Genotype Phenotype Coupling: Methods and Protocols [2 ed.] 1071632787, 9781071632789

Table of contents :
Preface
Contents
Contributors
Chapter 1: Quantitative Determination of Staphylococcus aureus Using Aptamer-Based Recognition and DNA Amplification Machinery
1 Introduction
2 Materials
2.1 Strains
2.2 Oligonucleotides
2.3 Media
2.4 Preparation and Modification of Gold Electrode
2.5 Detection of S. aureus
3 Methods
3.1 Cultivation and Preparation of Bacterial Strains
3.2 Preparation of Aptamer/W, AS/RP Duplexes
3.3 Formation of the Circular DNA
3.4 Verification Through Polyacrylamide-Gel Electrophoresis (PAGE)
3.5 Preparation and Modification of Gold Electrode
3.6 Target Incubation and Exo III Enzymatic Hydrolysis
3.7 Rolling Circle Amplification Reaction
3.8 Detection of S. aureus
3.9 Optimization of the Conditions for Electrochemical Detection of S. aureus
3.9.1 Optimization of the Formation of Circular DNA
3.9.2 Optimization of the Complementary Length of Aptamer/W and AS/RP Duplexes
3.9.3 Optimization of the Concentration Ratio of Aptamer/W and AS/RP Duplex
3.9.4 Optimization of the Concentration of Exo III and Reaction Time
3.9.5 Optimization of the Concentration of phi29 DNA Polymerase and Reaction Time
3.10 Field Emission Scanning Electron Microscopy Characterization of DNA Nanoflowers
3.11 Detection of S. aureus Using Aptamer-Based Recognition and DNA Amplification Machinery
3.12 Specificity Studies of the Detection Method
4 Notes
References
Chapter 2: Construction of Synthetic VHH Libraries in Ribosome Display Format
1 Introduction
2 Materials
2.1 Oligonucleotides
2.1.1 Primers Used for Generation of the 5′-Flanking Region of the Ribosome Display Construct and the Randomized Positions of ...
2.1.2 Primers for the Amplification of tolA Linker Encoded by pFP-RDV3
2.1.3 Primers for the Final Assembly of the 5′-Construct and tolA Linker
2.1.4 Primers for PCR to Prepare NGS Samples
2.2 PCR
3 Methods
3.1 Production of Input Library
3.1.1 Production of the VHH Library Fragments
3.1.2 Production of the tolA Fragment
3.1.3 Production of the Ribosome Display Construct
3.1.4 Next-Generation Sequencing
4 Notes
References
Chapter 3: Isolation of Adhirons Specific for Plant Protoplast Membrane Biomarkers Is Simplified by Phagemid Design
1 Introduction
2 Materials
2.1 Protoplast Isolation
2.2 Protoplast Assessments
2.3 Biopanning
2.4 Screening
3 Methods
3.1 Protoplast Isolation
3.2 Biopanning
3.2.1 Coating of Beads with the Antigen (See Note 10)
3.2.2 Bead and Library Blocking
3.2.3 Washing
3.2.4 Phage Elution and Amplification
3.2.5 Phage Production
3.2.6 Phage Precipitation
3.3 Second Round of Biopanning
3.3.1 Phage Extraction for Screening
3.3.2 Bacteria Transformation
3.4 Screening
3.4.1 Colony Amplification of Transformed Bacteria
3.4.2 In-Plate Phage Production
3.4.3 Protoplast Preparation
3.4.4 Blocking and Phage-Antigen Binding
3.4.5 Washing
3.4.6 Analysis by Flow Cytometer
3.4.7 Screening in Triplicates
4 Notes
References
Chapter 4: Facile One-Step Generation of Camelid VHH and Avian scFv Libraries for Phage Display by Golden Gate Cloning
1 Introduction
2 Materials
2.1 cDNA Synthesis and Gene-Specific Amplification of VHH or VH/VL for Library Generation
2.2 Library Construction Phage Display
2.2.1 Plasmids (Table 1)
2.2.2 Buffers, Enzymes, Primers, and Devices for Golden Gate Cloning
3 Methods
3.1 cDNA Synthesis
3.2 Amplification of Natural Variable Antibody Domain Repertoire
3.2.1 Amplification of Camelid Heavy Chain Only Variable Domain Repertoire
3.2.2 Amplification of Chicken Variable Antibody Domain Repertoire
3.3 Golden Gate Cloning for the Generation of Phage Display Libraries
3.3.1 GGC for Camelid VHH Libraries
3.3.2 GGC for Chicken scFv Libraries
4 Notes
References
Chapter 5: Identification of New Antibodies Targeting Tumor Cell Surface Antigens by Phage Display
1 Introduction
2 Materials
2.1 Immunization of Mice and Analysis of Antibody Titer in Serum
2.2 Generation of Murine scFv Antibody Library from Spleen
2.2.1 Preparation of Total RNA from Spleen
2.2.2 Generation of First Strand cDNA by Reverse Transcription
2.2.3 PCR Amplification of V-Regions and Assembly of scFvs
2.3 Phage Display with Cellular Panning
2.4 Whole Cell ELISA
3 Methods
3.1 Immunization of Mice and Analysis of Antibody Titer in Serum
3.2 Generation of Murine scFv Antibody Library from the Spleen
3.2.1 Preparation of Total RNA from the Spleen
3.2.2 Generation of the First Strand cDNA by Reverse Transcription
3.2.3 Amplification of Mouse V-Regions by PCR
3.2.4 Generation of VH and VL Sub-Libraries
3.2.5 Assembly of VH and VL to scFv
3.3 Phage Display with Cellular Panning
3.3.1 Preparation of Phages
3.3.2 Phage Titration
3.3.3 Cellular Panning
3.3.4 Binding Analysis of Polyclonal Phage Preparations by Flow Cytometry
3.4 Characterization of Monoclonal Phage Antibodies
3.4.1 Production of Monoclonal Phages
3.4.2 Whole Cell ELISA
3.5 Sequencing Analyses
3.5.1 Sanger Sequencing of Monoclonal Phages
3.5.2 Next-Generation Sequencing (NGS) of Libraries (Optional)
4 Notes
References
Chapter 6: Phage Display of Bovine Ultralong CDRH3
1 Introduction
2 Materials
2.1 Harvesting Bovine Lymphocytes
2.2 Lymphocyte RNA Extraction
2.3 Reverse Transcription PCR of Lymphocyte RNA
2.4 Primary PCR Amplification of CDRH3
2.5 Secondary PCR Amplification of Ultralong CDRH3
2.6 PCR Restriction Digestion
2.7 Plasmid Digestion and Ligation with Amplified Ultralong CDRH3
2.8 Precipitation of Ligation Product
2.9 Electroporation of E. coli and Plating Library
2.10 Phage Rescue
2.11 Biopanning
2.12 Monoclonal Phage Rescue
2.13 Phage ELISA
3 Method
3.1 Harvesting Bovine Lymphocytes
3.2 Lymphocyte RNA Extraction
3.3 Reverse Transcription PCR of Lymphocyte RNA
3.4 Primary PCR Amplification of CDRH3
3.5 Secondary PCR Amplification of Ultralong CDRH3s
3.6 PCR Restriction Digestion
3.7 Plasmid Construction
3.8 Plasmid Digestion and Ligation with Amplified Ultralong CDRH3´s
3.9 Precipitation of Ligation Product
3.10 Electroporation of E. coli and Plating Library
3.10.1 Electroporation
3.10.2 Titer Plates
3.10.3 Library Plate
3.11 Phage Rescue
3.11.1 Collecting Biomass in Liquid Media
3.11.2 Helper Phage Infection
3.11.3 Phage Precipitation
3.12 Biopanning
3.12.1 Phage Preparation
3.12.2 Streptavidin Beads Preparation
3.12.3 Phage Binding, Washing, and Elution
3.12.4 Infection and Titration
3.13 Monoclonal Phage Rescue
3.14 Phage ELISA
4 Notes
References
Chapter 7: Bacterial Cell Display for Selection of Affibody Molecules
1 Introduction
1.1 Affibody Molecules
1.2 Display of Affibody Molecules on E. coli
1.3 Display of Affibody Molecules on S. carnosus
2 Materials
2.1 MACS of Bacteria-Displayed Affibody Libraries
2.1.1 E. coli Display-Specific Reagents
2.1.2 S. carnosus Display-Specific Reagents
2.1.3 General Reagents
2.1.4 Equipment
2.2 FACS of Bacteria-Displayed Affibody Libraries
2.2.1 Reagents
2.2.2 Equipment
3 Methods
3.1 MACS of Bacteria-Displayed Affibody Libraries
3.2 FACS of Bacteria-Displayed Affibody Libraries
4 Notes
References
Chapter 8: Isolation of Antigen-Specific Unconventional Bovine Ultra-Long CDR3H Antibodies Using Cattle Immunization in Combin...
1 Introduction
2 Materials
2.1 Strains
2.2 Plasmids
2.3 Media and Buffers
2.4 PCR Amplification of Cow UL-CDR3H Regions
2.5 Digestion of the Destination Plasmid
2.6 Transformation of S. cerevisiae BJ5464 with LC Plasmid
2.7 Library Transformation of S. cerevisiae EBY100
2.8 Labeling and Selection of Yeast Cells with Fluorescence-Activated Cell Sorting (FACS)
2.9 Retransformation of Enriched Yeast Library to E. coli
2.10 General Equipment
3 Methods
3.1 General
3.2 Amplification of Bovine Ultra Long CDR3H Domains
3.3 Transformation of BJ5464 Yeast with the Fixed LC Plasmid
3.4 Digestion of the Destination Plasmid
3.5 Transformation of Yeast for CDR3H Library Generation
3.6 Cryopreservation for Long-Term Storage of Yeast
3.7 Mating of the CDR3H Yeast Library with the Fixed LC Clone
3.8 Induction of Expression for Antibody Yeast Surface Display
3.9 Fluorescence-Activated Cell Sorting for Detection and Selection of Antigen Binding Bovine UL-CDR3H-Based Fab Fragments
3.9.1 Staining of the Library and Controls for FACS Analysis
3.9.2 Cell Handling Following FACS Analysis
3.10 Sequencing the Display Vector from Yeast Libraries
3.11 Yeast Lysis and E. coli Transformation
4 Notes
References
Chapter 9: Selection of High-Affinity Heterodimeric Antigen-Binding Fc Fragments from a Large Yeast Display Library
1 Introduction
1.1 Bispecific Antibodies and mAb2 Antibody Molecules
1.2 Directed Evolution of Fcab Libraries
1.3 Fcab Fragment-Antigen Interaction
1.4 Design of Heterodimeric Fcab Libraries
1.5 Heterodimeric Fcab Libraries Constructed with Yeast Mating
1.6 Yeast Display Using Combined Genome-Integrated and Episomal Expression Cassette
1.7 Sorting of Heterodimeric Fcab Library and Identification of an Antigen-Specific Clone
1.8 Future Prospects
2 Materials
2.1 Reagents
2.2 Solutions and Buffers
2.3 Media
2.4 Kits
2.5 Equipment
2.6 Plasmids, Bacterial Strains, Yeast Strains, and Cell Lines
3 Methods
3.1 Construction of the Recipient Strain with the Genome-Integrated Fcab Expression Cassette
3.2 PCR Screening of the Yeast Colonies
3.3 Transformation of the Variant Heterodimer Chain Library
3.3.1 Library-Encoding PCR Fragment Preparation
3.3.2 Recipient Vector
3.3.3 Library Transformation
3.4 Quality Control and Sorting of Heterodimer Fc Yeast Display Libraries
3.4.1 Sequencing of Library Clones
3.4.2 Staining of Displayed Fc Heterodimer Mutants
3.4.3 Selection of Heterodimeric Fc Libraries
3.5 Thermal Stability of Yeast-Displayed Heterodimeric Fcab Fragments
3.6 Expression of Selected Fcab Clones in Mammalian Expression System
4 Notes
References
Chapter 10: A Two-Step Golden Gate Cloning Procedure for the Generation of Natively Paired YSD Fab Libraries
1 Introduction
2 Materials
2.1 Strains
2.2 Plasmids
2.3 Reagents for Nested PCR
2.4 Reagents for Gel Purification of VH-VL Insert
2.5 Reagents for First GGC Step
2.6 Reagents for E. coli Transformation
2.7 Reagents for Plasmid Preparation
2.8 Reagents for Second GGC Step
2.9 Reagents for S. cerevisiae Library Generation
3 Methods
3.1 Introduction of Esp3I Restriction Sites to RT-OE-PCR Product
3.2 Agarose Gel Electrophoresis
3.3 Gel Extraction of the VH-VL Insert
3.4 First GGC Step
3.5 E. coli Transformation
3.6 Plasmid Preparation
3.7 Second GGC Step
3.8 Yeast Transformation for Library Generation
4 Notes
References
Chapter 11: Single-Cell B-Cell Sequencing to Generate Natively Paired scFab Yeast Surface Display Libraries
1 Introduction
1.1 Background
1.2 Overview of Method
2 Materials
2.1 Mouse Immunization
2.2 Single-Cell Sequencing
2.3 Lead Clone Characterization and Refinement
2.4 Library Generation and YSD Expression
3 Methods
3.1 Mouse Immunization
3.1.1 Mouse Immunization
3.1.2 Harvesting B-Cells from Immune-Adapted Mice
3.1.3 Negative Selection for B-Cell Isolation
3.1.4 Positive Selection for B-Cell Isolation
3.2 Single-Cell Sequencing of BCRs
3.2.1 Prepare GEM Reaction Mix
3.2.2 Load the Chromium Next GEM Chip K
3.2.3 Post-GEM-RT Cleanup Using Dynabeads
3.2.4 cDNA Amplification
3.2.5 SPRIselect cDNA Cleanup
3.2.6 cDNA QC and Quantification
3.2.7 V(D)J Amplification 1
3.2.8 SPRIselect Post-V(D)J Amplification 1 Cleanup
3.2.9 V(D)J Amplification 2
3.2.10 SPRIselect Post-V(D)J Amplification 2 Cleanup
3.2.11 Fragmentation, End Repair, and A-Tailing
3.2.12 Adaptor Ligation
3.2.13 SPRIselect Post-Ligation Cleanup
3.2.14 Sample Index PCR
3.2.15 SPRIselect Post-Sample Index PCR Cleanup
3.2.16 GEX Library Fragmentation, End Repair, and A-Tailing
3.2.17 SPRIselect Post-Fragmentation, End Repair, and A-Tailing Cleanup
3.2.18 GEX Adaptor Ligation
3.2.19 SPRIselect Post-Ligation Cleanup
3.2.20 GEX Sample Index PCR
3.2.21 SPRIselect Post-GEX Sample Index PCR Cleanup
3.3 Lead Clone Characterization and Refinement
3.4 Library Generation and YSD Expression
3.4.1 Prepare VH and VL Inserts
3.4.2 Prepare Yeast Expression Vector
3.4.3 Ligation
3.4.4 Transformation
3.4.5 Prepare Plasmid
3.4.6 Transform Clones Into Yeast: Option A-Electroporation
3.4.7 Transform Clones Into Yeast: Option B-EZ-Yeast Transformation
4 Notes
References
Chapter 12: One-Pot Droplet RT-OE-PCR for the Generation of Natively Paired Antibody Immune Libraries
1 Introduction
2 Materials
2.1 Reagents for Preparation of Single-Cell Suspensions from Lymphatic Tissues
2.2 Reagents for B Cell Isolation
2.3 Reagents for B Cell Staining for Flow Cytometry
2.4 Oligonucleotides
2.5 Reagents for Droplet RT-OE-PCR Mix (2x)
2.6 Preparation of B Cells for Droplet Encapsulation
2.7 Droplet Generation
2.8 Microscopic Analysis
2.9 Droplet RT-OE-PCR
2.10 Purification of VH-VL PCR Product
2.11 Nested PCR
3 Methods
3.1 Preparation of Single-Cell Suspensions from Lymphatic Tissues
3.2 B Cell Isolation
3.3 B Cell Staining for Flow Cytometry
3.4 Preparation of Droplet RT-OE-PCR Mix (2x)
3.5 B Cell Preparation for Droplet Encapsulation
3.6 Droplet Generation
3.7 Microscopic Analysis of Droplet Generation and Stability
3.8 Droplet RT-OE-PCR
3.9 Purification of VH-VL PCR Product
3.10 Small-Scale Nested PCR Test
3.11 Large-Scale Nested PCR
4 Notes
References
Chapter 13: Affinity Maturation of the Natural Ligand (B7-H6) for Natural Cytotoxicity Receptor NKp30 by Yeast Surface Display
1 Introduction
2 Materials
2.1 Strains
2.2 Plasmids
2.3 Media
2.4 Reagents for Amplification of Diversified ΔB7-H6 DNA
2.5 Reagents for Destination Vector (pDest) Digestion
2.6 Reagents for ΔB7-H6 Transformation into EBY100
2.7 Reagents for YSD and FACS Analysis
2.8 Equipment
3 Methods
3.1 PCR Amplification of N-Terminal V-Like Domain of B7-H6 (ΔB7-H6) Diversity
3.2 Destination Vector (pDest) Digestion
3.3 Yeast Transformation for Library Generation
3.4 Sequence Analysis of Yeast Cell Display Vector
3.5 Cryopreservation for Long-Term Storage of Yeast Cells
3.6 Induction of ΔB7-H6 Yeast Surface Expression
3.7 Fluorescence-Activated Cell Sorting (FACS) Analysis for the Detection of Yeast Surface Display and NKp30 Binding
3.7.1 Surface Display Control
3.7.2 Library Staining for Affinity Maturation Purposes
3.7.3 Treatment of Sorted Cells After FACS Analysis
4 Notes
References
Chapter 14: Accessing Transient Binding Pockets by Protein Engineering and Yeast Surface Display Screening
1 Introduction
2 Materials
2.1 Protein Engineering
2.1.1 Random Mutagenesis
2.1.2 Site Saturation Mutagenesis
2.1.3 Primers for Protein Randomization
2.2 Yeast Surface Display (YSD) Library Generation and Sorting
2.2.1 Testing Correct Protein Cell Surface Display and Optimal Ligand Concentration for Library Screening
2.2.2 YSD Library Generation
2.2.3 Library Sorting by FACS
2.2.4 Single Clone Analysis
2.3 Production of Identified Protein Variants
3 Methods
3.1 Protein Engineering
3.1.1 Random Mutagenesis
3.1.2 Site Saturation Mutagenesis
3.2 Yeast Surface Display (YSD)
3.2.1 Testing Correct Protein Cell Surface Display and Optimal Ligand Concentration for Library Screening
3.2.2 Yeast Surface Display (YSD) Library Generation
3.2.3 Library Sorting by FACS
3.2.4 Single Clone Analysis
3.3 Production of Identified Protein Variants
3.4 Evaluation of Identified Protein Variants
4 Notes
References
Chapter 15: Tyrosine Phosphorylation Screening on the Yeast Surface by Magnetic Bead Selection and FACS
1 Introduction
2 Materials
2.1 Mutant Library Preparation
2.1.1 Media, Buffers, and Reagents
2.1.2 Equipment and Consumables
2.1.3 Cell Lines
2.2 Flow Cytometry Analysis and FACS of Yeast Displaying Mutant Substrate Libraries
2.2.1 Media, Buffers, and Reagents
2.2.2 Equipment and Consumables
2.2.3 Cell Lines
2.3 Sorting of Yeast Cells Displaying Phosphorylated Mutant Libraries Through Magnetic Bead Selection
2.3.1 Media, Buffers, and Reagents
2.3.2 Equipment and Consumables
2.3.3 Cell Lines
3 Methods
3.1 Mutant Library Preparation
3.1.1 Mutagenic PCR Primer Design
3.1.2 Mutant Substrate Generation Through Mutagenic PCR
3.1.3 Insert Preparation Through Assembly and Extension PCR
3.1.4 Insert DNA Concentration Through Ethanol Precipitation
3.1.5 Plasmid Digest for Substrate Incorporation
3.1.6 Plasmid DNA Concentration Through Ethanol Precipitation
3.1.7 Mutated Substrate Incorporation Through Yeast Electroporation Transformation
3.1.8 Cell Growth of Yeast Harboring Mutated Substrate Libraries
3.1.9 Induction of Protein Expression
3.2 Flow Cytometry Analysis and FACS of Yeast Displaying Mutant Substrate Libraries
3.2.1 Biotinylation of 4G10 Anti-Phosphotyrosine Antibody
3.2.2 Cell Labeling for Flow Cytometry Analysis of Phosphorylated Domains
3.2.3 Evaluation of Displayed Domains Through Flow Cytometry Analysis
3.2.4 Cell Labeling for FACS of Yeast Displaying Phosphorylated Domains
3.2.5 Yeast Surface Display Library Sorting Through FACS
3.2.6 Protein Induction of Yeast Harboring Sorted Mutated Substrate Libraries
3.3 Sorting of Yeast Cells Displaying Phosphorylated Mutant Libraries Through Magnetic Bead Selection
3.3.1 Cell Growth of Yeast Harboring Mutated Substrate Libraries for Magnetic Bead Sorting
3.3.2 Protein Induction of Yeast Harboring Mutated Substrate Libraries for Magnetic Bead Sorting
3.3.3 Biotinylation of 4G10 Anti-Phosphotyrosine Antibody
3.3.4 Magnetic Bead Preparation for Sorting of Phosphorylated Domains
3.3.5 Mutant Library Preparation for Magnetic Bead Sorting of Phosphorylated Domains
3.3.6 Magnetic Bead Sorting of Yeast Cells Displaying Phosphorylated Domains
3.3.7 Protein Induction of Yeast Harboring Magnetic Bead Sorted Mutated Substrate Plasmids
3.3.8 Cell Labeling for Flow Cytometry Analysis of Sorted Phosphorylated Domains
3.3.9 Yeast Surface Display Evaluation Through Flow Cytometry Analysis of Sorted Cells
4 Notes
References
Chapter 16: Bulk Reformatting of Antibody Fragments Displayed on the Surface of Yeast Cells to Final IgG Format for Mammalian ...
1 Introduction
2 Materials
2.1 Yeast Surface Display (YSD) Generation and Sorting
2.1.1 Library Generation for YSD
2.1.2 Library Sorting Using Fluorescence-Activated Cell Sorting (FACS)
2.1.3 Primers for YSD Generation
2.2 Reformatting
2.2.1 Transfer of Entire ORF from YSD to MD Vector
2.2.2 Bidirectional Promoter Exchange for Mammalian Expression
2.3 Production and Characterization
2.3.1 Transient Transfection and Purification of mAbs
3 Methods
3.1 Yeast Surface Display (YSD) Library Generation and Sorting
3.1.1 Library Generation for YSD
3.1.2 Library Sorting by FACS
3.2 Reformatting
3.2.1 Transfer of Both Entire ORFs from YSD to MD Vector
3.2.2 Bidirectional Promoter Exchange for Mammalian Expression
3.3 Production of mAbs
3.3.1 Transient Production and Purification of mAbs
4 Notes
References
Chapter 17: Antibody-Secreting Cell Isolation from Different Species for Microfluidic Antibody Hit Discovery
1 Introduction
2 Materials
2.1 Isolation from Mouse Tissue via MACS
2.2 Isolation from Human Peripheral Blood via MACS
2.3 Isolation from Rat Tissues via FACS
2.4 Storage and Revival of B Cells
2.5 ASC Staining with Cell Tracker or Anti-CD138 Antibody
2.6 Equipment
3 Methods
3.1 Isolation of Antibody-Secreting Cells
3.1.1 Isolation from Mouse Tissue via MACS
3.1.2 Isolation from Human Peripheral Blood via MACS
3.1.3 Isolation from Rat Tissues via FACS
3.2 Storage and Revival of B Cells
3.3 ASC Staining with Cell Tracker or Anti-CD138 Antibody
3.3.1 Staining with Cell Tracker
3.3.2 Anti-CD138 Detection
4 Notes
References
Chapter 18: Efficient Microfluidic Downstream Processes for Rapid Antibody Hit Confirmation
1 Introduction
2 Materials
2.1 Droplet-Based Microfluidic Antibody Discovery Using Cyto-Mine
2.2 Single-Cell Antibody Gene PCR Recovery
2.3 High-Throughput Antibody Reformatting
2.4 High-Throughput Transient Antibody Production
2.5 Single-Cell Sub-Cultivation
2.6 Supernatant Binding Confirmation via Biolayer Interferometry
2.7 Equipment
3 Methods
3.1 Droplet-Based Microfluidic Antibody Discovery Using Cyto-Mine
3.2 Single-Cell Antibody Gene PCR Recovery
3.2.1 cDNA Synthesis
3.2.2 First PCR
3.2.3 Nesting PCR
3.2.4 PCR Analysis via 2% Agarose Gel
3.3 High-Throughput Antibody Reformatting
3.4 High-Throughput Transient Antibody Production
3.5 Single-Cell Sub-Cultivation
3.6 Supernatant Binding Confirmation via Biolayer Interferometry
4 Notes
References
Chapter 19: Cell Line Development Using Targeted Gene Integration into MAR-Rich Landing Pads for Stable Expression of Transgen...
1 Introduction
2 Materials
2.1 Plasmids
2.2 Cell Culture
2.3 Transfection Using Neon Transfection System and Selection
2.4 Clone Isolation
2.5 Assessment of Productivity of Selected Clones and Clone Stability
3 Methods
3.1 Preparation of LP, Donor, and Helper Vectors for Transfection
3.2 Generation of LP Clones
3.3 Isolation of LP Clones
3.4 Evaluation of LP Clone Stability
3.5 Transfection with Donor and Helper Vectors for the Generation of mAb-Expressing Clones
3.6 Isolation and Evaluation of mAb-Expressing Clones
3.7 Evaluation of Product Stability
4 Notes
References
Chapter 20: Generation of Human 293-F Suspension NGFR Knockout Cells Using CRISPR/Cas9 Coupled to Fluorescent Protein Expressi...
1 Introduction
2 Materials
2.1 CRISPR/Cas9-sgRNA Expression Plasmids
2.2 Transient Expression
2.3 Tissue Culture
2.4 Flow Cytometry and Cell Sorting
3 Methods
3.1 Transient Transfection of 293-F Cells
3.2 Determination of Transfection Efficiency
3.3 Analysis of CRISPR/Cas9-Mediated NGFR Knockout
3.4 Depletion of NGFR- and FP-Positive Cells
4 Notes
References
Chapter 21: Antibody Display Technology (ADbody) to Present Challenging and Unstable Target Proteins on Antibodies
1 Introduction
2 Materials
2.1 Software for Construct Design
2.2 Expression Vectors
2.3 Tissue Culture
2.4 IMAC and In-Gel Fluorescent Imaging
3 Methods
3.1 Search Protein Model and Structure-Based Construct Design
3.2 Validate the Designed Model Computationally
3.3 Molecular Cloning
3.4 Small-Scale Protein Expression and Purification
3.4.1 Day 1
3.4.2 Day 2
3.5 IMAC and In-Gel Fluorescent Imaging
4 Notes
References
Chapter 22: SUMO: In Silico Sequence Assessment Using Multiple Optimization Parameters
1 Introduction
2 Materials
3 Methods
3.1 Sequence Annotation and Calculation of In Silico Properties
3.2 MHC-II Binding Predictions
3.3 Sequence Clustering
3.4 Generation of an Overview Table on the In Silico Developability Assessment of Multiple Antibody/VHH Sequences
3.5 Generation of a Detailed View on Specific Relevant Features of the Variable Antibody Sequence
3.6 Generation of PyMOL Session Files
4 Notes
References
Chapter 23: Streamlined Data Analysis Pipeline for Deep Sequence-Coupled Biopanning Identification of Pathogen-Specific Antibo...
1 Introduction
2 Materials
3 Methods
4 Notes
References
Index

Citation preview

Methods in Molecular Biology 2681

Stefan Zielonka · Simon Krah Editors

Genotype Phenotype Coupling Methods and Protocols Second Edition

METHODS

IN

MOLECULAR BIOLOGY

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

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

For over 35 years, biological scientists have come to rely on the research protocols and methodologies in the critically acclaimed Methods in Molecular Biology series. The series was the first to introduce the step-by-step protocols approach that has become the standard in all biomedical protocol publishing. Each protocol is provided in readily-reproducible step-by step 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 Second Edition

Edited by

Stefan Zielonka and Simon Krah Protein Engineering and Antibody Technologies (PEAT), Merck Healthcare KGaA, Darmstadt, Germany

Editors Stefan Zielonka Protein Engineering and Antibody Technologies (PEAT) Merck Healthcare KGaA Darmstadt, Germany

Simon Krah Protein Engineering and Antibody Technologies (PEAT) Merck Healthcare KGaA Darmstadt, Germany

ISSN 1064-3745 ISSN 1940-6029 (electronic) Methods in Molecular Biology ISBN 978-1-0716-3278-9 ISBN 978-1-0716-3279-6 (eBook) https://doi.org/10.1007/978-1-0716-3279-6 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2020, 2023 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Humana imprint is published by the registered company Springer Science+Business Media, LLC, part of Springer Nature. The registered company address is: 1 New York Plaza, New York, NY 10004, U.S.A.

Preface Among therapeutics, monoclonal antibodies (mAbs) and antibody-based molecules are one of the most important drug entities. This became apparent with the FDA-approval of the 100th mAb product in 2021. Since the 1970s, mAbs can be generated by hybridoma technology. For this, B-cells from immunized mice are fused with immortalized tumor cells and selected in HAT medium. Resulting hybridomas secrete antibodies of a defined specificity. In the last decades, tremendous progress in the discovery of antibodies has been made. Several methodologies are covered under the umbrella term “directed evolution,” for which the Nobel Prize was granted in 2018. Directed evolution mimics the process of natural selection and leads to the isolation of variants with desired functions after iterative rounds of in vitro mutagenesis and selection. One important representative in this field is phage display. Herein, filamentous phages that display the protein of interest (POI) and harbor its respective genetic information are generated after infection of phagemid carrying E. coli cells with helper phages. Because (in principle) each phage particle harbors the genetic information for one specific protein that is displayed on its surface, the genotype is linked to its phenotype. Amplification and in vitro mutagenesis allow for the generation of large DNA libraries that can be screened via phage display to identify variants with prescribed properties. In the field of antibody discovery, it enabled the generation of fully human antibodies, while antibodies derived from classical hybridoma campaigns were murine and needed to be humanized tediously. Moreover, phage display was used to further engineer binding proteins in terms of affinity, thermal stability, pH sensitivity, and much more. Besides phage display, other display technologies were developed that all harbor specific advantages and disadvantages. Methods and protocols for many of them have been reported in the first edition of Genotype Phenotype Coupling, including SELEX, cDNA, ribosomal, yeast, and mammalian display. Moreover, discovery and engineering were described not only for classical antibodies but also for antibody fragments such as VHHs and VNARs and alternative binding scaffolds like Affitins. The second edition of Genotype Phenotype Coupling aims at broadening the spectrum given in the first edition. In addition to classical display technologies, methodologies for the generation of natively paired antibody libraries, single cell technologies, alternative scaffolds, and in silico antibody sequence assessments are described. The application of those methods may allow for the generation of improved therapeutics and diagnostic reagents in a shorter time frame in the future. We thank all contributors that helped us to make this handbook with state-of-the-art technologies and methodologies, thereby providing an overview over an exciting and continuously evolving field. Darmstadt, Germany

Stefan Zielonka Simon Krah

v

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

1 Quantitative Determination of Staphylococcus aureus Using Aptamer-Based Recognition and DNA Amplification Machinery . . . . . . . . . . . . . . . . . . . . . . . . . . . . Nandi Zhou and Rongfeng Cai 2 Construction of Synthetic VHH Libraries in Ribosome Display Format . . . . . . . Audrey Guilbaud and Fre´de´ric Pecorari 3 Isolation of Adhirons Specific for Plant Protoplast Membrane Biomarkers Is Simplified by Phagemid Design. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Claudia D’Ercole and Ario de Marco 4 Facile One-Step Generation of Camelid VHH and Avian scFv Libraries for Phage Display by Golden Gate Cloning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Christina Bauer, Elke Ciesielski, Lukas Pekar, Simon Krah, Lars Toleikis, Stefan Zielonka, and Carolin Sellmann 5 Identification of New Antibodies Targeting Tumor Cell Surface Antigens by Phage Display . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Steffen Krohn, Matthias Peipp, and Katja Klausz 6 Phage Display of Bovine Ultralong CDRH3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Callum Joyce, Louise Speight, Alastair D. G. Lawson, Anthony Scott-Tucker, and Alex Macpherson 7 Bacterial Cell Display for Selection of Affibody Molecules. . . . . . . . . . . . . . . . . . . . Charles Dahlsson Leitao, Stefan Sta˚hl, and John Lo¨fblom 8 Isolation of Antigen-Specific Unconventional Bovine Ultra-Long CDR3H Antibodies Using Cattle Immunization in Combination with Yeast Surface Display. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Paul Arras, Jasmin Zimmermann, Britta Lipinski, Desislava Yanakieva, Daniel Klewinghaus, Simon Krah, Harald Kolmar, Lukas Pekar, and Stefan Zielonka 9 Selection of High-Affinity Heterodimeric Antigen-Binding Fc Fragments from a Large Yeast Display Library . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Filippo Benedetti, Gerhard Stadlmayr, Katharina Stadlbauer, ¨ ker, and Gordana Wozniak-Knopp Florian Ru 10 A Two-Step Golden Gate Cloning Procedure for the Generation of Natively Paired YSD Fab Libraries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Lena Vollmer, Simon Krah, Stefan Zielonka, and Desislava Yanakieva 11 Single-Cell B-Cell Sequencing to Generate Natively Paired scFab Yeast Surface Display Libraries. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Nathaniel Pascual, Theodore Belecciu, Sam Schmidt, Athar Nakisa, Xuefei Huang, and Daniel Woldring

vii

v xi

1 19

33

47

61 83

99

113

131

161

175

viii

12

13

14

15

16

17

18

19

20

21

22

Contents

One-Pot Droplet RT-OE-PCR for the Generation of Natively Paired Antibody Immune Libraries. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Desislava Yanakieva, Lena Vollmer, Satyendra Kumar, Stefan Becker, Lars Toleikis, Lukas Pekar, Harald Kolmar, Stefan Zielonka, and Simon Krah Affinity Maturation of the Natural Ligand (B7-H6) for Natural Cytotoxicity Receptor NKp30 by Yeast Surface Display. . . . . . . . . . . . . . . . . . . . . . Stefan Zielonka, Simon Krah, Paul Arras, Britta Lipinski, Jasmin Zimmermann, Ammelie Svea Boje, Katja Klausz, Matthias Peipp, and Lukas Pekar Accessing Transient Binding Pockets by Protein Engineering and Yeast Surface Display Screening . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jorge A. Lerma Romero and Harald Kolmar Tyrosine Phosphorylation Screening on the Yeast Surface by Magnetic Bead Selection and FACS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jose Ezagui and Lawrence A. Stern Bulk Reformatting of Antibody Fragments Displayed on the Surface of Yeast Cells to Final IgG Format for Mammalian Production . . . . . . . . . . . . . . . Stefania C. Carrara, Jan P. Bogen, David Fiebig, Julius Grzeschik, Bjo¨rn Hock, and Harald Kolmar Antibody-Secreting Cell Isolation from Different Species for Microfluidic Antibody Hit Discovery. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ramona Gaa, Qingyong Ji, and Achim Doerner Efficient Microfluidic Downstream Processes for Rapid Antibody Hit Confirmation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ramona Gaa, Hannah Melina Mayer, Daniela Noack, and Achim Doerner Cell Line Development Using Targeted Gene Integration into MAR-Rich Landing Pads for Stable Expression of Transgenes. . . . . . . . . . . . Claudia Oliviero, Steffen C. Hinz, Julius Grzeschik, Bjo¨rn Hock, Harald Kolmar, and Gerrit Hagens Generation of Human 293-F Suspension NGFR Knockout Cells Using CRISPR/Cas9 Coupled to Fluorescent Protein Expression . . . . . . . . . . . . Stefanie Schatz, Femke Harmina van Dijk, Aleksandra Elzbieta Dubiel, Tobias Cantz, Reto Eggenschwiler, and Jo¨rn Stitz Antibody Display Technology (ADbody) to Present Challenging and Unstable Target Proteins on Antibodies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Fu-Lien Hsieh and Tao-Hsin Chang SUMO: In Silico Sequence Assessment Using Multiple Optimization Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Andreas Evers, Shipra Malhotra, Wolf-Guido Bolick, Ahmad Najafian, Maria Borisovska, Shira Warszawski, Yves Fomekong Nanfack, Daniel Kuhn, Friedrich Rippmann, Alejandro Crespo, and Vanita Sood

213

231

249

275

291

313

327

343

361

373

383

Contents

23

ix

Streamlined Data Analysis Pipeline for Deep Sequence-Coupled Biopanning Identification of Pathogen-Specific Antibody Responses in Serum. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 399 Javier Leo, Susan B. Core, and Kathryn M. Frietze

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

407

Contributors PAUL ARRAS • Antibody Discovery and Protein Engineering, Merck Healthcare KGaA, Darmstadt, Germany; Protein Engineering and Antibody Technologies (PEAT), Merck Healthcare KGaA, Darmstadt, Germany CHRISTINA BAUER • Protein Engineering and Antibody Technologies (PEAT), Merck Healthcare KGaA, Darmstadt, Germany STEFAN BECKER • Protein Engineering and Antibody Technologies (PEAT), Merck Healthcare KGaA, Darmstadt, Germany THEODORE BELECCIU • Department of Chemical Engineering and Materials Science, Michigan State University, East Lansing, MI, USA; Institute for Quantitative Health Sciences and Engineering, Michigan State University, East Lansing, MI, USA FILIPPO BENEDETTI • Christian Doppler Laboratory for Innovative Immunotherapeutics, Institute of Molecular Biology, Department of Biotechnology, University of Natural Resources and Life Sciences (BOKU), Vienna, Austria JAN P. BOGEN • Institute for Organic Chemistry and Biochemistry, Technical University of Darmstadt, Darmstadt, Germany; Ferring Darmstadt Laboratories, Darmstadt, Germany AMMELIE SVEA BOJE • Stem Cell Transplantation and Immunotherapy, Division of Antibody-Based Immunotherapy, Department of Medicine II, Christian Albrechts University Kiel and University Medical Center Schleswig-Holstein, Kiel, Germany WOLF-GUIDO BOLICK • Ginkgo Analytics GmbH, Hamburg, Germany MARIA BORISOVSKA • Computational Chemistry & Biologics (CCB), EMD Serono, Billerica, MA, USA RONGFENG CAI • The Key Laboratory of Carbohydrate Chemistry and Biotechnology, Ministry of Education, School of Biotechnology, Jiangnan University, Wuxi, China TOBIAS CANTZ • Research Group Translational Hepatology and Stem Cell Biology, Department of Gastroenterology, Hepatology, and Endocrinology, Hannover Medical School, Hannover, Germany STEFANIA C. CARRARA • Institute for Organic Chemistry and Biochemistry, Technical University of Darmstadt, Darmstadt, Germany; Ferring Darmstadt Laboratories, Darmstadt, Germany TAO-HSIN CHANG • Department of Molecular Biology and Genetics, Johns Hopkins University School of Medicine, Baltimore, MD, USA; Howard Hughes Medical Institute, Baltimore, MD, USA ELKE CIESIELSKI • Protein Engineering and Antibody Technologies (PEAT), Merck Healthcare KGaA, Darmstadt, Germany SUSAN B. CORE • Department of Molecular Genetics and Microbiology, School of Medicine, University of New Mexico, Albuquerque, NM, USA ALEJANDRO CRESPO • Computational Chemistry & Biologics (CCB), EMD Serono, Billerica, MA, USA ARIO DE MARCO • Laboratory of Environmental and Life Sciences, University of Nova Gorica, Rozˇna Dolina, Nova Gorica, Slovenia CLAUDIA D’ERCOLE • Laboratory of Environmental and Life Sciences, University of Nova Gorica, Rozˇna Dolina, Nova Gorica, Slovenia

xi

xii

Contributors

CHARLES DAHLSSON LEITAO • Department of Protein Science, KTH – Royal Institute of Technology, Stockholm, Sweden ACHIM DOERNER • Protein Engineering and Antibody Technologies, Merck Healthcare KGaA, Darmstadt, Germany ALEKSANDRA ELZBIETA DUBIEL • Research Group Medical Biotechnology & Bioengineering, Faculty of Applied Natural Sciences, TH Ko¨ln – University of Applied Sciences, Campus Leverkusen, Leverkusen, Germany RETO EGGENSCHWILER • Research Group Translational Hepatology and Stem Cell Biology, Department of Gastroenterology, Hepatology, and Endocrinology, Hannover Medical School, Hannover, Germany ANDREAS EVERS • Computational Chemistry & Biologics (CCB), Merck Healthcare KGaA, Darmstadt, Germany JOSE EZAGUI • Department of Chemical, Biological and Materials Engineering, University of South Florida, Tampa, FL, USA DAVID FIEBIG • Institute for Organic Chemistry and Biochemistry, Technical University of Darmstadt, Darmstadt, Germany; Ferring Darmstadt Laboratories, Darmstadt, Germany YVES FOMEKONG NANFACK • Computational Chemistry & Biologics (CCB), EMD Serono, Billerica, MA, USA KATHRYN M. FRIETZE • Department of Molecular Genetics and Microbiology, School of Medicine, University of New Mexico, Albuquerque, NM, USA; Clinical and Translational Science Center, University of New Mexico Health Sciences, Albuquerque, NM, USA RAMONA GAA • Protein Engineering and Antibody Technologies, Merck Healthcare KGaA, Darmstadt, Germany JULIUS GRZESCHIK • Ferring Biologics Innovation Centre, Epalinges, Switzerland AUDREY GUILBAUD • Nantes Universite´, Univ Angers, INSERM, CNRS, Immunology and New Concepts in ImmunoTherapy, INCIT, UMR 1302/EMR6001, Nantes, France GERRIT HAGENS • Institute of Life Technologies, Haute Ecole d’Inge´nierie HES-SO Valais Wallis, Sion, Switzerland STEFFEN C. HINZ • Institute of Life Technologies, Haute Ecole d’Inge´nierie HES-SO Valais Wallis, Sion, Switzerland BJO¨RN HOCK • Institute for Organic Chemistry and Biochemistry, Technical University of Darmstadt, Darmstadt, Germany; Aerium Therapeutics, Biopoˆle, Epalinges, Switzerland FU-LIEN HSIEH • Department of Molecular Biology and Genetics, Johns Hopkins University School of Medicine, Baltimore, MD, USA; Howard Hughes Medical Institute, Baltimore, MD, USA XUEFEI HUANG • Institute for Quantitative Health Sciences and Engineering, Michigan State University, East Lansing, MI, USA; Department of Chemistry, Michigan State University, East Lansing, MI, USA QINGYONG JI • Protein Engineering and Antibody Technologies, EMD Serono, Billerica, MA, USA CALLUM JOYCE • Early Solutions, UCB Biopharma UK, Slough, UK KATJA KLAUSZ • Division of Antibody-Based Immunotherapy, Department of Internal Medicine II, University Medical Center Schleswig-Holstein and Christian-AlbrechtsUniversity Kiel, Kiel, Germany; Stem Cell Transplantation and Immunotherapy, Division of Antibody-Based Immunotherapy, Department of Medicine II, Christian Albrechts University Kiel and University Medical Center Schleswig-Holstein, Kiel, Germany

Contributors

xiii

DANIEL KLEWINGHAUS • Antibody Discovery and Protein Engineering, Merck Healthcare KGaA, Darmstadt, Germany HARALD KOLMAR • Institute for Organic Chemistry and Biochemistry, Technical University of Darmstadt, Darmstadt, Germany; Centre for Synthetic Biology, Technical University of Darmstadt, Darmstadt, Germany SIMON KRAH • Protein Engineering and Antibody Technologies (PEAT), Merck Healthcare KGaA, Darmstadt, Germany; Antibody Discovery and Protein Engineering, Merck Healthcare KGaA, Darmstadt, Germany STEFFEN KROHN • Division of Antibody-Based Immunotherapy, Department of Internal Medicine II, University Medical Center Schleswig-Holstein and Christian-AlbrechtsUniversity Kiel, Kiel, Germany DANIEL KUHN • Computational Chemistry & Biologics (CCB), Merck Healthcare KGaA, Darmstadt, Germany SATYENDRA KUMAR • Protein Engineering and Antibody Technologies (PEAT), EMD Serono Research and Development Institute, Billerica, MA, USA ALASTAIR D. G. LAWSON • Early Solutions, UCB Biopharma UK, Slough, UK JAVIER LEO • Department of Molecular Genetics and Microbiology, School of Medicine, University of New Mexico, Albuquerque, NM, USA JORGE A. LERMA ROMERO • Institute for Organic Chemistry and Biochemistry, Technical University of Darmstadt, Darmstadt, Germany BRITTA LIPINSKI • Antibody Discovery and Protein Engineering, Merck Healthcare KGaA, Darmstadt, Germany; Protein Engineering and Antibody Technologies (PEAT), Merck Healthcare KGaA, Darmstadt, Germany; Institute for Organic Chemistry and Biochemistry, Technische Universit€ at Darmstadt, Darmstadt, Germany ¨ JOHN LOFBLOM • Department of Protein Science, KTH – Royal Institute of Technology, Stockholm, Sweden ALEX MACPHERSON • Early Solutions, UCB Biopharma UK, Slough, UK SHIPRA MALHOTRA • Computational Chemistry & Biologics (CCB), EMD Serono, Billerica, MA, USA HANNAH MELINA MAYER • Protein Engineering and Antibody Technologies, Merck Healthcare KGaA, Darmstadt, Germany AHMAD NAJAFIAN • Computational Chemistry & Biologics (CCB), EMD Serono, Billerica, MA, USA ATHAR NAKISA • Institute for Quantitative Health Sciences and Engineering, Michigan State University, East Lansing, MI, USA; Department of Chemistry, Michigan State University, East Lansing, MI, USA DANIELA NOACK • Protein Engineering and Antibody Technologies, Merck Healthcare KGaA, Darmstadt, Germany CLAUDIA OLIVIERO • Institute of Life Technologies, Haute Ecole d’Inge´nierie HES-SO Valais Wallis, Sion, Switzerland NATHANIEL PASCUAL • Department of Chemical Engineering and Materials Science, Michigan State University, East Lansing, MI, USA; Institute for Quantitative Health Sciences and Engineering, Michigan State University, East Lansing, MI, USA FRE´DE´RIC PECORARI • Nantes Universite´, Univ Angers, INSERM, CNRS, Immunology and New Concepts in ImmunoTherapy, INCIT, UMR 1302/EMR6001, Nantes, France MATTHIAS PEIPP • Division of Antibody-Based Immunotherapy, Department of Internal Medicine II, University Medical Center Schleswig-Holstein and Christian-AlbrechtsUniversity Kiel, Kiel, Germany; Stem Cell Transplantation and Immunotherapy, Division

xiv

Contributors

of Antibody-Based Immunotherapy, Department of Medicine II, Christian Albrechts University Kiel and University Medical Center Schleswig-Holstein, Kiel, Germany LUKAS PEKAR • Protein Engineering and Antibody Technologies (PEAT), Merck Healthcare KGaA, Darmstadt, Germany; Antibody Discovery and Protein Engineering, Merck Healthcare KGaA, Darmstadt, Germany FRIEDRICH RIPPMANN • Computational Chemistry & Biologics (CCB), Merck Healthcare KGaA, Darmstadt, Germany FLORIAN RU¨KER • Christian Doppler Laboratory for Innovative Immunotherapeutics, Institute of Molecular Biology, Department of Biotechnology, University of Natural Resources and Life Sciences (BOKU), Vienna, Austria STEFANIE SCHATZ • Research Group Medical Biotechnology & Bioengineering, Faculty of Applied Natural Sciences, TH Ko¨ln – University of Applied Sciences, Campus Leverkusen, Leverkusen, Germany; Institute of Technical Chemistry, Leibniz University Hannover, Hannover, Germany SAM SCHMIDT • Department of Chemical Engineering and Materials Science, Michigan State University, East Lansing, MI, USA; Institute for Quantitative Health Sciences and Engineering, Michigan State University, East Lansing, MI, USA ANTHONY SCOTT-TUCKER • Early Solutions, UCB Biopharma UK, Slough, UK CAROLIN SELLMANN • Protein Engineering and Antibody Technologies (PEAT), Merck Healthcare KGaA, Darmstadt, Germany VANITA SOOD • Computational Chemistry & Biologics (CCB), EMD Serono, Billerica, MA, USA LOUISE SPEIGHT • Early Solutions, UCB Biopharma UK, Slough, UK KATHARINA STADLBAUER • Christian Doppler Laboratory for Innovative Immunotherapeutics, Institute of Molecular Biology, Department of Biotechnology, University of Natural Resources and Life Sciences (BOKU), Vienna, Austria GERHARD STADLMAYR • Christian Doppler Laboratory for Innovative Immunotherapeutics, Institute of Molecular Biology, Department of Biotechnology, University of Natural Resources and Life Sciences (BOKU), Vienna, Austria STEFAN STA˚HL • Department of Protein Science, KTH – Royal Institute of Technology, Stockholm, Sweden LAWRENCE A. STERN • Department of Chemical, Biological and Materials Engineering, University of South Florida, Tampa, FL, USA ¨ JORN STITZ • Research Group Medical Biotechnology & Bioengineering, Faculty of Applied Natural Sciences, TH Ko¨ln – University of Applied Sciences, Campus Leverkusen, Leverkusen, Germany LARS TOLEIKIS • Protein Engineering and Antibody Technologies (PEAT), Merck Healthcare KGaA, Darmstadt, Germany FEMKE HARMINA VAN DIJK • Research Group Medical Biotechnology & Bioengineering, Faculty of Applied Natural Sciences, TH Ko¨ln – University of Applied Sciences, Campus Leverkusen, Leverkusen, Germany LENA VOLLMER • Institute for Organic Chemistry and Biochemistry, Technische Universit€ at Darmstadt, Darmstadt, Germany SHIRA WARSZAWSKI • Computational Chemistry & Biologics, Merck KGaA, Yavne, Israel DANIEL WOLDRING • Department of Chemical Engineering and Materials Science, Michigan State University, East Lansing, MI, USA; Institute for Quantitative Health Sciences and Engineering, Michigan State University, East Lansing, MI, USA

Contributors

xv

GORDANA WOZNIAK-KNOPP • Christian Doppler Laboratory for Innovative Immunotherapeutics, Institute of Molecular Biology, Department of Biotechnology, University of Natural Resources and Life Sciences (BOKU), Vienna, Austria DESISLAVA YANAKIEVA • Antibody Discovery and Protein Engineering, Merck Healthcare KGaA, Darmstadt, Germany; Protein Engineering and Antibody Technologies (PEAT), Merck Healthcare KGaA, Darmstadt, Germany 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 Healthcare KGaA, Darmstadt, Germany; Antibody Discovery and Protein Engineering, Merck Healthcare KGaA, Darmstadt, Germany; Institute for Organic Chemistry and Biochemistry, Technische Universit€ at Darmstadt, Darmstadt, Germany JASMIN ZIMMERMANN • Antibody Discovery and Protein Engineering, Merck Healthcare KGaA, Darmstadt, Germany; Protein Engineering and Antibody Technologies (PEAT), Merck Healthcare KGaA, Darmstadt, Germany; Institute for Organic Chemistry and Biochemistry, Technische Universit€ at Darmstadt, Darmstadt, Germany

Chapter 1 Quantitative Determination of Staphylococcus aureus Using Aptamer-Based Recognition and DNA Amplification Machinery Nandi Zhou and Rongfeng Cai Abstract Staphylococcus aureus (S. aureus) is a common foodborne pathogen that threatens human health and safety. It is significant to develop sensitive detection methods for the monitoring of S. aureus contamination in food and environment. Herein, a novel machinery based on aptamer recognition, DNA walker, and rolling circle amplification (RCA) was designed, which can form unique DNA nanoflower and subsequently detect low-level S. aureus contamination in samples. To this end, two rationally designed DNA duplexes were modified on the surface of the electrode to identify S. aureus through the high affinity between aptamers and S. aureus. Combined with the repeated movement of DNA walker machinery on the electrode surface and RCA technology, a unique DNA nanoflower structure was formed. This can effectively transform the biological information of aptamer recognition of S. aureus into a significantly amplified electrochemical signal. Through reasonable design and optimization of the parameters of each part, the linear response range of the S. aureus biosensor is from 60 to 6 × 107 CFU/mL and the detection limit is as low as 9 CFU/ mL. Key words Staphylococcus aureus, Aptamer, DNA walker, Rolling circle amplification, Electrochemical detection

1

Introduction Staphylococcus aureus (S. aureus) is one of the most common foodborne pathogens. Food contaminated by S. aureus or S. aureus toxin can cause poisoning infections [1]. Traditional microbial detection techniques for S. aureus include microbial detection, instrumental analysis, immunological method, and molecular biology detection. These methods have high detection accuracy and good specificity. However, most of the detection methods require a long detection time, tedious steps, and high cost, which restrict their applications [2]. Biosensors are a kind of self-contained analytical devices consisting of biometric elements and signal

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2681, https://doi.org/10.1007/978-1-0716-3279-6_1, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

1

2

Nandi Zhou and Rongfeng Cai

converters [3]. They are usually superior in their rapid detection and high sensitivity, and thus have great potential in the detection of microorganisms [4, 5]. Aptamers are DNA or RNA obtained in vitro by the systematic evolution of ligands by exponential enrichment (SELEX), which have special spatial structures and can bind to specific targets [6]. The binding of aptamers to targets is similar to that of antigen-antibody binding, both of which can be used for the construction of bioanalytical platforms and are widely applied in the diagnosis and treatment of diseases [7, 8]. The binding of aptamers to targets is usually determined by their specific spatial structure [9]. When designing a biosensor, a single aptamer can bind to a specific number of targets (that is, the ratio of an aptamer to target is 1: n) [10]. However, this recognition mechanism has certain limitations on the sensitivity of detection, and a target molecule can only trigger a specific number of signal molecules [11]. To break through this limitation and improve the sensitivity of biosensors, different signal amplification technologies have been developed, including cyclic amplification [12], enzyme catalysis amplification [13], and nanomaterial-mediated amplification [14]. Among them, cyclic amplification can make the signal changes induced by a single molecule being recycled for multiple times to achieve the effect of signal amplification. In DNA-related biosensors, many amplification strategies have been developed. Commonly used recycling reactions include rolling circle amplification [15], strand displacement amplification [16], and DNA walker-based amplification [17]. Among them, DNA walker-based signal amplification has received extensive attention due to its ease of synthesis, sequence predictability, and programmability [18]. DNA walkers are nanoscale molecular devices driven by environmental stimuli [19], enzymatic reactions [20], or strand displacement reactions [21]. They can perform mechanical cyclic repetitive motions along DNA tracks composed of nucleic acids to achieve signal cascade amplification. Combined with biosensors, the DNA components in the DNA walkers can be triggered by specific targets. Then enzyme-mediated DNA substrate hydrolysis or toehold-mediated strand replacement occurs, leading to DNA hybridization in turn. Target binding-induced DNA walkers can activate hundreds of signaling molecules in response to a single, highly specific binding, providing a promising tool for designing signal amplification methods [22]. Herein, we present a quantitative determination method for S. aureus using aptamer-based recognition and DNA amplification machinery. As shown in Fig. 1, the gold electrode surface is modified with aptamer/DNA walker (W) and auxiliary sequence (AS)/ RCA reaction primer (RP) duplexes. By identifying S. aureus in the sample, aptamer recognition and DNA walker movement are triggered. The DNA walker walks along the electrode surface with the

Aptamer-Based Biosensing Amplification Machinery

3

Fig. 1 Schematic representation of the biosensor for S. aureus based on a DNA walker and DNA nanoflowers. (Reprinted by permission from ACS publications: Cai et al. [23])

help of Exo III and undergoes hydrolysis. Then the circular DNA and phi29 DNA polymerase initiate the rolling circle amplification. DNA nanoflowers with a high specific surface area are formed on the electrode surface due to the self-assembly of the sequence amplification. The loaded electroactive methylene blue provides a high electrochemical signal to realize the determination of S. aureus in the sample.

2

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

Strains

1. Staphylococcus aureus (S. aureus, ATCC 29213). 2. Enterococcus faecium (E. faecium, ATCC 19434). 3. Listeria monocytogenes (L. monocytogenes, BNCC 336877). 4. Pseudomonas aeruginosa (P. aeruginosa, preserved in our lab). 5. Enterococcus faecalis (E. faecalis, ATCC 19433). 6. Salmonella enteritidis (S. enteritidis, BNCC103134).

4

2.2

Nandi Zhou and Rongfeng Cai

Oligonucleotides

1. S. aureus aptamer: 5′- GCAATGGTACGGTACTTCCTCGG CACGTTCTCAGTAGCGCTCGCTGGTCATCCCACAGC TACGTCAAAAGTGCACGCTACTTTGCTAA-3′. 2. RCA reaction primer (RP): 5′-SH-(CH2)6-TTTTTTTTCGA CTCGACTCGACTGGCAAGTACCATTGC-3′ (see Note 1). 3. DNA walker 40 (W40): 5′-SH-(CH2)6-T40- GTAGCGTG CACTTTTGACGTAGCTGTGGGATGACCAGC GATTTTTTTTTT-3′ (see Note 2). 4. Auxiliary sequence 1 (AS1): 5′-T15- TTGCCAGTCGAGTC GAGTCGTCGCTGGTCATCCCA-3′ (see Note 3). 5. DNA walker 35 (W35): 5′-SH-(CH2)6-T40- GTAGCGTG CACTTTTGACGTAGCTGTGGGATGACCTTTTTTTTTT-3′. 6. Auxiliary sequence 2 (AS2): 5′-T15- TTGCCAGTCGAGTC GAGTCGGGTCATCCCACAGCT-3′ (see Note 4). 7. DNA walker 30 (W30): 5′-SH-(CH2)6-T40- TAGCGTG CACTTTTGACGTAGCTGTGGGATTTTTTTTTT-3′. 8. DNA walker 25 (W25): 5′-SH-(CH2)6-T40- TGCACTTTT GACGTAGCTGTGGGATTTTTTTTTT-3′. 9. DNA walker 20 (W20): 5′-SH-(CH2)6-T40- TTTTGACG TAGCTGTGGGATTTTTTTTTT-3′. 10. Auxiliary sequence 25 (AS25): 5′-T15-TTGCCAGTCGAGTC GAGTCGAAAAATCCCACAGCTACGTC-3′ (see Note 5). 11. Auxiliary sequence 20 (AS20): 5′-T15-TTGCCAGTCGAGTC GAGTCGTCCCACAGCTACGTC-3′. 12. Auxiliary sequence 15 (AS15): 5′-T15- TTGCCAGTC GAGTCGTCCCACAGCTACGTC-3′. 13. Auxiliary sequence 10 (AS10): 5′-T15- TTGCCAGTCGTCC CACAGCTACGTC-3′. 14. Circular template 1 (CT1): 5′-P-CTCGACTGGCGCAATGG TACTTGCCAGTCGAGTCGAGTCGGCAATGG TACTTGTTTGCCGACTCGA-3′ (see Note 6). 15. Circular template 2 (CT2): 5′-P- TGCCGACTCGAG CAATGGTACTTGCCAGTCGAGTCGAGTCGGCAATGG TACTTGTTCTCGACTGGC-3′ (see Note 7). 16. Circular template 2 ligation primer (CT2 LP): 5′-TCGAGTC GGCAGCCAGTCGAG-3′ (see Note 8).

2.3

Media

1. LB liquid media: Weigh 10 g peptone, 10 g NaCl, and 5 g yeast powder and transfer to a 1 L glass beaker. Add about 100 mL water to the beaker. Cover the beaker with plastic wrap (see Note 9). Stir with a magnetic stirrer. Make up to 1 L with water. Dispense the liquid media into 250 mL conical flasks, each containing 50 mL. Seal the conical flask with the sterile

Aptamer-Based Biosensing Amplification Machinery

5

filtration air-permeable sealing film and brown paper in turn. Sterilize for 20 min at 121 °C. 2. LB solid media: Weigh 10 g peptone, 10 g NaCl, 5 g yeast powder, 1.5 g agar, and transfer to a 1 L glass beaker. Add about 100 mL water to the beaker. Cover the beaker with plastic wrap. Stir with a magnetic stirrer. Make up to 1 L with water. Dispense the liquid media into 250 mL conical flasks, each containing 50 mL. Seal the conical flask with the sterile filtration air-permeable sealing film and brown paper in turn. Sterilize for 20 min at 121 °C. 2.4 Preparation and Modification of Gold Electrode

1. Piranha solution: Take 1 mL 30% H2O2 and 1 mL 98% H2SO4. Mix them at a ratio of 3:7. Stir with the vortex mixer (see Note 10). 2. 0.5 M H2SO4: Take 40 mL water into a 100 mL glass beaker. Add about 1.3889 mL of 98% H2SO4 into the beaker. Stir with a magnetic stirrer. Make up to 50 mL with water (see Note 11). 3. 10 mM phosphate-buffered saline (PBS, pH 7.0): Weigh 1.560 g NaH2PO4·2H2O and transfer to a 1 L glass beaker. Add about 100 mL water to the beaker. Cover the beaker with plastic wrap. Stir with a magnetic stirrer. Make up to 1 L with water to prepare the 10 mM NaH2PO4 solution. Weigh 3.581 g Na2HPO4·12H2O and transfer to another 1 L glass beaker. Add about 100 mL water to the beaker. Cover the beaker with plastic wrap. Stir with a magnetic stirrer. Make up to 1 L with water to prepare 10 mM Na2HPO4 solution. Weigh 7.455 g KCl and transfer to the third 1 L glass beaker. Mix NaH2PO4 solution and Na2HPO4 solution at the ratio of 38:62 in the third beaker and adjust pH to 7.0 with the two solutions. Stir with a magnetic stirrer. Make up to 1 L with the two solutions. Store at 4 °C until use (see Note 12). 4. 1 M 6-mercapto-1-hexanol (6-MCH) solution: Take about 2.89 mL PBS into a 5 mL centrifuge tube. Add about 0.4 μL 97% 6-MCH into the tube. Stir with vortex mixer (see Note 13).

2.5 Detection of S. aureus

1. 1 mM [Fe(CN)6]4-/3- solution: Weigh 0.0823 g K3[Fe (CN)6], 0.1056 g K4Fe(CN)6, and 1.8638 g KCl and transfer them to a 500 mL glass beaker. Add about 100 mL PBS to the beaker. Cover the beaker with plastic wrap. Stir with a magnetic stirrer. Make up to 250 mL with PBS. 2. 5 × TBE buffer (pH 8.0): Weigh 54 g Tris, 3.72 g Na2EDTA·2H2O, and 27.5 g boric acid, and transfer them to a 1 L glass beaker. Add about 100 mL water to the beaker. Cover the beaker with plastic wrap. Stir with a magnetic stirrer.

6

Nandi Zhou and Rongfeng Cai

Adjust pH to 7.0 with NaOH or HCl solution. Make up to 1 L with water (see Note 14). 3. Gelred nucleic acid staining solution: Weigh 0.585 g NaCl, and transfer to a 250 mL glass beaker. Add about 90 mL water and 30 μL 10,000 × Gelred to the beaker. Cover the beaker with plastic wrap. Stir with a magnetic stirrer. Make up to 100 mL with water. 4. 1 M NaOH: Weigh 4 g NaOH and transfer to a 250 mL glass beaker. Add about 90 mL water to the beaker. Cover the beaker with plastic wrap. Stir with a magnetic stirrer. Make up to 100 mL with water (see Note 15). 5. 0.1 M NaOH: Take about 80 mL water and transfer it to a 250 mL glass beaker. Add about 10 mL 1 M NaOH to the beaker. Cover the beaker with plastic wrap. Stir with a magnetic stirrer. Make up to 100 mL with water. 6. 1 M HCl: Take 50 mL water and transfer it to a 250 mL glass beaker. Add about 8.333 mL of 36–38% HCl to the beaker. Cover the beaker with plastic wrap. Stir with a magnetic stirrer. Make up to 100 mL with water (see Note 15). 7. 0.1 M HCl: Take about 80 mL water and transfer it to a 250 mL glass beaker. Add about 10 mL 1 M HCl to the beaker. Make up to 100 mL with water. 8. Methylene blue solution: Weigh 25.588 mg C16H18N3ClS and transfer to a 100 mL glass beaker. Add about 40 mL PBS to the beaker. Cover the beaker with plastic wrap. Stir with a magnetic stirrer. Make up to 50 mL with PBS. 9. Aqua regia: Take 2.5 mL HNO3 and transfer it to a 50 mL glass beaker. Slowly add 7.5 mL HCl to the beaker. Stir with a glass rod (see Note 16).

3

Methods

3.1 Cultivation and Preparation of Bacterial Strains

1. Take S. aureus competent cell suspension from -40 °C freezer. Thaw at room temperature. 2. Dip the bacterial solution with the inoculation loop and inoculate the bacteria on the sterilized LB solid media by the streak plate method (see Note 17). 3. Cultivate at 37 °C for 24 h at an agitation rate of 220 rpm. 4. Pick a single colony with an inoculating loop and inoculate it in sterilized LB liquid medium (see Note 18). Cultivate to OD600 = 0.2. 5. Centrifuge 1 mL medium at 4 °C for 5 min at 976 g. Discard the supernatant. Add 1 mL PBS into the bacterial precipitation.

Aptamer-Based Biosensing Amplification Machinery

7

Stir with a vortex mixer. Repeat the step and rinse the bacterial precipitation with PBS for three times. Re-dissolve the bacterial precipitation with 1 mL PBS. 6. Serially dilute the bacterial suspension in a tenfold gradient with PBS. 7. Take 100 μL of bacterial suspension and spread it on the plate (see Note 19). 8. Cultivate at 37 °C for 24 h. 9. Count the colonies on each plate (see Note 20). Calculate the final bacterial concentration using the formula: CFU/mL = Colonies (CFU) × Dilution factor/0.1 (mL). 10. Cultivate other bacteria for specificity studies under the same culture conditions. 3.2 Preparation of Aptamer/W, AS/RP Duplexes

1. Take 2 μL of 1 μM W and 4 μL of 1 μM S. aureus aptamer to the 0.5 mL centrifuge tube. Make up to 10 μL with PBS. Stir with a vortex mixer. 2. Take 1 μL of 10 μM RP, and 2 μL of 10 μM AS to the 0.5 mL centrifuge tube. Make up to 10 μL with PBS. Stir with a vortex mixer. 3. Denature the sequences in the two tubes at 95 °C for 10 min and place the tubes in the ice bath for 10 min to form aptamer/ W duplex and AS/RP duplex, respectively (see Note 21).

3.3 Formation of the Circular DNA

1. Take 8 μL of 100 μM CT1, and denature at 95 °C for 10 min. Cool down slowly to 37 °C and incubate at 37 °C for 2 h. 2. DNA circular structure reaction system (40 μL): 8 μL of 100 μM CT1, 1 μL of 40 U/μL T4 DNA ligase, 4 μL 10 × T4 DNA ligase buffer, and 27 μL ddH2O. Stir with a vortex mixer. 3. Incubate the reaction system at 16 °C for 12 h to complete the ligation. 4. Heat the ligation solution at 65 °C for 10 min to terminate the reaction. 5. Store the product at 4 °C for further use.

3.4 Verification Through Polyacrylamide-Gel Electrophoresis (PAGE)

1. Assemble the matching glass plate into the plastic mold (see Note 22). 2. 12% native PAGE gel (6 mL): 2355 μL water, 2400 μL 30% Acryl/Bis solution (29:1), 1200 μL 5 × TBE buffer, 36 μL of 10% ammonium persulfate, and 9 μL TEMED. Blow and mix with a pipette.

8

Nandi Zhou and Rongfeng Cai

3. Cast the gel into the glass plate and insert the comb slowly and smoothly (see Note 23). 4. Take out the gel from the mold after the gel is polymerized (see Note 24). Fix it in the electrophoresis tank. Slowly pull out the comb (see Note 25). 5. Add about 1 L 1 × TBE buffer to the electrophoresis tank to cover the gel plate. 6. Take 5 μL sample and 1 μL 6 × loading buffer, mix with a pipette. Load the mixture along the edge of the sample well. 7. Set the electrophoresis at 120 V for 45 min (see Note 26). 8. Turn off the power of the electrophoresis system. 9. Take out the gel from the tank (see Note 27). 10. Place the gel in a beaker containing Gelred staining solution. Wrap the beaker with tin foil. Shake continuously for 45 min at room temperature (see Note 28). 11. Photograph the electrophoresis results with a gel imager. 3.5 Preparation and Modification of Gold Electrode

1. Soak the gold electrode in piranha solution for 15 min to remove impurities on the surface of the electrode. 2. Rinse the surface of the electrode with ultrapure water. 3. Clean the electrode surface electrochemically with 0.5 M H2SO4 to activate the electrode. Set the sweep range of 0.35 to -1.5 V, scan rate at 0.1 V/s, sample interval of 0.01 V. 4. Clean the gold electrode with ultrapure water. Gently dry the electrode surface with ultrapure nitrogen. 5. Mix the aptamer/W and AS/RP duplexes. Stir with a vortex mixer. 6. Drop 10 μL duplexes mixture on the gold electrode surface. Incubate in the incubator at 30 °C for 12 h to modify the duplexes onto the electrode surface (see Note 29). 7. Rinse the electrode with ultrapure water for 15 s to remove unmodified DNA (see Note 30). 8. Put the modified electrode in [Fe(CN)6]4-/3- solution, and set the scan frequency of 0.1 Hz - 10 kHz. Use electrochemical impedance spectroscopy to characterize the modification process of the electrode. 9. Put each modified electrode into 100 μL of 1 M 6-MCH at room temperature for 60 min to avoid non-specific adsorption of DNA on the electrode surface.

3.6 Target Incubation and Exo III Enzymatic Hydrolysis

1. Slowly rinse the gold electrode surface with PBS. Put the electrode in 100 μL PBS containing different concentrations of S. aureus in the incubator at 37 °C for 60 min.

Aptamer-Based Biosensing Amplification Machinery

9

2. Exo III reaction system (10 μL): 0.75 μL of 10 U/μL Exo III, 1 μL 10 × Exo III buffer, and 8.25 μL ddH2O. Stir with a vortex mixer. 3. Drop 10 μL Exo III reaction system to the surface of the gold electrode, and incubate at 37 °C for 80 min (see Note 31). 3.7 Rolling Circle Amplification Reaction

1. Gently rinse the electrode with PBS after the Exo III digestion reaction. 2. Rolling circle amplification reaction system (100 μL): 2 μL ligated CT, 10 μL 10 × phi 29 DNA polymerase buffer, 1 μL of 10 U/μL phi 29 DNA polymerase, 5 μL of 25 μM dNTPs, 10 μL of 2 mg/mL BSA, and 72 μL methylene blue solution. Stir with a vortex mixer. 3. Incubate the gold electrode into the rolling circle amplification system at 30 °C for 3 h. 4. Gently rinse the gold electrode with PBS to remove excess methylene blue. Dry the electrode with ultrapure nitrogen.

3.8 Detection of S. aureus

1. Blow the PBS with ultrapure nitrogen for 30 min to remove the dissolved oxygen in the buffer. 2. Set the reacted gold electrode as the working electrode, the platinum electrode as the counter electrode, and silver–silver chloride electrode as the reference electrode. 3. Insert the three electrodes into PBS and connect them to the electrochemical workstation. 4. Use differential pulse voltammetry scan to detect the electrochemical signal. Set the scan voltage within -0.5 to 0.1 V.

3.9 Optimization of the Conditions for Electrochemical Detection of S. aureus 3.9.1 Optimization of the Formation of Circular DNA

1. Take 8 μL of 100 μM CT1, and denature at 95 °C for 10 min. Cool down slowly to 37 °C and incubate at 37 °C for 2 h. 2. DNA circular structure reaction system for CT1 (40 μL): 8 μL of 100 μM CT1, 1 μL of 40 U/μL T4 DNA ligase, 4 μL 10 × T4 DNA ligase buffer, and 27 μL ddH2O. Stir with a vortex mixer. 3. Take 8 μL of 100 μM CT2 and 8 μL of 100 μM CT2 LP to the 0.5 mL centrifuge tube. Denature at 95 °C for 10 min, slowly cool down to 37 °C, and incubate at 37 °C for 2 h to make CT2/CT2 LP duplex. 4. DNA circular structure reaction system for CT2 (40 μL): 16 μL CT2/CT2 LP duplex, 1 μL of 40 U/μL T4 DNA ligase, 4 μL 10 × T4 DNA ligase buffer, and 19 μL ddH2O. Stir with a vortex mixer. 5. Use PAGE to observe the effect of two circular DNA formation methods. When the gel concentration is relatively high, circular DNA cannot enter the gel, and the mobility is greatly reduced. Therefore, linear DNA and circular DNA with equivalent

10

Nandi Zhou and Rongfeng Cai

Fig. 2 PAGE image of circular DNA formation. Lane M: DNA marker; lane 1: linear CT1 (5 μM); lane 2: circular CT1 (5 μM); lane 3: CT2 LP (5 μM); lane 4: linear CT2 (5 μM); lane 5: circular CT2 (5 μM). (Reprinted by permission from ACS publications: Cai et al. [23])

molecular weight can be clearly distinguished by highconcentration PAGE (see Fig. 2). 6. Take two circular DNA for the rolling circle amplification reaction in S. aureus detection. Compare the signals of the experimental and control groups. 3.9.2 Optimization of the Complementary Length of Aptamer/W and AS/RP Duplexes

1. Design the aptamer/W and AS/RP duplexes with different lengths of the complementary regions according to the secondary structure (see Note 32). 2. The complementary lengths of aptamer and W are 20, 25, 30, 35, and 40 bps, respectively. The complementary lengths of AS and RP are 10, 15, 20, and 25 bps, respectively. 3. Compare the influence of duplexes with different complementary lengths on the detection and choose the optimal detection condition.

3.9.3 Optimization of the Concentration Ratio of Aptamer/W and AS/RP Duplex 3.9.4 Optimization of the Concentration of Exo III and Reaction Time

1. Set the concentration ratio of aptamer/W and AS/RP duplex as 1:1, 1:2, 1:4, 1:5, and 1:10, respectively. 2. Compare the difference in the detection signals under different concentration ratios. 1. Add 5, 7.5, 10, 12.5, and 15 U Exo III in the Exo III reaction system, respectively. 2. Compare the difference in the detection signals under different concentrations of Exo III.

Aptamer-Based Biosensing Amplification Machinery

11

3. Set the reaction time of Exo III as 40, 60, 80, 100, and 120 min, respectively. 4. Compare the difference in the detection signals under different reaction time. 3.9.5 Optimization of the Concentration of phi29 DNA Polymerase and Reaction Time

1. Add 5, 7.5, 10, 12.5, and 15 U phi29 DNA polymerase in the rolling circle amplification system, respectively. 2. Compare the detection signal and the signal growth rate under different concentrations of phi29 DNA polymerase. 3. Set the rolling circle amplification reaction time as 1, 3, 5, 7, and 9 h, respectively. 4. Compare the detection signal and the signal growth rate under different reaction time.

3.10 Field Emission Scanning Electron Microscopy Characterization of DNA Nanoflowers

1. Remove the protective film from the purchased silicon wafer. Soak it in freshly prepared aqua regia overnight. 2. Wash the silicon wafer with absolute ethanol for three times. Then, wash the silicon wafer with acetone for three times. Dry with ultrapure nitrogen. 3. Drop 5 μL rolling circle amplification product sample in the center of the silicon wafer. 4. Dry the silicon wafer in an oven at 28 °C for 2 h. Rinse the silicon wafer with 3 mL ddH2O (see Note 33). Put it in the desiccator overnight. 5. Characterize the sample by a field emission scanning electron microscope (see Fig. 3).

3.11 Detection of S. aureus Using Aptamer-Based Recognition and DNA Amplification Machinery

1. Modify the gold electrode with aptamer/W and AS/RP duplex. Then, block the electrode with 6-MCH to avoid nonspecific adsorption of DNA on the electrode surface. 2. Drop 100 μL PBS containing 60 to 6 × 108 CFU/mL S. aureus onto each electrode, respectively. Incubate the electrodes at 37 °C for 60 min. 3. Gently wash the electrodes with PBS and dry with ultrapure nitrogen. 4. Drop 10 μL Exo III reaction system on the electrode surface and incubate at 37 °C for 80 min. 5. Gently wash the electrode with PBS and dry with ultrapure nitrogen. 6. Drop 100 μL rolling circle amplification reaction system on the surface of the electrode. Incubate at 30 °C for 3 h. 7. Gently wash the electrode with PBS and dry with ultrapure nitrogen. 8. Use electrochemical impedance spectroscopy to characterize each reaction of the electrode (see Fig. 4).

12

Nandi Zhou and Rongfeng Cai

Fig. 3 FESEM of DNA nanoflowers produced by the RCA reaction with different RCA reaction times: (a) 1, (b) 3, (c) 5, (d) 7, and (e) 9 h. The patterns in the red box of (b) and (c) are the embryonic forms of DNA nanoflowers. (Reprinted by permission from ACS publications: Cai et al. [23])

9. Insert the electrodes into PBS and connect them to the electrochemical workstation for DPV scanning. Record the current response of each electrode (see Fig. 5). 10. Plot the current response to the logarithm of the concentration of S. aureus to obtain the calibration curve, which is y = 0.3206 × logC + 3.3536, R2 = 0.9924, where y represents the current response (μA), C represents the concentration of S. aureus (CFU/mL) (see Fig. 6). The current response has a linear relationship with the logarithm of the concentration of S. aureus in the range from 60 to 6 × 107 CFU/mL. The detection limit is 9 CFU/mL (S/N = 3). 11. Add different concentrations of S. aureus to the sterilized filtered lake water, tap water, and diluted honey to prepare real samples simulating S. aureus contamination. Repeat steps 1–8. Drop 100 μL real sample containing S. aureus on the electrode in step 2. Record the current response and calculate the concentration of S. aureus using the calibration curve. 3.12 Specificity Studies of the Detection Method

1. Cultivate several non-target bacteria such as E. faecium, L. monocytogenes, P. aeruginosa, E. faecalis, and S. enteritidis in LB medium. 2. Repeat steps 1–8 in Subheading 3.11. Add 100 μL of 3 × 109 CFU/mL non-target bacteria in step 2. 3. Compare the current responses of the biosensor for the detection of S. aureus and other non-target bacteria.

Aptamer-Based Biosensing Amplification Machinery

13

Fig. 4 EIS curves of each step: (a) bare electrode; (b) electrode modified with aptamer/W and AS/RP duplexes; (c) electrode modified with aptamer/W and AS/RP duplexes and blocked with 6-MCH; (d) electrode modified with aptamer/W and AS/RP duplexes, blocked with 6-MCH and digested by Exo III; (e) electrode modified with aptamer/W and AS/RP duplexes, blocked with 6-MCH, digested by Exo III, and elongated by phi29 DNA polymerase. The inset shows an enlarged view of curves (a), (b), and (d). EIS was modeled by using Randle’s equivalent circuit: C is the interfacial double-layer capacitance, Rs is the solution resistance, Ret is the electron transfer resistance, and Zw is the Warburg impedance. (Reprinted by permission from ACS publications: Cai et al. [23])

Fig. 5 DPV curves recorded at different concentrations of S. aureus: (a) 60, (b) 6 × 102, (c) 6 × 103, (d) 6 × 104, (e) 6 × 105, (f) 6 × 106, (g) 6 × 107, (h) 1.2 × 108, (i) 3 × 108, and (j) 6 × 108 CFU/mL. (Reprinted by permission from ACS publications: Cai et al. [23])

14

Nandi Zhou and Rongfeng Cai

Fig. 6 Relationship between the current response and the concentration of S. aureus. The inset shows the linear relationship between the current response and the logarithm of S. aureus concentration. (Reprinted by permission from ACS publications: Cai et al. [23])

4 Notes 1. The modified group “SH-(CH2)6” at the 5′ terminal of the sequence is used for modifying the gold electrode via the Au– S bond. 2. The number followed by “W” indicates the length of the complementary region of W and aptamer. The number followed by “T” indicates the repeat number of the base T. 3. AS sequence represents an auxiliary sequence complementary to RP. The 3′ terminal of AS1 is complementary to the 3′ terminal of W40. 4. The 3′ terminal of AS2 is complementary to the 3′ terminal of W35. 5. The number followed by “AS” indicates the length of the complementary region of AS and RP (except AS1 and AS2). 6. The modified group “P” at the 5′ terminal of the sequence is used for circular DNA formation. CT1 was designed for a selfcyclization strategy. 7. CT2 was designed for the primer-based cyclization strategy. 8. CT2 LP sequence represents the CT2 ligation primer complementary to CT2.

Aptamer-Based Biosensing Amplification Machinery

15

9. Adding 100 mL of water after weighing the chemicals can effectively prevent the powder from scattering. 10. Piranha solution needs to be freshly prepared. Add H2O2 dropwise into H2SO4, and the reaction process is accompanied by exothermic heat. Piranha solution is extremely corrosive. When preparing the solution, protective equipment is needed, and prepare it in a fume hood. Piranha solution can explode when mixed with organic solvents. Be careful and slow when adding substances that also have organic groups to the solution. 11. 0.5 M H2SO4 needs to be freshly prepared. Slowly add H2SO4 to H2O, and the reaction process is accompanied by exothermic heat. When preparing this solution, protective equipment is needed, and prepare it in a fume hood. 12. Buffer must be filtered with 0.22 μm membrane before storage. Reconstitute buffers every month to avoid buffer contamination due to microorganisms or other conditions. 13. 6-MCH has a pungent odor. Protective equipment is needed. The solution should be prepared and used in a fume hood. Avoid contact with skin and eyes. Avoid inhalation of vapor or mist droplets. 14. When the TBE buffer is prepared, it needs to be stirred overnight with a magnetic stirrer to prevent precipitation after a period of storage due to incomplete dissolution. Dilute five times when used as the electrophoresis solution. 15. Concentrated HCl or concentrated NaOH can be used first to close the gap from the starting pH to the desired pH. It is then best to use a series of lower concentrations of HCl or NaOH to avoid sudden drops or rises in pH. 16. Aqua regia needs to be freshly prepared. Aqua regia solution is extremely corrosive. When preparing the solution, protective equipment is required and prepare it in a fume hood. 17. Pay attention to avoid puncturing the solid medium when streaking the inoculation loop. 18. When picking individual colonies, pay attention to avoid contamination between colonies. 19. Number the plates according to the dilution factor, and make three replicates for each dilution factor. After adding the diluted bacterial solution, use a sterile spatula to spread the bacterial solution evenly on the plate. Use a sterile spatula for each dilution bacterial dilution. When changing the dilution of the bacterial solution, the spatula needs to be burned and sterilized. Put the smeared plate flat on the table for 20–30 min, so that the bacterial solution penetrates in the medium, and then invert the plate.

16

Nandi Zhou and Rongfeng Cai

20. The plate with colony numbers between 30 and 300 shall be selected for counting. 21. When the DNA sequence is denatured, it is necessary to open the centrifuge tube to exhaust and then cover the centrifuge tube cap. If not, the water vapor brought by the high temperature will lift the cap of the centrifuge tube. Then, the reaction system will evaporate. 22. After assembling the plastic mold, add water to it to check whether the mold is completely sealed. If it is not completely sealed, the prepared gel will leak when poured into it. 23. Combs are available in 0.75 and 1 mm sizes, corresponding to different glass plates. Carefully match the comb and the glass plate. The comb needs to be washed and dried before use, to prevent uneven gel concentration near the sample well. Be gentle when inserting the comb to avoid air bubbles. 24. It usually takes 1–2 h for the gel to coagulate. The coagulation speed becomes slower when the room temperature is lower, thus the amount of TEMED and ammonium persulfate can be appropriately increased. It is not advisable to place the gel for a long time after it has coagulated. It may result in water loss and affect the electrophoresis performance. 25. Pull out the comb slowly and vertically to avoid damaging the sample well. 26. The electrophoresis time can be appropriately adjusted according to the molecular weight of the sample. When the bromophenol blue dye appears at the bottom of the electrophoresis tank, the power can be turned off. 27. The thickness of the gel is small and care should be taken when removing it to avoid breaking the gel. 28. If the high-concentration gel is easy to break during the staining process, the amount of TEMED and amine persulfate can be appropriately reduced under the condition of high room temperature. 29. After the mixture is dropped on the surface of the gold electrode, it is buckled with a 0.5 mL centrifuge tube to avoid evaporation of the solution. 30. When using ultrapure water to clean the gold electrode modified with DNA, avoid rinsing the surface of the gold electrode with water flow, to prevent the damage of the modification of the gold electrode caused by a large impact force. The electrode cleaning can be performed by flowing water over the surface of the gold electrode.

Aptamer-Based Biosensing Amplification Machinery

17

31. After the Exo III reaction system is dropped on the surface of the gold electrode, it is buckled with a 0.5 mL centrifuge tube to avoid evaporation of the solution. 32. The secondary structure and Gibbs free energy predictions of duplexes can be obtained through online analysis services IDT (https://sg.idtdna.com/) and Nupack (http://www.nupack. org/). 33. Avoid direct rinsing when using water to clean the surface of the silicon wafer. The water should flow slowly over the surface of the silicon wafer. References 1. Li H, Tang T, Stegger M, Dalsgaard A, Liu T, Leisner JJ (2021) Characterization of antimicrobial-resistant Staphylococcus aureus from retail foods in Beijing, China. Food Microbiol 93:103603. https://doi.org/10. 1016/j.fm.2020.103603 2. Zhao Y, Xia D, Ma P, Gao X, Kang W, Wei J (2020) Advances in the detection of virulence genes of Staphylococcus aureus originate from food. Food Sci Hum Wellness 9:40–44. https://doi.org/10.1016/j.fshw.2019. 12.004 3. Jeong WJ, Choi SH, Lee HS, Lim YB (2019) A fluorescent supramolecular biosensor for bacterial detection via binding-induced changes in coiled-coil molecular assembly. Sens Actuators B Chem 290:93–99. https://doi.org/10. 1016/j.snb.2019.03.112 4. Mukama O, Wu J, Li Z, Liang Q, Yi Z, Lu X et al (2020) An ultrasensitive and specific point-of-care CRISPR/Cas12 based lateral flow biosensor for the rapid detection of nucleic acids. Biosens Bioelectron 159: 112143. https://doi.org/10.1016/j.bios. 2020.112143 5. Bu T, Yao X, Huang L, Dou L, Zhao B, Yang B et al (2020) Dual recognition strategy and magnetic enrichment based lateral flow assay toward Salmonella enteritidis detection. Talanta 206:120204. https://doi.org/10. 1016/j.talanta.2019.120204 6. Sharifi S, Vahed SZ, Ahmadian E, Dizaj SM, Eftekhari A, Khalilov R et al (2020) Detection of pathogenic bacteria via nanomaterialsmodified aptasensors. Biosens Bioelectron 150:111933. https://doi.org/10.1016/j. bios.2019.111933 7. Zhu A, Ali S, Xu Y, Ouyang Q, Chen Q (2021) A SERS aptasensor based on AuNPs functionalized PDMS film for selective and sensitive detection of Staphylococcus aureus. Biosens

Bioelectron 172:112806. https://doi.org/10. 1016/j.bios.2020.112806 8. Funari R, Chu KY, Shen AQ (2020) Detection of antibodies against SARS-CoV-2 spike protein by gold nanospikes in an opto-microfluidic chip. Biosens Bioelectron 169:112578. https://doi.org/10.1016/j.bios.2020. 112578 9. Odeh F, Nsairat H, Alshaer W, Ismail MA, Esawi E, Qaqish B et al (2019) Aptamers chemistry: chemical modifications and conjugation strategies. Molecules 25:3. https://doi. org/10.3390/molecules25010003 10. Yan M, Bai W, Zhu C, Huang Y, Yan J, Chen A (2016) Design of nuclease-based target recycling signal amplification in aptasensors. Biosens Bioelectron 77:613–623. https://doi. org/10.1016/j.bios.2015.10.015 11. Kim DM, Yoo SM (2020) DNA-modifying enzyme reaction-based biosensors for disease diagnostics: recent biotechnological advances and future perspectives. Crit Rev Biotechnol 40:787–803. https://doi.org/10.1080/ 07388551.2020.1764485 12. Ren R, Bi Q, Yuan R, Xiang Y (2020) An efficient, label-free and sensitive electrochemical microRNA sensor based on target-initiated catalytic hairpin assembly of trivalent DNAzyme junctions. Sens Actuators B Chem 304: 127068. https://doi.org/10.1016/j.snb. 2019.127068 13. Wang JR, Xia C, Yang L, Li YF, Li CM, Huang CZ (2020) DNA nanofirecrackers assembled through hybridization chain reaction for ultrasensitive SERS immunoassay of prostate specific antigen. Anal Chem 92:4046–4052. https://doi.org/10.1021/acs.analchem. 9b05648 14. Han S, Liu W, Zheng M, Wang R (2020) Label-free and ultrasensitive electrochemical DNA biosensor based on urchinlike carbon

18

Nandi Zhou and Rongfeng Cai

nanotube-gold nanoparticle nanoclusters. Anal Chem 92:4780–4787. https://doi.org/10. 1021/acs.analchem.9b03520 15. Huang R, He L, Li S, Liu H, Jin L, Chen Z et al (2020) A simple fluorescence aptasensor for gastric cancer exosome detection based on branched rolling circle amplification. Nanoscale 12:2445–2451. https://doi.org/10. 1039/c9nr08747h 16. Li X, Zhang P, Dou L, Wang Y, Sun K, Zhang X et al (2020) Detection of circulating tumor cells in breast cancer patients by nanopore sensing with aptamer-mediated amplification. ACS Sens 5:2359–2366. https://doi.org/10. 1021/acssensors.9b02537 17. Lv S, Zhang K, Zhu L, Tang D (2020) ZIF-8assisted NaYF4:Yb,Tm@ZnO converter with exonuclease III-powered DNA walker for near-infrared light responsive biosensor. Anal Chem 92:1470–1476. https://doi.org/10. 1021/acs.analchem.9b04710 18. Yang H, Xiao M, Lai W, Wan Y, Li L, Pei H (2020) Stochastic DNA dual-walkers for ultrafast colorimetric bacteria detection. Anal Chem 92:4990–4995. https://doi.org/10.1021/ acs.analchem.9b05149 19. Hu H, Zhou F, Wang B, Chang X, Dai T, Tian R et al (2021) Autonomous operation of 3D

DNA walkers in living cells for microRNA imaging. Nanoscale 13:1863–1868. https:// doi.org/10.1039/d0nr06651f 20. Yin Y, Chen G, Gong L, Ge K, Pan W, Li N et al (2020) DNAzyme-powered three-dimensional DNA walker nanoprobe for detection amyloid beta-peptide oligomer in living cells and in vivo. Anal Chem 92:9247–9256. https://doi.org/10.1021/acs.analchem. 0c01592 21. Oishi M, Saito K (2020) Simple single-legged DNA walkers at diffusion-limited nanointerfaces of gold nanoparticles driven by a DNA circuit mechanism. ACS Nano 14:3477–3489. https://doi.org/10.1021/acsnano.9b09581 22. Wang Y, Wang Y, Liu S, Sun W, Zhang M, Jiang L et al (2021) Toehold-mediated DNA strand displacement-driven super-fast tripedal DNA walker for ultrasensitive and label-free electrochemical detection of ochratoxin A. Anal Chim Acta 1143:21–30. https://doi. org/10.1016/j.aca.2020.11.013 23. Cai R, Zhang S, Chen L, Li M, Zhang Y, Zhou N (2021) Self-assembled DNA nanoflowers triggered by a DNA walker for highly sensitive electrochemical detection of Staphylococcus aureus. ACS Appl Mater Interfaces 13:4905– 4919

Chapter 2 Construction of Synthetic VHH Libraries in Ribosome Display Format Audrey Guilbaud and Fre´de´ric Pecorari Abstract Single-domain antibodies, or VHH, represent an attractive molecular basis to design affinity proteins with favorable properties. Beyond high affinity and specificity for their cognate target, they usually show high stability and high production yields in bacteria, yeast, or mammalian cells. In addition to these favorable properties, their ease of engineering makes them useful for many applications. Until the past few years, the generation of VHH involved the immunization of a Camelidae with the target antigen, followed by a phage display selection using phage libraries encoding the VHH repertoire of the animal blood sample. However, this approach is constrained by the accessibility to the animals, and the output relies on the animal’s immune system. Recently, synthetic VHH libraries have been designed to avoid the use of animals. Here, we describe the construction of VHH combinatorial libraries and their use for the selection of binders by ribosome display, a fully in vitro selection technique. Key words Ribosome display, Synthetic library, Single-domain antibody, VHH, Nanobody

1

Introduction Besides conventional IgGs, Camelidae also produces IgG2 and IgG3 devoid of light chain. Thus, these heavy-chain antibodies recognize their cognate antigen only via their VH domains. The fragment corresponding to the VH domain is called VHH or single-domain antibody (sdAb) and is also known under the commercial name “Nanobody”. Their discovery in 1993 [1] shifted the smallest antibody fragment to retain the antigen-binding specificity of a whole antibody from scFvs (25 kDa) to VHHs (14 kDa). Their small size is associated with a simpler structural organization than that of conventional antibodies (150 kDa), only one polypeptide chain and a unique disulfide bridge. These characteristics make their recombinant production possible with high yields not only in mammalian expression systems but also in cheaper bacterial systems [2]. Moreover, their simple structural organization makes

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2681, https://doi.org/10.1007/978-1-0716-3279-6_2, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

19

20

Audrey Guilbaud and Fre´de´ric Pecorari

them easier to engineer than full-length IgGs. The long CDR3 loop of VHHs allows them to recognize epitopes located in cavities on the surface of proteins, epitopes that are mostly inaccessible to classical antibodies. VHHs are also characterized by favorable properties, such as solubility, stability to pH, and temperatures [3, 4], and are thus attractive for various applications. For instance, VHHs have been a basis for developing a broad range of reagents useful for affinity purification or target capture [5, 6], for helping the crystallization of difficult proteins [7, 8], for imaging cells and tissues [9] or as high-resolution probes for electronic microscopy [10]. As VHHs also exhibit low or no immunogenicity [11, 12], they have been used for in vivo tumor imaging [13, 14] and for the development of a wide variety of immunotherapy approaches (see [15] for a recent review). In 2019, the FDA approved the first VHH, in fact, a VHH dimer, which is the Sanofi’s caplacizumab for the treatment of acquired thrombotic thrombocytopenic purpura [16]. This great interest in VHHs from many research teams has led to the exploration of several ways to generate them. An approach commonly used is to immunize with the target of interest a camelid, often a llama, and then collect blood to generate an immune library from the B lymphocytes [17]. This library is then used to perform phage display selections against the target to enrich for sequences with specificity for the target. Finally, a screening and characterization campaign is driven to identify the most interesting VHH clones regarding the final application. This process has proved successful and the vast majority of VHHs have been obtained in this way. However, it has several drawbacks. In addition to being time consuming and expensive, it is not possible for all laboratories interested in generating VHHs to have access to camelids. Furthermore, the VHHs obtained depend on the output of the animal’s immune system, which may bias the diversity of characteristics of the VHHs obtained. It is also obviously challenging to work with animals when dealing with toxic or non-immunogenic targets. An alternative approach is to use a synthetic library instead of an immune library in combination with selections by phage display [18–20] or yeast display [21]. Usually, these non-immune libraries are designed according to experimental feedback about stable and well-produced VHHs, or according to a consensus framework derived from llama genes. In order to improve the design of these libraries, it is also mandatory to carefully consider the length and composition of CDRs. Such libraries need to represent a much larger diversity than an immune library, at least 109 individual clones, to allow identification of high-affinity VHHs. This is restricting the selection techniques such as phage display or yeast display as they involve a cell-transformation step with the DNA library, a limiting step for the manipulation of large libraries.

