key: cord-1043175-7wdtr5hx authors: Chen, Chuan; Sun, Zehua; Liu, Xianglei; Li, Wei; Dimitrov, Dimiter S. title: Protocol for constructing large size human antibody heavy chain variable domain (VH) library and selection of SARS-CoV-2 neutralizing antibody domains date: 2021-05-29 journal: STAR Protoc DOI: 10.1016/j.xpro.2021.100617 sha: 660fd7ff45c251e0c940a7f7300bdb2d7f96e210 doc_id: 1043175 cord_uid: 7wdtr5hx This protocol is a comprehensive guide to phage display based selection of virus neutralizing VH antibody domains. It details three optimized parts including 1) Construction of large size (theoretically > 1011) naïve human antibody heavy chain domain library, 2) SARS-COV-2 antigen expression and stable cell line construction, and 3) Library panning for selection of SARS-CoV-2 specific antibody domains. Using this protocol, we identified a high-affinity neutralizing human VH antibody domain, VH ab8, which exhibits high prophylactic and therapeutic efficacy. Lead Contact *Correspondence: CHC316@pitt.edu, liwei171@pitt.edu, mit666666@pitt.edu Summary This protocol is a comprehensive guide to phage display based selection of virus neutralizing V H antibody domains. It details three optimized parts including 1) Construction of large size (theoretically > 10 11 ) naïve human antibody heavy chain domain library, 2) SARS-COV-2 antigen expression and stable cell line construction, and 3) Library panning for selection of SARS-CoV-2 specific antibody domains. Using this protocol, we identified a high-affinity neutralizing human V H antibody domain, V H ab8, which exhibits high prophylactic and therapeutic efficacy. For complete details on the use and execution of this protocol, please refer to (Li, Schafer et al. 2020) Before You Begin Total RNA isolation and CDNA synthesis Timing: 3-5 days 1. Collect peripheral blood mononuclear cells (PBMCs) from 12 healthy donors' blood samples before SARS-CoV-2 pandemic using Ficoll-Paque PLUS gradient (Sigma, Cat#GE17-1440-02) according to the manufacturer's protocol (https://www.sigmaaldrich.com/technicaldocuments/protocols/biology/isolation-of-mononuclear-cells/recommended-standardmethod.html ). 1.5 × 10 9 PBMCs were collected for RNA isolation. Timing: 3-5 days 6. Generation of an empty plasmid named pIW-Zeo ( Figure 2A ) that contains a CMV promotor, Intron, a secret signal peptide (SPE) for extracellular expression, multiple cloning site (MCS), internal ribosome entry site (IRES), zeocin resistant gene, woodchuck posttranscriptional regulatory elements, BGH poly A, origin of replication and ampicillin resistant gene (This plasmid was generated before starting the protocol). 7. Prepare the phagemid for library construction with Qiagen plasmids Maxi-Prep kit (Qiagen Maxi-prep, Cat#12663) . The phagemid we used is a modified pComb3X vector (Cat#VPT4012, Creative Biogene), in which the HA-tag was replaced by Flag-tag ( Figure 2C , 2D). 8. Maintain Expi293F™ cells (Thermo, Cat#A14527) with Expi293™ Expression Medium in a CO 2 resistant incubator at 135 rpm, 8% CO 2 , 95% humidity according to the manufacturer's protocol (https://www.thermofisher.com/document-connect/documentconnect.html?url=https%3A%2F%2Fassets.thermofisher.com%2FTFS-Assets%2FLSG%2Fmanuals%2FMAN0006283_Expi293_Cells_UG.pdf&title=VXNlciBHdWlkZTog RXhwaTI5M0YgQ2VsbHM= ). Note: Determine viability and cell clumping using the trypan blue dye exclusion method and make sure cell viability is higher than 97%. The Expi293F™ cells can maintain in Expi293™ J o u r n a l P r e -p r o o f CCTGGAGCCTGGCGGACCCA H2F167 TGGGTCCGCCAGGCTCCAGGACAASGSCTTGAGTGG CDR2 H2F2 TGGGTCCGCCAGGCTCCAGGGAAGGCCCTGGAGTGG H2F1345 TGGGTCCGCCAGGCTCCAGGGAAGGGNCTRGAGTGG H2R1 ATTGTCTCTGGAGATGGTGACCCTKYCCTGRAACTY H2R36 ATTGTCTCTGGAGATGGTGAATCGGCCCTTCACNGA H2R24 ATTGTCTCTGGAGATGGTGACTMGACTCTTGAGGGA