ribosome display ..............................19–40, 43, 304, 426

S Saccharomyces cerevisiae.............................................. v, 24, 32, 175, 179, 183, 225, 227, 235, 240, 242, 252, 253, 291, 309, 321–333, 337, 365 Selection ................................................................ 3, 4, 20, 43, 57, 80, 115, 144, 157, 175, 198, 224, 250, 268, 304, 336, 352, 373, 399, 426 Sequencing ....................................................................v, 8, 22, 24, 29, 30, 38, 75, 80, 84, 86, 99, 100, 104, 107, 137, 166–168, 170, 184, 187, 242, 260, 264, 292, 294, 299, 304, 321, 365–370, 381, 383–386, 388–390, 393, 404, 416, 418, 436, 437 Serum titer........................................................... 252, 270, 273, 275, 276, 283, 284 Shark antibodies ................................................... 191–207 Simultaneous secretion and surface display ....................................................... 321–333 Single B cell cloning............................................. 423–438 Single chain fragment variable (scFv) ................ 143–145, 147, 149, 151, 152, 268, 269, 290–300, 397, 427, 436 Single domain antibodies .................................... 173–188 Sleeping Beauty (SB)............................................. 353–354 Sortase A (SrtA) ........................................................80, 82 Species cross-reactivity ......................................... 250, 251 Sso7d ................................................................20, 21, 331

GENOTYPE PHENOTYPE COUPLING: METHODS

AND

PROTOCOLS Index 445

Stable transfection........................................352–354, 357 Staphylococcal surface display ..................................80–92 Staphylococcus aureus .................................................80–90 Streptavidin-ZZ.................................................... 337, 339 Systematic evolution of ligands by exponential enrichment (SELEX).................................. vi, 1–17

VHH ........................................................ 81, 82, 173–188 Virus-like particles (VLPs).................................. 157–159, 166, 168, 169, 351 VL ................................................................. 80, 252, 427, 429, 430, 432, 433, 435, 436 vNAR ............................................................192–207, 213

T

Y

T-cell receptors (TCRs) ......................192, 223–246, 399 TCR display .........................................225, 228, 243–245 Three-finger (3-F) scaffold ................................ 59, 72, 75 tolA fragment .................................................................. 29 Transposon vector................................................ 351–360

Yeast surface display (YSD) .............................v, 173–188, 196, 211–219, 250, 268, 269, 276, 289–301, 303–318, 322, 324, 335, 336, 366, 426 Yeast transformation ........................................... 183, 244, 294, 295, 300, 324, 327

V VH ................................................................ 80, 174, 187, 252, 340, 412, 427, 429, 430, 432, 433, 435, 436