VHH Libraries for Ribosome Display Selections

21

This is mainly why several synthetic VHH libraries have recently been designed for their use in the ribosome display selection technique [22, 23]. Being performed entirely in vitro, the ribosome display has the advantage to avoid the limiting step of having cells transformed with the DNA of the libraries. This allows to work with diversities of at least 1012 sequences. In this selection technique, the link between the phenotype and the genotype is maintained by the ribosome [18]. Our group uses this powerful technique for 15 years to identify Affitins with high affinity selected from synthetic libraries [24–26]. McMahon et al. [21] described the design of synthetic VHH libraries (1 × 108 variants) for their use in yeast display selections. In the following sections, we describe the generation of large VHH libraries (>1 × 1012 variants), adapted from McMahon et al., in a format suited to ribosome display.

2 2.1

Materials Oligonucleotides

2.1.1 Primers Used for Generation of the 5′Flanking Region of the Ribosome Display Construct and the Randomized Positions of the Gene Encoding VHH

VH-L1.1: 5′- CTTTAAGAAGGAGATATATCTATGGGATCC CAGGTGCAGCTGCAG GAAAGCGGCGGCGGCCTGGTGCAGGCGGGCGGCAG -3′. VH-L1.2: 5′GCCGCTCGCCGCGCAGCTCAGGCG CAGGCTGCCGCCCGCCTGC-3′. VH-L1.3: 5′CGCGGCGAGCGGCWMTATT TYTNNSNNSNNSNNSATGGGCTGGTATCGCCAGG-3′. VH-L1.4: 5′-TTCGCGTTCTTTGCCCGGCGCCTGGCGATAC CAGCCCAT-3′. VH-L1.5: 5′- CCGGGCAAAGAACGCGAAYTTGTTGCCR STATTRVTNNSGGTRSTANTACCWATTATGCGGA TAGCGTGAAAGGCC-3′. VH-L1.6: 5′GTTTTTCGCGTTATCGCGGCTAATGG TAAAGCGGCCTTTCACGCTATCCGCATA-3′. VH-L1.7 5′AGCCGCGATAACGCGAAAAACACCGTG TATCTGCAGATGAACAGCCTGAAACC-3′. VH-L1.8 5′CGCGCAATAATACACCGCGG TATCTTCCGGTTTCAGGCTGTTCATCTGCAGA-3′. VH-L1.9a 5′CCGCGGTGTATTATTGCGCG GYTNNSNNSNNSNNSNNSNNSNNSYWTNN STATTGGGGCCAGGGCACC-3′. VH-L1.9b 5′CCGCGGTGTATTATTGCGCG GYTNNSNNSNNSNNSNNSNNSNNSNNSNNSNNSNN SYWTNNSTATTGGGGCCAGGGCACC-3′.

22

Audrey Guilbaud and Fre´de´ric Pecorari

VH-L1.9c 5′CCGCGGTGTATTATTGCGCG GYTNNSNNSNNSNNSNNSNNSNNSNNSNNSNNSNNS NNSNNSNNSNNSYWTNNSTATTGGGGCCAGGGCACC3′. VH-L1.10: 5′GAATTCGGCCCCCGAGGCCATA TAAAGCTTGCCGCTGCTCACGGT CACCTGGGTGCCCTGGCCCCAATA-3′. T7C: 5′-ATACGAAATTAATACGACTCACTATAGGGAGACCA CAACGGTTTCCCTC-3′. SDA_RDV3: 5′- AGACCACAACGGTTTCCCTCTAGAAA TAATTTTGTTTAACTTTAAGAAGGAGATATATCTATG 3′. AF-link-R: 5′-GAATTCGGCCCCCGAGGCCATATAAAGC-3′. 2.1.2 Primers for the Amplification of tolA Linker Encoded by pFP-RDV3

AF-link-F: 5′- AAGCTTTATATGGCCTCGGGGGCCGAATTC 3′. AF-link-R: 5′-GAATTCGGCCCCCGAGGCCATATAAAGC-3′. TolAkurz: 5′CCGCACACCAGTAAGGTGTGCGGTTT CAGTTGCCGCTTTCTTTCT-3′.

2.1.3 Primers for the Final Assembly of the 5′Construct and tolA Linker

T7B: 5′-ATACGAAATTAATACGACTCACTATAGGGAGACCA CAACGG-3′.

2.1.4 Primers for PCR to Prepare NGS Samples

NGS-VHH_Fint: 5′- TCGTCGGCAGCGTCAGATGTGTATAA GAGACAGCTGAGCTGCGCGGCGAGC-3′.

TolAkurz: 5′CCGCACACCAGTAAGGTGTGCGGTTT CAGTTGCCGCTTTCTTTCT-3′.

NGS_VHH_Rint: 5′- GTCTCGTGGGCTCGGAGATGTGTA TAAGAGACAGGGGTGCCCTGGCCCCAATA-3′. 2.2

PCR

1. UHP water (various suppliers). 2. Phusion DNA Polymerase (Thermo Fisher Scientific). 3. dNTP solution: 10 mM of each dNTP. 4. 1 kb Plus DNA ladder. 5. 6× Orange DNA loading dye or equivalent. 6. Agarose. 7. GelRed nucleic acid gel stain (Biotium). 8. 1× TAE: 40 mM Tris–HCl base, 20 mM acetic acid, and 1 mM EDTA. 9. Wizard SV Gel and PCR Clean-Up System (Promega).

VHH Libraries for Ribosome Display Selections

3

23

Methods The VH-L1 library described here is adapted from the work of McMahon et al. [21]. This library corresponds to the randomization of CDR1, CDR2, and CDR3. The latter one has three lengths with 7, 11, and 15 randomized codons, resulting in libraries VH-L1a, VH-L1b, and VH-L1c, respectively. The randomized codon positions are the same as those described by McMahon et al. [21] (Fig. 1). Similarly to McMahon et al., to synthesize the library we perform a PCR to assemble a combination of nine non-degenerated sequences and three degenerated oligonucleotides that include WMT, TYT, YTT, RST, RVT, ANT, WAT, GYT, and YWT triplets to encode restricted sets of randomized amino acids. For the representation of the 20 amino acids, we use the NNS triplets instead of the trimer phosphoramidites to synthesize oligonucleotides used by McMahon et al., as these reagents are quite expensive and not accessible to all laboratories. The complete assembly of the VH-L1 library is performed by a final PCR to obtain sequences into the ribosome display format. It includes flanking regions with a T7 promoter, a ribosome binding site, a tolA linker, and stem loops at the ends to stabilize the construct, necessary for selection by ribosome display (Fig. 2a; see Notes 1 and 2). The tolA linker fragment is amplified by PCR from the ribosome display plasmid pFP-RDV3 (Fig. 2b; see Note 3) which encodes the part of the Escherichia coli tolA gene that is needed for the ribosome display construct [26]. The VH-L1 libraries (a, b, and c) thus obtained are validated by the next generation sequencing. The final DNA product does not contain a stop codon, and as a

Fig. 1 Scheme representing the CDR randomization strategy used for the synthesis of libraries. Triplets and the corresponding sets of encoded amino acids are indicated under CDR. The NNS triplet encodes the whole set of amino acids, which is represented by a “#”. (Adapted from McMahon et al. [21])

24

Audrey Guilbaud and Fre´de´ric Pecorari

Fig. 2 Schemes of the ribosome display constructs. (a) Scheme of the VH-L1a library in the ribosome display format with a zoom on the VHH region depicting the primers used to generate the VHH gene with randomized CDR. (b) Vector pFP-RDV3. It contains a β-lactamase gene for ampicillin resistance. The 487-bp region (152–638) possesses all functional regions required for cloning, in vitro transcription, and translation. A

VHH Libraries for Ribosome Display Selections

25

result the corresponding mRNAs, as this is important for ribosome to stall after in vitro translation of the library [27]. Thus, under ribosome display conditions, mainly high concentration of Mg2+ at 4 °C, the link between phenotype and genotype is created via the ternary complexes translated VHH-ribosome-mRNA, which are essential for selections. 3.1 Production of Input Library 3.1.1 Production of the VHH Library Fragments

1. The DNA product containing the randomized VHH gene is obtained by two distinct PCRs. The first one uses a combination of eight standards (T7C, SDA_RDV3, VH-L1.1, VH-L1.2, VH-L1.4, VH-L1.6, VH-L1.7, and VH-L1.8) and two degenerated oligonucleotides encoding wobble triplets (VH-L1.3 and VH-L1.5). The second PCR consists of the assembly of the degenerated (VH-L1.9a, b, or c) and the standard (VH-L1.10) oligonucleotides. The two products are then assembled by PCR to obtain the VHH fragment library (see Note 4). 2. For the first PCR, prepare 200 μL PCR mixture in PCR tubes (50 μL/tube) containing 8 pmol of each internal primer (0.8 μL of 10 μM primer SDA_RDV3, VH-L1.1, VH-L1.2, VH-L1.4, VH-L1.6, and VH-L1.7), 40 pmol of each external primer (4 μL of 10 μM primer T7C and VH-L1.8), 4 μL of dNTPs mix (containing 10 mM of each dNTP), 40 μL of 5× Phusion-GC buffer, and 4 U of Phusion DNA polymerase. 3. Use a thermocycler to perform the following PCR program: an initial denaturation step at 98 °C for 30 s, followed by 35 cycles of 98 °C for 10 s, 65 °C for 30 s, 72 °C for 20 s with a final elongation step of 72 °C for 5 min. 4. For the second PCR, prepare 200 μL PCR mixture in PCR tubes (50 μL/tube) containing 40 pmol of VH-L1.9a and VH-L1.10 primers (4 μL of 10 μM primer), 4 μL of dNTPs mix (containing 10 mM of each dNTP), 40 μL of 5× PhusionGC buffer, and 4 U of Phusion DNA polymerase. Prepare two more PCR mixes in the same way, replacing VH-L1.9a primer with VH-L1.9b or VH-L1.9c primer. 5. Use a thermocycler to perform the following PCR program: an initial denaturation step at 98 °C for 30 s, followed by 35 cycles of 98 °C for 10 s, 63 °C for 30 s, 72 °C for 15 s with a final elongation step of 72 °C for 5 min.

ä Fig. 2 (continued) BamHI/HindIII cloning cassette (244–275) encoding four stop codons in different reading frames is useful for sub-cloning DNA outputs during selection. Upstream of the VHH gene are located a ribosome binding site (RBS, 227–232), a hairpin-loop (178–198), and a T7 promoter (162–177). As an N-terminal fusion may be problematic for target recognition, the FLAG tag is located downstream of the VHH gene (315–338). After the FLAG tag is located a “tether” region, TolA (363–638) ending with a hairpin loop region (616–638)

26

Audrey Guilbaud and Fre´de´ric Pecorari

6. Prepare a 1.5% agarose gel. Mix 5 μL of each of the PCR reactions with 1 μL of 6× Orange DNA loading dye buffer and load these samples on the gel. On an adjacent lane, load 5 μL of 1 kb Plus DNA ladder and run the gel at 110 V for at least 40 min. 7. Image the gel and if the correct products corresponding to the expected sizes of 386 and 119 bps (131 and 143 bps for libraries b and c, respectively) are observed for the first and second PCR, respectively, load the rest of the PCR mixtures on separate 1.5% agarose gels to avoid cross-contaminations and excise the bands. 8. Purify the DNA product using Promega Wizard SV PCR and Gel purification kit according to the manufacturer’s specifications. Elute DNA with 38 μL UHP water. 9. Determine the concentration of the purified PCR products by UV absorbance. About 2.5 and 1.1 μg of purified DNA should be obtained for the first and second PCR, respectively. 10. To assemble DNA fragments from step 8, for each library, prepare 300 μL PCR mixture in PCR tubes (50 μL/tube) containing 450 and 150 ng of DNA from the first and second PCR, respectively, 150 pmol of each T7b and AF-link-R primer (15 μL of 10 μM primer), 6 μL of dNTPs mix (containing 10 mM of each dNTP), 60 μL of 5× Phusion-GC buffer, and 6 U of Phusion DNA polymerase. 11. Use a thermocycler to perform the following PCR program: an initial denaturation step at 98 °C for 30 s, followed by eight cycles of 98 °C for 10 s, 45 °C for 30 s, 72 °C for 20 s, followed by 30 cycles of 98 °C for 30 s, 55 °C for 30 s, 72 °C for 25 s with a final elongation step of 72 °C for 5 min. 12. Control the product on 1.5% agarose gel as in step 6. 13. Image the gel and if the correct products corresponding to the expected sizes of 485, 497, and 509 bps (for libraries a, b, and c, respectively) are observed, load the rest of the PCR mixtures on separate 1.5% agarose gels to avoid cross-contaminations and excise the bands. 14. Purify the DNA product as in step 8. 15. Determine the concentration of the purified PCR products by UV absorbance. About 2.6 μg of purified DNA should be obtained (see Note 5). 3.1.2 Production of the tolA Fragment

1. The tolA spacer is obtained in large quantities via PCR amplification from pFP-RDV3 vector (see Note 3). Prepare 500 μL PCR mixture in PCR tubes (50 μL/tube) containing 250 pmol of each primer (2.5 μL of 100 μM primer AF-link-F and tolAkurz), 250 ng of pFP-RDV3 vector (2.5 μL of 100 ng/μL),

VHH Libraries for Ribosome Display Selections

27

10 μL of dNTPs mix (containing 10 mM of each dNTP), 100 μL of 5× Phusion HF buffer, and 10 U of Phusion DNA polymerase. 2. Use a thermocycler to perform the following PCR program: 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 20 s, and final elongation at 72 °C for 5 min. 3. Control the product on 1.0% agarose gel as in step 6. The PCR should give an amplicon of 339 bps. 4. Purify the DNA product as in step 8. 5. Determine the concentration of the tolA linker by UV absorbance and store the product at -20 °C. About 3.2 μg of purified DNA should be obtained. 3.1.3 Production of the Ribosome Display Construct

1. For each library, prepare 500 μL PCR mixture in PCR tubes (50 μL/tube) containing 250 pmol of each primer (2.5 μL of 100 μM primer T7B and TolAkurz), 600 ng of VHH library from Subheading 3.1.1, 525 ng of tolA linker from Subheading 3.1.2, 10 μL of dNTPs mix (containing 10 mM of each dNTP), 100 μL of 5× Phusion HF buffer, and 10 U of Phusion polymerase. 2. Use a thermocycler to perform 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. Control the product on 1.5% agarose gel as in Subheading 3.1.1, step 6. The PCR should give amplicons of 824, 836, and 848 bps. 4. Purify the DNA product as in Subheading 3.1.1, steps 7 and 8. 5. Determine the concentration of DNA products by UV absorbance. About 7–11 μg of purified DNA should be obtained. 6. Each μg of the obtained libraries is equivalent to about 1.2 × 1012 molecules (see Notes 6 and 7). The library can be stored at -80 °C for several months or years.

3.1.4 Next-Generation Sequencing

1. For each library, prepare 50 μL PCR mixture in a PCR tube containing 25 pmol of each primer (2.5 μL of 10 μM adapter primer NGS-VHH_Fint and NGS-VHH_Rint) (Fig. 3a), 5 ng of VHH library from Subheading 3.1.3, 1 μL of dNTPs mix (containing 10 mM of each dNTP), 10 μL of 5× Phusion HF buffer, 3 μL of DMSO, and 1 U of Phusion polymerase. 2. Use a thermocycler to perform the following PCR program: an initial denaturation step at 98 °C for 30 s 98 °C for 10 s, followed by 20 cycles of PCR of 98 °C for 10 s, 66 °C for

28

Audrey Guilbaud and Fre´de´ric Pecorari

Fig. 3 (a) Principle of the preparation of samples for sequencing by Illumina NGS technology. The orange and blue sequences indicated on the PCR product obtained with primers NGS-VHH_Fint and NGS-VHH_Rint are used by the core-facility to perform a second PCR to index samples with Nextera adapters allowing multiplex sequencing. (b) Agarose gel electrophoresis run with samples obtained from PCR described in Subheading 3.1.4 (expected sizes are 347, 359, and 371 bps, for libraries a, b, and c, respectively)

30 s, 72 °C for 10 s with a final elongation step of 72 °C for 5 min. 3. Purify the products with Promega Wizard SV Gel and PCR Clean-up kit and determine the concentration of the purified PCR product by UV absorbance. 4. Control the homogeneity of the products on 1.5% agarose gel as in Subheading 3.1.1, step 6. 5. The products thus obtained are ready for a second PCR using Nextera index primers for Illumina next-generation sequencing (See Note 8). This second PCR is usually performed at the NGS core-facility (Fig. 3b).

VHH Libraries for Ribosome Display Selections

4

29

Notes 1. 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. 2. We use a construction similar to the one described by Amstutz et al [28] but with modifications to meet our needs, such as FLAG and RGS-His6 tags at C-terminus of VHH and different restriction sites. This system is compatible with prokaryotic in vitro translation with an E. coli S30 extract for ribosome display (Fig. 2a). 3. The ribosome display vector pFP-RDV3 (Fig. 2b) is obtained by cloning the following sequence of a synthetic DNA product in pUC18 plasmid (ATCC 37253) via NcoI and MluI restriction sites (highlighted below in bold): 5′-CCATGG ATACGAAATTAATACGACTCACTATAGG GAGACCACAACGGTTTCCCTCTAGAAA TAATTTTGTTTAACTTTAAGAAGGAGATATATC TATGGGATCCTAATGAGGTACCCTGAGTA GAAGCTTTATATGGCCTCGGGGGCCGAATTCTCT GAGCTCTCTGGGGACTACAAAGATGACGATGA CAAAGGCACCGGTTCCGGCGGTTCTGGCCA GAAGCAAGCTGAAGAGGCGGCAGC GAAAGCGGCGGCAGATGCTAAAGCGAAGGCC GAAGCAGATGCTAAAGCTGCGGAAGAAGCAGC GAAGAAAGCGGCTGCAGACGCAAAGAAAAAAGCA GAAGCAGAAGCCGCCAAAGCCGCAGCCGAAGCG CAGAAAAAAGCCGAGGCAGCCGCTGCGGCACT GAAGAAGAAAGCGGAAGCGGCAGAAGCAGCTG CAGCTGAAGCAAGAAAGAAAGCGGCAACT GAAACCGCACACCTTACTGGTGTGCGGTA AACGCGT-3′. 4. For the construction of high-quality libraries, it is recommended to use highly purified oligonucleotides (HPLC purification). This is to avoid as much as possible undesirable sequences due to n-1 products. 5. This DNA product corresponds to the gene of VHH, flanked by the 5′ sequence necessary for ribosome display and an additional 3′ sequence necessary for subsequent PCR assembly step with the tolA spacer (Fig. 2). 6. The upper limit for the size of the library at this step can be estimated to about 1.2 × 1012 variants when manipulating 1 μg of DNA. 7. It is crucial for the construction of libraries and their subsequent use for selections that PCR products obtained till

30

Audrey Guilbaud and Fre´de´ric Pecorari

this step are of high quality as judged by agarose electrophoresis (i.e. a single sharp band without any smear). 8. We usually request pair-end sequencing to cover the whole VHH sequence (Miseq, Flowcell Micro). Reads obtained are assembled and analyzed using Galaxy tools (https://usegalaxy. org/). NGS is useful to evaluate the quality of the libraries on large samples. As an indication, from 182225 VHH sequenced for libraries VH-L1, we determined that, by following this protocol for library preparation, 172400 sequences were different and that the largest cluster of sequences was composed of not more than four members (unpublished observation).

Acknowledgments The authors thank all previous members of the laboratory who helped to develop this protocol. References 1. Hamers-Casterman C, Atarhouch T, Muyldermans S et al (1993) Naturally occurring antibodies devoid of light chains. Nature 363:446– 448 2. de Marco A (2020) Recombinant expression of nanobodies and nanobody-derived immunoreagents. Protein Expr Purif 172:105645 3. Goldman ER, Liu JL, Zabetakis D, Anderson GP (2017) Enhancing stability of camelid and shark single domain antibodies: an overview. Front Immunol 8:865 4. Pia EAD, Martinez KL (2015) Single domain antibodies as a powerful tool for high quality surface plasmon resonance studies. PLoS One 10:e0124303 5. Huang C, Ren J, Ji F et al (2020) Nanobodybased high-performance immunosorbent for selective beta 2-microglobulin purification from blood. Acta Biomater 107:232–241 6. Verheesen P, ten Haaft MR, Lindner N et al (2003) Beneficial properties of single-domain antibody fragments for application in immunoaffinity purification and immuno-perfusion chromatography. Biochim Biophys Acta 1624: 21–28 7. Rasmussen SGF, Choi H-J, Fung JJ et al (2011) Structure of a nanobody-stabilized active state of the β(2) adrenoceptor. Nature 469:175–180 8. Lo¨w C, Yau YH, Pardon E et al (2013) Nanobody mediated crystallization of an archeal mechanosensitive channel. PLoS One 8: e77984

9. Yamagata M, Sanes JR (2018) Reporter–nanobody fusions (RANbodies) as versatile, small, sensitive immunohistochemical reagents. Proc Natl Acad Sci 115:2126–2131 10. Kijanka M, van Donselaar EG, Muller WH et al (2017) A novel immuno-gold labeling protocol for nanobody-based detection of HER2 in breast cancer cells using immuno-electron microscopy. J Struct Biol 199:1–11 11. Baral TN, Magez S, Stijlemans B et al (2006) Experimental therapy of African trypanosomiasis with a nanobody-conjugated human trypanolytic factor. Nat Med 12:580–584 12. Ackaert C, Smiejkowska N, Xavier C et al (2021) Immunogenicity risk profile of nanobodies. Front Immunol 12:632687 13. Bridoux J, Broos K, Lecocq Q et al (2020) Anti-human PD-L1 nanobody for immunoPET imaging: validation of a conjugation strategy for clinical translation. Biomol Ther 10: E1388 14. Xavier C, Vaneycken I, D’huyvetter M et al (2013) Synthesis, preclinical validation, dosimetry, and toxicity of 68Ga-NOTA-anti-HER2 Nanobodies for iPET imaging of HER2 receptor expression in cancer. J Nucl Med 54:776– 784 15. Chanier T, Chames P (2019) Nanobody engineering: toward next generation immunotherapies and immunoimaging of cancer. Antibodies (Basel) 8:13

VHH Libraries for Ribosome Display Selections 16. Morrison C (2019) Nanobody approval gives domain antibodies a boost. Nat Rev Drug Discov 18:485–487 17. 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 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. Yan J, Wang P, Zhu M et al (2015) Characterization and applications of Nanobodies against human procalcitonin selected from a novel naı¨ve Nanobody phage display library. J Nanobiotechnol 13:33 20. 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 21. McMahon C, Baier AS (2018) Yeast surface display platform for rapid discovery of conformationally selective nanobodies. Nat Struct Mol Biol 25:289–296 22. Zimmermann I, Egloff P, Hutter CAJ et al (2020) Generation of synthetic nanobodies

31

against delicate proteins. Nat Protoc 15: 1707–1741 23. Chen X, Gentili M, Hacohen N, Regev A (2021) A cell-free nanobody engineering platform rapidly generates SARS-CoV-2 neutralizing nanobodies. Nat Commun 12:5506 24. Mouratou B, Schaeffer F, Guilvout I et al (2007) Remodeling a DNA-binding protein as a specific in vivo inhibitor of bacterial secretin PulD. Proc Natl Acad Sci U S A 104: 17983–17988 25. Kalichuk V, Renodon-Corniere A, Behar G et al (2018) A novel, smaller scaffold for Affitins: showcase with binders specific for EpCAM. Biotechnol Bioeng 115:10 26. Kalichuk V, Kambarev S, Behar G et al (2020) Affitins: ribosome display for selection of Aho7c-based affinity proteins. Methods Mol Biol 2070:19–41 27. 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: 4937–4942 28. 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. Elsevier Academic Press, pp 497–509

Chapter 3 Isolation of Adhirons Specific for Plant Protoplast Membrane Biomarkers Is Simplified by Phagemid Design Claudia D’Ercole and Ario de Marco Abstract Phage display is an effective method to retrieve binders specific for a target epitope from a large clone library. Nevertheless, the panning process allows for the accumulation of some contaminant clones into the selected phage pool and, consequently, each clone requires individual screening to verify its actual specificity. This step is time-consuming, independently on the chosen method, and relies on the availability of reliable reagents. Since phages display a single binder responsible for the antigen recognition but their coat is formed by several repeats of the same proteins, the targeting of coat epitopes is often exploited to amplify the signal. Commercial anti-M13 antibodies are commonly labeled with peroxidase or FITC but customized antibodies might be necessary for specific applications. Here, we report a protocol describing the selection of anti-protoplast Adhirons that relies on the availability of nanobodies fused to a fluorescent protein to use during flow cytometry screening. Specifically, when preparing our Adhiron synthetic library, we designed a new phagemid that allows the expression of the clones fused to three tags. These can interact with a large variety of commercial and home-made reagents, selected according to the needs of the downstream characterization process. In the described case, we combined the ALFA-tagged Adhirons with an anti-ALFAtag nanobody fused with the fluorescent protein mRuby3. Key words Pea protoplasts, ALFAtag, Customized reagents, Adhirons, Protoplast biomarkers

1

Introduction Large collections of antibody fragments and synthetic binders of different origin have become largely accessible and have been used to recover reagents specific for almost any kind of molecule [1, 2]. We explored the possibility to exploit phage display to isolate Adhirons starting from a new library prepared in our laboratory. Adhirons are a relatively new class of binders, obtained by hypermutating two loops of a phytocystatin scaffold known for its outstanding stability [3]. Specifically, we wished to assess the suitability of this library for the recovery of binders specific for plant protoplast biomarkers due to the scarce availability of reagents selective for plant proteins and the peculiar challenges that are represented

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2681, https://doi.org/10.1007/978-1-0716-3279-6_3, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

33

Claudia D’Ercole and Ario de Marco

ALFAtag

SpyTag

Adhiron

mRuby3

6xHis

Nb

34

phage

Fig. 1 Characteristics of the Adhiron library clones. The Adhiron phage display library was built using a phagemid designed to express the Adhiron clones fused to a multi-tag (Top): a 6xHis sequence, an ALFAtag, and a SpyTag. Consequently, when displayed on the phages (Bottom), the Adhirons were decorated with tags that allow specific activities: (i) affinity purification (HisTag), (ii) covalent binding to SpyCatcher, which can be fused to further functional partners (SpyTag), and (iii) direct detection, using a fluorescent anti-ALFAtag nanobody (ALFAtag). The pink strips inside the Adhiron unit indicate the two hypervariable regions responsible for the target selective recognition

by protoplasts and by membrane proteins. Whereas plant components such as soluble proteins known to induce allergic reactions have been used to immunize animals and recover the corresponding antibody fragments by panning ad hoc prepared immune libraries [4, 5], there is no published report describing successful panning performed directly on protoplasts to identify binders targeting membrane epitopes. We collected experience in panning pre-immune nanobody libraries directly against whole mammalian cells, unicellular microalgae, cyanobacteria, and even extracellular vesicles [6–9]. In this work, we describe the use of a synthetic Adhiron library for performing blind panning on freshly isolated pea protoplasts with the aim of identifying binders able to recognize surface antigens displayed on the protoplast surface. The pairs of Adhirons/membrane biomarkers can be used for direct imaging, immunoprecipitation, the discovery of the antigen, and, in the middle term, their use can be even conceived for in vivo targeting/delivery. The library was cloned in a modified phagemid that provides a multi-tag for which a large array of inexpensive and customized reagents can be produced directly in the lab. These are then preferentially used according to the required application (Fig. 1). In the example described in the present contribution, we exploited the ALFAtag [10] to target the Adhiron construct with a recombinant, home-made anti-ALFAtag nanobody fused to fluorescent proteins. These detection reagents are particularly handy during flowcytometry screening because they allow the direct target labelling

Flow Cytrometric Screening of Anti-Protoplast Adhirons

35

without the necessity of a secondary antibody or the in vitro labeling of the primary. In our case, we initially produced two versions of the nanobody, one fused to mClover3 and the second to mRuby3. This availability allowed the direct comparison of the reagents and choosing the red fluorescent protein for the screening purpose because its signal better enabled to be distinguished from protoplast autofluorescence. It must be also underlined that the same anti-ALFAtag nanobody can be fused, according to the experimental requirements, also to other functional partners, such as enzymes or split-proteins. This flexibility enables to adapt the reagent to different diagnostic platforms or alternative screening protocols and provides even solutions that might be not available commercially. Finally, considering the usually high yields at which recombinant Adhirons are produced, we expect that the panning and screening approach described above could represent not only a protocol to secure binders selective for plant targets but a convenient generic approach to obtain inexpensive reagents for the scientific community.

2

Materials

2.1 Protoplast Isolation

1. Enzymatic digestion solution: 1 M mannitol, 100 mM MES-KOH, pH 5.6, 0.37% w/v macerozyme, and 1.5% w/v cellulase in a total volume of 10 mL (see Note 1). 2. Solution for enzymatic digestion quenching (W5): 2 mM MES, 154 mM NaCl, 125 mM CaCl2, and 5 mM KCl, pH 5.7. 3. Resuspension solution (MMG): 4 mM MES, 0.4 M mannitol, and 15 mM MgCl2, pH 5.7. 4. Polypropylene mesh (mesh pores: 105 μm; thickness: 121 μm) (see Note 2). 5. Scalpel. 6. Incubator. 7. Falcon tubes (15 mL). 8. Petri dishes 92 × 16 mm. 9. Beckman Allegra X-22R benchtop centrifuge with swinging bucket rotor, model SX4250 (see Note 3). 10. Fixed-angle centrifuge (see Note 4).

2.2 Protoplast Assessments

1. Epifluorescence and bright-field microscope (see Note 5). 2. Hemocytometer.

36

2.3

Claudia D’Ercole and Ario de Marco

Biopanning

1. MM medium: 10 mL M9 medium (5×), 100 μL 1 M MgSO4, 2.5 mL 20% glucose, 100 μL 1% thiamine, and milliQ water (sterile) to the final volume of 50 mL. 2. E. coli strain TG1. 3. Dynabeads® M-450 Epoxy (see Note 6). 4. Phosphate-buffered saline. 5. Phosphate-buffered saline containing 0.4 M mannitol (see Note 7). 6. Blocking solution: 1% milk in PBS/mannitol. 7. Adhiron library (diversity of 7 × 109). 8. 2xTY medium. 9. 2% glucose. 10. Ampicillin stock solution (100 mg/mL) dissolved in H2O. 11. Kanamycin stock solution (50 mg/mL) dissolved in H2O. 12. Elution solution: 0.2 M glycine-HCl, pH 2.2, 1 mg/mL BSA. 13. Neutralizing solution: 1 M Tris–HCl, pH 9.1. 14. Precipitation solution (PS): 30% PEG-600, 2.5 M NaCl. 15. M13K07 helper phages (stock solution 1 × 1011 pfu/mL). 16. Erlenmayer flask (250 mL). 17. Glycerol. 18. Cryotubes. 19. Microcentrifuge tubes (1.5 mL). 20. Falcon tubes (15 mL). 21. Magnetic rack. 22. Eppendorf MiniSpin centrifuge for 1.5 mL tubes. 23. Petri dish 150 × 20 mm. 24. Standard 96-well microplate, flat bottom.

2.4

Screening

1. MegaBlock 96 deep-well plates (volume: 2.2 mL). 2. Standard 96-well microplate, round bottom. 3. 2xTY medium. 4. Phosphate-buffered saline containing 0.4 M mannitol (see Note 7). 5. Blocking solution: 1% milk in PBS/mannitol (see Note 7). 6. M13K07 helper phages (stock solution 1 × 1011 pfu/mL). 7. Ampicillin stock solution (100 mg/mL) dissolved in H2O. 8. Kanamycin stock solution (50 mg/mL) dissolved in H2O. 9. Beckman Allegra X-22R benchtop centrifuge with microplate rotor, model S2096.

Flow Cytrometric Screening of Anti-Protoplast Adhirons

37

10. Anti-ALFAtag-mRuby3 nanobodies. 11. Filter tips. 12. Sterile toothpicks. 13. Incubator. 14. Guava® easyCyte™ instrument).

3

Flow

Cytometer

(or

equivalent

Methods

3.1 Protoplast Isolation

Day 0 1. Grow Pisum sativum plants in a growth chamber at 24–26 °C with a light/dark photoperiod of 16:8 (see Note 8). 2. Collect 1.5 g of leaves from 2-week-old plants (see Note 8) and cut 0.5–1 mm wide intra-veining strips. 3. Incubate strips with the digestion solution for 3:30 h, at 30 °C, with gentle shaking (70 rpm) (see Note 1). 4. Add W5 solution to stop the enzymatic reaction. 5. Filter the digested solution with polypropylene mesh (see Note 2). 6. Wash the digested cell suspension twice by resuspending in 2 mL of MMG solution and pelleting for 3 min at 100 × g. 7. Resuspend the pellet in 1 mL of MMG solution and keep it at 4 °C (see Note 9).

3.2

Biopanning

3.2.1 Coating of Beads with the Antigen (See Note 10)

Day 0—Preliminary Steps 1. Grow overnight TG1 E. coli cells in 50 mL of MM medium using a 250 mL flask at 37 °C and 220 rpm. 2. Resuspend by vortex (30 s to 1 min) the magnetic beads (Dynabeads® M-450 Epoxy) and add 50 μL/sample in two Eppendorf (1.5 mL). 3. Wash beads three times with 1 mL of PBS, collect them by means of a magnetic rack, and resuspend them in 50 μL of PBS. 4. Coat the beads with 500 μL of the depletion antigen milk (2 mg/mL). Label one tube “milk dep1” and the other one “milk dep2”. 5. Incubate under rotation 16–24 h, 4 °C. Day 1—First Round of Biopanning

38

Claudia D’Ercole and Ario de Marco

3.2.2 Bead and Library Blocking

1. Wash the coated beads (Day 0, steps 3 and 4) contained in both milk dep1 and milk dep2 tubes three times with 1 mL PBS/mannitol, and then add 1 mL of blocking solution. 2. In a new tube, put 1 mL of blocking solution and add 100 μL of the Adhiron library (see Note 11). 3. Incubate the tubes containing both beads and library for 30–45 min at 4 °C under constant rotation (30 rpm). 4. Remove the blocking solution from the first bead tube (milk dep1) after having recovered the beads by applying the magnetic rack. 5. Transfer the blocked Adhiron library to the first bead tube (milk dep1). 6. Incubate under rotation (30 rpm) for 1 h at 4 °C. 7. Remove the blocking solution from the second bead tube (milk dep2) after having recovered the beads by applying the magnetic rack. 8. Place the tube milk dep1 in the magnetic rack, recover the supernatant containing the depleted library, and transfer it to the second bead tube (milk dep2). 9. Incubate under rotation (30 rpm) for 1 h at 4 °C. 10. Transfer 1 mL of protoplasts (9 × 105 cells/mL) in a new tube, add 500 μL of blocking solution, and incubate under rotation (30 rpm) for 1 h, 4 °C. 11. Spin the tubes with protoplasts (100 × g, 2 min, 4 °C) and discard the blocking solution. 12. Place the tube milk dep2 in the magnetic rack, recover the supernatant containing the depleted library, and transfer it to the tube containing the protoplast pellet. 13. Incubate under rotation (30 rpm) 2 h at 4 °C. 14. Grow 500 μL of pre-cultured TG1 in 25 mL of 2xTY, 2% glucose. Grow until OD600nm = 0.5 (~2 h, 37 °C, 220 rpm).

3.2.3

Washing

1. Add 5 mL of blocking solution in two tubes of 15 mL and incubate for 1 h at 4 °C under rotation (30 rpm). 2. Discard the blocking solution from the 15 mL washing tubes by inversion. 3. Spin the protoplasts (100 × g, 2 min, 4 °C) and resuspend the pellet in 1 mL of PBS/mannitol. 4. Transfer the protoplasts to the first washing tube and perform 6/8 cycles of washing steps: • add 10 mL of cold PBS/mannitol, mix well and pellet cells (100 × g, 2 min, 4 °C), discard the supernatant, and repeat.

Flow Cytrometric Screening of Anti-Protoplast Adhirons

39

• at the third/fourth wash, add only 5 mL of PBS/mannitol, transfer it into the second tube, bring the volume to 10 mL, and proceed as above. 3.2.4 Phage Elution and Amplification

1. After the last washing step, add 0.9 mL of elution solution to the protoplast pellet, and incubate for 10 min at room temperature and 400 rpm. 2. Centrifuge the sample for 2 min at 100 × g and transfer the supernatant containing the phages to a new tube filled in with 250 μL of neutralizing solution. Label tube: “Eluted adhiron phages”. 3. Add 750 μL of elution fraction to 9.25 mL of TG1 (OD600nm = 0.5) and let the bacterial infection happening at 37 °C, for 30 min, with no stirring. 4. Add 10% glycerol to the remaining eluted phages, keep the sample at -80 °C. 5. Pellet the bacteria (3200 × g 10 min) and recover the pellet in 1.8 mL of 2xTY. 6. Spread 600 μL in each of the three Petri dishes of 2xTY, 1% glucose, 1X Amp. 7. Grow overnight TG1 E. coli at 37 °C and 220 rpm. Day 2

3.2.5

Phage Production

1. Recover the bacteria from the plates with 5 mL of 2xTY, 30% glycerol, and transfer 100 μL in 20 mL 2xTY, 20% glucose, 1X Amp in 250 mL flask. 2. Grow cells at 37 °C until OD600nm = 0.5. 3. Make two aliquots of the remaining sample and keep them at 80 °C in cryotubes. 4. When OD600nm = 0.5, transfer 10 mL of the fresh cell culture in a sterile tube and add 0.45 μL of helper phages. 5. Infect bacteria at 37 °C, 30 min, no stirring. 6. Spin-infected bacteria using the fixed-angle centrifuge (3200 × g, 10 min) and resuspend the pellet in 50 mL of 2xTY, 1X Amp, 1X Kan in 0.5 L flask. 8. Grow cells overnight at 30 °C and 220 rpm. 9. Coat magnetic beads as Day 0. Day 3

3.2.6

Phage Precipitation

1. Transfer bacteria from 0.5 L flask to 50 mL tube and centrifuge at 3200 × g for 10 min. 2. Transfer 40 mL of the supernatant to a clean tube, add 8 mL of cold PS solution, and mix. 3. Incubate on ice for 60 min.

40

Claudia D’Ercole and Ario de Marco

4. Pellet phages (3200 × g, 10 min, 4 °C). 5. Discard the supernatant and resuspend the pellet in 1.8 mL of PBS, 10% glycerol. 6. Transfer 100 μL of the recovered phages to a new tube and add 1 mL of blocking solution to start the second round of biopanning. 7. Make two aliquots of the remaining sample and keep at -80 °C in cryotubes. 3.3 Second Round of Biopanning 3.3.1 Phage Extraction for Screening

Follow the same procedure as in Day 1, then proceed to Day 4. Day 4 1. Recover the bacteria from the three plates using 5 mL of 2xTY, 30% glycerol. 2. Use 100 μL for miniprep to recover the phagemid to use for bacteria transformation. 3. Make aliquots of the remaining sample and keep at -80 °C in cryotubes.

3.3.2 Bacteria Transformation

1. Use 2 μL of the phagemid to transform 50 μL of E. coli TG1 competent cells grown in 1 mL of LB. 2. Use 50/75/100 μL of the transformed bacterial culture to plate three Petri dishes of 2xTY, 1% glucose, and 1X Amp. 3. Incubate the dishes overnight at 37 °C.

3.4

Screening

3.4.1 Colony Amplification of Transformed Bacteria

Day 1 1. Add 1 mL/well of 2xTY, 20% glucose, and 1X Amp in a sterile MegaBlock 96-deep well plate. 2. Pick 93 colonies from the transformation plate with sterile toothpicks or tips. 3. Leave three empty wells (medium only) for negative control. 4. Cover with sterile air-permeable film and grow overnight at 37 °C and 200 rpm. Day 2

3.4.2 In-Plate Phage Production

1. Add 120 μL/well of 2xTY, 50% glycerol in a sterile 96-well flat bottom plate for glycerol stock. 2. Add 1 mL/well of 2xTY, 20% glucose, and 1X Amp in a new sterile MegaBlock 96-deep well plate. 3. Transfer 80 μL of the preculture in the corresponding well of the new plates (first deep-well plate, then flat-bottom plate, using the same tips for the same row). 4. Grow at 37 °C for 2–2:30 h at 200 rpm.

Flow Cytrometric Screening of Anti-Protoplast Adhirons

41

5. When OD600nm = 0.5, infect with helper phages: add 45 μL/ well of 1:1000 dilution (stock solution 1 × 1011 pfu/mL) in 2xTY. 6. Grow at 37 °C, for 30 min, no stirring. 7. Pellet the micro-well plate containing the bacteria (1000 × g, 20 min at room temperature), discard the supernatant by inversion and gently resuspend the pellet in 1 mL of 2xTY, 1X Amp, 1X Kan. 8. Grow bacteria overnight at 30 °C and 200 rpm. 3.4.3 Protoplast Preparation 3.4.4 Blocking and Phage-Antigen Binding

As in Subheading 3.1 (see Note 12). Day 3 1. Add 50 μL/well of blocking solution (1% milk in PBS) in a 96-well round-bottom plate. 2. Pellet phage-infected bacteria from overnight production (1000 × g, 10 min). 3. Transfer 200 μL of phage supernatant in each of the corresponding wells of the plate with blocking solution and incubate for 1 h at room temperature. Preserve four empty wells (only blocking solution) as negative controls. 4. Add 5 μg/mL of anti.ALFAtag-mRuby3 to each well of the 96-well round-bottom microplate to prepare the diagnosis reagent fractions (see Note 13). 5. Incubate the microplate with the diagnosis reagent fractions for 1 h, at room temperature. 6. Pellet protoplasts purified in Day 2 (100 × g, 2 min, 4 °C). Discard the supernatant, and resuspend the pellet in the blocking solution at a concentration of 9 × 105 cell/mL. 7. Transfer 100 μL/well of protoplasts in a 96-well round-bottom plate and incubate them for 1 h at 4 °C. 8. Spin the blocked protoplasts (100 × g, 2 min, 4 °C), and discard the blocking solution. 9. Resuspend the protoplast pellets in each well with 100 μL of diagnosis reagent fractions, with the exception of the negative controls. 10. Incubate for 2 h at 4 °C.

3.4.5

Washing

1. Harvest protoplasts by centrifugation (100 × g, 2 min, 4 °C), discard the supernatant by gentle inversion, and resuspend the pellet with 200 μL/well of PBS/mannitol. 2. Repeat the washing step two times. 3. After the last washing step, resuspend the pellet in 100 μL/well of PBS/mannitol.

500 0

250

Counts

b)

0

Counts

a)

5 00

Claudia D’Ercole and Ario de Marco

2 50

42

100

101

102

103

Red Fluorescence

104

105

100

101

102

103

104

105

Red Fluorescence

Fig. 2 Flow cytometry-based screening of positive binders. Protoplasts possess a strong autofluorescence signal but protoplast-binding Adhirons induce a signal shift when targeted by red fluorescent anti-ALFAtag nanobodies (a). Such contribution is negligible when control Adhirons with no affinity for protoplasts are used (b). The protoplast background fluorescence is in orange, the signal of protoplasts treated with only the antiALFAtag-mRuby3 nanobody corresponds to the black line, while the blue signal is given by the combination of Adhiron plus the anti-ALFAtag-mRuby3 nanobody 3.4.6 Analysis by Flow Cytometer

The specific binding of Anti.ALFAtag-mRuby3 to the protoplasts will result in an increase of the fluorescence signal. 1. Acquire 50,000 events for each sample using the laser with the emission wavelength at 592 nm. 2. After data acquisition, signals from different samples are compared with controls by overlapping their spectra (Fig. 2). 3. A significant shift (threshold: fluorescence increase >3 × 101 log) between the spectrum peaks in the absence and presence of the anti-ALFAtag fluorescent nanobody identifies the positive clones (see Note 14).

3.4.7 Screening in Triplicates

4

Pick the selected positive clones from the glycerol 96-well microplate from “In-plate phage production” and repeat the test in triplicates to confirm the preliminary screening data.

Notes 1. Enzymatic digestion solution has been optimized for protoplast isolation from Pisum leaves [11]. Extraction of protoplasts from different types of plants might require some adjustments, it is necessary to take into account the variable structure of leaves. For instance, we adapted the solution to isolate protoplasts from Populus leaves that are thicker and more coriaceous. Such solution contained: 0.4 M mannitol,

Flow Cytrometric Screening of Anti-Protoplast Adhirons

43

20 mM KCl, 20 mM MES-KOH, pH 5.7, 0.8% w/v macerozyme, 3% w/v cellulase, 10 mM CaCl2, 5 mM 2-Mercaptoethanol, and 0.1% w/v BSA [12]. 2. Meshes of diameter minor than 100 μm allow the removal of most of the debris originated from the digested samples. However, a tissue gauze may serve as a rough alternative filter since small debris can be removed with the supernatant after protoplast centrifugation during the washing steps. Meshes with pores of smaller diameter (to choose according to the protoplast size) have the advantage of preventing any debris contamination and, therefore, the protocol will require a minor number of washing steps that can always damage the protoplasts and contribute to lower yields [13]. 3. After leaf digestion, the sample must be centrifuged to recover the protoplasts in the pellets. Since it is necessary to use low centrifugation speed to avoid damaging the protoplasts, the pellets are relatively unstable. By favoring the formation of a flat pellet on the tube bottom, swing-out rotors simplify the supernatant removal with respect to fixed-angle or vertical rotors, which determine the pellet accumulation along the tube wall. Of course, the step can be performed by adopting centrifuge systems different from ours, whether similar operational conditions are guaranteed. 4. All the steps requiring bacterial pelleting were performed using a fixed-angle rotor. Any model equivalent to ours will be suitable. 5. The epifluorescence microscope allows the detection of the protoplast autofluorescence in the red field. The signal is mostly due to chlorophyll that is excited by blue and green light and emits strongly in the red with a distinct bimodal emission with maxima at 685 and 720–730 nm. The brightfield microscope is used to estimate the total protoplasts spread on the surface of a hemocytometer. 6. The Dynabeads® M-450 Epoxy used in the experiments can be substituted by magnetic beads produced by other companies. However, unspecific binding should be verified. 7. Mannitol is used as an osmotic regulator. 8. In the case of Populus seedlings, they are grown in a mixture of low to medium decomposed peat and perlite. The electrical conductivity of water was 15 mS/m (±25%), and its pH in the range 5.5–6.5. NPK fertilizer (12:14:24) was added at a concentration of 0.5 kg/m3. Four-month Populus scions were used as the source for protoplast isolation.

44

Claudia D’Ercole and Ario de Marco

9. For the panning, we have used 900,000 protoplasts, corresponding to the whole production obtained applying the described protocols but did not try different concentrations. In contrast, we obtained similar number of protoplasts using the leaves of other plant species. 10. The panning is performed using a protoplast suspension but, since milk is present in the buffers, it is preferable to deplete the library against this potential antigen. Coating magnetic beads with milk is a handy solution because the beads, and the bound phages, can be easily removed by applying a magnetic bar. 11. Use always filter tips when using phages to prevent cross contamination. 12. Consider that you need roughly 9000 protoplasts for the analysis of each clone, therefore a format based on a 96-well plate will require almost one million of protoplasts. 13. The AntiALFAtag is a nanobody with picomolar affinity for its antigen [10]. It binds to the ALFAtag expressed by the phagemid in frame with Adhiron sequence. An irrelevant fluorescent nanobody can be used as a negative control. 14. Our experimental setting is not optimal, since our flow cytometer can work in a limited range of emissions. We made a preliminary screening using the detection nanobody fused to different fluorescent proteins but all having signals overlapping with the autofluorescence signals. Therefore, only an additive shift could be measured above the high background. However, we expect a significantly signal increase whether it would be possible to use: (i) a dye in the near infra-red spectrum for nanobody labelling; (ii) a compatible flow cytometer. References 1. Sˇkrlec K, Sˇtrukelj B, Berlec A (2015) Non-immunoglobulin scaffolds: a focus on their targets. Trends Biotechnol 33:408–418 2. Monegal A, Ami D, Martinelli C et al (2009) Immunological applications of single-domain llama recombinant antibodies isolated from a naı¨ve library. Protein Eng Des Sel 22:273–280 3. Tiede C, Tang AA, Deacon SE et al (2014) Adhiron: a stable and versatile peptide display scaffold for molecular recognition applications. Protein Eng Des Sel 27:145–155 4. Eeckhout D, Fiers E, Sienaert R et al (2000) Isolation and characterization of recombinant antibody fragments against CDC2a from Arabidopsis thaliana. Eur J Biochem 267:6775– 6783 5. Hu Y, Wu S, Wang Y et al (2021) Unbiased immunization strategy yielding specific

nanobodies against macadamia allergen of vicilin-like protein for immunoassay development. J Agric Food Chem 69:5178–5188 6. Cre´pin R, Gentien D, Duche´ A et al (2017) Nanobodies against surface biomarkers enable the analysis of tumor genetic heterogeneity in uveal melanoma patient-derived xenografts. Pigment Cell Melanoma Res 30:317–327 7. Mazzega E, Beran A, Cabrini M, de Marco A (2019) In vitro isolation of nanobodies for selective Alexandrium minutum recognition: a model for convenient development of dedicated immuno-reagents to study and diagnostic toxic unicellular algae. Harmful Algae 82: 44–51 8. Popovic M, Mazzega E, Toffoletto B, de Marco A (2018) Isolation of anti-extra-cellular vesicle single-domain antibodies by direct

Flow Cytrometric Screening of Anti-Protoplast Adhirons panning on vesicle-enriched fractions. Microb Cell Fact 17:6 9. Folorunsho OG, Oloketuyi SF, Mazzega E et al (2021) Nanobody-dependent detection of Microcystis aeruginosa by ELISA and thermal lens spectrometry. Appl Biochem Biotechnol 193:2729–2741 10. Go¨tzke H, Kilisch M, Martı´nez-Carranza M et al (2019) The ALFA-tag is a highly versatile tool for nanobody-based bioscience applications. Nat Commun 10:4403 11. Nanjareddy K, Arthikala MK, Blanco L et al (2016) Protoplast isolation, transient transformation of leaf mesophyll protoplasts and

45

improved Agrobacterium-mediated leaf disc infiltration of Phaseolus vulgaris: tools for rapid gene expression analysis. BMC Biotechnol 16:53 12. Guo J, Morrell-Falvey JL, Labbe´ JL et al (2012) Highly efficient isolation of Populus mesophyll protoplasts and its application in transient expression assays. PLoS One 7: e44908 13. Roth MG, Chilvers MI (2019) A protoplast generation and transformation method for soybean sudden death syndrome causal agents Fusarium virguliforme and F. brasiliense. Fungal Biol Biotechnol 6:7

Chapter 4 Facile One-Step Generation of Camelid VHH and Avian scFv Libraries for Phage Display by Golden Gate Cloning Christina Bauer, Elke Ciesielski, Lukas Pekar, Simon Krah, Lars Toleikis, Stefan Zielonka, and Carolin Sellmann Abstract Since its development in the 1980s, the Nobel Prize-awarded phage display technology has been one of the most commonly used in vitro selection technologies for the discovery of therapeutic and diagnostic antibodies. Besides the importance of selection strategy, one key component of the successful isolation of highly specific recombinant antibodies is the construction of high-quality phage display libraries. However, previous cloning protocols relied on a tedious multistep process with subsequent cloning steps for the introduction of first heavy and then light chain variable genetic antibody fragments (VH and VL). This resulted in reduced cloning efficiency, higher frequency of missing VH or VL sequences, as well as truncated antibody fragments. With the emergence of Golden Gate Cloning (GGC) for the generation of antibody libraries, the possibility of more facile library cloning has arisen. Here, we describe a streamlined one-step GGC strategy for the generation of camelid heavy chain only variable phage display libraries as well as the simultaneous introduction of heavy chain and light chain variable regions from the chicken into a scFv phage display vector. Key words VHH, scFv, Camelid, Chicken, Phage display, Golden Gate Cloning, Type IIs restriction enzyme, Antibody engineering, Protein engineering, Library generation, Single domain antibody

1

Introduction In this chapter, we provide a facile one-step cloning protocol for the generation of camelid single domain heavy chain only variable fragment (VHH from lama glama, Fig. 1a) and avian single chain variable fragment (scFv from gallus gallus domesticus, Fig. 1b) antibody phage display libraries after animal immunization employing type IIs restriction enzymes for Golden Gate cloning (GGC). The importance of monoclonal antibodies and their fragment-based formats (e.g., single domain antibodies, scFvs, and Fabs) in drug discovery is emphasized by over 130 therapeutic biological entities in eight therapeutic areas which were granted approval or which are in regulatory review in the United States or European Union [1].

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2681, https://doi.org/10.1007/978-1-0716-3279-6_4, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

47

48

Christina Bauer et al.

Fig. 1 3D representation of lama VHH (a) and chicken scFv antibody (b) as well as schematic representation of M13 bacteriophage with coat proteins p3–p9 with p3-VHH fusion protein displayed on the surface (c). (a, b) For structures, pdb entries 1i3v for lama VHH and 4p48 for chicken scFv were used (https://www.rcsb.org/) and CDR sequences were colored in orange for VHH, in green for chicken VH, and in blue for chicken VL according to IMGT (https://www.imgt.org/IMGTrepertoire/Proteins/) using PyMOL. Both structures highlight elongated CDR3 sequences in VHH and chicken VH. (c) Filamentous M13 bacteriophage is composed of coat proteins p3, p6, p7, p8, and dp9. For phage display, antibody fragment (here exemplarily a VHH) is fused to g3 which is the gene of the p3 coat protein for genotype-phenotype coupling. (Representation is based on Fukunaga and Taki from 2012 [39])

One of the oldest approaches for the isolation of antigenspecific antibodies is based on animal immunization (either naı¨ve or later humanized transgenic animals) in combination with the hybridoma technology pioneered by Ko¨hler and Milstein in 1975 [2]. In recent decades, several in vitro selection methods have been developed for the selection of antibodies from naı¨ve, synthetic, and immunization repertoires, including display technologies such as phage display and yeast display among others [3–9]. The key feature of all display technologies for the successful identification of desired antibodies is the so-called genotype-phenotype coupling, which means that the expressed and displayed antibody fragment (or protein of interest) is linked with its respective encoded genetic information. In 2018, the widely used phage technology for the discovery and manipulation of ligands, enzymes, peptides, and antibodies have been awarded the Nobel Prize in chemistry [10]. The roots of its discovery can be dated back in the 1980s [8] and since then the phage display technology facilitated the development of over 70 therapeutic antibodies entering clinical trials and 14 approved ones including the blockbuster anti-TNFα (tumor necrosis factor-alpha) antibody adalimumab (Humira®) [11, 12]. The basic principle of phage display is based on the groundbreaking work of Georg P. Smith on filamentous phage display by fusing the gene of interest to the minor envelope protein III gene of bacteriophage M13 (Fig. 1c) [8]. The resulting fusion

Lama VHH and Chicken scFv GGC for Phage Display Library Generation

49

protein is translated and transported to the surface and thereby linking the phenotype (antibody display) with the genotype (genetic information). In order to uncouple antibody fragmentpIII fusion protein expression from the phage replication and subsequent amplification, a phagemid system is commonly applied which contains a morphogenetic signal for packaging the vector into the assembled phage particles as well as elements for phagemid replication [5, 7]. Consequently, correctly assembled phages require co-transfection of the phagemid in E. coli with a helper phage devoid of replication elements [5]. The advantage of the phage display system is the possibility to generate huge antibody libraries and select them against toxic and evolutionary conserved antigens when applying naı¨ve libraries [13]. One limitation of the phage display technology is the dependence on the E. coli folding machinery including lack of glycosylation and restriction to smaller protein fragments instead of full-length antibodies with small exceptions [14, 15]. Thus, mainly smaller antibody fragments, e.g. Fab, scFv, and single domain antibodies, are employed for the construction of phage display libraries [16–20]. The usage of species with higher phylogenetic distance from humans, e.g. chicken or camelid, holds the potential to increase the antigen’s immunogenicity during immunization [21]. Chicken Ig gene diversification solely relies on somatic gene conversion events as a type of homologous recombination with rather uniform V-gene sequences [22, 23]. Thus, there is only the need for a single primer pair for heavy and light chain gene amplification after immunization and framework regions are amenable for humanization. Interestingly, the CDR3 loops of chicken antibodies are often rich in cysteines and are elongated compared to CDR3s in humans or mice which could lead to higher paratope complexity and CDR stability [24]. VHH represents the variable domain of the heavy chain only antibodies in camelids and similarly to chicken scFv, elongated and protruding CDR3 loops are described with increased stability due to non-canonical disulfide bonds [25, 26]. Due to their small size presumably leading to better tissue penetration, their high stability and their applicability to humanization as well as multiple reformatting options into bi- and multispecifics, VHHs emerged as a valuable building blog for the development of therapeutic antibodies [27–30]. The generation of phage display libraries in general is a tedious and labor-intensive process, especially when both heavy and light chain antibody diversities need to be introduced in the phagemid vector by conventional cloning [20]. This limitation could be overcome by the in 2008 presented Golden Gate Cloning approach utilizing type II restriction enzymes in combination with a T4 DNA ligase in a one-pot reaction to clone multiple DNA fragments in a plasmid within a predefined order [31]. Thereby, DNA overhangs (herein designated as signature sequences) guide the assembly of the digested fragments into the vector in a seamless manner so that

50

Christina Bauer et al.

restriction recognition sites are eliminated within the final product [31]. As a consequence, religation of empty vector is prevented, which ultimately leads to higher cloning efficiency. This elegant cloning strategy has been applied for the generation of Fab antibody fragment yeast surface display [32, 33] as well as phage display [34] libraries. In this chapter, we provide a protocol for a streamlined, one-step scFv or VHH phage display library generation employing GGC after chicken (Gallus gallus domesticus) or camelid (Lama glama) immunization, respectively. To this end, we utilize a phage display phagemid as a destination plasmid (pDest) for the introduction of either lama VHH or chicken light and heavy chain variable regions by a one-pot restriction- and ligation-based cloning.

2

Materials

2.1 cDNA Synthesis and Gene-Specific Amplification of VHH or VH/VL for Library Generation

1. SuperScript III First-Strand Synthesis System for Reverse Transcriptase PCR Kit (Invitrogen). (a) 50 μM Oligo(DT) Primer. (b) 10 mM dNTP mix. (c) Nuclease-free DEPC water. (d) 25 mM MgCl2. (e) RT Buffer 10×. (f) 0.1 M DTT. (g) RNaseH (2 U/μL). (h) Superscript™ III Reverse Transcriptase (200 U/mL). 2. Thermocycler. 3. Q5 High-Fidelity 2× Master Mix (NEB). 4. Device and reagents for agarose gel electrophoresis. 5. Wizard® SV Gel and PCR Clean-up System (Promega). 6. Nanodrop.

2.2 Library Construction Phage Display 2.2.1

Plasmids (Table 1)

Plasmid maps illustrating the most essential components are shown in Figs. 2 and 3. 1. pPD_Dest_SapI/BsmBI (destination plasmid for VHH (SapI) and scFv (BsmBI), respectively based on Hust and colleagues [16]) essential elements: ampicillin resistance gene for selection and plasmid propagation in E. coli., ColE1 and f1 origins of replication, pelB leader sequence for periplasmic antibody fragment expression, Stuffer sequence between SigA and SigZ is flanked by either SapI or BsmBI recognition sites (see Table 2), Myc: c-Myc epitope EQKLISEEDL-tag, His:

Lama VHH and Chicken scFv GGC for Phage Display Library Generation

51

Fig. 2 Schematic representation of GGC-mediated generation of camelid-derived single heavy only variable domain (VHH) phage display antibody libraries. SapI type II restriction endonuclease is employed to digest the destination plasmid (pPD_Dest_SapI) carrying a stuffer region flanked by respective recognition sites together with PCR-amplified VHH-repertoire derived by immunization with recognition sites in opposite orientation (S: GCTCTTCN, S: NGAAGAGC). Signature sequences (SigA, SigZ) remaining after SapI digestion are composed of complimentary overhangs facilitating pre-defined assembly during one-pot ligation. As a product, the final display phagemid (pPD_Expr) is created. ColE1: origin of replication. f1: origin of replication. AmpR: Ampicillin resistance gene, pelB: leader sequence. Linker: amino acids AAAGS, Myc: c-Myc epitope EQKLISEEDL, His: hexahistidine tag, TS: trypsin site KDIR, gIII. Gene encoding for phage protein pIII. Note that an amber stop codon is introduced between the hexahistidine tag and the trypsin site

hexahistidine tag, TS: trypsin site KDIR, gIII: Gene encoding for phage coat protein pIII, amber stop codon between hexahistidine tag, and the trypsin site. 2. pE_Linker_BsmBI (entry vector for scFv VH and VL cloning) essential elements: a glycine-serine linker (GS-linker) is framed with BsmBI restriction enzyme recognition sites as well as signature sequences (SigB and SigC), Kanamycin resistance marker for selection, and plasmid propagation in E. coli. Not shown: replication origin in E. coli (ColE1). 2.2.2 Buffers, Enzymes, Primers, and Devices for Golden Gate Cloning

1. SapI (NEB). 2. BsmBI (NEB). 3. HindIII (NEB). 4. T4 DNA Ligase (NEB). 5. T4 Ligase buffer (NEB). 6. ER2738 electrocompetent E. coli cells (Lucigen), genotype: [F ′proA+B+ lacIq Δ(lacZ)M15 zzf::Tn10 (tetr)] fhuA2 glnVΔ (lac-proAB) thi1Δ(hsdS-mcrB)5.

52

Christina Bauer et al.

Fig. 3 Schematic one-step process for the construction of scFv antibody fragment phage display libraries from avian antibody repertoire. In contrast to the VHH GGC process, Esp3I/BsmBI type II restriction endonucleases are employed for the design of the destination plasmid (pPD_Dest_BsmBI) and are also used for flanking the PCR-amplified VH and VL repertoires with recognition sites in opposite orientations (B: CGTCTCN, B: NGAGACG). Additionally, an entry plasmid is introduced to the one-pot digestion reaction carrying a glycine-serine linker (GS-linker) also framed with respective restriction enzyme recognition sites. Complimentary overhangs on each component named signature sequences (SigA, SigB, SigC, and SigZ) after BsmBI digestion enable the defined assembly of the four component system during ligation. This results in the final display phagemid vector (pPD_Expr). ColE1: origin of replication. f1: origin of replication. AmpR: Ampicillin resistance gene, pelB: leader sequence. Linker: amino acids AAAGS, Myc: c-Myc epitope EQKLISEEDL tag, His: hexahistidine tag, TS: trypsin site KDIR, gIII: Gene encoding for phage protein pIII. Note that an amber stop codon is introduced between the hexahistidine tag and the trypsin site Table 1 Plasmids Name

Description

pPD_Dest_SapI

Phagemid used for VHH GGC using SapI

pPD_Dest_BsmBI

Phagemid used for scFv GGC using BsmBI

pE_Linker_BsmBI

Entry plasmid containing the glycine-serine linker for the scFv for GGC

7. 0.2 cm cuvettes (Genepulser Xcell™). 8. MicroPulser electroporator (BioRad). 9. SOC medium (Lucigen). 10. Petri dishes (Thermo Scientific). 11. 2xTY-medium.

Lama VHH and Chicken scFv GGC for Phage Display Library Generation

53

Table 2 Signature sequences used for the GGC strategy Name

Sequence (5′ → 3′)

SapI (camelid VHH display) SigA

ATG

SigZ

GCG

BsmBI (chicken scFv display) SigA

CCAT

SigB

GGTG

SigC

CGGT

SigZ

GCGG

Table 3 Primers used for the variable antibody domain amplification Name

Sequence (5′ → 3′)

Gene-specific amplification of VHH regions as the template for library construction VHH_SapI_up

TAGCTAGCTCTTCTATGGCTGATGTGCAGCTGCAGGAGTC TGGGGGAGG

VHH_SapI_sh_lo

TAGCTAGCTCTTCTCGCGGGGTCTTCGCTGTGGTGCG

VHH_SapI_lh_lo

TAGCTAGCTCTTCTCGCTGGTTGTGGTTTTGGTGTCTTGGG

Gene-specific amplification of chicken scFv regions as the template for library construction Chicken_BsmBI_VH_up TAGCTACGTCTCTCCATGGCTGCCGTGACGTTGGACGAG Chicken_BsmBI_VH_lo TAGCTACGTCTCTCACCGGAACTGACGATGACTTCGGT Chicken_BsmBI_VL_up TAGCTACGTCTCTCGGTGCGCTGACTCAGCCGTCCTCG Chicken_BsmBI_VL_lo

TAGCTACGTCTCTCCGCTAGGACGGTCAGGGTTGTCCC

SapI or BsmBI recognition sites are shown in bold, and signature sequences are underlined

(a) 1.6% (w/v) tryptone. (b) 1% (w/v) yeast extract. (c) 0.5% (w/v) NaCl. 12. 1.5% (w/v) agar-agar. 13. 100 nM glucose (G). 14. 100 μg/mL ampicillin (A). 15. Petri dishes (Greiner). 16. RedTaqMix 1.1× (VWR). 17. Primers for variable antibody domain amplification (see Table 3).

54

3

Christina Bauer et al.

Methods In this protocol, we outline the generation of libraries based on immunized camelids or chickens starting from extracted RNA isolated from whole blood or tissues of immunized animals. Procedures for immunization and RNA preparation can be found elsewhere [35, 36].

3.1

cDNA Synthesis

Reverse transcription of RNA is described based on the SuperScript II First-Strand Synthesis System (Invitrogen). For one reaction, the volume is set to 20 μL. We suggest preparation of a master mix in case of multiple reactions. 1. Place a nuclease-free PCR tube on ice and add 2 μg of isolated total RNA. Add 1 μL of dNTP mix (10 mM, component of the kit) and 1 μL Oligo-dT-primer (50 μM). Add RNase-free water to a final volume of 10 μL. 2. Incubate at 65 °C for 5 min in a thermocycler, then subsequently let the reaction rest on ice for at least 1 min. 3. Add 1 μL RT buffer (10×), 2 μL MgCl2 (25 mM), 1 μL DTT (0.1 M), 1 μL RNAse (40 U/mL), and 1 μL Superscript III Reverse Transcriptase. 4. Mix by vortexing and centrifuge briefly. 5. Incubate for 50 min at 50 °C in a thermocycler for cDNA synthesis and terminate the reaction for 5 min at 85 °C. After 5 min incubation on ice and add 1 μL of RNaseH (2 U/μL) for RNA removal. Incubate 20 min at 37 °C in a thermocycler. The synthesized cDNA can be subjected to subsequent PCR amplification of natural antibody repertoires and can be stored at -80 °C.

3.2 Amplification of Natural Variable Antibody Domain Repertoire 3.2.1 Amplification of Camelid Heavy Chain Only Variable Domain Repertoire

For the generation of VHH libraries after immunization, the antibody framework repertoire of camelids (llama: Lama glama, alpaca: Vicugna pacos, huarizo: hybrid offspring of male llama and female alpaca) according to IMGT entries is used as a template. For this, 1 μL of previously described cDNA reaction is submitted to a PCR reaction mixture in a final volume of 50 μL. 1. Pre-chill a PCR tube on ice and add 22 μL of nuclease-free water. Add 1 μL of cDNA reaction as the template together with 1 μL of forward primer VHH_SapI_up and 1 μL of reverse primer VHH_SapI_sh_lo for the short hinge or VHH_SapI_lh_lo for the long hinge (from 10 μM stock, primer sequences listed in Table 1, see Notes 1 and 2). Add 25 μL of Q5® High-Fidelity 2× Master Mix.

Lama VHH and Chicken scFv GGC for Phage Display Library Generation

55

2. Start the PCR reaction with the following parameters: Initial denaturation 98 °C for 30 s. 30 cylces of 10 s at 98 °C, 20 s at 60 °C, and 40 s at 72 °C, followed by final elongation at 72 °C for 2 min. 3. Analyze the success of amplification using 1–2 μL PCR product by 1–2% (w/v) agarose gel electrophoresis and quantify concentration using Nanodrop. 4. Purify the PCR product using a PCR clean up kit according to the manufacturer’s instruction. PCR products might be stored at -20 °C. 3.2.2 Amplification of Chicken Variable Antibody Domain Repertoire

For the generation of scFv antibody fragment libraries after immunization, the IGMT antibody framework repertoire of chicken (Gallus gallus domesticus) is used as a template for the variable heavy chain and the variable light chain. Analogously to the VHH amplification, 1 μL of previously described cDNA reaction is submitted to a PCR reaction mixture in a final volume of 50 μL. 1. Pre-chill a PCR tube on ice and add 22 μL of nuclease-free water. Add 1 μL of cDNA reaction as the template together with 1 μL of forward primer and 1 μL of respective reverse primer (from 10 μM stock, primer sequences listed in Table 1). For VH amplification, the primer pair Chicken_BsmBI_VH_up and _lo and for the VL Chicken_BsmBI_VL_up and _lo are used. Add 25 μL of Q5® High-Fidelity 2× Master Mix. 2. Start the PCR reaction with the following parameters: Initial denaturation 98 °C for 30 s. 30 cylces of 10 s at 98 °C, 20 s at 58 °C, and 60 s at 72 °C, followed by final elongation at 72 °C for 2 min. 3. Analyze the success of amplification using 1–2 μL PCR product by 1–2% (w/v) agarose gel electrophoresis and quantify concentration using Nanodrop. 4. Purify the PCR product using a PCR clean up kit according to the manufacturer’s instruction. PCR products might be stored at -20 °C.

3.3 Golden Gate Cloning for the Generation of Phage Display Libraries 3.3.1 GGC for Camelid VHH Libraries

For the GGC reaction, the VHH PCR product from Subheading 3.3.1 is mixed with the destination phagemid and the type II restriction enzyme SapI (see Note 3). For a general description of phage display library generation, the reader is referred to Schirrmann and Hust [37]. The principle of Golden Gate Cloning for Fab phage display libraries has been previously described by Nelson and Valadon [34]. The herein used phagemid pPD_Dest_SapI is based on the phagemid pHAL14 constructed by the group of Du¨bel and co-workers for traditional two-step cloning procedure [16, 20].

56

Christina Bauer et al.

1. For the GGC reaction, pre-chill a PCR tube on ice and add 1 μg of destination plasmid (pPD_Dest_SapI), 500 ng of amplified VHH PCR product, 200 U SapI, 1000 U T4 DNA Ligase, and 10 μL 10× T4 DNA Ligase buffer with 10 mM dATP. Add nuclease-free water to a final volume of 100 μL. 2. Start the PCR thermocycler reaction with the following parameters: 40 cylces of 1 min at 37 °C and 1 min at 16 °C, followed by 10 min at 65 °C for enzyme inactivation. 3. Purify the GGC product using a PCR clean up kit according to the manufacturer’s instruction. GGC reaction products might be stored at -20 °C. 4. Thaw 50 μL of ER2738 electrocompetent cells (Lucigen) on ice and mix with 30 μL of the GGC reaction. 5. Transfer to an ice-cold electroporation cuvette (0.2 cm) and pulse with 1700 V electroporator. Immediately add 920 μL of pre-warmed SOC medium and incubate at 37 °C with 400 rpm for 1 h (see Note 4). 6. Remove 10 μL for dilution plating in order to calculate library complexity. Perform dilution plating on TY-GA plates. Incubate overnight at 37 °C. 7. Plate out the remaining 990 μL on two 25 cm square petri dishes of TY-GA agar. Incubate overnight at 37 °C. 8. Calculate library size (see Note 5). 3.3.2 GGC for Chicken scFv Libraries

Due to low presence of BsmBI (or Esp3I) recognition sites in chicken IgHV and IgLV sets, the GGC strategy for chicken scFv was based on this type II restriction enzyme (see Note 6) [38]. In summary, the VH and VL PCR products from Subheading 3.3.2 were mixed with the destination phagemid for scFv cloning (pPD_Dest_BsmBI) and the entry module for the incorporation of glycine-serine-linker (pE_Linker_BsmBI). Analogously to pPD_Dest_SapI, pPD_Dest_BsmBI is based on the phagemid pHAL14 form Du¨bel and colleagues [16]. The introduction of Hind III restriction enzyme recognition sites within the stuffer sequence of the destination plasmids allows the elimination of undigested residual destination plasmid during an additional restriction reaction with 5 μL Hind III. 1. For the GGC reaction, pre-chill a PCR tube on ice and add 1 μg of destination plasmid (pPD_Dest_BsmBI), 250 ng of amplified chicken VH PCR product, 250 ng of amplified chicken VL PCR product, 1.4 μg of pE_Linker_BsmBI, 400 U BsmBI, 50 U HindIII for complete digestion of the destination plasmid, 1000 U T4 DNA Ligase, and 10 μL 10× T4 DNA Ligase buffer with 10 mM dATP. Add nuclease-free water to a final volume of 100 μL.

Lama VHH and Chicken scFv GGC for Phage Display Library Generation

57

2. Start the PCR thermocycler reaction with the following parameters: 40 cylces of 1 min at 37 °C and 1 min at 16 °C, followed by 10 min at 80 °C for enzyme inactivation. 3. Purify the GGC product using a PCR clean up kit according to the manufacturer’s instruction. GGC reaction products might be stored at -20 °C. 4. Thaw 50 μL of ER2738 electrocompetent cells (Lucigen) on ice and mix with 30 μL of GGC reaction. 5. Transfer to an ice-cold electroporation cuvette (0.2 cm) and pulse with 1700 V electroporator. Immediately add 920 μL of pre-warmed SOC medium and incubate at 37 °C with 400 rpm for 1 h. 6. Remove 10 μL for dilution plating in order to calculate library complexity. Perform dilution plating on TY-GA plates. Incubate overnight at 37 °C. 7. Plate out the remaining 990 μL on two 25 cm square petri dishes of TY-GA agar. Incubate overnight at 37 °C. 8. Calculate library size. Procedures to assess library quality (colony PCR), library packaging, scFv production by E. coli, and antibody fragment selection have been elegantly described elsewhere [13, 20].

4

Notes 1. In this protocol, primers were utilized annealing to the hinge region of camelid-derived IgG2b and IgG3b isotypes. Alternatively, a reverse primer can be designed that anneals in framework region 4 of the VHH domain. Of note, this might result in the co-amplification of camelid-derived VH domains. 2. For VHH amplification using the oligonucleotides shown in Table 1, one co-amplifies also the VH-CH1 region of the conventional IgG repertoire of camelids. Hence, occasionally a faint side-product might appear on the agarose gel (of approximately 800–900 bp). Typically, we do not gel excise the VHH-specific band to avoid a potential bias due to re-amplification in a second PCR, since resulting libraries predominantly harbor VHH domains. 3. In addition to SapI, additional type IIs enzymes could potentially be exploited. However, one needs to take into consideration that recognition sites for that cognate enzyme might be present in the PCR-amplified VHH repertoire, like previously shown by our group for BsaI and BsmBI in case of amplified Lama glama-derived VHH diversities [38].

58

Christina Bauer et al.

4. One electroporation reaction using this cloning procedure typically results in approximately 1 × 106 to 1 × 107 unique clones. Parallelize electroporation reactions in order to obtain a sufficient library size. 5. We recommend sequencing the constructed library in order to assess the overall quality. To this end, adequate primers need to be designed. Typically, we send out a 96-well plate of single clones from constructed libraries for sequencing. Alternatively (or in parallel) one could also determine insert rates by colony PCR. Typically, insert rates for scFv and VHH immune libraries using this procedure range between 75% and 96%. 6. In this study, we decided to utilize BsmBI for chicken scFv library cloning. However, according to our data, SapI might be a viable alternative [38].

Acknowledgments We are grateful to Stefan Becker, Jonas Ku¨gler, Andre´ Frenzel, Bernhard Valldorf, and Michael Hust for the good collaboration. References 1. Kaplon H, Chenoweth A, Crescioli S, Reichert JM (2022) Antibodies to watch in 2022. MAbs 14(1):2014296. https://doi.org/10.1080/ 19420862.2021.2014296 2. Ko¨hler C, Milstein G (1975) Continuous cultures of fused cells secreting antibody of predefined specificity. Nature 256(Mopc 21): 495–497 3. Boder ET, Midelfort KS, Wittrup KD (2000) Directed evolution of antibody fragments with monovalent femtomolar antigen-binding affinity. Proc Natl Acad Sci 97(20):10701–10705. https://doi.org/10.1073/pnas.170297297 4. 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 5. Breitling F, Du¨bel S, Seehaus T, Klewinghaus I, Little M (1991) A surface expression vector for antibody screening. Gene 104(2):147–153. https://doi.org/10. 1016/0378-1119(91)90244-6 6. 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 7. McCafferty J, Griffiths AD, Winter G, Chiswell DJ (1990) Phage antibodies: filamentous

phage displaying antibody variable domains. Nature 348(6301):552–554. https://doi. org/10.1038/348552a0 8. Smith GP (1985) Filamentous fusion phage: novel expression vectors that display cloned antigens on the virion surface. Science 228: 1315–1317 9. Valldorf B et al (2022) Antibody display technologies: selecting the cream of the crop. Biol Chem 403(5–6):455–477. https://doi.org/ 10.1515/hsz-2020-0377 ˜ a E (2019) The 2018 10. Barderas R, Benito-Pen Nobel Prize in Chemistry: phage display of peptides and antibodies. Anal Bioanal Chem 411(12):2475–2479. https://doi.org/10. 1007/s00216-019-01714-4 11. 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 12. Alfaleh MA et al (2020) Phage display derived monoclonal antibodies: from bench to bedside. Front Immunol 11:1986. https://doi.org/10. 3389/fimmu.2020.01986 13. Hust M, Dubel S, Schirrmann T (2007) Selection of recombinant antibodies from antibody gene libraries. Methods Mol Biol 408:243–255

Lama VHH and Chicken scFv GGC for Phage Display Library Generation 14. Simmons LC et al (2002) Expression of fulllength immunoglobulins in Escherichia coli: rapid and efficient production of aglycosylated antibodies. J Immunol Methods 263(1–2): 133–147. https://doi.org/10.1016/S00221759(02)00036-4 15. Mazor Y, Van Blarcom T, Mabry R, Iverson BL, Georgiou G (2007) Isolation of engineered, full-length antibodies from libraries expressed in Escherichia coli. Nat Biotechnol 25(5):563–565. https://doi.org/10.1038/ nbt1296 16. Hust M et al (2011) A human scFv antibody generation pipeline for proteome research. J Biotechnol 152:159–170 17. De Haard HJ et al (1999) A large non-immunized human Fab fragment phage library that permits rapid isolation and kinetic analysis of high affinity antibodies. J Biol Chem 274(26):18218–18230. https://doi.org/10. 1074/jbc.274.26.18218 18. Vincke C, Gutie´rrez C, Wernery U, Devoogdt N, Hassanzadeh-Ghassabeh G, Muyldermans S (2012) Generation of single domain antibody fragments derived from camelids and generation of manifold constructs. Methods Mol Biol 907:145–176. https://doi.org/10.1007/978-1-61779974-7_8 19. Ubah OC, Barelle CJ, Buschhaus MJ, Porter AJ (2016) Phage display derived IgNAR V region binding domains for therapeutic development. Curr Pharm Des 22(43):6519–6526. h t t p s : // d o i . o r g / 1 0 . 2 1 7 4 / 1381612822666160907091708 20. Ku¨gler J et al (2015) Generation and analysis of the improved human HAL9/10 antibody phage display libraries. BMC Biotechnol 15(1):1–15. https://doi.org/10.1186/ s12896-015-0125-0 21. 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. https://doi.org/10.1016/0022-1759(95) 00143-X 22. Kim S, Humphries EH, Tjoelker L, Carlson L, Thompson CB (1990) Ongoing diversification of the rearranged immunoglobulin light-chain gene in a bursal lymphoma cell line. Mol Cell Biol 10(6):3224–3231. https://doi.org/10. 1128/mcb.10.6.3224-3231.1990 23. Schusser B et al (2013) Harnessing gene conversion in chicken B cells to create a human antibody sequence repertoire. PLoS One 8(11):e80108. https://doi.org/10.1371/jour nal.pone.0080108

59

24. Wu L et al (2012) Fundamental characteristics of the immunoglobulin VH repertoire of chickens in comparison with those of humans, mice, and camelids. J Immunol 188(1): 3 2 2 – 3 3 3 . h t t p s : // d o i . o r g / 1 0 . 4 0 4 9 / jimmunol.1102466 25. Hamers-Casterman C, Atarchouch T et al (1993) Naturally occurring antibodies devoid of light chains. Nature 363:446–448 26. Ko¨nning D 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 27. Chanier T, Chames P (2019) Nanobody engineering: toward next generation immunotherapies and immunoimaging of cancer. Antibodies 8(1):13. https://doi.org/10. 3390/antib8010013 28. Krah S, Schro¨ter C, Zielonka S, Empting M, Valldorf B, Kolmar H (2016) Single-domain antibodies for biomedical applications. Immunopharmacol Immunotoxicol 38(1):21–28. https://doi.org/10.3109/08923973.2015. 1102934 29. Pekar L et al (2020) Biophysical and biochemical characterization of a VHH-based IgG-like bi- and trispecific antibody platform. MAbs 12(1):1812210. https://doi.org/10.1080/ 19420862.2020.1812210 30. Vincke C, Loris R, Saerens D, MartinezRodriguez S, Muyldermans S, Conrath K (2009) General strategy to humanize a camelid single-domain antibody and identification of a universal humanized nanobody scaffold. J Biol Chem 284(5):3273–3284. https://doi.org/ 10.1074/jbc.M806889200 31. Engler C, Kandzia R, Marillonnet S (2008) A one pot, one step, precision cloning method with high throughput capability. PLoS One 3(11):e3647. https://doi.org/10.1371/jour nal.pone.0003647 32. Roth L et al (2019) Facile generation of antibody heavy and light chain diversities for yeast surface display by Golden Gate Cloning. Biol Chem 400:383–393. https://doi.org/10. 1515/hsz-2018-0347 33. Rosowski S et al (2018) A novel one-step approach for the construction of yeast surface display Fab antibody libraries. Microb Cell Factories 17(1):1–11. https://doi.org/10.1186/ s12934-017-0853-z 34. Nelson RS, Valadon P (2017) A universal phage display system for the seamless construction of Fab libraries. J Immunol Methods 450: 41–49. https://doi.org/10.1016/j.jim.2017. 07.011

60

Christina Bauer et al.

35. Grzeschik J et al (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 36. Roth L et al (2070) Isolation of antigenspecific VHH single-domain antibodies by combining animal immunization with yeast surface display. Methods Mol Biol 2020:173– 189. https://doi.org/10.1007/978-1-49399853-1_10 37. Schirrmann T, Hust M (2010) Construction of human antibody gene libraries and selection of

antibodies by phage display. Methods Mol Biol 651:177–209. https://doi.org/10.1007/ 978-1-60761-786-0_11 38. Sellmann C et al (2020) A one-step process for the construction of phage display scFv and VHH libraries. Mol Biotechnol 62(4): 228–239. https://doi.org/10.1007/s12033020-00236-0 39. Fukunaga K, Taki M (2012) Practical tips for construction of custom peptide libraries and affinity selection by using commercially available phage display cloning systems. J Nucleic Acids 2012:295719. https://doi.org/10. 1155/2012/295719

Chapter 5 Identification of New Antibodies Targeting Tumor Cell Surface Antigens by Phage Display Steffen Krohn, Matthias Peipp, and Katja Klausz Abstract The majority of therapeutic antibodies, bispecific antibodies, and chimeric antigen receptor (CAR) T cells in cancer therapy are based on an antibody or antibody fragment that specifically binds a target present on the surface of a tumor cell. Suitable antigens that can be used for immunotherapy are ideally tumor-specific or tumor-associated and stably expressed on the tumor cell. The identification of new target structures to further optimize immunotherapies could be realized by comparing healthy and tumor cells using “omics” methods to select promising proteins. However, differences in post-translational modifications and structural alterations that can be present on the tumor cell surface are difficult to identify or even not accessible by these techniques. In this chapter, we describe an alternative approach to potentially identify antibodies targeting novel tumor-associated antigens (TAA) or epitopes by using cellular screening and phage display of antibody libraries. Isolated antibody fragments can be further converted into chimeric IgG or other antibody formats to investigate the anti-tumor effector functions and finally identify and characterize the respective antigen. Key words Phage display, Immune library, Cellular panning, Immunotherapy, Monoclonal antibody

1

Introduction Today, antibodies targeting tumor-associated antigens (TAA) are an integral part of treatments for various cancer types, and their versatile application in different formats, such as bispecific antibodies and CAR T cells, significantly expanded the therapeutic options in the last 25 years. Many strategies to develop novel antibodies are based on the principle that an antibody or antibody fragment specifically binds a target structure, that is—in most cases—a surface protein on the tumor cell [1]. Although immunotherapies led to a significant improvement in survival for many patients, still not all patients benefit or relapse from current treatment options. In addition, the identification of new target structures on tumor cells represents a potential way to further optimize cancer immunotherapy.

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2681, https://doi.org/10.1007/978-1-0716-3279-6_5, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

61

62

Steffen Krohn et al.

An antigen that is suitable as the target structure for immunotherapy is in the ideal case tumor-specific or tumor-associated and stably expressed on the tumor cell surface. Furthermore, the antigen and even the epitope of a therapeutic antibody influence its set of anti-tumor effector functions. Both, direct Fab-mediated effector functions including the mediation of programmed cell death or ligand blocking and Fc-mediated effector mechanism like engaging immune cells or activation of the complement system depend on the three-dimensional structure of the antigen-antibody complex on tumor cell surface. The identification of new target structures for immunotherapy is frequently based on comparing healthy and tumor cells to select tumor-specific or tumor-associated surface proteins. This could be achieved by different “omics” methods comparing cells on genome, transcriptome, or proteome levels. Afterwards, antibodies can be generated against the most promising proteins and tested for their effector mechanism [2]. However, tumor-specific post-translational modifications and structural alterations that can appear on the tumor cell surface are more difficult to identify by these methods. In this chapter, we describe an alternative approach to identify promising antibodies against potential tumor targets by using antibody libraries and phage display (Fig. 1). With the generation of antibody libraries from mice immunized with tumor cells of interest, preferably freshly isolated from patients, antibodies with high affinities can be isolated since they’ve already being affinity maturated in vivo. Nevertheless, also antibody libraries from different origins, like naı¨ve libraries from human donors or synthetic libraries, can be used as a source [3, 4]. Subsequent cellular panning with intact, freshly prepared cells enables the enrichment of antibodies that preferentially bind to tumor cells. By using this approach, screening for antibodies binding TAAs on the tumor cells under physiological conditions also potentially allows the isolation of antibodies targeting post-translational modifications and structural alterations on tumor cell surfaces. Flow cytometry and whole cell ELISA experiments permit analyses of a broad panel of phage antibodies to unravel their tumor specificity (Fig. 3). After selection based on binding properties of the antibody fragments presented on the surface of the phages (here: single-chain fragment variable (scFv)), re-formatting and testing of desired anti-tumor effector functions can easily be performed since the genetic information of the antibody fragment is simultaneously encoded on the phagemid and is accessible by sequencing. V-region nucleotide sequences and clonotypes of the selected antibody fragments can be determined by Sanger sequencing of phagemids and online analysis tools like IMGT/V-QUEST or ARResT/Interrogate [5, 6]. Hereby, candidates that originate from a common B cell (and probably bind the same antigen) or

Identification of Antibodies Targeting Cell Surface Antigens

63

Fig. 1 Flow diagram for the identification of new antibodies targeting tumor cell surface antigens by phage display. The antibody library is generated from mice immunized with tumor cells (grey). To enrich antibody fragments that bind tumor-specific or tumor-associated antigens, phage display is performed repetitively in a cellular screening approach consisting of depletion/ preabsorption with cells of a healthy donor (blue) and positive selection with tumor cells (green). Separated monoclonal phages can be further characterized regarding binding (yellow) and nucleotide sequences of their V-regions (pink)

mutations that occurred during affinity maturation can be identified. In addition, next-generation sequencing (NGS) can be performed using appropriate systems (e.g., Illumina MiSeq) to deeply sequence the V-regions of all antibodies in the initial library and after each cellular panning step to (a) control the quality of the newly generated library and estimate its diversity, (b) monitor panning and follow the enrichment of individual candidates, and (c) identify rare candidates that show a clear enrichment but had not been selected so far. The integrated set of information helps fully exploiting phage display as it has been described also by other groups [7–9]. Subsequently, the variable regions of the most promising candidates can be converted into every desired antibody format e.g., IgG, intended to be used for immunotherapy to further investigate the antibodies modes of action. In addition, the antigen can be further characterized as a potential target structure and be identified. We used this method to isolate antibodies with preferential binding to malignant plasma cells but the protocol could be adjusted for other cell types of interest.

64

2

Steffen Krohn et al.

Materials

2.1 Immunization of Mice and Analysis of Antibody Titer in Serum

1. BALB/c mice (Charles River Laboratories). 2. Flow cytometry blocking buffer: 1% (w/v) bovine serum albumin (BSA) in phosphate-buffered saline (PBS), and 0.1% (w/v) sodium azide. Filtrate with a 0.22 μm bottle top filter. 3. Anti-mouse IgG FITC-labeled secondary antibody (Sigma).

2.2 Generation of Murine scFv Antibody Library from Spleen

1. TRIzol (Thermo Fisher Scientific).

2.2.1 Preparation of Total RNA from Spleen

4. Nuclease-free water.

2.2.2 Generation of First Strand cDNA by Reverse Transcription

1. Primer for cDNA Synthesis, oligo(dT)15 (Roche).

2. Isopropanol. 3. Chloroform.

2. RNasin Ribonuclease Inhibitors (Promega). 3. dNTP mix, 25 mM each (Biozym). 4. Superscript II reverse transcriptase, including 5
reaction buffer and dithiothreitol (Thermo Fisher Scientific). 5. Ribonuclease H (Thermo Fisher Scientific).

2.2.3 PCR Amplification of V-Regions and Assembly of scFvs

1. peqGOLD Pwo polymerase, including 10
reaction buffer and MgSO4 (VWR). 2. peqGOLD Taq polymerase (VWR). 3. Dimethyl sulfoxide. 4. Light and heavy chain V-region primer mix (Table 1, see Note 1). 5. PCR Nucleotide Mix, 10 mM each (Roche). 6. QIAquick Gel Extraction Kit (Qiagen). 7. SfiI, including 10
reaction buffer (New England Biolabs). 8. pUC19-MSC2017 (derivate of pUC19, produced in our laboratory) [10]. 9. Sodium acetate solution (3 M, pH 5.2), nuclease-free (Thermo Fisher Scientific). 10. Glycogen solution, molecular biological quality (Thermo Fisher Scientific). 11. Ethanol. 12. T4 DNA ligase, conc. (New England Biolabs). 13. n-Butanol. 14. Electrocompetent Escherichia coli (E. coli) XL1-Blue (Agilent).