H2R5 ATTGTCTCTGGAGATGGTGACSTGGCCTTGGAAGGA H2R7 ATTGTCTCTGGAGATGGTAAACCGTCCTGTGAAGCC FR3F ACCATCTCCAGAGACAATTCC FR3 FR3R GTCCTCGGCTCTCAGGCTG H3F3p AGCCTGAGAGCCGAGGACACRGCYTTRTATTACTGT CDR3 H3F1p257p AGCCTGAGAGCCGAGGACACAGCCAYRTATTACTGT H3Fother AGCCTGAGAGCCGAGGACACRGCYGTRTATTACTGT H3R GTGGCCGGCCTGGCCACTTGAGGAGACGGTGACC ALL-R GTCGCCGTGGTGGTGGTGGTGGTGGCCGGCCTGGCCACTTG EcoRV-RBD-P1 TCAGTGATATCCCAAATATAACCAATCTCTGCCCATTCG RBD-AviTag-His EcoRI-His-AviTag-P2 TCAGTGAATTCCTATTAGTGATGGTGGTGGTGATGGCTACCCTC GTGCCACTCG Degenerated bases: R = A,G; Y = C,T; M = A,C; K = G,T; S = C,G; N = A,C,G,T. The restriction sites were marked yellow, complementary sequences were marked underline, overlapping sequences were marked red ( Figure 1 and Figure 4 ). . These two reagents are critical for library construction. Alternatives: Throughout this protocol, we refer to several specific kit for many standard biology techniques. Investigators may substitute other commercially available kit as needed. Step-by-Step Method Details V H library construction High diversity and large size are the characteristics of a good library and the basic requirements for successful library panning leading to selection of high affinity binders. Due to the limited diversity of PBMCs B cell antibody gene, it is unlikely to generate a large size V H library by direct PCR amplification of V H region from the cDNA of PBMCs. Therefore, overlap-PCR was used to combine different antibody heavy chain complementarity-determining regions (CDRs) to increase the size and diversity of V H genes. In this protocol, a well-defined stable scaffold IGVH3-23 ( Figure 3A ) was chosen as basic scaffold for framework Region (FR): FR1, FR2, FR3 and FR4. The primer combinations used to amplify CDRs and overlap CDRs-FRs are listed in Table 1 and ii. Run all the PCR samples with 2% agarose gel to verify the size and purify the right size PCR products ( Figure 3B ) from the agarose gel with QIAquick Gel Extraction Kit. CRITICAL: The template cDNA is a mixture; the primers may have non-specific binding with the cDNA to PCR out some non-specific DNA bands. In the experiment, only the target bands will be purified for further overlapping PCR process to assembly full-length V H . b. Over-lapping PCR to assemble full-length V H with FR1, CDR1, CDR2, FR3 and CDR3 DNA purified from step 2a as template, ALL-F and ALL-R as primes using the following cycling conditions: i. 50 L PCR system: ii. Run all the PCR samples with 2% agarose gel to verify the size and purify the right size PCR products ( Figure 3B ) from the agarose gel with QIAquick Gel Extraction Kit. Note: If the over-lapping PCR productivity is low. Assemble the fragments with two steps (trouble shooting 1). . Stir with tips 20 times and keep on ice 5-10 min. b. Transfer the bacteria into ice pre-chilled 0.1 cm cuvette. c. Electroporate using the pre-set program with setting at 1.8 kV/ 200 ohms/25 F. d. Wash the cuvette with 1 mL pre-warm 2-YT medium three times and transfer the bacteria into 50 mL pre-warm 2-YT medium. Note: During large scale electroporation, 10 electroporated/transformed TG1 vials are resuspended with 500 mL pre-warm 2-YT medium in a 2L shake flask. Around 150 electroporation vials are needed for > 10 11 size library construction, thus 15 × 2L shake flasks with 7.5 L pre-warm 2-YT medium is needed to resuspend all the electroporated/transformed TG1 cells. 8. Recover the bacteria at 37℃, 200 rpm for 30 min, Aliquot 1/10 5 bacteria (5 l culture medium) from each bottle into a 1.5 ml centrifuge tube which contain 995 l fresh 2-YT medium for titration. 10-fold serial dilute the bacteria and take 1/10 7 , 1/10 8 and 1/10 9 of total bacteria (100 l of 10-, 10 2 -and 10 3 -fold diluted samples) from the dilutions, plate onto 2-YT-Agar plates with 100 g/mL ampicillin. Select the transformants by adding 100 g/mL ampicillin and 2% glucose, shaking at 37℃, 200 rpm for 2~3 hours till the OD6 00 reach to ~0.6-0.8. 