TTACTCGCGGCCCAGCCGGCCATGGCGGAKGTRMAGCTTCAGGAGTC TTACTCGCGGCCCAGCCGGCCATGGCGGAGGTBCAGCTBCAGCAGTC TTACTCGCGGCCCAGCCGGCCATGGCGCAGGTGCAGCTGAAGSARTC TTACTCGCGGCCCAGCCGGCCATGGCGGAGGTCCARCTGCAACARTC TTACTCGCGGCCCAGCCGGCCATGGCGCAGGTYCAGCTBCAGCARTC TTACTCGCGGCCCAGCCGGCCATGGCGCAGGTYCARCTGCAGCARTC TTACTCGCGGCCCAGCCGGCCATGGCGCAGGTCCACGTGAAGCARTC TTACTCGCGGCCCAGCCGGCCATGGCGGAGGTGAASSTGGTGGARTC TTACTCGCGGCCCAGCCGGCCATGGCGGAVGTGAWGSTGGTGGAGTC TTACTCGCGGCCCAGCCGGCCATGGCGGAGGTGCAGSTGGTGGARTC TTACTCGCGGCCCAGCCGGCCATGGCGGAKGTGCAMCTGGTGGARTC TTACTCGCGGCCCAGCCGGCCATGGCGGAGGTGAAGCTGATGGARTC TTACTCGCGGCCCAGCCGGCCATGGCGGAGGTGCARCTTGTTGARTC TTACTCGCGGCCCAGCCGGCCATGGCGGARGTRAAGCTTCTCGARTC TTACTCGCGGCCCAGCCGGCCATGGCGGAAGTGAARSTTGAGGARTC TTACTCGCGGCCCAGCCGGCCATGGCGCAGGTTACTCTRAAASARTC TTACTCGCGGCCCAGCCGGCCATGGCGCAGGTCCAACTVCAGCARCC TTACTCGCGGCCCAGCCGGCCATGGCGGATGTGAACTTGGAASARTC TTACTCGCGGCCCAGCCGGCCATGGCGGAGGTGAAGGTCATCGARTC

TTACTCGCGGCCCCCGAGGCCGCGGCCGCCACCACCAGAACCACCACCACC CGAGGAAACGGTGACCGTGGT TTACTCGCGGCCCCCGAGGCCGCGGCCGCCACCACCAGAACCACCACCACCCGAGGAGAC TGTGAGAGTGGT TTACTCGCGGCCCCCGAGGCCGCGGCCGCCACCACCAGAACCACCACCACCCGCAGAGA CAGTGACCAGAGT TTACTCGCGGCCCCCGAGGCCGCGGCCGCCACCACCAGAACCACCACCACCCGAGGAGA CGGTGACTGAGGT

TTACTCGCGGCCCAGCCGGCCGGCGGCCGCGGCGGCGGCGGCTCCGAYATCCAGCTGAC TCAGCC TTACTCGCGGCCCAGCCGGCCGGCGGCCGCGGCGGCGGCGGCTCCGAYATTGTTCTC

VH rev

VL for

Sequence (50 ! 30 )

VH for

Mix

Table 1 Primer sets

(continued)

2 2 3

10 10 10 10

3 3 2 2 4 3 2 3 4 2 3 2 2 3 3 3 3 2 2

μL

Identification of Antibodies Targeting Cell Surface Antigens 65

Mix

Table 1 (continued) μL 3 3 4 4 2 2 3 4 4 4 2 2 2 2 2.8

Sequence (50 ! 30 )

WCCCAGTC TTACTCGCGGCCCAGCCGGCCGGCGGCCGCGGCGGCGGCGGCTCCGAYATTGTGMTMAC TCAGTC TTACTCGCGGCCCAGCCGGCCGGCGGCCGCGGCGGCGGCGGCTCCGAYATTGTG YTRACACAGTC TTACTCGCGGCCCAGCCGGCCGGCGGCCGCGGCGGCGGCGGCTCCGAYATTGTRA TGACMCAGTC TTACTCGCGGCCCAGCCGGCCGGCGGCCGCGGCGGCGGCGGCTCCGAYATTMAGA TRAMCCAGTC TTACTCGCGGCCCAGCCGGCCGGCGGCCGCGGCGGCGGCGGCTCCGAYATTCAGATGA YDCAGTC TTACTCGCGGCCCAGCCGGCCGGCGGCCGCGGCGGCGGCGGCTCCGAYATYCAGA TGACACAGAC TTACTCGCGGCCCAGCCGGCCGGCGGCCGCGGCGGCGGCGGCTCCGAYATTGTTCTCA WCCAGTC TTACTCGCGGCCCAGCCGGCCGGCGGCCGCGGCGGCGGCGGCTCCGAYATTGWGC TSACCCAATC TTACTCGCGGCCCAGCCGGCCGGCGGCCGCGGCGGCGGCGGCTCCGAYATTSTRA TGACCCARTC TTACTCGCGGCCCAGCCGGCCGGCGGCCGCGGCGGCGGCGGCTCCGAYRTTKTGA TGACCCARAC TTACTCGCGGCCCAGCCGGCCGGCGGCCGCGGCGGCGGCGGCTCCGAYATTGTGA TGACBCAGKC TTACTCGCGGCCCAGCCGGCCGGCGGCCGCGGCGGCGGCGGCTCCGAYATTGTGATAAC YCAGGA TTACTCGCGGCCCAGCCGGCCGGCGGCCGCGGCGGCGGCGGCTCCGAYATTGTGA TGACCCAGWT TTACTCGCGGCCCAGCCGGCCGGCGGCCGCGGCGGCGGCGGCTCCGAYATTGTGA TGACACAACC

66 Steffen Krohn et al.

TTACTCGCGGCCCCCGAGGCCGCACGTTTKATTTCCAGCTTGG TTACTCGCGGCCCCCGAGGCCGCACGTTTTATTTCCAACTTTG TTACTCGCGGCCCCCGAGGCCGCACGTTTCAGCTCCAGCTTGG TTACTCGCGGCCCCCGAGGCCGCACCTAGGACAGTCAGTTTGG

19 9.5 9.5 2

Annealing regions of primer for mouse V and J genes with degenerated bases were taken from Burmester et al. [11]. Restriction sites (SfiI and NotI, bold) and linker sequences (underlined) were added to 50 ends of the primers. Mix primers (each 100 μM) according to the degree of degeneration using the indicated volumes. for: forward, rev: reverse, VH: variable heavy chain, VL: variable light chain; R ¼ A, G; Y ¼ C, T; M ¼ A, C; K ¼ G, T; S ¼ C, G; W ¼ A, T; H ¼ A, C, T; B ¼ C, G, T; V ¼ A, C, G; D ¼ A, G, T

VL rev

TTACTCGCGGCCCAGCCGGCCGGCGGCCGCGGCGGCGGCGGCTCCGAYATTTTGCTGAC TCAGTC TTACTCGCGGCCCAGCCGGCCGGCGGCCGCGGCGGCGGCGGCTCCGATGCTGTTGTGAC TCAGGAATC

Identification of Antibodies Targeting Cell Surface Antigens 67

68

Steffen Krohn et al.

15. Electroporation cuvette (Bio-Rad). 16. Super Optimal Broth (S.O.C.) medium (Thermo Fisher Scientific). 17. 2xYT medium: 16 g Tryptone, 10 g yeast extract, 5 g NaCl. Autoclave and store at room temperature. 18. Ampicillin stock: 100 mg/mL ampicillin in dH2O. Sterilize by filtration. 19. NucleoBond Xtra Maxi Kit (Macherey Nagel). 20. NucleoSpin Mini Kit (Macherey Nagel). 21. 2xYT agar plates: 2xYT medium, 1.5% (w/v) agar. Autoclave, add supplements, and prepare plates. Store at 4  C. 22. NotI-HF, including 10
reaction buffer (New England Biolabs). 23. AscI, including 10
reaction buffer (New England Biolabs). 24. pJB12 (kindly provided by Andreas Plu¨ckthun, University of Zu¨rich) [11]. 25. Chloramphenicol stock: 30 mg/mL chloramphenicol in ethanol. Sterilize by filtration. 26. Tetracycline stock: 5 mg/mL tetracycline in ethanol. Sterilize by filtration. 27. Freezing medium: 2xYT medium and 20% (v/v) glycerol. Autoclave and store at room temperature. 2.3 Phage Display with Cellular Panning

1. SB medium: 30 g/L tryptone, 20 g/L yeast extract, and 10 g/L MOPS (pH 7.0 NaOH). Autoclave and store at room temperature. 2. Glucose stock: 20% (w/v) glucose in dH2O. Sterilize by filtration. 3. Isopropyl β-d-1-thiogalactopyranoside (IPTG) stock: 1 M IPTG in dH2O. Sterilize by filtration. 4. M13KO7 helper phage (Thermo Fisher Scientific). 5. Kanamycin stock: 50 mg/mL kanamycin in dH2O. Sterilize by filtration. 6. PEG/NaCl solution: 20% (w/v) polyethylene glycol (PEG) 6000 in 2.5 M NaCl solution. Autoclave, stir, and store at room temperature. 7. Blocking buffer: 4% (w/v) BSA in PBS. Sterilize by filtration. 8. Washing buffer: 2% (w/v) BSA in PBS. Sterilize by filtration. 9. Trypsin solution: 50 mM TRIS (pH 7.4 HCl), 1 mM CaCl2, and 10 mg/mL trypsin (Sigma-Aldrich). Sterilize by filtration. Store in aliquots at 20  C.

Identification of Antibodies Targeting Cell Surface Antigens

69

10. Trypsin/PBS solution: Dilute trypsin solution 1:10 in PBS. Prepare immediately before use. 11. Sodium azide stock: 2% (w/v) sodium azide. Sterilize by filtration. 12. Anti-fd bacteriophage rabbit antibody (Sigma-Aldrich). 13. Anti-rabbit IgG F(ab)2 fragment-FITC (Jackson Immuno Research). 2.4

Whole Cell ELISA

1. 96-wellplates (flat bottom, Sarstedt). 2. Horseradish peroxidase (HRP)-conjugated anti-M13 mouse IgG1 antibody (Creative Diagnostics). 3. ELISA blocking buffer: 4% (w/v) BSA in PBS. Sterilize by filtration. 4. ELISA washing buffer: 0.1% (w/v) BSA in PBS. Sterilize by filtration. 5. ABTS substrate (Roche).

3

Methods

3.1 Immunization of Mice and Analysis of Antibody Titer in Serum

1. For immunization, inject 1
106 tumor cells (e.g., mononuclear cells from pleural effusion of a patient with plasma cell leukemia (95% CD38+/CD138+ plasma cells) intraperitoneally in BALB/c mice followed by three boosts (first boost at day 14, second boost at day 35, and third boost at day 56). 2. Draw test bleedings after each boost to analyze antibody titer. 3. Allow agglutination of blood (37  C for 1 h) and spin down at max. speed for 10 min. 4. Dilute serum 1:1000 with PBS. 5. Incubate 250,000 tumor cells/sample with diluted serum for 1 h on ice. 6. Add sample with serum of a non-immunized mouse as the control. 7. Wash cells three times with washing buffer and incubate with anti-mouse IgG FITC-labeled secondary antibody (1:20 with washing buffer) for 1 h on ice. 8. Wash cells three times and analyze by flow cytometry. 9. Seven days after the final boost, sacrifice mice and cryopreserve spleens for total RNA preparation (see Note 2).

70

Steffen Krohn et al.

3.2 Generation of Murine scFv Antibody Library from the Spleen

Carry out all procedures on ice and with pre-cooled reagents unless otherwise specified.

3.2.1 Preparation of Total RNA from the Spleen

1. Add the spleen to 6 mL cold TRIzol in a Dounce homogenizer and homogenize till the spleen fully disintegrates (see Note 3). 2. Split solution on six 1.5 mL tubes and perform further preparation of total RNA according to manufacturers’ guidelines. 3. Resolve each RNA pellet in 20 μL nuclease-free water, incubate at 55  C for 10 min, and pool samples. 4. Determine purity and concentration of total RNA by measuring absorbance at 260 and 280 nm according to the standard procedure. 5. Integrity of RNA can be analyzed by electrophoresis (e.g., Bioanalyzer from Agilent or agarose gel electrophoresis under denaturing conditions with formaldehyde according to the standard procedure). Two bands of ribosomal RNA (28S and 18S) with an intensity ratio of approximately 2:1 confirm the integrity of RNA (Fig. 2a). 6. Store RNA at 80  C.

3.2.2 Generation of the First Strand cDNA by Reverse Transcription

1. Add 10 μg of total RNA to 1 μg oligo(dT)15 and adjust the volume to 28 μL with nuclease-free water. 2. Place tubes in a thermo cycler and incubate at 70  C for 10 min. Cool down on ice. 3. Add 2 μL of RNasin (80 U), 10 μL 5
reaction buffer, 3 μL dNTP mix (25 mM each), 5 μL DTT (100 mM), and 2 μL SuperScript II (200 U/μL). 4. Incubate for 10 min at 21  C, 50 min at 42  C, and 5 min at 90  C for inactivation. Cool down on ice. 5. Add 1 μL Ribonuclease H (1 U/μL) and incubate for 20 min at 37  C to degrade the remaining RNA. 6. Store cDNA at 20  C.

3.2.3 Amplification of Mouse V-Regions by PCR

1. Mix primer sets for amplifying mouse variable heavy (VH) and light (VL) chain sequences according to Table 1. 2. Prepare at least three reactions each for VH and VL amplification. 3. Prepare a 5:1 mix of Pwo and Taq polymerase (see Note 4). 4. For one PCR reaction, add 5 μL cDNA, 5 μL 10
Pwo reaction puffer, 1 μL dNTP mix, 4 μL MgSO4, 2 μL DMSO, 1 μL forward primer mix, 1 μL reverse primer mix, and 2 μL polymerase mix to 29 μL nuclease-free water.

Identification of Antibodies Targeting Cell Surface Antigens

71

Fig. 2 Agarose gel electrophoresis images of total RNA, PCR products, and restriction endonuclease reactions. (a) Total RNA prepared from mouse spleen and analyzed by agarose gel electrophoresis under denaturing conditions with formaldehyde. Two bands of ribosomal RNA (28S and 18S) confirm integrity. (b) Amplification of V-regions (VH and VL) using gradient PCR with annealing temperatures between 45 and 55  C (lane 1–3). Bands below 300 base pairs (bp) are artifacts formed by the primers and can be seen in all samples including negative control (nc). (c) After digestion with the restriction enzymes AscI and NotI, the VL fragment (approximately 500 bp) is released from the plasmid of the VL sub-library (pUC-VL) and the plasmid of the VH sub-library (pUC-VH) is linearized for cloning (approximately 3200 bp). (d) After digestion with the

72

Steffen Krohn et al.

5. Place tubes in a thermo cycler and start program: 1
: 5 min at 92  C. 5
: 1 min at 92  C, 1 min at 45–55  C, and 1 min at 72  C (see Note 5). 25
: 1 min at 92  C, 1 min at 63  C, and 1 min at 72  C. 6. Apply reactions to 1.5% (w/v) agarose gel and separate by electrophoresis (Fig. 2b). 7. Purify approximately 400 base pairs (bps) VH and VL fragments by gel extraction (e.g., QIAquick Gel Extraction Kit) according to the manufacturers’ guidelines. Elute each column with 40 μL nuclease-free water. 3.2.4 Generation of VH and VL Sub-Libraries

1. Digest purified PCR products with SfiI. Adjust volumes to 44 μL with nuclease-free water and add 5 μL 10
reaction buffer and 1 μL SfiI (20 U/μL). Incubate overnight in a thermo cycler at 50  C with the heated lid. 2. Digest 1 μg pUC19-MSC2017 vector using the same setting as in step 1. 3. Gel-purify the digested vector (2888 bps) and V-regions (approximately 400 bps). 4. Adjust volumes of purified V-regions to 90 μL and add 10 μL sodium acetate solution and 2 μL glycogen solution. Mix by vortexing and add 250 μL ice-cold ethanol. Mix by vortexing again. 5. Incubate overnight at 20  C to precipitate the DNA. 6. Centrifuge at 4  C, 16,000
g for 20 min and remove the supernatant by pipetting. 7. Wash pellets with cold 70% (v/v) ethanol and repeat centrifugation. 8. Remove the supernatant thoroughly by pipetting. 9. Air-dry at room temperature until pellets appear translucent (max. 15 min). 10. Dissolve DNA in 5–7 μL nuclease-free water. 11. Determine the concentration of digested vector and concentrated V-regions (inserts). 12. Mix each of the V-regions with vector in a 5:1 ratio (150–400 ng/reaction) and adjust volume to 17 μL with nuclease-free water. Add 2 μL 10
ligase reaction buffer and

ä Fig. 2 (continued) restriction enzyme SfiI, the scFv fragment (approximately 800 bp) is released from the plasmid of the scFv library (pUC-scFv) for insertion in the phagemids of the pAK/pJB vector series (here pJB12; linearized 5312 bp fragment for cloning)

Identification of Antibodies Targeting Cell Surface Antigens

73

1 μL T4 DNA ligase (see Note 6). Incubate overnight at 16  C and afterwards inactivate 10 min at 65  C in a thermo cycler. 13. To de-salt ligation products, transfer reaction mixtures to 1.5 mL tubes and add 200 μL of n-butanol (10
volume of ligation reaction). Mix by vortexing and centrifuge with 25,000
g for 10 min at room temperature. Remove the supernatant. 14. Wash pellets with cold 70% (v/v) ethanol and repeat centrifugation. 15. Remove the supernatant thoroughly by pipetting. 16. Air-dry at room temperature until pellets appear translucent (max. 15 min). 17. Dissolve DNA pellets in 12 μL nuclease-free water. 18. Perform three electroporations with 4 μL DNA and 50 μL E. coli XL1-Blue for each VH and VL ligation reaction (see Note 7). 19. Resuspend bacteria from each electroporation immediately in 1 mL S.O.C. medium, pool bacteria of VH and VL cloning, respectively, and incubate at 37  C for 1 h with shaking. 20. Plate bacteria on at least 15 2xYT agar plates with 100 μg/mL ampicillin (150 mm diameter) per V-region. Furthermore, prepare serial dilutions and spread on plates to calculate the number of colony forming units (cfu) and to estimate the efficacy of ligation and transformation (see Note 8). 21. Incubate overnight at 37  C. 22. Wash bacterial colonies from agar plates using 2xYT medium and pellet bacteria by centrifugation (3,350
g, 20 min, 4  C). 23. Prepare plasmid DNA from bacteria using an appropriate isolation kit (e.g., NucleoBond Xtra Maxi) and determine purity and concentration of DNA by measuring absorbance at 260 and 280 nm according to the standard procedures (see Note 9). 24. Store plasmid DNA of VH and VL sub-libraries at 20  C. 25. In parallel, pick single bacterial colonies (at least 20) from titration plates and inoculate tubes with 3 mL 2xYT medium with 100 μg/mL ampicillin with each colony. Incubate at 37  C overnight with shaking. 26. Prepare plasmid DNA from 1 mL bacterial suspension using an appropriate isolation kit (e.g., NucleoSpin Mini) and determine the successful insertion of V-region by SfiI digestion as in step 1 following agarose gel electrophoresis. 27. At least 1
107 cfu and an insertion rate of at least 94% can be expected for VH and VL sub-libraries.

74

Steffen Krohn et al.

3.2.5 Assembly of VH and VL to scFv

1. Digest 3 μg of VH and VL sub-libraries with NotI-HF and AscI. Adjust volumes to 130 μL with nuclease-free water and add 15 μL 10
reaction buffer, 1.5 μL NotI-HF (20 U/μL) and 3 μL AscI (10 U/μL). Incubate overnight in a thermo cycler at 37  C with a heated lid. 2. Gel-purify the linearized pUC vector including the VH (approximately 3200 bp) and released VL fragments (approximately 500 bp) (Fig. 2c). 3. Concentrate the VL fragments and determine the concentrations as described in Subheading 3.2.4, steps 4–11. 4. Mix pUC vector (including VH) and VL fragments in a 5:1 ratio (150–400 ng/reaction) and adjust volume to 17 μL with nuclease-free water. Add 2 μL 10
reaction buffer and 1 μL T4 DNA ligase. Incubate overnight at 16  C and afterwards inactivate 10 min at 65  C in a thermo cycler. 5. Perform de-salting, transformation in E. coli XL1-Blue, plating and plasmid preparations as described in Subheading 3.2.4, steps 13–26. If insertion was successful, SfiI test-digestion should now release approximately 800 bp scFv fragments instead of the previous inserted approximately 400 bp V-regions. 6. Again, at least 1
107 cfu and an insertion rate of at least 94% can be expected for scFv libraries. 7. For the final cloning step, digest 3 μg of scFv library and phagemid vector pJB12 with SfiI. Adjust volumes to 132 μL with nuclease-free water and add 15 μL 10
reaction buffer and 3 μL SfiI. Incubate overnight in a thermo cycler at 50  C with a heated lid (see Note 10). 8. Gel-purify the linearized vector (5312 bp) and released scFv fragments (approximately 800 bp) (Fig. 2d). 9. Concentrate the scFv fragments and determine the concentrations as described in Subheading 3.2.4, steps 4–11. 10. Mix the vector and scFv fragments in a 5:1 ratio (150–400 ng/ reaction) and adjust volume to 17 μL with nuclease-free water. Add 2 μL 10
reaction buffer and 1 μL T4 DNA ligase. Incubate overnight at 16  C and afterward inactivate 10 min at 65  C in a thermo cycler. 11. Perform de-salting, transformation in E. coli XL1-Blue and plating as described in Subheading 3.2.4, steps 13–21 replacing ampicillin with 1% (w/v) glucose, 30 μg/mL chloramphenicol, and 10 μg/mL tetracycline. 12. Instead of plasmid preparation, wash the bacterial colonies from agar plates using 2xYT medium, pellet bacteria (1,900
g, 10 min, 4  C), wash pellet with freezing medium and resuspend in a small volume of freezing medium (4–7 mL).

Identification of Antibodies Targeting Cell Surface Antigens

75

13. Aliquot and freeze with liquid nitrogen. Store library at 80  C. 14. Again, separate single bacterial colonies for test-digestion with SfiI as described in Subheading 3.2.4, steps 25 and 26 replacing the ampicillin with 1% (w/v) glucose, 30 μg/mL chloramphenicol, and 10 μg/mL tetracycline. After successful ligation, approximately 800 bp scFv fragments should be released instead of the 2101 bp tet-cassette. 15. At least 1
107 cfu and an insertion rate of at least 94% can be expected. 3.3 Phage Display with Cellular Panning 3.3.1 Preparation of Phages

1. Inoculate 500 mL SB medium (supplemented with 1% (w/v) glucose, 30 μg/mL chloramphenicol, and 10 μg/mL tetracycline) with E. coli containing the final scFv library to adjust an OD600 of approximately 0.1. 2. Incubate at 37  C with shaking until OD600 reaches 0.5–0.8. 3. Transfer 50 mL to a pre-warmed reaction tube and add 25 μL IPTG solution and M13KO7 helper phage with a 20
multiplicity of infection (MOI) (see Notes 11 and 12). 4. Incubate for 30 min in a water bath at 37  C. 5. Transfer suspension to 500 mL SB medium (supplemented with 1% (w/v) glucose, 30 μg/mL chloramphenicol, and 0.5 mM IPTG). 6. Incubate for 1.5 h at 37  C with shaking and add kanamycin solution stock to adjust a final concentration of 25 μg/mL. 7. Incubate overnight at 30  C with shaking. 8. Split suspension on appropriate centrifugation tubes and centrifuge for 20 min at 4  C with 3,830
g. 9. Phages will be precipitated by mixing the supernatant with 1=4 volume ice-cold PEG/NaCl solution: Prepare fresh centrifugation tubes, add the necessary amount of PEG/NaCl and supernatant. Mix thoroughly and incubate on ice for at least 30 min. 10. Centrifuge for 20 min at 4  C with 12,400
g. 11. Carefully discard the supernatant and air-dry pellets for 10 min. 12. Remove remaining liquids and resuspend in 4 mL PBS. 13. Centrifuge the suspension at 25,000
g for 10 min to spin down residual bacterial debris. Transfer supernatant to new tubes. 14. Take a sample for phage titration and add sodium azide at a final concentration of 0.02% (w/v). Store phages at 4  C.

76 3.3.2

Steffen Krohn et al. Phage Titration

1. Inoculate 2xYT medium (supplemented with 10 μg/mL tetracycline) with E. coli XL1-Blue to adjust an OD600 of approximately 0.1 and grow until OD600 reaches 0.5. 2. Prepare a serial 100-fold dilution of the phage suspension (up to 1:1010) in 2xYT medium. 3. Add 10 μL of each dilution to 100 μL bacterial suspension, mix carefully and incubate at 37  C for 30 min without shaking. 4. Plate bacteria on 2xYT agar plates with 1% (w/v) glucose and 30 μg/mL chloramphenicol. 5. Incubate overnight at 37  C and count the cfu, representing the amount of phage particles containing phagemid. 6. About 1
1012 to 1
1013 cfu can be expected per preparation.

3.3.3

Cellular Panning

1. For depletion/preabsorption to remove unwanted binders, incubate 1
107 to 1
108 cells of a healthy donor (e.g., mononuclear cells (MNC) or leukocytes prepared from peripheral blood) in 2 mL blocking buffer for 30 min at 4  C under rotation. 2. Add at least 1
1012 cfu phages and adjust the final volume to 4 mL with blocking buffer. Incubate for 2 h at 4  C under rotation. 3. Meanwhile, block the cells of a second independent healthy donor as described in step 1. 4. Separate cells and unbound phages by centrifugation (300
g, 4 min, 4  C) and use the supernatant of the first donor/ depletion cycle to resuspend the cells of the second donor. Again, incubate for 2 h at 4  C under rotation. 5. Separate cells and unbound phages by centrifugation (300
g, 4 min, 4  C) and save the supernatant for positive selection. 6. Incubate 1
106 tumor cells in 500 μL blocking buffer for 30 min at 4  C under rotation. 7. Pellet tumor cells by centrifugation (300
g, 4 min, 4  C) and resuspend cells in supernatant from step 5. Incubate for 2 h at 4  C under rotation. 8. Wash cells five times with cold washing buffer and two times with cold PBS. 9. In parallel, inoculate 2xYT medium (supplemented with 10 μg/mL tetracycline) with E. coli XL1-Blue to adjust an OD600 of approximately 0.1 and grow until OD600 reaches 0.5 for infection. 10. Resuspend tumor cells in 1.5 mL fresh trypsin/PBS solution and incubate for 10 min under rotation at room temperature (see Note 13).

Identification of Antibodies Targeting Cell Surface Antigens

11. Separate cells from eluted (18,000
g, 10 min, 21  C).

phages

by

77

centrifugation

12. Add the eluted phages to 10 mL bacterial suspension, mix carefully, and incubate for 30 min in a water bath of 37  C. 13. Pellet bacteria (1,900
g, 10 min, 4  C), remove the supernatant, and resuspend in 2 mL 2xYT medium. 14. Plate bacteria on at least 10 2xYT agar plates with 1% (w/v) glucose, 30 μg/mL chloramphenicol, and 10 μg/mL tetracycline (150 mm diameter) (see Note 8). Furthermore, prepare serial dilutions and spread on plates. 15. Incubate overnight at 37  C. 16. Determine the number of cfu to estimate the phage titer. 17. Plates with serial dilutions can be used to pick separate bacterial colonies for the preparation of both monoclonal phages for ELISA and phagemid DNA for Sanger sequencing as described in Subheadings 3.4.1 and 3.5.1, respectively. 18. Wash bacterial colonies from agar plates using 2xYT medium and pellet bacteria by centrifugation (1,900
g, 10 min, 4  C). 19. Use bacteria for the next preparation of phages and/or freeze the remaining bacteria as described in Subheading 3.2.5, steps 12 and 13 (see Note 14). 3.3.4 Binding Analysis of Polyclonal Phage Preparations by Flow Cytometry

1. Incubate 5
105 target cells/sample in flow cytometry blocking buffer for 30 min on ice. 2. Pellet cells by centrifugation and resuspend them with phages diluted in flow cytometry blocking buffer (1
1010 to 1
1012 cfu/sample). Include phages with scFv that do not bind an antigen on your target cells as isotype control. 3. Incubate for 1 h on ice. 4. Wash cells three times with flow cytometry blocking buffer and resuspend cells with flow cytometry blocking buffer containing anti-fd bacteriophage rabbit antibody (1:100) (see Note 15). 5. Incubate for 1 h on ice. 6. Wash cells three times with flow cytometry blocking buffer and resuspend cells with flow cytometry blocking buffer containing anti-rabbit IgG F(ab)2 fragment-FITC (1:20) (see Note 15). 7. Wash cells three times with flow cytometry blocking buffer and measure fluorescence intensity in a flow cytometer (see Note 16).

78

Steffen Krohn et al.

3.4 Characterization of Monoclonal Phage Antibodies 3.4.1 Production of Monoclonal Phages

1. Use separated bacterial colonies after re-infection from Subheading 3.3.3, step 17 to inoculate tubes with 4 mL 2xYT medium (supplemented with 1% (w/v) glucose, 30 μg/mL chloramphenicol, and 10 μg/mL tetracycline). 2. Incubate at 37  C with shaking until OD600 reaches 0.5 and add helper phages with a 20
MOI (see Note 12). 3. Incubate at 37  C for 30 min without and for 30 min with shaking. 4. Pellet bacteria by centrifugation (1,900
g, 10 min, 4  C) and remove supernatant. 5. Resuspend the pellets in 5 mL 2xYT medium (supplemented with 1% (w/v) glucose, 30 μg/mL chloramphenicol, 0.5 mM IPTG, and 25 μg/mL kanamycin) and incubate overnight at 37  C with shaking. 6. Pellet bacteria by centrifugation (3,350
g, 20 min, 4  C) and transfer the supernatants into fresh reaction tubes. 7. Phage titer can be determined as described in Subheading 3.3.2 (see Note 17). 8. Store supernatant at 4  C.

3.4.2

Whole Cell ELISA

1. Block 96-well-plates overnight at 4  C with ELISA blocking buffer. 2. Prepare a sufficient amount of target cells in order to use 1
106 cells/well (100 μL). Incubate cells for 30 min in cold ELISA blocking buffer. Add different control cells (e.g., MNC from a healthy donor, other types of tumor cells, or stable transfected Chinese hamster ovary (CHO) cells expressing known TAA) to the panel to analyze the expression profile of bound antigens (Fig. 3). 3. Distribute the cells to wells, sediment cells by centrifugation (530
g, 20 min, 4  C), and remove supernatant. 4. Resuspend cells in 50 μL ELISA blocking buffer and add 150 μL of monoclonal phages (see Note 18). Mix gently by pipetting. 5. Wash cells three times with ELISA washing buffer and resuspend cells in 100 μL ELISA blocking buffer with HRP-conjugated anti-M13 antibody (1:2000) (see Note 15). 6. Wash cells three times with ELISA washing buffer and incubate with 100 μL ABTS substrate solution. 7. Signal usually appears after 5–30 min. 8. Measure absorbance at 405 nm (ref. 492 nm) with a microplate reader.

Identification of Antibodies Targeting Cell Surface Antigens

79

Fig. 3 Binding analysis of monoclonal phages by whole cell ELISA. Nine monoclonal phages that were isolated from a scFv library by screening on malignant plasma cells were tested on three plasma cell lines (L-363, U-266, and INA-6), Raji Burkitt’s lymphoma cell line, CEM T cell acute lymphoblastic leukemia cell line, and cells of healthy donors (mononuclear cells (MNC) from blood and HUVEC endothelial cells). A phage displaying a CD7-specific scFv was used as control. Monoclonal phages #5 and #23 were identified as specific for CD38 and ICAM-1 using stable transfected Chinese hamster ovary (CHO) cells expressing CD38 or CD54/ICAM-1. Color bar was set OD 405 nm ¼ 0 (white) to OD 405 nm > 4 (dark grey) 3.5 Sequencing Analyses 3.5.1 Sanger Sequencing of Monoclonal Phages

1. Use separated bacterial colonies after re-infection from Subheading 3.3.3, step 17 to inoculate tubes with 3 mL 2xYT medium supplemented with 1% (w/v) glucose, 30 μg/mL chloramphenicol, and 10 μg/mL tetracycline. 2. Incubate overnight at 37  C with shaking. 3. Prepare plasmid DNA from 1 mL bacterial suspension using an appropriate isolation kit (e.g., NucleoSpin Mini) and determine purity and concentration of DNA by measuring absorbance at 260 and 280 nm according to the standard procedures. 4. Perform Sanger sequencing using forward and reverse primer that bind to the phagemid sequence 50 and 30 of the scFv. In our case: forward: 50 -CGTATGTTGTGTGGAATTGTGAGCGG-30 reverse: 50 -CATAGCCCCCTTATTAGCGTTTGCC-30

80

Steffen Krohn et al.

5. The clonotypes of the V-regions (VH and/or VL) can be analyzed using online tools such as IMGT/V-QUEST or ARResT/Interrogate [5, 6]. 3.5.2 Next-Generation Sequencing (NGS) of Libraries (Optional)

1. Prepare phagemid DNA from pooled bacteria of the scFv library before panning and after each panning/re-infection using the plasmid isolation kit (e.g., NucleoBond Xtra Maxi) (see Note 9). 2. Determine purity and concentration of DNA by measuring absorbance at 260 and 280 nm according to the standard procedures. 3. Design primers suitable for the specific NGS device (e.g., Illumina MiSeq system) to both amplify VH and/or VL and adding adapters and indexes to 50 and 30 ends of the PCR products. 4. Perform runs on the NGS device using appropriate reagents (e.g., MiSeq Reagent Kit v3) (see Note 19). 5. The clonotypes of the V-regions (VH and/or VL) can be analyzed using online tools such as IMGT/HighV-QUEST or ARResT/Interrogate [5, 6]. 6. Calculate the percentage of reads for each clonotype in the initial library and after each panning to analyze the increase or decrease of frequencies. 7. Compare with clonotypes of the monoclonal phages.

4

Notes 1. Since these primers are long, we recommend polyacrylamide gel electrophoresis (PAGE) purified oligos to ensure a high percentage of full-length primers. 2. Place the spleen immediately in liquid nitrogen to prevent degradation of RNA due to high levels of RNases. 3. Handle RNA with care: Use RNase-free plastic ware and solutions, use pipette tips with filters, and sterilize glass ware by baking at 180  C for at least 3 h. 4. Adding a small amount of Taq polymerase to the proof-reading Pwo polymerase increases the yield of PCR products. 5. Splitting the (at least) three tubes on a gradient thermo cycler with annealing temperatures between 45 and 55  C increases the probability of amplifying a wide variety of V-regions. The PCR products of the first five cycles will be templates for the following 25 cycles of amplification. 6. Handle ligation reaction buffer with care: First, thaw and mix buffer at room temperature until it is fully dissolved and no precipitate can be observed. Then aliquot the buffer and properly store at 20  C to preserve the included ATP.

Identification of Antibodies Targeting Cell Surface Antigens

81

7. Store electroporation cuvettes at 20  C until use. After electroporation, immediately resuspend bacteria in S.O.C. medium to increase transformation efficiency. 8. An increased agar surface for plating allows a competition-free outgrowth of single bacterial colonies. 9. Measure the weight of the bacterial pellets before performing the plasmid/phagemid isolation and use the recommended amount noted in the manual of your isolation kit to avoid using too less or too much starting material. 10. Our inserted SfiI sites are compatible with the pAK/pJB vector series as described by Burmester et al. [11]. 11. Phages are viral particles with a high stability, handle with care: Use pipette tips with filters, prevent the spreading of liquids and the formation of aerosols, separate working with phages and other steps (e.g., generation of library) to avoid crosscontamination, use appropriate disinfection reagents. 12. Estimate the number of bacteria (OD600 of 1 equals 1
109 bacteria/mL). Use 20 times more phage particles (specified as plaque forming units (pfu)/mL) than number of bacteria for infection. 13. In our hands, the elution by trypsin digestion proved to be more efficient than pH shift (50 mM HCl). The pJB12 phagemid contains a trypsin site between scFv and coat protein pIII. 14. The number of cellular panning cycles is critical for the screening. On the one hand, phages binding tumor-specific or tumor-associated antigens need to be enriched. On the other hand, the diversity of antibodies will be reduced by every panning step. Properties of both, antigens (e.g., low abundance on cell) and scFv (e.g., production rate of the phages), lead to a competition while amplification and create a bias. The panning can be monitored using NGS as described in Subheading 3.5.2. In our hands, enrichment of individual antibodies was observed after the second panning round, although a third panning step can be helpful depending on antibody library and tumor cells. 15. Check whether your target cells have Fc receptors that may interact with the antibodies used for detection to avoid high background signals (e.g., anti-M13 mouse IgG2a antibody would bind to human high-affinity Fcγ receptor I (CD64)). 16. Alternatively, binding of the phage preparations can be measured by ELISA as described in Subheading 3.4.2. 17. Phage titer can be determined for each monoclonal phage to avoid a low or high signal due to different amounts of phages in the supernatants. Alternative methods for phage titration are described in the literature (e.g., ELISA [12]).

82

Steffen Krohn et al.

18. If phage preparations differ in their titers, adjust the volume of bacteria supernatant to use the same total amount of phages. Add a phage to the panel displaying a scFv that does not bind an antigen on the tumor cell as a negative control. 19. To ensure that the library is sequenced deep enough, generate more reads than cfu in the original antibody library before panning (max. diversity). The number of reads can be adjusted to the number of cfu after each panning.

Acknowledgments The pJB12 vector was kindly provided by Professor Andreas Plu¨ckthun. This work was supported by a research grant from the Else Kro¨ner-Fresenius-Stiftung to K.K. and a research grant from the Deutsche Krebshilfe e.V. to M.P. References 1. Carter PJ, Lazar GA (2018) Next generation antibody drugs: pursuit of the “high-hanging fruit”. Nat Rev Drug Discov 17(3):197–223 2. Carter P, Smith L, Ryan M (2004) Identification and validation of cell surface antigens for antibody targeting in oncology. Endocr Relat Cancer 11(4):659–687 3. Bakker AB, van den Oudenrijn S, Bakker AQ, Feller N, van Meijer M, Bia JA et al (2004) C-type lectin-like molecule-1: a novel myeloid cell surface marker associated with acute myeloid leukemia. Cancer Res 64(22):8443–8450 4. Ljungars A, Ma˚rtensson L, Mattsson J, Kovacek M, Sundberg A, Tornberg UC et al (2018) A platform for phenotypic discovery of therapeutic antibodies and targets applied on Chronic Lymphocytic Leukemia. NPJ Precis Oncol 2(1):18 5. Alamyar E, Duroux P, Lefranc MP, Giudicelli V (2012) IMGT((R)) tools for the nucleotide analysis of immunoglobulin (IG) and T cell receptor (TR) V-(D)-J repertoires, polymorphisms, and IG mutations: IMGT/V-QUEST and IMGT/HighV-QUEST for NGS. Methods Mol Biol 882:569–604 6. Bystry V, Reigl T, Krejci A, Demko M, Hanakova B, Grioni A et al (2017) ARResT/ Interrogate: an interactive immunoprofiler for IG/TR NGS data. Bioinformatics 33(3): 435–437 7. Ravn U, Didelot G, Venet S, Ng KT, Gueneau F, Rousseau F et al (2013) Deep

sequencing of phage display libraries to support antibody discovery. Methods 60(1): 99–110 8. Ravn U, Gueneau F, Baerlocher L, Osteras M, Desmurs M, Malinge P et al (2010) By-passing in vitro screening--next generation sequencing technologies applied to antibody display and in silico candidate selection. Nucleic Acids Res 38(21):e193 9. Ljungars A, Svensson C, Carlsson A, Birgersson E, Tornberg UC, Frendeus B et al (2019) Deep mining of complex antibody phage pools generated by cell panning enables discovery of rare antibodies binding new targets and epitopes. Front Pharmacol 10:847 10. Krohn S, Boje AS, Gehlert CL, Lutz S, Darzentas N, Knecht H et al (2022) Identification of new antibodies targeting malignant plasma cells for immunotherapy by nextgeneration sequencing-assisted phage display. Front Immunol 13:908093 11. Burmester J, Plu¨ckthun A (2001) Construction of scFv fragments from hybridoma or spleen cells by PCR assembly. In: Kontermann R, Du¨bel S (eds) Antibody engineering. Springer, Berlin, Heidelberg, pp 19–40 12. Kay BK, Winter J, McCafferty J (1996) Phage display of peptides and proteins: a laboratory manual. Elsevier Science

Chapter 6 Phage Display of Bovine Ultralong CDRH3 Callum Joyce, Louise Speight, Alastair D. G. Lawson, Anthony Scott-Tucker, and Alex Macpherson Abstract Phage display is an in vitro technique used in the discovery of monoclonal antibodies that has been used successfully in the discovery of both camelid VHH and shark variable new antigen receptor domains (VNAR). Bovines also contain a unique “ultralong CDRH3” with a conserved structural motif, comprising a knob domain and β-stalk. When removed from the antibody scaffold, either the entire ultralong CDRH3 or the knob domain alone, is typically capable of binding an antigen, to produce antibody fragments that are smaller than both VHH and VNAR. By extracting immune material from bovine animals and specifically amplifying knob domain DNA sequences by PCR, knob domain sequences can be cloned into a phagemid vector producing knob domain phage libraries. Target-specific knob domains can be enriched by panning the libraries against an antigen of interest. Phage display of knob domains exploits the link between phage genotype and phenotype and could prove to be a high throughput method to discover target-specific knob domains, helping to explore the pharmacological properties of this unique antibody fragment. Key words Antibody display, Antibody engineering, Bovine ultra-long CDR3H antibodies, Stalkknob, Phage display

1

Introduction Phage display is an in vitro technique used to investigate protein– ligand interactions and has been extensively used in the discovery of monoclonal antibodies. Since its inception in 1985, there have been 14 FDA-approved monoclonal antibodies discovered by phage display [1]. The link between the phage genotype and phenotype is key to any successful in vitro display method. The surface expression of antibody fragments is achieved alongside the packaging of DNA sequences into the filamentous phage virion via the use of helper phages. Often, smaller antibody formats such as Fab and scFv are displayed on the surface of the minor coat protein pIII, as they are more appropriate for expression in bacteria and successful display on phage [2]. Phage display remains a robust tool for the discovery of single domain heavy chain formats such as Camelid

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2681, https://doi.org/10.1007/978-1-0716-3279-6_6, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

83

84

Callum Joyce et al.

VHH (nanobodies) and shark variable new antigen receptor domains (VNARs), both of which display unique properties [3]. However, Camelids and cartilaginous fish are not the only animals identified that contain unique heavy chain structures. A small subset of bovine IgG and IgM antibodies contain an ultralong CDRH3 region, which was first described in 1997 [4] and first structurally characterized in 2013 [5]. Within these ultralong regions, small cysteine-rich globular structures, known as knob domains, sit atop two antiparallel β-strands, which hold the knob domain up to 45 Å above the neighboring CDRs [5, 6]. Interestingly, these 3–6 kDa knob domains can bind to antigens when removed from their antibody scaffolds [7], creating autonomous peptides, approximately three times smaller than both camelid VHH’s and shark IgNARs. These small binding domains could offer potentially useful pharmacological properties, including increased tissue penetrance and a short serum half-life, and may be particularly adept to allosteric modulation [8, 9]. Due to their small size, they are, unusually for antibodies, also amenable to production via solid-phase peptide synthesis [10]. Previously, antigen-specific knob domains have been discovered using a combination of fluorescence-activated cell sorting and deep sequencing of the CDR-H3 region [7]. The discovery of ultralong CDRH3 has also been described by yeast display [11] and we have recently published the first account of the construction of a knob domain phage library [12]. Knob domain phage libraries could provide a low-cost, high throughput in vitro method for the discovery of target-specific knob domains, incorporating the benefits of this robust antibody discovery platform, which are already well established for both VHHs and IgNARs [13, 14]. The principles of knob domain phage library construction are similar to the construction of other phage libraries. A schematic summary of the essential steps is given in Fig. 1. Once immune tissues have been harvested from a bovine animal, the repertoire of bovine immunoglobulin knob domain sequences is selectively amplified, using both the immunoglobulin framework regions and the knob domain stalk sequences. Primers designed to anneal to these regions are used in sequential PCRs. The amplified knob domain DNA is cloned upstream of the g3p gene within a phagemid vector. Exploiting the physical link between phage genotype and phenotype, the knob domain-pIII fusion proteins are displayed on the surface of the subsequent rescued phage library [15] and knob domains that are target specific can be discovered via biopanning [16]. In this chapter, we present a detailed method for the generation of antigen-enriched ultralong CDRH3 libraries, which may be performed for any antigen of interest. Ultralong CDRH3 sequences were selectively enriched against the antigen using phage display.

Phage Display of Bovine Ultralong CDRH3

85

Fig. 1 A schematic diagram highlighting the main steps in the construction of bovine ultralong CDRH3s. (1) Blood was taken from bovine animals and (2) B-cells were isolated from the extracted material. (3) cDNA was created in a reverse transcription PCR using PBMC total RNA as a template. (4) Sequential PCR’s using primers against the immunoglobulin framework regions and knob domain ascending and descending stalk regions. (5) Amplified knob domain sequences were cloned into a phagemid vector (6) which was subsequently transformed into bacteria. After rescuing using helper phage (7) a polyclonal population of phage can be precipitated from the culture supernatant. (8) The phage population can be enriched for target-specific knob domains via biopanning, and (9) sequencing of enriched phage identifies target binding sequences

2

Materials

2.1 Harvesting Bovine Lymphocytes

1. Leucosep™ tubes (Greiner Bio-one). 2. Lympholyte (Cedarlane). 3. RPMI media (Thermofisher).

2.2 Lymphocyte RNA Extraction

1. Tissueruptor 2 (Qiagen).

2.3 Reverse Transcription PCR of Lymphocyte RNA

1. Super script IV vilo Master Mix (Invitrogen).

2.4 Primary PCR Amplification of CDRH3

1. Bovine framework 3 primer. 2. Bovine framework 4 primer. 3. Phusion Green Hotstart II Master mix (Thermo Scientific). 4. PCR purification kit.

2.5 Secondary PCR Amplification of Ultralong CDRH3

1. Ascending stalk primer set. 2. Descending stalk primer set. 3. Phusion Green Hotstart II Master mix (Thermo Scientific). 4. PCR purification kit.

86

Callum Joyce et al.

2.6 PCR Restriction Digestion

1. NotI restriction enzyme (New England Biolabs).

2.7 Plasmid Digestion and Ligation with Amplified Ultralong CDRH3

1. NotI restriction enzyme (New England Biolabs).

2.8 Precipitation of Ligation Product

1. 10 mg/mL yeast tRNA (Thermofisher).

2.9 Electroporation of E. coli and Plating Library

1. Electrocompetent E. coli TG1 cells (Lucigen).

2. SfiI restriction enzyme (New Biolabs).

2. SfiI restriction enzyme (New England Biolabs). 3. Gel extraction kit.

2. Recovery medium (Lucigen). 3. 1 mm electroporation cuvettes. 4. LB medium (Tryptone 1% w/v, Yeast Extract 0.5% w/v, and NaCl 0.5% w/v, pH 7). 5. 5 × 90 mm petri dish with LB Agar + 1% Glucose + 100 μg/mL Carbenicillin. 6. Square bioassay dish (Thermo Scientific) with LB Agar + 1% Glucose + 100 μg/mL Carbenicillin.

2.10

Phage Rescue

1. 2TY media (Tryptone 1.6% w/v, Yeast Extract 1% w/v, and NaCl 0.5% w/v, pH 7). 2. 2TY + 40% glycerol. 3. M13K07 helper phage (kanamycin resistance). 4. 2TY media + 50 μg/mL Kanamycin + 100 μg/mL Carbenicillin. 5. ddH2O + 20% PEG-8000 + 2.5 M NaCl. 6. PBS (Sodium chloride 137 mM, Potassium chloride 2.7 mM, Disodium phosphate 10 mM, and Monopotassium phosphate 1.8 mM). 7. PBS + 40% glycerol. 8. Centrifuge fitted with, e.g. Sorvall SS-34 rotor for 225 mL falcon tubes, or Heraeus Megafuge fitted with Fibrelite rotor for 50 mL falcon tubes.

2.11

Biopanning

1. PBS + 2% Bovine serum albumin + 2% skimmed milk powder. 2. M-280 Streptavidin Dynabeads (Thermofisher). 3. Magnetic microtube rack. 4. PBS + 0.1% Tween 20 (SigmaAldrich). 5. 100 mM hydrochloric acid. 6. 1 M Tris–HCl.

Phage Display of Bovine Ultralong CDRH3

2.12 Monoclonal Phage Rescue

87

1. 96-deep well block. 2. 2TY media + 1% Glucose + 100 μg/mL Carbenicillin. 3. M13KO7 helper phage (Kanamycin resistance). 4. 2TY media + 50 μg/mL Kanamycin + 100 μg/mL Carbenicillin.

2.13

Phage ELISA

1. 96-well flat-bottom Nunc MaxiSorp (Thermofisher). 2. 9E10 Anti-myc tag antibody (AbCam). 3. PBS + 2% BSA + 2% skimmed milk powder. 4. PBS + 0.1% Tween 20. 5. Anti-M13-Horseradish (Sigma).

peroxidase

conjugate

antibody

6. One-step TMB solution (Thermofisher). 7. 1.5% NaF solution.

3

Method

3.1 Harvesting Bovine Lymphocytes

1. Take 500 mL of blood from a Holstein Friesian cow. Dilute the blood in the same volume of PBS. 2. Distribute the diluted blood evenly into Leucosep™ tubes (Greiner Bio-one) filled with 10 mL of Lympholyte (Cedarlane). 3. Centrifuge Leucosep™ tubes at 1000 × g for 20 min, ensuring that the centrifuge deceleration is slow. This will prevent the disruption of layers during deceleration. 4. Remove the lymphocyte layer and distribute evenly into 5 × 50 mL falcon tubes and dilute with 25 mL of PBS. Centrifuge falcon tubes at 1000 × g for 10 min. 5. Pool all lymphocytes by gently resuspending into a total of 30 mL of PBS. Take a small aliquot of cells to determine cell number and concentration. 6. Centrifuge the remaining cells at 1000 × g for 10 min. 7. Resuspend cells with RPMI media (Thermofisher) with 10% DMSO to 5 × 107 cells/mL. The cells can then be aliquoted into 1 mL cryotubes and stored at -80 °C until needed.

3.2 Lymphocyte RNA Extraction

1. Use an RNA extraction kit to extract total cell RNA according to the manufacturer’s instructions. Elute RNA in nuclease-free water. Perform any homogenization steps with a Tissueruptor 2 (Qiagen).

88

Callum Joyce et al.

Table 1 Reverse transcription PCR components (one reaction) Solution or component

Volume (μL)

Eluted lymphocyte RNA solution

4 (~800 ng)

Super script IV vilo Master Mix (Invitrogen)

8

Nuclease free water

28

2. Once cell RNA has been eluted, use it immediately in a reverse transcription PCR reaction. Any remaining eluted RNA can be stored at -80 °C. 3.3 Reverse Transcription PCR of Lymphocyte RNA

1. Perform eight reverse transcription PCR reactions. The contents of an individual 40 μL reaction are listed in Table 1. Pipette 4 μL of eluted lymphocyte RNA (this should be approximately 800 ng of total RNA, depending on the concentration of the eluted solution). Run the RT-PCR in a thermocycler set to the following: 25 °C for 10 min, 50 °C for 10 min, and 85 °C for 5 min. 2. Individual RT-PCR reactions can be pooled after cycling.

3.4 Primary PCR Amplification of CDRH3

The primers used in the following primary PCR reactions are as follows: Framework 3 (forward primer): 5′- GGACTCGGCCACMTAY TACTG-3′ (see Note 1). Framework 4 (reverse primer): 5′-GCTCGAGACGGTGAYCAG3′ (see Note 1) 1. Resuspend the lyophilized primers to a concentration of 100 μM. Create working stocks of primers at 10 μM. 2. Perform 8 × primary PCR reactions. The contents of an individual 50 μL reaction are listed in Table 2. Pipette 2 μL of RT-PCR product, 2.5 μL of DMSO, 2.5 μL of both primers, 25 μL of Phusion green hotstart II master mix, and 15.5 μL of nuclease-free water. Run the PCR in a thermocycler set to the following settings: 98 °C for 30 s, 98 °C for 10 s, 62 °C for 30 s, 72 °C for 30 s, and 72 °C for 5 min. The reaction was cycled 20 times. 3. Check the product by agarose gel electrophoresis (see Note 2). 4. Pool all reactions and clean using a PCR clean-up kit. Elute in a volume of 30–60 μL.

Phage Display of Bovine Ultralong CDRH3

89

Table 2 Primary PCR components (one reaction)

3.5 Secondary PCR Amplification of Ultralong CDRH3s

Solution or component

Volume (μL)

Nuclease-free water

15.5

Phusion Green Hotstart II Master mix (Thermo scientific)

25

Framework 3 primer—10 μM

2.5

Framework 4 primer—10 μM

2.5

RT-PCR product

2

DMSO

2.5

The primers used in the following secondary PCR reactions are as follows: Ascending set: 5′-CTCGCGGCCCAGCCGGCCATGGCCACTACTGTGCA CCAAAAAACA-3′ 5′-CTCGCGGCCCAGCCGGCCATGGCCACTACTGTGCA CCAAAGAACC-3′ 5′-CTCGCGGCCCAGCCGGCCATGGCCACTACTGTGCA CCAAAAAACG-3′ 5′-CTCGCGGCCCAGCCGGCCATGGCCACTACTGTGCA CCAACAAACT-3′ 5′-CTCGCGGCCCAGCCGGCCATGGCCACTACTGTGCA CCAACAGACC-3′ 5′-CTCGCGGCCCAGCCGGCCATGGCCACTACTGTGGT CCAGAAAACA-3′ 5′-CTCGCGGCCCAGCCGGCCATGGCCACTACTGTAGT CCAACGAACA-3′. Descending set: 5′-TGATGGGCGGCCGCGGCATCGACGTACCATTCGTA-3′ 5′-TGATGGGCGGCCGCGGTATCGACGTACCATTCGTA-3′ 5′-TGATGGGCGGCCGCGGCTTCGACGTACAATTCGTA-3′ 5′-TGATGGGCGGCCGCGGCATTGACGTAGAATTCGTA-3′ 5′-TGATGGGCGGCCGCGGCCTCGATGTCAAATTCGTA-3′ 5′-TGATGGGCGGCCGCGGTTTCGACGTGGTATTCGTA-3′. Descending primers contain NotI restriction site (GCGGCCGC) for cloning into a phagemid vector whilst ascending contain a SfiI site (GGCCCAGCCGGCC).

90

Callum Joyce et al.

Table 3 Secondary PCR components (1 × reaction) Solution or component

Volume (μL)

Nuclease-free water

15.5

Phusion Green Hotstart II Master mix (Thermo scientific)

25

Ascending primer—10 μM

2.5

Descending primer—10 μM

2.5

Primary PCR product

2 (~100 ng)

DMSO

2.5

1. Resuspend the lyophilized primers to a concentration of 100 μM. Create working stocks of ascending primers at 10 μM. Pool descending primers in a working stock so that each individual primer is at a concentration of 10 μM when pooled. 2. Ascending primers will be tested individually whilst the pooled descending primers will be used in all reactions (see Note 3). 3. Perform seven secondary PCR reactions (one per ascending primer). The contents of an individual 50 μL reaction are listed in Table 3. Pipette 2 μL of Primary PCR product, 2.5 μL of DMSO, 2.5 μL of both primers, 25 μL of Phusion green hotstart II master mix, and 15.5 μL of nuclease-free water. Run the PCR in a thermocycler set to the following: 98 °C for 30 s, 98 °C for 10 s, 62 °C for 30 s, 72 °C for 30 s, and 72 °C for 5 min. The reaction was cycled 30 times. 4. Check the product by agarose gel electrophoresis (see Note 2). 5. Pool all reactions and clean using a PCR clean-up kit. Elute in a volume of 30–60 μL. 3.6 PCR Restriction Digestion

3.7 Plasmid Construction

1. Digest all purified PCR products using NotI and SfiI restriction enzymes according to the manufacturer’s instructions (see Note 4) and column purify. A phagemid vector derived from PUC119 was used throughout library construction. The phagemid vector contains a pelB leader sequence, ahead of the ultralong CDRH3 sequence for display. This is followed by a poly-histidine and c-myc-tag, which is fused directly to the PIII coat protein. The entire open reading frame encoding the fusion protein is under the control of a glucoserepressible lac promoter. Upon superinfection with the helper phage, an M13 replication origin will result in the synthesis and packaging of single-stranded phagemid DNA, encoding the phage display construct within the phage virion. The phagemid vector

Phage Display of Bovine Ultralong CDRH3

91

contains an AmpR gene for resistance to Carbenicillin. The stuffer fragment was removed using NotI and SfiI restriction enzymes, complementary to the digestion of the amplified ultralong CDRH3 DNA. 3.8 Plasmid Digestion and Ligation with Amplified Ultralong CDRH3’s

1. Digest several micrograms of phagemid vector using NotI and SfiI restriction enzymes according to the manufacturer’s instructions (see Note 4). 2. Purify the digested phagemid vector by gel extraction. 3. In a thermocycler, ligate 3 μg of digested phagemid vector with digested ultralong CDRH3 PCR product in a 1:3 molar ratio. Incubate the reaction at 16 °C overnight and column purify after incubation. Elute in a final volume of 100 μL, this will aid in the maximum recovery of the ligated product. The reaction can then be precipitated to increase the concentration of DNA and reduce salt content.

3.9 Precipitation of Ligation Product

1. Add 1 μL of 10 mg/mL yeast tRNA (Thermofisher) and 11 μL of pH 5.2, 3 M sodium acetate to the eluted ligation product. 2. Add 280 μL of ice-cold ethanol and place on ice for 30 min. 3. Centrifuge at maximum speed for 15 min. 4. Remove the supernatant and add 1 mL of 70% ethanol. 5. Centrifuge at maximum speed for 5 min. 6. Remove the supernatant and allow pellets to air dry for 5–10 min. 7. Resuspend DNA pellets in 10 μL of nuclease-free water.

3.10 Electroporation of E. coli and Plating Library 3.10.1

Electroporation

1. Chill 6 × 1 mm electroporation cuvettes on ice for 30 min. Warm both recovery medium and plates to 37 °C. 2. Add the precipitated DNA to 180 μL of commercial electrocompetent TG1 cells (Luicgen) and mix. 3. Add 30 μL of cells to each cuvette, ensuring they are distributed evenly between the electrodes. 4. Shock each cuvette in an electroporator set to 10 μF, 600 Ω and 1800 V. Immediately add 1 mL of recovery medium to each cuvette. Resuspend the cells by pipetting and pool all transformations into a falcon tube. 5. Incubate the culture at 37 °C for 1 h to allow for recovery.

3.10.2

Titer Plates

1. Perform a five-point tenfold serial dilution of recovered cells into LB medium. 2. Plate 50 μL of each dilution onto a 90 mm petri dish containing LB Agar + 1% Glucose + 100 μg/mL Carbenicillin.

92

Callum Joyce et al.

3. Incubate overnight at 37 °C. 4. After incubation, titer plates can be used to calculate library size (see Note 5) and perform monoclonal phage rescue, if necessary. 3.10.3

Library Plate

1. Centrifuge the remaining recovered cells at 4500 rpm for 15 min and resuspend the pellet in 2 mL of LB medium. 2. Spread cells onto a square bioassay dish (Thermo Scientific) containing LB Agar + 1% Glucose + 100 μg/mL Carbenicillin and incubate at 30 °C overnight.

3.11

Phage Rescue

3.11.1 Collecting Biomass in Liquid Media

1. Add 20 mL of 2TY media to the square bioassay dish and scrape the bacteria into the liquid medium. 2. Use the collected culture to seed 200 mL of fresh 2TY media + 1% Glucose + 100 μg/mL Carbenicillin at an OD600 of 0.1. 3. Centrifuge the remaining culture at 6000 rpm for 10 min and resuspend the pellet in 2TY + 40% glycerol, equal to pellet volume by weight. Cells can be frozen at -80 °C.

3.11.2 Helper Phage Infection

1. Grow the 200 mL culture until it reaches an O.D600 of ~0.5. 2. Add M13K07 helper phage to reach an MOI of 20 and leave the culture static at 37 °C for 1 h to allow for infection. 3. Centrifuge the culture at 6000 rpm for 10 min and pour off the supernatant. 4. Resuspend the pellets with 200 mL of 2TY media + 50 μg/mL Kanamycin + 100 μg/mL Carbenicillin and leave shaking at 30 °C overnight.

3.11.3 Phage Precipitation

1. Centrifuge the culture at 6000 rpm for 10 min, decant the supernatant, and centrifuge it again. 2. Decant the supernatant and add 40 mL of 20% PEG-8000 + 2.5 M NaCl. Leave the supernatant on ice for 1 h. 3. Centrifuge the solution at 8000 rpm for 30 min. 4. Pour off the supernatant, resuspend the phage pellet in 20 mL of PBS, and transfer it to a Falcon tube. 5. Centrifuge at 4000 rpm for 10 min. 6. Decant the supernatant and add 4 mL of 20% PEG-8000 + 2.5 M NaCl. Incubate the solution for 1 h on ice. 7. Centrifuge the solution at 4500 rpm for 20 min. 8. Remove the supernatant and resuspend the phage pellet in 5 mL of PBS. Centrifuge the solution again at 4000 rpm for 5 min and decant the supernatant into a fresh falcon tube.

Phage Display of Bovine Ultralong CDRH3

93

9. Add 5 mL of 40% glycerol in PBS. Phage/mL can be estimated by measuring absorbance at 269 nm (see Note 6). Dilute the solution further if necessary to achieve ~1012 phage/mL. Precipitated phage can now be aliquoted and stored at -80 °C until needed. 3.12

Biopanning

3.12.1 Phage Preparation

1. Take 500 μL of precipitated phage at a concentration of ~1012 phage/mL and mix with 500 μL of 2% BSA + 2% skimmed milk powder in PBS. Incubate the mixture for 30 min at room temperature, continuously mixing. 2. After incubation, add biotinylated antigen at a concentration of ≤100 nM. Incubate at room temperature for 1 h, continuously mixing.

3.12.2 Streptavidin Beads Preparation

1. Take 100 μL of Streptavidin Dynabeads (Thermofisher), pellet them using a magnetic rack, and remove the storage solution. 2. Wash the beads once using 1 mL of PBS and a further three times using 1 mL of PBS + 1% BSA. Resuspend the beads in 1 mL of PBS + 1% BSA and incubate at room temperature for 1 h, mixing constantly.

3.12.3 Phage Binding, Washing, and Elution

1. After incubation, pellet the beads and remove the supernatant. Add the blocked phage solution and incubate for 10 min at room temperature, mixing as previous. 2. Pellet the beads and remove the supernatant. Wash the beads four times with 1 mL of 0.1% Tween 20 (Sigma) in PBS. During one of the washing steps, ensure to move the solution to a new microtube, this will help to reduce unspecific background phage. 3. After washing, pellet the streptavidin beads and remove the supernatant. Add 500 μL of 100 mM hydrochloric acid to elute antigen-specific phage from the beads. Incubate for 5 min with intermittent shaking. 4. Add 500 μL of 1 M Tris–HCl to neutralize the acid.

3.12.4 Infection and Titration

1. Add the eluted phage solution to 10 mL of mid-log TG1 cells in 2 × TY and incubate at 37 °C for 30 min without shaking to allow for infection. 2. Perform a five-point, tenfold serial dilution of infected cells into 2TY. Plate 50 μL of each dilution onto a 90 mm petri dish containing LB Agar + 1% Glucose + 100 μg/mL Carbenicillin. Incubate overnight at 37 °C. Output titer and monoclonal phage rescue can be calculated/performed from these plates.

94

Callum Joyce et al.

3. Centrifuge the remaining infected cells at 2500 rpm for 10 min. Resuspend the pellet in 1 mL of 2TY and spread across a 20 cm petri dish of LB Agar + 1% Glucose + 100 μg/mL Carbenicillin. Incubate overnight at 30 °C. After incubation, plates can be scraped down and a phage rescue can be performed, as described previously (see Subheading 3.11). 3.13 Monoclonal Phage Rescue

1. Pick individual colonies from output titer plates and place into a 96-deep well culture block containing 1 mL of 2TY media + 1% Glucose + 100 μg/mL Carbenicillin. Incubate the blocks overnight at 37 °C, shaking at 250 rpm. 2. To a fresh 96-deep well block containing 1 mL of 2TY media + 1% Glucose + 100 μg/mL Carbenicillin, add 100 μL of the overnight culture. Incubate the blocks at 37 °C, shaking at 250 rpm. 3. When the block reaches an approximate O.D600 of 0.5, add M13KO7 helper phage at an MOI of 20, and dilute the phage in 2TY media if necessary. 4. Incubate the block at 37 °C without shaking for 1 h to allow for infection. 5. Centrifuge the block at 3000 rpm for 15 min. 6. Tip the block upside down to decant supernatant and resuspend the pellets in 1 mL of 2TY media + 50 μg/mL Kanamycin + 100 μg/mL Carbenicillin. Incubate the block overnight at 30 °C. The resulting media can then be used in a phage ELISA to assess antigen binding of enriched ultralong CDRH3 phage libraries.

3.14

Phage ELISA

1. Coat a 96-well flat bottom Nunc MaxiSorp (Thermofisher) with 50 μL of antigen of interest at a concentration of 2.5 μg/mL. Coat two other plates with an anti-myc tag, to access the presence of displaying phage, and a control protein, to determine antigen specificity. Incubate overnight at 4 °C. 2. Centrifuge the monoclonal rescue block at 3000 rpm for 15 min and move the supernatant to another 96-deep well block containing an equal volume of 2% BSA + 2% skimmed milk powder in PBS. Incubate at room temperature for 1 h. 3. Wash 96-well flat bottom plates using PBS + 0.1% Tween 20. Repeat the washing three times. 4. Add 100 μL of blocked phage to each well and incubate shaking at room temperature for 1 h. 5. Wash the plates again, as previously described. 6. Add 100 μL per well of an anti-M13-HRP conjugate antibody (Merck) diluted in 1% BSA (see Note 7). Incubate shaking for 1 h at room temperature.

Phage Display of Bovine Ultralong CDRH3

95

Fig. 2 SPR single cycle kinetics data for two target-specific knob domains derived from phage display. Figure shows kinetic data for a GFP-specific knob domain with a 76 nM affinity for GFP

Fig. 3 SPR single cycle kinetics data for two target-specific knob domains derived from phage display. Figure shows kinetic data for a human Interleukin-2 specific knob domain with a 35 nM affinity for the target

7. Wash the plates again, as previously described. 8. Add 50 μL of TMB solution (Thermofisher) and incubate shaking at room temperature for 5 min. Add 50 μL of a 1.5% NaF solution to stop the reaction. 9. Measure the absorbance of each well at 630 and 490 nm, and use the difference between the two as the display result. Any positive results can be sequenced using PCR amplification and sequencing primers. This will give monoclonal antigen-specific knob domain sequences which may then be reformatted as a fusion protein or tested in isolation to assess its functionality outside a phage display setting. Figures 2 and 3 display SPR single cycle kinetics data for two target-specific knob domains derived from phage display using the methods described in this chapter.

96

Callum Joyce et al.

It is likely that multiple rounds of rescuing and biopanning will be needed to discover high-affinity antigen-specific knob domains. By rescuing and repeating biopanning with more stringent conditions (by reducing antigen concentration from 100 to 10, to 1 nM for example) the diversity of the library will reduce and increase the hit rate and prevalence of higher affinity knob domains.

4

Notes 1. Both framework 3 and framework 4 primers contain a degenerate position denoted by the letter “Y”. This position could be either a cytosine or a thymine. This degenerate position helps to cover a wider range of nucleotide combinations in these regions. 2. PCR-amplified products were ran on a 2% agarose gel. A successful PCR reaction should give a band approximately 200 bps. Negative control reactions (reactions run without any template DNA) can be used to help differentiate between primer dimerization and successful amplification of the template DNA. 3. During construction, ascending stalk primers were run in separate PCRs whilst descending stalk primers were pooled and used as a mix in all reactions. The ascending stalk has lower sequence identity between bovines compared to the descending stalk. It is more likely that the ascending primer set performance will vary between samples. Therefore, to access the performance of the individual ascending primers more accurately, they were used separately and not pooled. This helps to: (1) Reduce the overall number of PCRs that need to be performed by eliminating the need to run all descending and ascending primers in all possible combinations, and (2) Quickly identifying the variation in ascending primer performance across different samples. 4. Both the phagemid vector and amplified ultralong CDR3s contained NotI and SfiI restriction sites. Successful digestion of amplified ultralong CDR3s would be difficult to visualize on an agarose gel due to the small change in size. Setting up a parallel reaction with the vector, which can be visualized on a gel, gives a good indication of successful digestion. 5. Total volume recovered from transformation/Volume plated on titer plates = Y Colonies counted on a titer plate × Y × inverse of plate dilution = Library titer

Phage Display of Bovine Ultralong CDRH3

6.

97

ðA269 - A320Þ × 6 × 1016 = Virions=mL Size of phagemid vector ðbpÞ

7. The exact dilution of the anti-M13-HRP conjugate antibody will vary depending on the batch and the producer. Prior to performing a phage ELISA, a titration experiment with a wide range of dilutions may be necessary to determine the optimal dilution factor of the antibody. References 1. Alfaleh M, Alsaab H, Mahmoud A et al (2020) Phage display derived monoclonal antibodies: from bench to bedside. Front Immunol 11: 1986. https://doi.org/10.3389/fimmu. 2020.01986 2. McCafferty J, Griffiths AD, Winter G, Chiswell DJ (1990) Phage antibodies: filamentous phage displaying antibody variable domains. Nature 348(6301):552–554 3. Romao E, Morales-Yanez F, Hu Y et al (2017) Identification of useful nanobodies by phage display of immune single domain libraries derived from camelid heavy chain antibodies. Curr Pharm Des 22(43):6500–6518 4. Berens S, Wylie D, Lopez O (1997) Use of a single VH family and long CDR3s in the variable region of cattle Ig heavy chains. Int Immunol 9(1):189–199 5. Wang F, Ekiert DC, Ahmad I et al (2013) Reshaping antibody diversity. Cell 153(6): 1379–1393 6. Dong J, Finn J, Larsen P, Smith T, Crowe J (2019) Structural diversity of ultralong CDRH3s in seven bovine antibody heavy chains. Front Immunol 10:558. https://doi. org/10.3389/fimmu.2019.00558 7. Macpherson A, Scott-Tucker A, Spiliotopoulos A, Simpson C, Staniforth J et al (2020) Isolation of antigen-specific, disulphide-rich knob domain peptides from bovine antibodies. PLoS Biol 18(9):e3000821 8. Muttenthaler M, King G, Adams D, Alewood P (2021) Trends in peptide drug discovery. Nat Rev Drug Discov 20(4):309–325 9. Macpherson A, Laabei M, Ahdash Z et al (2021) The allosteric modulation of complement C5 by knob domain peptides. eLife 10: e63586. https://doi.org/10.7554/eLife. 63586

10. Macpherson A, Birthley JR, Broadbridge RJ et al (2021) The chemical synthesis of knob domain antibody fragments. ACS Chem Biol 16(9):1757–1769 11. Pekar L, Klewinghaus D, Arras P et al (2021) Milking the cow: cattle-derived chimeric ultralong CDR-H3 antibodies and their engineered CDR-H3-only Knobbody counterparts targeting epidermal growth factor receptor elicit potent NK cell-mediated cytotoxicity. Front Immunol 12:742418. https://doi.org/10. 3389/fimmu.2021.742418 12. Adams R, Callum J, Mikhail K et al (2023) Serum albumin binding knob domains engineered within a VH framework III bispecific antibody format and as chimeric peptides. Frontiers in Immunology 14. https://doi. org/10.3389/fimmu.2023.1170357 13. Schut M, Pepers B, Klooster R et al (2010) B10 Huntingtin specific camelid VHH selected from an immunised llama phage display library. J Neurol Neurosurg Psychiatry 81:A13. h t t p s : // d o i . o r g / 1 0 . 1 1 3 6 / j n n p . 2 0 1 0 . 222596.10 14. Dooley H, Flajnik M, Porter A (2003) Selection and characterization of naturally occurring single-domain (IgNAR) antibody fragments from immunized sharks by phage display. Mol Immunol 40(1):25–33 15. Kehoe J, Kay B (2005) Filamentous phage display in the new millennium. Chem Rev 105(11):4056–4072 16. McGuire MJ, Li S, Brown KC (2009) Biopanning of phage displayed peptide libraries for the isolation of cell-specific ligands. In: Methods in molecular biology: biosensors and biodetection, vol 504. Humana Press, pp 291–321

Chapter 7 Bacterial Cell Display for Selection of Affibody Molecules Charles Dahlsson Leitao, Stefan Sta˚hl, and John Lo¨fblom Abstract This review describes the principles for generation of affibody molecules using bacterial display on the Gram-negative Escherichia coli and the Gram-positive Staphylococcus carnosus, respectively. Affibody molecules are small and robust alternative scaffold proteins that have been explored for therapeutic, diagnostic, and biotechnological applications. They typically exhibit high-stability, affinity, and specificity with high modularity of functional domains. Due to the small size of the scaffold, affibody molecules are rapidly excreted through renal filtration and can efficiently extravasate from blood and penetrate tissues. Preclinical and clinical studies have demonstrated that affibody molecules are promising and safe complements to antibodies for in vivo diagnostic imaging and therapy. Sorting of affibody libraries displayed on bacteria using fluorescence-activated cell sorting is an effective and straightforward methodology and has been used successfully to generate novel affibody molecules with high affinity for a diverse range of molecular targets. Key words Bacterial display, Staphylococcus carnosus, Escherichia coli, Affibody molecules

1

Introduction Directed evolution by means of combinatorial protein engineering is a well-established and powerful strategy for the discovery of selective affinity proteins against essentially any molecular target. Phage display has been in the forefront of this technology for almost four decades [1], largely owed to its amenability for highthroughput screening and development of monoclonal antibodies using large molecular libraries. In phage display, bacteriophages displaying the protein library, through genetic fusion to coat proteins, undergo several rounds of biopanning to isolate variants capable of binding to a target antigen. Enrichment of binding populations is typically continuously evaluated using ELISA, and promising candidates revealed from sequencing are selected for downstream characterization. An alternative approach to the biopanning methodology of phage display is to rely on fluorescently labeled targets and the multivalent expression of library variants on the surface of cells

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2681, https://doi.org/10.1007/978-1-0716-3279-6_7, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

99

100

Charles Dahlsson Leitao et al.

(e.g., yeast or bacteria) to enable real-time sorting of cell populations using fluorescence-activated cell sorting (FACS) [2]. The protein library is anchored to the cell membrane while the encoding DNA is encapsulated within the cell, thus creating a physical link between phenotype and genotype [3]. The self-renewing property of cells, combined with the possibility to assess binding by quantitative flow cytometry, facilitates the selection and simplifies the downstream analysis of binding candidates. For example, both the equilibrium dissociation constant and the dissociation rate of isolated clones can be determined directly on the cell surface using flow cytometry following FACS, foregoing the need for subcloning procedures to produce soluble proteins during initial characterization [4]. Furthermore, it is possible to introduce specific protease-recognition sequences in the displayed protein construct to enable direct release by protease-mediated cleavage of functional proteins from the cell surface in amounts sufficient to perform various downstream assays [5]. During FACS-based selections, cell populations binding to a fluorescently labeled target can be distinguished and sorted from non-binding cell populations based on fluorescent signals, allowing for enrichment to be monitored and controlled in real-time (Fig. 1a). The multivalent expression on cells allows for quantitative analysis of the relative affinity, and the stringency of the selection is controlled by adjusting the dissociation time during washing, the target antigen concentration, and sort settings in the cell sorter. FACS-based selection is particularly useful for screening for dualfunctionality, such as bispecific target binding or cross-reactivity, by enabling interrogation of multiple fluorescent signals simultaneously [6]. Moreover, by including a reporter tag in the displayed protein construct and adding the corresponding fluorescent reporter molecule, such as an albumin-binding domain (ABD) and fluorescently labeled human serum albumin (HSA), the surface expression can be normalized which allows for effective discrimination of the relative affinity between variants [7, 8]. The utility of the platform also extends to applications other than affinity protein development, such as protease substrate profiling [9], and epitope mapping [10]. Typically, the diversity is initially too large for oversampling the library with FACS and is often preceded by magnetic activated cell sorting (MACS) to reduce complexity, which makes use of paramagnetic beads immobilized with the target antigen to capture binding cell populations. Examples of hosts used for bacterial display include the Gram-negative Escherichia coli [2] and the Gram-positive Staphylococcus carnosus [11, 12], both of which have been used for the engineering and development of affibody molecules.

Bacterial Cell Display for Selection of Affibody Molecules

A

I. Library transformation and display on cells

IV. Amplification

101

B

V. Sequencing and characterization

24 18

25

17 27 14

28

13 32

III. FACS screening

II. Incubation with fluorescently labeled target and reporter

11 10

35

Antigen binding

9

Helix 1

Helix 2

Helix 3

VDAK FNKEQQNAFYEILH LPNLN EEQRNAFIQSLKD DPSQ SANLLAEAKKLNDA QAPK

Surface expression

---- ----XXX-XX--XX ----- XX-XX---X--X- ---- -------------- ----

Fluorescently labeled antigen Fluorescently labeled HSA

C

S

Z

ABD AIDAc

D

S PP

Z

ABD ABD XM PP

Z Z

Cell wall

Cell wall

Inner membrane

Cell membrane

ABD ABD XM

ABD

Outer membrane

Fig. 1 (a) Schematic overview of bacterial display of affibody molecules. (I) Transformation of a host-specific expression vector, typically by electroporation, resulting in cell surface display of the protein library (red) fused to a reporter region (blue) used for surface normalization, typically an ABD for affibody libraries. (II) Cells are incubated with the target antigen and HSA labeled with different fluorophores. (III) Screening by FACS is used to isolate cell populations that demonstrate high-relative affinity for the target antigen from an applied sorting gate, while cells expressing non-binding protein variants are discarded. (IV) Amplification and enrichment of binding cell populations by cell growth in preparation for subsequent selection rounds. (V) Enrichment of the output from FACS-screening can be evaluated by DNA sequencing and soluble affibody variants can be produced for downstream characterization. (b) Structure of an affibody molecule with the 13 randomized positions on the first and second helix indicated. The amino acid sequence of the protein A-derived Z-domain is shown below the structure with randomized positions and the three helices indicated. (c) The expression vector used for E. coli displays affibody molecules with a schematic illustration of the anchoring and display on the cell membrane mediated by the AIDA-I autotransporter system shown below. (d) The expression vector used for S. carnosus displays affibody molecules utilizing the cell-wall anchoring sequence XM derived from staphylococcal protein A with a schematic illustration of the protein construct displayed on the cell surface shown below 1.1 Affibody Molecules

Affibody molecules are small (58 amino-acids, 6–7 kDa) and robust three-helical bundle proteins derived from staphylococcal protein A and have been developed using directed-evolution based methods to bind with high affinity to a plethora of target molecules [13]. Affibody molecules are typically characterized as having high thermal and chemical stability, and the small size provides benefits for both therapeutic and diagnostic applications, making it an interesting alternative to monoclonal antibodies. Affibody libraries

102

Charles Dahlsson Leitao et al.

are constructed by randomization of 13–15 surface-exposed residues on the first and second helix, generally excluding prolines (for its helix breaking propensity) and cysteines from the library design (Fig. 1b). The affibody scaffold lacks cysteine residues and its exclusion allows unique cysteines to be introduced post-selection as a chemical handle to enable site-specific functionalization of the affibody using thiol-coupling chemistry. Affibody libraries are typically synthesized using trinucleotide codons rather than degenerate NNK codons to minimize bias and introduction of stop codons [14]. Affibody molecules can be recombinantly produced in prokaryotic hosts as monomers or more complex constructs thus keeping the cost and complexity of manufacturing low. Furthermore, alternative administration routes, such as oral, subcutaneous, pulmonary, and ocular delivery, are being investigated [13]. Furthermore, peptide synthesis can be used as alternative to recombinant expression, thus offering a straightforward way for GMP production and to incorporate, for example, non-natural amino acids [13]. Modular extension by genetic fusion of additional functional domains typically has low impact on stability and affinity of affibody molecules, which enables innovative protein engineering and drug design. Orientation of domains in multispecific biologics has sometimes an impact on biological activity and function, as well as biodistribution of affibody molecules [15, 16], which necessitates comparative evaluation to determine the optimal design for a particular medical application. A fusion strategy commonly used for affibody molecules is the addition of an albumin-binding domain which confers the serum half-life and biodistribution of albumin for applications where prolonged residence time in blood is favorable [8]. Other examples include introducing bi- or multivalency [15] and multispecificity [17] by fusion of identical or different targeting domains, respectively. The small size is an important feature of affibody molecules, providing favorable tumor-targeting properties, such as efficient extravasation and tissue penetration [18], and perhaps the most advantageous aspect being rapid clearance from blood by renal excretion which improves the pharmacokinetic profile for diagnostic tumor imaging [19]. Furthermore, the small size and typically high solubility allows drug formulations with very high-drug concentrations per volume, which has proven advantageous in clinical settings (https://clinicaltrials.gov/ct2/show/NCT02690142), as it enables subcutaneous administration of the affibody-based drug, while monoclonal antibodies normally require intravenous injections performed at infusion center. Affibody molecules have been extensively studied in the context of radionuclide molecular imaging with a strong focus on developing and optimizing imaging probes for visualization of tumors expressing the HER-family of receptors [20]. The large size discrepancy compared to full-length antibodies, and antibody

Bacterial Cell Display for Selection of Affibody Molecules

103

fragments drastically improve upon the current standard of care, offering higher contrast and possibility for imaging at earlier time points. As an example, the HER3-targeting affibody molecule ZHER308698 was labeled with the short-lived radionuclide gallium-68 (half-life of 68 min) and was compared head-to-head with the HER3-targeting antibody Seribantumab and its corresponding F0 (ab)2 fragment labeled with the long-lived radionuclide zirconium-89 (half-life of 78.4 h) [21]. The affibody molecule exhibited overall superior imaging properties at 3 h postinjection compared to the antibody and the antibody fragment at 72 and 48 h, respectively. Thus, affibody molecules could enable same-day imaging for patients, which is currently not possible using antibodies, with benefits such as improved patient compliancy and possibility for more frequent and accurate monitoring of treatment responses. Another reported comparison between affibody molecules and antibodies involves visualization of EGFR-expression to guide resection of tumors in patients with recurrent glioma [22]. The authors concluded that the small size of the affibody molecule provided a more defined delineation of the tumor periphery and was superior for guiding resection, despite a 30-fold lower affinity compared to the antibody. Currently, one diagnostic affibody molecule called ABY-025 is being evaluated in phase II/III clinical trials as a gallium-68 labeled imaging probe for the stratification of patients with HER2-positive primary and metastatic cancers [23]. In terms of therapeutic development, the dimeric affibody ABY-035, binding and neutralizing IL-17A, is being evaluated in phase II/III for the treatment of psoriasis and other autoimmune conditions (www.clinicaltrials.gov/ct2/show/ NCT02690142), and has thus far demonstrated rapid and sustained efficacy without any adverse side-effects. 1.2 Display of Affibody Molecules on E. coli

The extensively studied Gram-negative bacterium E. coli is considered an attractive host for bacterial display of peptide and protein libraries due to properties such as high-expression levels, rapid growth rates, general ease of handling, and a wealth of knowledge concerning subcloning, recombinant expression, and secretion systems [24]. Additionally, high-transformation efficiency accommodates library sizes comparable to what is attainable for phage display [3]. However, the development of efficient display systems has been challenging for larger recombinant proteins due to problems traversing the two biological membranes of Gram-negative bacteria. For this reason, a technology called anchored periplasmic expression (APEx) was developed in which antibodies and antibody fragments are displayed on the inner membrane and exposed to the solution by permeabilization or complete disruption of the outer membrane following correct folding in the periplasmic space [25]. Disruption of the cell membrane affects the viability during FACS-sorting and prevents cell renewability, thus requiring rescue

104

Charles Dahlsson Leitao et al.

of the isolated clones by PCR with subsequent subcloning and transformation to new E. coli cells for each selection round, resulting in a quite laborious process. For displaying proteins on the outer membrane on E. coli, a promising approach is to use the autotransporter secretion pathway, also referred to as autodisplay [26]. Several different types of autotransporters have been used for display of recombinant proteins, and a handful have also been explored for library applications [27, 28]. For engineering of affibody molecules, a method based on the autotransporter Adhesin Involved in Diffuse Adherence (AIDA-I) has been used to efficiently express, secrete, and anchor affibody libraries on the surface of E. coli [29, 30]. AIDA-I employs a type Va secretion system [28, 29] and it consists of three parts, a signal peptide, a β-barrel domain, and a passenger sequence, which comprises the recombinant affibody molecules. The signal peptide conveys the translocation of unfolded proteins across the inner membrane. The β-barrel domain, called AIDAc, is folded into the outer membrane with assistance of chaperones in the periplasm. The passenger sequence is then shuttled through AIDAc and becomes anchored to the outer membrane (Fig. 1c). Optimization of the promoter and cultivation conditions for AIDA-I-based expression vectors revealed that the arabinose-controlled araBAD promoter produced the best results in terms of cell viability and controlled cell surface expression [31]. It was shown from a mock-selection that a 20,000-fold enrichment could be achieved in one round of FACS from a library of binders spiked 1:100,000 in a non-binding background, demonstrating the efficiency of the system. A large naı¨ve combinatorial affibody library was constructed and displayed on E. coli, using a modified version of the AIDA-I expression vector with a shorter natural linker and a single ABD with femtomolar affinity for HSA to reduce the overall size of the passenger sequence [27]. FACS-based selections of binders against HER2, HER3, IL3RA, CD69, and DGCR2 were performed, and binding clones could be isolated and verified as soluble proteins using flow cytometry and biosensor assays against all targets except IL3RA with affinities down in the nanomolar range. 1.3 Display of Affibody Molecules on S. carnosus

FACS-based screening of displayed protein libraries has also been evaluated for Gram-positive bacteria with Staphylococcal carnosus being the most established, presenting distinct advantages but also challenges compared to E. coli [11, 32]. Most notably, translocation of the displayed protein library to the bacterial surface is facilitated by the absence of a second membrane. The surface expression is also maintained at high levels due to low-extracellular protease activity. Moreover, the thick peptidoglycan cell wall has been shown to provide high tolerance to mechanical stress and thus enhanced viability when subjected to the harsh conditions of flow cytometry and cell sorting [2]. However, the

Bacterial Cell Display for Selection of Affibody Molecules

105

thicker cell wall negatively affects the transformation efficiency which limits the size of the practical protein library. Improvements were made to the transformation protocol which included a step involving heat-treatment of the cells to knock out host restriction enzymes [33]. This resulted in a 10,000-fold higher transformation frequency which allowed for the construction of high-complexity affibody libraries consisting of more than 109 variants with potential for both affinity maturation and naı¨ve selections. Following the improvements of the transformation efficiency, S. carnosus has been used to successfully perform affinity maturation of affibody molecules resulting in binders with affinity in the low-picomolar range. Examples include a human epidermal growth factor receptor 3 (HER3)-specific ligand-blocking affibody molecule with an affinity of around 20 pM [34, 35], an affibody dimer consisting of two identical genetically fused domains, which binds and sequesters amyloid beta peptide with an affinity of 300 pM for a potential treatment of Alzheimer’s disease [36, 37], and two distinct affibody molecules binding separate epitopes on the vascular endothelial growth factor receptor 2 (VEGFR2) formatted into a bi-paratopic binder with a remarkably slow dissociation rate [38, 39]. In addition to affibody molecules, S. carnosus display has been used to generate high-affinity binders for single-domain antibodies (nanobodies) [40] and scFv antibody fragments [41]. The vector used for S. carnosus display contains a promoter, secretion signal sequence (S), and a propeptide (PP) sequence from a gene expressing a membrane-bound lipase from Staphylococcal hyicus (Fig. 1d). A cloning site downstream of the PP is used for inserting the affibody library. Additionally, two engineered highaffinity ABDs are included downstream of the library in the displayed protein construct and used for normalization of surface expression. Origins of replication from both S. carnosus and E. coli as well as antibiotic genes expressing chloramphenicol acetyl transferase and beta-lactamase are included for stable replication and expression in S. carnosus and intermediary cloning in E. coli. At the C-terminus of the displayed protein construct is a cell-wall anchoring sequence (XM) derived from staphylococcal protein A [11]. The region is charged and traps the C-terminal end in the cytosol following translocation across the cell membrane. Endogenous sortase recognizes and cleaves a conserved motif in the XM-sequence and cross-links it to a peptide in the peptidoglycan cell wall, thus anchoring and displaying the protein library on the cell surface. The S. carnosus display system has also been used for applications other than generating specific affinity proteins. The potential for protease substrate profiling and discovery of new substrate sequences has been evaluated by screening a library of peptide linkers interconnecting an affibody molecule and an anti-idiotypic blocking domain [42]. Displayed affibody molecules connected to

106

Charles Dahlsson Leitao et al.

a linker sequence recognized by the proteases, specifically matrix metalloprotease (MMP)-1 and tobacco etch virus (TEV) protease, were able to bind to a fluorescently labeled soluble reporter protein upon addition of protease and the resulting removal of the blocking domain. From this method, the protease substrate profile for MMP-1 and TEV protease could be determined, and several new peptide sequences were identified to be processed with a catalytic activity up to eight-fold higher than previously reported substrates [9]. Epitope mapping has been successfully performed for both monoclonal and polyclonal antibodies against a panel of different antigens using S. carnosus by displaying a peptide library covering the entire sequence of an antigen in the form of peptide stretches of varying lengths [10, 43–45]. The epitope could be identified by screening the library against fluorescently labeled antibodies and analyzing the regions of the antigen where binding occurred. Brief description of materials and procedures for MACS and FACS of bacteria-displayed affibody libraries are given below. For more detailed protocols and information concerning the plasmids used to express the affibody-library on E. coli and S. carnosus, please see recent protocols in “Sta˚hl S, Lindberg H, Hjelm LC, Lo¨fblom J, Dahlsson Leitao C (2022) Engineering of affibody molecules. In: Silverman GJ, Rader C, Sidhu S (eds) Phage display: a laboratory manual, 2nd edn. Cold Spring Harbor Laboratory Press. Invited Chapter. In press.”

2

Materials

2.1 MACS of Bacteria-Displayed Affibody Libraries

1. E. coli cells harboring the vector pBAD2.2 [27] comprising the affibody library in fusion to AIDA-I autotransporter.

2.1.1 E. coli DisplaySpecific Reagents

3. Tryptose blood agar base (TBAB) plates with 100 μg/mL carbenicillin.

2. Carbenicillin, use 100 μg/mL in all experiments.

4. L-(+) arabinose, use 0.6% in all experiments. 5. LB medium. 2.1.2 S. carnosus Display-Specific Reagents

1. S. carnosus cells harboring the vector pSCZ2 [46] comprising the affibody library. 2. Chloramphenicol, use 10 μg/mL in all experiments. 3. Tryptose blood agar base (TBAB) plates with 10 μg/mL chloramphenicol. 4. TSB+Y medium.

2.1.3

General Reagents

1. Alexa Fluor 647 (HSA-AF647).

conjugated

2. Biotinylated target protein.

human

serum

albumin

Bacterial Cell Display for Selection of Affibody Molecules

107

3. Glycerol, 85%, sterile. 4. Streptavidin R-Phycoerythrin Conjugate (SA-PE), use 2 μg/ mL in all experiments (Invitrogen). 5. Streptavidin-coated paramagnetic beads (e.g., Dynabeads MyOne Streptavidin C1, Thermo Fischer). 6. 1  PBSP buffer sterile. Add 1 g Pluronic F108 NF surfactant (BASF Corp.), 150 mM NaCl, 8 mM Na2HPO4, 2 mM NaH2PO4 to a final volume of 1 L in sterile water. Adjust pH to 7.4. Sterile filter (0.2 μm). 2.1.4

Equipment

1. End-over-end rotamixer for 1.5 mL microcentrifuge tubes, set to 15 rpm. 2. Magnetic rack for 50 mL centrifuge tubes and 1.5 mL microcentrifuge tubes. 3. Spectrophotometer for optical density measurements, 600 nm (E. coli) and 578 nm (S. carnosus), and cuvettes.

2.2 FACS of Bacteria-Displayed Affibody Libraries

1. As in Subheading 2.1.

2.2.1

4. Sterile water.

Reagents

2. Agarose gel, 1%. 3. Biotinylated target protein. 5. DNA mass ladder. 6. dNTPs. 7. Low-fidelity DNA polymerase and buffer (e.g., Dreamtaq polymerase, NEB). 8. Primers for PCR-screening and sequencing.

2.2.2

Equipment

1. Cell sorter. 2. Flow cytometry tubes. 3. End-over-end rotamixer for 1.5 mL microcentrifuge tubes, set to 15 rpm. 4. Spectrophotometer for optical density measurements, 600 nm (E. coli) and 578 nm (S. carnosus), and cuvettes.

3

Methods

3.1 MACS of Bacteria-Displayed Affibody Libraries

1. Inoculate 1–100 the E. coli or S. carnosus library titer to appropriate cell growth liquid media supplemented with antibiotics, yielding a starting OD600 of around 0.1. Grow for 18–20 h at 37  C, 150 rpm shaking. 2. Re-inoculate 1–100 the library size from the overnight culture, yielding a starting OD600 of around 0.1. Incubate the

108

Charles Dahlsson Leitao et al.

culture at 37  C, 150 rpm. For E. coli display, at OD600 0.5, induce the cultures by adding arabinose. Incubate at 25  C, 150 rpm, overnight. 3. Next day, prepare magnetic beads for both the negative and the positive selection. For the negative selection, vortex streptavidin-coated beads and aliquot 100 μL to 1 mL PBSP. For the positive selection, prepare streptavidin-coated beads by coating with biotinylated target according to supplier’s recommendations. Capture the beads and remove the supernatant. Repeat the wash twice in 1 mL PBSP. To the beads for the negative selection, add 20 mL ice-cold PBPS and keep at 4  C until use. To the beads for the positive selection, pre-capture the biotinylated target on the beads, aiming for a coating density of around 70–100% (based on bead-binding capacity). Prepare a number of beads for a bead to cell ratio of around 1: 50 in first rounds and increase the proportion of beads in later rounds based on the flow-cytometric assessment of enrichment in step 11. Incubate at room temperature for 1 h with rotation. Instantly before positive selection, wash beads and dissolve in 20 mL ice-cold PBSP. 4. Measure OD600 of the library and prepare cells covering the library 1–100. Pellet cells by centrifugation at 4000  g, 10 min, 4  C. Wash three times in 50 mL ice-cold PBSP (see Note 1). 5. Perform a negative selection of the library by resuspending the pelleted cells in the 20 mL bead solution. Incubate with endover-end rotation, 30 min, at room temperature. 6. Capture the beads in the negative selection on the magnet at room temperature for 10 min. Collect the supernatant and transfer to a new tube. Repeat the capture of beads and collect the supernatant. Save 10 μL on ice for titration of the library input. Pellet the cells in the supernatant by centrifugation at 4000  g, 10 min, +4  C (see Note 2). 7. Resuspend the pelleted cells with the 20 mL target-coated magnetic beads (to a final concentration of approximately 5  1010 cells/mL) and incubate with end-over-end rotation, 2 h, room temperature. 8. Capture beads with bound affibody-cell complexes on the magnet for 10 min at room temperature. Aspire the supernatant and incubate on ice until titration. Wash beads four times in 10 mL ice-cold PBSP by inverting the tubes to release the beads and incubating on ice for 3 min prior to magnetic capture. Supernatants from the washes are centrifuged at 4000  g, 10 min, +4  C, and saved on ice for titration.

Bacterial Cell Display for Selection of Affibody Molecules

109

9. Resuspend the bead-affibody-cell output in 50 mL of appropriate cell growth liquid media supplemented with antibiotics and grow at 37  C, 150 rpm, overnight. Inoculate an aliquot of the unsorted library to 10 mL of appropriate cell growth liquid media supplemented with antibiotics, to use as control for flow-cytometric analysis of the enrichment. Titrate aliquots from all steps in dilution series (input 104–106; other 106– 109) and spread 100 μL on TBAB plates supplemented with appropriate antibiotics. 10. Next day, inoculate overnight cultivated cells to new culture with 5 mL appropriate cell growth liquid media supplemented with antibiotics, yielding a starting OD600 of 0.1, and proceed with step 2. Cultures will be used for both new rounds of MACS and for flow-cytometric analysis of the enrichment. As controls, prepare extra cultures of unsorted cells, and for E. coli prepare cells from the output without induction. Prepare glycerol stocks of the library output by dissolving cells in sterile glycerol to a final concentration of 15% stored at 80  C. Calculate the enrichment based on the colony number on the titration plates. 11. Analyze enrichment by flow cytometry. Transfer 10 μL of each cultivation to 1 mL PBSP and centrifuge at 2000  g, 4 min, 4  C. Wash twice in 1 mL PBSP. Discard the supernatant and resuspend cells in 50 μL PBSP containing 100 nM biotinylated target. Mix thoroughly and incubate for 1 h with end-over-end rotation at room temperature. 12. Wash cells with ice-cold PBSP and resuspend in 100 μL PBSP containing 40 nM HSA-AF647 and 2 μg/mL SA-PE. Incubate on ice for 30 min in the dark. Wash twice and resuspend cells in 200 μL ice-cold PBSP, transfer to flow cytometry tubes, and keep on ice in the dark until analysis. Analyze cell surface expression (monitored through HSA-AF647) and target binding (monitored through SA-PE) of >5  104 events in a flow cytometer using appropriate excitation lasers and emission filters. 13. Based on the titrations in step 10, perform additional rounds of MACS by repeating steps 1–9. When the output titer is below around 4  106, proceed using FACS for selection in Subheading 3.2 (see Note 3). 3.2 FACS of Bacteria-Displayed Affibody Libraries

1. Inoculate 10–100 of the library output from the MACS selection to appropriate cell growth liquid media supplemented with antibiotics, yielding a starting OD600 of 0.1. Proceed with steps 1 and 2 in Subheading 3.1.

110

Charles Dahlsson Leitao et al.

2. Aspire 10–100 the library output from the cultivation and wash as in step 11 in Subheading 3.1. Label cells in 500 μL PBSP containing 100 nM biotinylated target. Mix thoroughly and incubate with end-over-end rotation, 1 h, at room temperature. Proceed with washes and secondary incubations according to steps 11 and 12 in Subheading 3.1 in increased volumes of 500 μL (see Note 4). 3. Resuspend the labeled and pelleted cells in 600 μL ice-cold PBSP and transfer to a flow cytometry tube. Keep on ice in the dark until flow cytometric analysis. 4. Simultaneously, measure cell surface expression (monitored through HSA-AF647) and target binding (monitored through SA-PE) in a cell sorter using appropriate excitation lasers and emission filters. Gate cells showing surface expression and target binding, and screen cells corresponding to approximately 10 the library size. Sort cells into 1 mL appropriate cell growth liquid media. 5. Incubate the sorted cells with end-over-end rotation, 37  C, 1 h. Inoculate to appropriate cell growth liquid media supplemented with antibiotics and incubate at 37  C, 150 rpm, overnight. 6. Perform additional rounds of sorting by repeating steps 1–5. Prepare glycerol stocks of the library output by dissolving cells in sterile glycerol to a final concentration of 15%. Store at 80  C. 7. After the last round of FACS, plate the sorted cells onto TBAB plates supplemented with appropriate antibiotics and incubate at 37  C overnight. 8. Analyze library fragment length distribution and sequence enrichment by colony PCR screening and DNA sequencing using a low-fidelity polymerase and in 35 cycles to amplify the region of interest.

4

Notes 1. OD600 of the overnight culture should be >4. 2. Gently end-over-end rotate the magnet holder twice during the 10 min magnetic captures. 3. Typically, two or three rounds of MACS are used as pre-enrichment steps prior to FACS selections. 4. Target concentration, volume, as well as time and temperature for the selection will vary with library size and desired stringency.

Bacterial Cell Display for Selection of Affibody Molecules

111

Acknowledgments This work was supported by the Wallenberg Center for Protein Research (KWA 2019.0341) the VINNOVA grants 2019-0014 and 2017/02105 (CellNova), the Swedish Cancer Society (20 1090 PjF), and the Swedish Research Council (2019-05115). References 1. Smith GP (1985) Filamentous fusion phage: novel expression vectors that display cloned antigens on the virion surface. Science 228: 1315–1317 2. Lo¨fblom J (2011) Bacterial display in combinatorial protein engineering. Biotechnol J 6: 1115–1129 3. Sta˚hl S, Kronqvist N, Jonsson A et al (2013) Affinity proteins and their generation. J Chem Technol Biotechnol 88:25–38 4. Lo¨fblom J, Sandberg J, Werne´rus H et al (2007) Evaluation of staphylococcal cell surface display and flow cytometry for postselectional characterization of affinity proteins in combinatorial protein engineering applications. Appl Environ Microbiol 73:6714–6721 5. Kronqvist N, Lo¨fblom J, Severa D et al (2008) Simplified characterization through sitespecific protease-mediated release of affinity proteins selected by staphylococcal display. FEMS Microbiol Lett 278:128–136 ˚ strand M, Georgieva-Kotseva M 6. Nilvebrant J, A et al (2014) Engineering of bispecific affinity proteins with high affinity for ERBB2 and adaptable binding to albumin. PLoS One 9: e103094 7. Lo¨fblom J, Werne´rus H, Sta˚hl S (2005) Fine affinity discrimination by normalized fluorescence activated cell sorting in staphylococcal surface display. FEMS Microbiol Lett 248: 189–198 8. Jonsson A, Dogan J, Herne N et al (2008) Engineering of a femtomolar affinity binding protein to human serum albumin. Protein Eng Des Sel 21:515–527 9. Sandersjo¨o¨ L, Jonsson A, Lo¨fblom J (2017) Protease substrate profiling using bacterial display of self-blocking affinity proteins and flowcytometric sorting. Biotechnol J 12:1600365 10. Rockberg J, Lo¨fblom J, Hjelm B et al (2008) Epitope mapping of antibodies using bacterial surface display. Nat Methods 5:1039–1045 11. Lo¨fblom J, Rosenstein R, Nguyen MT et al (2017) Staphylococcus carnosus: from starter culture to protein engineering platform. Appl Microbiol Biotechnol 101:8293–8307

12. 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 13. Sta˚hl S, Gra¨slund T, Eriksson Karlstro¨m A et al (2017) Affibody molecules in biotechnological and medical applications. Trends Biotechnol 35:691–712 14. Arunachalam TS, Wichert C, Appel B et al (2012) Mixed oligonucleotides for random mutagenesis: best way of making them. Org Biomol Chem 10:4641–4650 15. Altai M, Leitao C, Rinne S et al (2018) Influence of molecular design on the targeting properties of ABD-fused mono- and bi-valent anti-HER3 affibody therapeutic constructs. Cell 7:164 16. Dahlsson Leitao C, Rinne S, Altai M et al (2020) Evaluating the therapeutic efficacy of mono- and bivalent affibody-based fusion proteins targeting HER3 in a pancreatic cancer xenograft model. Pharmaceutics 12:551 17. Malm M, Bass T, Gudmundsdotter L et al (2014) Engineering of a bispecific affibody molecule towards HER2 and HER3 by addition of an albumin-binding domain allows for affinity purification and in vivo half-life extension. Biotechnol J 9:1215–1222 18. Schmidt MM, Wittrup KD (2009) A modeling analysis of the effects of molecular size and binding affinity on tumor targeting. Mol Cancer Ther 8:2861–2871 19. Tolmachev V, Orlova A (2020) Affibody molecules as targeting vectors for PET imaging. Cancers (Basel) 12:651 20. Rinne SS, Orlova A, Tolmachev V (2021) PET and SPECT imaging of the EGFR family (RTK class I) in oncology. Int J Mol Sci 22:3663 21. Rinne SS, Leitao CD, Abouzayed A et al (2021) HER3 PET imaging: 68Ga-labeled affibody molecules provide superior HER3 contrast to 89Zr-labeled antibody and antibodyfragment-based tracers. Cancers (Basel) 13: 4791

112

Charles Dahlsson Leitao et al.

22. Sexton K, Tichauer K, Samkoe KS et al (2013) Fluorescent affibody peptide penetration in glioma margin is superior to full antibody. PLoS One 8:e60390 23. Velikyan I, Schweigho¨fer P, Feldwisch J et al (2019) Diagnostic HER2-binding radiopharmaceutical, [68Ga]Ga-ABY-025, for routine clinical use in breast cancer patients. Am J Nucl Med Mol Imaging 9:12–23 24. Daugherty PS (2007) Protein engineering with bacterial display. Curr Opin Struct Biol 17: 474–480 25. Harvey BR, Shanafelt AB, Baburina I et al (2006) Engineering of recombinant antibody fragments to methamphetamine by anchored periplasmic expression. J Immunol Methods 308:43–52 26. Jose J, Meyer TF (2007) The autodisplay story, from discovery to biotechnical and. Microbiol Mol Biol Rev 71:600–619 27. Andersson KG, Persson J, Sta˚hl S et al (2019) Autotransporter-mediated display of a naı¨ve affibody library on the outer membrane of Escherichia coli. Biotechnol J 14:1–8 28. Binder U, Matschiner G, Theobald I et al (2010) High-throughput sorting of an Anticalin library via EspP-mediated functional display on the Escherichia coli cell surface. J Mol Biol 400:783–802 29. Leo JC, Grin I, Linke D (2012) Type V secretion: mechanism(s) of autotransport through the bacterial outer membrane. Philos Trans R Soc Lond B Biol Sci 367:1088–1101 30. Fan E, Chauhan N, Udatha DBRKG et al (2016) Type V secretion systems in bacteria. Microbiol Spectr 4:1–24 31. Fleetwood F, Andersson KG, Sta˚hl S et al (2014) An engineered autotransporter-based surface expression vector enables efficient display of affibody molecules on OmpT-negative E. coli as well as protease-mediated secretion in OmpT-positive strains. Microb Cell Fact 13: 179 32. Samuelson P, Hansson M, Ahlborg N et al (1995) Cell surface display of recombinant proteins on Staphylococcus carnosus. J Bacteriol 177:1470–1476 33. Lo¨fblom J, Kronqvist N, Uhle´n M et al (2007) Optimization of electroporation-mediated transformation: Staphylococcus carnosus as model organism. J Appl Microbiol 102:736– 747 34. 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:e62791 35. Kronqvist N, Malm M, Go¨string L et al (2011) Combining phage and staphylococcal surface display for generation of ErbB3-specific affibody molecules. Protein Eng Des Sel 24:385– 396 36. Lindberg H, Johansson A, Ha¨rd T et al (2013) Staphylococcal display for combinatorial protein engineering of a head-to-tail affibody dimer binding the Alzheimer amyloid-β peptide. Biotechnol J 8:139–145 37. Lindberg H, Ha¨rd T, Lo¨fblom J et al (2015) A truncated and dimeric format of an affibody library on bacteria enables FACS-mediated isolation of amyloid-beta aggregation inhibitors with subnanomolar affinity. Biotechnol J 10: 1707–1718 38. Fleetwood F, Klint S, Hanze M et al (2015) Simultaneous targeting of two ligand-binding sites on VEGFR2 using biparatopic affibody molecules results in dramatically improved affinity. Sci Rep 4:7518 39. Fleetwood F, Gu¨ler R, Gordon E et al (2016) Novel affinity binders for neutralization of vascular endothelial growth factor (VEGF) signaling. Cell Mol Life Sci 73:1671–1683 40. 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 41. Hu FJ, Volk AL, Persson H et al (2018) Combination of phage and Gram-positive bacterial display of human antibody repertoires enables isolation of functional high affinity binders. New Biotechnol 45:80–88 42. Sandersjo¨o¨ L, Jonsson A, Lo¨fblom J (2015) A new prodrug form of Affibody molecules (pro-Affibody) is selectively activated by cancer-associated proteases. Cell Mol Life Sci 72:1405–1415 43. Hjelm B, Forsstro¨m B, Lo¨fblom J et al (2012) Parallel immunizations of rabbits using the same antigen yield antibodies with similar, but not identical, epitopes. PLoS One 7:e45817 44. Hjelm B, Fernández CD, Lo¨fblom J et al (2010) Exploring epitopes of antibodies toward the human tryptophanyl-tRNA synthetase. New Biotechnol 27:129–137 45. Rockberg J, Lo¨fblom J, Hjelm B et al (2010) Epitope mapping using gram-positive surface display. Curr Protoc Immunol 90:Unit9.9 46. Werne´rus H, Sta˚hl S (2002) Vector engineering to improve a staphylococcal surface display system. FEMS Microbiol Lett 212:47–54

Chapter 8 Isolation of Antigen-Specific Unconventional Bovine Ultra-Long CDR3H Antibodies Using Cattle Immunization in Combination with Yeast Surface Display Paul Arras, Jasmin Zimmermann, Britta Lipinski, Desislava Yanakieva, Daniel Klewinghaus, Simon Krah, Harald Kolmar, Lukas Pekar, and Stefan Zielonka Abstract Cattle are known for their repertoire of antibodies harboring extremely long CDR3H regions that form extensive “knob on stalk” cysteine-rich structures. The compact knob domain allows for the recognition of epitopes potentially not accessible to classical antibodies. To effectively access the potential of bovinederived antigen-specific ultra-long CDR3 antibodies, a straightforward and effective high-throughput method based on yeast surface display and fluorescence-activated cell sorting is described. Key words Antibody display, Antibody engineering, Bovine ultra-long CDR3H antibodies, Stalkknob, Yeast surface display

1

Introduction Nature offers a plethora of possibilities for the generation of specific immunoglobulin-based antibodies. In addition to classical heavyand light-chain antibodies of most vertebrates, other non-canonical formats, including camelids’ heavy-chain-only antibodies (HcAbs) and cartilaginous fish’s Immunoglobulin New Antigen Receptor (IgNAR), exhibit specific characteristics, which render them attractive for a range of applications [1–3]. Another variety of canonical antibodies is found in cattle and more recently in yaks [4, 5]. In these animals, a subset of approximately 10% of the antibody repertoire contains a heavy chain “ultralong” complementary determining region 3 (UL-CDR3H) with up to 70 amino acids [4, 6]. This extended CDR3H domain consists of a so-called “stalk” region formed by two beta sheets as well as of the “knob” region, which is rich in disulfide bonds and

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2681, https://doi.org/10.1007/978-1-0716-3279-6_8, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

113

114

Paul Arras et al.

protrudes from the main structure [6]. The other CDR regions of the heavy chain and the additional light chain (LC) primarily stabilize the paratope, while antigen contact is largely established via the CDR3H knob [6–8]. In comparison to humans or mice, cattle have a limited germline-based combinatorial repertoire which is compensated by increased post-combinatorial diversification [9]. Generally, the ultra-long CDR3 antibodies are the result of this constrained diversity in regards of VDJ-recombination, since they are based on only one V-segment, IGHV1-7 (VHBUL), and one D-segment, IGHD8-2, further pairing only with a restricted set of LCs encoded by Vλ1× genes [10]. To achieve a lager genetic and, thus, structural diversity enabling this antibody set binding to a multitude of antigen structures, a second diversification step is integrated via extensive somatic hypermutation [7, 11]. A high number of codons in the CDR3H sequence, termed “hot-spots,” can be altered to encode a cysteine by a single point mutation. This in combination with the four cysteines constitutively present in the D-segment enables a high-structural variety via different disulfide bond patterns [7]. The protruding and compact knob structures of the ultra-long CDR3H antibodies potentially allows them to recognize epitopes that are not accessible to the larger and eventual more planar paratopes of conventional antibodies. For example, ultra-long CDR3H antibodies obtained from immunized cattle were screened via fluorescence activated cell sorting (FACS) technology and yielded broad neutralizing antibodies (bnAb) to HIV envelope glycoprotein, which is a difficult target to address with antibodies obtained from more common species used for immunization like rabbit or monkey [12]. In other studies, vaccination and screening led to the discovery of potent ultra-long cow antibodies against complement component C5 [13] and epidermal growth factor receptor (EGFR) [14]. Additionally, it was shown that the knob structure could function autonomously as paratope without need for the antibody scaffold [13] or can be grafted directly onto the hinge region of a human IgG1 creating a low-molecular weight “Knobbody,” while retaining its major functionalities [14]. Finally, it was shown by our group that the UL-CDR3H repertoire can be harnessed to efficiently create ‘almost natural’ common light chain bispecific antibodies [15]. A platform technology that has proven to enable the efficient identification of complex proteins with prescribed properties is referred to as yeast surface display (YSD) [16–18]. This chapter provides straightforward protocols to set up and screen bovine ultra-long CDR3H libraries in a chimeric Fab scaffold using animal immunization in combination with classical YSD.

Isolation of Bovine Ultra-Long CDR3H Antibodies

115

Assuming that the regions outside the stalk-knob region of CDR3H as well as the entire LC are primarily relevant for structural integrity and orientation of the stalk-knob region and not for the antigen recognition per se, we describe the specific amplification of the epitope determining CDR3H segment from the cDNA of immunized cattle via PCR. The generated inserts are subsequently cloned into a synthetic IHGV1-7 scaffold employing yeast homologous recombination in yeast, referred to as gap repair cloning (GR), to create a yeast library harboring a defined antibody scaffold with diverse ultra-long CDR3H regions. Afterwards, this diversified CDR3H HC Fab repertoire is combined with a fixed LC by yeast mating, enabling functional Fab display. During the FACS screening process, full-length Fab expression with proper LC association is monitored via LC-directed detection antibodies simultaneously to antigen binding by staining via antigen-tag directed detection antibodies. Using this two-dimensional labeling strategy employing different fluorophores, an effective enrichment of target-specific ultra-long CDR3H bovine Fab structures can be achieved within several rounds of FACS.

2 2.1

Materials Strains

1. Saccharomyces cerevisiae strain EBY100 MATa AGA1::GAL1AGA1::URA3 ura3-52 trp1 leu2-delta200 his3-delta200 pep4::HIS3 prbd1.6R can1 GAL. 2. Saccharomyces cerevisiae strain BJ5464 MATalpha ura3-52 trp1 leu2-delta1 his3-delta200 pep4::HIS3 prb1-delta1.6R can1 GAL. 3. E. coli strain Top10 F-mcrA Δ(mrr-hsdRMS-mcrBC) Φ80LacZΔM15 Δ LacX74 recA1 araD139 Δ(araleu) 7697 galU galK rpsL (StrR) endA1 nupG.

2.2

Plasmids

The most frequent method to display in Fab format on the yeast surface requires the use of two separate plasmids. Herein, one plasmid (pYD_LEU_VL30) to encode for the fixed LC in the applied system, while another plasmid is serving as destination plasmid (pDest) for the CDR3H diversity, harbors the conserved heavy chain scaffold and a stuffer region, which is replaced by the CDR3H after digestion with specific restriction enzymes. Both plasmids are pYD derivatives containing auxotrophic markers for selection in yeast (tryptophan or leucin, respectively) as well as antibiotic selection markers (ampicillin or kanamycin, respectively) for amplification of the vectors in E. coli. Other essential features include the GAL1 promoter and replication origins for S. cerevisiae (ARS4/CEN6) and E. coli (ColE1) as well as required terminator sequences.

116

Paul Arras et al.

In the destination plasmid, the region encoding for the IHGV1-7 scaffold together with the corresponding IGHJ2-4 fragment and the human CH1 domain are fused in frame to the genes encoding for the Aga2p protein as well as the preceding corresponding Aga2p signal peptide sequence. The described Fab HC “dummy sequence” also contains a stuffer region with BamHI and NotI restriction sites in place of the CDR3H, which is digested and replaced by the amplified ultra-long CDR3H inserts via GR [16, 19–21]. To generate the required homology of the PCR amplicons to the pDest nucleotide sequence for incorporation, the CDR3H amplification is conducted with specific primers using cDNA from immunized cattle B lymphocytes as template. The primers target the IHGV1-7 region and IGHJ2-4 region adjacent to the CDR3H, respectively. Using multiple primers allows to account for genetic variability. The amplicons thus contain a 40–50 bp homology towards the pDest on each terminus of the insert, eventually allowing for a seamless in frame recombination of the CDR3H diversity into the VH scaffold. The appropriate LC plasmid contains a defined bovine VL30 sequence and a corresponding human lambda constant region, both fused in frame to the preceding App8 signal peptide. Figure 1 shows the schematic setup of the essential plasmid features. 2.3 Media and Buffers

1. Lysogeny Broth (LB)—Medium/Agar plates: Rich medium for E. coli liquid culture/plates. Composition for 1 L: 5 g yeast extract, 10 g peptone, 10 g NaCl. Dissolved in H2O. Adjust pH to 7.0. For plates, add 20 g agar and autoclave. After cooling down, add strain-specific antibiotic. 2. Yeast Extract Peptone Dextrose (YPD)—Medium/Agar plates: Full medium for yeast liquid culture/plates. Composition for 1 L: 20.00 g peptone, 10.00 g yeast extract, 20.00 g glucose. For plates, 20 g agar is added. Dissolve in 1 L H2O and autoclave (must for agar plates) or filter sterile. If autoclaved, glucose is autoclaved separately. 10 mL Pen-Strep is added after the solution has cooled down. 3. Synthetic Defined (SD) Medium—Trp/Leu/Trp-Leu/Agar plates: Defined medium for yeast liquid culture. Composition for 1 L: prepared as two solutions. Solution I: 26.70 g minimal SD-Base (Clontech). Dissolve in 890 mL H2O. For plates, 20 g agar is added. Autoclave (must for agar plates) or filter sterile. Solution 2: 1.92 g Drop-out-mix-Trp or 1.6 g DO-mix-Leu (Sigma Aldrich) or 1.54 g DO-mix-Trp/Leu (Sigma Aldrich), 8.56 g NaH2PO4 × H2O, 5.40 g Na2HPO4. Dissolved in 100 mL H2O. Autoclave or filter sterile. Combine both solutions and add 10 mL Pen-Strep (10,000 units/mL).

Isolation of Bovine Ultra-Long CDR3H Antibodies

117

Fig. 1 Schematic overview of the used plasmids and the experimental steps. (a) Schematic depiction if the destination plasmid (pDest) containing the necessary elements for amplification in E. coli (ColE1) and yeast (ARS4/CEN6) as well as the conserved domains of the display construct consisting of the Aga2p signal peptide gene-fused into the heavy chain’s VH (IgVH_Bull) and human CH1, while harboring a stuffer region instead of

118

Paul Arras et al.

4. Low-SD-Trp/Leu/Trp-Leu-Medium: Defined medium for yeast liquid culture with reduced glucose to starve library cells prior to freezing. Composition for 1 L: prepared as two solutions. Solution I: 5 g glucose, 6.7 g/L yeast nitrogen base without amino acids. Dissolve in 890 mL H2O. Solution 2: 1.92 g Drop-out-mix-Trp or 1.6 g DO-mix-Leu (Sigma Aldrich) or 1.54 g DO-mix-Trp/Leu (Sigma Aldrich), 8.56 g NaH2PO4 × H2O, 5.40 g Na2HPO4. Dissolved in 100 mL H2O. Autoclave or filter sterile. Combine both solutions and add 10 mL Pen-Strep (10,000 units/mL). 5. SG-Trp/Leu/Trp-Leu-Medium: Defined medium with galactose/raffinose to induce the GAL1 promotor system for yeast liquid culture. Composition for 1 L: Prepared as two solutions. Solution I: minimal SD-Base + Raf/Gal (Clontech) 37.00 g. Solution 2: 1.92 g Drop-out-mix-Trp or 1.6 g DO-mix-Leu (Sigma Aldrich) or 1.54 g DO-mix-Trp/Leu (Sigma Aldrich), 8.56 g NaH2PO4 × H2O, 5.40 g Na2HPO4. Dissolved in 100 mL H2O. Autoclave or filter sterile. Combine both solutions and add 10 mL Pen-Strep (10,000 units/mL). 6. Library freezing solution: Cryoprotectant for -80 °C storage of S. cerevisiae cells. Composition: 2% (v/v) glycerin, 0.67% (w/v) yeast nitrogen base. Dissolve in H2O and filter sterile. 7. Electroporation buffer: Buffer for electroporation of yeast according to Benatuil et al. [19]. Composition: 1 M sorbitol and 1 mM CaCl2. Dissolve in H2O and filter sterile. 8. Lithium-Acetate (Li/Ac) buffer: Buffer for conditioning the yeast cells prior to transformation. Composition: 100 mM lithium-acetate and 10 mM dithiothreitol (DTT). Dissolve in H2O and filter sterile. Prepare fresh on the day of use.

ä Fig. 1 (continued) the CDR3H region. Surface presentation is mediated by additional C-terminal fusion of a linker sequence and Aga2p gene. To allow for selection in respective species, ampicillin cassette as well as a tryptophan auxotrophic marker was implemented in the destination vector. (b) Schematic illustration of the corresponding light chain plasmid encoding for the fixed VL30 LC, while otherwise harboring similar plasmid features as pDest except using a complimentary antibiotic selection marker. (c) Mating type a (Mat-a) yeast strain is transformed with the linearized destination plasmid and diverse UL-CDR3H sequences possessing PCR incorporated homologous ends facilitating gap repair cloning, thus generating the CDR3H library. The corresponding fixed light chain plasmid is transformed into yeast strain with matching alpha mating type (Matα). Subsequent mating of both yeast strains enables generation of the final Fab based display library in haploid yeast cells. (d) Induction of Fab expression by carbon source change in media enables the surface display of the bovine CDRH3 library which can be examined in two-dimensional FACS analysis using a fluorescence labeled anti-light chain antibody in combination with an appropriate fluorescence labeled antibody against the antigen tag

Isolation of Bovine Ultra-Long CDR3H Antibodies

119

Table 1 Primers used for amplification of UL-CDR3H regions Name

Sequence

VHBULL1_CDR3_GR_up AGCAGCGTGACAACTGAGGACTCGGCCACATACTACTGTAC TACTGTG VHBULL2_CDR3_GR_up AGCAGCGTGACAACTGAGGACTCGGCCACATACTACTGTAC TACTGTGCAC VHBULL4_CDR3_GR_up AGCAGCGTGACAACTGAGGACTCGGCCACATACTACTGTAC TACTGTGCACCAG VHBULL5_CDR3_GR_lo GCCCTTGGTACTAGCTGAGGAGACGGTGACCAGGAGTCC TTGGCCCCA VHBULL6_CDR3_GR_lo GCCCTTGGTACTAGCTGAGGAGACGGTGACCAGGAGTCC TTGGCCCCAGGCATC

2.4 PCR Amplification of Cow UL-CDR3H Regions

1. Q5 High-Fidelity 2× Master Mix (New England Biolabs). 2. Nuclease-free water. 3. Primers for UL-CDR3H regions with pDest homology for gap repair cloning, see Table 1. 4. PCR clean-up kit.

2.5 Digestion of the Destination Plasmid

1. BamHI and NotI restriction enzymes, 20,000 units/mL. 2. 10× CutSmart Buffer. 3. Nuclease-free water. 4. DNA Clean-Up Kit. 5. Destination plasmid, as schematically depicted in Fig. 1a.

2.6 Transformation of S. cerevisiae BJ5464 with LC Plasmid

1. Frozen-EZ Yeast Transformation II Kit (Zymo Research).

2.7 Library Transformation of S. cerevisiae EBY100

1. Electroporation buffer.

2. LC Plasmid (pYD_LEU_VL30), as schematically depicted in Fig. 1b.

2. Li/Ac buffer. 3. 1 M sorbitol in H2O (Sterile). 4. Digested pDest (Fig. 1c). 5. Cow UL-CDR3H PCR amplified inserts (Fig. 1c).

2.8 Labeling and Selection of Yeast Cells with Fluorescence—Activated Cell Sorting (FACS)

1. Phosphate buffered saline (PBS). 2. Anti-penta-his Alexa Fluor 647 conjugate antibody. 3. Anti-λ-LC–phycoerythrin (PE)–conjugate antibody. 4. Target protein his-tagged.

120

Paul Arras et al.

2.9 Retransformation of Enriched Yeast Library to E. coli

1. MasterPure Yeast DNA Purification Kit (Lucigen).

2.10 General Equipment

1. Cryogenic vials.

2. One Shot™ TOP10 electro competent E. coli (Invitrogen). 3. Glycerol.

2. Sterile filtration units (0.22 μm), e.g., SteriTop. 3. Baffled shaking flasks (0.15–3 L volume). 4. Electroporator. 5. 0.2 cm electroporation cuvettes. 6. Shaking incubator (20, 30, and 37 °C). 7. Flow Cytometer and cell sorter (e.g., BD FACS Aria™ Fusion). 8. Thermocycler. 9. Agarose gel electrophoresis setup. 10. Benchtop centrifuge. 11. Cell density meter. 12. Petri dishes. 13. Spectrometer for DNA quantification.

3

Methods Starting material for the herein described protocol is cDNA reverse-transcribed from total RNA obtained from the PBMC repertoire of cattle after immunization with an antigen of choice.

3.1

General

3.2 Amplification of Bovine Ultra Long CDR3H Domains

All yeast centrifugation steps are performed at 2000 × g for 3 min. It is assumed for the calculation that an OD600 value of 1 corresponds to 107 yeast cells/mL. 1. On ice, prepare a 50 μL PCR reaction in a 0.2 mL PCR vial. Add 1 μL of the prepared cow cDNA (approx. 100 ng/μL) as template, add forward and backward primer (see Table 1) to a final concentration of 500 nM, fill up to 25 μL with nucleasefree water, and add 25 μL of 2× Q5 high-fidelity Master Mix (see Notes 1 and 2). 2. Using the aforementioned primers from Table 1, the PCR is conducted with the following parameters: Initial denaturation for 3 min at 98 °C, 35 successive cycles of denaturation at 98 °C for 10 s, annealing at 72 °C 30 s and elongation at

Isolation of Bovine Ultra-Long CDR3H Antibodies

121

72 °C for 20 s followed by a final extension step at 72 °C for 2 min. 3. The success of the PCR reaction is analyzed via 1.5% (w/v) agarose gel electrophoresis. A distinct band should be visible between 300 and 500 bp (see Note 3). 4. Purify the DNA with Wizard® SV Gel and PCR Clean-Up System or similar system according to the manufacturer’s instruction and determine the DNA concentration spectrometrically at 260 nm absorption using a NanoDrop or similar device. The purified PCR products can be stored at 4 to 20 °C. 3.3 Transformation of BJ5464 Yeast with the Fixed LC Plasmid

BJ5464 (mating type α) are transformed with the pYD_LEU_VL30 LC plasmid. The transformed cells are mated with the EBY100 strain after CDR3H library construction, to generate diploid yeast cells harboring the genetic information for the desired Fab fragments. The transformation can be performed with any convenient method, e.g., chemically with the Frozen-EZ Yeast Transformation II Kit (Zymo Research) according to the manufacturer’s manual. Store the resulting transformants at -80 °C as described in Subheading 3.6.

3.4 Digestion of the Destination Plasmid

1. In 6× 1.5 mL reaction tubes, combine each 14 μg of the destination plasmid pDest_VHBULL_CDR3, 70 μL of 10× CutSmart reaction buffer, 140 Units of NotI-HF, and 140 Units BamHI-HF with water to a final volume of 700 μL. 2. Incubate the digestion mix at 37 °C for 1 h. 3. Analyze the success of the restriction via 0.5–1% (w/v) agarose gel electrophoresis. A distinct band should be visible at approximately 6500 bp. 4. Purify the DNA with a PCR Clean-Up kit or a similar system according to the manufacturer’s instruction and determine the DNA concentration 260 nm absorption using a NanoDrop or similar device. The purified PCR products can be stored at +4 to -20 °C.

3.5 Transformation of Yeast for CDR3H Library Generation

The yeast transformation to generate the yeast CDR3H library is based on a modified transformation protocol for S. cerevisiae by Benatuil et al. [19] 1. Grow a culture of EBY100 (mating type a) cells overnight to saturation in a baffled flask at 30 °C and 120 rpm in YPD medium. 2. From the overnight culture, inoculate 500 mL of fresh YPD medium to an OD600 of 0.3. The cells are incubated at 30 °C

122

Paul Arras et al.

and 120 rpm until the culture reaches an OD600 of ~1.6 (3–4 h). 3. Pellet cells by centrifugation. 4. Wash pellet twice by (resuspending and subsequent pelleting by centrifugation) with 250 mL ice-cold H2O and once with electroporation buffer (1 M sorbitol +1 mM CaCl2). 5. Pellet cells by centrifugation. 6. Resuspend cells in 20 mL Li/Ac buffer. 7. Transfer the suspension to a baffled culture flask and incubate at 30 °C and 120 rpm for 30 min. 8. Collect the cells by centrifugation and wash once with 250 mL electroporation buffer. 9. Resuspend the cells in 5 mL electroporation buffer. 10. 40 μg of the digested destination plasmid is added to the cells and carefully mixed. 11. 400 μL of the suspension is separated for backbone re-ligation control. 12. 120 μg of the PCR amplicon is added to the remaining solution. 13. 400 μL of the cell suspension is added to a 2 mm electroporation cuvette and electroporated at 2.5 kV, 200 Ω, and 25 μF. After electroporation reaction, the cells are immediately taken up in 1 mL 1:1 v/v YPD:sorbitol (1 M) and transferred to a total 8 mL 1:1 v/v YPD:sorbitol per reaction. 14. The previous step is repeated for the whole cell solution and the backbone control. 15. Incubate the transformed cells in YPD:sorbitol for 1 h at 30 °C at 120 rpm. 16. Collect the cells by centrifugation and resuspended in SD (-Trp) to a final volume of 10 mL (Backbone control accordingly to 1 mL). 17. Use 100 μL (1% of the volume) of the solution for a serial dilution and plate the equivalent of the 107th–109th part SD-Trp plates to determine the library titer. Incubate the plates for 2 days at 30 °C before counting the colonies and calculating the library size. 18. Transfer the remaining library to 1 L SD(-Trp) and incubate at 30 °C and 120 rpm. 19. Incubate the library for 1–2 days. 20. Transfer the library to 1 L SD-Low-Trp at an initial OD600 of 1. 21. Incubate for 1 day at 30 °C and 120 rpm.

Isolation of Bovine Ultra-Long CDR3H Antibodies

3.6 Cryopreservation for Long-Term Storage of Yeast

123

Prior to mating, the library is frozen and stored at -80 °C. 1. Collect cells from a fresh SD- (+ corresponding dropout mix) culture by centrifugation and discard the supernatant completely. 2. Wash twice with ice-cold H2O. 3. Pellet the cells and resuspend in library freezing solution at approximately 1010 cells/mL. 4. Create appropriate aliquots of cells in cryogenic vials and store at -80 °C.

3.7 Mating of the CDR3H Yeast Library with the Fixed LC Clone

1. Start a SD-Trp culture from the deep-frozen library EBY100 cells. Inoculate at an OD600 = 1 and incubate at 30 °C and 120 rpm overnight. Of note, we recommend an oversampling of the library size of at least tenfold. 2. Prepare a SD-Leu culture from the frozen BJ5464 fixed LC transformants. Inoculate at an OD600 = 1 and incubate at 30 ° C and 120 rpm overnight. 3. The next day, measure the cell density of both cultures and combine cells in tenfold excess (or at least 109 cells) in regard to the EBY100 library size from each culture in a 50 mL falcon tube. 4. Collect the cells by centrifugation. 5. Resuspend the cells in 1 mL YPD medium. 6. Prepare 5 YPD-Agar plates and pipet two spots of 100 μL of the yeast solution on each of the plates (do not spread out!). 7. Let the solution dry on the plate prior to incubation for 24 h at 30 °C. 8. The next day, wash the cell layer on each pate down using 2.5–5 mL PBS. Pool cell suspension in a 50 mL falcon tube. 9. Collect the cells via centrifugation, discard the supernatant. 10. Resuspend in 10 mL SD-Trp-Leu. 11. Use 100 μL (1% of the volume) of the solution for a serial dilution plating on SD-Trp-Leu plates to determine the final library titer. Incubate the plates for 2 days at 30 °C before counting the colonies and calculating the final library size. 12. Transfer remaining cells into 1 L SD-Trp-Leu and grow overnight. 13. Transfer the library to 1 L SD-Low-Trp at an initial OD600 of 1. 14. Incubate for 1 day at 30 °C and 120 rpm. The library is now ready for FACS-based screening or longterm cryopreservation (as described in Subheading 3.6).

124

Paul Arras et al.

3.8 Induction of Expression for Antibody Yeast Surface Display

1. Thaw a cryopreserved vial of the library at room temperature and resuspend the cells in an appropriate volume of SD-Trp medium roughly corresponding to 1–2 × 107 cells/mL (OD600 = ~1–2). To ensure the coverage of the diversity, the number of transferred cells for the initial culture should exceed the estimated library size at least tenfold. 2. Incubate the cells overnight at 30 °C and 120 rpm. 3. Measure OD600 and harvest cells (at least library diversity ×10) via centrifugation. 4. Resuspend cells in SG-Trp induction medium with an initial cell density of OD600 = 1. 5. Incubate the cells at 20 °C and 120 rpm for 40–96 h to allow for the expression of the antibody and display on the yeast surface.

3.9 FluorescenceActivated Cell Sorting for Detection and Selection of Antigen Binding Bovine ULCDR3H-Based Fab Fragments

This section will give an overview of possible labeling strategies in preparation for the FACS screening of the generated bovine UL-CDR3H library (Fig. 2). The two-dimensional strategy employed herein uses one dimension to control the full-length Fab display on the cells, which is achieved via a fluorescently labeled [e.g., Phycoerythrin (PE)] antibody against the human constant part of the LC of the displayed Fab. For the detection of the binding to the antigen, an indirect staining via an anti-penta-his antibody (labeled with, e.g., AF-647) targeting the antigen’s his-tag is utilized, establishing the second dimension (see Notes 4 and 5). For all cytometric measurements and sorts, a set of staining controls are necessary to ensure the functionality of the labeling strategy and facilitate the setting of sorting gates. Relevant controls are as follows: 1. Untreated cells for the general baseline. 2. A display and fluorophore binding control, which is only treated with the labeling antibodies but without the antigen. This allows the proper adjustment of the sorting gates and controls for possible binding of the antigen-detection antibody to the cell by unspecific interactions as well as unwanted interaction due to cross-reactivity of the utilized detection antibodies. 3. Negative control for which an unrelated (his-tagged) protein is used to evaluate the level of unspecific binding by the displayed antibodies and to allow for the adjustment of the gates accordingly. 4. Optional controls can include positive controls with an actively displaying single clone against the target antigen (if available), to control the functionality of the used reagents, FACS device, and experimental setup (e.g., lasers, scatters).

Isolation of Bovine Ultra-Long CDR3H Antibodies

125

Fig. 2 FACS analysis of the bovine derived ultra-long CDR3H diversified YSD library. Yeast cells were stained for proper surface expression of the Fab fragments with PE anti lambda LC in combination with AF647-antiHis-Tag (first sorting round)/APC-anti-His (second sorting round) antibody staining for simultaneously target antigen binding detection. Initial library displayed on yeast cells (left panel) and enriched library prior sorting round 2 (right panel), incubated with incubated with 1 μM target antigen (upper row) or DPBS for gate adjustment (lower row). Cell populations with corresponding values as well as the applied sorting gate are indicated. R4.0.3 with flowCore2.2.0 and flowViz1.54.0. packages was used for plot illustration

126

Paul Arras et al.

3.9.1 Staining of the Library and Controls for FACS Analysis

The number of cells required for a sufficient control varies depending on the library size and the employed cytometer. Here, 2 × 107 cells are used to prepare the control measurements for antigen specificity, proper Fab surface display, and exclusion of unwanted cross-reactivity of the used detection antibodies, ultimately allowing for ideal device settings and sorting gate adjustment. The number of cells required for the FACS sort varies depending on the calculated library size but can be prepared accordingly by scaling up the following steps. To theoretically cover the whole library size, we recommend oversampling the library titer at least 10 times for sorting purposes. 1. Pellet 2 × 107 yeast cells by centrifugation and decant supernatant. 2. Wash the cells twice with PBS. 3. Resuspend cells in. (a) 40 μL of 1 μM his-tagged protein antigen diluted in PBS for sorting purposes (see Note 6) (b) 40 μL of PBS for Fab display control. (c) 40 μL of 1 μM unrelated his-tagged protein in PBS for antigen specificity control). 4. Incubate for 30 min on ice. 5. Wash the cells twice with PBS prior resuspension in 40 μL of detection reagent mix, composed of a 1:20 dilution of antipenta-his AF-647 conjugate antibody as well as anti-LC antibody (PE) (see Note 7). Incubate samples shielded from light for 30 min on ice. 6. Wash the cells two times with PBS and resuspend in 600 μL PBS (see Note 8). Store samples on ice and protected from light until FACS analysis.

3.9.2 Cell Handling Following FACS Analysis

1. Transfer sorted cells in a baffled flask containing appropriate volume of SD-Trp medium (about 20 mL). Incubate for 2 days at 30 °C and 120 rpm. 2. Count cell number and induce Fab expression by media exchange as described in Subheading 3.8. 3. Remaining cells can be cryopreserved as described in Subheading 3.6. 4. If needed, enrich binding population by several consecutive FACS rounds. 5. When the desired ratio of binders is enriched (rule of thumb: >50% of total displaying cells), the DNA can be isolated and sequenced by a method of choice.

Isolation of Bovine Ultra-Long CDR3H Antibodies

127

3.10 Sequencing the Display Vector from Yeast Libraries

To gather information about the genotypes of present (enriched) libraries, the yeast cells are initially lysed, followed by plasmid DNA isolation and subsequent transformation of E. coli (Top10, Invitrogen) via electroporation. By using plates with ampicillin as selection marker, resulting E. coli single clones contain only the variable heavy chain plasmid.

3.11 Yeast Lysis and E. coli Transformation

1. The lysis and extraction of DNA can be performed using any convenient method, e.g., with MasterPure Yeast DNA Purification Kit (Lucigen) according to the manufacturer’s manual. 2. The isolated plasmid DNA is subsequently utilized to transform 50 μL E. coli (Top10, Invitrogen) cells via electroporation reactions (1.8 kV) in ice-cold 1 mm cuvettes according to the manufacturer’s manual using 3–4 μL of yeast plasmid DNA solution. 3. Incubate the cells with the kit’s outgrow medium as described in the manufacturer’s manual. 4. Plate 5 μL of the total cell solution diluted in 50 mL PBS on a 10 cm LB-Agar plate (with suitable selection antibiotic, ampicillin). Prepare two plates per electroporation reaction to ensure a suitable number of colonies. 5. From the E. coli colonies grown on the agar plates, an appropriate number (we recommend at least 96 clones per library) is picked and transferred into a 96-deep-well plate with 600 μL LB medium with ampicillin. The plate is incubated overnight at 37 °C and 700 rpm. 6. The next day, 20% glycerol stocks are prepared from this plate by mixing 60 μL E. coli culture with 40 μL 50% glycerol in 96 well plates. 7. These glycerol stocks can be stored at -80 °C and used for sequencing. The evaluation of the resulting sequences can be performed using standard bioinformatics software.

4

Notes 1. A suitable amount of insert DNA for the construction of a library with 10 electroporation reactions requires about 96 PCR reactions a` 50 μL. 2. To cover the genetic diversity in the primer binding regions, a combination of three forward primers and two matching reverse primers are used. The resulting six primer pairs are used to conduct separate PCR reactions in equal ratios.

128

Paul Arras et al.

3. After successful PCR, the amplification results from all primer pairs are pooled. 4. To avoid a potential enrichment of binders against the detection antibody, it is advisable to use a different fluorophore and clonotype of the detection antibody in consecutive screening rounds. 5. To avoid bleaching of the fluorophores, the reagents as well as the stained cells should be shielded from direct light and the labeling should be performed on ice. 6. Reduced antigen concentrations can be used to increase sort stringency. 7. The dilution of the fluorescence labeled detection reagents can be adjusted but should exceed antigen concentration to avoid limitations in antigen binding detection. 8. The final volume can be adjusted according to the manufacturer’s instructions, depending on the used device. References 1. 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 2. Zielonka S, Empting M, Grzeschik J et al (2015) Structural insights and biomedical potential of IgNAR scaffolds from sharks. MAbs. 7(1):15–25. https://doi.org/10. 4161/19420862.2015.989032. PMID: 25523873; PMCID: PMC4622739 3. Krah S, Schro¨ter C, Zielonka S et al (2016) Single-domain antibodies for biomedical applications. Immunopharmacol Immunotoxicol 38(1):21–28. https://doi.org/10.3109/ 08923973.2015.1102934. Epub 2015 Nov 9. PMID:26551147 4. Saini SS, Allore B, Jacobs RM et al (1999) Exceptionally long CDR3H region with multiple cysteine residues in functional bovine IgM antibodies. Eur J Immunol 29:2420–2426 5. Wu M, Zhao H, Tang X et al (2022) Organization and complexity of the yak (Bos Grunniens) immunoglobulin loci. Front Immunol 13: 2111 6. Wang F, Ekiert DC, Ahmad I et al (2013) Reshaping antibody diversity. Cell 153:1379– 1393 7. Haakenson JK, Huang R, Smider VV (2018) Diversity in the cow ultralong CDR H3 antibody repertoire. Front Immunol 9:1–10 8. Deiss TC, Vadnais M, Wang F et al (2019) Immunogenetic factors driving formation of ultralong VH CDR3 in Bos taurus antibodies. Cell Mol Immunol 16:64–75

9. Zhao Y, Jackson SM, Aitken R (2006) The bovine antibody repertoire. Dev Comp Immunol 30:175–186 10. Saini SS, Farrugia W, Ramsland PA et al (2003) Bovine IgM antibodies with exceptionally long complementarity-determining region 3 of the heavy chain share unique structural properties conferring restricted VH + Vλ pairings. Int Immunol 15:845–853 11. Saini SS, Kaushik A (2002) Extensive CDR3H length heterogeneity exists in bovine foetal VDJ rearrangements. Scand J Immunol 55: 140–148 12. Sok D, Le KM, Vadnais M et al (2017) Rapid elicitation of broadly neutralizing antibodies to HIV by immunization in cows. Nature 548: 108–111 13. Macpherson A, Scott-Tucker A, Spiliotopoulos A et al (2020) Isolation of antigen-specific, disulphide-rich knob domain peptides from bovine antibodies. PLoS Biol 18:1–21 14. Pekar L, Klewinghaus D, Arras P et al (2021) Milking the cow: cattle-derived chimeric ultralong CDR-H3 antibodies and their engineered CDR-H3-only knobbody counterparts targeting epidermal growth factor receptor elicit potent NK cell-mediated cytotoxicity. Front Immunol 12:742418 15. Klewinghaus D, Pekar L, Arras P et al (2022) Grabbing the bull by both horns: bovine ultralong CDR-H3 paratopes enable engineering of ‘Almost Natural’ common light chain bispecific antibodies suitable for effector cell redirection. Front Immunol 12:801368. https://doi.org/

Isolation of Bovine Ultra-Long CDR3H Antibodies 10.3389/fimmu.2021.801368. PMID: 35087526; PMCID: PMC8787767 16. Boder ET, Wittrup KD (2000) Yeast surface display for directed evolution of protein expression, affinity, and stability. Methods Enzymol 328:430–444 17. Valldorf B, Hinz SC, Russo G et al (2022) Antibody display technologies: selecting the cream of the crop. Biol Chem 403:455–477 18. Doerner A, Rhiel L, Zielonka S et al (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. Epub 2013 Nov 26. PMID: 24291259 19. Benatuil L, Perez JM, Belk J et al (2010) An improved yeast transformation method for the

129

generation of very large human antibody libraries. Protein Eng Des Sel 23:155–159 20. Pekar L, Klausz K, Busch M et al (2021) Affinity maturation of B7-H6 translates into enhanced NK cell-mediated tumor cell lysis and improved proinflammatory cytokine release of bispecific immunoligands via NKp30 engagement. J Immunol 206(1):225– 236.https://doi.org/10.4049/jimmunol. 2001004. Epub 2020 Dec 2. PMID: 33268483; PMCID: PMC7750860 21. Roth L, Grzeschik J, Hinz SC et al (2019) Facile generation of antibody heavy and light chain diversities for yeast surface display by Golden Gate Cloning. Biol Chem 400 (3):383–393. https://doi.org/10.1515/hsz2018-0347. PMID: 30465712

Chapter 9 Selection of High-Affinity Heterodimeric Antigen-Binding Fc Fragments from a Large Yeast Display Library Filippo Benedetti, Gerhard Stadlmayr, Katharina Stadlbauer, Florian Ru¨ker, and Gordana Wozniak-Knopp Abstract Antigen-binding Fc (Fcab™) fragments, where a novel antigen binding site is introduced by the mutagenesis of the C-terminal loops of the CH3 domain, function as parts of bispecific IgG-like symmetrical antibodies when they replace their wild-type Fc. Their homodimeric structure typically leads to bivalent antigen binding. In particular, biological situations monovalent engagement, however, would be preferred, either for avoiding agonistic effects leading to safety issues, or the attractive option of combining a single chain (i.e., one half) of an Fcab fragment reactive with different antigens in one antibody. We present the strategies for construction and selection of yeast libraries displaying heterodimeric Fcab fragments and discuss the effects of altered thermostability of the basic Fc scaffold and novel library designs that lead to isolation of highly affine antigen binding clones. Key words Bispecific antibodies, Fcab, mAb2, Heterodimeric antibody, Knobs-into-Holes, Directed evolution, Yeast display

1

Introduction

1.1 Bispecific Antibodies and mAb2 ™ Antibody Molecules

In the past decades, bispecific antibodies have been in the limelight of the antibody-based therapeutics, and with more than 110 candidates in clinical development they are expected to make a huge contribution to human health [1]. Since the clinical approval of the first candidate in 2009 [2], there are currently six approved bispecific antibodies, four intended for treatment of tumor conditions [3–6], one for combating age-related macular degeneration [7], and one for enzyme replacement therapy [8]. Their ability of spatial or temporal simultaneous targeting of two antigens is extremely advantageous in certain biological situations. They exhibit unprecedented functionality in recruiting and activating immune cells to exert their cytotoxic effect, blocking of redundant signaling pathways, blocking of multiple immune checkpoints, which often

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2681, https://doi.org/10.1007/978-1-0716-3279-6_9, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

131

132

Filippo Benedetti et al.

resolved new biological interactions, and inciting formation of protein complexes [9]. In addition, bispecific antigen binding can enhance selectivity and hence improve efficacy and safety of antibody therapy [10]. The success of bispecifics results not only from the thorough study of target combinations, but also from the plethora of available formats which can incorporate the two specificities: these include fragment-based covalently connected binding proteins, heterodimeric IgG-like proteins, and fusion formats with binding entities appended to IgG. Indeed, the architecture of the targeting agent and the positioning of the binding sites for distinct paratopes have often been shown to affect the activity [11]. Bispecific antibodies with included Fc fragment share the advantages of IgGs, such as long half-life in vivo owing to the pH-dependent interaction with the neonatal Fc receptor, and the ability to exert effector functions such as antibody-dependent cellular cytotoxicity, antibody-dependent cellular phagocytosis, or complementmediated cytotoxicity. Symmetrical IgG-like formats are very popular because of their similarity to unmodified IgG, and hence more predictable manufacturing features and pharmacokinetics [12– 14]. Among those, the mAb2 ™ bispecific antibody format presents an attractive alternative, with currently three molecules of this format in clinical testing [15–17]. In this format, antigen binding is mediated by the Fab-arms as well as by the mutated Fc fragment (Fc fragment with antigen binding or Fcab™ fragment) [18]. Typically, the modified C-terminal loops of the CH3 domains form the novel antigen binding site and enable a strong antigen interaction. 1.2 Directed Evolution of Fcab™ Libraries

Fcab fragments are generated by construction of large yeast [18] or phage display libraries [19] and subsequent selection of the clones of interest using directed evolution. Yeast display methods have in the past been successfully used not only for isolation of antigenbinding clones, but also for the improvement of structural properties of the displayed protein, such as thermostability [20]. Characterization of phenotypic properties of library-displayed clones, induced under selection pressure, has contributed to optimization of Fcab library designs [21]. Interestingly, even the interaction mode of the antigen and Fcab fragment could be influenced by the selection protocol, as in the example of construction of pH-dependent antigen binding clones [22]. Importantly, affinity maturation of Fcab clones can be efficiently performed in the yeast system due to the possibility of normalization, i.e., simultaneous staining with antigen and a reporter anti-tag antibody, which enables the differentiation of antigen-binding cells by their display level of the Fcab clones [23]. Moreover, the anti-Fc antibodies that will only recognize correctly folded Fc fragments can be applied as reporter molecules to preferentially isolate clones with favorable biophysical properties [20]. Depending on the desired properties of the Fcab clone, fluorescently labeled ligands of Fc fragment, such as protein A or the effector molecule CD64 (FcγRI), can be used to

Discovery of Heterodimeric Fcab™ Fragments

133

steer the selection towards the binders that will be amenable for up-scalable chromatographic purification, or engagement of the effector cells, respectively [24]. 1.3 Fcab™ Fragment–Antigen Interaction

The interaction of Fcab fragments and their cognate antigens has until now been described with several methods, including affinity determination with surface plasmon resonance, biolayer interferometry, isothermal titration calorimetry [25], but also with crystal structures [26, 27]. In all publicly available structures, Fcab binding is bivalent: each of the Fcab monomers can bind to one antigen molecule, due to the homodimeric nature of the Fc fragment. In the case of VEGF-specific Fcab clones, the dimeric nature of the VEGF antigen itself prompts formation of a larger Fcab fragmentantigen complexes. In some biological situations, however, monovalent antigen engagement may prove advantageous. This is the case for CD3, which is a part of the T-cell receptor (TCR) and is activated upon the TCR contact with peptide-loaded major histocompatibility complex (p-MHC) on antigen-presenting cells in an immunological synapse [28]. Bivalent engagement of CD3 can elicit massive cytokine release, causing acute systemic inflammatory syndrome that can ultimately result in multiple organ failure and death [29], while toxic effects have been much less potent when CD3 was targeted with monovalent binding agents, such as single-chain Fv or Fab fragments, as parts of bispecific therapeutics [30]. This could be a valuable application for the Fcab fragments designed particularly to mediate monovalent binding. Further, while low-affinity antigen interaction might require bivalency to achieve a certain level of biological activity, distinct strong monovalent binders could possibly be combined in an Fcab fragment and hence mediate binding to a further antigen molecule. In the recent years, trispecific antibodies have often proven superior to their bispecific counterparts by demonstrating novel modes of action [31–33]. We were therefore interested to explore the ability of the Fcab yeast display platform for the ability to select heterodimeric Fcab clones, which could permit monovalent antigen interaction. We present the protocols applied for design, construction, and selection of such libraries.

1.4 Design of Heterodimeric Fcab™ Libraries

There are several heterodimerization motifs that have already proven successful for the formation of heterodimeric Fc fragment, including such that induce the preferential pairing of the heterologous CH3 domains due to steric compatibility [34, 35], electrostatic steering [36], or introduction of completely novel residues that form orthologous surfaces [37]. We have chosen the wellcharacterized “Knobs-into-Holes” (KiH) mutagenesis strategy, which involves only two mutations: T366Y and Y407T [34] (EU numbering [38]). This mutation typically decreases the melting temperature of the Fc fragment by about 20 °C [39].

134

Filippo Benedetti et al.

1.5 Heterodimeric Fcab™ Libraries Constructed with Yeast Mating

Yeast mating has been successfully used in the past for creation of heterodimeric Fab libraries. In this system, two yeast strains of opposite mating types, each harboring a library encoding one of the heterodimer chains, form diploids in the mating process and are typically selected in a medium deficient for two medium additives relevant to each of the auxotrophic markers: tryptophan and leucine [40]. However, the KiH-heterodimerized wild-type Fc construct expressed in this system was surprisingly detectable only in 40% of the yeast population, half of what is expected from heterologous protein-displaying haploid yeast display cultures. The expression could hardly be detected under the stress conditions at an induction temperature of 37 °C. Although the Fcab sequences used to form a heterodimer pair were de-homologized to minimize the undesired recombination events, other elements of the expression cassette might have contributed to eventually decrease the efficiency of the display system.

1.6 Yeast Display Using Combined Genome-Integrated and Episomal Expression Cassette

Our next strategy was to incorporate the “Hole” chain of the heterodimer into the genome of Saccharomyces cerevisiae EBY100, and the “Knob” chain was expressed from a pYD1-based vector. Binding to soluble CD64 was considered particularly important, as its binding site on an Fc fragment depends on the dimer formation [41]. The first design tested involved randomization of the residues 358–362 in the AB-loop and 413–415 as well as 418–422, located in the EF-loop of the CH3 domain of the “Knob” chain (Fig. 1a). Comparing with the diploid system, the expression and correct folding of the recombinant proteins was more favorable, as we found that 22% of the cells reacted with CD64, 12% with the structure-reporting anti-CH2 antibody, and 39% with protein A when induced at 20 °C. Stress induction conditions at 37 °C nevertheless notably reduced the percentage of positive cells (Fig. 1b). In the Fcab molecules, all three C-terminal loops as well as the amino acid residues of the C-terminus can be mutated to mediate antigen binding [25], but in many examples more restricted mutagenesis is sufficient, typically involving AB- and EF-loops [18, 42, 43]. We have tested the importance of binding determinants for an anti-VEGF clone with mutations in AB-, CD-, EF-loops, and the C-terminus in a loop-reversion experiment where altered parts of the sequence have been reset to wild-type residues. Interestingly, we found that the mutated residues in AB-loop and C-terminus were sufficient for an antigen-binding Fc fragment (Fig. 2a). Although the antigen affinity was much lower than could be detected for the original VEGF-binding clone, this protein in soluble form exhibited a higher thermostability than could be established for a wild-type Fc, which can be led back to stabilizing motifs in the modified C-terminus of the VEGF-binding clone [27] (Fig. 2b). Thermostability is a measure of the degree of order in a protein and it is of course favorable not to further decrease the

Discovery of Heterodimeric Fcab™ Fragments

135

Fig. 1 (a) Front and side view of the first heterodimeric library design with surfaced randomized residues 358–362 in the AB-loop and 413–415 as well as 418–422 in the EF-loop of both heterodimer chains (PDB: 1OQO). Grey: CH2 domains, blue: “Knob” CH3 domain, magenta: “Hole” CH3 domain, green: carbohydrate. The side chains of the residues causing heterodimerization Y366 and T407 are depicted with sticks. The figure was prepared using PyMOL (Schro¨dinger LLC). (b) FACS analysis of this library using two different induction temperatures and staining with Fc ligands. Percentage of positive cells is indicated

KiH-induced impairment, which we also observed after heating the expressing cells and subsequent staining with the soluble CD64 (Fig. 2c). Therefore, we have designed a library with mutations of residues 359 and 360 and an insertion of five random residues in the AB-loop and mutagenized residues 440–447 in C-terminus as a starting point for heterodimeric Fcab libraries (Fig. 2d). We hence combined the integrated “Hole” chain with episomally encoded “Knob” Fc-library with modifications in the AB-loop and the C-terminus as described above, and with an insertion of five amino acid residues in the AB-loop. The library was constructed at the size of 2.7 × 108 independent members. The staining for reactivity with structural markers after an induction at 37 °C has revealed 40% CD64-, 34% anti-CH2-antibody-, and 49% of protein A-positive yeast cells, which was considered sufficient for sorting of properly folded antigen-reactive Fcab clones (Fig. 2e).

136

Filippo Benedetti et al.

Fig. 2 (a) Binding of VEGF by double loop reversion mutants of the CT6 Fcab clone analyzed with FACS (labels indicate mutagenized loops; C-ter: C-terminus). (b) Thermostability of the positive reversion mutant compared with wild-type Fc as determined by differential scanning calorimetry (midpoint of transition of the CH3 domain is indicated). (c) comparison of CD64 binding of yeast displaying wild-type Fc and a “KiH” heterodimerized Fc, heated to different temperatures, using median fluorescence. (d) Front and side view of the alternative heterodimeric library design with surfaced residues randomized in AB-loop and in C-terminus of both heterodimer chains (PDB: 1HZH). Grey: CH2 domains, blue: “Knob” CH3 domain, magenta: “Hole” CH3 domain, green: carbohydrate. The side chains of the residues causing heterodimerization Y366 and T407 are depicted with sticks. The figure was prepared using PyMOL (Schro¨dinger LLC). (e) FACS analysis of this library after induction at 37 °C and staining with Fc ligands and an anti-Xpress tag antibody. Percentage of positive cells is indicated

Discovery of Heterodimeric Fcab™ Fragments

1.7 Sorting of Heterodimeric Fcab™ Library and Identification of an Antigen-Specific Clone

137

For antigen selection, the clinically relevant immune-oncology checkpoint programmed-death ligand-1 (PD-L-1) [44] was chosen as a bait, as it can be found as a monomer on the cancer cell surface [45]. Regarding the large library size, the library was first processed using magnetic activating cell sorting (MACS), and the following selection rounds were performed with fluorescent-activated cell sorting (FACS) procedure. In each FACS round, the sorting gate for the CH2-specific antibody-positive and antigen-positive cells was set after selecting for the morphology of yeast population, single yeast cells, and excluding auto-fluorescent cells. After three selection rounds, an enrichment of antigen-binding clones could be observed, and single yeast colonies were plated out for screening (Fig. 3). A prominently represented clone sequence (80% of the positive screened clones), named F1, was chosen for reformatting into a soluble heterodimeric Fcab clone and could be well expressed in HEK293 system with a yield of over 50 mg/L culture as a protein of favorable biophysical properties with a monomeric profile in the size-exclusion chromatography (SEC) in native conditions (Fig. 4a). This Fcab clone could specifically react with the antigen in ELISA with an EC50 of 136 nM (comparing with 0.8 nM determined for the clinical antibody atezolizumab) (Fig. 4b) and recognized immobilized PD-L-1 with an affinity constant of 170 nM in a biolayer interferometry experiment (Fig. 4c). Thermostability of the clone was comparable to the wild-type KiH-heterodimerized Fc fragment, as determined by the differential scanning calorimetry (Fig. 4d). To test if the moderate ability of antigen binding of the Fcab clone can be improved, a homodimeric molecule with two mutated CH3 domains was produced. With the expression level of 50 mg/L in HEK system and monodisperse SEC profile, this Fcab clone could bind to PD-L-1 at an EC50 of 32 nM in ELISA (Fig. 4e) and 17 nM in BLI (Fig. 4c). It can hence be concluded that like for several antibodies, the avid binding resulting from two homodimeric binding sites contributed to high-affinity-antigen binding of the Fcab clone. Also, the thermostability of this clone was high,

Fig. 3 Antigen selection of the heterodimeric Fcab yeast display library. Bivariate plots show the reactivity of yeast cells with the antigen (y-axis) and anti-CH2-antibody (x-axis), the numbers in the gates refer to the percentage of collected cells, and “in” and “out” the number of processed and collected cells, respectively

138

Filippo Benedetti et al.

Fig. 4 (a) Size exclusion chromatography in native conditions reveals monomeric status of wild-type “KiH” Fc, heterodimerized anti-PD-L-1 Fcab clone F1 and its homodimeric version. MWS: molecular weight standard with proteins of 670, 158, 44, 17, and 1.35 kDa. (b) Reactivity of the F1 Fcab clone with immobilized PD-L-1 in ELISA in comparison with atezolizumab (upper panel) and the non-reactivity with the control antigens (lower panel). (c) Biolayer interferometry determination of KD of heterodimeric and homodimeric F1 clone for binding to immobilized PD-L-1. (d) Thermal denaturation of “KiH” Fc, heterodimeric F1 clone, wild-type Fc, and homodimeric F1 clone measured with differential scanning calorimetry. (e) Binding of heterodimeric and homodimeric version of F1 to PD-L-1 in ELISA (upper panel) and binding of homodimeric F1 to PD-L-1 and control antigens (lower panel). (f) Model of the F1-AB-loop showing the postulated novel salt bridge. The figure was prepared using PyMOL (Schro¨dinger LLC)

Discovery of Heterodimeric Fcab™ Fragments

139

differing from the wild-type only for 5 °C in the midpoint of transition (Tm) corresponding to melting of CH3 domains (Fig. 4d). Molecular modelling of the F1-clone was performed with SWISS-MODEL workspace [46] and energy minimization with YASARA (RRID:SCR_017591) proposed a structure very similar to the wild-type Fc, and the extended AB-loop region exhibited a short alpha-helical motif. Interestingly, it was proposed that the newly introduced residue D359 forms a salt bridge with the wildtype K414, which could positively affect the stability of the F1 clone (Fig. 4f). 1.8

2 2.1

Future Prospects

Realizing that the stability of the displayed construct is crucial to the successful selection of antigen-binding heterodimeric Fcab clones, we tested if heterodimerization motifs assigning the Fc a higher stability level also have a positive influence on display parameters of Fcab fragments. For this purpose, steric-clash-based ZW1-heterodimerization motif which does not lower the thermostability of the Fc fragment [35], introducing four mutations in the “Knob” chain and four mutations in the “Hole” chain, was used. The apparent thermostability of the ZW-1-heterodimerized Fc on the yeast surface indeed superseded the KiH-heterodimerized version (Fig. 5a). Heterodimeric libraries in ZW-1 context with the mutagenized “Knob”-chain, carrying the randomizations in AB loop (358–362) and EF-loop (413–415 and 418–422) (Fig. 5b), expressed from an episomal plasmid, also featured a large proportion of cells binding to structure-reporting ligand molecules after induction at 37 °C (Fig. 5c). In summary, stabilized Fc-scaffold offers extended options for the introduction of mutagenized regions and thereby expanded possibilities for the selection of Fcab fragments with diverse architectures.

Materials Reagents

1. EZ-Link™ Scientific).

Sulfo-NHS-LC-LC-Biotin

(Thermo

Fisher

2. Anti-Xpress antibody (Thermo Fisher Scientific). 3. Anti-CH2 antibody (Bio-RAD). 4. Protein A-biotin (Calbiochem). 5. CD64 (Fc gamma RI) (his-tagged) (R&D systems). 6. Anti-biotin antibody-APC conjugate (MACS Miltenyi). 7. Anti-pentahis antibody Alexa Fluor-488 conjugate (QIAgen). 8. Anti-pentahis antibody Alexa Fluor-647 conjugate (QIAgen). 9. Anti-mouse IgG (Sigma-Aldrich).

(Fab-specific)

F(ab′)2-FITC

conjugate

Fig. 5 (a) Comparison of CD64 binding of yeast displaying “ZW1”- and a “KiH”-heterodimerized Fc, heated to different temperatures, using the percentage of positive cells. (b) Front and side view of ZW-1 heterodimeric library design with surfaced randomized residues 358–362 in the AB-loop and 413–415 as well as 418–422 in the EF-loop of both heterodimer chains (PDB: 1OQO). Grey: CH2 domains, blue: “Knob” CH3 domain, magenta: “Hole” CH3 domain, green: carbohydrate. The side chains of the residues causing heterodimerization (“Knob” chain: V350, L366, 392L, 394W; “Hole” chain: V350, Y351, A405, V407) are depicted with sticks. The figure was prepared using PyMOL (Schro¨dinger LLC). (c) FACS analysis of this library after induction at 37 °C and staining with Fc ligands and an anti-Xpress tag antibody. Percentage of positive cells is indicated

Discovery of Heterodimeric Fcab™ Fragments

141

10. Goat anti-mouse IgG (H + L) cross-adsorbed secondary antibody-PE conjugate (Thermo Fisher Scientific). 11. Neutravidin-phycoerythrin (PE) (Thermo Fisher Scientific). 12. Streptavidin-Alexa Fluor-647 conjugate (Thermo Fisher Scientific). 13. High-fidelity proofreading polymerase (such as Q5 Hi-Fidelity 2× MasterMix, New England Biolabs). 14. MyTaq™ Red Mix (Bioline). 15. Restriction enzymes. 16. T4 ligase and 10× ligase buffer (New England Biolabs). 17. QuikChange (Agilent). 2.2 Solutions and Buffers

Lightning

Site-Directed

Mutagenesis

Kit

1. 50% PEG3350. 2. 1 M Li-acetate solution. 3. 2 mg/mL salmon sperm DNA. 4. 500 mM EDTA, pH 8.0. 5. Buffer for yeast media: 1 M potassium phosphate buffer, pH 6.0. 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: 10% bovine serum albumin (BSA) in PBS. 10. Candor solution (Candor Bioscience). 11. MACS buffer: 0.25% BSA in PBS with 2 mM EDTA, pH 7.2. 12. FACS staining solution: 2% BSA in PBS. 13. Freezing buffer: 30% glycerol, sterilized with filtration. 14. Polyethyleneimine (PEI) (Polysciences).

2.3

Media

1. YPAD: 20 g/L peptone, 10 g/L yeast extract, 2% glucose, 0.125 g/L adenine hemisulfate. 2. SD-CAA: 1% CAA solution with 0.67% YNB with ammonium sulfate, 100 mM KH2PO4/ K2HPO4 buffer, pH 6.0, 2% glucose. 3. SG/R-CAA: 1% CAA solution with 0.67% YNB with ammonium sulfate, 100 mM KH2PO4/K2HPO4 buffer, pH 6.0, 2% galactose, 1% raffinose.

142

Filippo Benedetti et al.

4. SD-Leu/-Trp with glucose: 1× Drop-out Supplements solution (DO-Leu/-Trp), 0.67% YNB with ammonium sulfate, 100 mM KH2PO4/K2HPO4 buffer, 2% glucose. 5. SD-Leu/-Trp with galactose/raffinose: 1× Drop-out Supplements solution (DO-Leu/-Trp), 0.67% YNB with ammonium sulfate, 100 mM KH2PO4/K2HPO4 buffer, 2% galactose, 1% raffinose. 6. YPAD solid media: 1.5% agar with 20 g/L peptone, 10 g/L yeast extract, 2% glucose, and 0.125 g/L adenine hemisulfate. 7. SD-Leu-Trp solid media: 1.5% agar with 1× Drop-out supplements solution (DO-Leu/-Trp), 0.67% YNB with ammonium sulfate, 100 mM KH2PO4/ K2HPO4 buffer, and 2% glucose. 8. Complete F17 medium: F17 medium (Thermo Fisher Scientific) with 4 mM L-glutamine, 1% Pluronic® F68, and 0.25 μg/ mL G418. 9. 20% TN-1 solution: 200 g TN-1 (Organotechnie) in 1 L complete F17 medium. 10. LB medium: 1.5% agar with 10 g/L peptone, 5 g/L yeast extract, and 10 g/L NaCl. 11. SOC medium: 0.5% yeast extract, 2% peptone, 10 mM NaCl, 2.5 mM KCl, 10 mM MgCl2, 10 mM MgSO4, and 20 mM glucose. 12. 100× pen-strep solution: 100000 U/mL penicillin and 10 mg/mL streptomycin. 13. 1000× ampicillin solution: 100 mg/mL ampicillin solution. 2.4

Kits

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

2.5

Equipment

1. 96 U-well microtiter plates. 2. 1.5 mL microcentrifuge tubes. 3. 4.5 and 1.0 mL cryotubes. 4. 15 and 50 mL conical tubes. 5. 125, 500, 1000, and 2000 mL Erlenmeyer flasks. 6. 24-well-culture plates, round well bottom (Whatman® UNIPLATE). 7. Semi-permeable plate-sealing membranes. 8. 90 mm Petri dishes. 9. Mr. Frosty isopropanol bath.

Discovery of Heterodimeric Fcab™ Fragments

143

10. Pipettes from P10 to P1000 range with respective tips and multichannel pipettes. 11. Rotating wheel. 12. -80 °C freezer. 13. Incubator with shaking platform and temperature set to 20, 30, and 37 °C. 14. Heating blocks and water baths. 15. SuperMACS magnetic bead separator. 16. High-speed cell sorter such as ARIA I (Becton-Dickinson). 17. FACS analysis apparatus. 18. Gradient thermal cycler. 19. Nanodrop spectrophotometer. 2.6 Plasmids, Bacterial Strains, Yeast Strains, and Cell Lines

1. pCM218 vector. 2. pYD1 yeast display vector kit (Thermo Fisher Scientific). 3. pTT5 vector (Canadian National Research Council). 4. E. coli TOP 10 (F– mcrA Δ(mrr-hsdRMS-mcrBC) φ80lacZΔM15 ΔlacX74 recA1 araD139 Δ(ara-leu)7697 galU galK λ– rpsL(StrR) endA1 nupG) (Thermo Fisher Scientific). 5. EBY100 (MATa GAL1-AGA1::URA3 ura3-52 trp1 leu2Δ1 his3Δ200 pep4::HIS2 prb1_1.6R can1 GAL) (ATCC® MYA4941™). 6. HEK293-6E™ (Canadian National Research Council).

3

Methods

3.1 Construction of the Recipient Strain with the GenomeIntegrated Fcab™ Expression Cassette

As only the “Knob” chain of the heterodimer was first selected for mutagenesis, the invariant wild-type Fc “Hole” variant was integrated into the yeast genome (sequences of the constructs in Table 1). Vector pCM218 [47] was used, with a mutated tetR’ moiety under the control of cytomegalovirus promoter. The amplicon with the expression cassette containing the galactose-inducible GAL1, 10-promoter, a yeast leader peptide sequence to guide soluble expression, and the sequence encoding the desired “Hole” chain, is cloned into the polylinker site (sequences of all oligonucleotides in Table 2). The EcoRV-linearized plasmid is intended to integrate into the chromosomal mutated LEU2 locus. 1. Amplify the expression cassette containing the GAL1, 10-promotor sequence, leader peptide sequence and the sequence of the wild-type Fc “Hole” chain with a PCR using primers that allow cloning within the pCM218 polylinker region (incorporating, e.g., HindIII and EcoRI restriction sites).

144

Filippo Benedetti et al.

Table 1 Amino acid and nucleotide sequences of wild-type constructs and Fcab clones Wild type Fc-Amino acid sequence TCPPCPAPELLGGPSVFLFPPKPKDTLMISRTPEVTCVVVDVSHEDPEVKFNWYVDGV EVHNAKTKPREEQYNSTYRVVSVLTVLHQDWLNGKEYKCKVSNKALPAPIEKTISKAK GQPREPQVYTLPPSRDELTKNQVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPPVL DSDGSFFLYSKLTVDKSRWQQGNVFSCSVMHEALHNHYTQKSLSLSPGK “Knob” heterodimer – nucleotide sequence ACGTGTCCCCCATGTCCCGCCCCTGAGCTGCTGGGCGGCCCTTCCGTGTTCCTGTTCC CTCCCAAGCCAAAGGACACCCTGATGATCTCCCGGACCCCTGAGGTGACCTGTGTGGT GGTGGACGTGAGCCACGAGGACCCAGAGGTGAAGTTCAACTGGTACGTGGACGGCGTG GAGGTGCACAACGCCAAGACCAAGCCTAGAGAGGAGCAGTACAACAGCACCTACCGCG TGGTGAGCGTGCTGACCGTGCTGCACCAGGATTGGCTGAATGGCAAGGAGTACAAGTG CAAGGTGAGCAACAAGGCCCTGCCTGCCCCCATCGAGAAGACCATCTCCAAGGCCAAG GGCCAGCCTCGAGAACCACAGGTGTACACCCTGCCCCCATCCCGGGATGAGCTGACCA AGAACCAGGTCAGCCTGTACTGCCTGGTCAAAGGCTTCTATCCCAGCGACATCGCCGT GGAGTGGGAGAGCAATGGGCAGCCGGAGAACAACTACAAGACCACGCCTCCCGTGCTG GACTCCGACGGCTCCTTCTTCCTCTACAGCAAGCTCACCGTGGACAAGAGCAGGTGGC AGCAGGGGAACGTCTTCTCATGCTCCGTGATGCATGAGGCTCTGCACAACCACTACAC ACAGAAGAGCCTCTCCCTGTCTCCGGGTAAA “Hole” heterodimer-nucleotide sequence ACATGCCCACCGTGCCCAGCACCTGAACTCCTGGGGGGACCGTCAGTCTTCCTCTTCC CCCCAAAACCCAAGGACACCCTCATGATCTCCCGGACCCCTGAGGTCACATGCGTGGT GGTGGACGTGAGCCACGAAGACCCTGAGGTCAAGTTCAACTGGTACGTGGACGGCGTG GAGGTGCATAATGCCAAGACAAAGCCGCGGGAGGAGCAGTACAACAGCACGTACCGTG TGGTCAGCGTCCTCACCGTCCTGCACCAGGACTGGCTGAATGGCAAGGAGTACAAGTG (continued)

Discovery of Heterodimeric Fcab™ Fragments

145

Table 1 (continued) CAAGGTCTCCAACAAAGCCCTCCCAGCCCCCATCGAGAAAACCATCTCCAAAGCCAAA GGGCAGCCTCGAGAACCGCAGGTTTATACTCTGCCTCCGAGCCGTGACGAACTGACTA AAAATCAGGTTTCACTGACGTGTCTGGTGAAAGGTTTTTACCCGTCTGATATTGCAGT TGAATGGGAAAGTAACGGTCAGCCTGAAAATAACTACAAAACAACCCCACCGGTTCTG GATAGTGATGGTAGCTTTTTTCTGACGTCCAAACTGACTGTTGATAAATCTCGTTGGC AGCAGGGTAATGTTTTTAGCTGTAGCGTTATGCATGAAGCCCTGCATAATCATTATAC CCAGAAATCGCTGAGCCTGAGTCCAGGCAAA F1 Fcab clone – Knob variant Amino acid sequence TCPPCPAPELLGGPSVFLFPPKPKDTLMISRTPEVTCVVVDVSHEDPEVKFNWYVDGV EVHNAKTKPREEQYNSTYRVVSVLTVLHQDWLNGKEYKCKVSNKALPAPIEKTISKAK GQPREPQVYTLPPSRDELDGTWRGNNQVSLYCLVKGFYPSDIAVEWESNGQPENNYKT TPPVLDSDGSFFLYSKLTVDKSRWQQGNVFSCSVMHEALHNHYTQKSMMYARVR Nucleotide sequence ACGTGTCCCCCATGTCCCGCCCCTGAGGCAGCTGGCGGCCCTTCCGTGTTCCTGTTCC CTCCCAAGCCAAAGGACACCCTGATGATCTCCCGGACCCCTGAGGTGACCTGTGTGGT GGTGGACGTGAGCCACGAGGACCCAGAGGTGAAGTTCAACTGGTACGTGGACGGCGTG GAGGTGCACAACGCCAAGACCAAGCCTAGAGAGGAGCAGTACAACAGCACCTACCGCG TGGTGAGCGTGCTGACCGTGCTGCACCAGGATTGGCTGAATGGCAAGGAGTACAAGTG CAAGGTGAGCAACAAGGCCCTGCCTGCCCCCATCGAGAAGACCATCTCCAAGGCCAAG GGCCAGCCTCGAGAACCACAGGTGTACACCCTGCCCCCATCCCGGGATGAGCTGGATG GTACTTGGCGTGGTAATAACCAGGTCAGCCTGTACTGCCTGGTCAAAGGCTTCTATCC CAGCGACATCGCCGTGGAGTGGGAGAGCAATGGGCAGCCGGAGAACAACTACAAGACC ACGCCTCCCGTGCTGGACTCCGACGGCTCCTTCTTCCTCTACAGCAAGCTCACCGTGG ACAAGAGCAGGTGGCAGCAGGGGAACGTCTTCTCATGCTCCGTGATGCATGAGGCTCT GCACAACCACTACACACAGAAGAGTATGATGTATGCGCGGGTGAGGTGATAA For F1 clone, the modifications of the Fc sequence are highlighted: mutated and inserted residues in AB-loop in blue, mutated residues in C-terminus in red, and the “Knob” mutation T366Y in bold lettering

146

Filippo Benedetti et al.

Table 2 Primers for cloning of Fcab expression cassettes, preparation of library fragments, and recloning for the expression in animal cells Yeast display Cloning to pCM218 PCR-amplification of Fc “Hole” FChet_cass_hind1

acgtaagctt acggattaga agccgccgag c

FChet_cass_ecor2

agcggaattc aattctctta ggattcgatt c

Sequencing of pCM218-cloned constructs alphafo

ctacaacaga agatgaaacg

218 seqback

gattaagttg ggtaacgcc

Yeast library Randomized: AB loop: 359–360 + 5 inserted amino acids, C-terminus: 440–447 PCR-amplification of library fragment for recombination with pYD1dem, linearized with Esp3I and NotI Primer1

caccctgccc ccatcccggg atgagctgnn knnknnknnk nnknnknnka accaggtcag cctgtac

Primer2

cgaagggccc tctagactcg atcgagcggc cgcttatcam nnmnnmnnmn nmnnmnnmnn mnncttctgt gtgtagtggt tg

Randomized: AB loop: 358–362, EF-loop: 413–415 and 418–422 PCR-amplification of library fragment for recombination with pYD1dem, linearized with Esp3I Primer1

gaaccacagg tgtacaccct gcccccatcc cgggatgagn nknnknnknn knnkgtcagc ctgtactgcc tggtcaaag

Primer2

gtggttgtgc agagcctcat gcatcacgga gcatgagaam nnmnnmnnmn nmnnccacct mnnmnnmnnc acggtgagct tgctgtaga

Sequencing of pYD1-cloned constructs pYD fwd

agtaacgttt gtcagtaatt gc

pYD rev

gtcgattttg ttacatctac ac

Conversion of the “Knob” chain to homodimer Mutagenesis of recipient vector with cloned F1; reset of Y366 to wildtype T F1_Y366T

aaccaggtca gcctgacctg cctggtcaaa ggc

F1_Y366Ta

gcctttgacc aggcaggtca ggctgacctg gtt

(continued)

Discovery of Heterodimeric Fcab™ Fragments

147

Table 2 (continued) HEK expression Cloning to pTT5 with wild-type Fc, restricted with BsrGI and BamHI ABnest

ccaagggcca gcctcgagaa ccacaggtgt acac

Ytohbam2

acgtggatcc gaagggccct ctagactcga tcgag

Sequencing of pTT5-cloned constructs pTT forward

gccatacact tgagtgacaa tgacatc

pTT reverse

ccaaatgatt tgccctccca tatgtc

Relevant restriction sites are indicated in italics (BsrGI site in the ABnest primer occurs naturally in the Fc sequence)

2. Clone the amplicon into pCM218. 3. Linearize the vector with EcoRV using incubation of 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 3.3 and downscale the transformation components volume by 10. Plate the transformation reaction to SD-Trp/Leu plates supplemented with tryptophan. Incubate at 30 °C for 3 days. 3.2 PCR Screening of the Yeast Colonies

1. Prepare 30 μL Solution 1 (Digestion buffer) with 1 μL Zymolyase (both Zymo Research) per colony intended for analysis. 2. Use this solution to resuspend approximately 2 μL of a wellseparated transformed yeast colony and incubate at 37 °C with vigorous shaking for 45 min. 3. Incubate at 95 °C for 5 min. 4. Freeze at -80 °C. 5. Let thaw and briefly spin down the cell debris in a microfuge. Lysates can be stored for later use at -20 °C. 6. Use the supernatant as a template in a PCR reaction. For screening, an inexpensive polymerase, such as MyTaq™ Red, can be used. PCR reaction mix consists of 1× MyTaq™ Red mastermix, 10 pmol of each oligonucleotide, and 1 μL of lysate in a 10 μL-reaction volume. Amplification works well with the protocol in Table 3.

148

Filippo Benedetti et al.

Table 3 PCR cycling conditions for amplification with MyTaq™ Red Segment

Cycles

1

1

2

35

3

1

Temperature (°C)

Time

Initial denaturation

95

3 min

Denaturation

95

15 s

Annealing

55

15 s

Extension

72

15 s

Final extension

72

5 min

3.3 Transformation of the Variant Heterodimer Chain Library

For the construction of large libraries, gap repair-driven homologous recombination in yeast is an efficient mechanism that enables of final genetic constructs based only on the 25 bases of homology between the PCR fragment and the recipient vector. To optimize the recombination of library inserts with different designs, the recipient vector was linearized at different positions to minimize the residual sequences and hence reduce the required length of the oligonucleotides, and thereby the probability of errors resulting from the synthesis. The introduction of the Type IIS restriction enzyme recognition sites has proven optimally suitable for this purpose, such as in the case of the pYD1dem vector [25] used for the construction of heterodimer libraries described here.

3.3.1 Library-Encoding PCR Fragment Preparation

Depending on the desired library design, the randomized oligonucleotides will contain: – An Fc-fragment template matching sequence for annealing of minimum 15 nucleotides. – Randomized regions. – Homologous end joining region of minimum 25 nucleotides to enable annealing to the recipient vector or other fragments intended for insertion if a multi-fragment transformation is performed. For the described library designs, the randomized stretches of amino acids were encoded on two oligonucleotides of less than a total length of 100 nucleotides each. Because of the relatively long primers and the presence of randomized regions, the PCR steps were performed with a touch-down protocol described below and a proofreading polymerase mix to minimize the errors. 1. Perform a PCR with 1× Q5 High Fidelity Polymerase mix, 100 pmol of each primer, and 10 ng of the template Fc-encoding plasmid per 100 μL reaction mix.

Discovery of Heterodimeric Fcab™ Fragments

149

Table 4 PCR cycling conditions for amplification with Q5 Hi-Fidelity polymerase Segment

Cycles

1

1

2

6

3

29

4

1

Temperature (°C)

Time

Initial denaturation

98

5 min

Denaturation

98

20 s

Annealing

64–1.5/cycle

20 s

Extension

72

20 s

Denaturation

98

20 s

Annealing

55

20 s

Extension

72

20 s

Final extension

72

5 min

2. Each polymerase is provided with the recommendation of the producer regarding the optimal annealing and extension temperature and extension time. An example of optimized protocol with touch-down steps for use with Q5 Hi-Fidelity Polymerase is presented in Table 4. 3. Analyze the products of the reaction using agarose gel electrophoresis to decide whether to further optimize the PCR protocol or gel purify the fragment. The latter is typically accompanied with about double the loss of product comparing with liquid PCR-product purification. 4. Determine DNA spectrophotometer.

concentration

using

Nanodrop

5. Store the DNA at -20 °C. 3.3.2

Recipient Vector

1. Perform the linearization of recipient vector with digestion of 50 μg vector DNA with 50 U of each restriction enzyme in an appropriate buffer. 2. Control the linearization using agarose gel electrophoresis. Use an aliquot of uncut vector incubated in the mixture as required for the digest, but without the restriction enzymes, as a control. 3. Purify the vector backbone using a DNA-purification kit. Use gel extraction if the excised fragment is greater than 100 base pairs (see Note 1). 4. Determine DNA spectrophotometer.

concentration

5. Store the DNA at -20 °C.

using

Nanodrop

150

Filippo Benedetti et al.

3.3.3 Library Transformation

A positive strain obtained when following the protocol described in Subheading 3.1 can be propagated and transformed with a library of variants of the “Knob” chain of the Fc heterodimer, cloned into a vector assigning tryptophan prototrophy such as pYD1. The protocol describes the transformation with 100 μg linearized vector and 100 μg library insert, which leads to about 108 colonies. 1. Start an overnight culture from a fresh streak of the recipient strain in 10 mL YPAD on an orbital shaker at 30 °C (see Note 2). 2. On the next day, determine the OD600 of the culture and dilute to OD600 of 0.4 in a 500 mL pre-warmed YPAD medium. 3. Incubate the yeast culture on an orbital shaker at 30 °C until it reaches an OD600 between 1.5 and 2. 4. Distribute the culture to 50 mL-conical tubes. 5. Pellet the yeast by centrifugation at 1000 g, 5 min, at room temperature (RT). 6. Wash the pellets with 25 mL sterile distilled water. 7. Collect the cells by centrifugation at 1000 g, 5 min, at RT. 8. Resuspend each pellet in 3 mL of 200 mM Li-acetate solution and let shake at 30 °C for 15 min. 9. Collect the cells by centrifugation at 1000 g, 5 min, at RT and discard the Li-acetate solution. 10. Resuspend the cells in the dregs of the Li-acetate as the pelleted cells are difficult to resuspend in transformation mixture containing PEG3350. 11. Add 2.4 mL 50% PEG3350, 360 μL 1 M Li-acetate, 500 μL heat-shocked (5 min at 95 °C) salmon sperm DNA, and 10 μg of each insert and vector DNA in a volume that should not exceed 340 μL. Let shake at 30 °C for 30 min. 12. Incubate at 42 °C in a water bath for 45 min. Invert the tubes every 10 min. 13. Pellet the yeast and remove the transformation solution. 14. Add 5 mL YPAD medium and incubate with shaking at 30 °C for 30 min. 15. Collect the cells by centrifugation at 1000 g, 5 min, at RT. Inoculate into 500 mL SD-Leu/-Trp medium supplemented with pen/strep. 16. 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. 17. Incubate the yeast library on an orbital shaker at 30 °C for 24 h.

Discovery of Heterodimeric Fcab™ Fragments

151

18. Passage 25–475 mL fresh SD-Leu/-Trp medium with pen/strep for the selection to proceed with shaking at 30 °C for 24 h. 19. Collect yeast cells with centrifugation at 1000 g, 5 min at 4 °C, discard medium, and mix with an equal volume of freezing buffer. About 20 mL of yeast suspension will be produced. 20. To determine the number of viable cells, remove an aliquot of the culture and dilute in SD-Leu/-Trp before spreading on SD-Leu/-Trp plates and incubating for 3 days at 30 °C. Between 109 and 1010 cells/mL, cryostock are expected. 21. Aliquot to 1 mL cryotubes, place them into a Mr. Frosty isopropanol bath cooled to 4 °C, transfer to -80 °C, and keep them incubated overnight. Afterwards, they can be stored at -80 °C. 3.4 Quality Control and Sorting of Heterodimer Fc Yeast Display Libraries 3.4.1 Sequencing of Library Clones 3.4.2 Staining of Displayed Fc Heterodimer Mutants

The constructed library is checked for level of correctness on the genotypic level using PCR with primers recommended in Table 2 and sequencing. The plasmid DNA can either be purified using Zymoprep II kit (see Notes 3 and 4), or crude lysate prepared as described in Subheading 3.2 can serve as a template.

After the selection, yeast libraries are induced in minimal medium containing 2% galactose as the inductor and 1% raffinose as the carbon source, to control the expression via potential N- and C-terminal tags fused with the Fc, as well as the properties of the displayed Fc fragments that can be judged from their binding to structure-sensitive antibodies and other ligands (Table 5 includes recommended dilutions). The induction temperature can vary from 20 to 37 °C, with the higher values intended to interrogate the effect of stressed yeast protein folding machinery and enhanced protease activity on the display level and the integrity on the displayed proteins (see Note 5). For the yeast display of heterodimeric Fc, induction at 37 °C should proceed overnight not to critically impair the viability, while the cultures should be incubated at 20 °C for 48 h to achieve full induction. 1. Inoculate single colonies into 2 mL SD-CAA medium with pen/strep (see Note 6) in 24-round bottom well plates. Cover the plate with a semi-permeable membrane. 2. Incubate overnight at 30 °C with shaking at 180 rpm. 3. Determine OD600 by diluting the cultures at 1:20–1:50 in PBS, to achieve the OD600 values from 0.2 to 0.8, where the cell concentration is in linear correlation with the absorption. Dilute the culture to OD600 of 1 into 2 mL SG/R-CAA medium with pen/strep. Cover the plate with a semipermeable membrane.

152

Filippo Benedetti et al.