9. Add M13KO7 helper phage into the cells with multiplicity of infection (MOI) = 10:1, incubate at 37℃, 45 min. Mix every 15 min during incubation. Note: MOI means the ratio of phages added to bacteria. OD 600 of 1 corresponds to approximately 5 × 10 8 TG1 cells per ml. If the OD 600 ≈ 0.6, total TG1 cells equals 0.6 × 5 × 10 8 /mL × 7500 mL ≈ 2.25 × 10 12 , and ≈ 2.25 × 10 13 M13KO7 helper phage are needed for the infection. 10. Centrifuge the bacteria at 5,000 g for 5 min at 4℃. Resuspend the bacterial pellet with fresh 2-YT contain 100 g/mL ampicillin, 50 g/mL kanamycin. 11. Shaking at 30℃, 200 rpm overnight (12~15 hours). 12. Spin down the bacteria at 8,000 g for 15 min at 4℃, transfer the supernatant into new bottles, add 25% volume of PEG/NaCl solution into the supernatant and incubate on ice for 1 hour. 13. Centrifugation of the mixture at 11,000 g for 20 min at 4℃. Discard supernatant and resuspend the pellet with 50 mL ice cold DPBS per liter of culture. 14. Centrifugation again at 10,000 g for 10 min at 4℃ to eliminate the bacterial contamination. CRITICAL: This bacteria elimination step is critical before storage, it will eliminate most of the bacteria in the phage library. The remaining bacteria will be eliminated by multiple rounds of wash during the phage panning process. 15. Transfer all the supernatant into a new bottle, add 20% glycerol and aliquot into 1 mL/vial, stock at -80℃ for long-term storage. Determine the phage titer by detecting OD 268 (1 OD ≈ 5 × 10 12 phage) (Durr, Nothaft et al. 2010 ). The phage titer should above 5 × 10 12 /mL CRITICAL: 50 mL ligation with 375 g digested pComb3X and 125 g V H is needed for a > 10 11 size library. ~150 g ligation DNA can be recovered with DNA Clean & Concentrator-5 kit (efficiency is ~30%). 1 g ligation is used for one electroporation. Each electroporation will result in > 10 9 colonies, 150 electroporation will produce > 1.5 × 10 11 colonies in total. According to the Sanger sequencing results, ~30% of the plasmid are empty vector, junk DNA or V H with stop codon. So, 50 mL ligation should get a > 1.05 × 10 11 size library. 16. Quality checks of the library: a. Randomly pick > 100 colonies and scale up in 4 mL 2-YT medium with 100 g/mL ampicillin, shaking at 200 rpm overnight (12~15 hours). b. Purify the plasmid with QIAprep Spin Miniprep Kit and elute with 30 L water. c. Send all the plasmid for sequencing and analyze all the sequences. CRITICAL: All the V H sequences should be different from each other. ~30% of the plasmid are empty vector, junk DNA or V H with stop codon. Note: If repeat V H s are detected, the quality of the library is not good and library size might be smaller than 10 11 , so re-construction of the library to make the size >10 11 is needed. Note: We are using two asymmetric SfiI sites for library construction. Sequencing results showed there is no issue of orientation of V H insert in our pComb3X system. Note: Beside randomly picked > 100 colonies for sequencing, we also evaluated the library quality and performance by panning with more than 5 antigens. For a high-quality library, several dozens of different V H binders binding to each antigen were enriched. Multiple binders and the diversity of the selected V H binders revealed good quality of library. Compare with the microbial expression system, mammalian cell expression has the advantage of protein expression, more advanced peptide folding and post-translational modifications (Gray 2001 , Khan 2013 . In general, antigens expressed from mammalian cells which have proper folding and post-translational modifications are essential for full biological activity and successful selection of high affinity specific binders. Here we used the SARS-CoV-2 receptor-binding domain (RBD) as an example. This section includes molecular cloning, expression, and purification (Figure 1, Figure 2 and Figure 5 ). Timing: 3 days 17. Clone the RBD into expression plasmid pIW-Zeo ( Figure 2A Verification of the plasmid by DNA sequencing and keep the right clones (pIW-Zeo-RBD, Figure 2B ). 18. Transfection and Protein Expression (30 mL expression, Figure 5A ): a. The day before transfection, seed the cells at a density of 2.0 × 10 6 viable cells/mL and incubate at 37℃ in incubator shaker rotating at 135 rpm with 8% CO 2 and 85% humidity. b. On the day of transfection, determine number and viability of the cells using an automated cell counter. Dilute the cells to 3 × 10 6 viable cells/mL with Expi293™ Expression Medium. c. Add 27 mL cell suspension into a 125 mL Erlenmeyer shaker flask. Return the cells to the incubator. d. prepare DNA-PEI complexes as follows: i. Dilute 30 g of plasmid DNA into 1.5 mL Expi293™ Expression Medium, mix gently and incubate for 5 min at RT (20~25℃). ii. Dilute 120 g of PEI into 1.5 mL Expi293™ Expression Medium, mix gently and incubate for 5 min at RT (20~25℃). iii. After 5 min incubation, mix the plasmid DNA with the PEI. Incubate at RT (20~25℃) for 10-20 min. iv. After the DNA-PEI complex incubation is complete, add the complex into shaker flask from Step 18c. Gently swirl the flask. v. Return and incubate the cells in the incubator. Maintain 7 days at 37℃. Note: Some expressed proteins might have degradation, denaturation, or aggregation during several days culturing at 37℃. Short time culture can protect protein from degradation, denaturation or aggregation to get proteins of higher quality, while long time culturing has higher yield. For the RBD-AviTag expression, the yield of 3 days culture is ≈ 5mg/L, and of 7 days is higher than 10 mg/L. We also found the activity of RBD-AviTag purified at day 3 is better than that of day 7. Thus, according to our experience, 3 days of culture is sufficient for RBD-AviTag expression. e. (optional) 24 hours after transfection, the transfected cells can be used for stable cell pool selection with the following steps ( Figure 5A ): i. Take 5 mL transfected cells into 50 mL centrifugation tube. Centrifugation at 300 g, 3 min. Return the supernatant into the expression bottle. ii. Resuspend the cells with 5 mL Expi293™ Expression Medium contain 250 g/mL Zeocin. Return and incubate the cells in the incubator. iii. Change medium at day 1, day 2, day 3 and day 4 with 5 mL fresh Expi293™ Expression Medium containing 250 g/mL Zeocin. Note: Because of the transfection efficiency variation, the viability of the cells at day 4 might be different. If the viability is < 30% at day 4, reduce the Zeocin concentration to 50 g/mL for the following three days selection. In our RBD-AviTag stable cell pool selection process, the viability is ~64% at day 4. So, 250 g/mL Zeocin was used for 7 days selection. Expression Medium with 50 g/mL Zeocin. Add the suspension cell into a 125 mL Erlenmeyer shaker flask, Return and incubate the cells in the incubator maintain 3 to 6 days (Stable cell pool is ready for expression). Timing: 1 day 19. Protein purification a. Prepare the Ni-NTA gravity column: i. Transfer 1 mL His Pur™ Ni-NTA Resin (10 mg protein/ 1 mL beads) into a gravity column. ii. Allow the resin to settle, then let the excess buffer drain through the column by gravity flow. iii. Wash the resin with 20 mL Milli Q water. iv. Wash the resin with 20 mL binding buffer. b. Cell culture media preparation: i. Transfer the expression medium at step 18 into a 50 mL conical tubes, spin down at 300 g for 5 min at 4℃. ii. Pour supernatant into new conical tubes. Spin down at 12,000 g for 10 min