Table 5 Dilutions of detection reagents used for analysis of yeast-displayed heterodimeric Fc libraries and clones Final dilution (μg/mL) Primary reagents Anti-Xpress antibody

1

Anti-CH2 antibody

5

Protein A – biotin

10

CD64

1

Conjugates Anti-pentahis-Alexa Fluor® 488 ®

1

Anti-pentahis-Alexa Fluor 647

1

Anti-mouse IgG (Fab-specific) F(ab′)2-FITC conjugate

5

Goat anti-mouse IgG (H + L) cross-adsorbed antibody-PE

1

Neutravidin-PE

1.25 ®

Streptavidin-Alexa Fluor 647

1

Anti-biotin antibody-APC

1

4. Incubate at the desired temperature with shaking at 180 rpm. 5. Dilute the induced cultures to OD600 of 1 in 2% BSA-PBS and block for 30 min at RT. 6. Distribute the cells in 100 μL aliquots into a 96-well plate. 7. Collect the cells using centrifugation at 1000 g, 5 min at 20 °C, and remove the supernatant. 8. Add 100 μL of the primary antibodies or Fc ligands, diluted in 2% BSA-PBS, and incubate for 30 min at RT. 9. If a two-step staining is used, place the plate on ice for 5 min. 10. Collect the cells using centrifugation at 1000 g, 5 min at 4 °C, and remove the supernatant. 11. Perform a wash step by adding 200 μL ice-cold PBS. 12. Pellet the cells using centrifugation at 1000 g, 5 min at 4 °C, and remove the supernatant. 13. Add 100 μL of the solution containing fluorescently labeled reagents in ice-cold 2% BSA-PBS. 14. Incubate for 30 min on ice, protected from light. 15. Collect the cells using centrifugation at 1000 g, 5 min at 4 °C, and remove the supernatant. 16. Resuspend in 200 μL ice-cold PBS and incubate on ice until flow cytometry analysis.

Discovery of Heterodimeric Fcab™ Fragments 3.4.3 Selection of Heterodimeric Fc Libraries

153

Although the FACS-based methods allow excellent visual discrimination of the yeast cells that should be propagated in the selection rounds, including not only their antigen binding properties, but also reactivity with the expression normalization markers and structure-sensitive ligand, as well as the singlet status, libraries intended for de novo antigen selection are mostly too large to be processed at 20-fold coverage in the initial selection step. Therefore, MACS protocols are used to collect antigen-binding clones and proceed using its output of manageable size. Previously to the selection, staining is performed as described in Subheading 3.2, step 2, only with a representative number of yeast cells, corresponding to 10–20-fold of the number of independent library members, and including biotinylated or otherwise tagged antigen in the first step and streptavidin magnetic beads for MACS procedure or streptavidin-based detection reagent for FACS-based selection in the second step. The choice of blocking agent can be tailored to steer the stringency of the selection. 1. Perform a MACS selection starting with 4 × 109 cells per LS column. Inoculate the expected output of 106–107 cells to 50 mL SD-CAA with pen/strep. For enumeration, plate out the aliquots of this solution to SD-Leu/-Trp plates and incubate at 30 °C for 3 days. Incubate the rest of the culture at 30 ° C with shaking at 180 rpm. 2. On the next day, the OD600 of the liquid culture should exceed 10. Proceed with freezing or/and induction by resuspending a representative number of cells in SG/R-CAA medium with pen/strep at an OD600 of 1. 3. For FACS-based selection, perform the staining as described in Subheading 3.4.2 and include the antigen as a bait. 4. Before sorting, strain the cell suspension through 40 μM-nylon mesh capped tubes not to plug the nozzle with cell clumps. 5. For the sorting procedure, set a hierarchy of gates, defining first yeast population in FSC/SSC scatter, selecting for singlet cells in FSC/BSC, gating out the auto-fluorescent cells (see Note 7), and finally collecting top 0.5% of antigen-positive and CH2specific antibody (or another structural marker)-reactive cells. At least 20-fold output of the previous sorting round should be processed. Helpful are the following controls: – cells stained with each antigen and normalization (or folding) marker molecule alone, together with their corresponding secondary reagent: to assess the effect of the spectral overlap of the fluorophores. – cells stained with all reagents except the antigen: to assess the background of the staining and potentially set the gate to a certain percent of “false positive” population (see Note 8).

154

Filippo Benedetti et al.

3.5 Thermal Stability of Yeast-Displayed Heterodimeric Fcab™ Fragments

We have applied the heating and subsequent anti-CH2- antibody staining of yeast displayed wild-type Fc variants with heterodimerization motifs that in soluble form lead to differences in thermal stability, to assess the correlation with the properties of yeastexpressed molecules. The protocol can be applied to model libraries with different designs to deliver a rapid assessment of thermostability of displayed proteins on library scale [21, 48]. 1. Dilute a sample of induced library culture in PBS at OD600 of 1. Put aside an untreated aliquot that will be used as control. 2. Distribute the sample in 100 μL-aliquots into 200 μL-PCR tubes. 3. Heat up a gradient cycler to the range of temperatures chosen for testing. 4. Incubate the samples for 30 min, then let cool down. 5. Add 100 μL 4% BSA-PBS solution and transfer into a 96 Uwell plate. Perform the blocking step at RT for 30 min. 6. Centrifuge at 1000 g, 5 min at RT and remove the supernatant. 7. Resuspend in 100 μL of 1 μg/μL soluble his-tagged CD64 in 2% BSA-PBS. 8. Incubate with shaking for 30 min at RT. Place on ice. 9. Centrifuge at 1000 g, 5 min at 4 °C and remove the supernatant. 10. Perform the wash step by resuspending the cells in 200 μL ice-cold PBS. 11. Centrifuge at 1000 g, 5 min at 4 °C and remove the supernatant. 12. Resuspend in 100 μL anti-His-AlexaFluor 488 conjugate, diluted to 0.5 μg/mL, or anti-His-AlexaFluor 647 conjugate, diluted to 0.25 μg/mL, in ice-cold 2% BSA-PBS. 13. Incubate for 30 min on ice. 14. Centrifuge at 1000 g, 5 min at 4 °C and remove the supernatant. 15. Resuspend in 200 μL ice-cold PBS and proceed with FACS analysis. 16. Record the median fluorescence values of yeast population and plot against the incubation temperature on the x-axis.

3.6 Expression of Selected Fcab™ Clones in Mammalian Expression System

After the identification of antigen-positive yeast clones, the variant heterodimer chain can be rapidly reformatted for the protein to be expressed in suspension HEK293-6E cells. Apart of the heterodimeric PD-L-1 specific Fcab clone, a homodimer can simply be produced by reverting the T366Y mutation to wild-type using Quikchange Lightning mutagenesis kit and oligonucleotides listed in Table 2.

Discovery of Heterodimeric Fcab™ Fragments

155

1. Produce the PCR fragment encoding the mutated Fcab CH3 domain sequence with primers listed in Table 2 using a proofreading polymerase. 2. Clone into correspondingly cut pTT5 vector with cloned Fc fragment sequence and transform to E. coli. 3. Select on LB-amp medium. Verify the presence of insert of the correct length and sequence with primers listed in Table 2. 4. Purify plasmid DNA using a mini-preparation kit. 5. For PEI-mediated transfection, prepare 1 μg of total DNA (mix of the heterodimer chains in the ratio of 1:1) and 2 μg of PEI per mL HEK293-6E culture intended for transfection, dilute each in 1/20 volume of F17 complete medium, combine the solutions and incubate for 15 min at room temperature. 6. For screening, deliver the solution dropwise onto mammalian cells (HEK293-6E) at a density of 1.5–2.0 × 106/mL using 2 mL-cultures in 12-well-cell culture plates. Cover the plates with a semi-permeable membrane and incubate at 37 °C in humidified atmosphere with 5% CO2 with shaking at 130 rpm for 48 h. 7. Add 0.5 mL complete F17 medium with 50 μL 20% tryptone TN-1 solution to the culture. Cultivate for further 48 h (see Note 9). 8. Harvest the supernatant by centrifugation at 1000 g, 15 min, 4 °C. Supernatant can now be used for SDS-PAGE analysis to estimate Fcab expression. 9. This protocol can be used at a larger scale for production of larger quantities of the supernatant, which is processed using Protein A chromatography to obtain purified Fcab proteins.

4

Notes 1. Typically, the DNA fragments smaller than 100 base pairs cannot be efficiently purified with DNA purification kits, which is why this preparation step can be used to remove them from the desired fragment. Nevertheless, the efficiency varies between the different products, and it can be recommended that products of different manufacturers are tested to find the one best suited for the experimental set-up. 2. Although S. cerevisiae EBY100 strain is not deficient in adenine metabolism, the addition of adenine hemisulfate to the YPD medium enhances the growth rate of the strain, which is important especially when preparing of exponential cultures required

156

Filippo Benedetti et al.

for transformation. This phenomenon can probably be led back to the relatively common occurrence of spontaneous mutations influencing the adenine synthesis pathway (10-5–10-6/ division). 3. Zymoprep II protocol can be used exactly according to the manufacturer’s instructions, but the yield can be improved if the incubation with the lysis buffer (Solution II) is prolonged from 5 to 15 min. Further, the centrifugation step to remove genomic DNA and proteins from plasmid DNA extended to 10 min results in a better clarification of the solution and prevents clogging of the Zymoprep column. 4. As there is no RNA-se treatment in the Zymoprep II protocol, the determination of the concentration of the yeast plasmid by measuring A280 cannot deliver reliable values. Gel electrophoresis of the preparation reveals large quantities of genomic RNA and the concentration of the single-copy-per-cell plasmid is low. 5. Induction temperature of 37 °C can strongly impair the sensitivity of yeast cells to subsequent shear stress, such as delivered upon sorting, and diminish the viability of the cells if they are plated out afterwards. Only 10% of the expected cell count will form colonies on selective medium. 6. The use of a strain with genome-integrated vector assigning leucine prototrophy permits use of SD-CAA instead of defined synthetic double-selective media, with the same efficiency in heterodimeric Fc expression. The growth rate of the cultures as well as the viability of the cells after mechanical stress is, however, improved. 7. The percentage of cells with high autofluorescence can amount to few percent in the cultures grown and induced in synthetic double-negative medium, and without excluding these cells which also exhibit a fluorescence that they have not been stained for, they appear very similar to the desired doublepositive cells that should be selected. 8. “False positive cells” is a description of a population that appears stained with the structural or normalization marker and presents a certain top percent of antigen-positive cells, although no difference can be observed in comparison with the sample incubated without antigen. 9. Due to a relatively large ratio of the surface area to volume, medium addition is required for the small-scale HEK expression in plates, to compensate for the evaporation. For the same reason, expression should not exceed 4 days in total.

Discovery of Heterodimeric Fcab™ Fragments

157

Acknowledgments The financial support by the company F-star Therapeutics, Christian Doppler Society, Austrian Federal Ministry for Digital and Economic Affairs and the National Foundation for Research, Technology and Development is gratefully acknowledged (CD Laboratory for innovative Immunotherapeutics, grant to GWK). FB was a fellow of the international PhD program “BioToP-Biomolecular Technology of Proteins,” funded by the Austrian Science Fund (FWF) (W1224). This project was also supported by EQ-BOKU VIBT GmbH and BOKU Core Facility for Biomolecular and Cellular Analysis. ™ Fcab and mAb2 are the trademarks of F-star Therapeutics Limited (Cambridge, United Kingdom). References 1. Ma J, Mo Y, Tang M, Shen J, Qi Y, Zhao W, Huang Y, Xu Y, Qian C (2021) Bispecific antibodies: from research to clinical application. Front Immunol 12. https://doi.org/10. 3389/fimmu.2021.626616 2. Sebastian M, Kuemmel A, Schmidt M, Schmittel A (2009) Catumaxomab: a bispecific trifunctional antibody. Drugs Today 45:589– 597. https://doi.org/10.1358/dot.2009.45. 8.1401103 3. Sanford M (2015) Blinatumomab: first global approval. Drugs 75:321–327. https://doi. org/10.1007/S40265-015-0356-3 4. Kang C (2022) Mosunetuzumab: first Approval. Drugs 82:1229–1234. https://doi. org/10.1007/S40265-022-01749-5 5. Moreau P, Garfall AL, van de Donk NWCJ et al (2022) Teclistamab in relapsed or refractory multiple myeloma. N Engl J Med 387:495– 5 0 5 . h t t p s : // d o i . o r g / 1 0 . 1 0 5 6 / nejmoa2203478 6. Parums DV (2021) Editorial: Global regulatory initiatives deliver accelerated approval of the first bispecific therapeutic monoclonal antibody for advanced non-small cell lung cancer (NSCLC). Med Sci Monit 27:10.12659/ MSM.934854 7. Nicolo` M, Ferro Desideri L, Vagge A, Traverso CE (2021) Faricimab: an investigational agent targeting the Tie-2/angiopoietin pathway and VEGF-A for the treatment of retinal diseases. Expert Opin Investig Drugs 30:193–200. https://doi.org/10.1080/13543784.2021. 1879791 8. Weyand AC, Pipe SW (2019) New therapies for hemophilia. Blood 133:389–398. https:// doi.org/10.1182/blood-2018-08-872291

9. Labrijn AF, Janmaat ML, Reichert JM, Parren PWHI (2019) Bispecific antibodies: a mechanistic review of the pipeline. Nat Rev Drug Discov 18:585–608. https://doi.org/10. 1038/S41573-019-0028-1 10. Sellmann C, Doerner A, Knuehl C et al (2016) Balancing selectivity and efficacy of bispecific Epidermal Growth Factor Receptor (EGFR) × c-MET antibodies and antibody-drug conjugates. J Biol Chem 291:25106–25119. https://doi.org/10.1074/jbc.M116.753491 11. Moore PA, Zhang W, Rainey GJ et al (2011) Application of dual affinity retargeting molecules to achieve optimal redirected T-cell killing of B-cell lymphoma. Blood 117:4542– 4551. https://doi.org/10.1182/blood-201009-306449 12. Orcutt KD, Ackerman ME, Cieslewicz M, Quiroz E, Slusarczyk AL, Frangioni JV, Wittrup KD (2010) A modular IgG-scFv bispecific antibody topology. Protein Eng Des Sel 23:221. https://doi.org/10.1093/PRO TEIN/GZP077 13. Yu L, Huang N, Ge L, Sun H, Fu Y, Liu C, Wang J (2021) Structural design of tetravalent T-cell engaging bispecific antibodies: improve developability by engineering disulfide bonds. J Biol Eng 15:18. https://doi.org/10.1186/ s13036-021-00272-7 14. Wu X, Sereno AJ, Huang F et al (2015) Fab-based bispecific antibody formats with robust biophysical properties and biological activity. MAbs 7:470–482. https://doi.org/ 10.1080/19420862.2015.1022694 15. Kraman M, Faroudi M, Allen NL et al (2020) FS118, a bispecific antibody targeting lag-3 and PD-L1, enhances T-cell activation

158

Filippo Benedetti et al.

resulting in potent antitumor activity. Clin Cancer Res 26:3333–3344. https://doi.org/ 10.1158/1078-0432.CCR-19-3548 16. Gaspar M, Pravin J, Rodrigues L, Uhlenbroich S, Everett KL, Wollerton F, Morrow M, Tuna M, Brewis N (2020) CD137/OX40 bispecific antibody induces potent antitumor activity that is dependent on target coengagement. Cancer Immunol Res 8: 7 8 1 – 7 9 3 . h t t p s : // d o i . o r g / 1 0 . 1 1 5 8 / 2326-6066.CIR-19-0798 17. Lakins MA, Koers A, Giambalvo R et al (2020) FS222, a CD137/PD-L1 tetravalent bispecific antibody, exhibits low toxicity and antitumor activity in colorectal cancer models. Clin Cancer Res 26:4154–4167. https://doi.org/10. 1158/1078-0432.CCR-19-2958 18. Wozniak-Knopp G, Bartl S, Bauer A et al (2010) Introducing antigen-binding sites in structural loops of immunoglobulin constant domains: Fc fragments with engineered HER2/neu-binding sites and antibody properties. Protein Eng Des Sel 23:289–297. https:// doi.org/10.1093/protein/gzq005 19. Everett KL, Kraman M, Wollerton FPG, Zimarino C, Kmiecik K, Gaspar M, Pechouckova S, Allen NL, Doody JF, Tuna M (2019) Generation of Fcabs targeting human and murine LAG-3 as building blocks for novel bispecific antibody therapeutics. Methods 154: 60–69. https://doi.org/10.1016/j.ymeth. 2018.09.003 20. Traxlmayr MW, Faissner M, Stadlmayr G, Hasenhindl C, Antes B, Ru¨ker F, Obinger C (2012) Directed evolution of stabilized IgG1Fc scaffolds by application of strong heat shock to libraries displayed on yeast. Biochim Biophys Acta Proteins Proteomics 1824:542–549. https://doi.org/10.1016/j.bbapap.2012. 01.006 21. Hasenhindl C, Traxlmayr MW, WozniakKnopp G, Jones PC, Stadlmayr G, Ru¨ker F, Obinger C (2013) Stability assessment on a library scale: a rapid method for the evaluation of the commutability and insertion of residues in C-terminal loops of the CH3 domains of IgG1-Fc. Protein Eng Des Sel 26:675–682. https://doi.org/10.1093/protein/gzt041 22. Traxlmayr MW, Lobner E, Hasenhindl C, Stadlmayr G, Oostenbrink C, Ru¨ker F, Obinger C (2014) Construction of pH-sensitive Her2-binding IgG1-Fc by directed evolution. Biotechnol J 9:1013–1022. https://doi.org/ 10.1002/biot.201300483 23. 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

24. Sa´dio F, Stadlmayr G, Stadlbauer K, Ru¨ker F, Wozniak-Knopp G (2019) Yeast surface display and cell sorting of antigen-binding Fc fragments. Methods Mol Biol 1923:287–308. https://doi.org/10.1007/978-1-4939-90245_13 25. Wozniak-Knopp G, Stadlmayr G, Perthold JW, Stadlbauer K, Woisetschl€ager M, Sun H, Ru¨ker F (2017) Designing Fcabs: well-expressed and stable high affinity antigen-binding Fc fragments. Protein Eng Des Sel 30:567–581. https://doi.org/10.1093/protein/gzx042 26. Lobner E, Humm AS, Go¨ritzer K, Mlynek G, Puchinger MG, Hasenhindl C, Ru¨ker F, Traxlmayr MW, Djinovic´-Carugo K, Obinger C (2017) Fcab-HER2 interaction: a Me´nage a` Trois. Lessons from X-ray and Solution studies. Structure 25:878–889.e5. https://doi.org/ 10.1016/J.STR.2017.04.014 27. Lobner E, Humm A-S, Mlynek G, Kubinger K, Kitzmu¨ller M, Traxlmayr MW, Djinovic´-Carugo K, Obinger C (2017) Two-faced Fcab prevents polymerization with VEGF and reveals thermodynamics and the 2.15 Å crystal structure of the complex. MAbs 9:1088–1104. https://doi.org/10.1080/19420862.2017. 1364825 28. Birnbaum ME, Berry R, Hsiao YS et al (2014) Molecular architecture of the αβ T cell receptor-CD3 complex. Proc Natl Acad Sci U S A 111:17576–17581. https://doi.org/10. 1073/PNAS.1420936111 29. Li J, Piskol R, Ybarra R et al (2019) CD3 bispecific antibody-induced cytokine release is dispensable for cytotoxic T cell activity. Sci Transl Med 11:eaax8861 30. Nagorsen D, Baeuerle PA (2011) Immunomodulatory therapy of cancer with T cellengaging BiTE antibody blinatumomab. Exp Cell Res 317:1255–1260. https://doi.org/10. 1016/J.YEXCR.2011.03.010 31. Gleason MK, Verneris MR, Todhunter DA, Zhang B, McCullar V, Zhou SX, PanoskaltsisMortari A, Weiner LM, Vallera DA, Miller JS (2012) Bispecific and trispecific killer cell engagers directly activate human NK cells through CD16 signaling and induce cytotoxicity and cytokine production. Mol Cancer Ther 11: 2674–2684. https://doi.org/10.1158/ 1535-7163.MCT-12-0692/85205/AM/ BISPECIFIC-AND-TRISPECIFIC-KILLERCELL-ENGAGERS 32. Bogen JP, Carrara SC, Fiebig D, Grzeschik J, Hock B, Kolmar H (2021) Design of a trispecific checkpoint inhibitor and natural killer cell engager based on a 2 + 1 common light chain antibody architecture. Front Immunol 12. https://doi.org/10.3389/fimmu.2021. 669496

Discovery of Heterodimeric Fcab™ Fragments 33. Natale V, Stadlmayr G, Benedetti F, Stadlbauer K, Ru¨ker F, Wozniak-Knopp G (2022) Trispecific antibodies produced from mAb2 pairs by controlled Fab-arm exchange. Biol Chem 403:509–523. https://doi.org/10. 1515/hsz-2021-0376 34. Carter P, Ridgway JBB, Presta LG (1996) ‘Knobs-into-holes’ provides a rational design strategy for engineering antibody CH3 domains for heavy chain heterodimerization. Immunotechnology 2:73. https://doi.org/ 10.1016/1380-2933(96)80685-3 35. Von Kreudenstein TS, Escobar-Carbrera E, Lario PI et al (2013) Improving biophysical properties of a bispecific antibody scaffold to aid developability: quality by molecular design. MAbs 5:646–654. https://doi.org/10.4161/ mabs.25632 36. Gunasekaran K, Pentony M, Shen M et al (2010) Enhancing antibody Fc heterodimer formation through electrostatic steering effects: applications to bispecific molecules and monovalent IgG. J Biol Chem 285: 19637–19646. https://doi.org/10.1074/ JBC.M110.117382 37. Skegro D, Stutz C, Ollier R, Svensson E, Wassmann P, Bourquin F, Monney T, Gn S, Blein S (2017) Immunoglobulin domain interface exchange as a platform technology for the generation of Fc heterodimers and bispecific antibodies. J Biol Chem 292:9745–9759. https://doi.org/10.1074/jbc.M117.782433 38. Edelman GM, Cunningham BA, Gall WE, Gottlieb PD, Rutishauser U, Waxdal MJ (1969) The covalent structure of an entire gammaG immunoglobulin molecule. Proc Natl Acad Sci U S A 63:78–85. https://doi. org/10.1073/pnas.63.1.78 39. Atwell S, Ridgway JBB, Wells JA, Carter P (1997) Stable heterodimers from remodeling the domain interface of a homodimer using a phage display library. J Mol Biol 270:26–35. https://doi.org/10.1006/jmbi.1997.1116 40. Weaver-Feldhaus JM, Lou J, Coleman JR, Siegel RW, Marks JD, Feldhaus MJ (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

159

41. Shields RL, Namenuk AK, Hong K et al (2001) High resolution mapping of the binding site on human IgG1 for FcγRI, FcγRII, FcγRIII, and FcRn and design of IgG1 variants with improved binding to the FcγR. J Biol Chem 276:6591–6604. https://doi.org/10.1074/ jbc.M009483200 42. Leung KM, Batey S, Rowlands R et al (2015) A HER2-specific Modified Fc Fragment (Fcab) induces antitumor effects through degradation of HER2 and apoptosis. Mol Ther 23:1722– 1 7 3 3 . h t t p s : // d o i . o r g / 1 0 . 1 0 3 8 / m t . 2015.127 43. Stadlbauer K, Andorfer P, Stadlmayr G, Ru¨ker F, Wozniak-Knopp G (2022) Bispecific mAb2 antibodies targeting CD59 enhance the complement-dependent cytotoxicity mediated by rituximab. Int J Mol Sci 23:5208. https:// doi.org/10.3390/IJMS23095208 44. Pardoll DM (2012) The blockade of immune checkpoints in cancer immunotherapy. Nat Rev Cancer 12(4):252–264. https://doi.org/10. 1038/nrc3239 45. Terawaki S, Tanaka Y, Nagakura T et al (2007) Specific and high-affinity binding of tetramerized PD-L1 extracellular domain to PD-1expressing cells: possible application to enhance T cell function. Int Immunol 19: 8 8 1 – 8 9 0 . h t t p s : // d o i . o r g / 1 0 . 1 0 9 3 / INTIMM/DXM059 46. Waterhouse A, Bertoni M, Bienert S et al (2018) SWISS-MODEL: homology modelling of protein structures and complexes. Nucleic Acids Res 46:W296–W303. https://doi.org/ 10.1093/NAR/GKY427 47. Bellı´ G, Garı´ E, Piedrafita L, Aldea M, Herrero E (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 48. Orr BA, Carr LM, Wittrup KD, Roy EJ, Kranz DM (2003) Rapid method for measuring ScFv thermal stability by yeast surface display. Biotechnol Prog 19:631–638. https://doi.org/ 10.1021/bp0200797

Chapter 10 A Two-Step Golden Gate Cloning Procedure for the Generation of Natively Paired YSD Fab Libraries Lena Vollmer, Simon Krah, Stefan Zielonka, and Desislava Yanakieva Abstract In vitro antibody display libraries have emerged as powerful tools for a streamlined discovery of novel antibody binders. While in vivo antibody repertoires are matured and selected as a specific pair of variable heavy and light chains (VH and VL) with optimal specificity and affinity, during the recombinant generation of in vitro libraries, the native sequence pairing is not maintained. Here we describe a cloning method that combines the flexibility and versatility of in vitro antibody display with the advantages of natively paired VH–VL antibodies. In this regard, VH–VL amplicons are cloned via a two-step Golden Gate cloning procedure, allowing the display of Fab fragments on yeast cells. Key words Yeast surface display, Fab library, Golden Gate cloning, Restriction enzyme type IIs, Bidirectional promoter, Paired VH–VL antibody library

1

Introduction Six hypervariable loops (CDRs) form the antigen binding site (paratope) of an antibody and determine its affinity and specificity towards an antigen. Three CDRs are distributed on each of the variable domains of the heavy (VH) and light (VL) chains [1]. The conformation of the CDRs and the assembly of VH–VL play a crucial role in the binding and biophysical properties of an antibody [2]. Although remarkable progress has been made in the field of antibody cloning and in vitro display library generation [3], for a long time, recovery of cognate VH–VL sequences was limited to hybridoma technology [4] and single B cell cloning approaches [5]. This changed with the emergency of droplet-based microfluidics. Herein, single B cells are encapsulated in individual compartments, allowing for amplification and subsequent linkage of VH– VL via reverse transcription-overlap extension (RT-OE) PCR. These PCR products can be used for the generation of paired antibody display libraries. It was demonstrated that such libraries

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2681, https://doi.org/10.1007/978-1-0716-3279-6_10, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

161

162

Lena Vollmer et al.

are of higher sensitivity and specificity than equivalent randomly paired approaches [6, 7]. Single chain variable fragments (scFvs) and fragments antigen binding (Fabs) are the most commonly used antibody formats for in vitro phage [8–10] or yeast surface display (YSD) [11, 12]. While scFvs can be easily cloned in a single step into corresponding display vectors, the procedures for the generation of Fab libraries are more challenging because they necessitate the expression of two different polypeptide chains (heavy and light). In contrast, reformatting of scFvs into an IgG format often leads to stability issues and reduced affinity [13]. Therefore, Fab display seems to be more suitable for therapeutic antibody discovery [14]. Saccharomyces cerevisiae Fab libraries can be generated by co-transfection of one yeast strain with two different plasmids or by mating of two haploid yeast strains harboring plasmids for each chain [12, 15]. Using Golden Gate Cloning (GGC), one plasmid encoding for both heavy and light chain Fabs can be efficiently generated. GGC utilizes type IIs restriction enzymes capable of cleaving DNA outside their recognition site. This property enables the specific assembly of multiple fragments in one coding frame in a single step using alternating digestion and ligation cycles [16], which increases the cloning efficiency significantly [17]. To construct a natively paired Fab YSD library, we developed a two-step GGC procedure (Fig. 1) for vector-integration of the cognate antibody VH–VL sequences, generated via droplet RT-OE-PCR. Starting with a nested PCR, restriction sites are specifically introduced at the ends of framework four (FR4) of VH and VL (Fig. 1a). These restriction sites facilitate a one-direction “in frame” integration of the PCR product to a destination vector, which carries CL and CH1-Aga2p domains (Fig. 1b) [18]. Afterwards, the assembled destination plasmid is used for the transformation of Escherichia coli cells for plasmid multiplication. In the second GGC step, the linker between VH and VL is cleaved with another type IIs enzyme (SapI) (Fig. 1c) and a cassette consisting of a bidirectional galactose promoter (GAL1/ 10) and S. cerevisiae secretion leaders is inserted. Yeast cells are subsequently transformed with the two-step GGC product and selected using tryptophan autotrophy marker.

2

Materials Prepare all solutions using ultrapure sterile water (18 MΩ cm) obtained from water purification system and autoclaved for 20 min at 120 °C, as well as analytical grade reagents. Store all reagents at 4 °C or -20 °C when indicated. Follow all waste disposal regulations when disposing waste materials.

2-Step GGC for Native YSD Libraries

163

Fig. 1 Scheme of a two-step GGC procedure for the generation of VH–VL paired Fab libraries for YSD. (a) Nested PCR is performed using FR4-specific primers and a paired VH–VL PCR product. Esp3I-restriction sites and specific 4 nt overhangs are introduced to facilitate the cloning procedure. (b) In the first GGC (1.GGC) step, the paired VH–VL construct is incorporated in the pDest vector, which codes for the CL and a CH1-Aga2p fusion protein. E. coli cells are transformed with the cloned vectors, which enables the rapid isolation of large amounts of library plasmids. (c) A pEntry vector is used as a donor of a bidirectional GAL1/10 promoter and secretion peptides (app8 SP and Aga2p SP), flanked by SapI-restriction sites and complementary ligation overhangs facilitating the second Golden Gate cloning step (2.GGC). SapI-restriction sites with corresponding ligation overhangs have been introduced to the VH–VL linker during the droplet RT-OE-PCR 2.1

Strains

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

2.2

Plasmids

Architectures of the destination (pDest) and entry (pEntry) vectors are illustrated in Fig. 1b, c, respectively. pDest comprises a pYD-derived backbone, providing the E. coli and S. cerevisiae replication origins ColE1 and ARS4/CEN6, respectively. An ampicillin resistance marker (AmpR) enables the selection of transformed E. coli cells, while a tryptophan auxotrophic marker (Trp) enables the selection of yeast. Further modules of pDest are the constant domains of the heavy (CH1) and light chains (Cκ or Cƛ), as well as a Aga2p, C-terminally fused to CH1. Esp3I-restriction sites flank a stuffer sequence of about 100 bp, which serves as a cloning site for

164

Lena Vollmer et al.

the paired VH–VL construct after nested PCR. Using a Golden Gate reaction, the VH–VL amplicon is ligated in frame with the constant CL and CH1 domains. The second plasmid (pEntry) serves as a donor for a bidirectional yeast promoter (GAL1/10) and secretion signal sequences (app8 and Aga2p-SP), which are flanked by SapI-restriction sites. Additionally, the plasmid carries a kanamycin resistance gene (kanR) and a E. coli replication origin (ColE1). 2.3 Reagents for Nested PCR

Primers for the generation of a VH–VL insert for GGC (nested PCR) are shown in Tables 1 and 2. An amplification scheme is illustrated in Fig. 1a. Platinum II Hot-Start PCR Master Mix (2×) (Invitrogen) is used for the PCR reaction.

Table 1 Oligonucleotide primer mix (kappa) for nested PCR Name

Sequence (5′–3′) capitalized = primer overhang

Nested JH_01

CGAGTAGCGTATCGTCTCTTAGTtgaggagacagtgaccgtgg

Nested JH_02

CGAGTAGCGTATCGTCTCTTAGTtgaggagacagtgaccagggtg

Nested JH_03

CGAGTAGCGTATCGTCTCTTAGTtgaggagacagtgaccagggt

Nested JH_04h

CGAGTAGCGTATCGTCTCTTAGTtgaagagacartgaccattgtcc

Fab nested JK1

ACTGACGTAGCTCGTCTCTTTCTtttgatctccaccttggtccctccgccgaamgt

Fab nested JK2

ACTGACGTAGCTCGTCTCTTTCTtttgatttccaccttggtcccttggccgaacgt

Fab nested JK3

ACTGACGTAGCTCGTCTCTTTCTtttaatctccagtcgtgtcccttggccgaaggt

Fab nested JK4

ACTGACGTAGCTCGTCTCTTTCTtttgatatccactttggtcccagggccgaaagt

Fab nested JK5

ACTGACGTAGCTCGTCTCTTTCTtttgatctccagcttggtcccctggccaaaast

Table 2 Oligonucleotide primer mix (lamda) for Nested 2 PCR Name

Sequence (5′–3′) capitalized = primer overhang

Nested JH_01

CGAGTAGCGTATCGTCTCTTAGTtgaggagacagtgaccgtgg

Nested JH_02

CGAGTAGCGTATCGTCTCTTAGTtgaggagacagtgaccagggtg

Nested JH_03

CGAGTAGCGTATCGTCTCTTAGTtgaggagacagtgaccagggt

Nested JH_04h

CGAGTAGCGTATCGTCTCTTAGTtgaagagacartgaccattgtcc

Fab nested JL1

ACTGACGTAGCTCGTCTCTGTCCtaggacggtcagcttggtccctccgccgaayac

Fab nested JL2

ACTGACGTAGCTCGTCTCTGTCCgaggrcggtcagctgggtgcctcctccgaacac

Fab nested JL3

ACTGACGTAGCTCGTCTCTGTCCtaggacggtgaccttggtcccagttccgaagac

Fab nested JL4

ACTGACGTAGCTCGTCTCTGTCCgaggacggtcaccttggtgccactgccgaacac

2-Step GGC for Native YSD Libraries

2.4 Reagents for Gel Purification of VH–VL Insert

165

1. UltraPure Low-melting-point Agarose (Invitrogen). 2. Gel Red Nucleic Acid Stain (Biotium). 3. 6× Gel Loading Dye (New England Biolabs). 4. DNA ladder. 5. 1× TAE buffer. 6. UV Gel Imaging System (BioRad).

2.5 Reagents for First GGC Step

1. pDest (destination vector). 2. VH–VL insert from Subheading 3.3. 3. Esp3I (New England Biolabs). 4. T4 DNA Ligase (New England Biolabs). 5. rCutSmart buffer (New England Biolabs). 6. ATP (New England Biolabs). 7. OmniPur water, sterile, nuclease free (Sigma-Aldrich). 8. ReliaPrep DNA (Promega).

2.6 Reagents for E. coli Transformation

Clean-Up

and

Concentration

System

1. Ligated pDest with VH–VL insert from Subheading 3.4. 2. One Shot TOP10 Electrocompetent E. coli (Invitrogen). 3. Cold electroporation cuvette (0.1 cm) (Sigma-Aldrich). 4. Electroporation Gene Pulser Xcell (Bio-Rad). 5. LB-Amp media: Dissolve 5 g yeast extract, 10 g NaCl and 10 g peptone in 1 L deionized H2O prior sterilization by autoclaving. Chill medium to approximately 50 °C, then add 1 mL of sterile filtered ampicillin solution (100 mg/mL in deionized H2O). 6. LB-Amp plates: Dissolve 15 g agar, 10 g NaCl, 10 g peptone, and 5 g yeast extract in 1 L deionized H2O prior sterilization by autoclaving. Chill medium to approximately 50 °C, then add 1 mL of sterile filtrated ampicillin solution (100 mg/mL in deionized H2O) and prepare plates.

2.7 Reagents for Plasmid Preparation

1. Transformed E. coli from Subheading 3.5.

2.8 Reagents for Second GGC Step

1. pEntry (GAL1/10 entry vector).

2. GenElute Plasmid Midiprep Kit (Sigma Aldrich).

2. Purified pDest with VH–VL insert from Subheading 3.6. 3. SapI (New England Biolabs). 4. T4 DNA Ligase (New England Biolabs).

166

Lena Vollmer et al.

5. T4 Ligase buffer (New England Biolabs). 6. ATP (New England Biolabs). 7. OmniPur water, sterile, nuclease free (Sigma-Aldrich). 8. Wizard SV gel and PCR clean-up system (Promega). 2.9 Reagents for S. cerevisiae Library Generation

1. YPD media: Dissolve 10 g yeast extract, 20 g D(+)-glucose, and 20 g peptone in 1 L deionized H2O prior sterilization by autoclaving. Afterwards, add 10 mL of Penicillin-Streptomycin (10,000 Units/mL). Remove particles by sterile filtration utilizing a 0.22 μm bottle top filter. 2. LiAc buffer: 100 mM lithium acetate, 10 mM DTT (sterile filtered). 3. Electroporation buffer: 1 M Sorbitol, 1 mM CaCl2 × 2 H2O (autoclaved). 4. Cold electroporation cuvette (0.2 cm) (Sigma-Aldrich). 5. SD-Trp media: Dissolve 26.7 g minimal SD-Base in 890 mL deionized H2O prior sterilization by autoclaving. In parallel, dissolve 1.92 g Dropout-mix-Trp (Sigma-Aldrich) and 8.6 g NaH2PO4 × H2O and 5.4 g Na2HPO4 in deionized H2O and adjust the volume to 100 mL prior sterilization by autoclaving. Combine both solutions, add 10 mL of PenicillinStreptomycin (10,000 Units/mL). Remove particles by sterile filtration utilizing a 0.22 μm bottle top filter. 6. SD-Trp plates: Dissolve 23.35 g of minimal SD-Agar Base in 445 mL deionized H2O prior sterilization by autoclaving. In parallel, dissolve 0.96 g Dropout-mix-Trp (Sigma-Aldrich and 4.28 g of NaH2PO4 × H2O and 2.7 g of Na2HPO4 in deionized H2O and adjust the volume to 50 mL prior sterilization by autoclaving. Combine both solutions, add 10 mL of Penicillin-Streptomycin (10,000 Units/mL) and prepare plates. 7. SD Low-Trp medium: Dissolve 5 g dextrose and 6.7 g yeast nitrogen base (w/o amino acids) in 890 mL deionized H2O prior sterilization by autoclaving. In parallel, dissolve 1.92 g Dropout-mix-Trp (Sigma-Aldrich) and 8.56 g NaH2PO4 × H2O and 5.4 g Na2HPO4 in deionized H2O and adjust the volume to 100 mL. Sterilize all solutions by autoclaving prior combination. Afterwards, add 10 mL of Penicillin-Streptomycin (10,000 Units/mL). Remove particles by sterile filtration utilizing a 0.22 μm bottle top filter. 8. SG-Trp media: Dissolve 37 g of minimal SD-Base + Gal/Raf in 490 mL deionized H2O. In parallel, dissolve 1.92 g Dropoutmix-Trp (Sigma-Aldrich) and 8.6 g of NaH2PO4 × H2O and 5.4 g of Na2HPO4 in deionized H2O and adjust the volume to

2-Step GGC for Native YSD Libraries

167

100 mL. Furthermore, dissolve 110 g of PEG8000 in deionized H2O and adjust the volume to 400 mL. Sterilize all solutions by autoclaving prior combination. Afterwards, add 10 mL of Penicillin-Streptomycin (10,000 Units/mL). Remove particles by sterile filtration utilizing a 0.22 μm bottle top filter. 9. Yeast library freezing solution: Dissolve 0.67 g of yeast nitrogen base and 2 g of glycerol 100 mL deionized H2O prior sterile filtration of the solution.

3

Methods This section describes a GGC-based cloning procedure for generating YSD Fab libraries with natively paired VH and VL chains. The generation of the paired amplicon is described elsewhere.

3.1 Introduction of Esp3I Restriction Sites to RT-OE-PCR Product

1. Use purified PCR product as a template for a nested PCR reaction to introduce overhangs enabling the subsequent two-step GGC. 2. Mix the following components: 200 μL Platinum II Hot Start PCR Master mix, 8 μL nested primer mix (kappa or lambda), 400 ng purified droplet-PCR template, add to 400 μL with OmniPur water. 3. Perform PCR with the following program: A step of initial denaturation at 94 °C for 30 s, followed by 20 cycles of 30 s denaturation at 94 °C, 30 s primer annealing at 60 °C and 20 s elongation at 72 °C, ultimately finalized by a prolonged elongation step at 72 °C for 7 min.

3.2 Agarose Gel Electrophoresis

1. Prepare a 2% (w/v) ultra-low point melting agarose in 1× TAE buffer. Add Gel Red Nucleic Acid Stain to the melted agarose at 1× final concentration. 2. Cast gel in a tray containing a gel comb (2 lanes, 1 mm thickness). 3. Mix 400 μL sample with loading dye at 1× final concentration. 4. Load ~250 μL sample per lane along with a ladder as a reference. 5. Run gel in TAE buffer for 1–1.5 h at 110 V and 400 mA. 6. Visualize bands under UV light (Fig. 2).

3.3 Gel Extraction of the VH–VL Insert

1. Identify the correct DNA band according to the DNA ladder (approx. 750 bp).

168

Lena Vollmer et al.

Fig. 2 Agarose gel electrophoresis for analysis of the nested PCR reaction. Paired VH–VL (50 ng/μL) was amplified by a nested PCR. The PCR product was purified using gel electrophoresis and gel extraction, and used as an insert for the Golden Gate reaction

2. Cut out the desired band using a clean scalpel and dissect into little pieces. 3. Transfer the gel slices microcentrifuge tube.

into

a

weighed

1.5

mL

4. Weight the tube again to obtain the mass of the gel slices. 5. Extract the DNA using Wizard SV Gel and PCR clean-up system according to manufacturer’s instructions. 6. Elute DNA with 50 μL OmniPur water after incubation for 5 min at room temperature and centrifugation at 16,000 × g for 2 min. 7. Determine the DNA concentration. 8. Store the sample at -20 °C or proceed with the cloning procedure. 3.4

First GGC Step

This section describes the experimental procedure for “in frame” insertion of the natively paired VH–VL construct in a YSD destination vector. 1. Thaw all reagents on ice. 2. Prepare a 100 μL reaction by mixing the following reagents: 1000 ng pDest (kappa or lambda), 500 ng of template from Subheading 3.3, 200 U Esp3I, 4000 U T4 DNA Ligase, 10 μL 10× rCut Smart, 1 mM ATP, add OmniPur H2O to 100 μL. 3. Perform Golden Gate reaction in a thermocycler using the following parameters: 40 cycles of 5 min at 37 °C, 5 min at 16 °C followed by a final cycle at 37 °C for 10 min.

2-Step GGC for Native YSD Libraries

169

4. Purify GGC reaction using ReliaPrep DNA Clean-Up and Concentration System according to the manufacturer’s protocol (see Notes 1 and 2). 5. Elute DNA by adding 30 μL OmniPur water, incubation for 5 min at room temperature and centrifugation at 10000 × g for 1 min (see Note 3). 6. Purified product might be stored at -20 °C or directly used for subsequent steps. 3.5 E. coli Transformation

1. Thaw electrocompetent E. coli 10 min on ice. 2. Add 100 ng of VH:VL product from Subheading 3.4 into 1 vial of E. coli cells (2 vials in total are used). 3. Transfer cell-DNA suspension into a cold cuvette (0.1 cm). 4. Electroporate cells with 1800 V, 25 μF, and 200 Ω. 5. Immediately add 900 μL pre-warmed S.O.C medium. 6. Transfer suspension in a 1.5 mL Eppendorf tube. 7. Incubate cells for 1 h at 37 °C and 700 rpm on a thermo-block. 8. For library diversity estimation, take 100 μL of cell suspension and prepare a 1:10 serial dilution in 900 μL PBS. Plate 100 μL of diluted cell suspensions on LB-Amp agar plates (see Note 4). 9. Incubate agar plates overnight at 37 °C and count colonies on the next day (see Note 5). 10. Pellet the rest of the transformed cell suspension (~900 μL) by centrifugation for 4 min at 4000 × g and resuspend in a 1 mL sterile PBS. 11. Plate 2× 500 μL each on a large square LB-Amp agar plate. 12. Grow cells overnight at 37 °C. 13. On the next day, harvest cells using 20 mL PBS per plate and collect the cell suspension in a 50 mL tube. 14. To check the efficiency of the cloning procedure, perform a colony-PCR (see Note 6).

3.6 Plasmid Preparation

1. Centrifuge the cell suspension from Subheading 3.5 at 5000 × g for 10 min and remove supernatant. 2. Isolate plasmids from E. coli using GeneElute Plasmid DNA Midiprep Kit following manufacturer’s protocol and elute the plasmid DNA in 1 mL OmnuPur water. 3. Combine eluates from all tubes and determine DNA concentration and purity on Nanodrop.

170

Lena Vollmer et al.

3.7 Second GGC Step

This section describes the second cloning step for integration of a bidirectional yeast promoter and secretion leader sequences into the linker region between the VH and VL domains in the previously generated YSD plasmid. 1. Thaw all reagents on ice. 2. Prepare 12× 100 μL reactions by mixing the following reagents: 12 μg isolated plasmid from Subheading 3.6, 16.8 μg pEntry vector, 1200 U SapI, 30,000 U T4 DNA Ligase, 120 μL 10× rCut Smart, 1 mM ATP, add OmniPur H2O to 1200 μL. 3. Transfer reaction mix to 12 PCR tubes (100 μL/tube). 4. Perform Golden Gate reaction in a thermocycler using the following parameters: 40 cycles of 5 min at 37 °C, 5 min at 16 °C followed by a final cycle at 37 °C for 10 min. 5. Pool samples and purify GGC reaction using Wizard SV Gel and PCR clean-up system according manufacturer’s instructions—use two columns (see Notes 1 and 2). 6. Elute DNA by adding 30 μL OmniPur water to each column, incubation for 5 min at room temperature and centrifugation at 10000 × g for 1 min (see Note 3). 7. Purified product might be stored at -20 °C or directly used for subsequent steps.

3.8 Yeast Transformation for Library Generation

The experimental procedure for library generation follows the protocol from Benatuil and colleagues for improved S. cerevisiae yeast transformation [19]. All centrifugation steps to pellet yeast cells are performed for 3 min at 4000 × g. 1. Grow EBY100 in 50 mL YPD media at 120 rpm and 30 °C overnight to reach a stationary phase. 2. Inoculate 300 mL fresh YPD media with the overnight culture to an OD600 of about 0.5 (see Note 7). 3. Incubate cells at 120 rpm and 30 °C until the OD600 value reaches about 1.6–1.9. 4. Pellet cells by centrifugation and decant supernatant. 5. Wash cells twice with 150 mL ice-cold water and once with 150 mL ice-cold electroporation buffer. 6. Resuspend cells in 60 mL LiAc-buffer and incubate at 30 °C and 120 rpm for 30 min in a baffled flask. 7. Pellet cells by centrifugation and wash once with 150 mL ice-cold electroporation buffer. 8. Resuspend pellet in electroporation buffer to a final volume of approximately 3 mL. Using 400 μL electro-competent EBY100 per cuvette, conduct six electroporation reactions.

2-Step GGC for Native YSD Libraries

171

Table 3 Primers for colony-PCR Name

Sequence (5′–3′)

pDest CH1 seq

ACCAAACAACCCAAAGCAGC

pDest Cl seq

ACACCAGGGTGGCCTTGTTG

pDest Ck seq

GACAAACAACAGAAGCAGTAC

9. Mix 2.5 mL electro-competent cells with the purified GGC reaction from Subheading 3.7. 10. Fill cell-DNA mix in ice-cold 0.2 cm electroporation cuvettes (400 μL/cuvette) and perform electroporation reaction with 2.5 kV, 25 μF, and 200 Ω. Transfer cells immediately after reaction into 16 mL of a YPD medium and 1 M sorbitol mixture (1:1 ratio). 11. Incubate the cells at 120 rpm for 1 h at 30 °C (see Note 8). 12. Pellet cells by centrifugation and resuspend cell pellet in 10 mL PBS. 13. Perform a serial 1:10 dilution using 100 μL cell suspension and 900 μL PBS for estimation of number of transformants (see Note 9). Plate 100 μL cell suspension from each dilution on a SD-Trp agar plate. Incubate for 72 h at 30 °C and count the number of transformants to calculate the maximal library size (see Note 10). 14. Transfer remaining cells in 500 mL SD-Trp media and incubate at 120 rpm for 48 h at 30 °C. 15. Transfer library to SD Low-Trp medium at an OD600 of ~1.0. For proper oversampling, transfer at least a ten-fold excess of cells as calculated by dilution plating. 16. Incubate for another 48 h at 120 rpm and 30 °C. 17. The final yeast display library can be directly induced for FACS using SG-Trp media for 24–48 h and cryo-preserved for longterm storage at -80 °C (see Note 11). 18. To access the quality of the yeast display library, at least 100 clones should be sequenced using primers from Table 3 (authors recommendation).

4

Notes 1. Prior to clean up, incubate the PCR reaction at 65 °C for 10 min – this facilitates the dissociation of DNA-binding proteins (etc. T4 ligase) and increses the plasmid DNA yield.

172

Lena Vollmer et al.

2. Dry the column by vacuum centrifuging for 3 min. 3. For better yield, add eluate second time on the same column and centrifuge again at 10000 × g for 1 min. 4. Plate 4 LB-Amp agar plates with 10-5–10-8 dilutions. 5. Diversity should lie in the range of 107–108 transformants. 6. Perform a colony-PCR as follows: pick single E. coli clones and resuspend each clone in 30 μL water. Heat suspension at 98 °C for 5 min and centrifuge briefly. Take 2 μL of the lysate as template in a colony-PCR reaction. Use primer annealing to the vector backbone (pDest) (see Table 3). Analyze PCR product by agarose gel electrophoresis. 7. We assume that OD600 = 1 corresponds to approx. 1 × 107 yeast cells/mL. 8. Approximately 1 × 107 transformants can be obtained per electroporation reaction. To ensure coverage of larger diversities, several electroporation reactions can be parallelized. 9. Plate four SD-Trp agar plates with 10-5–10-8 dilutions. 10. Diversity should be in the range of 107–108 transformants. 11. Pellet 1010 yeast cells per vial and resuspend in 2 mL freezing solution. Shock-freeze yeast by directly storing at -80 °C. References 1. Charles J, Janeway A, Travers P, Walport M, Shlomchik MJ (2001) The structure of a typical antibody molecule. In: The immune system in health and disease, 5th edn. Garland Science 2. Ferna´ndez-Quintero ML et al (2020) Antibodies exhibit multiple paratope states influencing VH–VL domain orientations. Commun Biol 3(1):1–14 3. Hoogenboom HR (2005) Selecting and screening recombinant antibody libraries. Nat Biotechnol 23(9):1105–1116 4. Chen Y, Kim SH, Shang Y, Guillory J, Stinson J, Zhang Q, Ho¨tzel I, Hoi KH (2018) Barcoded sequencing workflow for high throughput digitization of hybridoma antibody variable domain sequences. J Immunol Methods 455:88–94 5. Tiller T, Meffre E, Yurasov S, Tsuiji M, Nussenzweig MC, Wardemann H (2008) Efficient generation of monoclonal antibodies from single human B cells by single cell RT-PCR and expression vector cloning. J Immunol Methods 329(1–2):112–124 6. Rajan S et al (2018) Recombinant human B cell repertoires enable screening for rare, specific, and natively paired antibodies. Commun Biol 1(1):1–8

7. Adler AS et al (2018) A natively paired antibody library yields drug leads with higher sensitivity and specificity than a randomly paired antibody library. mAbs 10(3):431–443 8. Vaughan TJ et al (1996) Human antibodies with sub-nanomolar affinities isolated from a large non-immunized phage display library. Nat Biotechnol 14(3):309–314 9. Ledsgaard L, Kilstrup M, Karatt-Vellatt A, McCafferty J, Laustsen AH (2018) Basics of antibody phage display technology. Toxins 10(6):236 10. Almagro JC, Pedraza-Escalona M, Arrieta HI, Pe´rez-Tapia SM (2019) Phage display libraries for antibody therapeutic discovery and development. Antibodies 8(3):44 11. 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 12. Baek D-S, Kim Y-S (2014) Construction of a large synthetic human Fab antibody library on yeast cell surface by optimized yeast mating. J Microbiol Biotechnol 24(3):408–420 13. Steinwand M, Droste P, Frenzel A, Hust M, Du¨bel S, Schirrmann T (2014) The influence of antibody fragment format on phage display

2-Step GGC for Native YSD Libraries based affinity maturation of IgG. MAbs 6(1): 204–218 14. Sivelle C, Sierocki R, Ferreira-Pinto K, Simon S, Maillere B, Nozach H (2018) Fab is the most efficient format to express functional antibodies by yeast surface display. MAbs 10(5):720–729 15. Weaver-Feldhaus JM, Lou J, Coleman JR, Siegel RW, Marks JD, Feldhaus MJ (2004) Yeast mating for combinatorial Fab library generation and surface display. FEBS Lett 564(1–2): 24–34 16. Engler C, Kandzia R, Marillonnet S (2008) A one pot, one step, precision cloning method

173

with high throughput capability. PLoS One 3(11):e3647 17. Engler C, Marillonnet S (2013) Combinatorial DNA assembly using Golden Gate cloning. In: Synthetic biology. Humana Press, Totowa, pp 141–156 18. Boder ET, Wittrup KD (1997) Yeast surface display for screening combinatorial polypeptide libraries. Nat Biotechnol 15(6):553–557 19. 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(4):155–159

Chapter 11 Single-Cell B-Cell Sequencing to Generate Natively Paired scFab Yeast Surface Display Libraries Nathaniel Pascual, Theodore Belecciu, Sam Schmidt, Athar Nakisa, Xuefei Huang, and Daniel Woldring Abstract The immune cell profiling capabilities of single-cell RNA sequencing (scRNA-seq) are powerful tools that can be applied to the design of theranostic monoclonal antibodies (mAbs). Using scRNA-seq to determine natively paired B-cell receptor (BCR) sequences of immunized mice as a starting point for design, this method outlines a simplified workflow to express single-chain antibody fragments (scFabs) on the surface of yeast for high-throughput characterization and further refinement with directed evolution experiments. While not extensively detailed in this chapter, this method easily accommodates the implementation of a growing body of in silico tools that improve affinity and stability among a range of other developability criteria (e.g., solubility and immunogenicity). Key words Single-cell sequencing, Yeast surface display (YSD), Antibodies, Mouse immunization, scFab library generation, B-cell receptors

1 1.1

Introduction Background

Single-cell sequencing technologies offer researchers a versatile toolkit to identify the genetic differences between individual cells in a population, giving us a better understanding of the role of individual cells within their populations and cellular evolutionary relationships [1, 2]. In the context of immunology, single-cell RNA sequencing (scRNA-seq) has offered significant insights into how both adaptive and innate immune cells develop their individual responses to antigens [1–4]. With advances in single-cell sequencing technologies that allow for the pairing of variable heavy and variable light domain sequences of B-cell receptors and the pairing of each receptor to specific antigens, there has been an accelerated pace in applying single-cell immune profiling technologies (e.g., LIBRA-seq) toward theranostic applications [5, 6]. To leverage these insights in this method, scRNA-seq will be used to design

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2681, https://doi.org/10.1007/978-1-0716-3279-6_11, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

175

176

Nathaniel Pascual et al.

antibodies for further characterization and refinement using information derived from immune-adapted B cells. Using the B-cell receptor sequences determined from scRNAseq, computational tools can be implemented to further characterize and refine monoclonal antibody designs to reduce the resource burdens of laboratory expression and analysis (see Note 1). While a wide variety of tools are available, this method provides a simple and straightforward protocol for translating specific clonotypes into yeast surface display (YSD) libraries. Despite the availability of many surface display techniques such as phage display, mammalian cell surface display, and bacterial display, we chose yeast surface display for our work for ease of cloning, applicability to flow cytometry, eukaryotic posttranslational modifications, and relative affordability [7]. The phenotype–genotype linkage common to cell display technologies allows for the effect on the fitness of specific mutations to be observed [8]. 1.2 Overview of Method

Here, we report a procedure for generating YSD libraries to rapidly characterize monoclonal antibodies designed from the single-cell immune profiling data produced from immunized mice (Fig. 1). After immunizing mice against a target, B cells can be isolated from

Fig. 1 Method overview. After immunizing mice against the antigen (a), B-cell receptor sequences can be isolated via scRNA-seq (b). After reducing the total number of clonotypes with the use of in silico assays (c), the mAbs can be reformatted as humanized scFabs that can be expressed on the surface of yeast. Successive rounds of cell sorting can then be implemented to characterize and improve binding affinity (e)

YSD Antibody Library from Single-Cell Sequencing

177

harvested spleens. Using 10× Genomics’ Chromium Platform to isolate single cells onto gel beads to prepare sequencing libraries, the resulting cDNA libraries will be enriched with mouse B-cellspecific primer mixes. After sequencing the enriched cDNA libraries, we outline a simplified pipeline to express single-chain antibody fragments (scFabs) for YSD expression from the raw sequencing data.

2

Materials

2.1 Mouse Immunization

1. BALB/c mice (at least 8 mice). 2. 3.3 nM antigen solution. This can include any antigen you want to produce a library of antibodies against. 3. 1 mg/mL monophosphoryl lipid A (MPLA) in DMSO. 4. 1× phosphate-buffered saline (PBS) solution, pH 7.4. 5. 0.05 M carbonate–bicarbonate buffer: 0.05 M NaHCO3/ Na2CO3, 0.02% NaN3, adjust to pH 9.6. 6. Spleen isolation solution: 10% fetal bovine serum (FBS), 10 mM HEPES, 50 μM mercaptoethanol, 2 mM L-glutamine, 100 μg/mL penicillin G, and 1% penicillin–streptomycin, suspended in RPMI 1640 culture medium. 7. Bovine serum albumin (BSA). 8. 1 mL syringe with 23-gauge needle. 9. 5 mL syringe 10. 0.1% BSA/PBS solution. 11. 10× PBS/0.5% Tween-20. 12. BSA antigen conjugate. These will differ depending on the antigen of interest, so these may have to be custom-made in the laboratory. 13. HRP-conjugated goat anti-mouse IgG (Jackson ImmunoResearch Laboratory, ID: AB_10015289). 14. 0.5 M H2SO4. 15. 96-well, high-binding flat-bottom microtiter plates. 16. TMB solution: 5 mg 3,3′,5,5′-tetramethylbenzidine (TMB) in 2 mL DMSO and 18 mL of citric acid buffer containing 20 μL of H2O2. 17. Microplate reader. 18. 70% ethanol. 19. Scalpel. 20. CO2 tank. 21. Sealed chamber (i.e., euthanasia box).

178

Nathaniel Pascual et al.

22. Forceps. 23. Petri dishes. 24. RPMI 1640 culture medium. 25. 10% fetal bovine serum (FBS). 26. 10 mM HEPES. 27. 50 μM mercaptoethanol. 28. 2 mM L-glutamine. 29. 100 μg/mL penicillin G. 30. 1% penicillin–streptomycin solution. 31. 50 mL centrifuge tubes. 32. Pipettes. 33. Centrifuge. 34. 1× RBC lysis buffer. 35. 70 μm Nylon mesh. 36. 0.4% trypan blue solution. 37. Hemocytometer. 38. BioLegend™ Biotin-Antibody Cocktail. 39. 5× MojoSort™ Buffer, Cat. #: 480017. 40. MojoSort™ Magnet, Cat. #: 480019. 41. 14 mL polypropylene tubes. 42. Streptavidin nanobeads. 43. Biotinylated BSA antigen conjugate. Like the BSA antigen conjugate, this will differ depending on the antigen of interest and may need to be synthesized in the laboratory. 44. Dry ice. 45. 10% DMSO. 2.2 Single-Cell Sequencing

1. Dual Index Chromium Next GEM Single Cell 5′ Kit v2, PN-1000263. 2. 10× Genomics Library Construction Kit, PN-1000190. 3. Chromium Next GEM Single Cell 5′ Gel Bead Kit v2, 16 rxns, PN-1000264. 4. Chromium Single Cell Mouse BCR Amplification Kit, 16 rxns, PN-1000255. 5. Dynabeads MyOne Silane, PN-2000048. 6. Chromium Next GEM Chip K Single Cell Kit, 16 rxns, PN-1000287. 7. Dual Index Kit TT Set A, 96 rxns, PN-1000215. 8. 10× Vortex Adapter, PN-330002.

YSD Antibody Library from Single-Cell Sequencing

179

9. Chromium Next GEM Secondary Holder, PN-3000332. 10. 10× magnetic separator, PN-230003. 11. Thermal cycler. 12. PCR tubes (10× Genomics recommends Eppendorf, USA Scientific, or Thermo Fisher Scientific PCR 8-tube strips). 13. 1.5 mL tubes. 14. 2.0 mL tubes. 15. Nuclease-free water. 16. Pure ethanol. 17. SPRIselect Reagent Kit, Cat No.: B23318 (see Note 2). 18. 10% Tween-20. 19. 50% (v/v) glycerin (glycerol) in aqueous solution. 20. Qiagen Buffer EB, Cat. #: 19086. 21. Vortex mixer. 22. Divided polystyrene reservoirs. 23. Mini-centrifuge. 24. ThermoMixer. 25. Thermoblock for 1.5 mL tubes. 26. Agilent 4200 TapeStation (see Note 3). 27. NovaSeq 300 Cycle Reagent Kit, v 1.5. 28. RT GEM Master Mix: 18.8 μL RT reagent B, 7.3 μL poly-dT RT primer, 1.9 μL reducing agent B, and 8.3 μL RT enzyme C. 29. Dynabeads Cleanup Mix: 5 μL nuclease-free water, 182 μL cleanup buffer, 8 μL Dynabeads MyOne Silane, and 5 μL reducing agent B. 30. Dynabeads Elution Solution 1: 98 μL buffer EB, 1 μL 10% Tween-20, and 1 μL reducing agent B. 31. cDNA Amplification Mix: 50 μL Amp Mix and 15 μL cDNA primers. 32. V(D)J Amplification 1 Reaction Mix: 50 μL Amp Mix and 48 μL B Cell Mix 1 v2. 33. V(D)J Amplification 2 Reaction Mix: 50 μL Amp Mix and 15 μL B Cell Mix 2 v2. 34. Fragmentation Mix: 15 μL nuclease-free water, 5 μL fragmentation buffer, and 10 μL fragmentation enzyme. 35. Adaptor Ligation Mix: 20 μL ligation buffer, 10 μL DNA ligament, and 20 μL Adaptor Oligos.

180

Nathaniel Pascual et al.

2.3 Lead Clone Characterization and Refinement

1. Workstation with Linux/CentOS. 2. MATLAB R2022a. 3. bcl2fastq2 (v. 2.20). 4. Cellranger multi (v. 6.1.2).

2.4 Library Generation and YSD Expression

1. Primers specific to VH and VL inserts of interest or VH and VL domain DNA sequences can be synthesized from your vendor of choice. 2. Standard cloning reagents, instruments, and materials (e.g., Phusion Polymerase, HF buffer, nuclease-free water, and thermal cyclers). 3. pCTCon2-based yeast expression vector [9–12] (Addgene: 41843). 4. NotI-HF. 5. NheI-HF. 6. 10× CutSmart Buffer. 7. PCR/DNA Cleanup Kit. 8. Gel Extraction Kit. 9. Luria Broth (LB) Media. 10. LB agar. 11. 100-mm petri dish. 12. Ampicillin. 13. Plasmid Preparation Kit. 14. 10× T4 DNA ligase buffer. 15. T4 DNA ligase. 16. NEB5α competent cells. 17. SOB or SOC media. 18. EBY100 yeast. 19. YPD growth media: 10.0 g/L yeast extract, 20.0 g/L bacto peptone, and 20.0 g/L dextrose. 20. YPD agar: 10.0 g/L yeast extract, 20.0 g/L bacto peptone, 20.0 g/L dextrose, and 18 g/L bacto agar. 21. Buffer E: 1.0 M sorbitol powder, 1.0 mM CaCl2, filter sterilized. 22. Lithium acetate solution: 100 mM lithium acetate, 10 mM Tris-HCl buffer, pH 7.5, 1 mM EDTA, 30% PEG 8000, and 10 mM DTT. 23. Electroporation cuvette. 24. Electroporator. 25. SD-/Trp growth media.

YSD Antibody Library from Single-Cell Sequencing

181

26. SD/-Trp agar. 27. 1-L baffled flask. 28. EZ-Yeast Transformation Kit. 29. Shaking incubator.

3

Methods

3.1 Mouse Immunization

To obtain the B cells that will undergo single-cell sequencing, mice will need to be immunized with the antigen of interest (Fig. 2). We use BALB/c mice, which have demonstrated faster spleen B-cell proliferation and a better overall humoral immune response than the commonly used C57BL/6 strain [13]. However, any mouse strain used in immunological research should work with this method. Since this protocol is meant for a variety of possible antigens, the antibody titer of the mice has to be determined via indirect ELISA after the immunization [14]. The day of peak antibody titer is determined by pooling the sera of six immunized mice after immunization. It is not absolutely necessary to determine the exact day of peak antibody titer for an individual mouse; we only want to approximate it so we can obtain a sufficient number

Fig. 2 Antigen-specific B cells are isolated from the spleens of immunized mice and prepared for single-cell sequencing. Six mice are first immunized (a), and the day of peak antibody titer for the antigen of interest is determined by pooling their sera and performing an ELISA over time (b). Two separate mice are then immunized, and their spleens are then extracted to harvest B cells and prepare a cell suspension (c–e). After lysing red blood cells, the B cells are isolated via negative and positive selection and sent to single-cell sequencing (f–g). Cell viability and cell count were verified by hemocytometry before preparing sequencing libraries in Subheading 3.3

182

Nathaniel Pascual et al.

of viable antigen-specific B cells for the 10× Genomics single-cell sequencing workflow (10,000 cells, 90% viability, and 1000 cells per μL). After the day of peak antibody titer is determined for your antigen of interest, it is recommended to immunize two separate mice for spleen extraction. A sufficient number of B cells may be obtained with one spleen, but mouse spleens can vary quite significantly in size. The single-cell suspension obtained from the spleens then undergoes RBC lysis and negative selection to remove cells other than B cells. Positive selection then has to be performed in order to isolate only the B cells that bind the antigen of interest. In this protocol, we use biotinylated BSA conjugated to the antigen and streptavidin nanobeads for positive selection. The isolated B cells are then sequenced through the 10× Genomics workflow in Subheading 3.2. 3.1.1 Mouse Immunization

1. Prepare 6 BALB/c mice to inject with the antigen of interest. 2. Prepare 200 μL of vaccine for each injection, containing 3.3 nM antigen solution, 20 μL of MPLA solution, and 172.4 μL PBS. 3. The day before inoculating the mice, draw approximately 40 μL of blood from each of the mice and pool the sera to run an indirect ELISA (as a starting point for the antibody titer) (see Note 4). 4. Inject the vaccine subcutaneously under the scruff of the mice. 5. Perform a blood draw the next week, and then, every other week afterward (i.e., perform blood draws on days 7, 21, 35, and 49). Perform two booster vaccinations on all of the mice on the weeks after the second and third blood draws (i.e., on days 14 and 28). 6. To run the indirect ELISA, apply the following procedure adopted from Wu et al. 2021 [14]: (a) Add 10 μg/mL of BSA antigen conjugate in 0.05 M carbonate–bicarbonate buffer [0] to the 96-well plates. Incubate these plates overnight at 4 °C. (b) Wash these plates with PBS/0.5% Tween-20 and then add 100 μL of 1% BSA/PBS to each well at 22 °C. Wait 1 h for the block to occur. (c) Wash the plates with PBS/0.5% Tween-20 and then incubate them with mouse sera dilutions in 0.1% BSA/PBS (add 100 μL per well of this diluted sera and use three wells per dilution). Incubate the plates at 37 °C. (d) Let the plates incubate for 2 h and then wash them with PBS/0.5% Tween-20. (e) Add a 1:2000 dilution of the HRP-conjugated goat antimouse IgG in 100 μL of 0.1% BSA/PBS to the wells and incubate the plates for 1 h at 37 °C.

YSD Antibody Library from Single-Cell Sequencing

183

(f) Wash the plates with PBS/0.5% Tween-20. To each of the plates, add 20 mL of the TMB solution. (g) After about 15 min of letting the color form, add 50 μL 0.5 M H2SO4 to quench. (h) Measure the readout at 450 nm. Use regression analysis to determine the titer (see Note 5). Perform all samples in triplicate. 7. Plot the antibody titer from the six mice over time to determine the day when levels reach their peak. 3.1.2 Harvesting B-Cells from Immune-Adapted Mice

1. Once the day of peak antibody titer is determined, obtain two separate BALB/c mice and inoculate them as described above, administering booster injections on days 14 and 28 after the initial inoculation. There is no need to perform blood draws for ELISA this time since the day of peak antibody titer is known. 2. Euthanize the mice with CO2 on the day determined to have peak antibody titer (see Note 6). 3. Sterilize the euthanized mice with 70% ethanol. 4. Create incisions on the left ventral sides of the mice (approximately 2.5 cm in length) between the last rib and the hip joint using sterilized surgical scissors. Cut the skin, but do not cut the peritoneal wall yet. 5. Obtain two sterile petri dishes and fill them each with 10 mL of spleen isolation solution (see Subheading 2.1). 6. Using a second sterilized pair of surgical scissors, create 2.5 cm long incisions in the exposed peritoneal walls with the same orientations as the previous skin incisions. 7. Use sterilized forceps to pull the spleens through these incisions. Cleanly remove the connective tissue from the spleen using forceps or scissors and place the spleens in the prepared petri dishes. 8. Using the top side of a 5 mL syringe plunger, grind the spleens gently until they become a homogeneous slurry (see Note 7). 9. Pipette the medium along with the spleen contents from the petri dishes to a single 50 mL centrifuge tube. 10. Add 2 mL of the RPMI 1640 culture medium back into each of the petri dishes to rinse them and then add the rinse to the centrifuge tube. 11. Centrifuge the contents of the 50 mL tube containing the spleens for 5 min at 1600 RPM and 4 °C. 12. Aspirate the supernatant from the tube and then loosen the cell pellets by gently flicking the tube. 13. Add 5 mL of red blood cell lysis buffer to the centrifuge tube. Incubate it on ice while gently shaking for 5 min.

184

Nathaniel Pascual et al.

14. Dilute the contents of the centrifuge tube with PBS until the 50 mL mark is reached. Centrifuge for 5 min at 1600 RPM and 4 °C. 15. Aspirate the supernatant again and loosen the cell pellets as mentioned in step 12 of Subheading 3.1.2. Add 20 mL of the RPMI 1640 culture medium to resuspend the cell pellets. 16. Acquire another 50 mL centrifuge tube and a 70 μm nylon mesh. Filter the contents of the original tube containing the cell pellets by pouring it into the new tube through the nylon mesh. 17. Count the cells with a hemocytometer (or cell counter) and use 0.4% trypan blue solution to check cell viability. Ensure that the cell viability is above 90%. You will need to immunize another pair of mice and extract their spleens on the day of peak antibody titer if the cell viability is less than 90%. 3.1.3 Negative Selection for B-Cell Isolation

1. Centrifuge the cell suspension for 5 min at 1600 RPM and 4 °C. Resuspend in an appropriate volume of MojoSort™ buffer. Adjust the cell concentration to 1 × 108 cells per mL using the buffer. 2. Perform a slightly modified version of the MojoSort™ protocol for negative selection, starting with the fourth step in the MojoSort™ protocol, but begin by aliquoting 300 μL of cell suspension instead of the 100 μL specified [15]. This is to ensure a sufficient number of cells for the sequencing later in Subheading 3.2. (a) Aliquot the cell suspension into a new 14 mL polypropylene tube (do not use a standard centrifuge tube). (b) Instead of adding 10 μL of the Biotin-Antibody cocktail and 10 μL of BioLegend™ streptavidin nanobead solution, as specified in the protocol, add 30 μL of each. Do not discard the supernatant—these are your cells of interest. (c) Conduct everything else as specified in the protocol. Negative selection should remove most cells which are not B cells. 3. Perform a cell count and viability check on the supernatant cells obtained from the negative selection using a hemocytometer and 0.4% trypan blue. Ensure well over 10,000 cells are present with a viability of over 90%. (a) If fewer than 10,000 cells are present, repeat step 17 by aliquoting another 300 μL of cell suspension. (b) If the viability is less than 90%, another pair of mice needs to be immunized and have their spleens extracted on the day of peak antibody titer.

YSD Antibody Library from Single-Cell Sequencing 3.1.4 Positive Selection for B-Cell Isolation

185

1. Perform a modified version of the available MojoSort™ positive selection protocol beginning with its fourth step, but do not aliquot 100 μL of cell suspension as mentioned. Simply use the supernatant from the negative selection [16]. (a) Also, do not use the biotin-conjugated antibody supplied in the MojoSort™ kit, since your desired cells are only the B cells that bind your antigen of interest. Instead, use your antigen of interest conjugated to biotinylated BSA (see Subheading 2.1). (b) The volume and concentration of biotinylated BSA antigen solution added should be determined through titration for optimal results. (c) Add 10 μL of BioLegend™ Streptavidin Nanobead solution for every 107 cells. (d) Follow everything else as listed in the protocol. 2. Ensure that the buffer of the cells contains no calcium or magnesium. This is needed for the 10× Genomics single-cell sequencing. If you are using a buffer with those ingredients, resuspend your cells in a buffer lacking them, such as a PBS + BSA buffer. 3. Perform a cell count and viability check using a hemocytometer and 0.4% trypan blue. Ensure you have at least 10,000 cells, with a concentration of 1000 cells per microliter and a viability of above 90%. (a) If you meet the viability but not the cell count criterion, start by aliquoting another 300 μL of the cell suspension obtained prior to negative selection (Subheading 3.1.3, step 2). (b) If the viability criterion is not met, start by immunizing another pair of mice. 4. Add an appropriate amount of 10% DMSO to your cells as an anti-freeze agent and pack them in dry ice for shipping and storage prior to single-cell sequencing.

3.2 Single-Cell Sequencing of BCRs

Single-cell libraries for sequencing were generated using 10× Genomics’ Chromium Single Cell Immune Profiling platform (Fig. 3) (see Note 8 for alternatives). Sequencing was achieved using Illumina NovaSeq 6000 SP platform [17–19]. Unless otherwise stated, the manufacturer’s recommended protocols throughout this section were implemented. First, gel beads-in-emulsion (GEMs) are generated to separate the collection of thousands of cells into pairs of single cells with unique barcoded gel beads suspended in partitioning oil. The vast majority of gel beads are unpaired to ensure only a single cell is paired to each gel bead barcode. Following the generation of GEMs, reverse transcription

186

Nathaniel Pascual et al.

Fig. 3 Sequencing library generation with 10× Genomics Chromium Platform. (a) Suspended cell mixture produced in Subheading 3.2 is separated into individual cell–gel bead complexes using 10× Genomics Chip K Platform. (b) cDNA from each cell is produced after cell lysis using the primers fixed on the surface of each gel bead. (TSO: template switch oligo; UMI: unique molecular identifier; BC: 10× Genomics cell-specific bar code; R1: read 1 primer) (c) BCR sequences can then be enriched from the pool of cDNA using mouse-specific B-cell primer mixes. (d) Finally, the cDNA libraries are prepared for sequencing by adding Illumina-specific read 2 sequence and sampling indexes. (GEX: gene expression library; V(D)J: B-cell cDNA library)

(RT) is performed to form cDNA from the lysed cells. The RT primer consists of Illumina’s read 1 (R1) sequencing primer, a cellspecific 10× barcode sequence, a transcript-specific unique molecular identifier (UMI), and a common template switch oligo (TSO) (see Fig. 3b). Reagents and primers from RT are removed from cDNA generated in this step using silane magnetic beads. Using mouse B-cell-specific primers that bind to the constant regions of the BCR, the cDNA libraries can be enriched. V(D)J sequencing libraries for BCRs were generated to include Illumina’s read 2 primer sequences through adaptor ligation after performing end repair and A-tailing on the enriched cDNA samples. Furthermore, using primers P5 and P7, the sampling indexes i5 and i7 were added to each molecule. Samples were cleaned up using SPRIselect Reagent Kit. Gene expression (GEX) sequencing libraries are then generated in a similar manner to the previous section. The resulting libraries must then be pooled for sample processing using Illumina’s NovaSeq platform, following the manufacturer’s protocol for the XP workflow. 3.2.1 Prepare GEM Reaction Mix

1. Prepare the RT GEM master mix according to the number of samples being prepared (see Note 9). 2. Transfer 36.3 μL of RT GEM master mix into each tube of a PCR 8-tube strip on ice or cooling tube rack. 3. Assemble the Chromium Next GEM Chip K according to the manufacturer’s instructions. 4. Based on the cell count and cell stock concentrations, transfer the appropriate volume of cell suspension and nuclease-free water to the master mix (see Note 10). Each tube should have 75 μL in total. Pipette to mix.

YSD Antibody Library from Single-Cell Sequencing 3.2.2 Load the Chromium Next GEM Chip K

187

1. For each unused well, add 50% glycerol solution. 45 μL for row 3, 50 μL for row 2, and 70 μL for row 1. 2. Pipette 70 μL of master mix with cells to each well in row 1. Ensure that no bubbles are introduced. 3. Prepare the gel beads by assembling the 10× vortex adapter to the tube strip holder with the gel bead strip. Vortex for 30 s and then centrifuge the gel bead strip for 5 s. Ensure there are no bubbles in the tubes and secure the gel bead strip back in the holder with the lid. 4. After puncturing the foil seal of the gel bead tube, load row 2 with 50 μL of gel beads. Wait 30 s. 5. Add 45 μL of partitioning oil into row 3. 6. Attach gasket and add chip K to chromium controller or chromium X/iX. Run the samples. While samples are being processed, place a PCR tube strip on ice. 7. After the approximately 18 min cycle, remove the chip and discard the gasket. Ensure the volumes in rows 1 and 2 are equal to check if there are any clogs. During the cycle, place a PCR 8-tube strip on ice. 8. Slowly transfer 100 μL of GEMS from row 3 to the PCR tube strip. The liquid should appear as a uniform, opaque solution across all channels. 9. Transfer the tube strip to a thermal cycle, and run the samples with the following protocol.

3.2.3 Post-GEM-RT Cleanup Using Dynabeads

Lid temperature

Reaction volume

Run time

53 °C

125 μL

~55 min

Step

Temperature

Time

1

53 °C

45 min

2

85 °C

5 min

3

4 °C

Hold

1. Add 125 μL of the recovery agent (dyed pink) to each sample at room temperature. Allow the mixture to separate into two phases for 2 min (see Note 11). 2. Remove 125 μL of the recovery agent/oil phase (dyed pink) from the bottom of the tube without aspirating the sample within the clear aqueous phase (see Note 12). 3. Prepare the Dynabeads Cleanup Mix according to adjusting the total volume to the number of samples being prepared: (a) Before adding the silane beads, vortex thoroughly for at least 30 s. If clumps remain, pipette to mix suspension.

188

Nathaniel Pascual et al.

4. Vortex Dynabeads Cleanup Mix and add 200 μL of the mix to each sample. Pipette to mix. Incubate for 10 min at room temperature (with caps open). 5. While samples incubate, prepare Dynabeads elution solution 1, adjusting the total volume to the number of samples being prepared. Vortex to mix. 6. After samples incubate for 10 min, place the tube strip on the 10× magnetic separator set to high until the solution is clear. (A white boundary layer between the two phases may form.) Remove the supernatant. 7. With the tubes on the magnetic separator, add 300 μL 80% ethanol to each. Wait for 30 s and then remove ethanol. 8. Centrifuge before placing the sample on the 10× magnetic separator low position with the magnet set to the low setting. 9. Remove any remaining ethanol and allow the pellet to air dry for at least 2 min. 10. Immediately add 35.5 μL of Dynabeads elution solution 1 (from Subheading 3.2.3, step 5) to the sample after removing the tube from the magnet. Pipette to mix until beads are fully resuspended (without introducing bubbles). 11. After incubating the sample for 1 min at room temperature, place the samples on the magnetic separator set to low. 12. Once the solution clears, transfer 35 μL of each sample to a new 8-tube strip. 3.2.4

cDNA Amplification

1. Prepare the cDNA amplification mix, adjusting the total volume to the number of samples being prepared (see Note 13). 2. Add 65 μL of the cDNA amplification mix to each 35 μL sample from step 12 of Subheading 3.2.3. 3. Pipette to mix, centrifuge briefly, and incubate samples with the following cycle in a thermal cycler. Lid temperature

Reaction volume

Run time

105 °C

100 μL

~25–50 min

Step

Temperature

Time

Cycles

Initial denaturation

98 °C

45 s

1

Denaturing Annealing Extension

98 °C 63 °C 72 °C

20 s 30 s 1 min

11–16 (See step 3a for more details)

Final extension

72 °C

1 min

1

Hold

4 °C

Hold

1

YSD Antibody Library from Single-Cell Sequencing

189

(a) The number of amplification cycles should be optimized based on the targeted cell recovery and the cell type. Suggested initial guesses for the number of cycles for a given targeted cell recovery and cell type are provided in the following table.

Targeted cell recovery

Low RNA content cells (e.g., primary cells) Total cycles

High RNA content cells (e.g., cell lines) Total cycles

500–2000

16

14

2001–6000

14

12

6001–10,000

13

11

4. Store the sample (see Note 14), or proceed to the following step. 3.2.5 SPRIselect cDNA Cleanup

1. Add 60 μL of 0.6× SPRIselect reagent to each sample and pipette to mix. 2. After incubating for 5 min at room temperature, place the tube strips on the magnetic separator with the magnet set to high. Once the solution is clear, remove the supernatant. 3. Add 200 μL of 80% ethanol to the pellet, wait 30 s, and then remove the ethanol. Repeat this wash step. 4. Centrifuge the tube strip and place it on the magnetic separator with the magnet set to low. 5. Remove any residual ethanol and air dry for at least 2 min. 6. After removing the tubes from the magnet, add 45.5 μL buffer EB. Pipette to mix. 7. After incubating for 2 min at room temperature, place the tube strip on the magnetic separator with the magnet set to high. 8. Once the solution clears, transfer 45 μL of the sample to the new 8-tube strip. 9. Store the sample (see Note 15), or proceed to the following step.

3.2.6 cDNA QC and Quantification

1. Analyze the samples by running 1 μL of the undiluted sample on a bioanalyzer (see Note 16). (a) For cells with low RNA content, run 1 μL, but for high RNA content cells, use 1 μL of a 1:10 dilution in ultrapure water. (b) Quantify the cDNA yield for each sample.

190

Nathaniel Pascual et al.

3.2.7 V(D)J Amplification 1

1. Place the cDNA product on ice. 2. Transfer 2 μL of each sample to a tube strip. 3. Prepare the V(D)J Amplification 1 Reaction Mix on ice, adjusting the total volume to the number of samples being prepared. 4. Add 98 μL of the V(D)J Amplification 1 Reaction Mix to each 2 μL sample. Pipette to mix and centrifuge briefly. 5. Incubate samples with the following cycle. Lid temperature

Reaction volume

Run time

105 °C

100 μL

~25–30 min

Step

Temperature

Time

Cycles

Initial denaturation

98 °C

45 s

1

Denaturing Annealing Extension

98 °C 62 °C 72 °C

20 s 30 s 1 min

8

Final extension

72 °C

1 min

1

Hold

4 °C

Hold

1

6. Store the samples at 4 °C for up to 72 h, or proceed to the following step. 3.2.8 SPRIselect Post-V (D)J Amplification 1 Cleanup

1. Add 50 μL 0.5× SPRIselect reagent to each sample. Pipette to mix. 2. After incubating samples at room temperature for 5 min, place the tube strip on the magnetic separator set to high. 3. Once the solution clears, transfer 145 μL of the supernatant to a new tube strip. Do not discard the supernatant. 4. Add 30 μL 0.8× SPRIselect reagent to each sample. Pipette to mix. 5. After incubating for 5 min at room temperature, place the tube strip on the magnetic separator set to high. 6. Once the solution is clear, remove 170 μL supernatant, making sure not to aspirate any of the beads. 7. Add 200 μL 80% ethanol, wait 30 s, and aspirate and discard the ethanol. Repeat this ethanol washing step two more times. 8. Centrifuge samples and then place the sample on the magnetic separator set to low. 9. Remove any residual ethanol and add 35.5 μL buffer EB. Pipette to mix. 10. After incubating for 2 min at room temperature, place the tube strip on the magnetic separator set at low.

YSD Antibody Library from Single-Cell Sequencing

191

11. Once the solution supernatant is clear, transfer 35 μL of the sample to a new tube strip. 12. Store the sample (see Note 17) or proceed to the next step. 3.2.9 V(D)J Amplification 2

1. Prepare V(D)J Amplification 2 Reaction Mix on ice, adjusting the total volume to the number of samples being prepared. 2. Add 65 μL of V(D)J Amplification 2 Reaction Mix to each sample. Pipette to mix and centrifuge briefly. 3. Run the samples with the following thermal cycle. Lid temperature

Reaction volume

Run time

105 °C

100 μL

~25–30 min

Step

Temperature

Time

Cycles

Initial denaturation

98 °C

45 s

1

Denaturing Annealing Extension

98 °C 62 °C 72 °C

20 s 30 s 1 min

8

Final extension

72 °C

1 min

1

Hold

4 °C

Hold

1

4. Store the sample at 4 °C for up to 72 h, or proceed to the next step. 3.2.10 SPRIselect Post-V (D)J Amplification 2 Cleanup

1. Add 50 μL 0.5× SPRIselect reagent to each sample. Pipette to mix. 2. After incubating samples at room temperature for 5 min, place the tube strip on the magnetic separator set to high. 3. Once the solution is clear, transfer 145 μL of the supernatant to a new tube strip. Do not discard the supernatant. 4. Add 30 μL 0.8× SPRIselect reagent to each sample. Pipette to mix. 5. After incubating for 5 min at room temperature, place the tube strip on the magnetic separator set to high. 6. Once the solution is clear, remove 170 μL supernatant, making sure not to aspirate any of the beads. 7. Add 200 μL 80% ethanol, wait 30 s, and remove the ethanol. Repeat this ethanol washing step two more times. 8. Centrifuge samples and then place samples on the magnetic separator set to low. 9. Remove any residual ethanol, and add 45.5 μL buffer EB. Pipette to mix. 10. After incubating for 2 min at room temperature, place the tube strip on the magnetic separator set at low.

192

Nathaniel Pascual et al.

11. Once the solution is clear, transfer 45 μL of the sample to a new tube strip. 12. Store the sample (see Note 18) or proceed to the next step. 13. Perform QC on the V(D)J Amplification with an Agilent Bioanalyzer High Sensitivity chip (see Note 19). (a) Run 1 μL of each sample at a 1:5 dilution using ultra-pure water. If using RNA-rich cells, dilute the samples further. (b) Determine the yield for each sample. 3.2.11 Fragmentation, End Repair, and A-Tailing

1. Before starting step 5 of Subheading 3.2.11, pre-chill a thermal cycler to 4 °C. 2. Pipette 50 ng of each sample into a tube strip set on ice. (a) Adjust volume to 20 μL with nuclease-free water if the volume of 50 ng is less than 20 μL. (b) If 50 ng of the sample exceeds 20 μL, only use 20 μL of the sample. 3. Prepare the fragmentation mix, adjusting to the number of samples being prepared. 4. Add 30 μL of the fragmentation mix into each tube containing 20 μL of the sample. Pipette to mix and centrifuge briefly. Thoroughly vortex the fragmentation buffer before adding. 5. Transfer the samples into the thermal cycler and run the following protocol.

3.2.12

Adaptor Ligation

Lid temperature

Reaction volume

Run time

65 °C

50 μL

~35 min

Step

Temperature

Time

Pre-cool block

4 °C

Hold

Fragmentation

32 °C

2 min

End repair and A-tailing

65 °C

30 min

Hold

4 °C

Hold

1. While the samples undergo fragmentation, prepare the adaptor ligation mix, adjusting to the number of samples being prepared. 2. After the protocol in step 5 of Subheading 3.2.11 is finished, add 50 μL of the adaptor ligation mix to each of the samples. Pipette to mix and centrifuge briefly. 3. Run the samples with the following protocol in the thermal cycler.

YSD Antibody Library from Single-Cell Sequencing

3.2.13 SPRIselect PostLigation Cleanup

193

Lid temperature

Reaction volume

Run time

30 °C

100 μL

15 min

Step

Temperature

Time

1

20 °C

15 min

2

4 °C

Hold

1. Add 80 μL of 0.8× SPRIselect reagent to each sample. Pipette to mix. 2. After incubating for 5 min at room temperature, place the tube strip on the magnetic separator set on high. Remove the supernatant when the solution is clear. 3. Add 200 μL 80% ethanol to the pellet, wait 30 s, and remove the ethanol. Repeat this ethanol washing step. 4. Centrifuge the tube strips, and place the tubes on the magnetic separator set to low. Remove any residual ethanol, and air dry for 2 min. 5. Add 30.5 μL buffer EB. Pipette to mix until beads are thoroughly resuspended. 6. After incubating for 2 min at room temperature, place the tube strips on the magnetic separator set to low. 7. Once the solution is clear, transfer 30 μL of the sample to a new tube strip.

3.2.14 PCR

Sample Index

1. Add 50 μL of the Amp Mix to 30 μL sample. 2. Add 20 μL of an individual Dual Index TT Set A to each well and record the well ID used. Ensure that the well IDs are not repeated in a single run. Pipette to mix and centrifuge briefly. 3. Incubate the samples with the following protocol in a thermal cycler. Lid temperature

Reaction volume

Run time

105 °C

100 μL

~30 min

Step

Temperature

Time

Cycles

Initial denaturation

98 °C

45 s

1

Denaturing Annealing Extension

98 °C 54 °C 72 °C

20 s 30 s 20 s

8

Final extension

72 °C

1 min

1

Hold

4 °C

Hold

1

4. Store the samples at 4 °C for up to 72 h or proceed to the following step.

194

Nathaniel Pascual et al.

3.2.15 SPRIselect PostSample Index PCR Cleanup

1. Add 80 μL of 0.8× SPRIselect reagent to each sample. Pipette to mix. 2. After incubating for 5 min at room temperature, place the tube strip on the magnetic separator set on high. Remove the supernatant. 3. Once the solution is clear, add 200 μL 80% ethanol to the pellet, wait 30 s, and remove the ethanol. Repeat this ethanol washing step two more times. 4. Centrifuge the tube strips and place the tubes on the magnetic separator set to low. 5. Remove any residual ethanol, and air dry for 2 min. 6. Add 35.5 μL buffer EB. Pipette to mix until beads are thoroughly resuspended. 7. After incubating for 2 min at room temperature, place the tube strips on the magnetic separator set to low. 8. Once the solution is clear, transfer 35 μL of the sample to a new tube strip. 9. Store the sample at 4 °C for up to 72 h or at -20 °C for longterm storage. 10. Analyze 1 μL sample diluted by 1:10 on an Agilent Bioanalyzer High Sensitivity Chip (see Note 20). (a) Determine the average fragment size from the trace for library quantification.

3.2.16 GEX Library Fragmentation, End Repair, and A-Tailing

1. Before starting step 5 of Subheading 3.2.16, pre-cool the thermal cycler block to 4 °C. 2. Add 50 ng of the GEX sample into a tube within a tube strip that has been chilled on ice. (a) If the total volume is less than 20 μL, add enough nuclease-free water for a total volume of 20 μL. (b) If the volume of 50 ng is greater than 20 μL, only add 20 μL of the sample into the tube. 3. Vortex the fragmentation buffer before preparing the fragmentation mix, adjusting the total volume to the number of samples being prepared. 4. Transfer 30 μL of fragmentation mix into each of the 20 μL samples. 5. Run the samples with the following protocol on a thermal cycler.

YSD Antibody Library from Single-Cell Sequencing

3.2.17 SPRIselect PostFragmentation, End Repair, and A-Tailing Cleanup

195

Lid temperature

Reaction volume

Run time

65 °C

50 μL

~35 min

Step

Temperature

Time

Pre-cool block

4 °C

Hold

Fragmentation

32 °C

5 min

End repair and A-tailing

65 °C

30 min

Hold

4 °C

Hold

1. After the protocol is complete, add 30 μL 0.6× SPRIselect reagent to each sample. Pipette to mix. 2. After incubating the SPRIselect beads with the samples for 5 min at room temperature, place the tube strip on the magnetic separator set to high. Do not discard the supernatant. 3. Once the solution is clear, transfer 75 μL of the supernatant to a new tube strip. 4. Add 10 μL of 0.8× SPRIselect reagent to each sample. Pipette to mix. 5. After incubating for 5 min at room temperature, place the tube strip on the magnetic separator set to high. 6. Once the solution is clear, remove 80 μL of the supernatant. Do not discard any of the beads. 7. With the tube strip on the magnetic separator, add 125 μL 80% ethanol to the pellet and remove the ethanol after 30 s. Repeat this washing step one more time. 8. Remove the ethanol and then remove the tube strip from the separator. 9. Add 50.5 μL buffer EB. Pipette to mix. Incubate for 2 min at room temperature. 10. Place the tube strip on the magnetic separator set to high. Once the solution is clear, transfer 50 μL of each sample to a new tube strip.

3.2.18 GEX Adaptor Ligation

1. Prepare the adaptor ligation mix, adjusting the total volume to the number of samples being prepared. 2. Add 50 μL of the adaptor ligation mix to each sample, pipetting to mix. 3. Incubate samples with the following protocol on a thermal cycler.

196

Nathaniel Pascual et al.

3.2.19 SPRIselect PostLigation Cleanup

Lid temperature

Reaction volume

Run time

30 °C

100 μL

15 min

Step

Temperature

Time

1

20 °C

15 min

2

4 °C

Hold

1. Add 80 μL 0.8× SPRIselect reagent to each sample and pipette to mix. 2. After incubating the samples for 5 min at room temperature, place the tube strip on the magnetic separator set to high. Once the solution is clear, remove the supernatant. 3. Add 200 μL 80% ethanol to the pellet, wait 30 s, and remove the ethanol. Repeat this washing step one more time. 4. Centrifuge briefly and place the tube strip on the magnetic separator set to low. 5. Remove any remaining ethanol and air dry for 2 min. 6. Remove the tubes from the magnet, and add 30.5 μL buffer EB, pipetting to mix. 7. Incubate the samples for 2 min at room temperature and place the tubes on the magnetic separator set to low. 8. Once the solution is clear, transfer 30 μL of each sample to a new tube strip.