at 4℃. c. Protein loading and elution: i. Load all the media into the column at step 19a. ii. Once the medium has completely entered the column, wash the column with 40 mL wash buffer. iii. Elution the RBD-AviTag with 4 mL elution buffer. iv. Change buffer with DPBS with 10 kD ultra-filter. Follow instructions in the manufacturer's protocol (https://www.thermofisher.com/documentconnect/documentconnect.html?url=https%3A%2F%2Fassets.thermofisher.com%2FTFS-Assets%2FLSG%2Fmanuals%2FMAN0015695_2162596_PierceProteinConcent rator_2_6mLPI.pdf&title=VXNlciBHdWlkZTogUGllcmNlIFByb3RlaW4gQ29uY2V udHJhdG9yLCBQRVMgLSAzSywgMTBLLCAzMEsgYW5kIDEwMEsgTVdDTzsgMi 02bUw= ). v. Measure the protein concentration on NanoDrop Lite. Run SDS-PAGE gel to characterize the purified RBD-AviTag ( Figure 5B . Adapted from ). Note: The productivity of RBD-AviTag is ~10 mg/L. 1 L purified RBD-AviTag sample is needed for concentration detection with NanoDrop Lite. After obtaining the RBD-AviTag antigen, its quality was checked by human angiotensin-converting enzyme 2 (ACE2) binding ELISA. Then it is ready for bio-panning. This section describes streptavidinmagnetic beads-based library panning strategy. 20. Biotin-label the RBD-AviTag (RBD-Bio) antigen with BirA Biotin-Protein Ligase Kit according to the manufacture's protocol (https://www.avidity.com/showpdf.asp?N=B9B6C28E-1976-4CFE-89DE-E37D1D8DB2CA ). 21. Library panning with streptavidin magnetic Beads: a. Thaw a phage library aliquot (1 mL/aliquot, ≈ 5 × 10 12 phage/mL). Add 250 L (V PEG/NaCl :V phage = 1:4) PEG/NaCl solution into the phage, incubate on ice for 20 min. b. Centrifugation at 12,000 g for 10 min at 4℃. Discard supernatant and resuspend the pellet with 200 l DPBS. c. Take 2 × 10 12 library phage diluted into 1 mL 3% BSA-DPBS. Add 10 g RBD-Bio antigen, rotation 1.5 hours at RT (20~25℃). CRITICAL: The amount of antigen used for panning are different in each round while the incubation time did not change. In our protocol we used 10 g for 1 st round, 2 nd round 5 g, 3 rd round 1 g, 4 th round 0.2 g. (Use high concentration antigen at 1 st round can help to enrich more binders. Lower concentration of antigens in the following panning process will help to enrich high affinity V H domains). d. At the same time, pick single colony from TG1 bacterial plate (LB-agar plate) or directly take 1 L TG1 from commercial stock and scale up in 20 mL 2-YT medium at 37℃, 200 rpm until OD 600 ≈ 0.5. e. Take 25 L streptavidin beads into a 1.5 mL centrifugation tube, wash the beads with 1 mL DPBS twice with magnets stand. Note: Blocking the beads with 3% BSA-DPBS for 1 hour at RT (20~25℃) before use is good for panning. Because we have an extra step of depletion with streptavidin beads before panning (start from 2 nd round), Streptavidin beads blocking is optional. f. Transfer the phage at step 21c into washed streptavidin beads, rotate the tubes at 10 RPM/min for 1 hour at RT (20~25℃). g. Wash the beads with 1 mL 0.05% PBST (0.05% Tween-20 in DPBS) for 5 times, then wash with DPBS twice. Note: Wash the beads with 0.05% PBST as follows: 1 st round 5 times, 2 nd round 8 times, 3 rd round 15 times and 4 th wash 20 times. (Increasing the washing numbers throughout the phage display selection can help to decrease low affinity binders. So that high affinity antibodies can be enriched efficiently). h. Resuspend the beads with 5 mL TG1 cells (OD 600 ≈ 0.5) in 15 ml culture tube, incubate the mixture at 37℃, 45 min. Mix every 15 min during incubation. CRITICAL: The phage on the beads can infect TG1 cells effectively, so the elution step is not required. i. Take 1/10 3 infected TG1 cells (5 l culture medium) into a 1.5 ml centrifuge tube which contain 995 l fresh 2-YT medium. 10-fold serial dilute the bacteria with 2-YT medium and take 1/10 4 , 1/10 5 and 1/10 6 of total bacteria plate onto 2-YT-Agar plates with 100 g/mL ampicillin. The rest add 100 g/mL ampicillin and 2% glucose, shaking at 37℃, 200 rpm for 2 hours. j. Add 10 L M13KO7 helper phage (Cat#18311019, Thermo) into the cells (10 L/5 ml, phage titer: 10 11 /mL), incubate at 37℃, 45 min. Mix every 15 min during incubation. k. Centrifuge at 4,000 g for 10 min at RT (20~25℃). Discard supernatant and resuspend the pellet with 50 mL 2-YT contain 100 g/mL ampicillin, 50 g/mL kanamycin. shaking at 30℃, 200 rpm 12-15 hours. 22. Phage purification and start of the next round: a. Check the colonies the next day. b. Transfer the cell culture into a 50 mL centrifuge tube, Spin down the bacteria at 8,000 g for 10 min at 4℃ and collect the supernatant with a new 50 mL centrifugation tube. c. Add 25% volume of PEG/NaCl solution into the supernatant, mix and incubate on ice for 1 hour. d. Centrifuge at 11,000 g for 20 min. Discard supernatant and resuspend the pellet with 2 mL cold DPBS. e. Separate the phage into two 1.5 mL tubes and centrifugation at 15,000 g for 1 min at 4℃ to eliminate the bacteria. Note: This bacterial eliminate step is critical before storage, it will eliminate most of the bacteria in the phage. The rest bacteria will be eliminated by multiple rounds of wash during the next round of phage panning. f. Transfer the phage to two new tubes and determined the titer by detecting OD 268 (1 OD ≈ 5 × 10 12 phage). CRITICAL: If the phage titer is lower than 10 11 /mL, do not start next round of panning. Check the quality of TG1 cell and M13KO7 helper phage to confirm both are good for experiment. Repeat the first round of panning and culture 15 hours before purifying the phage. g. Add 20% glycerol into one tube and stock in -80℃. Take 10 12 phage from the other tube for next round of panning. Note: If the phage purified from previous round is lower than 10 12 , use > 10 11 phage for panning. The phage input can be less in the later rounds of panning due to the decreased diversity. Pause point: add 20% glycerol into both tubes and stock in -80℃ for long term storage till next round of panning. 23. Complete round 2, round 3 and round 4 (optional) phage panning. CRITICAL: Before starting next round of panning, take 10 11 to 10 12 previous round purified phage, dilute the phage into 1 mL 5% Milk-DPBS (second and fourth rounds panning blocking the phage with 5% Milk-DPBS, third round panning blocking the phage with 3% BSA-DPBS) and incubate with 50 L streptavidin-magnetic beads at 10 rpm for 1 hour at RT (20~25℃). Clean the beads with magnets. The beads depleted phage is used for next round of panning. Note: Depending on the enrichment, in some cases round 4 is necessary to enrich high-affinity binders and narrow down the whole hits. (Optional) polyclonal phage ELISA to detect the enrichment of binders: a. Coat the ELISA plates with RBD-AviTag (5 g/mL in DPBS, 50 L/well) overnight (12~15 hours) at 4℃. b. Wash plates 3 times with 0.05% PBST, block with 100 L 5% Milk-DPBS 2 hours. c. Wash plates 3 times with 0.05% PBST. d. Add 100 L 5% Milk-DPBS diluted phage (~10 11 phage from each round), shake 1 hour with 200 rpm/min at RT (20~25℃). e. Wash plates with 0.05% PBST, 4 times. f. Add 100 L Anti-M13 Antibody (HRP) (Sino Biological Inc, Cat#11973-MM05T-H, 1:2000 diluted in 5% Milk-DPBS), shake 45 min with 200 rpm/min at RT (20~25℃). g. Wash plates with 0.05% PBST, 5 times. h. Add 50 μL TMB into each well and reaction 5-10 min. i. Stop the reaction with 50 L 2 M H 2 SO 4 , read the optical density (OD) with a microplate reader at 450 nm. After three rounds of panning, colonies will be picked for expression of heavy chain antibody domains which will be further screened by supernatant ELISA ( Figure 6A . ). a. Add 180 L 2-YT medium with 100 g/mL ampicillin to