3.2.20 PCR

GEX Sample Index

1. Add 50 μL Amp Mix to each sample, and add 20 μL of Dual Index TT Set A to each well. Pipette to mix. 2. Incubate the samples with the following protocol. Lid temperature

Reaction volume Run time

105 °C

100 μL

~30 min

Step

Temperature

Time

Cycles

45 s

1

Initial denaturation 98 °C Denaturing Annealing Extension

98 °C 54 °C 72 °C

20 s 30 s 1 min

14 or 16 See table below for details

Final extension

72 °C

1 min

1

Hold

4 °C

Hold

1

YSD Antibody Library from Single-Cell Sequencing

197

cDNA input

Recommended number of cycles

1–25 ng

16

26–50 ng

14

3. Store the sample at 4 °C for up to 72 h, or proceed to the next step. 3.2.21 SPRIselect PostGEX Sample Index PCR Cleanup

1. Add 60 μL 0.6× SPRIselect reagent to each sample. Pipette to mix. 2. After incubating for 5 min at room temperature, place the tube strip on the magnetic separator set to high. Do not discard the supernatant. 3. Once the solution is clear, transfer 150 μL of the supernatant to a new tube strip. 4. Add 20 μL 0.8× SPRIselect reagent to each sample. Pipette to mix. 5. After incubating for 5 min at room temperature, place the tubes on the magnetic separator. 6. Once the solution is clear, remove 165 μL of the supernatant, and ensure that none of the beads are aspirated. 7. With the tubes still on the magnet, wash the pellet with 200 μL 80% ethanol. Wait 30 s before removing the ethanol. Repeat this wash step one more time. 8. Centrifuge samples briefly before placing tubes back on the magnetic separator set to low. 9. Remove the remaining ethanol before adding 35.5 μL buffer EB. 10. Remove the tube strip on the magnet, and pipette to mix. 11. After incubating for 2 min at room temperature, return the tubes to the magnetic separator set to low. 12. Once the solution is clear, transfer 35 μL of each sample to a new tube strip. 13. Store at 4 °C for up to 72 h or -20 °C for long-term storage. 14. Analyze 1 μL of the sample diluted 1:10 on an Agilent Bioanalyzer High Sensitivity Chip (see Note 21). 15. The V(D)J library and GEX library are subsequently pooled in a 1:7 ratio using a single SP 2× 150 bp paired-end lane from Illumina’s NovaSeq 6000 SP platform using the XP workflow (see Note 22).

198

Nathaniel Pascual et al.

To limit the burden of experimentally characterizing each potential clone, this section entails an overview of how the raw sequencing data produced by Illumina sequencing can be interpreted as a list of clonotypes (Fig. 4). After converting the BCL file into the FASTQ file format (Table 1), 10× Genomics’ Cell Ranger program can be used to generate sequences and counts of paired clonotypes from V

3.3 Lead Clone Characterization and Refinement

Fig. 4 Analyzing raw sequencing data to isolate natively paired B-cell receptor sequences. The raw sequencing data contained with .bcl files are first converted to .fastq files using bcl2fastq2. Using the .fastq file as an input, cellranger multi-pipeline simultaneously analyzes background gene expression (from the GEX cDNA library) and B-cell library sequences (from the VDJ cDNA library) to assemble sequences of paired clonotypes. Once the framework and CDR sequences are annotated, selected clonotypes can be cloned into yeast for high-throughput characterization Table 1 Documentation for programs/workflow used Tool/program/ workflow

Version Use

Link

Cellranger multi

6.1.2

Analyze GEX and V(D)J sequencing data [step 3 of Subheading 3.3]

https://support.10xgenomics.com/ single-cell-vdj/software/ pipelines/latest/using/multi

Bcl2fastq2

2.20

Creating a FASTQ sequencing file from raw data [step 1 of Subheading 3.3]

https://support.illumina.com/ content/dam/illumina-support/ documents/documentation/ software_documentation/bcl2 fastq/bcl2fastq2-v2-20-softwareguide-15051736-03.pdf

“SequenceBuilder” R2022a Obtaining VH and VL sequences https://github.com/ MATLAB script for incorporation into scFab WoldringLabMSU/ [step 5 of Subheading 3.3] SequenceProcessing10X

YSD Antibody Library from Single-Cell Sequencing

199

(D)J libraries after performing alignment, filtering, and counting operations [17, 19]. While not extensively detailed in this method, several computational tools and workflows can be implemented at this stage to reduce the number of clones to characterize experimentally based on various developability criteria. In the Notes section, we provide some suggestions to incorporate tools to consider humanness (see Note 23), binding affinity (see Note 24), and developability broadly (see Note 25). 1. The FASTQ sequencing file can be generated from the raw data produced in the sequencing workflow using bcl2fastq2 with the PE150 format. After installing bcl2fastq2 into a compatible Linux/CentOS workstation, the following command can be run with the necessary input base call files (.cbcl), filter files (.filter), cluster location files (s.locs), and run info file defined with the --input-dir option (see Note 26): nohup /path/to/bcl2fastq --runfolder-dir /path/ to/run_folder \ --output-dir /path/to/output_dir \ --input-dir /path/to/input_directory

2. Prepare the following lines in a CSV file to use as an input for the cellranger multi pipeline (see Note 27): [gene-expression] reference, /path/to/gex_data [vdj] reference, /path/to/vdj_data [libraries] fastq_id, fastqs, lanes, feature_types, subsample_rate GEX_fastqs_id, /path/to/GEX_fastq, 1|2, Gene Expression, VDJ_B_fastqs_id, /path/to/vdj_fastqs, 1|2, VDJ-B,

3. Run the cellranger multi pipeline (for documentation, see Note 28): cellranger multi --id=[run ID string (see Note 29)] \ --csv=/path/to/csv_input_file

4. Much of the 10× Genomics sequencing data obtained from the cellranger multi pipeline can be downloaded in CSV format, and the amino acid or DNA sequences of the framework regions (FWRs) and complimentary determining regions (CDRs) can be extracted from the output file “consensus_annotations.csv” (see Note 30). Once these sequences are extracted, concatenate them in their numbered order (i.e.,

200

Nathaniel Pascual et al.

FWR1, CDR1, FWR2, CDR2, FWR3, CDR3, and FWR4) to form VH and VL chains. 5. Pair these newly formed VH and VL chains with their respective clonotype ID (e.g., clonotype 1 and clonotype 2), chain ID (e.g., IGK, IGH, and IGL), and the number of reads for each chain. 6. Export these data to a separate CSV file containing the clonotype ID in the first column, the concatenated VH or VL sequence in the second, the reads per chain in the third, and the chain ID in the fourth. Once this CSV file is ready, you can perform the steps that follow to generate a chimeric singlechain Fab provided you also have the sequence of a human IgG1 scFab that you can obtain CH, CL, and linker sequences from. 7. Run the available SequenceBuilder MATLAB script on the CSV file made in step 6 above, making sure to include the correct name of the CSV file in the script (the default name is 10XGEN. xlsx) (see Note 31). Running the script will generate the corresponding VH and VL chains for each clonotype in separate CSV outputs. This script will go through the data for each of the clonotypes, and if any of them happen to have multiple light chains listed, it will select the one with the most reads. 8. Obtain the amino acid or DNA sequences of the constant heavy (CH), constant light (CL), and linker regions of the human IgG1 scFab. Make sure the antibody numbering system used in the 10× Genomics sequencing data is consistent with the numbering system used by any software that identifies the CH, CL, and linker regions of the human IgG1 scFab. 9. Assemble all of the obtained amino acid or DNA sequences in the following order to obtain the amino acid or DNA sequences of the chimeric scFabs: VL, CL, linker, VH, and CH. Three-dimensional structures of the scFabs can now be obtained using homology modeling software. These chimeric scFabs can offer a starting point for further computational refinement and development of expressed antibodies with increased humanness and reduced immunogenicity. 10. For generating scFab sequences with greater humanness, you can choose to extract only the CDR regions from the “consensus_annotations.csv” file. These CDR regions will have to be concatenated with the framework regions of the human IgG1 scFab (in the same order as presented in step 4 above) so that VH and VL chains are obtained. Make a separate CSV with these VH or VL sequences, chain IDs, clonotype IDs, and reads as mentioned in steps 5 and 6 above. Make sure the numbering systems for the 10× Genomics data and the IgG1 Fab are the same. Run the MATLAB script and then follow steps 8 and 9 as mentioned above.

YSD Antibody Library from Single-Cell Sequencing

201

Fig. 5 Yeast surface display expression of single-chain antibody fragment (scFab). (a) scFabs are expressed on the surface of yeast using the Aga1-Aga2 system. With Aga1p anchored to the surface, the scFab along with hemagglutinin (HA) and c-Myc (Myc) epitope tags can be fused to Aga2p, which is attached to Aga1p via disulfide bonds. (b) Several different antibody fragment formats may be inserted in the Aga1-Aga2 system. The simplified outline of scFab and single-chain variable fragment (scFv) is illustrated here. (c) Expression of Aga2p-scFab fusion protein is driven by the galactose-inducible GAL1 (pGAL1) promoter. Relative placement of HA and Myc tags is shown along with the cut sites for NheI and NotI

3.4 Library Generation and YSD Expression

3.4.1 Prepare VH and VL Inserts

After determining which antibodies to experimentally characterize, the VH and VL domains identified in Subheading 3.3 can be isolated to be cloned into yeast display vectors (Fig. 5). The VH and VL inserts can be either directly amplified from cDNA libraries generated in Subheading 3.2 or purchased as synthetic DNA fragments. A standard T4 ligation-based strategy is presented below, but other cloning methods may be substituted. After propagating plasmid in competent E. coli (NEB5α), the expression vector is transformed into yeast via electroporation or EZ-Yeast transformation (see Note 32). 1. VH and VL domains can be prepared by direct PCR amplification from cDNA libraries or ordered as synthetic DNA fragments (see Note 33). In either case, the insert should be flanked with NheI and BamHI for downstream steps. 2. Prepare the inserts by incubating the following restriction digest mix overnight at 37 °C.

202

Nathaniel Pascual et al.

Component

Amount

Units

Insert DNA

Up to 10

μg

NheI-HF

5

U

BamHI-HF

5

U

CutSmart (10×)

5

μL

ddH2O

To a total reaction volume of 50 μL

3. Isolate the desired fragment with a DNA cleanup kit or gel extraction kit (after performing gel electrophoresis on RE digest products) according to the manufacturer’s instructions. 4. Quantify insert concentration. 3.4.2 Prepare Yeast Expression Vector

1. Pick a single colony of pCTCon2 expression vector and inoculate 7 mL of LB-amp overnight (see Note 34). 2. Isolate plasmid from the overnight culture with a plasmid preparation kit of your choice, following the manufacturer’s instructions. 3. Quantify plasmid concentration and purity. 4. Prepare the following restriction digest mix to linearize the vector. Component

Amount

Units

Backbone plasmid DNA

Up to 10

μg

NheI-HF

5

Units

BamHI-HF

5

Units

CutSmart (10×)

5

μL

ddH2O

To a total reaction volume of 50 μL

5. Incubate RE digest mix overnight at 37 °C. 6. Isolate the desired linearized DNA with a DNA Cleanup Kit or Gel Extraction Kit (after performing gel electrophoresis on RE digest products) according to the manufacturer’s instructions. 7. Quantify linearized DNA concentration and purity via A260: A280 ratio. 8. Store products at -20 °C for long-term storage or proceed to the following step.

YSD Antibody Library from Single-Cell Sequencing 3.4.3

Ligation

203

1. Verify the concentration and purity of insert and linearized backbone DNA. 2. Prepare the following ligation reaction mix. Component

Amount

Units

T4 DNA ligase buffer (10×)

1.5

μL

T4 DNA ligase

400

units

Linearized DNA backbone

100

ng

Insert DNA

3× to 10× molar excess

ddH2O

To a total volume of 15 μL

3. Incubate at room temperature for 10 min. 3.4.4

Transformation

1. Transform 5 μL of ligation reaction mix product into 50 μL NEB5α competent cells, following the manufacturer’s recommended protocol [20]. Store the remaining ligation product at -20 °C after heat inactivating T4 ligase by incubating the reaction mix at 65 °C for 10 min. 2. Thaw NEB5α cells on ice. 3. Pipette 5 μL of the ligation reaction mix to 50 μL of NEB5α competent cells. 4. Incubate the ligation mix–cell mixture on ice for 30 min. 5. Heat shock cell mixture at 42 °C for exactly 30 s. 6. Place cells on ice for 5 min. 7. Add 950 μL SOB (alternatively SOC) media and incubate the mixture at 37 °C for 1–2 h in a shaking incubator set to 250 rpm. 8. Plate 50–200 μL of cells on an LB agar plate. 9. Incubate overnight at 37 °C. 10. Verify clones by colony PCR and Sanger sequencing.

3.4.5

Prepare Plasmid

1. Inoculate 7 mL of LB supplemented with ampicillin with a single colony from the LB–ampicillin agar plate from step 8 in Subheading 3.4.4. 2. Incubate overnight at 37 °C. 3. Isolate the desired plasmid with a plasmid prep kit of your choice, following the manufacturer’s suggested protocol. Quantify plasmid yield and purity.

3.4.6 Transform Clones Into Yeast: Option A— Electroporation

1. Grow an overnight colony of EBY100 yeast on YPD media at 30 °C, shaking at 250 rpm.

204

Nathaniel Pascual et al.

2. The following morning, for each sample, inoculate 10 × 107 cells into 100 mL of YPD. 3. Grow cells at 30 °C, 250 rpm until a density of 1.3–1.5 × 107 cells/mL. (This step should take ~6–8 h, assuming a doubling time of 1.5 h) (see Note 35). 4. Pellet cells (2250 rcf for 3 min) and discard the supernatant. 5. Wash cells with cold dH2O twice, pelleting cells at 2250 rcf for 3 min to discard the supernatant. 6. Wash the cells once with 25 mL cold buffer E. 7. Resuspend cells in 25 mL of lithium acetate solution (see Note 36). 8. Incubate cells for 30 min at 30 °C, shaking at 250 rpm. 9. Centrifuge cells at 2250 × g at 4 °C for 3 min. Discard the supernatant. 10. Wash cells with 25 mL col buffer E. After centrifuging cells at 2250 × g at 4 °C for 3 min, discard the supernatant. 11. Resuspend each sample in 1 mL of cold buffer E, and transfer samples to 1.5 mL tubes. 12. Centrifuge at 5000 × g at 4 °C for 1 min. Discard the supernatant. 13. Wash cells twice with 1 mL cold buffer E, centrifuging samples at 5000 × g at 4 °C for 1 min. 14. After discarding the supernatant from the last wash, resuspend to cells in 300 μL of cold buffer E. 15. Add 6 μg of your desired clone (or ~1.5 pmol) to the cells. 16. Transfer cell–DNA mixture to electroporation cuvette. Incubate on ice for 5 min. 17. Remove any condensation from the outside of the cuvette with a paper towel. 18. Pulse cuvette at 25 μF, 1.2 kV with a time constant of ~4–45 ms. 19. Immediately add 1 mL of room temperature YPD to the cuvette, and chill the cuvette on ice. 20. Transfer cell–DNA mixture to 15 mL conical tube with 4 mL YPD. 21. Incubate at 30 °C for 1–2 h in a shaking incubator set to 250 rpm. 22. Centrifuge cells at 1300 × g for 1 min. 23. Resuspend cells in 1 mL SD-/Trp and transfer cells to a baffled flask with 100 mL SD-/Trp.

YSD Antibody Library from Single-Cell Sequencing

205

24. Plate cells on SD/-Trp plates, creating serial dilutions ranging from 100× to 2000× dilutions to determine transformation efficiency. 25. Incubate plates at 30 °C. Cells in baffled flasks should be incubated for at least 16 h at 30 °C and 250 rpm. 3.4.7 Transform Clones Into Yeast: Option B—EZYeast Transformation

1. Grow EBY100 cells in 10 mL YPD broth to an OD600 of 0.8–1.0 (mid-log phase). 2. Pellet cells at 500 × g for 4 min. Discard the supernatant. 3. Wash cells with 10 mL EZ 1 solution, pellet cells at 500 × g for 4 min, and discard the supernatant. 4. Resuspend the pellet with 1 mL EZ 2 solution. 5. Add 0.5–2 μg (less than 5 μL total volume) to 50 μL of competent cells with 500 μL EZ 3 solution. Mix thoroughly. 6. Incubate at 30 °C for 45 min, vertexing 2–3 times to mix. 7. Pellet cells at 1500 × g for 3 min. Discard the supernatant. 8. Wash cells with 1 mL PBSA, pellet cells at 1500 × g for 3 min, and discard the supernatant. 9. Resuspend cells in 3 mL of YPD media. 10. Incubate cells at 30 °C for 1 h. 11. Pellet cells at 1500 × g for 3 min and remove supernatant. 12. Wash cells with 1 mL of PBSA, pellet cells at 1500 × g for 3 min, and resuspend cells in 5 mL of SD. 13. Plate 50 μL of cells suspended in SD media to characterize transformation efficiency.

4

Notes 1. In silico affinity maturation protocols can be used alongside in silico antibody–antigen docking to selectively mutate the antibody paratope and generate sequences with improved binding to an antigen of interest. This work can be combined with other computational protocols that optimize antibody structural stability through selective mutations. The Rosetta software suite has been successfully applied to such tasks, allowing researchers to focus on smaller, more optimized antibody libraries prior to affinity maturation in a laboratory setting [21–23]. Furthermore, other developability criteria can be considered at this stage. For example, computational tools such as Hu-mAb and BioPhi can be used in the process of humanizing antibodies generated from animal models to reduce their immunogenicity [24, 25]. In contrast, developability can be considered more broadly through workflows that

206

Nathaniel Pascual et al.

incorporate multiple developability criteria in a single workflow to identify candidates that exhibit suboptimal properties [26]. 2. For all steps using SPRIselect beads, thoroughly vortex SPRIselect reagent mix to ensure beads are fully resuspended. 3. For certain steps, Agilent Bioanalyzer or Qubit dsDNA Fluorometer Kit can be used instead. 4. It is advisable to practice an immunization and blood draw with a mouse in advance, to ensure the procedure can be done quickly and without causing undue stress to the animal. 5. The ELISA protocol described here uses BSA conjugated to any antigen, while the work by Wu et al. uses BSA conjugated to a specific antigen. The identity of the antigen should not affect the results obtained. 6. You may be able to obtain the needed number of B cells with one mouse spleen, but the size of the spleen can vary significantly with the age of the mouse. Thus, two spleens can guarantee a sufficient number of cells for the sequencing that comes later. 7. White adipose tissue in the spleens may be quite difficult to grind down. It is not necessary to homogenize such tissue completely. 8. In addition to 10× Genomics’ Chromium Platform, Bio-Rad (ddSEQ), Dolomite (Drop-seq), Takara Bio (ICELL8), and BD (Rhapsody) offer alternative kits to generate single-cell sequencing libraries [27, 28]. Yamawaki et al. (2021) compare the offerings of 10× Genomics, Bio-Rad, Dolomite, and Takara, whereas Gao et al. (2020) offer a comparison between BD and 10× Genomics. If desired, a greater degree of customizability is offered by Smart-seq2 (at the expense of time needed to optimize the workflow) compared to the standardized protocols offered in traditional kits [29, 30]. See et al. (2018) provide a comprehensive review of key experimental factors to consider (e.g., desired cell sequencing yield, sequencing depth, and experimental design) and a generalized comparison of available scRNA-seq methods [31]. For this method, we implemented 10× Genomics’ Chromium due to the prior established implementation of the platform at our institution. 9. For steps involving creating master mixes (i.e., super mix and ready mix), we recommend including at least 10% buffer to account for any pipetting inaccuracies. 10. Optimal cell recovery ranges from 1000 to 10,000 cells, with cell stocks having concentrations ranging from 700 to 1200 cells/μL. We recommend consulting with a research support facility (e.g., Genomics Core) to optimize the amount being loaded in each well. The actual number of cells required can

YSD Antibody Library from Single-Cell Sequencing

207

vary based on the complexity of the cell type and the desired results of the experiments. For the experimental protocol in this chapter, we targeted 10,000 cells for recovery at a concentration of 1000 cells/μL. 11. If the mixture fails to separate, gently mix by inverting the tube five times, and proceed to the following step after allowing the mixture to separate. 12. A smaller aqueous phase volume (relative to the recovery agent/partitioning oil phase) indicates a clog during GEM generation. Clogs are generally caused by improper handling of the gel beads. If there is a clog, you should rerun the samples. If this is not possible (e.g., due to limited sample cells), you can proceed to the following steps, but the expected cDNA yield would decrease. 10× Genomics has, at the time of writing, a warranty policy where the costs of replacement reagents and chips can be reimbursed for appropriately documented claims: https://kb.10xgenomics.com/hc/en-us/arti cles/217266006-Should-I-proceed-if-there-is-a-sample-clog13. The primer composition is significantly different between versions and desired applications, and ensure you use the appropriate solution for this step. 14. Samples can be stored at 4 °C for 72 h or -20 °C for less than a week. 15. Samples can be stored at 4 °C for 72 h or at -20 °C for up to 4 weeks. 16. 10× Genomics recommends Agilent Bioanalyzer High Sensitivity Chip, Agilent TapeStation, or Qubit Fluorometer with the dsDNA HS Assay to perform QC on cDNA produce. 17. Store the sample at 4 °C for up to 72 h or at -20 °C for up to a week. 18. Store the sample at 4 °C for up to 72 h or at -20 °C for up to a week. 19. Alternatively, samples can be analyzed with an Agilent TapeStation, LabChip, or Qubit Fluorometer using the dsDNA HS Assay Kit. 20. Alternatively, an Agilent TapeStation or LabChip can be used to analyze the samples. 21. Alternatively, samples can be analyzed with an Agilent TapeStation or LabChip. 22. The 1:7 ratio was determined in consultation with 10× Genomics technical support and MSU’s Genomics Core based on the specifics of our experiment. We highly recommend consulting with 10× Genomics support and (if available) institution/

208

Nathaniel Pascual et al.

campus-specific genomics core staff to ensure the appropriate parameters are chosen for sequencing. 23. Humanness/Immunogenicity: As noted in Subheading 3.3, chimeric scFabs were generated in this method. To address potential concerns with immunogenicity, these sequences can be further humanized with computational tools. One can simply graft the annotated CDR regions onto human antibody frameworks with high homology to the source sequence [32]. Alternatively, more powerful machine learning classifiers (e.g., Hu-mAb and BioPhi) can be used to identify mutations that result in significant improvements in estimated “humanness” [24, 25]. 24. Affinity Maturation: In silico affinity maturation can be performed to improve the binding affinities of antibody candidates obtained from single-cell sequencing before laboratory expression. To accomplish this, scFab sequences obtained from the single-cell sequencing data can be transformed into threedimensional PDB structures using homology modeling software like AlphaFold [33]. These PDB structures can then undergo in silico protein–protein or protein–small molecule (i.e., ligand) docking to the antigen of interest in a software suite such as Rosetta. Prior to docking, the homologymodeled scFab structures need to be relaxed into Rosetta’s force field and the structures of the antigen’s conformers need to be generated if it is a small molecule [34, 35]. The docked scFab-antigen pose with the lowest energy is obtained after docking, and this complex is used as an input to Rosetta’s affinity maturation protocol. In the affinity maturation protocol, the user specifies the residues of the scFab they want to be mutated and what amino acid substitutions they want in those mutated positions in an input called a “resfile.” The scFab will then be mutated using Rosetta’s single-state design protocol, and the binding energy of the designed scFab structures can be compared to the non-mutated control. Rosetta’s design protocol generates mutations that improve both stability and binding affinity. Using Python scripts available within the Meiler Lab single-state design workshop, plots can be generated that show the energetic contribution of each mutation to the binding and stability scores [22]. The amino acid substitutions that contributed the most to the stabilization and binding affinity of the scFab can be incorporated into the sequences that will ultimately be expressed in yeast. If none of the designed scFab structures have binding energy and stability scores that meet the user’s expectations, the lowest energy designed scFab structure (the strongest binder generated through mutagenesis) can be subjected to another round of affinity maturation, where it will serve as a control. Multiple

YSD Antibody Library from Single-Cell Sequencing

209

iterations of this affinity maturation protocol on the lowest scoring designed models may help the user approach an energy minimum for their scFab-antigen complex. Expression of these optimized sequences may greatly expedite antibody development. 25. Developability Assessments in General: While most tools developed to computationally predict developability properties consider each property alone, the impacts of mutating a candidate mAb to improve one characteristic have been reported to impact other properties (e.g., mutations to improve binding affinity resulting in decreases to thermostability) [36, 37]. Furthermore, there are a wide variety of tools to consider, yet only a subset of these tools may be applicable to a specific project. One approach to this challenge is the combinatorial triage approach [26]. Due to the computational demands of many of these programs, less computationally demanding programs are first used to narrow down the list of potential mAbs, saving the most computationally demanding workflows for the final stage of characterization. Researchers must decide which aspects are most important to their specific projects, but there is general agreement that “orthogonal combinations” of diverse algorithms should be considered to address biases incurred by the specific methods implemented. 26. Bcl2fastq2 (v. 2.20) documentation can be found here: https://support.illumina.com/content/dam/illumina-sup port/documents/documentation/software_documentation/ bcl2fastq/bcl2fastq2-v2-20-software-guide-15051736-03. pdf 27. Sample CSV configuration files can be found in the following webpage: https://support.10xgenomics.com/single-cell-vdj/ software/pipelines/latest/using/multi#examples 28. Cellranger multi documentation: https://support.10 xgenomics.com/single-cell-vdj/software/overview/welcome 29. The run ID should be the same name as the output folder. This name should be less than 64 characters. 30. Depending on which options are used to run Cellranger, the consensus_annotations.csv file is typically located in the following path: /path/to/output/data/[run ID string]/ outs/vdj_b/ 31. The SequenceBuilder MATLAB script can be found at our G i t H u b l i n k e d h e r e : h t t p s : // g i t h u b . c o m / WoldringLabMSU/SequenceProcessing10X 32. Electroporation versus EZ-Yeast transformation: Electroporations typically yield higher transformation efficiencies than methods such as Zymo Research’s Frozen-EZ-Yeast

210

Nathaniel Pascual et al.

Transformation II Kit, but they require more time and instrumentation. EZ-Yeast is faster and requires less optimization, at the cost of transformation efficiency. Clonal populations only require one successful transformant and so efficiency is much less important, and EZ-Yeast can be used without any detriment [38]. 33. Desired VH and VL domain sequences can be directly amplified from pooled sequence libraries via PCR, or the entire Fab cassette can be ordered [39]. While both methods are acceptable, only 81% of a population of ~94,000 cells could be isolated by PCR amplification. Though, the shorter turnaround time of producing oligonucleotides and the potential time required to optimize PCR conditions should be compared to the higher costs and longer turnaround time typically required to produce synthetic gene fragments. 34. Alternatively, an overnight culture can be prepared by directly inoculating LB media from frozen glycerol stock [40]. 35. Alternatively, the 100 mL culture can be inoculated with less cells for a longer incubation time. 36. DTT should be prepared at the time of the experiment. We recommend dissolving 0.077 g DTT into 1 mL of Tris-HCl buffer, pH 7.5, and subsequently filter sterilizing the solution into the conical tube with cells.

Acknowledgments We would like to thank the Genomics Core at Michigan State University for assisting us with the documentation for the sequencing workflow. We would also like to thank Dr. Kevin Childs and Emily Crisovan of Michigan State University for the informative discussion on the 10× Genomics sequencing process and Michael Cartwright of 10× Genomics for providing advice on the cloning of antibodies from cDNA. Some of the figures produced above were created with BioRender.com. References 1. Tang X, Huang Y, Lei J et al (2019) The singlecell sequencing: new developments and medical applications. Cell Biosci 9:1–9 2. Cao Y, Qiu Y, Tu G et al (2020) Single-cell RNA sequencing in immunology. Curr Genomics 21:564–575 3. Goldstein LD, Chen YJJ, Wu J et al (2019) Massively parallel single-cell B-cell receptor sequencing enables rapid discovery of diverse

antigen-reactive antibodies. Commun Biol 2: 1–10 4. Wen W, Su W, Tang H et al (2020) Immune cell profiling of COVID-19 patients in the recovery stage by single-cell sequencing. Cell Discov 6:1–18 5. He B, Liu S, Wang Y et al (2021) Rapid isolation and immune profiling of SARS-CoV-2 specific memory B cell in convalescent COVID-19

YSD Antibody Library from Single-Cell Sequencing patients via LIBRA-seq. Signal Transduct Target Ther 6:1–12 6. Setliff I, Shiakolas AR, Pilewski KA et al (2019) High-throughput mapping of B cell receptor sequences to antigen specificity. Cell 179: 1636–1646.e15 7. Valldorf B, Hinz SC, Russo G et al (2022) Antibody display technologies: selecting the cream of the crop. Biol Chem 403:455–477 8. 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 9. Julian MC, Rabia LA, Desai AA et al (2019) Nature-inspired design and evolution of antiamyloid antibodies. J Biol Chem 294:8438– 8451 10. Boder ET, Wittrup KD (1997) Yeast surface display for screening combinatorial polypeptide libraries. Nat Biotechnol 15:553–557 11. Chao G, Lau WL, Hackel BJ et al (2006) Isolating and engineering human antibodies using yeast surface display. Nat Protoc 1:755–768 12. Stern LA, Schrack IA, Johnson SM et al (2016) Geometry and expression enhance enrichment of functional yeast-displayed ligands via cell panning. Biotechnol Bioeng 113:2328–2341 ˜ azu´ N, Aoki MP et al (2007) 13. Pellegrini A, Guin Spleen B cells from BALB/c are more prone to activation than spleen B cells from C57BL/6 mice during a secondary immune response to cruzipain. Int Immunol 19:1395–1402 14. Wu X, Ye J, DeLaitsch AT et al (2021) Chemoenzymatic synthesis of 9NHAc-GD2 antigen to overcome the hydrolytic instability of O-acetylated-GD2 for anticancer conjugate vaccine development. Angew Chemie - Int Ed 60:24179–24188 15. MojoSort™ Isolation Kits Protocol 1, https://www.biolegend.com/en-us/ protocols/mojosort-isolation-kits-protocol-1 16. MojoSort™ Streptavidin Nanobeads Protocol - Positive Selection, https://www.biolegend. com/protocols/mojosor t-streptavidinnanobeads-protocol-positive-selection/4748/ 17. Analyzing V(D)J, Gene Expression & Feature Barcode with cellranger multi, https://sup port.10xgenomics.com/single-cell-vdj/soft ware/pipelines/6.1/using/multi 18. 10X Genomics (2021) Chromium Next GEM Single Cell 5′ Reagent Kits v2 (Dual Index) (CG000331• Rev C), 19. bcl2fastq2 Conversion Software v2.20 Software Guide (15051736), www.illumina.com/ company/legal.html

211

20. NEB High Efficiency Transformation Protocol (C2987H/C2987I), https://international. neb.com/protocols/0001/01/01/high-effi ciency-transformation-protocol-c2987 21. Xu J, Tack D, Hughes RA et al (2014) Structure-based non-canonical amino acid design to covalently crosslink an antibodyantigen complex. J Struct Biol 185:215–222 22. Schoeder CT, Schmitz S, Adolf-Bryfogle J et al (2021) Modeling immunity with Rosetta: methods for antibody and antigen design. Biochemistry 60:825–846 23. Warszawski S, Katz AB, Lipsh R et al (2019) Optimizing antibody affinity and stability by the automated design of the variable lightheavy chain interfaces. PLoS Comput Biol 15: e1007207 24. Marks C, Hummer AM, Chin M et al (2021) Humanization of antibodies using a machine learning approach on large-scale repertoire data. Bioinforma 37:4041–4047 25. Prihoda D, Maamary J, Waight A et al (2022) BioPhi: a platform for antibody design, humanization, and humanness evaluation based on natural antibody repertoires and deep learning. MAbs 14:2020203 26. Khetan R, Curtis R, Deane CM et al (2022) Current advances in biopharmaceutical informatics: guidelines, impact and challenges in the computational developability assessment of antibody therapeutics. MAbs 14:2020082 27. Yamawaki TM, Lu DR, Ellwanger DC et al (2021) Systematic comparison of highthroughput single-cell RNA-seq methods for immune cell profiling. BMC Genomics 22:1– 18 28. Gao C, Zhang M, Chen L (2020) The comparison of two single-cell sequencing platforms: BD rhapsody and 10x genomics chromium. Curr Genomics 21:602–609 29. Picelli S, Bjo¨rklund ÅK, Faridani OR et al (2013) Smart-seq2 for sensitive full-length transcriptome profiling in single cells. Nat Methods 10:1096–1100 30. Wang X, He Y, Zhang Q et al (2021) Direct comparative analyses of 10X genomics chromium and smart-seq2. Genom Proteom Bioinforma 19:253–266 31. See P, Lum J, Chen J et al (2018) A single-cell sequencing guide for immunologists. Front Immunol 9:2425 32. Safdari Y, Farajnia S, Asgharzadeh M et al (2013) Antibody humanization methods - a review and update. Biotechnol Genet Eng Rev 29:175–186

212

Nathaniel Pascual et al.

33. Jumper J, Evans R, Pritzel A et al (2021) Highly accurate protein structure prediction with AlphaFold. Nature 596:583–589 34. Nivo´n LG, Moretti R, Baker D (2013) A Pareto-optimal refinement method for protein design scaffolds. PLoS One 8:e59004 35. Mendenhall J, Brown BP, Kothiwale S et al (2021) BCL::Conf: improved open-source knowledge-based conformation sampling using the crystallography open database. J Chem Inf Model 61:189–201 36. Rabia LA, Desai AA, Jhajj HS et al (2018) Understanding and overcoming trade-offs between antibody affinity, specificity, stability and solubility. Biochem Eng J 137:365–374

37. Raybould MIJ, Marks C, Krawczyk K et al (2019) Five computational developability guidelines for therapeutic antibody profiling. Proc Natl Acad Sci U S A 116:4025–4030 38. 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 39. Horns F, Quake SR (2020) Cloning antibodies from single cells in pooled sequence libraries by selective PCRF. PLoS One 15:e0236477 40. Thean RKR, Ong DXY, Heng ZSL et al (2021) To plate or to simply unfreeze, that is the question for optimal plasmid extraction. J Biomol Tech 32:57–62

Chapter 12 One-Pot Droplet RT-OE-PCR for the Generation of Natively Paired Antibody Immune Libraries Desislava Yanakieva, Lena Vollmer, Satyendra Kumar, Stefan Becker, Lars Toleikis, Lukas Pekar, Harald Kolmar, Stefan Zielonka, and Simon Krah Abstract Classical yeast surface display (YSD) antibody immune libraries are generated by a separate amplification of heavy- and light-chain antibody variable regions (VH and VL, respectively) and subsequent random recombination during the molecular cloning procedure. However, each B cell receptor comprises a unique VH-VL combination, which has been selected and affinity matured in vivo for optimal stability and antigen binding. Thus, the native variable chain pairing is important for the functioning and biophysical properties of the respective antibody. Herein, we present a method for the amplification of cognate VH-VL sequences, compatible with both next-generation sequencing (NGS) and YSD library cloning. We employ a single B cell encapsulation in water-in-oil droplets, followed by a one-pot reverse transcription overlap extension PCR (RT-OE-PCR), resulting in a paired VH-VL repertoire from more than a million B cells in a single day. Key words Droplet PCR, Natively heavy-light-chain paired antibody repertoires, Immune library, Next-generation sequencing, Yeast surface display

1

Introduction Droplet microfluidics has emerged as a versatile technology for single-cell-based assays [1]. By high-throughput encapsulation of single cells in monodisperse aqueous droplets, compartmentalization and simultaneous sample miniaturization are achieved [2]. This approach has enabled the development of a row of cutting-edge technologies such as single-cell phenotyping [3], directed enzyme evolution [4], DNA quantitation by droplet digital PCR (ddPCR) [5], single-cell multi-omics analysis [6], and generation of natively paired antibody libraries [7, 8]. While in vitro antibody display platforms like phage display and YSD have found wide application by standardizing and accelerating the antibody discovery process [9–11], advances in next-generation

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2681, https://doi.org/10.1007/978-1-0716-3279-6_12, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

213

214

Desislava Yanakieva et al.

DNA sequencing (NGS) opened the door to in silico-based antibody selection by sequencing whole antibody repertoires (Ig-seq) [12, 13]. In this regard, the native VH-VL pairing is of particular importance for the selection of clonally expanded antibody variants with a high probability of antigen specificity [14]. Thus, a significant effort has been put into developing strategies for the recovery of cognate VH-VL sequences [15–18]. At the same time, the implementation of these techniques has been reported for the generation of natively paired in vitro display libraries. Compared to randomly paired approaches, natively paired libraries are of higher sensitivity and specificity [7, 8]. Furthermore, the repertoire coverage of in vitro libraries is limited by the transformation efficiencies of the antibodyexpressing host cells and screening/selection capabilities [19]. Therefore, in the case of combinatorial libraries the theoretical antibody diversity quickly exceeds the technically manageable library size due to random heavy-light-chain shuffling. In contrast, natively paired libraries reconstitute the initial antibody repertoire more precisely at significantly smaller library sizes. Here, we describe a method for the generation of PCR products encoding for natively paired human antibody sequences from immunized transgenic rodents (primers were designed according to OmniRat antibody diversities; by using adapted primer sequences, the protocol can also be applied to generate PCR products from different transgenic rodents, wild-type mice, or humans). Native pairing is archived by linking cognate VH-VL sequences in a one-pot single B cell reverse transcription overlap extension PCR (RT-OE-PCR) (Fig. 1). The generated amplicons are compatible with both NGS and in vitro antibody display library generation via Golden Gate cloning (GGC).

2

Materials Prepare all solutions using ultrapure sterile water (18 MΩ cm) obtained from a water purification system and autoclave for 20 min at 120 °C. Use analytical grade reagents and store them at 4 °C or at -20 °C when indicated. Follow all waste disposal regulations when disposing of waste materials.

2.1 Reagents for Preparation of SingleCell Suspensions from Lymphatic Tissues

1. Tissue dissociation medium (DPBS + 1 mM EDTA): Dilute 1 mL 0.5 M ethylenediaminetetraacetic acid disodium salt solution in 500 mL Gibco™ DPBS (no Ca2+, no Mg2+) under sterile conditions. Store at 4 °C. 2. Culture dishes: 100 mm. 3. Syringes: 3 cc.

Droplet-Based VH-VL Paired Immune Library Generation

215

Fig. 1 Scheme of the one-pot droplet-based RT-OE-PCR procedure for generation of natively paired VH-VL amplicons: B cells are isolated from human blood or lymphatic tissues from immunized animals. Enriched B cells are encapsulated in water-in-oil droplets together with an RT-OE-PCR mix. Cell lysis, reverse transcription, and overlap extension PCR are facilitated in a one-pot reaction using a standard PCR cycling device. Following the PCR, the emulsion is broken, and the PCR products are purified and subjected to a second amplification in a nested PCR

4. Cell strainer: 70 μm. 5. Conical tubes: 50 mL. 6. Serological pipettes: 5 mL and 10 mL. 2.2 Reagents for B Cell Isolation

1. Cell counter (e.g., Vi-CELL Blue from Beckman Coulter). 2. 5 mL round-bottom polystyrene tube. 3. EasySep Rat B Cell Isolation Kit (STEMCELL Technologies). 4. Separation buffer: RoboSep buffer (STEMCELL Technologies) or DPBS + 2% heat-inactivated FBS (Gibco) + 1 mM EDTA. 5. Separation magnet: Technologies).

EasySep

Magnet

6. DNA LoBind tubes 5 mL (Eppendorf).

(STEMCELL

216

Desislava Yanakieva et al.

2.3 Reagents for B Cell Staining for Flow Cytometry

1. 96-well round (U) bottom plate. 2. PE anti-rat CD3 antibody (BioLegend). 3. APC anti-rat CD45RA antibody (BioLegend). 4. Staining buffer (PBS + 0.5% BSA): Dissolve 0.1 g bovine serum albumin (Sigma-Aldrich) in 20 mL Gibco DPBS (no Ca2+, no Mg2+). Vortex vigorously and keep at 4 °C (see Note 1).

2.4

Oligonucleotides

2.5 Reagents for Droplet RT-OE-PCR Mix (2×)

Primers for the generation of amplicons encoding for natively paired antibodies from transgenic rodents (OmiRats) are shown in Table 1 (RT-OE-PCR) and Table 2 (nested PCR) (see Note 2, Note 3, and Note 4). The amplification scheme is illustrated in Fig. 2. 1. RTX(-exo) polymerase: expression and purification according to Bhadra et al. (2021) protocol [20]. 2. 10×RTX buffer: 600 mM Tris–HCl, 250 mM (NH4)2SO4, 100 mM KCl, 20 mM MgSO4, pH 8.4. Add 10 mL of water to a 500 mL glass baker. Weigh 9.45 g Tris base and transfer it to the baker, and add 3.30 g (NH4)2SO4, 0.75 g MgSO4, and 0.12 g KCl. Add water to 80 mL and adjust pH to 8.4 with HCl. Finally, fill the baker to 100 mL with water. Aliquot at 10 mL and store the buffer at -20 °C for long-term storage and 4 °C for short-term storage (1 month) (see Note 5). 3. 5 M Betaine = Q Solution (Qiagen) (see Note 4). 4. 10 mM dNTP solution mix (see Note 4). 5. 200 mM DTT solution: Weigh 0.309 g 1,4-dithiothreitol in a 50 mL tube. Fill with 10 mL DEPC-H2O and vortex to dissolve. Filter through a 0.22 μm sterile filter, aliquot in 1 mL tubes, and store at 20 °C (see Note 6). 6. Tween-20: Aliquot 1 mL Tween-20. Store at room temperature under light exclusion (see Note 7). 7. 50 mg/mL Ultrapure BSA (Invitrogen) (see Note 4). 8. 40 U/μL RNAseOUT (Invitrogen) (see Note 4). 9. Primers according to Table 1. 10. DEPC-H2O (Invitrogen): Prepare 1 mL aliquots and store at -20 °C.

2.6 Preparation of B Cells for Droplet Encapsulation

1. DNA LoBind tubes 1.5 mL (Eppendorf). 2. Cell wash buffer (2×RTX buffer): Dilute 2 mL 10×RTX buffer with 8 mL sterile water. Filter through a 0.22 μm sterile filter. Store at 4 °C. 3. OptiPrep density gradient medium (Sigma-Aldrich). Store at 4 °C.

Droplet-Based VH-VL Paired Immune Library Generation

217

Table 1 Oligonucleotide primers for OmniRat droplet PCR (5′-3′) (capitalized = primer overhang) Fab-VH in mix

Fab hVH1-OE-in

TCTAGAAGAGCTAACGTATGCTCTTCTGGT caggtccagctkgtrcagtctgg Fab hVH157-OE- TCTAGAAGAGCTAACGTATGCTCTTCTGGT in caggtgcagctggtgsartctgg Fab hVH2-OE-in TCTAGAAGAGCTAACGTATGCTCTTCTGGT cagrtcaccttgaaggagtctg Fab hVH3-OE-in TCTAGAAGAGCTAACGTATGCTCTTCTGGT gaggtgcagctgktggagwcy Fab hVH4-OE-in TCTAGAAGAGCTAACGTATGCTCTTCTGGT caggtgcagctgcaggagtcsg Fab hVH4-DP63- TCTAGAAGAGCTAACGTATGCTCTTCTGGT OE-in caggtgcagctacagcagtggg Fab hVH6-OE-in TCTAGAAGAGCTAACGTATGCTCTTCTGGT caggtacagctgcagcagtca Fab hVH3N-OE- TCTAGAAGAGCTAACGTATGCTCTTCTGGT in tcaacacaacggttcccagtta

Fab-VK in mix

Fab hVK1-OE-in Fab hVK2-OE-in Fab hVK3-OE-in Fab hVK5-OE-in

ACCAGAAGAGCATACGTTAGCTCTTCTAGA gacatccrgdtgacccagtctcc ACCAGAAGAGCATACGTTAGCTCTTCTAGA gatattgtgmtgacbcagwctcc ACCAGAAGAGCATACGTTAGCTCTTCTAGA gaaattgtrwtgacrcagtctcc ACCAGAAGAGCATACGTTAGCTCTTCTAGA gaaacgacactcacgcagtctc

Fab-VL in mix

Fab hVL1-OE-in

ACCAGAAGAGCATACGTTAGCTCTTCTAGA cagtctgtsbtgacgcagccgcc Fab hVL1459ACCAGAAGAGCATACGTTAGCTCTTCTAGA OE-in cagcctgtgctgactcaryc Fab hVL15910ACCAGAAGAGCATACGTTAGCTCTTCTAGA OE-in cagccwgkgctgactcagccmcc Fab hVL2-OE-in ACCAGAAGAGCATACGTTAGCTCTTCTAGA cagtctgyyctgaytcagcct Fab hVL3-OE-in ACCAGAAGAGCATACGTTAGCTCTTCTAGA tcctatgwgctgacwcagccaa Fab hVL-DPL16- ACCAGAAGAGCATACGTTAGCTCTTCTAGA OE-in tcctctgagctgastcaggascc Fab hVL3-38-OE- ACCAGAAGAGCATACGTTAGCTCTTCTAGA in tcctatgagctgayrcagcyacc Fab hVL6-OE-in ACCAGAAGAGCATACGTTAGCTCTTCTAGA aattttatgctgactcagcccc Fab hVL78-OE-in ACCAGAAGAGCATACGTTAGCTCTTCTAGA cagdctgtggtgacycaggagcc

Fab-k out mix

Fab hIgKC-OE out Fab rIgHG-OEout Fab rIgHM-OEout

GCGGATAACAATTTCACACAGG ctgctcatcagatggcgggaagatgaagacagatggtgcag CGCAGTAGCGGTAAACGGCcgctggacagggctccagagttcca CGCAGTAGCGGTAAACGGCcttcagtgttgttctggtagttccaggag (continued)

218

Desislava Yanakieva et al.

Table 1 (continued) Fab-λ out mix

Fab hIgLC-OE out Fab rIgHG-OEout Fab rIgHM-OEout Common fwd Common rev

GCGGATAACAATTTCACACAGGttgragctcctcagaggagggygggaa CGCAGTAGCGGTAAACGGCcgctggacagggctccagagttcca CGCAGTAGCGGTAAACGGCcttcagtgttgttctggtagttccaggag cgcagtagcggtaaacggc gcggataacaatttcacacagg

Table 2 Oligonucleotide primers for nested PCR (5′-3′) Nested-k mix

Fab nested hIgK Fab nested rIgHG Fab nested rIgHM

agatggtgcagccacagttc ggatagacagatggggctgttgtt ggggaagacagttggggaggact

Nested- λ mix

Fab nested hIgL Fab nested rIgHG Fab nested rIgHM

gagggygggaacagagtgac ggatagacagatggggctgttgtt ggggaagacagttggggaggact

4. Cell encapsulation buffer (2×RTX buffer + 20% OptiPrep): Dilute 1 mL 10×RTX buffer and 1 mL OptiPrep with 3 mL DEPC-H2O. Filter through a 0.22 μm sterile filter and store at 4 °C. 5. Flowmi cell strainer—40 μm. 2.7 Droplet Generation

1. μEncapsulator microfluidics system (Dolomite Bio). 2. μEncapsulator 2-reagent droplet chip—50 μm, fluorophilic (Dolomite Bio). 3. μEncapsulator sample reservoir chip—100 μm (Dolomite Bio). 4. QX200 droplet generation oil for EvaGreen (Bio-Rad). 5. HFE-7500 3 M Novec Engineered Fluid (FluoroChem). 6. 20 mL glass vials with screw caps (DWK WHEATON). 7. DNA LoBind tubes 5 mL (Eppendorf). 8. Microchip washing solution 1 (1% SDS solution): Dilute 3 mL 10% sodium dodecyl sulfate (SDS) solution (Fisher Scientific) with 27 mL sterile water. Filter through a 0.22 μm syringe filter and store it at room temperature.

Droplet-Based VH-VL Paired Immune Library Generation

219

Fig. 2 Paired VH-VL amplification strategy in a one-pot droplet PCR. In the first step, VH and VL are reversetranscribed using primer binding to FR1 of VH and VL and the constant domains of heavy and light chains. The FR1 primer carries a complement linker sequence, which promotes the overlap extension reaction. The out primer targeting the constant domains inserts small overhangs, coding for the common primer, which drives the amplification of the linked VH-VL product. Following droplet PCR, emulsion breakage, and DNA purification, a nested PCR is performed using primers binding inside the generated VH-VL product targeting the 5′ regions of the constant domains

9. Microchip washing solution 2 (H2O): Filter 30 mL sterile water through a 0.22 μm syringe filter and store it at room temperature. 10. Microchip washing solution 3 (isopropanol): Filter 30 mL 99.9% 2-propanol (Sigma-Aldrich) through a 0.22 μm syringe filter and store at room temperature. 2.8 Microscopic Analysis

1. Cell counting slides for TC10/TC20 cell counter (Bio-Rad).

2.9 Droplet RT-OEPCR

1. Twin tec PCR plate 96 150 μL (Eppendorf).

2. Inverted microscope (20 × objective).

2. Easy Pierce Heat Sealing Foil (Thermo Scientific). 3. ALPSTM 25 manual heat sealer (Thermo Scientific). 4. C1000 Touch Thermal Cycler (Bio-Rad).

220

Desislava Yanakieva et al.

2.10 Purification of VH-VL PCR Product

1. 1H,1H,2H,2H-Perfluoro-1-octanol (PFO) (Sigma-Aldrich). 2. DNA LoBind tubes 1.5 mL (Eppendorf). 3. Wizard SV Gel and PCR Clean-Up System (Promega). 4. OmniPur water, sterile, nuclease-free (Sigma-Aldrich).

2.11

Nested PCR

1. Platinum™ II Hot-Start PCR Master Mix (2X) (Invitrogen). 2. Primers according to Table 2.

3

Methods The method section describes the procedure of generating PCR products encoding for natively paired antibody VH-VL domains starting with tissue from immunized transgenic rodents (OmniRats). PCR products can be utilized for NGS analysis or antibody display library generation. The amplification process is schematically illustrated in Fig. 2.

3.1 Preparation of Single-Cell Suspensions from Lymphatic Tissues

1. Harvest spleen, lymph nodes, and bone marrow from immunized OmniRats. 2. Transfer the tissues into a sterile 100 mm culture dish with 10 mL dissociation medium. 3. Use the flat end of the plunger from a 3 cc syringe to grind the tissues by crushing them in gentle circular motions (five times). 4. Prime a 70 μm cell strainer with 2 mL dissociation medium. 5. Dissociate the minced lymphatic tissues with a 10 mL serological pipette until a homogeneous mixture is obtained. 6. Transfer the mixture to the primed cell strainer on top of a sterile 50 mL conical tube. Wash the culture dish with 5 mL dissociation buffer and apply it to the cell strainer as well. 7. Use the plunger of a new 3 cc syringe to gently pass the dissociated tissue through the cell strainer (press the flat end with a circular motion). 8. Wash the cell strainer with 5 mL dissociation buffer and discard it. 9. Fill up the 50 mL tube with DPBS and centrifuge at 300 × g for 10 min. 10. Carefully remove and discard the supernatant (see Note 8).

3.2

B Cell Isolation

1. Resuspend the dissociated lymphatic tissue in 20 mL RoboSep buffer and centrifuge at 120 × g for 10 min. 2. Discard the supernatant carefully and resuspend the cells in the remaining volume (ca. 0.5 mL).

Droplet-Based VH-VL Paired Immune Library Generation

221

3. Measure cell density (cell counter) and adjust to 5 × 107 cells/ mL with RoboSep buffer (see Note 9) to a final volume of 0.5–2 mL (see Note 10). 4. Transfer 1 × 105 cells to a 96-well plate (U-bottom) and follow Subheading 3.3. 5. Transfer the remaining cells to a 5 mL polystyrene roundbottom tube and add 50 μL/mL rat B-cell isolation cocktail. 6. Mix the cells gently and incubate at room temperature for 10 min. 7. Add 25 μL/mL RapidSpheres (see Note 11) to the sample and fill the tube up to 2.5 mL with RoboSep buffer. 8. Remove the lid of the tube and incubate for 3 min with the magnet. 9. Pour the sample (without removing the tube from the magnet) into a new 5 mL tube and try to collect the whole supernatant. 10. Put a new tube into the magnet and incubate again for 3 min. 11. Collect the negative fraction by pouring the supernatant into a 5 mL DNA LoBind tube and determine the cell count. 12. Centrifuge at 300 × g for 10 min (see Note 8). 3.3 B Cell Staining for Flow Cytometry

During this step, the efficiency of the B-cell isolation is evaluated. The B cell population should be above 80% of the total cells. 1. Transfer 1 × 105 purified B cells to the same 96-well plate as mentioned in Subheading 3.2 (point 4). 2. Add 20 ng APC anti-rat CD45RA antibody and 40 ng PE antirat CD3 antibody (see Note 12). 3. Fill wells to 100 μL with staining buffer and incubate for 1 h at 4 °C. 4. Centrifuge the plate at 500 × g for 5 min at 4 °C. 5. Wash twice with 150 μL cold staining buffer. 6. Resuspend in 150 μL staining buffer and analyze via cell cytometer. An example of FACS plots is shown in Fig. 3.

3.4 Preparation of Droplet RT-OE-PCR Mix (2×)

1. Thaw all reagents on ice. Place the RTX(-exo) polymerase and the RNAseOUT inhibitor in a -20 °C cooling rack. 2. Mix the following amounts of each reagent to reach a 2×PCR mix with a final volume of 500 μL: 1. 25 μL 10×RTX buffer. 2. 200 μL Betaine (Q Solution). 3. 20 μL dNTPs. 4. 7.2 μL Tween-20 (see Note 13).

222

Desislava Yanakieva et al.

Fig. 3 Flow cytometric analysis of rat B cell enrichment using negative MACS selection. Fresh lymphatic tissues (spleen, bone marrow, and lymph nodes) from a lambda OmniRat animal were processed, and B cell enrichment was performed as described in Subheading 3.2. Rat B cells (CD45RA-positive, CD3-negative) were stained with fluorescently labeled detection antibodies and gated as shown

5. 12.2 μl DTT stock solution. 6. 10 μL Ultrapure BSA solution. 7. Primer (premixed primers with 10 μM of each): (i) 40 μL common primer mix. (ii) 5 μL Fab-VH in mix. (iii) 5 μL Fab-VK in mix or Fab-VL in mix. (iv) 5 μL Fab-κ out mix or Fab-λ out mix. 8. 25 μL RNAseOUT. 9. RTX(-exo) enzyme to a final concentration of 20 μg/mL. 10. Add DEPC-H2O to 500 μL. 3. Mix the PCR mix thoroughly and store on ice until encapsulation. 3.5 B Cell Preparation for Droplet Encapsulation

1. Wash cells from Subheading 3.2 with 1.5 mL cold 2×RTX buffer. 2. Transfer the cell suspension in a 1.5 mL DNA LoBind tube (see Note 14) and centrifuge at 300 × g and 4 °C for 10 min. 3. Carefully remove the supernatant. 4. Resuspend the cells in 300 μL cold encapsulation buffer and measure the cell density (see Note 15).

Droplet-Based VH-VL Paired Immune Library Generation

223

5. Adjust the cell density to 2.2 × 106 cells/mL with encapsulation buffer (see Note 16). 6. Filter the cells through a 40 μM Flowmi cell strainer to remove cell aggregates and other big particles, which may clog the microfluidic chip. 7. Store cell suspension on ice until encapsulation. 3.6 Droplet Generation

1. Turn on all modules of the μEncapsulator system. 2. Place the microfluidic and sample chips on the temperature controller (TSO). 3. Set the TSO to 4 °C and let the microfluidic and sample chips cool down for 5–10 min. 4. Fill two glass vials with 10 mL sterile-filtered (0.22 μm) HFE-7500 fluid. 5. Place the droplet generation oil in the oil pump (see Note 17) and the vials with HFE-7500 in both aqueous phase pumps. 6. Calibrate all pumps using the Novec-7500 setting and tare flow rates and pressure. 7. Apply 100 μL 2×RT-OE-PCR mix and 100 μL cell suspension in the channels of the sample chip. 8. Purge the tubing by setting the pumps to 2000 mbar (see Note 18). 9. Start the encapsulation process by initiating the cell channel first, followed by activation of the 2×RT-OE-PCR mix channel and the oil channel (see Note 19). 10. Start collecting the emulsion after stable droplets without air bubbles are formed. For this, place a 5 mL DNA LoBind tube on ice and collect the droplet emulsion. 11. After the 2 × 100 μL samples are encapsulated, remove the HFE-7500 oil from the sample chip and apply new samples. 12. Repeat the encapsulation process until all cells are emulsified. 13. After the encapsulation process is finished, wash the sample and microfluidic chips sequentially with 100 μL of the three washing solutions (see Note 20). 14. Remove the microfluidic and sample chips from the device and store them protected from dust.

3.7 Microscopic Analysis of Droplet Generation and Stability

1. Directly after droplet generation, apply 2 μL droplet emulsion on a plastic cell counter slide. Wait until the emulsion distributes and add 2 μL droplet generation oil to dilute the droplets and distribute them more evenly. 2. Observe the generated droplets on an inverted microscope using a 20× objective and take images.

224

Desislava Yanakieva et al.

Fig. 4 Microscopy images of cells encapsulated in water-in-oil emulsion directly after droplet generation (on the left) and after RT-OE-PCR cycling program (on the right). Most of the droplets retain their size and monodispersity during the PCR. A small fraction of larger droplets occurs due to droplet coalescence at high temperatures

3. Repeat the procedure after the RT-OE-PCR cycling. 4. Compare the monodispersity and size of the droplets before and after the PCR (exemplarily shown in Fig. 4). 5. Count the cells in a fixed number of droplets (e.g., 100) to calculate the average droplet occupancy. 3.8 Droplet RT-OEPCR

1. Following droplet generation, remove some of the excess oil (lower clear phase in the tube). 2. Distribute the emulsion (opaque phase) in the wells of a 96-well PCR plate—50 μL/well (see Note 21). 3. Seal the plate hermetically using heat sealing foil (see Note 22). 4. Place the plate in a PCR cycling machine and start the protocol as shown in Table 3.

3.9 Purification of VH-VL PCR Product

1. Observe the samples for emulsion breakage after the RT-OEPCR (see Note 23). 2. Pool the samples from the 96-well PCR plate in a 2 mL reaction tube. 3. Remove excess oil from the bottom of the tube (clear layer on the bottom). 4. Add an equivalent volume PFO and invert to break the emulsion. 5. Centrifuge briefly at 1000 × g to separate the aqueous phase (on the top) from the encapsulation oil + PFO mix (on the bottom). 6. Carefully remove most of the bottom layer with a 1 mL pipette (see Note 24).

Droplet-Based VH-VL Paired Immune Library Generation

225

Table 3 Droplet RT-OE-PCR protocol Reverse transcription

68 °C

30 min

Initial denaturing

94 °C

2 min

94 °C 50 °C 68 °C

30 s 30 s 2 min

4 cycles

94 °C 55 °C 68 °C

30 s 30 s 2 min

4 cycles

94 °C 60 °C 68 °C

30 s 30 s 2 min

22 cycles

Final elongation

68 °C

7 min

hold

4 °C

1

7. Transfer the aqueous phase to a new tube, by making sure that no oil is carried over (see Note 25). 8. Mix the sample with an equivalent volume membrane binding solution (Wizard SV Gel and PCR Clean-Up System). 9. Apply 700 μL to a DNA-binding column and centrifuge for 1 min at 16000 × g. Repeat this step until the whole sample has been applied. 10. Wash the column with 1 × 750 μL and 1 × 500 μL by centrifugation at 16000 × g for 1 min. 11. Dry the column by centrifuging for 5 min without lid. 12. For DNA elution—add 50 μL OmniPur water and incubate for 5 min at room temperature. 13. Elute by centrifugation at 16000 × g for 2 min. 14. Store the sample at -20 °C or proceed with nested PCR directly. 3.10 Small-Scale Nested PCR Test

1. Test nested PCR conditions by preparing 3–4 × 50 μL reactions with different template concentrations (1–10%). 2. Mix the following components for each 50 μL PCR: 1. 25 μL Platinum II Hot-Start PCR Master Mix. 2. 1 μL nested primer mix (κ or λ). 3. 0.5–5 μL droplet PCR template. 4. Add to 50 μL with OmniPur water. 3. Run PCR with the protocol shown in Table 4. 4. Analyze PCR using a 2% agarose gel (exemplarily shown in Fig. 5a).

226

Desislava Yanakieva et al.

Table 4 Nested PCR protocol 94 °C

3 min

94 °C 60 °C 72 °C

15 s 15 s 30 s

Final elongation

72 °C

5 min

hold

4 °C

1

Initial denaturing

30 cycles

Fig. 5 Agarose gel electrophoresis of nested PCR. (a) Different template (purified droplet PCR product) amounts (2–10%) were tested in small reaction volumes (50 μL). (b) A large-scale nested PCR (400 μL) using 1 and 2% template was compared by agarose gel electrophoresis. The product was purified and used for further applications 3.11 Large-Scale Nested PCR

1. After identifying the optimal template concentration for nested PCR, perform a large-scale (400 μL) PCR. 2. Mix the following components: 1. 200 μL Platinum II Hot-Start PCR Master Mix. 2. 8 μL nested primer mix (κ or λ). 3. 4–35 μL droplet PCR template. 4. Add 400 μL with OmniPur water. 3. Transfer to 8×PCR tubes (50 μL/tube) and run PCR with the protocol shown in Table 4. 4. Check the PCR product on a 2% agarose gel (exemplarily shown in Fig. 5b). 5. Pool samples and purify product using Wizard SV Gel and PCR Clean-Up System as described in Subheading 3.9.

Droplet-Based VH-VL Paired Immune Library Generation

4

227

Notes 1. Prepare a fresh solution each time on the same day and discard the rest after the experiment. 2. Order all oligonucleotides HPLC purified (for sequences longer than 50 bp, PAGE purification is recommended). 3. Dissolve the primers at 100 μM concentration in DEPC-H2O and prepare 10 μM dilution by mixing the primers as indicated and adding DEPC-H2O. 4. Store at -20 °C. 5. Filter the 10×RTX buffer each time you prepare a 2×RTX buffer dilution using a 0.22 μm sterile filter. 6. DTT is unstable at room temperature. One aliquot should not be subjected to multiple freeze–thaw cycles (max. five times). 7. Store Tween-20 in a dark tube (1.5 mL, brown from Eppendorf) or cover the tube with aluminum foil. 8. You can store the cells by resuspension in heat-inactivated FBS + 10% DMSO and freezing overnight at -80 °C in a Mr. Frosty freezing container (Thermo Scientific) filled with 99.9% isopropanol. The cells should be transferred in a liq. nitrogen container for long-term storage the next day. 9. If you observe cell aggregation after centrifugation, resuspend cell aggregates with a 1 mL pipette tip and pass the cells again through a pre-wetted 70 μm cell strainer. 10. If there are less than 2.5 × 107 cells, resuspend in 0.5 mL buffer and continue with the protocol using the same amounts of the kit’s components. If there are more than 1 × 108 cells—adjust cell density to 5 × 107 cells/mL and split the sample (>2 mL) into two 5 mL round-bottom polystyrene tubes. 11. Prior use, vortex the RapidSpheres tube for at least 30 s. at high speed and observe the suspension—the magnetic particles should appear evenly dispersed and without aggregates. 12. Dilute the antibodies 1:10 in DPBS and store them at 4 °C in the dark. Use 1 μL APC anti-CD45RA and 2 μL PE anti-rat CD3 diluted antibodies for 1 × 105 cells in 100 μL staining buffer. 13. Vortex vigorously for 10 s after adding Tween-20 to the mix. 14. Use 0.5 mL buffer to resuspend the B cells, transfer the cells to the 1.5 mL tube, and wash the walls of the 5 mL tube with 1 mL buffer to collect all remaining B cells. 15. Use a 1:10 dilution of the cells in 2xRTX buffer to measure the cell density.

228

Desislava Yanakieva et al.

16. This cell density in combination with a 50 μm two-channel microfluidic chip results in an average distribution of one cell per 20 droplets (λ = 0.5), and only 0.14% of the droplets contain two or more cells corresponding to ~2.8% mispaired amplicons. If higher throughput is required, the cell density can be increased to up to 4.4 × 106 cells/mL corresponding to one cell per 10 droplets and 0.56% double-occupied droplets (5.7% mispaired amplicons). 17. Use a fresh vial of droplet generation oil (7 mL) each time to avoid contamination and clogging of the chip by dust. You can pool and sterile filter (0.22 μm) the rest of the multiple vials. 18. Start the pumps one after another and wait until the oil starts dropping. Then, turn off the pumps in the same order and quickly screw the tubing adaptor to the chip holder. 19. Start by setting the aqueous phase pumps to 4 μL/min and the oil pump to 10 μL/min. Wait until the flow of the aqueous sample is visible and gradually increase the flow rate of the oil until stable droplet formation is detected. A flow rate ratio of 1: 10 aqueous:oil phase works good as a starting point. Fine-tune the flow rate ratios until monodispersed droplets are generated. Always keep the same flow rate for both aqueous channels (cells and PCR mix) to make sure they are mixed at a 50:50 ratio. A line at the interface of the two solutions (in the droplets) should be visible and can help to adjust the flow rates so a 50: 50 mixture is achieved. 20. Fill each channel of the sample chip with 100 μL washing solution. Remove the oil and the HFE-7500 vials from the pumps. Purge the pumps with air to remove the oil from the tubing (use 4000 mbar to simultaneously dry the tubing). Connect the tubing with the chip holder and wash the microfluidic chip by setting the aqueous channel pumps at 4000 mbar and the oil pump (only air) at 2000 mbar. Wait until the whole sample is floated through the channel and repeat the procedure with the next washing solution. After washing with isopropanol, wait for 5 min while floating air through the channels to dry the chips. 21. Some of the droplet emulsion will stick to the walls of the tube. Use some droplet generation oil to wash the emulsion from the walls and transfer it to the 96-well PCR plate. 22. Sealing the PCR plate with an adhesive foil or regular caps will result in increased emulsion breakage during the PCR due to evaporation processes. 23. Emulsion breakage can occur during the PCR if some of the wells were not sealed completely hermetically. In this case, an aqueous layer on top of the opaque emulsion layer is visible. The emulsion volume is reduced, and the color of the emulsion is more transparent than at the beginning.

Droplet-Based VH-VL Paired Immune Library Generation

229

24. Press the pipette down to its first stop, push the pipette tip through the aqueous phase, and press down to the second stop to flush any sample from the tip. Wait for a couple of seconds for the two phases to settle down and slowly collect most of the oil. Leave 20–50 μL oil so the phase of the aqueous sample remains clearly visible. 25. The remaining oil interferes with the subsequent DNA purification steps. If you by accident carry over some of the oil, centrifuge briefly and transfer the top layer to a new tube. References 1. Joensson HN, Andersson Svahn H (2012) Droplet microfluidics—a tool for single-cell analysis. Angew Chem Int Ed 51(49): 12176–12192 2. Sohrabi S, Kassir N, Moraveji MK (2020) Droplet microfluidics: fundamentals and its advanced applications. RSC Adv 10(46): 27560–27574 3. Brower KK et al (2020) Double emulsion Picoreactors for high-throughput single-cell encapsulation and phenotyping via FACS. Anal Chem 92(19):13262–13270 4. Stucki A, Vallapurackal J, Ward TR, Dittrich PS (2021) Droplet microfluidics and directed evolution of enzymes: an intertwined journey. Angew Chem Int Ed 60(46):24368–24387 5. Hindson CM et al (2013) Absolute quantification by droplet digital PCR versus analog realtime PCR. Nat Methods 10(10):1003–1005 6. Baron M, Yanai I (2017) New skin for the old RNA-Seq ceremony: the age of single-cell multi-omics. Genome Biol 18(1):1–3 7. Adler AS et al (2018) A natively paired antibody library yields drug leads with higher sensitivity and specificity than a randomly paired antibody library. MAbs 10(3):431–443 8. Rajan S et al (2018) Recombinant human B cell repertoires enable screening for rare, specific, and natively paired antibodies. Commun Biol 1(1):1–8 9. Ministro J, Manuel AM, Goncalves J (2019) Therapeutic antibody engineering and selection strategies. In: Current applications of pharmaceutical biotechnology. Springer, Cham, pp 55–86 10. Laustsen AH, Greiff V, Karatt-Vellatt A, Muyldermans S, Jenkins TP (2021) Animal immunization, in vitro display technologies, and machine learning for antibody discovery. Trends Biotechnol 39(12):1263–1273

11. Klemm J, Pekar L, Krah S, Zielonka S (2021) Antibody display systems. In: Introduction to antibody engineering. Springer, Cham, pp 65–96 12. Ravn U et al (2010) By-passing in vitro screening—next generation sequencing technologies applied to antibody display and in silico candidate selection. Nucleic Acids Res 38(21):e193–e193 13. Parola C, Neumeier D, Reddy ST (2018) Integrating high-throughput screening and sequencing for monoclonal antibody discovery and engineering. Immunology 153(1):31–41 14. Curtis NC, Lee J (2020) Beyond bulk singlechain sequencing: getting at the whole receptor. Curr Opin Syst Biol 24:93–99 15. Tanno H et al (2020) A facile technology for the high-throughput sequencing of the paired VH:VL and TCRβ:TCRα repertoires. Sci Adv 6:eaay9093 16. Setliff I et al (2019) High-throughput mapping of B cell receptor sequences to antigen specificity. Cell 179(7):1636–1646. e15, 12 17. Goldstein LD et al (2019) Massively parallel single-cell B-cell receptor sequencing enables rapid discovery of diverse antigen-reactive antibodies. Commun Biol 2(1):1–10 18. Zhang Z et al (2022) Interpreting the B-cell receptor repertoire with single-cell gene expression using Benisse. Nat Mach Intell 4(6):596–604 19. Feldhaus MJ, Siegel RW (2004) Yeast display of antibody fragments: a discovery and characterization platform. J Immunol Methods 290(1–2):69–80 20. Bhadra S, Maranhao AC, Paik I, Ellington AD (2021) A one-enzyme RTqPCR assay for SARS-CoV-2, and procedures for reagent production. Bio-Protoc 11:e3898–e3898

Chapter 13 Affinity Maturation of the Natural Ligand (B7-H6) for Natural Cytotoxicity Receptor NKp30 by Yeast Surface Display Stefan Zielonka, Simon Krah, Paul Arras, Britta Lipinski, Jasmin Zimmermann, Ammelie Svea Boje, Katja Klausz, Matthias Peipp, and Lukas Pekar Abstract In recent years, the development of bispecific antibodies (bsAbs) has experienced tremendous progress for disease treatment, and consequently, a plethora of bsAbs is currently scrutinized in clinical trials. Besides antibody scaffolds, multifunctional molecules referred to as immunoligands have been developed. These molecules typically harbor a natural ligand entity for the engagement of a specific receptor, while binding to the additional antigen is facilitated by an antibody-derived paratope. Immunoligands can be exploited to conditionally activate immune cells, e.g., natural killer (NK) cells, in the presence of tumor cells, ultimately causing target-dependent tumor cell lysis. However, many ligands naturally show only moderate affinities toward their cognate receptor, potentially hampering killing capacities of immunoligands. Herein, we provide protocols for yeast surface display-based affinity maturation of B7-H6, the natural ligand of NK cell-activating receptor NKp30. Key words Affinity maturation, B7-H6, Bispecifics, Fluorescence-activated cell sorting, Immunoligand, NKp30, Protein engineering, Yeast surface display

1

Introduction Antibodies (Abs), secreted by matured and activated B lymphocytes, represent the humoral immune response of the adaptive immune system. Antibodies evolved to recognize specific foreign antigen structures. Using several species like rodents or camelids for immunization and subsequently for the antibody selection processes, specific antibodies with therapeutic potential can be readily isolated [1–5]. Additionally, Abs can be engineered to fulfill different modes of action; e.g., Abs can help dampen or enhance immune reactions. In this regard, canonical antibodies are for instance applied as therapeutics to bind antigen structures on altered cells, thus opsonizing and marking them for destruction by the hosts

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2681, https://doi.org/10.1007/978-1-0716-3279-6_13, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

231

232

Stefan Zielonka et al.

immune cells [6]. Additionally, antibodies themselves are capable to induce antitumoral effects by blocking receptor–ligand interactions and thereby preventing signal transduction in the occupied cell [7]. Alternatively, specific antibodies can also be used for the opposite purpose of inducing receptor downstream signaling, which results in a desired cellular response [8]. However, as diseases are often multifactorial in their origin and manifestation, e.g., heterogeneity of target receptor distribution in solid tumors, the use of canonical monoclonal antibodies (mAbs) is often limited due to their single specificity and constrained modes of action. Consequently, multispecific antibodies have gained continuous attention in the past decade, render them attractive as the “next generation” of antibody therapeutics. Advantageously, they afford binding to different target structures employing a single molecule, potentially resulting in more precise therapies with improved safety profiles and efficacy compared to classical mAbs [9–12]. Furthermore, binding to distinct targets can be exploited to link immune cells via (co-)stimulating receptor engagement to tumor cells addressed via tumor-specific or associated antigens (TAAs), referred to as immune cell redirection [13]. Cross-linking of the immune cell and tumor cell by the bsAb culminates into the conditional activation of the effector cell, which can ultimately result in direct lysis of the target cell, the release of immunomodulatory chemo- and cytokines, and the induction of proliferation or cell differentiation of the immune cell subset. As of today, several bispecific entities with one bispecific fusion protein among them are approved either by FDA or EMA, and a multitude is currently scrutinized in clinical trials [14–16]. For the majority, the engaged immune cells comprise effector T cells, preferentially CD8+ T lymphocytes, although other immune cell subsets, like natural killer (NK) cells, gain substantial interest. NK cells were discovered in the early 1970s due to their ability to spontaneously lyse tumor cells without the need for prior antigen sensitization [17]. NK cells express a variety of germline-encoded activating and inhibitory receptors, whose sophisticated interplay defines the fate of the NK cell [18, 19]. The most prominent activating receptors comprise CD16, NKG2D, DNAM-1, and the natural cytotoxicity receptors (NCRs, i.e., NKp30, NKp44, and NKp46), while distinctive inhibitory receptors are NKG2A and killer cell immunoglobulin-like receptors (KIRs) [19–21]. Eventually, the transduced signals of these opposing activating and inhibitory receptors determine whether the physiologically adjusted balance, leading to a self-tolerant state of the NK cells, gets shifted toward activation, resulting in efficient eradication of the scrutinized cell [17]. In this context, stressed cells, e.g., due to infection or genetic alteration, express stress-induced ligands, such as B7-H6, whose N-terminal V-like domain (“ΔB7-H6”) is recognized by natural cytotoxicity triggering receptor 3 (NKp30, Fig. 1) on NK cells, also

Affinity Maturation of B7-H6 by Yeast Surface Display

233

Fig. 1 Co-crystal structure of NKp30 extracellular domain (gray) with N-terminal V-like domain of B7-H6 (ΔB7H6, blue). Considered residues of ΔB7-H6 at the binding interface with NKp30 for focused randomization are indicated in yellow. Software PyMOL v0.99 based on pdb entry 3PV6 was used for illustration

favoring immune cell activation and target cell lysis [22]. To mimic this induced-self activation state of NK cells, natural ligands have successfully been tested in preclinical settings for robust effector cell recruitment and conditional activation in a target-dependent manner [23, 24]. Equipping multispecific engager molecules with natural ligands as agonistic elements harnesses the evolutionary evolved interaction of the structure to its cognate receptor, thus exploiting a defined and potentially known activation pathway. Natural ligand-derived fusion proteins, termed immunoligands, are accordingly composed of at least one ligand peptide structure commonly fused to an antibody structure, i.e., scFvs or VHHs, for the anchorage to the desired TAA [25]. In addition to redirecting NK cells via ligands of activating NK cell receptors, also bispecific antibodies harboring two antibody-based paratopes are being heavily exploited for triggering the cytotoxic capacity of this effector cell type [26–29]. For that purpose, the utilized target-specific binding moieties are often obtained by screening platform technologies, such as yeast surface display (YSD), enabling the screening of huge diversities to identify desired Hits [5, 30]. YSD technology was initially described by Boder and Wittrup in 1997, employing the yeast strain Saccharomyces cerevisiae for the presentation of the protein of interest (POI) fused to the mating adhesion receptor Aga2p [31]. Natural attachment of the POI fusion molecule Aga2p after expression by disulfide bridges to the cell surface anchor Aga1p thereby allowing for the reliable coupling of the genotype to the phenotype

234

Stefan Zielonka et al.

Fig. 2 Schematic depiction of vector assembly for ΔB7-H6 display plasmid and translation into ΔB7-H6 protein structure presentation system on yeast surface. (a) Gap repair cloning scheme for generation of yeast surface display ΔB7-H6 diversity by homologous recombination. Restriction enzyme-digested destination plasmid 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. (b) Schematic illustration of yeast surface display of ΔB7H6 protein (blue) attached to the cell surface via Aga1p-Aga2p system with C-terminally fused HA epitope enabling simultaneous detection of proper surface expression resulting in possible 2D FACS analysis

(Fig. 2b). Moreover, the eukaryotic background of S. cerevisiae affords the benefit of intrinsic machinery for proper protein folding and consequently degradation of misfolded or aggregated proteins, referred to as unfolded protein response [30]. The compatibility of YSD with fluorescence-activated cell sorting (FACS) renders this technology also attractive for specific screening purposes such as affinity maturations (AFMs) of proteins toward their cognate target structures, as FACS analysis enables online measurements in real time of individual library candidates. Hence, we previously set out to identify variants with increased binding affinities using YSD, as the NK cell-activating receptor ligand B7-H6 shows only moderate affinity for its cognate receptor NKp30 [32]. Thereby, we were able to substantially increase killing capacities of bispecific immunoligands harboring optimized ΔB7-H6 domains and an EGFRdirected Fab for tumor targeting [33]. In this chapter, we provide related protocols for the AFM of NKp30 natural ligand ΔB7-H6 structure using YSD. To this end, the amplification of a synthetically diversified ΔB7-H6 repertoire in a one-step PCR and the engraftment of the PCR products via homologous recombination into a yeast surface display vector, termed gap repair cloning (Fig. 2a), are described. Moreover, the AFM process using three consecutive rounds of FACS selection (Fig. 3a), resulting in the isolation of affinity maturated natural ligand variants compared to ΔB7-H6 wild type (Fig. 3b), is emphasized.

Affinity Maturation of B7-H6 by Yeast Surface Display

235

Fig. 3 Enrichment of affinity enhanced ΔB7-H6 variants by yeast surface display and FACS selection. Simultaneous binding to NKp30 and full-length ΔB7-H6 expression was monitored using a two-dimensional staining strategy. (a) Three consecutive rounds of sorting with 1 μM, 100 nM, and 50 nM NKp30, respectively, were conducted to enrich high-affinity variants. Sorted populations are indicated by applied sorting gates and corresponding percental values, respectively. (b) Direct comparison of final sort round 3 output with parental ΔB7-H6 wild-type variant at 50 nM NKp30 concentration. Percental values of double-positive populations (full-length ΔB7-H6 display and antigen binding) indicated

2 2.1

Materials Strains

1. Saccharomyces cerevisiae strain EBY100 (MATa) (Invitrogen). (URA3-52 trp1 leu2Δ1 his3Δ200 pep4::HIS3 prb1Δ1.6R can1 GAL) (pIU211:URA3). 2. Escherichia coli strain TOP10 distributed by Invitrogen (F-mcrA Δ(mrr-hsdRMS-mcrBC) Φ80lacZΔM15 ΔlacX74 recA1 araD139 Δ(ara leu) 7697 galU galK rpsL (StrR) endA1 nupG).

2.2

Plasmids

Essential features of the final ΔB7-H6 display plasmid are schematically illustrated in Fig. 4. As destination vector (pDest), a pYD-derived plasmid backbone is used harboring substantial features like a replication origin in E. coli (ColE1) and S. cerevisiae (ARS4/CEN6), an ampicillin resistance marker (AmpR) for selection in E. coli and a tryptophan auxotrophic marker for selection in

236

Stefan Zielonka et al.

Fig. 4 Schematic illustration of generated yeast surface display plasmid. The main genetic components of the system are shown. GAL1, GAL1 promoter; Aga2 SP, Aga2 signal peptide; ΔB7-H6, N-terminal V-like domain of NKp30 natural ligand B7-H6; G/S linker, glycine–serine linker; Myc, Myc epitope; Aga2p, Aga2p cell adhesion molecule; HA, hemagglutinin epitope 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 a frame on the used plasmid

EBY100, as well as terminator sequences (not shown). The diversified ΔB7-H6 domain is genetically fused in frame following a GAL1 promotor and an Aga2p signal peptide by digestion of a stuffer region with BsaI and subsequent replacement via gap repair cloning. To this end, the ΔB7-H6 gene is framed with homologous overhangs to the C-terminal glycine-serine linker and the N-terminal Aga2p signal sequence (see Note 1) enabling the insertion by homologous recombination (Fig. 2a). Additionally, the Aga2p sequence for surface attachment and a following hemagglutinin (HA) epitope tag allow for the detection of proper ΔB7-H6 surface expression. 2.3

Media

1. YPD media: Dissolve 10 g yeast extract, 20 g D(+)-glucose, and 20 g peptone in 1 L deionized H2O prior sterilization by autoclaving. Afterward, add 10 mL of penicillin–streptomycin (10,000 Units/mL). Remove particles by sterile filtration utilizing a 0.22 μm bottle top filter.

Affinity Maturation of B7-H6 by Yeast Surface Display

237

2. SD-Trp media: Dissolve 26.7 g minimal SD base in 890 mL deionized H2O prior sterilization by autoclaving. Parallelly, dissolve 1.92 g dropout-mix-Trp (Sigma-Aldrich) and 8.6 g NaH2PO4×H2O and 5.4 g Na2HPO4 in deionized H2O and adjust the volume to 100 mL prior sterilization by autoclaving. Combine both solutions, and add 10 mL of penicillin–streptomycin (10,000 Units/mL). Remove particles by sterile filtration utilizing a 0.22 μm bottle top filter. 3. SD-Trp plates: Dissolve 23.35 g of minimal SD-agar base in 445 mL deionized H2O prior sterilization by autoclaving. Parallelly, dissolve 0.96 g dropout-mix-Trp (Sigma-Aldrich) and 4.28 g of NaH2PO4×H2O and 2.7 g of Na2HPO4 in deionized H2O and adjust the volume to 50 mL prior sterilization by autoclaving. Combine both solutions, add 10 mL of penicillin–streptomycin (10,000 Units/mL), and prepare plates. 4. SG-Trp media: Dissolve 37 g of minimal SD base + Gal/Raf in 490 mL deionized H2O. Parallelly, dissolve 1.92 g dropoutmix-Trp (Sigma-Aldrich) and 8.6 g of NaH2PO4×H2O and 5.4 g of Na2HPO4 in deionized H2O and adjust the volume to 100 mL. Furthermore, dissolve 110 g of PEG8000 in deionized H2O and adjust the volume to 400 mL. Sterilize all solutions by autoclaving prior combination. Afterward, add 10 mL of penicillin–streptomycin (10,000 Units/mL). Remove particles by sterile filtration utilizing a 0.22 μm bottle top filter. 5. SD low-Trp medium: Dissolve 5 g dextrose and 6.7 g yeast nitrogen base (w/o amino acids) in 890 mL deionized H2O prior sterilization by autoclaving. Parallelly, dissolve 1.92 g dropout-mix-Trp (Sigma-Aldrich) and 8.56 g NaH2PO4×H2O and 5.4 g Na2HPO4 in deionized H2O and adjust the volume to 100 mL. Sterilize all solutions by autoclaving prior combination. Afterward, add 10 mL of penicillin– streptomycin (10,000 Units/mL). Remove particles by sterile filtration utilizing a 0.22 μm bottle top filter. 6. Yeast library freezing solution: Dissolve 0.67 g of yeast nitrogen base and 2 g of glycerol 100 mL deionized H2O prior sterile filtration of the solution. 7. LB Amp media: Dissolve 5 g yeast extract, 10 g NaCl, and 10 g peptone in 1 L deionized H2O prior sterilization by autoclaving. Chill medium to approximately 50 °C, then add 1 mL of sterile-filtered ampicillin solution (100 mg/mL in deionized H2O). 8. LB Amp plates: Dissolve 15 g agar, 10 g NaCl, 10 g peptone, and 5 g yeast extract in 1 L deionized H2O prior sterilization by autoclaving. Chill medium to approximately 50 °C, then add

238

Stefan Zielonka et al.

1 mL of sterile filtrated ampicillin solution (100 mg/mL in deionized H2O), and prepare plates. 2.4 Reagents for Amplification of Diversified ΔB7-H6 DNA

1. Primer sequences for ΔB7-H6 gap repair cloning are given in Table 1. 2. Q5 High-Fidelity 2× Master Mix (New England BioLabs). 3. Nuclease-free water. 4. Wizard SV gel and PCR Clean-Up system (Promega).

2.5 Reagents for Destination Vector (pDest) Digestion

1. pDest according to Subheading 2.2. 2. Restriction enzyme BsaI High-Fidelity (20,000 Units/mL, New England BioLabs). 3. Cut Smart Buffer 10× (New England BioLabs). 4. Nuclease-free water. 5. Wizard SV gel and PCR Clean-Up System (Promega).

2.6 Reagents for ΔB7-H6 Transformation into EBY100

1. BsaI digested pDest according to Subheading 3.2. 2. ΔB7-H6 PCR amplicons according to Subheading 3.1. 3. Electroporation buffer: 1 M sorbitol, 1 mM CaCl2 × 2 H2O (autoclaved). 4. LiAc buffer: 100 mM LiAc, 10 mM DTT (sterile-filtered). 5. 1 M sorbitol (autoclaved).

2.7 Reagents for YSD and FACS Analysis

1. Dulbecco’s phosphate-buffered saline (DPBS). 2. Anti-HA tag antibody (PE) (Abcam). 3. Penta-His Alexa Fluor 647 conjugate antibody (Qiagen). 4. Target protein, His-tagged.

2.8

Equipment

1. 0.22 μm Steriflip and Steritop filtration units. 2. Thermocycler. 3. BioSpec (VWR) Nano or equivalent instrumentation. 4. Device and reagents for agarose gel electrophoresis. 5. Benchtop centrifuge. 6. Baffled flasks (150 mL–3 L volume). 7. Shaking incubator (20 °C, 30 °C, and 37 °C). 8. Cell density meter. 9. 0.2 cm electroporation cuvettes (Bio-Rad). 10. Electroporator Gene Pulser Xcell™ (Bio-Rad). 11. Flow cytometry device (e.g., Cell Sorter SH800S, Sony).

Affinity Maturation of B7-H6 by Yeast Surface Display

239

Table 1 Oligonucleotide primers used in this study Name

Sequence (5′–3′)

B7-H6_up

ATGCAGTTACTTCGCTGTTTTTCAATATTTTCTG

B7-H6_lo

GATTTCTGAAGAAGATTTG

12. MasterPure Yeast DNA Purification Kit (Lucigen). 13. 9-cm petri dishes. 14. Cryogenic vials. 15. Freezing container.

3

Methods The method section describes the generation procedure of a yeast surface display library with subsequent FACS analysis for an in silico designed and synthetically constructed ΔB7-H6 diversity in order to isolate ΔB7-H6 variants with increased affinity for their cognate receptor. In this example, the amino acid randomization scheme was designed based on the co-crystal structure of the NK cellactivating receptor NKp30 extracellular domain (ECD) in its bound orientation with the natural ligand B7-H6 (pdb: 3PV6). In more detail, from this co-crystal structure, eight potentially relevant amino acid residues at the binding interface of the N-terminal V-like domain of B7-H6 and NKp30 were considered for focused randomization procedure via TRIM technology at GeneArt (Thermo Fisher Scientific) (Fig. 1). For an insightful comparison of designed and observed frequencies at the specific amino acid positions, Sanger sequencing was used (Fig. 5). In general, all reagents, concentrations, volumes, and protocols in the following can be utilized for the introduction of synthetic diversities into yeast surface display libraries, if designed properly (see Note 2). Moreover, in Subheading 3.7, we present a selection strategy, which allows for the isolation of affinity optimized ΔB7H6 variants employing a two-dimensional FACS analysis.

3.1 PCR Amplification of NTerminal V-Like Domain of B7-H6 (ΔB7-H6) Diversity

1. Add 25 μL Q5 High-Fidelity 2X Master Mix to a PCR tube placed on ice. Add 1 μL (approx. 60 ng/μL) ΔB7-H6 encoding template DNA and 1 μL corresponding forward and reverse primers, respectively (out of a 10 μM stock, primer sequences are given in Table 1), as well as 22 μL nuclease-free water. The whole ΔB7-H6 diversity was amplified using a defined forward and reverse primer set (see Note 2).

240

Stefan Zielonka et al.

Fig. 5 Comparison of the amino acid distribution between designed and observed frequencies (values on the y-axis in %) at randomized positions of ΔB7-H6. Different colors indicate the respective amino acids expected and verified for each considered position. Amino acids are shown in the conventional 3-letter code. X represents ambiguous result. The randomization process was performed via TRIM technology at GeneArt (Thermo Fisher Scientific). Amino acid frequencies were determined using Sanger sequencing analysis

2. Perform PCR as follows: A step of initial denaturation at 98 °C for 30 s, followed by 35 cycles of 10 s denaturation at 98 °C, 20 s primer annealing at 55 °C, and 45 s elongation at 72 °C, ultimately finalized by a prolonged elongation step at 72 °C for 5 min. 3. Analyze PCR products by 1–2% (w/v) agarose gel electrophoresis. A distinct band at approximately 500 bp should indicate an amplified ΔB7-H6 region. Use the Wizard® SV Gel and PCR Clean-Up System according to the manufacturer’s instruction to purify PCR products. Determine DNA concentration via BioSpec Nano or equivalent device and store PCR products at -20 °C if needed. 3.2 Destination Vector (pDest) Digestion

1. Add 10 μg of pDest (1 mg/mL), 2 μL Cut Smart 10× Buffer, and 0.5 μL BsaI-HF in a final volume of 20 μL (add nucleasefree water to achieve this volume) to a PCR tube placed on ice. 2. Perform digestion reaction protected from light at room temperature overnight.

Affinity Maturation of B7-H6 by Yeast Surface Display

241

3. Analyze digested pDest by 1–2% (w/v) agarose gel electrophoresis. A distinct band at approximately 6000 bp should indicate successful digestion. Purify reaction products using Wizard® SV Gel and PCR Clean-Up System according to the manufacturer’s instruction. 3.3 Yeast Transformation for Library Generation

In this section, we provide, in brief, the experimental steps for the library construction as a modified protocol from Benatuil and colleagues for improved S. cerevisiae yeast transformation [34]. Here, all centrifugation steps to pellet yeast cells are performed for 3 min at 4000 × g. 1. Incubate EBY100 in YPD media at 120 rpm and 30 °C overnight to stationary phase. 2. Inoculate 500 mL fresh YPD media with the overnight culture to an OD600 of about 0.3 (see Note 3). 3. Incubate cells at 120 rpm and 30 °C until the OD600 value is about 1.6–1.9. 4. Pellet cells by centrifugation and remove supernatant. 5. Wash cells twice by resuspension in 250 mL ice-cold water followed by a washing step with 250 mL ice-cold electroporation buffer. 6. Resuspend cells in 100 mL LiAc buffer and incubate at 30 °C and 120 rpm for 30 min in a baffled flask. 7. Pellet cells by centrifugation prior washing once with 250 mL ice-cold electroporation buffer. 8. Resuspend the pellet in electroporation buffer to a final volume of approximately 5 mL. Using 450 μL electrocompetent EBY100 per electroporation reaction, a number of at least ten reactions can be conducted (see Note 4). 9. Mix 450 μL electrocompetent cells with 1 μg digested pDest and 4 μg ΔB7-H6 PCR per reaction. 10. Fill cell-DNA mix in ice-cold 0.2 cm electroporation cuvette and perform electroporation reaction with 2500 V. Transfer cells immediately after reaction into 8 mL of a YPD medium and 1 M sorbitol mixture (1:1 ratio). Incubate the cells at 120 rpm for 1 h at 30 °C (see Note 5). 11. Pellet cells by centrifugation and resuspend cell pellet afterward in 10 mL SD-Trp media to eventually calculate library titer by dilution plating on SD-Trp agar plates (100 μL of stock solution for highest cell concentration). Incubate for 72 h at 30 °C prior counting the number of transformants for calculation of library size. 12. Decant remaining cells in 1 L in SD-Trp media and incubate at 120 rpm for 48 h at 30 °C.

242

Stefan Zielonka et al.

13. Transfer library to SD Low-Trp medium at an OD600 of ~1.0. For proper oversampling, transfer at least a ten-fold excess of cells as calculated by dilution plating. 14. Incubate for another 48 h at 120 rpm and 30 °C. 15. The final yeast display library can be analyzed by FACS (Subheadings 3.6 and 3.7) and cryo-preserved for long-term storage (Subheading 3.5). Of note, to ensure an appropriate quality of the yeast display library, at least 100 clones should be sequenced (authors’ recommendation, Note 6). 3.4 Sequence Analysis of Yeast Cell Display Vector

1. Either pick single clones from serial dilution or directly use cell stock from freshly grown initial yeast library for inoculation of an appropriate volume of SD-Trp medium in a baffled flask. Incubate the cells for 2 days at 30 °C and 120 rpm prior display plasmid isolation using the MasterPure Yeast DNA Purification Kit according to the manufacturer’s instructions. 2. Use 1 μL isolated plasmid for the transformation of 50 μL electrocompetent E. coli TOP10 cells. 3. Incubate the E. coli cells for 1 h in SOC medium without antibiotics prior plating cells on LB Amp agar plates and overnight incubation at 37 °C. Next day, pick single colonies and incubate cells in 600 μL LB Amp medium in an adequate deep well microplate overnight at 37 °C and 700 rpm. 4. Mix 40 μL glycerol solution (50% v/v) with 60 μL of E. coli culture in an appropriate microplate prior deep freezing at 80 °C until shipping for sequencing.