each well of 96-well plate. b. Pick single colony into each well, Incubate the 96-well plates at 37℃, 200 rpm until OD 600 ≈ 0.5. Note: Take 200 l OD 600 = 0.2, 0.4, 0.6, 0.8 TG1 cells into a 96 well plates, detect the OD 600 with Synergy HTX Multi-Mode Reader to get a standard curve. Compare the absorbance with the standard curve. Once most wells reach to OD 600 ≈ 0.5, add IPTG to induce V H expression. c. Add 20 L IPTG stock (10 mM) into each well (final IPTG concentration = 1 mM) to induce V H expression. Incubate the plates at 30℃, 200 rpm 12 to 15 hours. d. Coat the ELISA plates with RBD-AviTag (2 g/mL in DPBS, 50 L/well) overnight (12~15 hours) at 4℃. e. Wash plates 3 times with 0.05% PBST, and block with 100 L 3% BSA-DPBS 2 hours. f. Wash plates 3 times with 0.05% PBST. g. Add 50 L 6% BSA-DPBS into each well of the RBD-AviTag coated plates. Spin the bacteria expression plates (step 24c) at 4,000 g for 5 min at 4℃, transfer 50 L supernatant into each well, shake 2 hours with 200 rpm/min at RT (20~25℃). 25. V H antibody domain expression and purification a. Transform HB2151 competent cells with selected plasmid DNA by heat-shock 1 min at 42℃. Recovering 45-60 min at 37℃, 200 rpm, plate the transformed cells onto 2-YT-Agar plates with 100 g/mL ampicillin and 1% glucose. Incubate at 37℃ overnight (12~15 hours). CRITICL: The pComb3X has an amber stop codon (TAG) between flag tag and gene III ( Figure 2D ). TG1 is an amber codon (TAG) suppressor strain, allowing translation to read through the codon and to produce a full-length V H -gene III fusion protein. For V H antibody domain library construction, each electroporation will result in > 10 9 colonies and 150 electroporation will make > 10 11 size V H library. Transient expression productivity of RBD-AviTag with Expi293 system should yield > 10 mg/L and the RBD-AviTag expression stable cell pool will be generated within 10 days. High affinity RBD V H binders which compete with human ACE2 for binding to RBD will be selected after three rounds of panning. We have got 16 unique V H binders with this protocol and the equilibrium dissociation constant of these binders is from 300 nM to 4 nM. Compared with single human B cell isolation, phage display is based on bacterial-expression system. In general, it has limitations on protein expression, folding and post-translational modification. Our V H antibody domain library is generated from healthy human donors with the CDRs naturally grafted from human PBMC cDNA. It may lower the possibility of non-specific binding to human cells compare with synthetic library. However, non-specific binding was found in some of the selected V H antibody domains. Compare with scFv, Fab and VHH libraries, due to large size of our V H library, we have selected out many high affinity V H binders (nM range affinity). There are no significant affinity limitations compare with other libraries. However, the V H antibody domains are much easier to aggregation and aggregations are detected in most of the selected antibody domains. So, characterization of the selected domains one by one to figure out the best functional candidates for further therapeutic development is needed. Troubleshooting Problem 1: The productivity of full-length V H assemble by over-lapping PCR is low (step 2). Potential Solution: Assemble the full-length V H with two steps: 1. Over-lapping PCR to assemble FR1, CDR1 and CDR2 with ALL-F/H2R1, H2R24, H2R36, H2R57 primers to get FR1-CDR1-CDR2-FR3. Assemble FR3 and CDR3 with FR3F/All-R primers to get FR3-CDR3-FR4 using the following cycling conditions: 50 Run all the PCR samples with 2% agarose gel to verify the size and purify the right size PCR products from the agarose gel with QIAquick Gel Extraction Kit. Problem 2: Low efficiency of electroporation, hard to generate large size phage library (step 7). Potential Solution: The most common reason of low efficiency is the poor quality of digested V H antibody domains (step 3) or DNA degradation by over-digestion and ligation (step 4). Long extra protection bases (such as 15 bp) in primers can be added in front of the SfiI restriction site. This will help improve the digestion of SfiI to get high quality digested V H . Run agarose gel to check the ligation DNA quality. If degradation is detected, shorten the ligation time to 48 hours. Low affinity of the selected V H antibody domains (step 25f). Potential Solution: Use lower concentration of antigens for panning and increase the washing times to enrich high affinity domains. Selected V H domain candidates aggregate (Step 25g). Human variable domains rapidly aggregate when heated to 80 ~ 85 °C (this condition well above their melting temperatures). The aggregated phage can be eliminated by centrifugation while the nonaggregation phage remains in the supernatant. Use the heat-treated phage supernatant for panning can help to get non-aggregating V H domains (Dudgeon, Rouet et al. 2012) . Before panning (start from second round), heat the phage at 80℃ for 10min, then keep on ice for 10 min and centrifugation for 10 min at 15,000 g (white pellet can be found at the bottom of the centrifuge tube after centrifugation). Collect the supernatant into a new tube to eliminate the aggregated V H s for panning. Problem 5: J o u r n a l P r e -p r o o f During Expi293F™ cell maintain, cell clumping was detected, and cell viability is lower than 97% (In "Before You Begin, step 8"). Thaw a new Expi293F™ should always be the first choice. If clumping still detected, check the shake speed of the CO 2 Resistant incubator, and subculture the cells every two days with fresh medium. Check the cell viability with trypan blue. If viability is lower than 97%, do not let the cell density over 5 × 10 6 /mL and do not maintain the cells more than 5 days without subculture. Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contacts, Chuan Chen (CHC316@pitt.edu), Wei Li (liwei171@pitt.edu), and Dimiter S. Dimitrov (mit666666@pitt.edu). For cell lines and plasmid please contact Dimiter S. Dimitrov (mit666666@pitt.edu). All other materials are available commercially. This study did not generate any unique datasets or code. Step 1: Molecular cloning of the pIW-Zeo-RBD plasmid (4 days were needed). Step 2: Overnight transfection of the Expi293 TM cell with pIW-Zeo-RBD plasmid (1 day was needed). Step 3: After transfection, cells were maintained at 37 ℃ in incubator shaker rotating at 135 rpm with 8% CO2 and 85% humidity for 3~7 days for RBD- General strategy for the generation of human antibody variable domains with increased aggregation resistance The Escherichia coli glycophage display system Overview of protein expression by mammalian cells Gene expression in Mammalian cells and its applications High Potency of a Bivalent Human VH Domain in SARS-CoV-2 Animal Models Enhanced elicitation of potent neutralizing antibodies by the SARS-CoV-2 spike receptor binding domain Fc fusion protein in mice Potent neutralization of SARS-CoV-2 by human antibody heavy-chain variable domains isolated from a large library with a new stable scaffold We would like to thank the members of our group, Dontcho Jelev, Megan Shi, Cynthia Adams, Du-San Baek, Ye-Jin Kim, and Xiaojie Chu for their helpful discussions. We also thank our previous colleagues Drs. Weizao Chen, Tianlei Ying and Zhongyu Zhu for their helpful discussion. This work was supported by the University of Pittsburgh Medical Center. Chuan Chen drafted the manuscript, edited by Wei li, Zehua Sun, Xianglei Liu and Dimiter S. Dimitrov. The authors declare no competing interests. J o u r n a l P r e -p r o o f