3.5 Cryopreservation for Long-Term Storage of Yeast Cells

1. Grow yeast cells to saturation in SD-Trp medium before harvesting by centrifugation. 2. Resuspend cells in SD Low-Trp medium with an OD600 of 1 and incubate for 48–72 h at 120 rpm and 30 °C. (The authors recommend using an appropriate volume, and consequently, cell number, to ensure library diversity oversampling by at least a factor of 10). 3. Pellet cells by centrifugation and discard the supernatant. 4. Wash cells once with DPBS. Pellet cells subsequently by centrifugation and remove supernatant. 5. Resuspend and adjust cells to approximately 1 × 1010 cells per mL in yeast library freezing solution. 6. Transfer suspensions into cryogenic vials and deep-freeze vials at -80 °C.

Affinity Maturation of B7-H6 by Yeast Surface Display

3.6 Induction of ΔB7-H6 Yeast Surface Expression

243

1. Resuspend thawed aliquot of yeast cells in SD-Trp medium. (The authors recommend that the total number of cells in the starting culture should exceed the calculated library size at least by a factor of 10). 2. Cultivate overnight at 120 rpm and 30 °C. 3. Pellet cells by centrifugation and discard the supernatant. For this, the authors recommend utilizing at least a ten-fold excess of cells compared to the calculated library diversity. 4. Resuspend cells in SG-Trp medium to OD600 of 1 and incubate cells afterward for 2 days at 120 rpm and 20 °C for surface expression of ΔB7-H6.

3.7 FluorescenceActivated Cell Sorting (FACS) Analysis for the Detection of Yeast Surface Display and NKp30 Binding

3.7.1 Surface Display Control

This section provides a labeling strategy to monitor NKp30 ECD binding using an indirect fluorescence staining of the NKp30 ECD histidine tag via a Penta-His Alexa Fluor 647 Conjugate antibody in combination with simultaneous detection of proper surface expression of ΔB7-H6 molecules employing a phycoerythrin (PE)labeled anti-HA tag antibody (see Note 7). Of note, the authors recommend using a series of controls for all performed flow cytometric analyses and FACS to ensure the functionality of employed reagents, ensure adequate surface expression levels of the yeast library, and enable appropriate sorting gate adjustments (Fig. 3). Therefore, staining a positive control, i.e., target-positive molecule (here ΔB7-H6 wild type), a negative control, i.e., target-negative single clone or library, an antibody display control (without antigen), and an untreated control (without reagents) should be performed. It is also noteworthy that all labeling steps are performed on ice and fluorophores are protected from direct light. Furthermore, the herein-described process relates exemplarily to 2 × 107 yeast cells per sample but can simply be scaled up for library sorting (see Note 8). 1. Pellet 2 × 107 cells by centrifugation and discard the supernatant. 2. Wash cells twice with DPBS prior resuspension in 40 μL DPBS, followed by incubation on ice for about 30 min. 3. Wash cells twice with DPBS prior resuspension in 40 μL DPBS containing the detection reagents comprising Penta-His Alexa Fluor 647 Conjugate antibody (diluted 1:20) and anti-HA tag antibody (PE) (diluted 1:20), followed by incubation on ice for about 30 min. 4. Wash cells twice with DPBS prior final resuspension in 600 μL DPBS. Keep the sample on ice and shielded from light until flow cytometric analysis.

244

Stefan Zielonka et al.

3.7.2 Library Staining for Affinity Maturation Purposes

The following section describes the staining protocol for the protein of interest displaying yeast cells regarding ΔB7-H6 surface expression and target binding for affinity maturation purposes. Hence, sorting stringencies were increased during the screening campaign to enrich ΔB7-H6 variants with higher affinities for NKp30 ECD (see Notes 9 and 10). This protocol can be utilized to label positive controls for comparison with the scrutinized diversities and libraries in general (sorting stringency should be adapted regarding desired binder characteristics and protein specificities, Notes 9 and 10). Scale-up for sorting purposes if needed (see Note 8). 1. Pellet 2 × 107 cells by centrifugation and discard the supernatant. 2. Wash cells twice with DPBS prior to resuspension in 40 μL DPBS containing: (a) Antigen with a concentration of 1 μM for the initial sorting round and detection reagents (see Notes 7, 9, and 10). (b) Antigen with a concentration of 100 nM for sorting round 2 and detection reagents. (c) Antigen with a concentration of 50 nM for sorting round 3 and detection reagents. 3. Incubate cells on ice for 60 min shielded from light. 4. Wash cells once in 40 μL DPBS and resuspend cell pellet: (a) In 600 μL DPBS, followed by immediate FACS analysis in “normal” sorting mode for the initial sorting round. (b) In 600 μL DPBS. Incubate on ice for 30 min prior to FACS analysis in “purity” sorting mode for sorting round 2. (c) In 6 mL DPBS. Incubate on ice for 30 min prior to FACS analysis in “ultra-purity” sorting mode for sorting round 3.

3.7.3 Treatment of Sorted Cells After FACS Analysis

This section describes the subsequent proceeding with the sorted cells after FACS analysis for long-term storage and successive cytometric analysis rounds. 1. Inoculate appropriate volume of SD-Trp medium (approximately 1 × 106 cells per 20 mL media) with sorted cells collected in FACS tube. 2. Incubate for 2 days at 120 rpm and 30 °C. 3. Determine cell number (see Note 3) and induce ΔB7-H6 surface expression (see Note 8) by changing the media, according to Subheading 3.6.

Affinity Maturation of B7-H6 by Yeast Surface Display

245

4. Prepare remaining cells for long-term storage as already described in Subheading 3.5.

4

Notes 1. Target overhang regions were sequences of conserved nucleotides on the destination vector (pDest) facilitating the insertion of PCR amplicons via homologous recombination. Therefore, the N-terminal amplicon overhang comprised nucleotides identical to a part of the Aga2p signal peptide, while the overhang at the C-terminus was homologous to the G/S linker sequence. Consequently, the ΔB7-H6 molecules were expressed and presented with a free N-terminus and attached to the surface via C-terminal Aga2p fusion. 2. The synthetic diversity was designed to already harbor the constant N-terminal signal peptide (Fig. 4; Aga2p SP) and the glycine–serine linker and the Myc tag. The used forward primer was conceived to anneal in the signal peptide sequence, while the reverse primer was designed to anneal at the Myc tag encoding region. 3. We suppose that an absorbance value of 1 at 600 nm corresponds to approximately 1 × 107 yeast cells per mL. 4. We recommend adjustment of electrocompetent yeast cells to approximately 5 mL while using about 450 μL per reaction to ensure at least ten complete reactions due to dead volumes and loss during pipetting steps. 5. A number of approximately 1 × 107 transformants can be achieved per electroporation reaction. To cover large diversities, several electroporation reactions can be parallelized to scale up the gap repair cloning process and yield an adequate library size of transformed EBY100 clones. 6. To avoid insufficient results due to quality issues of the generated EBY100 transformants, the authors recommend sequencing the initial library. To this end, a sequencing primer annealing approximately 60 bp upstream of the region of interest (here ΔB7-H6) should be designed to ensure coverage of the desired nucleotide sequence. For reliable sequencing of larger nucleotide regions, designing additional sequencing primers should be considered to ensure coverage of the complete desired region. 7. Different combinations of detection reagents and fluorophores are possible, depending on the application and setup of the utilized cytometer. However, as there is a multitude of labeling reagents commercially available, there should be considered for the selection of fluorescent dyes that no overlapping of

246

Stefan Zielonka et al.

emission spectra occurs, which otherwise need to be compensated properly on the FACS device to ensure correct measurements. 8. To scale-up labeling steps for library sorting purposes, hence allowing for FACS analysis of an adequate number of clones, increase the number of cells, volumes, and amounts of labeling reagents proportionally as the multitude of the “standard” sample with 2 × 107 cells in a volume of 40 μL. Of note, the authors recommend oversampling of calculated diversities by at least a factor of ten. As the screening of large initial libraries (e.g., based on synthetic diversification or non-immunized specimen) is often limited by FACS throughput, a maximum number of cells should be sorted in these cases. However, for successive sorting rounds with reduced diversities, the recommendation with oversampling of at least a factor of ten can be utilized. 9. As the affinity of ΔB7-H6 and NKp30 is only moderate (~0.5–1 μM) and mainly driven by the fast off-rate, the authors used in this particular case a combined incubation step for the ΔB7-H6 presenting yeast cells together with antigen and detection reagents to avoid antigen loss during washing steps, in contrast to the normally separated and step-wise incubations of antigen and detection reagents, respectively. 10. One possibility to enhance stringency is a significant reduction in antigen concentration. However, the authors recommend using an antigen concentration of 1 μM for the initial sorting round and to reduce the concentration gradually when the antigen-binding cell population is enriched and/or combine this antigen reduction with other methods to enhance stringency like increased buffer volumes to force antigen dissociation.

Acknowledgements We would like to thank Michael Busch, Bernhard Valldorf, Harald Kolmar, Daniela Wesch, Hans-Heinrich Oberg, Steffen Krohn, Ammelie Svea Boje, Carina Lynn Gehlert, Lars Toleikis, Simon Krah, Tushar Gupta, and Brian Rabinovich for scientific advice, guidance on overall strategy, and technical assistance. References 1. Lu R-M et al (2020) Development of therapeutic antibodies for the treatment of diseases. J Biomed Sci 27(1). https://doi.org/10. 1186/s12929-019-0592-z

2. Ko¨nning D et al (2017) Camelid and shark single domain antibodies: structural features and therapeutic potential. Curr Opin Struct

Affinity Maturation of B7-H6 by Yeast Surface Display Biol 45:10–16. https://doi.org/10.1016/j. sbi.2016.10.019 3. Grzeschik J et al (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 4. Roth L et al (2019) Facile generation of antibody heavy and light chain diversities for yeast surface display by Golden Gate Cloning. Biol Chem 400(3):383–393. https://doi.org/10. 1515/hsz-2018-0347 5. Valldorf B et al (2022) Antibody display technologies: selecting the cream of the crop. Biol Chem 403(5–6):455–477. https://doi.org/ 10.1515/hsz-2020-0377 6. Lu LL, Suscovich TJ, Fortune SM, Alter G (2018) Beyond binding: antibody effector functions in infectious diseases. Nat Rev Immunol 18(1):46–61. https://doi.org/10. 1038/nri.2017.106 7. Li S, Schmitz KR, Jeffrey PD, Wiltzius JJW, Kussie P, Ferguson KM (2005) Structural basis for inhibition of the epidermal growth factor receptor by cetuximab. Cancer Cell 7(4):301–311. https://doi.org/10.1016/j. ccr.2005.03.003 8. Mayes PA, Hance KW, Hoos A (2018) The promise and challenges of immune agonist antibody development in cancer. Nat Rev Drug Discov 17(7):509–527. https://doi. org/10.1038/nrd.2018.75 9. Brinkmann U, Kontermann RE (2021) Bispecific antibodies. Science 372(6545):916–917. https://doi.org/10.1126/science.abg1209 10. Brinkmann U, Kontermann RE (2017) The making of bispecific antibodies. MAbs 9(2): 1 8 2 – 2 1 2 . h t t p s : // d o i . o r g / 1 0 . 1 0 8 0 / 19420862.2016.1268307 11. Ma J et al (2021) Bispecific antibodies: from research to clinical application. Front Immunol 12. https://doi.org/10.3389/fimmu.2021. 626616 12. Krah S, Kolmar H, Becker S, Zielonka S (2018) Engineering IgG-like bispecific antibodies—an overview. Antibodies 7(3):28. https://doi. org/10.3390/antib7030028 13. Wang S et al (2021) The state of the art of bispecific antibodies for treating human malignancies. EMBO Mol Med 13(9). https://doi. org/10.15252/emmm.202114291 14. Esfandiari A, Cassidy S, Webster RM (2022) Bispecific antibodies in oncology. Nat Rev Drug Discov 21(6):411–412. https://doi. org/10.1038/d41573-022-00040-2

247

15. Gera N (2022) The evolution of bispecific antibodies. Expert Opin Biol Ther 22(8):945–949. https://doi.org/10.1080/14712598.2022. 2040987 16. Labrijn AF, Janmaat ML, Reichert JM, Parren PWHI (2019) Bispecific antibodies: a mechanistic review of the pipeline. Nat Rev Drug Discov 18(8):585–608. https://doi.org/10. 1038/s41573-019-0028-1 17. Abel AM, Yang C, Thakar MS, Malarkannan S (2018) Natural killer cells: development, maturation, and clinical utilization. Front Immunol 9. https://doi.org/10.3389/fimmu.2018. 01869 18. Myers JA, Miller JS (2021) Exploring the NK cell platform for cancer immunotherapy. Nat Rev Clin Oncol 18(2):85–100. https://doi. org/10.1038/s41571-020-0426-7 19. Shimasaki N, Jain A, Campana D (2020) NK cells for cancer immunotherapy. Nat Rev Drug Discov 19(3):200–218. https://doi.org/10. 1038/s41573-019-0052-1 20. Paul S, Lal G (2017) The molecular mechanism of natural killer cells function and its importance in cancer immunotherapy. Front Immunol 8. https://doi.org/10.3389/ fimmu.2017.01124 21. Pende D et al (2019) Killer Ig-Like Receptors (KIRs): their role in NK cell modulation and developments leading to their clinical exploitation. Front Immunol 10. https://doi.org/10. 3389/fimmu.2019.01179 22. Matta J et al (2013) Induction of B7-H6, a ligand for the natural killer cell–activating receptor NKp30, in inflammatory conditions. Blood 122(3):394–404. https://doi.org/10. 1182/blood-2013-01-481705 23. Kellner C et al (2012) Mimicking an induced self phenotype by coating lymphomas with the NKp30 ligand B7-H6 promotes NK cell cytotoxicity. J Immunol 189(10):5037–5046. h t t p s : // d o i . o r g / 1 0 . 4 0 4 9 / j i m m u n o l . 1201321 24. Peipp M et al (2015) HER2-specific immunoligands engaging NKp30 or NKp80 trigger NK-cell-mediated lysis of tumor cells and enhance antibody-dependent cell-mediated cytotoxicity. Oncotarget 6(31). https://doi. org/10.18632/oncotarget.5135 25. Peipp M, Klausz K, Boje AS, Zeller T, Zielonka S, Kellner C (2022) Immunotherapeutic targeting of activating natural killer cell receptors and their ligands in cancer. Clin Exp Immunol. https://doi.org/10.1093/cei/ uxac028 26. Ellwanger K et al (2019) Redirected optimized cell killing (ROCK®): a highly versatile

248

Stefan Zielonka et al.

multispecific fit-for-purpose antibody platform for engaging innate immunity. MAbs 11(5): 8 9 9 – 9 1 8 . h t t p s : // d o i . o r g / 1 0 . 1 0 8 0 / 19420862.2019.1616506 27. Gauthier L et al (2019) Multifunctional natural killer cell engagers targeting NKp46 trigger protective tumor immunity. Cell 177(7): 1701–1713.e16. https://doi.org/10.1016/j. cell.2019.04.041 28. Klewinghaus D et al (2022) Grabbing the bull by both horns: Bovine Ultralong CDR-H3 Paratopes enable engineering of ‘almost natural’ common light chain bispecific antibodies suitable for effector cell redirection. Front Immunol 12. https://doi.org/10.3389/ fimmu.2021.801368 29. Klausz K et al (2022) Multifunctional NK cell– engaging antibodies targeting EGFR and NKp30 elicit efficient tumor cell killing and proinflammatory cytokine release. J Immunol 209(9):1724–1735. https://doi.org/10. 4049/jimmunol.2100970 30. 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 31. 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 32. Li Y, Wang Q, Mariuzza RA (2011) Structure of the human activating natural cytotoxicity receptor NKp30 bound to its tumor cell ligand B7-H6. J Exp Med 208(4):703–714. https:// doi.org/10.1084/jem.20102548 33. Pekar L et al (2021) Affinity maturation of B7-H6 translates into enhanced NK cell– mediated tumor cell lysis and improved proinflammatory cytokine release of bispecific immunoligands via NKp30 engagement. J Immunol 206(1):225–236. https://doi.org/ 10.4049/jimmunol.2001004 34. Benatuil L et al (2010) An improved yeast transformation method for the generation of very large human antibody libraries. Protein Engineering, Design & Selection (PEDS) 23 (4):155–159

Chapter 14 Accessing Transient Binding Pockets by Protein Engineering and Yeast Surface Display Screening Jorge A. Lerma Romero and Harald Kolmar Abstract The binding pocket of some therapeutic targets can acquire multiple conformations that, to some extent, depend on the protein dynamics and the interaction with other molecules. The inability to reach the binding pocket can impose a substantial or even insurmountable barrier for the de novo identification or optimization of small-molecule ligands. Herein, we describe a protocol for the engineering of a target protein and a yeast display FACS sorting strategy to identify protein variants with a stable transient binding pocket with improved binding for a cryptic site-specific ligand. This strategy may facilitate drug discovery using the resulting protein variants with accessible binding pockets for ligand screening. Key words Transient binding pockets, Protein engineering, Yeast surface display, Cell cytometry

1

Introduction In the early stages of drug research and development, it is crucial to have knowledge about the active site of a disease-related protein. Particularly important is to have an NMR, X-ray crystal structure, or a comparative homology model of the protein and information of the ligand binding site localization [1, 2]. The binding site or also called binding pocket is a cavity usually located on the surface or inside the protein, which in most cases binds a natural ligand [3, 4]. While most protein binding pockets are accessible in their ligand-free state and easily visualized by NMR or X-ray crystallography, some proteins present no apparent binding pocket [5– 8]. However, in the presence of a ligand, a cryptic binding site of these proteins can be exposed [5, 7, 9]. Mostly, the discovery of cryptic sites is unforeseen and is only found after the crystal structures of ligand-bound proteins show that the ligand binds in a new transient binding pocket or in a previously known pocket that underwent a conformational change [4, 9]. Small-molecule drug

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2681, https://doi.org/10.1007/978-1-0716-3279-6_14, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

249

250

Jorge A. Lerma Romero and Harald Kolmar

discovery for targets belonging to a large protein family is being hindered due to low selectivity and adverse off-target effects [4, 10, 11]. Proteins with transient binding pockets, which for a long time had been considered undruggable, are now attractive drug targets and give an alternative to redesigning drugs with low selectivity [11]. Unfortunately, finding and characterizing new cryptic sites are not easy tasks. There are several in silico methods, which attempt to find allosteric sites by molecular dynamics simulations [1]. The first hurdle when using computational simulations was to overcome the initial lock-and-key theory of protein–ligand interaction, which set a protein as a rigid molecule that was able to bind a ligand without any kind of conformational changes [6, 12]. Actually, there are many available structure prediction algorithms that take into consideration the protein and ligand flexibility that facilitate the search for potential druggable transient binding pockets for therapeutic targets like β-lactamase, interleukin-2, diverse kinases, FKBP, and heat shock protein 90 [11, 13–20]. Some exemplary methods for new allosteric site identification are virtual high-throughput screening, which involves the screening of thousands of compounds against a therapeutic target and de novo drug design, a structure-based approach that demands a protein structure [21]. Likewise, peptide phage display [22] and tethering (site-directed ligand discovery) [23, 24] are some practical approaches to identifying cryptic sites and low-affinity binders. Once a transient binding pocket was identified, high-affinity ligands development is the next logic step. Even when the localization of a cryptic site in a protein is known, the screening of highaffinity binders is a challenging task. Therefore, a general strategy to identify new or analogs of low-affinity ligands for cryptic sites is required (Fig. 1). To facilitate the access of a new small-molecule library to the transient binding pocket, a protein variant with a stabilized and accessible transient binding pocket (in the absence of a ligand) is advantageous. Herein, we describe a method combining protein engineering and fluorescence-activated cell sorting (FACS) of a yeast display library of the target protein of interest. This method is designed to find variants of a selected protein with a stabilized transient binding pocket with the goal to ease the discovery of ligands that selectively interact with them. Additionally, the variants can be further studied and characterized to gain a better understanding of the protein–ligand interaction, and the dynamics and plasticity of the transient binding pocket of the protein of interest.

Accessing Transient Binding Pockets by YSD and Protein Engineering

251

Fig. 1 Schematic representation of a protein with a confirmed transient binding pocket. To study the cryptic site of a protein, it is necessary to overcome high energy barriers or to stabilize the transient binding pocket by modifying key amino acids in the protein. A protein variant with a stabilized open conformation is helpful to facilitate ligand discovery. Created with BioRender.com

2 Materials 2.1 Protein Engineering 2.1.1 Random Mutagenesis

1. GeneMorph II Technologies).

Random

Mutagenesis

Kit

(Agilent

2. dNTPs. 3. Nuclease-free water. 4. Thermocycler. 5. DpnI (New England BioLabs). 6. Wizard SV Gel and PCR Clean-Up System (Promega) or similar. 7. 1% agarose gel and device for gel electrophoresis. 8. OneTaq® DNA Polymerase (New England BioLabs).

2.1.2 Site Saturation Mutagenesis

1. OneTaq® DNA Polymerase (New England BioLabs). 2. dNTPs. 3. Nuclease-free water. 4. Thermocycler. 5. DpnI (New England BioLabs). 6. Wizard SV Gel and PCR Clean-Up System (Promega) or similar. 7. 1% agarose gel and device for gel electrophoresis.

252

Jorge A. Lerma Romero and Harald Kolmar

Table 1 Utilized primers for random mutagenesis, SSM, amplification, and sequencing of the FKBP51 coding sequence in pCT vector Name

5′-3′ sequence

F67_deg_Fw

GGAAAATTGTCAAATGGAAAGAAGNNKGATTCCAGTCATG

F67_deg_Rv

CATTTCTATCATGACTGGAATCMNNCTTCTTTCCATTTGAC

pCT_FKBP51_fw AGTGGTGGTGGTGGTTCTGGTGGTGGTGGTTCTGGTGGTGGTGGTTC TGCTAGCATGAC pCT_FKBP51_rv TGTTGTTATCAGATCTCGAGCTATTACAAGTCCTCTTCAGAAATAAGC TTTTGCTCGGATCC pCT_seq_up

TACCCATACGACGTTCCAGACTAC

pCT_seq_lo

CAGTGGGAACAAAGTCGATTTTGTTAC

Degenerate positions for oligonucleotides follow the subsequent code: N = G + T + A + C; K = G + T NNK is a degenerated codon with N = any nucleotide and K = G or C. MNN is the complementary codon with M = A or T

2.1.3 Primers for Protein Randomization

See Table 1.

2.2 Yeast Surface Display (YSD) Library Generation and Sorting

SD-Trp: 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 glucose, and 75 μg/mL kanamycin (+14 g/L agar agar for agar plates).

2.2.1 Testing Correct Protein Cell Surface Display and Optimal Ligand Concentration for Library Screening

1. SG-Trp: 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 75 μg/mL kanamycin. 2. Spectrophotometer. 3. Centrifuge. 4. PBS: phosphate-buffered saline, pH 7.4. 5. Anti-c-Myc–Biotin antibody (Miltenyi Biotec). 6. Streptavidin-APC (eBioscience). 7. SAFit-FL [25] (see Note 1). 8. BD Influx™ cell sorter or similar device. 9. CytoFLEX Flow Cytometer (Beckman Coulter) or similar device. 10. Thermocycler. 11. Wizard SV Gel and PCR Clean-Up System (Promega) or similar.

Accessing Transient Binding Pockets by YSD and Protein Engineering

253

12. 1% agarose gel and device for gel electrophoresis. 13. Yeast cells: S. cerevisiae EBY100 [MATa URA3-52 trp1 leu2Δ1 his3Δ200 pep4::HIS3 prb1Δ1.6R can1 GAL (pIU211: URA3)] (Thermo Fisher Scientific). 2.2.2 YSD Library Generation

1. BamHI-HF (New England BioLabs). 2. NheI-HF (New England BioLabs). 3. Wizard SV Gel and PCR Clean-Up System (Promega) or similar. 4. Yeast peptone dextrose (YPD): 20 g/L peptone-casein, 20 g/L glucose, and 10 g/L yeast extract. 5. Platform flask shaker. 6. Spectrophotometer. 7. Deionized water. 8. Electroporation buffer: 1 M sorbitol and 1 mM CaCl2. 9. Conditioning buffer: 0.1 M LiAc and 10 mM DTT. 10. 2 mm Bio-Rad Gene Pulser Cuvette. 11. Bio-Rad Gene Pulser Xcell. 12. Yeast pCT_entry vector (see Note 2 and Fig. 5). 13. 1 M sorbitol. 14. SD-Trp: 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 glucose, and 75 μg/mL kanamycin (+14 g/L agar agar for agar plates). 15. Yeast cells: S. cerevisiae EBY100 [MATa URA3-52 trp1 leu2Δ1 his3Δ200 pep4::HIS3 prb1Δ1.6R can1 GAL (pIU211: URA3)] (Thermo Fisher Scientific).

2.2.3 Library Sorting by FACS

1. SD-Trp: 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 glucose, and 75 μg/mL kanamycin (+14 g/L agar agar for agar plates). 2. SG-Trp: 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 75 μg/mL kanamycin. 3. Spectrophotometer. 4. Platform flask shaker. 5. Centrifuge. 6. PBS: phosphate-buffered saline, pH 7.4. 7. Anti-c-Myc–Biotin antibody (Miltenyi Biotec).

254

Jorge A. Lerma Romero and Harald Kolmar

8. Streptavidin-APC (eBioscience). 9. SAFit-FL [25]. 10. BD Influx™ cell sorter or similar device. 11. Yeast cells: S. cerevisiae EBY100 (generated library). 2.2.4 Single Clone Analysis

1. SD-Trp: 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 glucose, and 75 μg/mL kanamycin (+14 g/L agar agar for agar plates). 2. SG-Trp: 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 75 μg/mL kanamycin. 3. Platform flask shaker. 4. BD Influx™ cell sorter or similar device. 5. CytoFLEX Flow Cytometer (Beckman Coulter) or similar device. 6. 20 nM NaOH aqueous solution. 7. Thermoblock. 8. OneTaq® DNA Polymerase (New England BioLabs). 9. dNTPs. 10. Nuclease-free water. 11. Thermocycler. 12. Wizard SV Gel and PCR Clean-Up System (Promega) or similar. 13. 1% agarose gel and device for gel electrophoresis.

2.3 Production of Identified Protein Variants

1. Q5® High-Fidelity DNA Polymerase (NEB). 2. dNTPs. 3. Nuclease-free water. 4. Thermocycler. 5. Wizard SV Gel and PCR Clean-Up System (Promega) or similar. 6. DpnI (New England BioLabs). 7. 2 mm Bio-Rad Gene Pulser Cuvette. 8. Bio-Rad Gene Pulser Xcell. 9. dYT: 5 g/L NaCl, 16 g/L tryptone, 10 g/L yeast extract (+14 g/L agar agar for agar plates). 10. 75 mg/mL kanamycin. 11. 100 mg/mL ampicillin.

Accessing Transient Binding Pockets by YSD and Protein Engineering

255

12. Bacterial cells: electrocompetent Escherichia coli top 10. 13. Bacterial cells: electrocompetent Escherichia coli BL21 (DE3). 14. Incubator at 37 °C. 15. 1 M isopropyl β-D-1-thiogalactopyranoside (IPTG) stock solution. 16. Centrifuge. 17. IMAC A buffer: 50 mM Tris–HCl, 600 mM NaCl, 20 mM imidazol. 18. IMAC B buffer: 50 mM Tris–HCl, 600 mM NaCl, 500 mM imidazol. 19. Ultrasonic cell disruptor (Bandelin Sonopuls) or another cell disruption device. 20. Cell culture shaker. ¨ KTA Pure 25 M (Cytiva) or similar chromatography system. 21. A 22. HisTrap HP column (Cytiva) or comparable Ni-NTA column. 23. PBS, pH 7.4. 24. Dialysis membrane. 25. 15% SDS-PAGE gel and electrophoretic chamber.

3

Methods This chapter describes a workflow for the generation of protein variants through random mutagenesis and site-directed mutagenesis (see Fig. 2 for general process guidance). The randomized coding sequences of the target protein are used to generate a yeast surface display library, which is screened with a fluorescent conformation-specific ligand to obtain protein variants with a stabilized transient binding pocket. The positive variants can then be expressed and analyzed to determine the binding affinity improvement compared to the wild-type protein. In this exemplary study, the FK506-binding protein 51 (FKBP51) was mutated and screened with a published conformation-specific tracer. The protein engineering and screening strategy resulted in over a dozen of FKBP51 variants with improved affinities up to 34-fold compared to the wild type [26].

3.1 Protein Engineering 3.1.1 Random Mutagenesis

The target gene is randomized following the protocol of the GeneMorph II random mutagenesis kit. 1. Prepare a 50 μL reaction with the forward and reverse primer of the target DNA (pCT_FKBP51_fw and pCT_FKBP51_rv) and adjust the initial amount of target DNA depending on the desired mutation rate as shown in Table 2.

256

Jorge A. Lerma Romero and Harald Kolmar

Fig. 2 Flowchart for generation of protein variants with stabilized transient binding pockets

Accessing Transient Binding Pockets by YSD and Protein Engineering

257

Table 2 Random mutagenesis reaction setup dependent on the required mutation rate. For more information, refer to the manufacturer’s protocol Low rate

Medium rate

High rate

10× Mutazyme II Reaction Buffer

5 μL

5 μL

5 μL

40 mM dNTP mix

1 μL

1 μL

1 μL

pCT FKBP51 primers (10 mM each)

1 μL

1 μL

1 μL

Mutazyme II DNA polymerase

1 μL

1 μL

1 μL

Template

900 ng

250 ng

50 ng

Nuclease-free water

Fill up to 50 μL

Fill up to 50 μL

Fill up to 50 μL

Table 3 Thermocycling conditions for random mutagenesis PCR Temperature

Duration [mm:ss]

Cycles

Initial denaturation

95 °C

2:00

X1

Denaturation Annealing Extension

95 °C 64 °C 72 °C

00:30 00:30 1:00

X30

Final extension

72 °C

10:00

X1

The PCR cycling conditions are described in Table 3. Adjust annealing temperature and extension time depending on the target DNA primers and template length. 2. Cleave methylated parental DNA with five units of the restriction endonuclease DpnI for at least 3 h at 37 °C. 3. Verify successful amplification by 1% (w/v) agarose gel electrophoresis and purify the amplicons using Wizard SV Gel and PCR Clean-Up System (Promega) or similar. 4. Use 1 μL of the amplicon as a template for amplification with OneTaq polymerase. In this step, primers (pCT_FKBP51_fw and pCT_FKBP51_rv) are added that introduce overhangs that are also present in the enzymatically cleaved recipient vector to allow for gap repair cloning in yeast. Perform 25 × 100 μL PCRs in parallel as follows: • 20 μL 5× OneTaq reaction buffer. • 2 μL 10 mM dNPT mix. • 2 μL each primer (10 mM each).

258

Jorge A. Lerma Romero and Harald Kolmar

Table 4 Thermocycling conditions for mutagenized insert amplification using OneTaq polymerase Temperature

Duration [mm:ss]

Cycles

Initial denaturation

94 °C

2:00

X1

Denaturation Annealing Extension

94 °C 64 °C 68 °C

00:20 00:50 00:35

X30

Final extension

68 °C

5:00

X1

• 0.5 μL OneTaq DNA polymerase. • 1 μL template. • Fill up to 50 μL with nuclease-free water. The PCR cycling conditions are described in Table 4. 5. Purify amplified DNA using Wizard SV Gel and PCR CleanUp System (Promega) or similar. At least 240 μg of the insert is required for 20 electroporation reactions for the library generation. If a larger library is required, 12 μg of additional insert DNA is required for each electroporation step. 6. Store DNA insert at -20 °C until further use. 3.1.2 Site Saturation Mutagenesis

For the site saturation mutagenesis, a two-step PCR is performed. Two separate PCR reactions are performed to generate two spliced DNA molecules of the FKBP51 gene with a single mutation and an overlap extension PCR to fuse the two fragments together. 1. First PCR step: Prepare a 50 μL reaction where the forward degenerated primer (Table 1) is paired with the reverse primer of the target DNA and vice versa (e.g., pCT_FKBP51_fw + F67_deg_Rv and pCT_FKBP51_rv + F67_deg_fw). The PCR samples were prepared for a volume of 50 μL each as follows: • 10 μL 5× OneTaq reaction buffer. • 1 μL 10 mM dNPT mix. • 1 μL each primer (10 mM each). • 0.25 μL OneTaq DNA polymerase. • ~ 20 ng template. • Fill up to 50 μL with nuclease-free water. The PCR cycling conditions are described in Table 5. 2. Purify both amplified fragments using Wizard SV Gel and PCR Clean-Up System (Promega) or similar.

Accessing Transient Binding Pockets by YSD and Protein Engineering

259

Table 5 Thermocycling conditions for SSM using OneTaq polymerase Temperature

Duration [mm:ss]

Cycles

Initial denaturation

95 °C

02:00

X1

Denaturation Annealing Extension

95 °C 52–56 °C 68 °C

00:20 00:50 00:25

X30

Extension

68 °C

05:00

X1

Table 6 Thermocycling conditions for overlap extension PCR using OneTaq polymerase Temperature

Duration [mm:ss]

Cycles

Initial denaturation

95 °C

02:00

X1

Denaturation Annealing Extension

95 °C 64 °C 68 °C

00:20 00:50 00:35

X30

Extension

68 °C

05:00

X1

3. Second PCR step: Use 1 μL of each purified product of the first PCR step for an overlap extension PCR. The reaction is made with the target DNA primers (pCT_FKBP51_fw + pCT_FKBP51_rv). • 10 μL 5× OneTaq reaction buffer. • 1 μL 10 mM dNPT mix. • 1 μL primers (10 mM each). • 0.25 μL OneTaq DNA polymerase. • 1 μL of each fragment as template. • 35.75 μL nuclease-free water. The PCR cycling conditions are described in Table 6. 4. Cleave methylated parental DNA with five units of the restriction endonuclease DpnI overnight at room temperature. 5. Purify both amplified fragments using Wizard SV Gel and PCR Clean-Up System (Promega) or similar. 6. Use 1 μL of the purified DNA as a template for amplification with OneTaq polymerase. Perform 25 × 100 μL PCRs in parallel shown in Subheading 3.1.1. Divide the reaction by the number of modified amino acid positions of the protein

260

Jorge A. Lerma Romero and Harald Kolmar

(e.g., eight positions modified, 3 × 100 μL PRCs of each modified residue). 7. Verify successful amplification by 1% (w/v) agarose gel electrophoresis and purify amplified DNA using Wizard SV Gel and PCR Clean-Up System (Promega) or similar. At least 240 μg of the insert is required for 20 electroporation reactions for the library generation. If a larger library is required, 12 μg of additional insert DNA is required for each electroporation step. 8. Store DNA insert at -20 °C until further use. 3.2 Yeast Surface Display (YSD) 3.2.1 Testing Correct Protein Cell Surface Display and Optimal Ligand Concentration for Library Screening

Before generating a yeast library with modified amino acid sequences, it should be confirmed that the wild-type protein can be successfully displayed over the yeast surface of S. cerevisiae EBY100. 1. Transform yeast cells with a pCT vector containing the target gene fused to yeast surface anchored protein Aga2p following the protocol by Benatuil et al. [27–29]. Only one electroporation reaction is necessary. 2. Grow the transformed cells in 50 mL SD-Trp medium overnight at 30 °C and 180 rpm. 3. Induce the transformed yeast cells by inoculating 50 mL SG-Trp at an OD600 of 1.0 with the grown cells in SD-Trp medium overnight at 30 °C, 180 rpm. 4. The following day, measure the OD600 of the cell culture and aliquot in 5 × 1.5 mL reaction tubes 2 × 107 induced yeast cells per tube (1OD = 2 × 107 cells/mL). 5. Harvest the cells by centrifugation, discard the supernatant, and wash the yeast cells with 1 mL of PBS. 6. Incubate the washed yeast cells with a biotin-conjugated c-Myc antibody (diluted 1:75) on ice for 30 min (see Note 3). 7. Wash the yeast cells with 1 mL of PBS. 8. Stain with the secondary labeling reagent Streptavidin-APC (eBioscience) diluted 1:75. Incubate on ice for 30 min. Additionally, add different concentrations of the fluorescently labeled protein-specific ligand (SAFit-FL in the specific case of FKBP51) to four of the five reaction tubes. The fifth reaction tube without tracer serves as a negative control. 9. Wash the yeast cells with 1 mL of PBS and resuspend the cell pellet in 600 μL PBS. 10. Analyze cells using a cell sorter such as a BD Influx™ cell sorter, CytoFLEX Flow Cytometer (Beckman Coulter), or similar devices. If a cell population presents APC fluorescence, the yeast display system has successfully expressed and displayed the protein of interest on the cell surface.

Accessing Transient Binding Pockets by YSD and Protein Engineering

261

11. Sorting gates should be set appropriately, with a low (100 μg of empty pCT with 50 units of BamHI-HF and 50 units of NheI-HF and incubate for 3 h at 37 °C and subsequently at room temperature overnight. 2. Purify the digested pCT plasmid using Wizard SV Gel and PCR Clean-Up System (Promega) or similar. At least 80 μg of digested vector backbone is required for 20 electroporation reactions. 3. Preparation of electrocompetent S. cerevisiae cells (EBY100) starts by inoculating a shaking flask with 1 L YPD medium with the overnight culture at a 0.3 OD600. 4. Cells are washed three times with ice-cold deionized water and twice with ice-cold electroporation buffer. 5. Cells are reconditioned in 200 mL of 0.1 M LiAc/ 10 mM DTT. 6. Collect the conditioned cells by centrifugation in 6 × 50 mL-centrifuge tubes, wash three times with 50 mL ice-cold electroporation buffer, and resuspended the cell pellet with electroporation buffer to reach a final volume of 10 mL. 7. For 20 electroporation reactions, create a master mix with 80 μg of digested vector backbone and 240 μg of the DNA insert (4 μg of digested vector backbone and 12 μg of the DNA insert per electroporation reaction). 1/20 of the total volume of the master mix is to be used for one electroporation reaction. 8. Transfer the cells into 80 mL of 1:1 mix of 1 M sorbitol: YPD media. Incubate on a platform shaker at 225 rpm and 30 °C for 1 h. 9. Collect cells by centrifugation and resuspend in 1 L of SD-Trp media and grow cells at 30 °C overnight. 3.2.3 Library Sorting by FACS

1. Induce the yeast library and the yeast cells transformed with the Wt protein in 50 mL of SG-Trp media overnight at 30 °C, 180 rpm at an OD600 of 1.0. 2. The following day measure the OD600 of the cell cultures and aliquot in 1.5 mL reaction tubes 2 × 109 induced yeast cells (library) per tube. 1OD = 2 × 107 cells/mL The screening should be of at least 10× of the yeast library size determined by plating the transformed yeast cells after the library generation. 3. Harvest 2 × 107 induced yeast cells (Wt) per tube. This will follow the same incubation and staining procedure like the yeast library and will serve as a negative control for FACS sorting. 4. Wash the cells with 1 mL PBS. 5. Incubate the washed yeast cells with a biotin-conjugated c-Myc antibody (diluted 1:75) on ice for 30 min. 6. Wash the yeast cells with 1 mL of PBS.

Accessing Transient Binding Pockets by YSD and Protein Engineering

263

Fig. 4 FACS plots from library screening against FKBP51. As shown exemplarily for this, FKBP51 library cells show surface presentation (c-Myc-tag detection) on the y-axis and ligand binding signal (SAFit-FL) on the x-axis. The sorting gate was set to capture approximately 1% of the tracer binding population

7. Stain with the secondary labeling reagent Streptavidin-APC (eBioscience) diluted 1:75. Incubate on ice for 30 min. Additionally, the amount of the fluorescently labeled proteinspecific ligand (SAFit-FL in the specific case of FKBP51) is determined in Subheading 3.2.1. 8. Wash the yeast cells with 1 mL of PBS and resuspend the cell pellet of three to four reaction tubes in 1 mL PBS. 9. Sort cells using a cell sorter such as a BD Influx™ cell sorter or similar devices. Set a sorting gate to capture approximately 1% of the tracer binding population. An exemplary FACS plot is shown in Fig. 4. 10. Plate the sorted yeast cells on SD-Trp agar plates and then incubate at 30 °C for 48 h. 11. Transfer cells to SD-Trp media and incubate further overnight at 30 °C. 12. Inoculate 50 mL SG-Trp media at an initial OD600 of 1.0 and induce the cells overnight at 30 °C. 13. Repeat the screening procedure as required (see Note 4). 3.2.4 Single Clone Analysis

1. After the last sorting round plate the sorted yeast cells were on SD-Trp agar plates (50 μL per plate) and incubate at 30 °C for 48 h.

264

Jorge A. Lerma Romero and Harald Kolmar

2. Pick single colonies from the agar plate and inoculate 5 mL SG-Trp media and induce the cells overnight at 30 °C and 180 rpm. 3. Follow the steps of Subheading 3.2.3 to stain the selected clones and the induced yeast cells displaying the Wt protein. 4. Set the proper sorting gates and confirm positive clones by a strong shift of the yeast population to the tracer-positive quadrant of a dot plot. 5. From the liquid culture of the positive clones, harvest 20–30 μL by centrifugation and resuspend in 25 μL of 20 nM NaOH aqueous solution. Alternatively, pick the same colony of the positive clone and resuspend in 25 μL of 20 nM NaOH aqueous solution. 6. Incubate for 20 min at 98 °C. 7. Perform a 25 μL PCR for each positive clone using the pCT_seq_up and pCT_seq_lo primers as follows: • 5 μL 5× OneTaq reaction buffer. • 0.5 μL 10 mM dNPT mix. • 0.5 μL each primer (10 mM each). • 0.12 μL OneTaq DNA polymerase. • 1 μL template. • Fill up to 25 μL with nuclease-free water. The PCR cycling conditions are described in Table 7. 8. Verify successful amplification by 1% (w/v) agarose gel electrophoresis and purify amplified DNA using Wizard SV Gel and PCR Clean-Up System (Promega) or similar. 9. Sequence the purified PCR products with the pCT_seq_up or pCT_seq_lo primers to evaluate which mutation(s) in the protein is responsible for the enhanced ligand binding.

Table 7 Thermocycling conditions for single clone check PCR using OneTaq polymerase Temperature

Duration [mm:ss]

Cycles

Initial denaturation

94 °C

2:00

X1

Denaturation Annealing Extension

94 °C 54 °C 68 °C

00:20 00:50 00:45

X30

Final extension

68 °C

5:00

X1

Accessing Transient Binding Pockets by YSD and Protein Engineering

265

Fig. 5 Plasmid map of the yeast pCT_entry vector. The plasmid contains an ampicillin resistance gene for bacterial selection, tryptophan auxotroph gene for yeast selection, BamHI and NheI restriction sites for gap repair cloning of the gene of interest, an Aga2p gene for the surface presentation of the protein of interest, and a Myc tag to verify the correct protein expression 3.3 Production of Identified Protein Variants

1. Sub-clone the mutated target DNA to a proper destination vector depending on the expression organism of choice (pET30 was used for FKBP51 variants). Alternatively, insert a point mutation in an existent plasmid containing the Wt protein (pET30_FKBP51_FK1 Wt. See Note 2 and Fig. 6) by full plasmid amplification with Q5 High-Fidelity DNA Polymerase and no-overlapping primers (e.g., F67E_FKBP51_Fw and F67E_FKBP51_Rv; Table 8) as follows: • 10 μL 5× Q5 Reaction Buffer. • 2 μL 10 mM dNPT mix.

266

Jorge A. Lerma Romero and Harald Kolmar

Fig. 6 Plasmid map of the pET30_FKBP51_FK1 Wt. The plasmid contains a kanamycin resistance gene for bacterial selection, T7 promoter, and lac repressor gene lacI for IPTG induction of the protein of interest (FK1 domain of FKBP51)

• 2.5 μL each primer (10 mM each). • 0.5 μL Q5 High-Fidelity DNA Polymerase. • 20 ng pET30_FKBP51_FK1 Wt. • Fill up to 50 μL with nuclease-free water. The PCR cycling conditions are described in Table 9. 2. Purify the mutated plasmid using Wizard SV Gel and PCR Clean-Up System (Promega) or similar. 3. Cleave methylated parental DNA with five units of the restriction endonuclease DpnI for at least 3 h at 37 °C and overnight at room temperature.

Accessing Transient Binding Pockets by YSD and Protein Engineering

267

Table 8 Utilized primers for full plasmid amplification (no overlapping) for single amino acid exchange of FKBP51 coding sequence in pET30 Name

5′-3′ sequence

F67E_FKBP51_Fw

CGGCAAAAAA GAA GATAGCAGCCATGATCGTAATG

F67E_FKBP51_Rv

TTGCTCAGTTTGCCTTTATAGTGCACATACACTTTATCACC

Table 9 Thermocycling conditions for full plasmid site-directed mutagenesis using Q5 polymerase Temperature

Duration [mm:ss]

Cycles

Initial denaturation

98 °C

1:00

X1

Denaturation Annealing Extension

98 °C 70 °C 72 °C

00:10 00:30 3:15

X25

Final extension

72 °C

5:00

X1

4. Purify the plasmid variant using Wizard SV Gel and PCR Clean-Up System (Promega) or similar. 5. Use 1 μL of the purified plasmid to transform electrocompetent Escherichia coli top 10 cells by electroporation in a 2 mm Bio-Rad Gene Pulser Cuvette with a Gene Pulser Xcell from Bio-Rad at 2500 V and 25 μF. 6. Resuspend electroporated cells in 1 mL of dYT media and incubate at 37 °C for 1 h. 7. Plate cells on a dYT agar plate with 75 μg/mL kanamycin (or appropriate antibiotic for the used plasmid) and incubate at 37 °C overnight. 8. On the next day, pick a couple of colonies, generate a 5 mL dYT pre-culture inoculated with a positive colony, and incubate overnight at 37 °C and 170 rpm. 9. On the next morning, isolate the plasmid from the E. coli overnight culture with the Wizard® Plus Miniprep DNA Purification System Kit from Promega following the manufacturer’s instruction. 10. Use 1 μL of the purified plasmid to transform electrocompetent E. Coli BL21(DE3) cells by electroporation in a 2 mm Bio-Rad Gene Pulser Cuvette with a Gene Pulser Xcell from Bio-Rad at 2500 V and 25 μF. 11. Resuspend electroporated cells in 1 mL of dYT media and incubate at 37 °C for 1 h.

268

Jorge A. Lerma Romero and Harald Kolmar

12. Plate cells on a dYT agar plate with 75 μg/mL kanamycin (or appropriate antibiotic for the used plasmid) and incubate at 37 °C overnight. 13. On the next day pick a colony and inoculate a shaking flask with 50 mL dYT + 75 μg/mL kanamycin. Incubate overnight at 37 °C and 170 rpm. 14. Inoculate a shaking flask with 1 L dYT + 75 μg/mL kanamycin to a 0.1 OD600 and incubate at 37 °C and 190 rpm until a 0.6–0.8 OD600 was reached. 15. Induce gene expression by the addition of 1 mM isopropyl β-D-1-thiogalactopyranoside (IPTG) and incubate at 30 °C and 190 rpm overnight. 16. Harvest the cells by centrifugation (6000 rpm, 10 min) and resuspend the pellet in 25 mL IMAC A buffer. 17. Transfer the suspension into 50 mL centrifuge tube and disrupt the cells by sonication using an ultrasonic cell disruptor (Bandelin Sonopuls) or similar device. 18. Centrifuge at 13500 rpm for 15 min to remove cell debris, filter the supernatant through a 0.45 μm syringe filter, and purify the expressed protein by Ni-NTA affinity chromatography. 19. Dialyze collected protein fractions in PBS pH 7.4. 20. Perform a 15% SDS–PAGE analysis to evaluate the presence and purity of the protein. 3.4 Evaluation of Identified Protein Variants

1. Purified variants can be characterized by determining thermal stability and comparing it to the Wt protein. 2. The affinity increase in the protein can be determined by fluorescence polarization or similar techniques. Conditions are dependent on the protein and ligand of choice. 3. Protein co-crystallization with the ligand can give a better understanding of the molecular basis for the allosteric stabilization /destabilization, which leads to the binding affinity enhancement.

4

Notes 1. If a crystal structure of the protein is available, virtual screening can be used as a first step to identify potential ligands for in vitro screening [13]. 2. Vector sequences.

Accessing Transient Binding Pockets by YSD and Protein Engineering

269

> Yeast pCT_entry vector GACGAAAGGGCCTCGTGATACGCCTATTTTTATAGGTTAATGTCATGATAATAATGGTTTCTTAGGACGGATC GCTTGCCTGTAACTTACACGCGCCTCGTATCTTTTAATGATGGAATAATTTGGGAATTTACTCTGTGTTTATTT ATTTTTATGTTTTGTATTTGGATTTTAGAAAGTAAATAAAGAAGGTAGAAGAGTTACGGAATGAAGAAAAAA AAATAAACAAAGGTTTAAAAAATTTCAACAAAAAGCGTACTTTACATATATATTTATTAGACAAGAAAAGCA GATTAAATAGATATACATTCGATTAACGATAAGTAAAATGTAAAATCACAGGATTTTCGTGTGTGGTCTTCTA CACAGACAAGATGAAACAATTCGGCATTAATACCTGAGAGCAGGAAGAGCAAGATAAAAGGTAGTATTTGT TGGCGATCCCCCTAGAGTCTTTTACATCTTCGGAAAACAAAAACTATTTTTTCTTTAATTTCTTTTTTTACTTTC TATTTTTAATTTATATATTTATATTAAAAAATTTAAATTATAATTATTTTTATAGCACGTGATGAAAAGGACCC AGGTGGCACTTTTCGGGGAAATGTGCGCGGAACCCCTATTTGTTTATTTTTCTAAATACATTCAAATATGTAT CCGCTCATGAGACAATAACCCTGATAAATGCTTCAATAATATTGAAAAAGGAAGAGTATGAGTATTCAACAT TTCCGTGTCGCCCTTATTCCCTTTTTTGCGGCATTTTGCCTTCCTGTTTTTGCTCACCCAGAAACGCTGGTGAA AGTAAAAGATGCTGAAGATCAGTTGGGTGCACGAGTGGGTTACATCGAACTGGATCTCAACAGCGGTAAGAT CCTTGAGAGTTTTCGCCCCGAAGAACGTTTTCCAATGATGAGCACTTTTAAAGTTCTGCTATGTGGCGCGGTA TTATCCCGTATTGACGCCGGGCAAGAGCAACTCGGTCGCCGCATACACTATTCTCAGAATGACTTGGTTGAGT ACTCACCAGTCACAGAAAAGCATCTTACGGATGGCATGACAGTAAGAGAATTATGCAGTGCTGCCATAACCA TGAGTGATAACACTGCGGCCAACTTACTTCTGACAACGATCGGAGGACCGAAGGAGCTAACCGCTTTTTTTC ACAACATGGGGGATCATGTAACTCGCCTTGATCGTTGGGAACCGGAGCTGAATGAAGCCATACCAAACGACG AGCGTGACACCACGATGCCTGTAGCAATGGCAACAACGTTGCGCAAACTATTAACTGGCGAACTACTTACTC TAGCTTCCCGGCAACAATTAATAGACTGGATGGAGGCGGATAAAGTTGCAGGACCACTTCTGCGCTCGGCCC TTCCGGCTGGCTGGTTTATTGCTGATAAATCTGGAGCCGGTGAGCGTGGGTCTCGCGGTATCATTGCAGCACT GGGGCCAGATGGTAAGCCCTCCCGTATCGTAGTTATCTACACGACGGGCAGTCAGGCAACTATGGATGAACG AAATAGACAGATCGCTGAGATAGGTGCCTCACTGATTAAGCATTGGTAACTGTCAGACCAAGTTTACTCATA TATACTTTAGATTGATTTAAAACTTCATTTTTAATTTAAAAGGATCTAGGTGAAGATCCTTTTTGATAATCTCA TGACCAAAATCCCTTAACGTGAGTTTTCGTTCCACTGAGCGTCAGACCCCGTAGAAAAGATCAAAGGATCTTC TTGAGATCCTTTTTTTCTGCGCGTAATCTGCTGCTTGCAAACAAAAAAACCACCGCTACCAGCGGTGGTTTGT TTGCCGGATCAAGAGCTACCAACTCTTTTTCCGAAGGTAACTGGCTTCAGCAGAGCGCAGATACCAAATACT GTCCTTCTAGTGTAGCCGTAGTTAGGCCACCACTTCAAGAACTCTGTAGCACCGCCTACATACCTCGCTCTGC TAATCCTGTTACCAGTGGCTGCTGCCAGTGGCGATAAGTCGTGTCTTACCGGGTTGGACTCAAGACGATAGTT ACCGGATAAGGCGCAGCGGTCGGGCTGAACGGGGGGTTCGTGCACACAGCCCAGCTTGGAGCGAACGACCT ACACCGAACTGAGATACCTACAGCGTGAGCATTGAGAAAGCGCCACGCTTCCCGAAGGGAGAAAGGCGGAC AGGTATCCGGTAAGCGGCAGGGTCGGAACAGGAGAGCGCACGAGGGAGCTTCCAGGGGGGAACGCCTGGTA TCTTTATAGTCCTGTCGGGTTTCGCCACCTCTGACTTGAGCGTCGATTTTTGTGATGCTCGTCAGGGGGGCCGA GCCTATGGAAAAACGCCAGCAACGCGGCCTTTTTACGGTTCCTGGCCTTTTGCTGGCCTTTTGCTCACATGTTC TTTCCTGCGTTATCCCCTGATTCTGTGGATAACCGTATTACCGCCTTTGAGTGAGCTGATACCGCTCGCCGCAG CCGAACGACCGAGCGCAGCGAGTCAGTGAGCGAGGAAGCGGAAGAGCGCCCAATACGCAAACCGCCTCTCC CCGCGCGTTGGCCGATTCATTAATGCAGCTGGCACGACAGGTTTCCCGACTGGAAAGCGGGCAGTGAGCGCA ACGCAATTAATGTGAGTTACCTCACTCATTAGGCACCCCAGGCTTTACACTTTATGCTTCCGGCTCCTATGTTG TGTGGAATTGTGAGCGGATAACAATTTCACACAGGAAACAGCTATGACCATGATTACGCCAAGCTCGGAATT AACCCTCACTAAAGGGAACAAAAGCTGGGTACCCGACAGGTTATCAGCAACAACACAGTCATATCCATTCTC AATTAGCTCTACCACAGTGTGTGAACCAATGTATCCAGCACCACCTGTAACCAAAACAATTTTAGAAGTACTT TCACTTTGTAACTGAGCTGTCATTTATATTGAATTTTCAAAAATTCTTACTTTTTTTTTGGATGGACGCAAAGA AGTTTAATAATCATATTACATGGCATTACCACCATATACATATCCATATACATATCCATATCTAATCTTACTTA TATGTTGTGGAAATGTAAAGAGCCCCATTATCTTAGCCTAAAAAAACCTTCTCTTTGGAACTTTCAGTAATAC GCTTAACTGCTCATTGCTATATTGAAGTACGGATTAGAAGCCGCCGAGCGGGTGACAGCCCTCCGAAGGAAG

270

Jorge A. Lerma Romero and Harald Kolmar

ACTCTCCTCCGTGCGTCCTCGTCTTCACCGGTCGCGTTCCTGAAACGCAGATGTGCCTCGCGCCGCACTGCTC CGAACAATAAAGATTCTACAATACTAGCTTTTATGGTTATGAAGAGGAAAAATTGGCAGTAACCTGGCCCCA CAAACCTTCAAATGAACGAATCAAATTAACAACCATAGGATGATAATGCGATTAGTTTTTTAGCCTTATTTCT GGGGTAATTAATCAGCGAAGCGATGATTTTTGATCTATTAACAGATATATAAATGAATTCTACTTCATACATT TTCAATTAAGATGCAGTTACTTCGCTGTTTTTCAATATTTTCTGTTATTGCTTCAGTTTTAGCACAGGAACTGA CAACTATATGCGAGCAAATCCCCTCACCAACTTTAGAATCGACGCCGTACTCTTTGTCAACGACTACTATTTT GGCCAACGGGAAGGCAATGCAAGGAGTTTTTGAATATTACAAATCAGTAACGTTTGTCAGTAATTGCGGTTC TCACCCCTCAACAACTAGCAAAGGCAGCCCCATAAACACACAGTATGTTTTTAAGGACAATAGCTCGACGAT TGAAGGTAGATACCCATACGACGTTCCAGACTACGCTCTGCAGGCTAGTGGTGGTGGTGGTTCTGGTGGTGG TGGTTCTGGTGGTGGTGGTTCTGCTAGCGTCATCAAGGCATGGGACATTGGGGTGGCTACCATGAAGAAAGG AGAGATATGCCATTTACTGTGCAAACCAGAATATGCATATGGCTCGGCTGGCAGTCTCCCTAAAATTCCCTCG AATGCAACTCTCTTTTTTGAGATTGAGCTCCTTGATTTCAAAGGAGAGGGATCCGAGCAAAAGCTTATTTCTG AAGAGGACTTGTAATAGCTCGAGATCTGATAACAACAGTGTAGATGTAACAAAATCGACTTTGTTCCCACTG TACTTTTAGCTCGTACAAAATACAATATACTTTTCATTTCTCCGTAAACAACATGTTTTCCCATGTAATATCCT TTTCTATTTTTCGTTCCGTTACCAACTTTACACATACTTTATATAGCTATTCACTTCTATACACTAAAAAACTA AGACAATTTTAATTTTGCTGCCTGCCATATTTCAATTTGTTATAAATTCCTATAATTTATCCTATTAGTAGCTA AAAAAAGATGAATGTGAATCGAATCCTAAGAGAATTGAGCTCCAATTCGCCCTATAGTGAGTCGTATTACAA TTCACTGGCCGTCGTTTTACAACGTCGTGACTGGGAAAACCCTGGCGTTACCCAACTTAATCGCCTTGCAGCA CATCCCCCCTTCGCCAGCTGGCGTAATAGCGAAGAGGCCCGCACCGATCGCCCTTCCCAACAGTTGCGCAGC CTGAATGGCGAATGGCGCGACGCGCCCTGTAGCGGCGCATTAAGCGCGGCGGGTGTGGTGGTTACGCGCAGC GTGACCGCTACACTTGCCAGCGCCCTAGCGCCCGCTCCTTTCGCTTTCTTCCCTTCCTTTCTCGCCACGTTCGC CGGCTTTCCCCGTCAAGCTCTAAATCGGGGGCTCCCTTTAGGGTTCCGATTTAGTGCTTTACGGCACCTCGAC CCCAAAAAACTTGATTAGGGTGATGGTTCACGTAGTGGGCCATCGCCCTGATAGACGGTTTTTCGCCCTTTGA CGTTGGAGTCCACGTTCTTTAATAGTGGACTCTTGTTCCAAACTGGAACAACACTCAACCCTATCTCGGTCTA TTCTTTTGATTTATAAGGGATTTTGCCGATTTCGGCCTATTGGTTAAAAAATGAGCTGATTTAACAAAAATTTA ACGCGAATTTTAACAAAATATTAACGTTTACAATTTCCTGATGCGGTATTTTCTCCTTACGCATCTGTGCGGTA TTTCACACCGCAGGCAAGTGCACAAACAATACTTAAATAAATACTACTCAGTAATAACCTATTTCTTAGCATT TTTGACGAAATTTGCTATTTTGTTAGAGTCTTTTACACCATTTGTCTCCACACCTCCGCTTACATCAACACCAA TAACGCCATTTAATCTAAGCGCATCACCAACATTTTCTGGCGTCAGTCCACCAGCTAACATAAAATGTAAGCT TTCGGGGCTCTCTTGCCTTCCAACCCAGTCAGAAATCGAGTTCCAATCCAAAAGTTCACCTGTCCCACCTGCT TCTGAATCAAACAAGGGAATAAACGAATGAGGTTTCTGTGAAGCTGCACTGAGTAGTATGTTGCAGTCTTTT GGAAATACGAGTCTTTTAATAACTGGCAAACCGAGGAACTCTTGGTATTCTTGCCACGACTCATCTCCATGCA GTTGGACGATATCAATGCCGTAATCATTGACCAGAGCCAAAACATCCTCCTTAGGTTGATTACGAAACACGC CAACCAAGTATTTCGGAGTGCCTGAACTATTTTTATATGCTTTTACAAGACTTGAAATTTTCCTTGCAATAACC GGGTCAATTGTTCTCTTTCTATTGGGCACACATATAATACCCAGCAAGTCAGCATCGGAATCTAGAGCACATT CTGCGGCCTCTGTGCTCTGCAAGCCGCAAACTTTCACCAATGGACCAGAACTACCTGTGAAATTAATAACAG ACATACTCCAAGCTGCCTTTGTGTGCTTAATCACGTATACTCACGTGCTCAATAGTCACCAATGCCCTCCCTCT TGGCCCTCTCCTTTTCTTTTTTCGACCGAATTAATTCTTAATCGGCAAAAAAAGAAAAGCTCCGGATCAAGAT TGTACGTAAGGTGACAAGCTATTTTTCAATAAAGAATATCTTCCACTACTGCCATCTGGCGTCATAACTGCAA AGTACACATATATTACGATGCTGTCTATTAAATGCTTCCTATATTATATATATAGTAATGTCGTTTATGGTGCA CTCTCAGTACAATCTGCTCTGATGCCGCATAGTTAAGCCAGCCCCGACACCCGCCAACACCCGCTGACGCGCC CTGACGGGCTTGTCTGCTCCCGGCATCCGCTTACAGACAAGCTGTGACCGTCTCCGGGAGCTGCATGTGTCAG AGGTTTTCACCGTCATCACCGAAACGCGCGA AmpR Aga2p Myc-Tag Gal1-Promoter Trp1 BamHI-HF NheI-HF

Accessing Transient Binding Pockets by YSD and Protein Engineering

271

See the plasmid map in Fig. 5. > Bacterial pET30_FKBP51_FK1 Wt TGGCGAATGGGACGCGCCCTGTAGCGGCGCATTAAGCGCGGCGGGTGTGGTGGTTACGCGCAGCGTGACCGC TACACTTGCCAGCGCCCTAGCGCCCGCTCCTTTCGCTTTCTTCCCTTCCTTTCTCGCCACGTTCGCCGGCTTTCC CCGTCAAGCTCTAAATCGGGGGCTCCCTTTAGGGTTCCGATTTAGTGCTTTACGGCACCTCGACCCCAAAAAA CTTGATTAGGGTGATGGTTCACGTAGTGGGCCATCGCCCTGATAGACGGTTTTTCGCCCTTTGACGTTGGAGT CCACGTTCTTTAATAGTGGACTCTTGTTCCAAACTGGAACAACACTCAACCCTATCTCGGTCTATTCTTTTGAT TTATAAGGGATTTTGCCGATTTCGGCCTATTGGTTAAAAAATGAGCTGATTTAACAAAAATTTAACGCGAATT TTAACAAAATATTAACGTTTACAATTTCAGGTGGCACTTTTCGGGGAAATGTGCGCGGAACCCCTATTTGTTT ATTTTTCTAAATACATTCAAATATGTATCCGCTCATGAATTAATTCTTAGAAAAACTCATCGAGCATCAAATG AAACTGCAATTTATTCATATCAGGATTATCAATACCATATTTTTGAAAAAGCCGTTTCTGTAATGAAGGAGAA AACTCACCGAGGCAGTTCCATAGGATGGCAAGATCCTGGTATCGGTCTGCGATTCCGACTCGTCCAACATCA ATACAACCTATTAATTTCCCCTCGTCAAAAATAAGGTTATCAAGTGAGAAATCACCATGAGTGACGACTGAA TCCGGTGAGAATGGCAAAAGTTTATGCATTTCTTTCCAGACTTGTTCAACAGGCCAGCCATTACGCTCGTCAT CAAAATCACTCGCATCAACCAAACCGTTATTCATTCGTGATTGCGCCTGAGCGAGACGAAATACGCGATCGC TGTTAAAAGGACAATTACAAACAGGAATCGAATGCAACCGGCGCAGGAACACTGCCAGCGCATCAACAATA TTTTCACCTGAATCAGGATATTCTTCTAATACCTGGAATGCTGTTTTCCCGGGGATCGCAGTGGTGAGTAACC ATGCATCATCAGGAGTACGGATAAAATGCTTGATGGTCGGAAGAGGCATAAATTCCGTCAGCCAGTTTAGTC TGACCATCTCATCTGTAACATCATTGGCAACGCTACCTTTGCCATGTTTCAGAAACAACTCTGGCGCATCGGG CTTCCCATACAATCGATAGATTGTCGCACCTGATTGCCCGACATTATCGCGAGCCCATTTATACCCATATAAA TCAGCATCCATGTTGGAATTTAATCGCGGCCTAGAGCAAGACGTTTCCCGTTGAATATGGCTCATAACACCCC TTGTATTACTGTTTATGTAAGCAGACAGTTTTATTGTTCATGACCAAAATCCCTTAACGTGAGTTTTCGTTCCA CTGAGCGTCAGACCCCGTAGAAAAGATCAAAGGATCTTCTTGAGATCCTTTTTTTCTGCGCGTAATCTGCTGC TTGCAAACAAAAAAACCACCGCTACCAGCGGTGGTTTGTTTGCCGGATCAAGAGCTACCAACTCTTTTTCCGA AGGTAACTGGCTTCAGCAGAGCGCAGATACCAAATACTGTCCTTCTAGTGTAGCCGTAGTTAGGCCACCACTT CAAGAACTCTGTAGCACCGCCTACATACCTCGCTCTGCTAATCCTGTTACCAGTGGCTGCTGCCAGTGGCGAT AAGTCGTGTCTTACCGGGTTGGACTCAAGACGATAGTTACCGGATAAGGCGCAGCGGTCGGGCTGAACGGGG GGTTCGTGCACACAGCCCAGCTTGGAGCGAACGACCTACACCGAACTGAGATACCTACAGCGTGAGCTATGA GAAAGCGCCACGCTTCCCGAAGGGAGAAAGGCGGACAGGTATCCGGTAAGCGGCAGGGTCGGAACAGGAG AGCGCACGAGGGAGCTTCCAGGGGGAAACGCCTGGTATCTTTATAGTCCTGTCGGGTTTCGCCACCTCTGACT TGAGCGTCGATTTTTGTGATGCTCGTCAGGGGGGCGGAGCCTATGGAAAAACGCCAGCAACGCGGCCTTTTT ACGGTTCCTGGCCTTTTGCTGGCCTTTTGCTCACATGTTCTTTCCTGCGTTATCCCCTGATTCTGTGGATAACC GTATTACCGCCTTTGAGTGAGCTGATACCGCTCGCCGCAGCCGAACGACCGAGCGCAGCGAGTCAGTGAGCG AGGAAGCGGAAGAGCGCCTGATGCGGTATTTTCTCCTTACGCATCTGTGCGGTATTTCACACCGCATATATGG TGCACTCTCAGTACAATCTGCTCTGATGCCGCATAGTTAAGCCAGTATACACTCCGCTATCGCTACGTGACTG GGTCATGGCTGCGCCCCGACACCCGCCAACACCCGCTGACGCGCCCTGACGGGCTTGTCTGCTCCCGGCATCC GCTTACAGACAAGCTGTGACCGTCTCCGGGAGCTGCATGTGTCAGAGGTTTTCACCGTCATCACCGAAACGC GCGAGGCAGCTGCGGTAAAGCTCATCAGCGTGGTCGTGAAGCGATTCACAGATGTCTGCCTGTTCATCCGCG TCCAGCTCGTTGAGTTTCTCCAGAAGCGTTAATGTCTGGCTTCTGATAAAGCGGGCCATGTTAAGGGCGGTTT TTTCCTGTTTGGTCACTGATGCCTCCGTGTAAGGGGGATTTCTGTTCATGGGGGTAATGATACCGATGAAACG AGAGAGGATGCTCACGATACGGGTTACTGATGATGAACATGCCCGGTTACTGGAACGTTGTGAGGGTAAACA ACTGGCGGTATGGATGCGGCGGGACCAGAGAAAAATCACTCAGGGTCAATGCCAGCGCTTCGTTAATACAGA TGTAGGTGTTCCACAGGGTAGCCAGCAGCATCCTGCGATGCAGATCCGGAACATAATGGTGCAGGGCGCTGA

272

Jorge A. Lerma Romero and Harald Kolmar

CTTCCGCGTTTCCAGACTTTACGAAACACGGAAACCGAAGACCATTCATGTTGTTGCTCAGGTCGCAGACGTT TTGCAGCAGCAGTCGCTTCACGTTCGCTCGCGTATCGGTGATTCATTCTGCTAACCAGTAAGGCAACCCCGCC AGCCTAGCCGGGTCCTCAACGACAGGAGCACGATCATGCGCACCCGTGGGGCCGCCATGCCGGCGATAATGG CCTGCTTCTCGCCGAAACGTTTGGTGGCGGGACCAGTGACGAAGGCTTGAGCGAGGGCGTGCAAGATTCCGA ATACCGCAAGCGACAGGCCGATCATCGTCGCGCTCCAGCGAAAGCGGTCCTCGCCGAAAATGACCCAGAGC GCTGCCGGCACCTGTCCTACGAGTTGCATGATAAAGAAGACAGTCATAAGTGCGGCGACGATAGTCATGCCC CGCGCCCACCGGAAGGAGCTGACTGGGTTGAAGGCTCTCAAGGGCATCGGTCGAGATCCCGGTGCCTAATGA GTGAGCTAACTTACATTAATTGCGTTGCGCTCACTGCCCGCTTTCCAGTCGGGAAACCTGTCGTGCCAGCTGC ATTAATGAATCGGCCAACGCGCGGGGAGAGGCGGTTTGCGTATTGGGCGCCAGGGTGGTTTTTCTTTTCACCA GTGAGACGGGCAACAGCTGATTGCCCTTCACCGCCTGGCCCTGAGAGAGTTGCAGCAAGCGGTCCACGCTGG TTTGCCCCAGCAGGCGAAAATCCTGTTTGATGGTGGTTAACGGCGGGATATAACATGAGCTGTCTTCGGTATC GTCGTATCCCACTACCGAGATGTCCGCACCAACGCGCAGCCCGGACTCGGTAATGGCGCGCATTGCGCCCAG CGCCATCTGATCGTTGGCAACCAGCATCGCAGTGGGAACGATGCCCTCATTCAGCATTTGCATGGTTTGTTGA AAACCGGACATGGCACTCCAGTCGCCTTCCCGTTCCGCTATCGGCTGAATTTGATTGCGAGTGAGATATTTAT GCCAGCCAGCCAGACGCAGACGCGCCGAGACAGAACTTAATGGGCCCGCTAACAGCGCGATTTGCTGGTGA CCCAATGCGACCAGATGCTCCACGCCCAGTCGCGTACCGTCTTCATGGGAGAAAATAATACTGTTGATGGGT GTCTGGTCAGAGACATCAAGAAATAACGCCGGAACATTAGTGCAGGCAGCTTCCACAGCAATGGCATCCTGG TCATCCAGCGGATAGTTAATGATCAGCCCACTGACGCGTTGCGCGAGAAGATTGTGCACCGCCGCTTTACAG GCTTCGACGCCGCTTCGTTCTACCATCGACACCACCACGCTGGCACCCAGTTGATCGGCGCGAGATTTAATCG CCGCGACAATTTGCGACGGCGCGTGCAGGGCCAGACTGGAGGTGGCAACGCCAATCAGCAACGACTGTTTGC CCGCCAGTTGTTGTGCCACGCGGTTGGGAATGTAATTCAGCTCCGCCATCGCCGCTTCCACTTTTTCCCGCGTT TTCGCAGAAACGTGGCTGGCCTGGTTCACCACGCGGGAAACGGTCTGATAAGAGACACCGGCATACTCTGCG ACATCGTATAACGTTACTGGTTTCACATTCACCACCCTGAATTGACTCTCTTCCGGGCGCTATCATGCCATACC GCGAAAGGTTTTGCGCCATTCGATGGTGTCCGGGATCTCGACGCTCTCCCTTATGCGACTCCTGCATTAGGAA GCAGCCCAGTAGTAGGTTGAGGCCGTTGAGCACCGCCGCCGCAAGGAATGGTGCATGCAAGGAGATGGCGC CCAACAGTCCCCCGGCCACGGGGCCTGCCACCATACCCACGCCGAAACAAGCGCTCATGAGCCCGAAGTGGC GAGCCCGATCTTCCCCATCGGTGATGTCGGCGATATAGGCGCCAGCAACCGCACCTGTGGCGCCGGTGATGC CGGCCACGATGCGTCCGGCGTAGAGGATCGAGATCGATCTCGATCCCGCGAAATTAATACGACTCACTATAG GGGAATTGTGAGCGGATAACAATTCCCCTCTAGAAATAATTTTGTTTAACTTTAAGAAGGAGATATACATATG AGCTATTATCATCATCACCATCACCACGATTATGATATTCCGACCACCGAAAACCTGTATTTTCAAGGCGCCC CTATGACCACCGATGAAGGTGCAAAAAACAATGAAGAAAGCCCGACCGCAACCGTTGCAGAACAGGGTGAA GATATTACCAGCAAAAAAGATCGTGGCGTGCTGAAAATTGTTAAACGTGTTGGTAATGGTGAAGAAACGCCG ATGATTGGTGATAAAGTGTATGTGCACTATAAAGGCAAACTGAGCAACGGCAAAAAATTCGATAGCAGCCAT GATCGTAATGAACCGTTTGTTTTTAGCCTGGGTAAAGGCCAGGTTATTAAAGCATGGGATATTGGTGTTGCCA CCATGAAAAAAGGTGAAATTGCACATCTGCTGATCAAACCGGAATATGCCTATGGTAGCGCAGGTAGCCTGC CGAAAATTCCGAGCAATGCAACCCTGTTTTTTGAAATTGAACTGCTGGATTTCAAAGGCGAATAAGTCGACA AGCTTGCGGCCGCACTCGAGCACCACCACCACCACCACTGAGATCCGGCTGCTAACAAAGCCCGAAAGGAA GCTGAGTTGGCTGCTGCCACCGCTGAGCAATAACTAGCATAACCCCTTGGGGCCTCTAAACGGGTCTTGAGG GGTTTTTTGCTGAAAGGAGGAACTATATCCGGAT FKBP51-FK1 KanR T7 Promoter LacI F1Ori Position F67

See the plasmid map in Fig. 6.

Accessing Transient Binding Pockets by YSD and Protein Engineering

273

3. The gene of interest contains a Myc tag at the C-terminus to verify the correct expression of the protein on the surface of the yeast cells. 4. In the exemplified study, only three screening rounds were performed as a sufficient number of positive clones was obtained. Nonetheless, more screening rounds with decreasing tracer concentration may be performed if necessary/desired. References 1. Laurie ATR, Jackson RM (2006) Methods for the prediction of protein-ligand binding sites for structure-based drug design and virtual ligand screening. Curr Protein Pept Sci 7: 395–406 2. Gao M, Skolnick J (2013) A comprehensive survey of small-molecule binding pockets in proteins. PLoS Comput Biol 9:e1003302 3. Kokh DB, Czodrowski P, Rippmann F, Wade RC (2016) Perturbation approaches for exploring protein binding site flexibility to predict transient binding pockets. J Chem Theory Comput 12:4100–4113 4. Stank A, Kokh DB, Fuller JC, Wade RC (2016) Protein binding pocket dynamics. Acc Chem Res 49:809–815 5. Beglov D, Hall DR, Wakefield AE, Luo L, Allen KN, Kozakov D, Whitty A, Vajda S (2018) Exploring the structural origins of cryptic sites on proteins. Proc Natl Acad Sci 115:E3416–E3425 6. Durrant JD, McCammon JA (2011) Molecular dynamics simulations and drug discovery. BMC Biol 9:71 7. Shan Y, Mysore VP, Leffler AE, Kim ET, Sagawa S, Shaw DE (2022) How does a small molecule bind at a cryptic binding site? PLoS Comput Biol 18:e1009817 8. Arkin MR, Randal M, DeLano WL et al (2003) Binding of small molecules to an adaptive protein–protein interface. Proc Natl Acad Sci 100:1603–1608 9. Bowman GR, Geissler PL (2012) Equilibrium fluctuations of a single folded protein reveal a multitude of potential cryptic allosteric sites. PNAS. https://doi.org/10.1073/pnas. 1209309109/-/DCSupplemental 10. Huggins DJ, Sherman W, Tidor B (2012) Rational approaches to improving selectivity in drug design. J Med Chem 55:1424–1444 11. Umezawa K, Kii I (2021) Druggable Transient Pockets in Protein Kinases. Molecules 26:651 12. Nussinov R, Ma B (2012) Protein dynamics and conformational selection in bidirectional signal transduction. BMC Biol 10:2

13. Eyrisch S, Helms V (2007) Transient pockets on protein surfaces involved in protein-protein interaction. J Med Chem 50:3457–3464 14. Kokh DB, Richter S, Henrich S, Czodrowski P, Rippmann F, Wade RC (2013) TRAPP: a tool for analysis of Transient binding Pockets in Proteins. J Chem Inf Model 53:1235–1252 15. Meiler J, Baker D (2006) ROSETTALIGAND: protein-small molecule docking with full sidechain flexibility. Proteins: Structure Function Bioinformatics 65:538–548 16. Zacharias M (2004) Rapid protein-ligand docking using soft modes from molecular dynamics simulations to account for protein deformability: binding of FK506 to FKBP. Proteins: Structure, Function Bioinformatics 54:759–767 17. Oleinikovas V, Saladino G, Cossins BP, Gervasio FL (2016) Understanding cryptic pocket formation in protein targets by enhanced sampling simulations. J Am Chem Soc 138:14257– 14263 18. Kumar S, Ma B, Tsai C-J, Wolfson H, Nussinov R (1999) Folding funnels and conformational transitions via hinge-bending motions. Cell Biochem Biophys 31:141–164 19. Ma B, Kumar S, Tsai C-J, Nussinov R (1999) Folding funnels and binding mechanisms. Protein Eng Des Sel 12:713–720 20. Teague SJ (2003) Implications of protein flexibility for drug discovery. Nat Rev Drug Discov 2:527–541 21. Rath VL, Ammirati M, Danley DE et al (2000) Human liver glycogen phosphorylase inhibitors bind at a new allosteric site. Chem Biol 7: 677–682 22. Maun HR, Eigenbrot C, Lazarus RA (2003) Engineering exosite peptides for complete inhibition of factor VIIa using a protease switch with substrate phage. J Biol Chem 278:21823– 21830 23. Hardy JA, Lam J, Nguyen JT, O’Brien T, Wells JA (2004) Discovery of an allosteric site in the caspases. Proc Natl Acad Sci 101:12461– 12466

274

Jorge A. Lerma Romero and Harald Kolmar

24. Braisted AC, Oslob JD, Delano WL, Hyde J, McDowell RS, Waal N, Yu C, Arkin MR, Raimundo BC (2003) Discovery of a potent small molecule IL-2 inhibitor through fragment assembly. J Am Chem Soc 125:3714–3715 25. Gaali S, Kirschner A, Cuboni S et al (2015) Selective inhibitors of the FK506-binding protein 51 by induced fit. Nat Chem Biol 11:33– 37 26. Lerma Romero JA, Meyners C, Christmann A, Reinbold LM, Charalampidou A, Hausch F, Kolmar H (2022) Binding pocket stabilization by high-throughput screening of yeast display libraries. Front Mol Biosci. https://doi.org/ 10.3389/fmolb.2022.1023131 27. 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 28. Bogen JP, Grzeschik J, Krah S, Zielonka S, Kolmar H (2020) Rapid generation of chicken immune libraries for yeast surface display. Methods Mol Biol 2070:289–302 29. Becker S, Schmoldt HU, Adams TM, Wilhelm S, Kolmar H (2004) Ultra-highthroughput screening based on cell-surface display and fluorescence-activated cell sorting for the identification of novel biocatalysts. Curr Opin Biotechnol 15:323–329 30. 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

Chapter 15 Tyrosine Phosphorylation Screening on the Yeast Surface by Magnetic Bead Selection and FACS Jose Ezagui and Lawrence A. Stern Abstract The ability to understand and characterize phosphorylation is important to the study of cell signaling and to synthetic biology approaches. Current methods for characterizing kinase–substrate interactions are limited by their inherently low throughput and the heterogeneity of samples analyzed. Recent advances in yeast surface display techniques provide new opportunities for studying individual kinase–substrate interactions in a stimulus-independent fashion. Here, we describe techniques for building substrate libraries into fulllength domains of interest that, when co-localized intracellularly with individual kinases, result in the display of phosphorylated domains on the yeast surface, as well as fluorescence-activated cell sorting and magnetic bead selection techniques for enriching from these libraries based on phosphorylation state. Key words Yeast surface display, High-throughput screening, Phosphorylation, Kinase

1

Introduction Numerous cellular processes are tightly correlated to a cell’s surrounding environment [1, 2], where environmental cues have the capacity to activate cell signal transduction through intricate groups of pathways [3, 4]. These massive pathways work as a chain of commands that are often controlled by posttranslational modifications, which are alterations to specific protein motifs mediated by enzymes [5]. As an example, protein phosphorylation consists of the kinase-mediated addition of a phosphate group from ATP to a specific amino acid [6]. These enzymes and their inverse, phosphatases, are thought to play an important role in all major cellular functions [7], making the understanding of their interaction patterns and behavior extremely important. Constraints in the current standards of evaluation and their capacity to innately generate posttranslational modifications limit the capacity of analyzing large libraries of kinase–substrate pairs [8, 9]. Yeast surface display [10–12] and more specifically the

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2681, https://doi.org/10.1007/978-1-0716-3279-6_15, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

275

276

Jose Ezagui and Lawrence A. Stern

yeast endoplasmic reticulum sequestration screening (YESS) system have proven to be a quality platform to evaluate enzymatic modifications of short peptides [13–17]. The system results in the co-localization and retention of an enzyme and substrate pair in the endoplasmic reticulum (ER) through ER-targeted signal sequences and peptides able to bind to the ER-resident KDEL receptors. This increases the interaction period between the pair of interest, enhancing the possibility of a posttranslational modification of the substrate by the enzyme of interest. The incorporation of an N-terminal Aga2p to the substrate for surface display together with the application of phosphorylation-based high-throughput screening methods such as fluorescence-activated cell sorting (FACS) [18] or magnetic bead sorting [19] provides an efficient strategy for library-based protein engineering and screening [20]. Here, we present a facile substrate specificity profiling method through the generation of yeast libraries containing locationspecific mutations coupled with sorting techniques to enable the characterization of phosphorylation-enriched mutant libraries. These methods include a protocol for FACS and an optimized magnetic bead capture strategy as sorting methods. The presented compilation of methods provides an effective strategy for kinasesubstrate reaction profiling.

2

Materials

2.1 Mutant Library Preparation

1. General forward primer for sequence of interest.

2.1.1 Media, Buffers, and Reagents

3. Mutagenic forward primer for sequence of interest.

2. General reverse primer for sequence of interest. 4. Reverse primer that overlaps forward mutagenic primer in unmutated region. 5. pCT (Addgene Plasmid #41843) or other yeast surface display plasmid containing a kinase of interest (example published in [20]). 6. 10% Elution buffer: dilute 100 μL of elution buffer (Buffer EB from QIAGEN kits or equivalent) in 900 μL of dH2O. 7. Taq reaction buffer. 8. Taq DNA polymerase. 9. 10 mM dNTPs. 10. Template plasmid with DNA sequence to be mutated. 11. Ethanol: 100% and 70% diluted in dH2O. 12. 5 M sodium acetate pH 5.2. 13. Restriction enzyme activity buffer. 14. NheI restriction enzyme.

Tyrosine Phosphorylation Screening by Yeast Surface Display

277

15. XhoI restriction enzyme. 16. YPD medium (yeast extract peptone dextrose):10 g/L yeast extract, 20.0 g/L peptone, and 20.0 g/L dextrose in deionized water. Autoclave. Store at room temperature. 17. Buffer E: 182.2 g/L sorbitol, 0.147 g/L calcium chloride in deionized water. Sterile filter. Store at room temperature. 18. Dithiothreitol (DTT). 19. Lithium acetate buffer: 10.2 g/L lithium acetate, 1.24 g/L Tris–HCl, 0.26 g/L Tris base, and 0.372 g/L ethylenediaminetetraacetic acid (EDTA) in deionized water. Sterile filter. Store at room temperature. 20. SD-CAA medium (selective growth medium for yeast):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 dissolved in deionized water. Sterile filter. Store at room temperature. 21. SD-CAA plate medium—buffer components: 16.8 g/L sodium citrate dihydrate, 3.9 g/L citric acid, and 15 g/L agar in 900 mL deionized water. Autoclave. Nutrient components: 20.0 g/L dextrose, 6.7 g/L yeast nitrogen base, and 5.0 g/L casamino acids dissolved in 100 mL deionized water. Sterile filter into buffer components once buffer components cool to below 50 °C. Pour 20–25 mL into an individual petri dish. Allow plates to solidify at room temperature. Store upside down at 4 °C. 22. SRG-CAA medium (selective medium for protein expression in yeast): 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, 20 g/L raffinose, 6.7 g/L yeast nitrogen base, and 5.0 g/L casamino acids in deionized water. Sterile filter. Store at room temperature. 2.1.2 Equipment and Consumables

All plastic and glassware should be sterilized prior to their use. 1. Personal computer with Internet access. 2. Vortex mixer. 3. Tube rotator or rocking table. 4. 1.7/2.0 mL vials. 5. 5 mL polystyrene tubes. 6. Thermal cycler. 7. 0.2 mL PCR tubes. 8. -20 °C freezer. 9. -80 °C freezer. 10. Pipette controller.

278

Jose Ezagui and Lawrence A. Stern

11. Serological pipettes. 12. Single-channel pipette set (e.g., 1–10 μL, 10–100 μL, and 100–1000 μL). 13. 14 mL polystyrene round-bottom tubes (for liquid yeast culture growth). 14. 250 mL baffled flasks (for liquid yeast culture growth). 15. 1 L media bottles. 16. 3 mL plastic cuvettes. 17. 0.2 cm electroporation cuvettes. 18. Micropulser electroporator. 19. Shaking incubator at 30 °C, 220 rpm (for liquid yeast culture growth). 20. Static incubator at 30 °C (for plated yeast growth). 21. Spectrophotometer capable of reading the optical density at 600 nm (OD600nm). 22. Microcentrifuge with rotor for 1.7/2.0 mL vials (for pelleting yeast cells). 23. Centrifuge with bucket rotor for 15/50 mL conical tubes (for washing/preparing yeast). 24. Petri dishes. 25. Top loading balance. 26. Precision balance. 27. Kimwipes or other delicate task wipers. Cell Lines

1. Saccharomyces cerevisiae yeast surface display strain EBY100 (available from ATCC).

2.2 Flow Cytometry Analysis and FACS of Yeast Displaying Mutant Substrate Libraries

This section shares entries 5, 20, 21, and 22 from Subheading 2.1.1, with the following additions:

2.1.3

2.2.1 Media, Buffers, and Reagents

1. PBS: 1× phosphate-buffered saline (PBS), pH 7.4. Sterile filter. Store at room temperature. 2. PBSA: 1× phosphate-buffered saline (PBS), pH 7.4 with 0.1% bovine serum albumin by weight. Sterile filter. Store at 4 °C. 3. NHS-PEG4-biotin or equivalent biotin conjugation kit. 4. Mouse anti-phosphotyrosine antibody 4G10. 5. Goat anti-Myc tag antibody. 6. Donkey anti-goat FITC. 7. Streptavidin Alexa Fluor 647.

Tyrosine Phosphorylation Screening by Yeast Surface Display 2.2.2 Equipment and Consumables

279

This section shares entries 3, 4, 8–15, and 19–25 from Subheading 2.1.2, with the following additions: 1. 0.5 mL desalting spin columns with molecular weight cutoff lower than the molecular weight of recombinant target proteins of interest. 2. Permanent markers. 3. Flow cytometry analyzer. 4. Fluorescence-activated cell sorter.

2.2.3

Cell Lines

2.3 Sorting of Yeast Cells Displaying Phosphorylated Mutant Libraries Through Magnetic Bead Selection

1. Same as Subheading 2.1.3. This section shares all entries with Subheading 2.2.1 with the addition of the following: 1. Dynabeads Biotin Binder or equivalent streptavidin-coated magnetic beads.

2.3.1 Media, Buffers, and Reagents 2.3.2 Equipment and Consumables

This section shares entries 3, 4, 8–15, and 19–25 from Subheading 2.1.2 and entries 1–3 with Subheading 2.2.2 with the following additions: 1. DynaMag-2 or equivalent magnet capable of holding 1.7/2.0 mL vials.

2.3.3

3

Cell Lines

1. Same as Subheading 2.1.3.

Methods

3.1 Mutant Library Preparation

The following steps describe the generation of libraries with motifspecific mutations, which empower profiling of substrate specificity through co-localization in the endoplasmic reticulum with a kinase of interest. Figure 1 illustrates the mutagenesis scheme to be used.

3.1.1 Mutagenic PCR Primer Design

1. Define the substrate motif to be mutated and substitute each amino acid with an NNK randomized codon. 2. Extend the mutagenic primer from step 1 in both directions by including the 20–25 base pairs on each side corresponding to the unmutated sequence. Add up to 3 base pairs in each direction with the purpose of having cytosine or guanine on the primer ends.

280

Jose Ezagui and Lawrence A. Stern

Fig. 1 Mutagenesis scheme for substrate libraries. DNA encoding a gene of interest (top) is used as the template for a two-step mutagenic PCR. In two separate reactions, template DNA is primed by either a mutagenic forward primer and overall reverse primer (middle, left) or an overall forward primer and a non-mutagenic reverse primer that overlaps the mutagenic forward primer outside of the site of mutation (middle, right). These two fragments are then combined by assembly PCR and yielded a mutated substrate gene (bottom)

3. Evaluate the hairpins and melting temperature to ensure primer quality at the designed cycling conditions. 4. Pair the mutagenic primer with an opposite-direction general primer sitting outside the sequence of interest. 5. Evaluate the differences in melting temperatures (maximum difference of preferably 3°C) from both primers and modify the length of the mutagenic primer accordingly. 6. Copy the reverse complimentary sequence of the 20–25 base pairs extension 5′ to the mutated motif included in step 2. This sequence will work as the overlapping complimentary primer to the mutagenic primer.

Tyrosine Phosphorylation Screening by Yeast Surface Display

281

7. Same as in step 3. 8. Pair the overlapping complimentary primer with an oppositedirection general primer sitting outside the sequence of interest. 9. Evaluate the differences in melting temperatures (maximum difference of preferably 3°C) from both primers and modify the length of the overlapping primer accordingly. 3.1.2 Mutant Substrate Generation Through Mutagenic PCR

1. Solubilize the mutation containing oligonucleotides in 10% elution buffer to a final concentration of 100 μM and vortex for at least 30 s (see Note 1). 2. Incubate solubilized oligonucleotides for 30 min at room temperature with constant agitation on a rotator or a rocking table. 3. Transfer an aliquot of solubilized oligonucleotides to a new 1.7 mL vial. Dilute to 10 μM using deionized water. 4. Generate mutated sequence-containing overlapping pairs through a two-reaction mutagenic PCR. (a) In a PCR tube, for a final standard volume of 50 μL, combine 5 μL of Taq reaction buffer (1× final concentration), 5 μL of 10 μM forward primer sitting 5′ to the sequence of interest, 5 μL of 10 μM reverse primer that overlaps the mutagenic primer to be used, 1 μL of dNTPs, 1 μL of plasmid containing the gene to be mutated (template DNA), and 0.5 μL of Taq DNA polymerase. Complete to final volume with deionized water (see Notes 2 and 3). (b) In a PCR tube, for a final standard volume of 50 μL, combine 5 μL of Taq reaction buffer (1× final concentration), 5 μL of 10 μM reverse primer sitting 3′ to the sequence of interest, 5 μL of 10 μM forward mutagenic primer, 1 μL of dNTPs, 1 μL of plasmid containing the gene to be mutated (template DNA), and 0.5 μL of Taq DNA polymerase. Complete to final volume with deionized water. 5. Thermally cycle the reactions for a total of 30 cycles. Annealing temperature should be established depending on primer design and the extension time each cycle is 1 min per kb of DNA to be generated. 6. The total quantity of resulting DNA (50 μL per reaction) is confirmed through its band size and brightness with gel electrophoresis and purified through gel extraction kit using the manufacturer’s protocol (see Note 4). Samples can be stored at -20°C.

282

Jose Ezagui and Lawrence A. Stern

3.1.3 Insert Preparation Through Assembly and Extension PCR

1. In a PCR tube, for a final volume of 25 μL, combine 2.5 μL of Taq reaction buffer (1× final concentration), 5 μL of product from step 4a (see Subheading 3.1.2), 5 μL of product from step 4b (see Subheading 3.1.2), 1 μL of dNTPs, and 0.5 μL of Taq DNA polymerase. Complete with deionized water to final volume. 2. Thermally cycle the reactions described for a total of 15 cycles. Annealing temperature should be established depending on primer design, and the extension time each cycle is 1 min per kb of DNA to be generated. 3. Scale up the reaction in a PCR tube for a final volume of 200 μL by mixing 20 μL of Taq reaction buffer (1× final concentration), 20 μL of 10 μM forward primer sitting 5′ from the sequence of interest used in step 4a (see Subheading 3.1.2), 20 μL of 10 μM reverse primer sitting 3′ from the sequence of interest used in step 4b (see Subheading 3.1.2), 25 μL from the assembly PCR reaction from step 2, 4 μL dNTPs, and 0.5 μL Taq DNA polymerase. Complete to final volume with deionized water. 4. Same as step 5 (see Subheading 3.1.2). 5. Confirm product through its band size using 5 μL of the PCR product in gel electrophoresis.

3.1.4 Insert DNA Concentration Through Ethanol Precipitation

1. Transfer DNA sample into a 1.7 mL vial and add 10% volume of 3 M NaAc (pH: 5.2). Mix thoroughly. 2. Add 3× volume of 100% ethanol into the mixture and incubate at 4°C for 10 min (see Note 5). 3. Centrifuge at 15,000 g for 20 min at 4°C. Discard the supernatant, making sure to not disturb the pellet formed in the bottom of the vial (see Note 6). 4. Add 500 μL of 70% ethanol into the vial and vortex for 10 s. 5. Repeat step 3. 6. Add 500 μL of 100% ethanol into the vial and vortex for 10 s. 7. Repeat step 3. 8. Place vial inside a sterile environment and leave open to air dry.

3.1.5 Plasmid Digest for Substrate Incorporation

1. In a 1.7 mL vial, combine 50 μg of plasmid DNA used in step 4 (see Subheading 3.1.2), 10× restriction enzyme activity buffer (to a final concentration of 1×), and 5 U restriction enzyme (NheI and XhoI) per μg of plasmid to be digested. The final volume of restriction enzyme added should not surpass 5% of the total volume. Fill to final volume with deionized water and incubate overnight at 37°C.

Tyrosine Phosphorylation Screening by Yeast Surface Display

283

3.1.6 Plasmid DNA Concentration Through Ethanol Precipitation

1. Same as Subheading 3.1.4.

3.1.7 Mutated Substrate Incorporation Through Yeast Electroporation Transformation

1. Using a 10 μL pipette or similar, transfer an EBY100 yeast colony from an agar plate into 5 mL of YPD media in a 14 mL culture tube and grow overnight at 30°C with constant agitation (250 rpm). Multiple cultures can be grown simultaneously to ensure sufficient yeast is available for subsequent steps. 2. Introduce 1 × 108 yeast into 100 mL of YPD (OD = 0.1) per sample to be transformed. Incubate culture at 30°C with constant agitation (250 rpm) to OD600nm between 0.8 and 1.5 (approximately 3–4 h). For calculation, 1 OD600nm = 1 × 107 yeast/mL. 3. In 50 mL conical tubes, centrifuge the yeast culture for 3 min at 2500 g and 4°C. Using a serological pipette or similar, remove the supernatant without disturbing the formed pellet. 4. Resuspend pelleted cells in 25 mL of cold water (4°C). Repeat step 3 after each resuspension for cell pelleting. Repeat this step once for a total of two washes with cold water. 5. Resuspend pelleted cells in 25 mL of cold buffer E (4°C). Repeat step 3 for cell pelleting. 6. Resuspend pelleted cells in 25 mL of lithium acetate buffer. 7. Measure 0.0385 g of dithiothreitol (DTT) and dilute with 0.5 mL of lithium acetate buffer. Add DTT solution to resuspended cells for a final concentration of 10 mM. 8. Incubate yeast at 30°C and constant agitation (220 rpm) for 30 min. 9. Repeat step 3. 10. Repeat step 5. 11. Resuspend cells in 1 mL of cold buffer E and transfer them into a 1.7 mL microcentrifuge tube. 12. Centrifuge the sample for 1 min at 5000 g and 4°C. Using a 1 mL pipette or similar, remove the supernatant without disturbing the formed pellet. 13. Resuspend pelleted cells in 1 mL of cold buffer E. Repeat step 12. 14. Resuspend pelleted cells to final volume of 300 μL per sample (yeast + buffer) to be electroporated in cold buffer E. 15. Resuspend ethanol precipitated mutant DNA insert from Subheading 3.1.4 in 20 μL of buffer E.

284

Jose Ezagui and Lawrence A. Stern

16. In a cold microcentrifuge tube kept on ice, mix 300 μL of prepared EBY100 cells with 6 μg of digested vector from Subheading 3.1.6 (less than 6 μL or 1.5 pmol) and 20 μL of resuspended mutant DNA insert. Gently pipette repeatedly to mix (see Note 7). 17. Transfer mixture to a cold 0.2 cm electroporation cuvette on ice and incubate for 5 min on ice. 18. Using a Kimwipe, dry the exterior of the 0.2 cm electroporation cuvette and place it into electroporator. Pulse the sample at 25 μF and 1.2 kV. Remove the cuvette from the electroporator and add 1 mL of room temperature YPD media to the cuvette. Pipette to mix and transfer yeast to a 14 mL culture tube. 19. Add an additional 1 mL of room temperature YPD media to the cuvette and pipette to mix. Transfer yeast to the same 14 mL culture tube. 3.1.8 Cell Growth of Yeast Harboring Mutated Substrate Libraries

1. Incubate yeast at 30°C with constant agitation (220 rpm) for 2 h. 2. Centrifuge the culture for 1 min at 1300 g. Using a serological pipette, remove the supernatant without disturbing the formed pellet. 3. Prepare a 250 mL flask with 100 mL of SD-CAA media. Resuspend the pelleted cells in 1 mL of SD-CAA media and inoculate the prepared 100 mL. 4. Serially dilute 10 μL of culture to 10× and 100× dilutions. Without piercing the agar, distribute 10 μL of each dilution onto SD-CAA plates and incubate statically for 48 h at 30°C to determine transformation efficiency. 5. Incubate 100 mL cell culture at 30°C and 220 rpm for at least 18 h. Monitor the optical density after 16 h and after 20 h.

3.1.9 Induction of Protein Expression

1. Count the number of colonies grown on the dilution plates from step 4 (see Subheading 3.1.8) to determine 1× library diversity. Add a volume from step 5 (see Subheading 3.1.8) of yeast equivalent to 20× library diversity to 50 mL conical tubes. 2. Centrifuge the culture for 3 min at 2500 g. Discard the supernatant without disturbing the yeast pellet. 3. Resuspend the yeast pellet in SRG-CAA to a final OD600nm less than 1 ( YSD_Entry_Insert. GCTCTTCAAGCCAAGGCAATATTAGTTAATCCCAACAATATTGTGAACAGGTAGGTAAAATG AGCGAAAGATAATGTCATGGTGGCTGCGACGGTCTTCGGTGAATTTTCAAAAATTCTTACTT TTTTTTTGGATGGACGCAAAGAAGTTTAATAATCATATTACATGGCATTACCACCATATACAT ATCCATATACATATCCATATCTAATCTTACTTATATGTTGTGGAAATGTAAAGAGCCCCATTA TCTTAGCCTAAAAAAACCTTCTCTTTGGAACTTTCAGTAATACGCTTAACTGCTCATTGCTAT ATTGAAGTACGGATTAGAAGCCGCCGAGCGGGTGACAGCCCTCCGAAGGAAGACTCTCCTC CGTGCGTCCTCGTCTTCACCGGTCGCGTTCCTGAAACGCAGATGTGCCTCGCGCCGCACTGC TCCGAACAATAAAGATTCTACAATACTAGCTTTTATGGTTATGAAGAGGAAAAATTGGCAGT AACCTGGCCCCACAAACCTTCAAATGAACGAATCAAATTAACAACCATAGGATGATAATGC GATTAGTTTTTTAGCCTTATTTCTGGGGTAATTAATCAGCGAAGCGATGATTTTTGATCTATT AACAGATATATAAATGCAAAAACTGCATAACCACTTTAACTAATACTTTCAACATTTTCGGTT TGTATTACTTCTTATTCAAATGTAATAAAAGTATCAACAAAAAATTGTTAATATACCTCTATA CTTTAACGTCAAGGAGAAAAAACCGAAGACATCGCTTGCCACCATGACATTATCTTTCGCTC ATTTTACCTACCTGTTCACAATATTGTTGGGATTAACTAATATTGCCTTGGCATGAAGAGC Aga1P Signal Pepde for LC, Gal1,10 promoter, Aga1P Signal Pepde for HC

Bulk Reformatting of Antibody Fragments to Final IgG Format

303

> YSD_pDest_Kappa CTAAATTGTAAGCGTTAATATTTTGTTAAAATTCGCGTTAAATTTTTGTTAAATCAGCTCATTT TTTAACCAATAGGCCGAAATCGGCAAAATCCCTTATAAATCAAAAGAATAGACCGAGATAG GGTTGAGTGGCCGCTACAGGGCGCTCCCATTCGCCATTCAGGCTGCGCAACTGTTGGGAAG GGCGTTTCGGTGCGGGCCTCTTCGCTATTACGCCAGCTGGCGAAAGGGGGATGTGCTGCAA GGCGATTAAGTTGGGTAACGCCAGGGTTTTCCCAGTCACGACGTTGTAAAACGACGGCCAG TGAGCGCGACGTAATACGACTCACTATAGGGCGAATTGGCGGAAGGCCGTCAAAAGATATT CTTTATTGAAAAATAGCTTGTCACCTTACGTACAATCTTGATCCGGAGCTTTTCTTTTTTTGCC GATTAAGAATTAATTCGGTCGAAAAAAGAAAAGGAGAGGGCCAAGAGGGAGGGCATTGG TGACTATTGAGCACGTGAGTATACGTGATTAAGCACACAAAGGCAGCTTGGAGTATGTCTG TTATTAATTTCACAGGTAGTTCTGGTCCATTGGTGAAAGTTTGCGGCTTGCAGAGCACAGAG GCCGCAGAATGTGCTCTAGATTCCGATGCTGACTTGCTGGGTATTATATGTGTGCCCAATAG AAAGAGAACAATTGACCCGGTTATTGCAAGGAAAATTTCAAGTCTTGTAAAAGCATATAAA AATAGTTCAGGCACTCCGAAATACTTGGTTGGCGTGTTTCGTAATCAACCTAAGGAGGATGT TTTGGCTCTGGTCAATGATTACGGCATTGATATCGTCCAACTGCATGGAGATGAGTCGTGGC AAGAATACCAAGAGTTCCTCGGTTTGCCAGTTATTAAAAGACTCGTATTTCCAAAAGACTGC AACATACTACTCAGTGCAGCTTCACAGAAACCTCATTCGTTTATTCCCTTGTTTGATTCAGAA GCAGGTGGGACAGGTGAACTTTTGGATTGGAACTCGATTTCTGACTGGGTTGGAAGGCAA GAGAGCCCCGAAAGCTTACATTTTATGTTAGCTGGTGGACTGACGCCAGAAAATGTTGGTG ATGCGCTTAGATTAAATGGCGTTATTGGTGTTGATGTAAGCGGAGGTGTGGAGACAAATGG TGTAAAAGACTCTAACAAAATAGCAAATTTCGTCAAAAATGCTAAGAAATAGGTTATTACTG AGTAGTATTTATTTAAGTATTGTTTGTGCACTTGCCTATGCGGTGTGAAATACCGCACAGAT GCGTAAGGAGAAAATACCGCATCAGGAAATTGTAAACGTTAATATTTTGTTAAAATTCGCGT TAAATTTTTGTTAAATCAGCTCATTTTTTAACCAATAGGCCGAAATCGGCAAAATCCCTTATA AATCAAAAGAATAGACCGAGATAGGGTTGAGTGTTGTTCCAGTTTGGAACAAGAGTCCACT ATTAAAGAACGTGGACTCCAACGTCAAAGGGCGAAAAACCGTCTATCAGGGCGATGGCCC ACTACGTGAACCATCACCCTAATCAAGTTTTTTGGGGTCGAGGTGCCGTAAAGCACTAAATC GGAACCCTAAAGGGAGCCCCCGATTTAGAGCTTGACGGGGAAAGCCGGCGAACGTGGCGA GAAAGGAAGGGAAGAAAGCGAAAGGAGCGGGCGCTAGGGCGCTGGCAAGTGTAGCGGT CACGCTGCGCGTAACCACCACACCCGCCGCGCTTAATGCGCCGCTACAGGGCGCGTCCATT CGCCATTCAGGCTGCGCAACTGTTGGGAAGGGCGATCGGTGCGGGCCTCTTCGCTATTACG CCAGCTGAATTGGAGCGACCTCATGCTATACCTGAGAAAGCAACCTGACCTACAGGAAAGA GTTACTCAAGAATAAGAATTTTCGTTTTAAAACCTAAGAGTCACTTTAAAATTTGTATACACT TATTTTTTTTATAACTTATTTAATAATAAAAATCATAAATCATAAGAAATTCGCTTATTTAGAA GTGTCAACAACGTATCTACCAACGATTTGACCCTTTTCCATCTTTTCGTAAATTTCTGGCAAG GTAGACAAGCCGACAACCTTGATTGGAGACTTGACCAAACCTCTGGCGAAGAATTGTTAAT TAAGAGCTCGTCTCCAGATCTATTAACACTCTCCCCTGTTGAAGCTCTTTGTGACGGGCGAG CTCAGGCCCTGATGGGTGACTTCGCAGGCGTAGACTTTGTGTTTCTCGTAGTCTGCTTTGCT CAGCGTCAGGGTGCTGCTGAGGCTGTAGGTGCTGTCCTTGCTGTCCTGCTCTGTGACACTCT CCTGGGAGTTACCCGATTGGAGGGCGTTATCCACCTTCCACTGTACTTTGGCCTCTCTGGGA TAGAAGTTATTCAGCAGGCACACAACAGAGGCAGTTCCAGATTTCAACTGCTCATCAGATG GCGGGAAGATGAACACAGATGGTGCAGCCACAGTTCGTGAAGAGCTAGGTGTAGACGGCT TCAAGTCATGCTATGGCAGCTCTTCTGCCTCTACTAAAGGTCCATCTGTTTTTCCATTGGCTC

304

Stefania C. Carrara et al.

CATCTTCTAAATCTACATCTGGTGGTACTGCTGCTTTGGGTTGTTTGGTTAAGGATTATTTTC CAGAACCAGTCACCGTTTCTTGGAATTCTGGTGCTTTGACTTCTGGTGTTCATACTTTTCCAG CCGTATTGCAATCTTCTGGCTTGTATTCTTTGTCCTCTGTTGTTACTGTTCCCTCTTCTTCTTTG GGTACTCAAACTTACATCTGCAACGTTAACCATAAGCCATCTAACACCAAGGTTGATAAGAG AGTCGAACCTAAGTCTTGTGATAAGACTCATACCGAGACGGGTGGTGGTGGTTCTGGTGGT GGTGGTTCTGGTGGTGGTGGTTCTGGTGGTGGTGGTTCTCAGGAACTGACAACTATATGCG AGCAAATCCCCTCACCAACTTTAGAATCGACGCCGTACTCTTTGTCAACGACTACTATTTTGG CCAACGGGAAGGCAATGCAAGGAGTTTTTGAATATTACAAATCAGTAACGTTTGTCAGTAA TTGCGGTTCTCACCCCTCAACAACTAGCAAAGGCAGCCCCATAAACACACAGTATGTTTTTT GATAAGATCCGCTCTAACCGAAAAGGAAGGAGTTAGACAACCTGAAGTCTAGGTCCCTATT TATTTTTTTATAGTTATGTTAGTATTAAGAACGTTATTTATATTTCAAATTTTTCTTTTTTTTCT GTACAGACGCGTGTACGCATGTAACATTATACTGAAAACCTTGCTTGAGAAGGTTTTGGGA CGCTCGAAGATCCAGCTGCATTAATGAATCGAACAAATAGGGGTTCCGCGCACATTTCCCCG AAAAGTGCCACCTGAACGAAGCATCTGTGCTTCATTTTGTAGAACAAAAATGCAACGCGAG AGCGCTAATTTTTCAAACAAAGAATCTGAGCTGCATTTTTACAGAACAGAAATGCAACGCGA AAGCGCTATTTTACCAACGAAGAATCTGTGCTTCATTTTTGTAAAACAAAAATGCAACGCGA GAGCGCTAATTTTTCAAACAAAGAATCTGAGCTGCATTTTTACAGAACAGAAATGCAACGCG AGAGCGCTATTTTACCAACAAAGAATCTATACTTCTTTTTTGTTCTACAAAAATGCATCCCGA GAGCGCTATTTTTCTAACAAAGCATCTTAGATTACTTTTTTTCTCCTTTGTGCGCTCTATAATG CAGTCTCTTGATAACTTTTTGCACTGTAGGTCCGTTAAGGTTAGAAGAAGGCTACTTTGGTG TCTATTTTCTCTTCCATAAAAAAAGCCTGACTCCACTTCCCGCGTTTACTGATTACTAGCGAA GCTGCGGGTGCATTTTTTCAAGATAAAGGCATCCCCGATTATATTCTATACCGATGTGGATT GCGCATACTTTGTGAACAGAAAGTGATAGCGTTGATGATTCTTCATTGGTCAGAAAATTATG AACGGTTTCTTCTATTTTGTCTCTATATACTACGTATAGGAAATGTTTACATTTTCGTATTGTT TTCGATTCACTCTATGAATAGTTCTTACTACAATTTTTTTGTCTAAAGAGTAATACTAGAGAT AAACATAAAAAATGTAGAGGTCGAGTTTAGATGCAAGTTCAAGGAGCGAAAGGTGGATGG GTAGGTTATATAGGGATATAGCACAGAGATATATAGCAAAGAGATACTTTTGAGCAATGTT TGTGGAAGCGGTATTCGCAATATTTTAGTAGCTCGTTACAGTCCGGTGCGTTTTTGGTTTTTT GAAAGTGCGTCTTCAGAGCGCTTTTGGTTTTCAAAAGCGCTCTGAAGTTCCTATACTTTCTAG AGAATAGGAACTTCGGAATAGGAACTTCAAAGCGTTTCCGAAAACGAGCGCTTCCGAAAAT GCAACGCGAGCTGCGCACATACAGCTCACTGTTCACGTCGCACCTATATCTGCGTGTTGCCT GTATATATATATACATGAGAAGAACGGCATAGTGCGTGTTTATGCTTAAATGCGTACTTATA TGCGTCTATTTATGTAGGATGAAAGGTAGTCTAGTACCTCCTGTGATATTATCCCATTCCATG CGGGGTATCGTATGCTTCCTTCAGCACTACCCTTTAGCTGTTCTATATGCTGCCACTCCTCAA TTGGTGGGCCTTCCGCTCACTGCCCGCTTTCCAGTCGGGAAACCTGTCGTGCCAGCTGCATT AACATGGTCATAGCTGTTTCCTTGCGTATTGGGCGCTCTCCGCTTCCTCGCTCACTGACTCGC TGCGCTCGGTCGTTCGGGTAAAGCCTGGGGTGCCTAATGAGCAAAAGGCCAGCAAAAGGC CAGGAACCGTAAAAAGGCCGCGTTGCTGGCGTTTTTCCATAGGCTCCGCCCCCCTGACGAG CATCACAAAAATCGACGCTCAAGTCAGAGGTGGCGAAACCCGACAGGACTATAAAGATACC AGGCGTTTCCCCCTGGAAGCTCCCTCGTGCGCTCTCCTGTTCCGACCCTGCCGCTTACCGGAT ACCTGTCCGCCTTTCTCCCTTCGGGAAGCGTGGCGCTTTCTCATAGCTCACGCTGTAGGTATC TCAGTTCGGTGTAGGTCGTTCGCTCCAAGCTGGGCTGTGTGCACGAACCCCCCGTTCAGCCC GACCGCTGCGCCTTATCCGGTAACTATCGTCTTGAGTCCAACCCGGTAAGACACGACTTATC GCCACTGGCAGCAGCCACTGGTAACAGGATTAGCAGAGCGAGGTATGTAGGCGGTGCTAC AGAGTTCTTGAAGTGGTGGCCTAACTACGGCTACACTAGAAGAACAGTATTTGGTATCTGC GCTCTGCTGAAGCCAGTTACCTTCGGAAAAAGAGTTGGTAGCTCTTGATCCGGCAAACAAA CCACCGCTGGTAGCGGTGGTTTTTTTGTTTGCAAGCAGCAGATTACGCGCAGAAAAAAAGG

Bulk Reformatting of Antibody Fragments to Final IgG Format

305

ATCTCAAGAAGATCCTTTGATCTTTTCTACGGGGTCTGACGCTCAGTGGAACGAAAACTCAC GTTAAGGGATTTTGGTCATGAGATTATCAAAAAGGATCTTCACCTAGATCCTTTTAAATTAA AAATGAAGTTTTAAATCAATCTAAAGTATATATGAGTAAACTTGGTCTGACAGTTACCAATG CTTAATCAGTGAGGCACCTATCTCAGCGATCTGTCTATTTCGTTCATCCATAGTTGCCTGACT CCCCGTCGTGTAGATAACTACGATACGGGAGGGCTTACCATCTGGCCCCAGTGCTGCAATG ATACCGCGAGAACCACGCTCACCGGCTCCAGATTTATCAGCAATAAACCAGCCAGCCGGAA GGGCCGAGCGCAGAAGTGGTCCTGCAACTTTATCCGCCTCCATCCAGTCTATTAATTGTTGC CGGGAAGCTAGAGTAAGTAGTTCGCCAGTTAATAGTTTGCGCAACGTTGTTGCCATTGCTAC AGGCATCGTGGTGTCACGCTCGTCGTTTGGTATGGCTTCATTCAGCTCCGGTTCCCAACGAT CAAGGCGAGTTACATGATCCCCCATGTTGTGCAAAAAAGCGGTTAGCTCCTTCGGTCCTCCG ATCGTTGTCAGAAGTAAGTTGGCCGCAGTGTTATCACTCATGGTTATGGCAGCACTGCATAA TTCTCTTACTGTCATGCCATCCGTAAGATGCTTTTCTGTGACTGGTGAGTACTCAACCAAGTC ATTCTGAGAATAGTGTATGCGGCGACCGAGTTGCTCTTGCCCGGCGTCAATACGGGATAAT ACCGCGCCACATAGCAGAACTTTAAAAGTGCTCATCATTGGAAAACGTTCTTCGGGGCGAA AACTCTCAAGGATCTTACCGCTGTTGAGATCCAGTTCGATGTAACCCACTCGTGCACCCAAC TGATCTTCAGCATCTTTTACTTTCACCAGCGTTTCTGGGTGAGCAAAAACAGGAAGGCAAAA TGCCGCAAAAAAGGGAATAAGGGCGACACGGAAATGTTGAATACTCATACTCTTCCTTTTTC AATATTATTGAAGCATTTATCAGGGTTATTGTCTCATGAGCGGATACATATTTGAATGTATTT AGAAAAATAAACAAATAGGGGTTCCGCGCACATTTCCCCGAAAAGTGCCAC

TRP1 promoter, TRP1, f1 ori, CLκ, Stuffer, CH1, Paral Hinge, GS Linker, Aga2p, Cyc Terminator, 2u, ori, AmpR

> Mammalian_Dest (MD) vector TGTATTTAGAAAAATAAACAAATAGGGGTTCCGCGCACATTTCCCCGAAAAGTGCCACCTGA CGTCGACGGATCGGGAGATCTCCCGATCCCCTATGGTGCACTCTCAGTACAATCTGCTCTGA TGCCGCATAGTTAAGCCAGTATCTGCTCCCTGCTTGTGTGTTGGAGGTCGCTGAGTAGTGCG CGAGCAAAATTTAAGCTACAACAAGGCAAGGCTTGACCGACAATTGCATGAAGAATCTGCT TAGGGTTAGGCGTTTTGCGCTGCTTCGCGGCAGTGAAAAAAATGCTTTATTTGTGAAATTTG TGATGCTATTGCTTTATTTGTAACCATTATAAGCTGCAATAAACAAGTTAACAACAACAATTG CATTCATTTTATGTTTCAGGTTCAGGGGGAGGTGTGGGAGGTTTTTTAAAGCAAGTAAAACC TCTACAAATGTGGTATGGCTGATTATGAGCTAGAGATCGAGACGTAGGTGTAGACGGCTTC AAGTCATGCTATGGCACGTCTCCATACCTGTCCCCCTTGTCCT GCCCCTGAAGCCGCCGGCGGACCTTCCGTGTTCCTGTTCCCCCCAAAGCCCAAGGATACCCT GATGATCAGCCGGACCCCCGAAGTGACCTGCGTGGTGGTGGATGTGTCCCACGAGGACCCT GAAGTGAAGTTCAATTGGTACGTGGACGGCGTGGAAGTGCACAACGCCAAGACCAAGCCC AGAGAGGAACAGTACAACAGCACCTACCGGGTGGTGTCCGTGCTGACAGTGCTGCATCAG GACTGGCTGAACGGCAAAGAGTACAAGTGCAAGGTGTCCAACAAGGCCCTGCCTGCCCCCA TCGAGAAAACCATCAGCAAGGCCAAGGGCCAGCCCCGCGAACCCCAGGTGTACACACTGCC TCCCAGCAGGGACGAGCTGACCAAGAACCAGGTGTCCCTGACCTGTCTCGTGAAAGGCTTC TACCCCTCCGATATCGCCGTGGAATGGGAGAGCAATGGCCAGCCCGAGAACAACTACAAGA CCACCCCCCCTGTGCTGGACAGCGACGGCTCATTCTTCCTGTACAGCAAGCTGACCGTGGAC AAGTCCCGGTGGCAGCAGGGCAACGTGTTCAGCTGCAGCGTGATGCACGAGGCCCTGCAC AACCACTACACCCAGAAGTCCCTGAGCCTGAGCCCCGGCAAATAATGATCCAGAGGATCAT

306

Stefania C. Carrara et al.

AATCAGCCATACCACATTTGTAGAGGTTTTACTTGCTTTAAAAAACCTCCCACACCTCCCCCT GAACCTGAAACATAAAATGAATGCAATTGTTGTTGTTAACTTGTTTATTGCAGCTTATAATG GTTACAAATAAAGCAATAGCATCACAAATTTCACAAATAAAGCATTTTTTTCACTGCCCCGTG CGGTGGGCTCTATGGCTTCTGAGGCGGAAAGAACCAGCTGGGGCTCTAGGGGGTATCCCC ACGCGCCCTGTAGCGGCGCATTAAGCGCGGCGGGTGTGGTGGTTACGCGCAGCGTGACCG CTACACTTGCCAGCGCCCTAGCGCCCGCTCCTTTCGCTTTCTTCCCTTCCTTTCTCGCCACGTT CGCCGGCTTTCCCCGTCAAGCTCTAAATCGGGGGCTCCCTTTAGGGTTCCGATTTAGAGCTT TACGGCACCTCGACCGCAAAAAACTTGATTTGGGTGATGGTTCACGTAGTGGGCCATCGCC CTGATAGACGGTTTTTCGCCCTTTGACGTTGGAGTCCACGTTCTTTAATAGTGGACTCTTGTT CCAAACTGGAACAACACTCAACCCTATCGCGGTCTATTCTTTTGATTTATAAGGGATTTTGCC GATTTCGGCCTATTGGTTAAAAAATGAGCTGATTTAACAAATATTTAACGCGAATTTTAACA AAATATTAACGTTTACAATTTAATTTGAGCAAAAGGCCAGCAAAAGGCCAGGAACCGTAAA AAGGCCGCGTTGCTGGCGTTTTTCCATAGGCTCCGCCCCCCTGACGAGCATCACAAAAATCG ACGCTCAAGTCAGAGGTGGCGAAACCCGACAGGACTATAAAGATACCAGGCGTTTCCCCCT GGAAGCTCCCTCGTGCGCTCTCCTGTTCCGACCCTGCCGCTTACCGGATACCTGTCCGCCTTT CTCCCTTCGGGAAGCGTGGCGCTTTCTCATAGCTCACGCTGTAGGTATCTCAGTTCGGTGTA GGTCGTTCGCTCCAAGCTGGGCTGTGTGCACGAACCCCCCGTTCAGCCCGACCGCTGCGCCT TATCCGGTAACTATCGTCTTGAGTCCAACCCGGTAAGACACGACTTATCGCCACTGGCAGCA GCCACTGGTAACAGGATTAGCAGAGCGAGGTATGTAGGCGGTGCTACAGAGTTCTTGAAG TGGTGGCCTAACTACGGCTACACTAGAAGAACAGTATTTGGTATCTGCGCTCTGCTGAAGCC AGTTACCTTCGGAAAAAGAGTTGGTAGCTCTTGATCCGGCAAACAAACCACCGCTGGTAGC GGTGGTTTTTTTGTTTGCAAGCAGCAGATTACGCGCAGAAAAAAAGGATCTCAAGAAGATC CTTTGATCTTTTCTACGGGGTCTGACTTCAAATATGTATCCGCTCATGAGACAATAACCCTGA TAAATGCTTCAATAATATTGAAAAAGGAAGAGTATGGAAAAAAAAATCACCGGCTATACCA CCGTGGATATTAGCCAGTGGCATCGTAAAGAACATTTTGAAGCGTTTCAGAGCGTGGCGCA GTGCACCTATAACCAGACCGTGCAGCTGGATATCACCGCGTTTCTGAAAACCGTGAAAAAA AACAAACACAAATTCTACCCGGCGTTTATTCATATTCTGGCCCGTCTGATGAACGCGCATCC GGAATTTCGTATGGCCATGAAAGATGGCGAACTGGTGATTTGGGATAGCGTGCATCCGTGC TATACCGTGTTTCATGAACAGACCGAAACCTTTAGCAGCCTGTGGAGCGAATATCATGATGA TTTTCGCCAGTTCCTGCATATTTATAGCCAGGATGTGGCGTGCTATGGCGAAAACCTGGCCT ATTTTCCGAAAGGCTTCATCGAAAACATGTTCTTTGTGAGCGCGAATCCGTGGGTGAGCTTT ACCAGCTTCGATCTGAACGTGGCGAACATGGATAACTTTTTTGCGCCGGTGTTTACCATGGG CAAATATTATACCCAGGGCGATAAAGTGCTGATGCCGCTGGCCATTCAGGTGCATCATGCG GTGTGTGATGGCTTTCATGTGGGCCGTATGCTGAACGAACTGCAGCAGTATTGTGATGAAT GGCAGGGCGGTGCGTAATAACTGTCAGACCAAGTTTACTCATATATACTTTAGATTGATTTA AAACTTCATTTTTAATTTAAAAGGATCTAGGTGAAGATCCTTTTTGATAATCTCATGACCAAA ATCCCTTAACGTGAGTTTTCGTTCCACTGAGC

SV40 polyA signal, Stuffer, Hinge, CH2 domain, CH3 domain, f1 ori, colE1 origin, CamR cassee

> Mammalian_Entry

Bulk Reformatting of Antibody Fragments to Final IgG Format

307

ACGGATTAGAAGCCGCCGAGCGGGTGACAGCCCTCCGAAGGAAGACTCTCCTCCGTGCGTC CTCGTCTTCACCGGTCGCGTTCCTGAAACGCAGATGTGCCTCGCGCCGCACTGCTCCGAACA ATAAAGATTCTACAATACTAGCTTTTATGGTTATGAAGAGGAAAAATTGGCAGTAACCTGGC CCCACAAACCTTCAAATGAACGAATCAAATTAACAACCATAGGATGATAATGCGATTAGTTT TTTAGCCTTATTTCTGGGGTAATTAATCAGCGAAGCGATGATTTTTGATCTATTAACAGATAT ATAAATGCAAAAACTGCATAACCACTTTAACTAATACTTTCAACATTTTCGGTTTGTATTACTT CTTATTCAAATGTAATAAAAGTATCAACAAAAAATTGTTAATATACCTCTATACTTTAACGTC AAGGAGAAAAAACCATGCAGTTACTTCGCTGTTTTTCAATATTTTCTGTTATTGCTAGCGTTT TAGCAGGGGAACAAAAGCTTATTTCTGAAGAGGACTTGGGATCCGCGCGGAAGACATGCG ACGCTAGAGATCCGTTTAAACTTGGacctgggagtggacacctgtggagagaaaggcaaagtggatgtca gtcactcaagtgtatggccagatcgggccaggtgaatatcaaatcctcctcgtggaaactgacaatcagcgcaga agtaatgcccgcgagagggagtactcaccccaacagctggatctcaagcctgccacacctcacctcgaccatccgc cgtctAaagaccgcctactaaacatcatcagcagcacctccgccagaaacaaccccgaccgccacccgctgccgc ccgccacggtgctcagcctaccgcgactgtgactggagacgcctctcgagaggccgatccggtcgatgcgg actcgctcaggtccctcggtggcggagtaccgcggaggccgacgggtccgatccaagagtactggaaagaccgcg aagagtgtcctcaaccgcgagcccaacagctggccctcgcagacagcgatgcggaagagagtgaggatctgacgg cactaaacgagctctgcatatagacctcccaccgtacacgcctaccgcccatgcgtcaacggggcggggaacg acaggaaagtcccggaggtgccaaaacaaactcccagacgtcaatggggtggagacggaaatc cccg tgagtcaaaccgctatccacgcccaggtgtactgccaaaaccgcatcaccatggtaatagcgatgactaatacgtaga tgtactgccaagtaggaaagtcccgtaaggtcatgtactgggcataatgccaggcgggccataccgtcagacgtca atagggggcggacggcatatgatacacgatgtactgccaagtgggcagtaccgtaaatactccacccagacgt caatggaaagtccctaggcgactatgggaacatacgtcaagacgtcaatgggcgggggtcggggcggtca gccaggcgggccataccgtaagatgtaacgcggaactccatatatgggctatgaactaatgaccccgtaagaa ctaaataactagtcaataatcaatgtcaacatggcggtcataggacatgagccaatataaatgtaccatctaaagta tatatgagtaaacggtctgacagaccaatgcaatcagtgaggcacctatctcagcgatctgtctatcgcatcc ataggcctgactccccgtcgtgtagataactacgatacgggagggcaccatctggccccagtgctgcaatgataccg cgagatccacgctcaccggctgtacatataggctcatgtccaatatgaccgccatg gacagaagactagt taaatagtaatcaaacggggtcaagca tagcccatatatggagccgcgacataacacggtaaatggcc cgcctggctgaccgcccaacgacccccgcccagacgtcaataatgacgtatgcccatagtaacgccaatagggact ccagacgtcaatgggtggagtatacggtaaactgcccacggcagtacatcaagtgtatcatatgccaagtccg ccccctagacgtcaatgacggtaaatggcccgcctggcaatgcccagtacatgaccacgggactcctacgg cagtacatctacgtaagtcatcgctaaccatggtgatgcggggcagtacaccaatgggcgtggatagcggtg actcacggggatccaagtctccaccccagacgtcaatgggagtgggcaccaaaatcaacgggactccaa aatgtcgtaataaccccgccccggacgcaaatgggcggtaggcgtgtacggtgggaggtctatataagcagagctcg tagtgaaccgtcagatcctcactctcccgcatcgctgtctgcgagggccagctggggctcgcgggaggacaaa ctccgcggtctccagtactcggatcggaaacccgtcggcctccgaacggtactccgccaccgagggacctgagcg agtccgcatcgaccggatcggaaaacctctcgagaaaggcgtctaaccagtcacagtcgcaaggtaggctgagcaccg tggcgggcggcagcgggtggcggtcgggggtctggcggaggtgctgctgatgatgtaaaaagtaggcggtcT agacggcggatggtcgaggtgaggtgtggcaggcgagatccagctgggggtgagtactccctctcaaaagcgggc aacctgcgctaagagtcagtccaaaaacgaggaggatgatacacctggcccg atctggccatacacg agtgacaatgacatccactgcctctctccacaggtgtccactcccaggtCCAAGTTTAAACGGATCTCTAG CGCGCTATGTCTTCCGCGCGGCTCAGGGGGGGGTGGCTCAGGTGGCGGTGGATCGGGAG GAGGTGGGAGTGGTGGCGGTGGTTCACAGGAACTGACAACTATATGCGAGCAAATCCCCT CACCAACTTTAGAATCGACGCCGTACTCTTTGTCAACGACTACTATTTTGGCCAACGGGAAG GCAATGCAAGGAGTTTTTGAATATTACAAATCAGTAACGTTTGTCAGTAATTGCGGTTCTCA CCCCTCAACAACTAGCAAAGGCAGCCCCATAAACACACAGTATGTTTTTTAATAGCGCCCGT CACAAAGAGCTTCAACAGGGGAGAGTGTTAATAGGCCGCAGCCGAACAGAAGTTGATTTCC GAAGAAGACCTCCCATGGTAGTGAGTTTAAACCCGCTGATCTGATAACAACAGTGTAGATG TAACAAAATCGACTTTGTTCCCACTGTACTTTTAGCTCGTACAAAATACAATATACTTTTCATT

308

Stefania C. Carrara et al.

TCTCCGTAAACAACATGTTTTCCCATGTAATATCCTTTTCTATTTTTCGTTCCGTTACCAACTTT ACACATACTTTATATAGCTATTCACTTCTATACACTAAAAAACTAAGACAATTTTAATTTTGCT GCCTGCCATATTTCAATTTGTTATAAATTCCTATAATTTATCCTATTAGTAGCTAAAAAAAGAT GAATGTGAATCGAATCCTAAGAGAATTGGGCAAGTGCACAAACAATACTTAAATAAATACT ACTCAGTAATAACGAATCGTAGTTTCATGATTTTCTGTTACACCTAACTTTTTGTGTGGTGCC CTCCTCCTTGTCAATATTAATGTTAAAGTGCAATTCTTTTTCCTTATCACGTTGAGCCATTAGT ATCAATTTGCTTACCTGTATTCCTTTACATCCTCCTTTTTCTCCTTCTTGATAAATGTATGTAGA TTGCGTATATAGTTTCGTCTACCCTATGAACATATTCCATTTTGTAATTTCGTGTCGTTTCTAT TATGAATTTCATTTATAAAGTTTATGTACAAATATCATAAAAAAAGAGAATCTTTTTAAGCAA GGATTTTCTTAACTTCTTCGGCGACAGCATCACCGACTTCGGTGGTACTGTTGGAACCACCT AAATCACCAGTTCTGATACCTGCATCCAAAACCTTTTTAACTGCATCTTCAATGGCCTTACCTT CTTCAGGCAAGTTCAATGACAATTTCAACATCATTGCAGCAGACAAGATAGTGGCGATAGG GTTGACCTTATTCTTTGGCAAATCTGGAGCAGAACCGTGGCATGGTTCGTACAAACCAAATG CGGTGTTCTTGTCTGGCAAAGAGGCCAAGGACGCAGATGGCAACAAACCCAAGGAACCTG GGATAACGGAGGCTTCATCGGAGATGATATCACCAAACATGTTGCTGGTGATTATAATACC ATTTAGGTGGGTTGGGTTCTTAACTAGGATCATGGCGGCAGAATCAATCAATTGATGTTGA ACCTTCAATGTAGGGAATTCGTTCTTGATGGTTTCCTCCACAGTTTTTCTCCATAATCTTGAA GAGGCCAAAACATTAGCTTTATCCAAGGACCAAATAGGCAATGGTGGCTCATGTTGTAGGG CCATGAAAGCGGCCATTCTTGTGATTCTTTGCACTTCTGGAACGGTGTATTGTTCACTATCCC AAGCGACACCATCACCATCGTCTTCCTTTCTCTTACCAAAGTAAATACCTCCCACTAATTCTCT GACAACAACGAAGTCAGTACCTTTAGCAAATTGTGGCTTGATTGGAGATAAGTCTAAAAGA GAGTCGGATGCAAAGTTACATGGTCTTAAGTTGGCGTACAATTGAAGTTCTTTACGGATTTT TAGTAAACCTTGTTCAGGTCTAACACTACCGGTACCCCATTTAGGACCACCCACAGCACCTA ACAAAACGGCATCAGCCTTCTTGGAGGCTTCCAGCGCCTCATCTGGAAGTGGAACACCTGT AGCATCGATAGCAGCACCACCAATTAAATGATTTTCGAAATCGAACTTGACATTGGAACGAA CATCAGAAATAGCTTTAAGAACCTTAATGGCTTCGGCTGTGATTTCTTGACCAACGTGGTCA CCTGGCAAAACGACGATCTTCTTAGGGGCAGACATTAGAATGGTATATCCTTGAAATATATA TATATATATTGCTGAAATGTAAAAGGTAAGAAAAGTTAGAAAGTAAGACGATTGCTAACCA CCTATTGGAAAAAACAATAGGTCCTTAAATAATATTGTCAACTTCAAGTATTGTGATGCAAG CATTTAGTCATGAACGCTTCTCTATTCTATATGAAAAGCCGGTTCCGGCGCTCTCACCTTTCC TTTTTCTCCCAATTTTTCAGTTGAAAAAGGTATATGCGTCAGGCGACCTCTGAAATTAACAAA AAATTTCCAGTCATCGAATTTACTCCAAGCTGCCTTTGTGTGCTTAATCACGTATACTCACGT GCTCAATAGTCACCAATGCCCTCCCTCTTGGCCCTCTCCTTTTCTTTTTTCGACCGAATTTCTT GAAGACGAAAGGGCCTCGTGATACGCCTATTTTTATAGGTTAATGTCATGATAATAATGGTT TCTTAGGACGGATCGCTTGCCTGTAACTTACACGCGCCTCGTATCTTTTAATGATGGAATAAT TTGGGAATTTACTCTGTGTTTATTTATTTTTATGTTTTGTATTTGGATTTTAGAAAGTAAATAA AGAAGGTAGAAGAGTTACGGAATGAAGAAAAAAAAATAAACAAAGGTTTAAAAAATTTCA ACAAAAAGCGTACTTTACATATATATTTATTAGACAAGAAAAGCAGATTAAATAGATATACA TTCGATTAACGATAAGTAAAATGTAAAATCACAGGATTTTCGTGTGTGGTCTTCTACACAGA CAAGATGAAACAATTCGGCATTAATACCTGAGAGCAGGAAGAGCAAGATAAAAGGTAGTA TTTGTTGGCGATCCCCCTAGAGTCTTTTACATCTTCGGAAAACAAAAACTATTTTTTCTTTAAT TTCTTTTTTTACTTTCTATTTTTAATTTATATATTTATATTAAAAAATTTAAATTATAATTATTTT TATAGCACGTGATGAAAAGGACCCAGGTGGCACTTTTCGGGGAAATGTGCGCGGAACCCCT ATTTGTTTATTTTTCTAAATACATTCAAATATGTATCCGCTCATGAGACAATAACCCTGATAA ATGCTTCAATAATATTGAAAAAGGAAGAGTTTAGAAAAACTCATCGAGCATCAAATGAAAC TGCAATTTATTCATATCAGGATTATCAATACCATATTTTTGAAAAAGCCGTTTCTGTAATGAA GGAGAAAACTCACCGAGGCAGTTCCATAGGATGGCAAGATCCTGGTATCGGTCTGCGATTC

Bulk Reformatting of Antibody Fragments to Final IgG Format

309

CGACTCGTCCAACATCAATACAACCTATTAATTTCCCCTCGTCAAAAATAAGGTTATCAAGTG AGAAATCACCATGAGTGACGACTGAATCCGGTGAGAATGGCAAAAGTTTATGCATTTCTTTC CAGACTTGTTCAACAGGCCAGCCATTACGCTCGTCATCAAAATCACTCGCATCAACCAAACC GTTATTCATTCGTGATTGCGCCTGAGCGAGACGAAATACGCGATCGCTGTTAAAAGGACAA TTACAAACAGGAATCGAATGCAACCGGCGCAGGAACACTGCCAGCGCATCAACAATATTTT CACCTGAATCAGGATATTCTTCTAATACCTGGAATGCTGTTTTCCCGGGGATCGCAGTGGTG AGTAACCATGCATCATCAGGAGTACGGATAAAATGCTTGATGGTCGGAAGAGGCATAAATT CCGTCAGCCAGTTTAGTCTGACCATCTCATCTGTAACATCATTGGCAACGCTACCTTTGCCAT GTTTCAGAAACAACTCTGGCGCATCGGGCTTCCCATACAATCGATAGATTGTCGCACCTGAT TGCCCGACATTATCGCGAGCCCATTTATACCCATATAAATCAGCATCCATGTTGGAATTTAAT CGCGGCCTAGAGCAAGACGTTTCCCGTTGAATATGGCTCATAACACCCCTTGTATTACTGTT TATGTAAGCAGACAGTTTTATTGTTCATGACCAAAATCCCTTAACGTGAGTTTTCGTTCCACT GAGCGTCAGACCCCGTAGAAAAGATAACTGTCAGACCAAGTTTACTCATATATACTTTAGAT TGATTTAAAACTTCATTTTTAATTTAAAAGGATCTAGGTGAAGATCCTTTTTGATAATCTCAT GACCAAAATCCCTTAACGTGAGTTTTCGTTCCACTGAGCGTCAGACCCCGTAGAAAAGATCA AAGGATCTTC TTGAGATCCTTTTTTTCTGCGCGTAATCTGCTGCTTGCAAACAAAAAAACCACCGCTACCAGC GGTGGTTTGTTTGCCGGATCAAGAGCTACCAACTCTTTTTCCGAAGGTAACTGGCTTCAGCA GAGCGCAGATACCAAATACTGTCCTTCTAGTGTAGCCGTAGTTAGGCCACCACTTCAAGAAC TCTGTAGCACCGCCTACATACCTCGCTCTGCTAATCCTGTTACCAGTGGCTGCTGCCAGTGG CGATAAGTCGTGTCTTACCGGGTTGGACTCAAGACGATAGTTACCGGATAAGGCGCAGCGG TCGGGCTGAACGGGGGGTTCGTGCACACAGCCCAGCTTGGAGCGAACGACCTACACCGAA CTGAGATACCTACAGCGTGAGCATTGAGAAAGCGCCACGCTTCCCGAAGGGAGAAAGGCG GACAGGTATCCGGTAAGCGGCAGGGTCGGAACAGGAGAGCGCACGAGGGAGCTTCCAGG GGGGAACGCCTGGTATCTTTATAGTCCTGTCGGGTTTCGCCACCTCTGACTTGAGCGTCGAT TTTTGTGATGCTCGTCAGGGGGGCCGAGCCTATGGAAAAACGCCAGCAACGCGGCCTTTTT ACGGTTCCTGGCCTTTTGCTGGCCTTTTGCTCACATGTTCTTTCCTGCGTTATCCCCTGATTCT GTGGATAACCGTATTACCGCCTTTGAGTGAGCTGATACCGCTCGCCGCAGCCGAACGACCG AGCGCAGCGAGTCAGTGAGCGAGGAAGCGGAAGAGCGCCCAATACGCAAACCGCCTCTCC CCGCGCGTTGGCCGATTCATTAATGCAGCTGGCACGACAGGTTTCCCGACTGGAAAGCGGG CAGTGAGCGCAACGCAATTAATGTGAGTTACCTCACTCATTAGGCACCCCAGGCTTTACACT TTATGCTTCCGGCTCCTATGTTGTGTGGAATTGTGAGCGGATAACAATTTCACACAGGAAAC AGCTATGACCATGATTACGCCAAGCTCGGAATTAACCCTCACTAAAGGGAACAAAAGCTGG CTAGT

GAL1 promoter, Aga2p signal sequence, Myc tag, SA misc, Enh MLP misc, SD misc, TPL CDS, CMV promoter, CMV enhancer, Stuffer, Aga2p, MAT alpha transcripon terminaon, LEU2, TRP1 promoter, CEN/ARS, KanR, ori

2. Other commonly used immunization hosts, such as humans, mice, rats, or chicken, may also be used. The primers for different immune repertoires only require the Golden Gate assembly overhangs described in Tables 1 and 2. Of special note is to ensure no GGC sites are found in the germline sequences, in order to not lose diversity during this cloning procedure. 3. The primers listed are for use with OmniRat immunizations, with unique forward primers to the leader sequence of VH and

310

Stefania C. Carrara et al.

VL kappa with reverse primers annealing to rat CH1 or CL domains. Other primer combinations can be used if other immunization hosts were utilized. 4. In this study, only CLκ antibodies were screened. Nonetheless, this workflow may also be performed for CLλ antibodies by exchanging the YSD_pDest_Kappa for YSD_pDest_Lambda, replacing only the constant region and using the corresponding VL Lambda primers for library generation. 5. In the exemplified study, only two screening rounds were performed as a sufficient binding of nearly the entire population was observed. Nonetheless, more screening rounds with increasing antigen stringency may be performed if necessary/ desired. 6. For improved characterization of, e.g., different effector functions, the Mammalian_entry vector can be designed to contain different mutations in CH2/CH3 domains. This modular toolbox would allow to sub-clone directly into, e.g., silenced or unsilenced Fc regions. 7. While this procedure was developed for YSD libraries, similar workflows have already been described for phage display or other display technologies. 8. The inoculation of 50 mL dYT medium is sufficient for a Midiprep Plasmid preparation; however, scaling up or down is also possible. For example, 5–7 mL cultures can be inoculated for Minipreps to sequence a larger number of colonies before larger preparations of the interesting candidates. 9. Sanger sequencing of isolated clones can be performed after the purification of plasmid DNA. If possible, E. coli colony sequencing can be performed before purification by, e.g., NightSeq sequencing (Microsynth). The number of sequenced clones can be defined depending on the diversity and differences in the sequences. 10. Other mammalian host systems may also be used for the production of full-length IgG molecules with the 2xeCMV promoter cassette, such as CHO cells. References 1. Kaplon H, Chenoweth A, Crescioli S, Reichert JM (2022) Antibodies to watch in 2022. MAbs 14. https://doi.org/10.1080/19420862. 2021.2014296 2. Mullard A (2021) FDA approves 100th monoclonal antibody product. Nat Rev Drug Discov 20:491–495. https://doi.org/10.1038/ d41573-021-00079-7 3. Schaffitzel C, Hanes J, Jermutus L, Plu¨ckthun A (1999) Ribosome display: an in vitro method

for selection and evolution of antibodies from libraries. J Immunol Methods 231:119–135. https://doi.org/10.1016/S0022-1759(99) 00149-0 4. 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

Bulk Reformatting of Antibody Fragments to Final IgG Format 5. Josephson K, Ricardo A, Szostak JW (2014) mRNA display: from basic principles to macrocycle drug discovery. Drug Discov Today 19:388–399. https://doi.org/10.1016/j. drudis.2013.10.011 6. Parthiban K, Perera RL, Sattar M et al (2019) A comprehensive search of functional sequence space using large mammalian display libraries created by gene editing. MAbs 11:884–898. https://doi.org/10.1080/19420862.2019. 1618673 7. 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 8. Wittrup DK, Kieke MC, Kranz DM et al (1999) Yeast cell surface display of proteins and uses thereof. pp 1–74. US6699658B1; h t t p s : // p a t e n t s . g o o g l e . c o m / p a t e n t / US6699658B1/en 9. 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/s41594-0180028-6 10. Rosowski S, Becker S, Toleikis L et al (2018) A novel one-step approach for the construction of yeast surface display Fab antibody libraries. Microb Cell Factories 17:3. https://doi.org/ 10.1186/s12934-017-0853-z 11. Sivelle C, Sierocki R, Ferreira-Pinto K et al (2018) Fab is the most efficient format to express functional antibodies by yeast surface display. MAbs 10:720–729. https://doi.org/ 10.1080/19420862.2018.1468952 12. Uchan´ski T, Zo¨gg T, Yin J et al (2019) An improved yeast surface display platform for

311

the screening of nanobody immune libraries. Sci Rep 9:382. https://doi.org/10.1038/ s41598-018-37212-3 13. Osborn MJ, Ma B, Avis S et al (2013) Highaffinity IgG antibodies develop naturally in Igknockout rats carrying germline human IgH / Ig κ / Ig λ loci bearing the rat C H region. J Immunol:1481–1490. https://doi.org/10. 4049/jimmunol.1203041 14. Fiebig D, Bogen JP, Carrara SC et al (2022) Streamlining the transition from yeast surface display of antibody fragment immune libraries to the production as IgG format in mammalian cells. Front Bioeng Biotechnol 10. https://doi. org/10.3389/fbioe.2022.794389 15. Carrara SC, Fiebig D, Bogen JP et al (2021) Recombinant antibody production using a dual-promoter single plasmid system. Antibodies 10:18. https://doi.org/10.3390/ antib10020018 16. 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 17. Bogen JP, Carrara SC, Fiebig D et al (2020) Expeditious generation of biparatopic common light chain antibodies via chicken immunization and yeast display screening. Front Immunol 11. https://doi.org/10.3389/ fimmu.2020.606878 18. Bogen JP, Carrara SC, Fiebig D et al (2021) Design of a trispecific checkpoint inhibitor and natural killer cell engager based on a 2 + 1 common light chain antibody architecture. Front Immunol 12. https://doi.org/10. 3389/fimmu.2021.669496

Chapter 17 Antibody-Secreting Cell Isolation from Different Species for Microfluidic Antibody Hit Discovery Ramona Gaa, Qingyong Ji, and Achim Doerner Abstract The recent advent of microfluidic-assisted antibody hit discovery as standard methodology accelerated pharmaceutical research. While work on compatible recombinant antibody library approaches is ongoing, the major source of antibody-secreting cells (ASCs) remains to be primary B cells of mostly rodent origin. As fainting viability and secretion rates can lead to false-negative screening results, careful preparation of these cells is an essential prerequisite for successful hit discovery. We here describe procedures to enrich plasma cells from relevant tissues of mice and rats and plasmablasts from human blood donations. Although freshly prepared ASCs yield the most robust results, suitable freezing and thawing protocols to preserve the viability and antibody secretory function can circumvent extensive process time and allow transferring of samples between laboratories. An optimized procedure is described to yield similar secretion rates after prolonged storage when compared to freshly prepared cells. Finally, the identification of ASC-containing samples can increase the probability of success of droplet-based microfluidics—two methods for pre- or in-droplet staining are described. In summary, the preparative methods described herein can facilitate robust and successful microfluidic antibody hit discovery. Key words Plasma cell, Plasmablast, Primary cell isolation, Antibody-secreting cell (ASC), Microfluidics

1

Introduction The plethora of therapeutic antibodies in clinical use or development indicates their remarkable efficacy and safety in several therapeutic areas. While historically most candidates were identified via phage [1] or yeast [2] libraries and B-cell cloning [3, 4] and hybridoma technology, recent development of commercially available microfluidics devices, [5, 6] services, and enhanced screening methodologies empowers accelerated yet robust hit discovery [7– 9]. Microfluidics are also applied for production of cell lines [8] or the identification of screening tools [6], as we report exemplary sorting plots for such applications in Fig. 2. Microfluidic antibody screening mainly relies on antibody-secreting cells (ASCs), due to

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2681, https://doi.org/10.1007/978-1-0716-3279-6_17, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

313

314

Ramona Gaa et al.

their ability to secrete antibodies. Most ASCs in rodents and humans are plasma cells [8], but plasmablasts differentiated from memory B cells [10] or secretion libraries [11] are also applied. Although published hit discovery processes [6, 8] with microfluidics deliver broad antibody panels, several levers for improvement remain in the optimization of initial immunization steps, the sampling time, interrogated organs [12, 13], and adapted protocols [14] for the respective host species. In addition, as faint viability and secretion rate can lead to false-negative screening results, careful preparation of ASCs is an essential prerequisite for successful hit discovery. We here describe procedures to enrich plasma cells and plasmablasts from relevant tissues of mice and rats (Fig. 1) and plasmablasts from human blood donations. Although freshly prepared ASCs yield the most robust results [15], cryopreservation while retaining functionality [16] can enable high flexibility of assay time and allow transferring of samples between laboratories. An optimized procedure is described to yield similar secretion rates after prolonged storage when compared to freshly prepared cells (Fig. 2). Finally, we report two methods for pre- or in-droplet ASC staining to identify ASCs in samples, with exemplary experiments shown in Fig. 3. In summary, the preparative methods described herein can facilitate robust and successful microfluidic antibody hit discovery.

2

Materials

2.1 Isolation from Mouse Tissue via MACS

1. EasySep™ Mouse CD138 Positive Selection Kit for RoboSep. 2. Primary mouse cells are from lymph nodes, bone marrow, and spleen. 3. 1× Dulbecco’s phosphate-buffered saline (DPBS). 4. 50 μM MACS Filter. 5. B-cell media: IMDM, GlutaMAX™ supplement medium supplemented with 10% heat-inactivated FBS, 1× non-essential amino acid solution (NEAA), 1× penicillin–streptomycin (Pen/Strep), 1× sodium pyruvate, and 1× 2-mercaptoethanol.

2.2 Isolation from Human Peripheral Blood via MACS

1. Human blood sample ~ 297 mL (in EDTA). 2. SepMate™-50 (IVD). 3. Lymphoprep. 4. Cell scrapers. 5. Washing buffer: 1× Dulbecco’s phosphate-buffered saline (DPBS) and 2% [v/v] heat-inactivated fetal bovine serum (FBS). 6. 50 μM MACS Filter.

ASC Isolation

315

Fig. 1 Isolation of antibody-secreting cells (red color) using two methods. (a) MACS isolates mouse plasma cells and plasmablasts using a mouse CD138 Positive Selection Kit from STEMCELL Technologies. (b) FACS sorts rat plasma cells and plasmablasts with an anti-rat CD138 antibody using a flow sorter. (c) An exemplary gating strategy for sorting rat plasma cells and plasmablasts on SONY MA900 sorter

7. EasySep™ Human Pan-B Cell Enrichment Kit for RoboSep. 8. EasySep™ Human Biotin Positive Selection Kit II for EasySep. 9. CD38 Monoclonal Antibody (HIT2), Biotin, eBioscience™. 10. 50 mL conical centrifugation tube. 2.3 Isolation from Rat Tissues via FACS

1. Anti-rat CD138 antibody (biotinylated or conjugated to a fluorochrome). 2. Primary rat cells from lymph nodes, bone marrow, and spleen. 3. FACS buffer containing 1× DPBS and 2% FBS. 4. Mouse anti-rat CD32 (BD Biosciences, cat. no. 550271).

316

Ramona Gaa et al.

Fig. 2 Exemplary antibody secretion sorting plots indicating the presence of antibody-secreting cells (ASCs). (a) Scheme of in-droplet FRET assay setup. (b) Sorting plot applying human plasmablasts obtained from vaccinated healthy donors. Three distinct populations within the sorting gate represent single droplets (left) and doublet and triplet droplet fusions (ascending populations) due to in this instance prolonged incubation time of 16 h. (c) Sort for antibody secretion as in (a), applying frozen and revived plasmablasts 10 months after isolation

5. PE-conjugated mouse anti-rat CD5 (BD Biosciences, cat. no. 554851). 6. PE-conjugated mouse anti-rat CD11b (BD Biosciences, cat. no. 554862). 7. PE-Cy-7-conjugated mouse anti-rat CD45R (Thermo Fisher, cat. no. 25-0460-82). 8. APC-conjugated streptavidin. 9. 1× Dulbecco’s phosphate-buffered saline (DPBS). 10. 5 mL round bottom polystyrene tube with cell strainer snap cap. 11. 7-AAD. 12. 50 mL conical centrifugation tube. 13. B-cell media: IMDM, GlutaMAX™ supplement medium supplemented with 10% heat-inactivated FBS, 1× non-essential amino acid solution (NEAA); 1× penicillin–streptomycin (Pen/Strep), 1× sodium pyruvate, and 1× 2-mercaptoethanol. 2.4 Storage and Revival of B Cells

1. CryoStor CS10 cell freezing medium. 2. Cryogenic vial. 3. Cryo 1 °C freezing container. 4. 50 mL conical centrifugation tube. 5. Cell media or buffer required in subsequent experiments.

2.5 ASC Staining with Cell Tracker or Anti-CD138 Antibody

1. CellTracker™ Orange CMRA Dye. 2. 50 mL conical centrifugation tube. 3. 1× Dulbecco’s phosphate-buffered saline (DPBS).

ASC Isolation

317

Fig. 3 Pre- or in-droplet ASC staining for sorting toward binding to cellular targets on co-encapsulated cells. Schematic of CytoTracker prestaining (a) or anti-CD138 in-droplet detection (b) applying green peak versus red peak sorting mode. (c) Proof of concept spiking experiment sorting plot. Anti-EGFR-specific Cetuximab IgG1 secreting cells were spiked 1:100 in non-secreting cells, stained with CytoTracker, co-encapsulated with EGFR-positive A431 cells, and sorted for cellular binding. (c) Similarly, EGFR-specific Mab108 hybridoma was mixed 1:1 with irrelevant OKT3 hybridoma cells and sorted for binding to EGFR-positive A431 tumor cells. Cell binding detection was realized by the addition of goat anti-mouse H+L-AF488 (2.5 μg/mL), while ASC signal was generated by PE/Dazzle 594 rat anti-mouse CD138 (2.5 μg/mL) and 1 μg/mL mouse Fc block

4. Growth media depending on the cell line. 5. PE/Dazzle 594 rat anti-mouse CD138 antibody. 6. AF488 AffiniPure Fab Fragment goat anti-mouse IgG (H + L). 7. Mouse BD Fc Block. 2.6

Equipment

1. 37 °C incubator for cell lines. 2. RoboSep. 3. Vi-CELL.

318

Ramona Gaa et al.

4. EasySep magnet. 5. Centrifuge for 50 mL conical vials. 6. Eppendorf Safe-Lock tube 1.5 mL. 7. Eppendorf Safe-Lock tube 2 mL. 8. Pipette controller. 9. Serological pipettes. 10. Piston-operated pipette (different sizes). 11. Pipette tips (different sizes). 12. -80 °C freezer. 13. Tank with liquid nitrogen. 14. Water bath. 15. A cell sorter such as SONY MA900, FACS Aria, or similar instruments.

3

Methods

3.1 Isolation of Antibody-Secreting Cells 3.1.1 Isolation from Mouse Tissue via MACS

For the isolation of murine antibody-secreting cells (ASC), primary cells from bone marrow, lymph nodes, and splenocytes can be harvested and prepared. Mice are immunized with the antigen of interest, and cells are isolated 5 days post-last boost. Cells can be resuspended in 1 mL RoboSep buffer directly after isolation and stored overnight at 4 °C before MACS cell isolation (see Note 1). Cells are enriched for ASCs by EasySep™ Mouse CD138 Positive Selection Kit according to the RoboSep manual. 1. Measure the cell titer of tissues separately. Therefore, make a 1: 30–1:100 dilution in 600 μL 1× DPBS and measure titer via Vi-CELL with settings for PMBC (see Note 2). 2. Mix tissues to get enough cells for one isolation or mix only bone marrow and lymph nodes (see Note 3). 3. For EasySep™ Mouse CD138 Positive Selection Kit, a cell titer of 1e+08 cells per mL is required. Dilute the sample in RoboSep buffer to adjust cell titer. 4. Filter cells through 50 μM MACS separation filter and transfer directly in a RoboSep tube (see Note 4). 5. Run RoboSep Protocol for EasySep™ Mouse CD138 Positive Selection Kit. 6. Take up the cell pellet containing CD138-positive mouse cells in 1 mL 1× DPBS or B-cell media. 7. Count the cell number of CD138-positive cells (a 1:10 dilution is recommended).

ASC Isolation

319

8. CD138-positive cells can be used directly for any further analysis such as FACS and microfluidics or frozen and stored in liquid nitrogen (view Subheading 3.2). 3.1.2 Isolation from Human Peripheral Blood via MACS

Human plasmablasts were isolated from 297 mL peripheral blood with ethylenediaminetetraacetic acid (EDTA) obtained by a donation from healthy individuals in this instance 6–7 days postvaccination to ascertain the highest levels of target-specific plasmablasts. Preparation of peripheral blood mononuclear cells (PBMCs) was performed following the SepMate™-50 (IVD; STEMCELL) manual, using 15 mL Lymphoprep (STEMCELL) as density gradient medium. Afterward, Pan-B cells were enriched using EasySep™ Human Pan-B Cell Enrichment Kit (STEMCELL), followed by CD38-positive cell enrichment for antibody-secreting cells. 1. Fill SepMate tubes with 15 mL room temperature Lymphoprep. Spin down shortly at 300 × g for 2 min. 2. Briefly add 15 mL washing buffer into SepMate tubes. Avoid mixing of washing buffer with Lymphoprep. 3. Carefully transfer 20 mL peripheral blood into each SepMate tube and avoid subsidence of blood under the separating disk (see Note 5). 4. Centrifuge at 1200 × g for 10 min with brake off. 5. Remove the supernatant to 30 mL labeling on the SepMate tube using a pipette or vacuum aspiration. (see Note 6). 6. Before the transfer of layer with enriched PBMCs, scratch attached cells off the tube wall with a cell scraper and pour of supernatant containing PBMCs into the new 50 mL tube (see Note 7). 7. Spin down at 300 × g, for 8 min with the brake on. 8. Discard supernatant, pool pellets of two 50 mL tubes, and wash with 40 mL washing buffer. 9. Centrifuge again with 120 × g for 10 min (brake off) to remove platelets. 10. Remove the supernatant again and pool all pellets into one tube. 11. Resuspend PBMCs in 40 mL washing buffer. 12. Measure the cell titer. Therefore, make a 1:10 dilution in 600 μL 1× PBS and measure titer via Vi-CELL with settings for PMBC. 13. Spin down at 300 × g, for 8 min with brake on. 14. For Pan-B enrichment with EasySep™ Human Pan-B Cell Enrichment Kit via RoboSep, adjust PBMCs to the required cell titer according to the kit manual with RoboSep buffer.

320

Ramona Gaa et al.

15. Filter cells through 50 μM MACS separation filter and transfer directly in RoboSep tube. 16. Run RoboSep Protocol for EasySep™ Human Pan-B Cell Enrichment Kit (see Note 8). 17. Measure the cell titer of isolated Pan-B cells. 18. Adjust cell titer again according to EasySep™ Human Biotin Positive Selection Kit II manual. 19. Follow instructions of EasySep™ Human Biotin Positive Selection Kit II using biotinylated CD38 monoclonal antibody (HIT2) in a concentration of 0.5 μg/100 μL. 20. Isolated CD38-positive blood cells are ready for use. 3.1.3 Isolation from Rat Tissues via FACS

Rats are chosen in some instances as the most appropriate species for immunization to achieve optimal cross-reactivities to certain species of interest or when human variable domain transgenic rats are applied to yield fully human antibody hit panels [17]. Isolation of rat plasma cells seems to be a challenge because no commercial kits are available on the market currently. An anti-rat CD138 antibody was generated in-house by immunizing mice and used to sort rat plasma cells by FACS for microfluidics-based antibody screening. Compared to MACS, the FACS method takes a longer time to obtain a large number of interest cell population and needs sorting equipment. However, FACS gives a higher purity of specific cell population, especially for rare cells. To isolate rat plasma cells, primary cells are prepared from different tissues similarly as described above for mice and resuspended in B-cell medium. Red blood cells from the spleen and bone marrow are preferably lysed with ammonium chloride-based lysing buffer. Cells are then stained with anti-rat CD138 antibody, fluorescence-labeled anti-rat CD11b for monocyte/macrophages, anti-rat CD5 for T cells, and anti-rat CD45R for B cells. CD45R+CD138+ plasmablasts and CD45R-CD138+ plasma cells are then sorted on SONY cell sorter or similar instruments. 1. Measure the cell titer of different tissues, respectively. Make a 1: 5–1:10 dilution in 600 μL with 1× DPBS and count the cell number via Vi-CELL with settings for PMBC. 2. Mix cells from different tissues to get enough cells for one isolation or pool cells from the lymph nodes of several rats (see Note 9). 1 ~ 2e + 08 cells are usually used for rat plasma cell isolation. 3. Spin down cells in a 50 mL Falcon tube at 500 × g for 5 min at 4 °C. 4. Wash cells once with FACS buffer and then resuspend cells to 2e + 07/mL in FACS buffer.

ASC Isolation

321

5. Add mouse anti-rat CD32 antibody to 1 μg/mL and incubate at room temperature for 5 min to block Fc receptors. 6. Add biotinylated anti-rat CD138 to 1 μg per mL and incubate the cells at 4 °C for 30 min (see Note 10). 7. Wash the cells with FACS buffer and spin down the tube at 500 × g for 5 min at 4 °C. 8. Dump the supernatant and add a cocktail of fluorochromelabeled antibodies containing PE-conjugated mouse anti-rat CD11b, PE-conjugated mouse anti-rat CD5, PE-Cy-7 conjugated mouse anti-rat CD45R, and APC-conjugated streptavidin (see Note 11). 9. Incubate the cells at 4 °C for 30 min. 10. Wash the cells twice with FACS buffer. 11. Resuspend the cells to 1 ~ 2e + 07 per mL in FACS buffer containing 7-AAD. Filter cells through a cell strainer cap to remove cell clumps. 12. Run stained cells on a flow cytometer. Bulk sort CD11b-CD5-CD45R+CD138+ plasmablasts and CD11b-CD5-CD45R-CD138+ plasma cells into a 5 mL polystyrene tube containing 1 mL of B-cell medium (Fig. 1) (see Note 12). 13. Sorted cells can be used directly for downstream screening analysis. 3.2 Storage and Revival of B Cells

1. Centrifuge remaining cells at 500 × g for 5 min. 2. Discard supernatant. 3. Resuspend cells in CryoStor CS10 with 1e + 06 – 1e + 07 cells per mL. 4. Pipette cells in the cryogenic vial (with ~1 mL per vial) and place tubes in Nalgene freezing container at -80 °C overnight. 5. Transfer in liquid nitrogen for long-term storage. 6. For thawing, warm up the cryo vial in 37 °C water bath for ~5 min. 7. Transfer cells in a new 50 mL centrifugation tube and slowly fill up with pre-warmed cell medium as recently described by Fecher et al. [16]. 8. Spin down for 5 min, 500 × g. 9. Resuspend cells in buffer, which is required in the next steps.

3.3 ASC Staining with Cell Tracker or Anti-CD138 Antibody

Depending on the available number of secreting cells and to ensure signal from noise differentiation, ASCs can either be prestained with appropriate dyes or detected within the microfluidic device by appropriate detection antibodies. Cell pre-labeling necessitates

322

Ramona Gaa et al.

thorough washing steps leading to an undesired loss of cells and library diversity, on the other hand generates a robust signal and avoids the use of additional secondary antibodies in-droplet-based closed homogenous reaction setups. For the Cyto-Mine device with 488 nm excitation and 594 nm emission spectrum, prestaining with 5 μM CellTracker™ Orange CMRA Dye (Thermo Fisher Scientific) was found to be most appropriate. For in-droplet staining of murine plasma cells or hybridoma cells, an anti-CD138 detection antibody not competing with preceding MACS reagents was added to the droplet generation mixture. The binding of these B-cell-specific detection antibodies generates a cell-focused fluorescence signal enabling peak mode screening as schematically presented in Fig. 3b. 3.3.1 Staining with Cell Tracker

1. Measure the cell titer of antibody-secreting cells. 2. Take up the fourfold excess of the final required cell amount. 3. Spin down the cells at 500 × g for 5 min. 4. Discard the supernatant. 5. Optional: Wash cells with 1× DPBS, spin down again, and remove the supernatant. 6. Resuspend cell pellet in 1× DPBS or serum-free media with a cell titer of 1e+06 viable cells/mL. 7. Add 5 μM CellTracker™ Orange CMRA Dye and mix gently (see Note 13). 8. Incubate for 30 min under growth conditions (e.g., 37 °C, 5% CO2, and 80% humidity). 9. Centrifuge again and remove the staining solution. 10. Wash once with fresh media. 11. Measure the cell titer again and use the required number of stained cells for the next steps.

3.3.2 Anti-CD138 Detection

Exemplary selection of detection antibodies for murine ASC for selection of antibody hits bound to a tumor target cell line as applied in a Cyto-Mine sort. 1. Use 2.5–5 μg/mL PE/Dazzle 594 rat anti-mouse CD138 antibody for the selection of cell surface binding antibodysecreting cells. 2. 2.5 μg/mL AF488 AffiniPure Fab Fragment goat anti-mouse IgG (H + L) for target-specific mAb detection on the target cell surface. 3. 1 μg/mL Mouse BD Fc Block to block Fcγ receptors on antibody-secreting cells (procedure according to manufacturer’s manual).

ASC Isolation

4

323

Notes 1. The faster the cells are prepared after isolation the better, an overnight storage or shipment is acceptable. 2. Measure PMBC cell titer with the following Vi-CELL settings: Minimum diameter (microns)

5

Maximum diameter (microns)

50

Number of images

50

Aspirate cycles

1

Tr ypan blue mixing cycles

3

Cell brightness (%)

85

Cell sharpness

100

Viable cell spot brightness (%)

65

Viable cell spot area (%)

5

Minimum circularity

0

Decluster degree

Medium

3. If cell availability is sufficient, it is recommended to use only primary cells from bone marrow and lymphocytes to obtain a higher enrichment of antibody-secreting cells. 4. Alternatively, all RoboSep protocols can be performed manually with EasySep magnets from STEMCELL. For any isolations, it is important to use the provided tubes by STEMCELL due to an optimized tube surface. RoboSep buffer must be pre-chilled (+4 °C) for MACS. 5. Pour the blood slowly to avoid mixing the density gradient layers. It could be helpful to hold the blood sample tube closer to the edge of the SepMate tube and not directly above the hole of the separating disk. After a few minutes, the erythrocytes automatically start to settle down. 6. After the centrifugation step, all unwanted blood cells, which do not belong to PBMCs, are located under the separation disk. Below and over the separation disk are phases of Lymphoprep. The top layer is plasma, and the PBMCs are located between the plasma and Lymphoprep layers. 7. Transfer the supernatant carefully but quickly, otherwise the blood parts below the partition will start to leak out. 8. Human Pan-B cells are enriched via a negative selection. Undesired cells are labeled and removed with magnetic particles; desired cells were transferred into a new tube.

324

Ramona Gaa et al.

9. In immunized rats, most of the antigen-specific plasmablasts or plasma cells are from lymph nodes although bone marrow contains much more CD138+ plasma cells. 10. Anti-rat CD138 antibody can be labeled a fluorochrome and used to stain plasma cells directly. 11. Fluorochrome-labeled antibodies need to be titrated, and the best dilution is chosen for staining. 12. A small aliquot of cells can be included and stained with an isotype control antibody for anti-rat CD138 antibody to set a sorting gate for plasma cells for the first few times. The percentage of CD138+ cells varies from immunization in transgenic OmniRat, ranging from 0.1 to 1% of total cells. 13. Working concentration may vary from cell line to cell line and must be tested out before. Also, the staining time could be adapted.

Acknowledgements We would like to thank Ralf Guenther and Saurabh Joshi for generating anti-rat CD138 antibody. Two figures were created using BioRender. References 1. Frenzel A, Schirrmann T, Hust M (2016) Phage display-derived human antibodies in clinical development and therapy. MAbs 8: 1177–1194 2. Krah S, Grzeschik J, Rosowski S et al (2018) A streamlined approach for the construction of large yeast surface display fab antibody libraries. Methods Mol Biol 1827:145–161 3. Carbonetti S, Oliver BG, Vigdorovich V et al (2017) A method for the isolation and characterization of functional murine monoclonal antibodies by single B cell cloning. J Immunol Methods 448:66–73 4. Zhang D, Jiang F, Zaynagetdinov R et al (2020) Identification and characterization of M6903, an antagonistic anti–TIM-3 monoclonal antibody. Onco Targets Ther 9(1): 1744921 5. Josephides D, Davoli S, Whitley W et al (2020) Cyto-mine: an integrated, picodroplet system for high-throughput single-cell analysis, sorting, dispensing, and monoclonality assurance. SLAS Technol 25(2):177–189 6. Winters A, McFadden K, Bergen J et al (2019) Rapid single B cell antibody discovery using

nanopens and structured light. MAbs 11: 1025–1035 7. Mazutis L, Gilbert J, Ung WL et al (2013) Single-cell analysis and sorting using dropletbased microfluidics. Nat Protoc 8:870–891 8. Ge´rard A, Woolfe A, Mottet G et al (2020) High-throughput single-cell activity-based screening and sequencing of antibodies using droplet microfluidics. Nat Biotechnol 38:715– 721 9. Wang Y, Jin R, Shen B et al (2021) Highthroughput functional screening for nextgeneration cancer immunotherapy using droplet-based microfluidics. Sci Adv 7:1–14 10. Rogers TF, Zhao F, Huang D et al (2020) Isolation of potent SARS-CoV-2 neutralizing antibodies and protection from disease in a small animal model. Science (80- ) 369:956– 963 11. Yanakieva D, Elter A, Bratsch J et al (2020) FACS-based functional protein screening via microfluidic co-encapsulation of yeast secretor and mammalian reporter cells. Sci Rep 10:1–13 12. Eyer K, Castrillon C, Chenon G et al (2020) The quantitative assessment of the secreted

ASC Isolation IgG repertoire after recall to evaluate the quality of immunizations. J Immunol 205:1176– 1184 13. Eyer K, Doineau RCL, Castrillon CE et al (2017) Single-cell deep phenotyping of IgG-secreting cells for high-resolution immune monitoring. Nat Biotechnol 35:977–982 14. Asensio MA, Lim YW, Wayham N et al (2019) Antibody repertoire analysis of mouse immunization protocols using microfluidics and molecular genomics. MAbs 11:870–883 15. Bounab Y, Eyer K, Dixneuf S et al (2020) Dynamic single-cell phenotyping of immune

325

cells using the microfluidic platform DropMap. Nat Protoc 15(9):2920–2955. Erratum in: Nat Protoc. 2021 Aug;16(8):4108 16. Fecher P, Caspell R, Naeem V et al (2018) B cells and B cell blasts withstand cryopreservation while retaining their functionality for producing antibody. Cell 7:50 17. 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

Chapter 18 Efficient Microfluidic Downstream Processes for Rapid Antibody Hit Confirmation Ramona Gaa, Hannah Melina Mayer, Daniela Noack, and Achim Doerner Abstract Microfluidics has been recently applied to better understand the spatial and temporal progression of the immune response in several species, for tool and biotherapeutic production cell line development and rapid antibody hit discovery. Several technologies have emerged that allow interrogation of large diversities of antibody-secreting cells in defined compartments such as picoliter droplets or nanopens. Mostly primary cells of immunized rodents but also recombinant mammalian libraries are screened for specific binding or directly for the desired function. While post-microfluidic downstream processes appear as standard steps, they represent considerable and interdependent challenges that can lead to high attrition rates even if original selections had been successful. In addition to next-generation sequencing recently described in depth elsewhere, this report aims at in detail explanations of exemplary droplet-based sorting followed by single-cell antibody gene PCR recovery and reproduction or single-cell sub-cultivation for crude supernatant confirmatory studies. Key words Microfluidics, Antibody hit discovery, PCR recovery, Small-scale expression, Single-cell cultivation

1

Introduction Microfluidics has become a standard methodology in several relevant scientific fields. Microfluidic approaches are for example applied in basic science to broaden the understanding of spatial and temporal progression of the immune response in different species [1, 2], as well in very applied technologies for pharmaceutical research such as tool generation [3], therapeutic biotherapeutic production clone identification (Fig. 1a), or hit discovery (Fig. 1b). Microfluidics for antibody discovery originated in academic laboratories [4] and are now a routine antibody discovery platform in pharmaceutical companies. A comprehensive review by Fitzgerald and Leonard covers the multitude of current approaches for antibody discovery [4]. Antibodies secreted from single cells are interrogated after compartmentalization in nanopens [3],

Stefan Zielonka and Simon Krah (eds.), Genotype Phenotype Coupling: Methods and Protocols, Methods in Molecular Biology, vol. 2681, https://doi.org/10.1007/978-1-0716-3279-6_18, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

327

328

Ramona Gaa et al.

Fig. 1 Exemplary antibody-secreting cell sorting plots via Cyto-Mine droplet microfluidic system. (a) Sorting hybridoma for secretion with the option to select only top secreting clones from the upper right part of the indicated gate. (b) Sort of primary human plasmablasts for specific binding to tetanus toxoid (TT). (c) Schematic of sorting pre-stained antibody-secreting cells (ASCs, red) for binding to cellular targets on co-encapsulated cells applying peak versus average fluorescence detection. (d–f) Full gating strategy for such a cellular binding sort. Anti-EGFR-specific cetuximab IgG1-secreting cells (ASCs) were spiked 1:100 in non-secreting cells, stained with CytoTracker, co-encapsulated with EGFR-positive A431 cells, and sorted for cellular binding. (d) Red signal indicates ASC-containing droplets, whereas green peak signals indicate secreted and target cell-bound antibody. (e) Dispensing gate similar to sorting gate in (a) with subsequent gating of green peak versus green average to exclude fused doublet but target binding negative droplets (light blue population above dispensing gate in (f))

microcapillaries [5–7], or picoliter droplets [8]. The antibody hit discovery presented herein bases on the droplet-based, commercially available Cyto-Mine stand-alone device [9]. A discussion on the advantages of the respective technologies can be found in Gaa et al. [10]. Most workflows focus on plasma cell interrogation obtained after immunization of wild-type or transgenic rodents [11] (Fig. 1), but also differentiated memory B cells, human plasma blasts [12, 13], or mammalian secretion libraries based on HEK293 or CHO [14] are applied. Within the past decade, CHO libraries have been generated by, among others, lentiviral delivery [15], nucleases [16], recombinases [17, 18], and episomal approaches [19]. Sorting can aim at identifying binders to recombinantly produced antigen or cells [20], internalizing clones or antibodies mediating a desired function indicated by an appropriate activated fluorescence reporter cell line [14, 21, 22]. Antibodies screened by microfluidics retain their native chain pairing, and several post-sort downstream processes have been applied. Although often believed

Microfluidic DSP

329

to be known standard methods, these steps represent an underestimated challenge that can lead to high attrition rates even if preceding selections had been successful. We would therefore here supply methods for two of three mostly described approaches. Application of next-generation sequencing (NGS) follows bulk export and enables evaluation of the full diversity of all sorted hit candidates. While common light chain or domain antibody repertoires comprising only one coding sequence of interest can be analyzed without additional efforts, evaluation of native VH + VL pairs necessitates elaborative cloning procedures, followed by full re-synthesis of potential hit antibody genes. This methodology unleashes its great potential when high-throughput capacities for both sequencing and synthesis are available, as exemplified and in detail described by Ge´rard et al. [11]. In case one of these factors is limited, single clones can be exported to individual microtiter plate wells, and potential antibody hit genes can be recovered by singlecell PCR (Fig. 2a). A detailed description of recovery, variable domain sub-cloning, and small-scale expression is given below. In case libraries comprising antibody hit-secreting mammalian cells are applied and sufficiently viable after selection, another option is single-cell sub-cultivation rather than sub-cloning. This requires comparatively low efforts, consumables, or labware, and supernatant analyses can yield confirmatory results [23]. High success rates for both single-cell PCR recovery and sub-cultivation are dependent on several factors such as cell stability; adherence state; and time spent on a microfluidic device correlating with resulting viability and require skilled laboratory personnel. The choice for the optimal microfluidic downstream workflow should therefore include an evaluation of all aforementioned relevant factors.

2

Materials

2.1 Droplet-Based Microfluidic Antibody Discovery Using CytoMine

1. Full growth cell culture media. 2. OptiPrep™ Density Gradient Medium. 3. 10% Pluronic™ F-68. 4. Cyto-Cartridge. 5. Flowmi. 6. Antibody-secreting cells. 7. DyLight 488/Alexa Fluor 488 labeled antibodies/antigens used as donor. 8. DyLight 594/Alexa Fluor 594-labeled reagents as acceptor. 9. 96-well PCR 384-well plates.

330

Ramona Gaa et al.

Fig. 2 Antibody hit candidate gene PCR recovery, reproduction, and binding confirmation. (a) Exemplary agarose gel of second PCR for VH amplification and corresponding VK gene. Wells with VH and VK band at corresponding height were purified for ligation into expression vectors. (b) Confirmation of anti-CRA antibodies’ production from transient expression tested via loading on AHC (antihuman capture) biosensors (BLI). Green: anti-CRA reference antibody; blue: irrelevant isotype control (cetuximab); red: supernatant from Expi293F cells with ExpiFectamine reagents; purple: fresh Expi293F media. All samples (black) with higher loading than the media control (red) were regarded as successfully produced. (c) Kinetic measurements of four selected anti-CRA binders, indicating high specificities and affinities. Dark blue, positive reference antibody. No unspecific binding of irrelevant mAb on biotinylated recombinant CRA ECD, no unspecific binding of samples on negative control biotinylated human EGFR, and no unspecific binding of samples on unloaded biosensor tips was observed (data not shown) 2.2 Single-Cell Antibody Gene PCR Recovery

1. Lysis buffer: 0.8 μL 0.1 M dithiothreitol (DTT); 0.4 μL RNaseOUT™ Recombinant Ribonuclease Inhibitor; and 6.8 μL nuclease-free water for a total of 8 μL lysis buffer per well. 2. Denaturing master mix containing 0.58 μL 50 μM random hexamers, 0.34 μL RNaseOUT™ Recombinant Ribonuclease Inhibitor, 1 μL 10% IGEPAL CA-630, and 4.08 μL nucleasefree water. 3. Reverse transcription (RT) master mix: 5.58 μL 5× SuperScript IV buffer, 2 μL 0.1 M DTT, 0.66 μL SuperScript™ IV Reverse Transcriptase, 0.84 μL 10 mM dNTP mix, 0.42 μL RNaseOUT™ Recombinant Ribonuclease Inhibitor, and 4.5 μL nuclease-free water. 4. First PCR Master Mix: 25 μL Platinum™ II Hot-Start PCR Master Mix (2×), 0.5 μL 10 μM forward primer, and 0.5 μL

Microfluidic DSP

331

10 μM reverse primer, filled with nuclease-free water to a total of 45 μL per well. For potential primer sequences, please see for example supportive information section of Gaa et al. [10]. 5. Nesting PCR master mix: 5 μL 10× AccuPrime PCR buffer 1, 0.2 μL AccuPrime™ Taq DNA Polymerase, high fidelity, 0.1 μL forward and 0.1 μL reverse primer (100 μM stock), and 39.6 μL nuclease-free water for each well. 6. E-Gel™ 48 Agarose Gels, 2%. 7. Perfect DNA™ Markers, 0.05–10 kbp. 2.3 High-Throughput Antibody Reformatting

1. PCR amplicons from nested PCR. 2. Wizard® SV 96 PCR Clean-Up System. 3. 95% ethanol. 4. Human pTT-IgH, pTT-IgK, or pTT-IgL vector backbone. 5. MultiShot™ StripWell Mach1™ T1 Phage-Resistant Chemically Competent E. coli. 6. SOC medium. 7. KingFisher™ Plastics for 96 deep-well format. 8. AeraSeal film. 9. Plate incubator for E. coli with shaking function. 10. Super Broth medium. 11. 100 mg/L ampicillin solution. 12. QIAprep 96 Turbo Kit.

2.4 High-Throughput Transient Antibody Production

1. ExpiFectamine™ 293 Transfection Kit. 2. Opti-Mem, reduced serum media. 3. Expi293F cells. 4. Adequate shaking flask for suspension cells. 5. Expi293 Expression Medium. 6. KingFisher™ Plastics for 96 deep-well plate. 7. AeraSeal film.

2.5 Single-Cell SubCultivation

1. DMEM/HamsF12. 2. Conditioned DMEM/HamsF12 medium. 3. ProCHO5. 4. Conditioned ProCHO5 medium. 5. Penicillin–streptomycin. 6. Water bath. 7. Greiner 384-well plate, flat-bottom, μ-Clear, PS, black.

332

Ramona Gaa et al.

8. Incucyte Incubator. 9. Incucyte 2021C Software. 10. Multichannel pipette. 11. Pipette tips (different sizes). 12. Cyto-Mine dispensed cells. 13. Breathe-easy sealing tape. 2.6 Supernatant Binding Confirmation via Biolayer Interferometry

1. Octet Red96 system. 2. 96-well plate for Octet. 3. Supernatant samples. 4. Adequate controls. 5. 1× Dulbecco’s phosphate-buffered saline (DPBS). 6. Biosensors for Octet RED96 system, e.g., AHC (antihuman capture). 7. Kinetics buffer: 1× DPBS with 0.1% Tween-20 and 1% BSA. 8. Antigen of interest. 9. ForteBio data analysis software.

2.7

Equipment

1. Cyto-Mine. 2. Thermal cycler. 3. Vi-CELL. 4. 37 °C incubator for cell lines (non-shaking and shaking). 5. Centrifuge for 50 mL conical vials. 6. Centrifuge for 1.5–2 mL tubes. 7. Eppendorf Safe-Lock tube 1.5 mL. 8. Eppendorf Safe-Lock tube 2 mL. 9. Pipette controller. 10. Serological pipettes. 11. Piston-operated pipette (different sizes). 12. Pipette tips (different sizes). 13. Water bath.

3

Methods

3.1 Droplet-Based Microfluidic Antibody Discovery Using CytoMine

Data reported herein were generated using the automated CytoMine workflow for droplet-based high-throughput antibody screening with pre-manufactured cartridge design and assay principles described in Fig. 1. Generally, up to two million 450 pl droplets can be generated in a 1 mL input volume containing up to one

Microfluidic DSP

333

million antibody-secreting cells (ASCs), and practically maximally 1.944 million droplets can be screened carrying Poissondistributed approximately 720,000 ASCs. Fo¨rster resonance energy transfer (FRET)-based IgG detection was used for the successful sorting of ASCs from hybridoma mixtures (Fig. 1a) or peripheral blood of vaccinated humans (Fig. 1b), exemplifying the applicability for the identification of high-secreting production cell line clones in pharmaceutical research and development. Figure 1c schematically explains a reporter setup for sorting toward cellular binding to a co-encapsulated tumor cell as example for a highly relevant application in therapeutic antibody hit discovery. A full gating strategy for a proof-of-concept spiking and re-enrichment experiment is shown in Fig. 1d–f. 1. Prepare basic for encapsulation freshly by combining 830 μL cell cultivation medium with 160 μL OptiPrep™ Density Gradient Medium and 10 μL 10% Pluronic™ F-68. 2. Remove the amount of basic encapsulation media, which will be replaced with the amount of detection antibodies, labeled antigens, or other necessary reagents, add them, and mix gently (see Note 1). 3. Measure the titer of cells of interest. 4. Take the required number of cells, spin down at 500 × g for 5 min, and wash with fresh full growth media (see Note 2). 5. Spin down again and discard the supernatant. 6. Resuspend cells of interest in encapsulation media (see Note 3). 7. Before loading the Cyto-Cartridge, filter cell suspension through 40 μM Flowmi® cell strainer. 8. Cells in 1 mL were encapsulated in two million droplets and incubated at 37 °C for 2–18 h, depending on the respective assay setup. 9. Sorting and dispensing in 96-well or 384-well plates are conducted following the manufacturer’s protocol (see Note 4). 3.2 Single-Cell Antibody Gene PCR Recovery

The Cyto-Mine device can dispense droplets containing one antibody-secreting cell. They can be either dispensed in 96- or 384-well plates, and droplet properties such as size, shape, and cell contents can be evaluated by visual inspection of dispensing channel images stored on the device. Also, other cell sorters such as Sony SH800 or BD FACS Aria can be used for single-cell dispensing. For a subsequent PCR analysis, a 96-well PCR plate can be used as a target plate containing 8 μL lysis buffer. Primers applied in subsequent nesting PCRs need to be designed with appropriate overhang sequences for reformatting into an expression vector backbone of choice (Fig. 2a).

334 3.2.1

Ramona Gaa et al. cDNA Synthesis

1. Prepare 96-well PCR plate containing 8 μL lysis buffer per well. 2. Dispense one cell or droplet per well. 3. Store plates at -80 °C (see Note 5). 4. Thaw plates on ice. 5. Add 6 μL denaturing master mix and incubate for 5 min at 65 ° C in thermal cycler. 6. Transfer 14 μL reverse transcription (RT) master mix per well. 7. Spin down for 30 s at 1500 rpm. 8. Place the plate in thermal cycler with the following settings: 20 °C for 15 min, 42 °C for 5 min, 25 °C for 10 min, 42 °C for 55 min, 90 °C for 10 min, and hold on 4 °C. 9. cDNA can be used directly or stored at -20 °C.

3.2.2

First PCR

1. Pipette 45 μL of First PCR Master Mix into each well of a 96-well PCR plate (see Note 6). 2. Add 5 μL of each cDNA to each prepared well of a 96-well PCR plate. 3. Spin down for 30 s at 1500 rpm. 4. Set the PCR conditions as follows: initial denaturation 95 °C for 2 min, 30 cycles of 95 °C for 30 s denature, 55 °C (VH) or 50 °C (for VK/VL) for 30 s anneal, 72 °C for 40 s extend, final elongation at 72 °C for 10 min, and hold 4 °C. 5. Store at -20 °C.

3.2.3

Nesting PCR

1. Keep the V-genes still separated. 2. Add 45 μL of nesting PCR master mix into a new PCR plate for each V-gene. 3. Transfer 5 μL PCR amplicon of the 1st PCR into each well. 4. Spin down for 30 s at 1500 rpm. 5. Use subsequent program settings: 1 cycle at 95 °C for 3 min, 5 cycles at 94 °C for 30 s, 42 °C for 30 s, and 72 °C for 30 s, followed by 30–50 cycles at 94 °C for 30 s, 55 °C for 30 s (for VH, VK, and VL), and 72 °C for 40 s, finalized by 72 °C for 5 min, and hold 4 °C.

3.2.4 PCR Analysis via 2% Agarose Gel

1. Prepare an E-Gel™ 48 Agarose Gels, 2% (Invitrogen). 2. Load 5 μL product onto each line. 3. Load 5 μL Perfect DNA™ Markers, 0.05–10 kbp. 4. Runtime 15–20 min.

Microfluidic DSP

3.3 High-Throughput Antibody Reformatting

335

All wells yielding an agarose gel lane of relevant height are considered hits. These hits are re-arrayed into a fresh PCR plate and purified using a Wizard® SV 96 PCR Clean-Up System. Afterward, the purified amplicons are reformatted into a human pTT-IgH, pTT-IgK, or pTT-IgL vector backbone via homologs recombination using chemically competent E. coli strain. 1. Perform PCR clean-up according to the manufacturer’s protocol with 95% ethanol for wash steps (PCR amplicon