Convalescent and uninfected human blood samples

Volunteers aged 23 to 93 with a history of convalescent COVID-19 were enrolled from April 2020 to January 2021. Blood samples were collected on the day or one day before discharging from the hospital after symptom resolution. Duration is the time between PCR positive and blood sample collection. All blood samples used in this study were collected before taking any SARS-CoV-2 vaccination. Uninfected healthy volunteers aged 36 to 62 who do not have severe immunological symptoms such as immunodeficiency, autoimmune, and allergic diseases were enrolled, and we confirmed uninfected/unvaccinated donors by their clinical history and ELISA titer. Detailed information on the cohort is in Supplementary Table1. PBMCs and plasma samples were isolated by density gradient centrifugation with Ficoll-Paque PLUS (GE Healthcare) and stored at 80C until use. The study was approved by the Ethical Committee for Epidemiology of Hiroshima University (E-2011) for studies involving humans. Informed consent was obtained from all subjects involved in the study.

For single-cell sorting, PBMCs were treated with FcX blocking antibodies (BioLegend, #4422302) to reduce non-specific labeling of the cells. PBMCs were stained with S-trimer-Strep-tag, CD19-APC-Cy7 (BioLegend, #302217), and IgD-FITC (BioLegend, #348206) for 20min on ice. After washing, cells were stained with Strep-Tactin XT-DY649 (IBA) for 20min on ice. The cells were resuspended in FACS buffer (PBS containing 1% FCS, 1mM EDTA, and 0.05% NaN3) supplemented with 0.2g/ml propidium iodide (PI) to exclude dead cells. Cell sorting was performed on Special Order System BD FACSAria II (BD Biosciences) to isolate S-trimer+ CD19+ IgD cells from the PI live cell gate. Cells were directly sorted into a 96-well PCR plate. Plates containing single-cells were stored at 80C until proceeding to RT-PCR. Flow cytometric data were acquired on BD LSRFortessa (BD Biosciences) or CytoFLEX S (Beckman Coulter). Flow cytometric data were analyzed using BD FACSDiva (v8.0.2, BD Biosciences), CytExpert software (v2.4, Beckman Coulter), or FlowJo software (v10.8.1, BD Biosciences).

Single-cell sorted PCR plates were added to each well by 2l of pre-RT-PCR mix containing the custom reverse primers. After heating at 65C for 5min, plates were immediately cooled on ice. 2l of the pre-RT-PCR2 (PrimeScript II Reverse Transcriptase, Takara Bio) mix was added to each well. For RT reaction, samples were incubated at 45C for 40min followed by heating at 72C for 15min, then cooled on ice. For PCR amplification of full-length immunoglobulin heavy and light chain genes, PrimeSTAR DNA polymerase (Takara Bio) and custom primers were used. For first PCR, the initial denaturation at 98C for 1min was followed by 25 cycles of sequential reaction of 98C for 10s, 55C for 5s, and 72C for 1.5min. For second PCR, the initial denaturation at 98C for 1min was followed by 35 cycles of sequential reaction of 98C for 10s, 58C for 5s, and 72C for 1.5min. PCR fragments were assembled into a linearized pcDNA vector using NEBuilder HiFi DNA Assembly Master Mix (New England Biolabs) according to the manufacturers instructions. The pcDNA3 (Invitrogen) vectors containing an Ig light chain gene and the pcDNA4 (Invitrogen) vectors containing an Ig heavy chain gene were simultaneously transfected into Expi293 cells using Expi293 Expression System Kit (Thermo Fisher Scientific). Four days after the transfection, the culture supernatants were collected and subjected to ELISA.

For the production of recombinant S-trimer, soluble S protein (amino acids 11213), including the T4 foldon trimerization domain, a histidine tag, and a strep-tag, was cloned into the mammalian expression vector. The protein sequence was modified to remove the polybasic cleavage site (RRAR to A), and two stabilizing mutations were also introduced (K986P and V987P; wild-type numbering)34,35. The human codon-optimized nucleotide sequence encoding for the S protein of SARS-CoV-2 (GenBank: MN994467) was synthesized commercially (Eurofins Genomics). A soluble version of the S protein (amino acids 11213), including the T4 foldon trimerization domain, a histidine tag, and a strep-tag, was cloned into the mammalian expression vector pCMV. The protein sequence was modified to remove the polybasic cleavage site (RRAR to A), and two stabilizing mutations were also introduced (K986P and V987P; wild-type numbering)35. The gene encoding RBD of SARS-CoV-2, Wuhan-Hu-1, was synthesized and cloned into vector pcDNA containing a human Ig leader sequence and C-terminal 6xHis tag. RBD mutants were generated by overlap PCR using primers containing mutations. The vector was transfected into Expi293 cells and incubated at 37C for 4 days. Supernatants were purified using Capturem His-Tagged Purification kit (Takara Bio), then dialyzed by PBS buffer overnight. Protein purity was confirmed by SDS-PAGE. Protein concentration was determined spectrophotometrically at 280nm.

MaxiSorp ELISA plates (Thermo Fisher Scientific) were coated with 2g/ml purified spike RBD or trimer in 1xBBS (140mM NaCl, 172mM H3BO3, 28mM NaOH) overnight at 4C, and then blocked with blocking buffer containing 1% BSA in PBS for 1h. Antibodies diluted in Reagent Diluent (0.1% BSA, 0.05% Tween in Tris-buffered Saline) were added and incubated for 2h. HRP-conjugated antibodies were added and incubated for 2h. Wells were reacted with the TMB substrate (KPL) and the reaction was stopped using 1M HCl. The absorbance at 450nm was measured on iMark Microplate Reader (Bio-Rad) and analyzed on MPM 6 software (Bio-Rad). Antigen-specific Ig titers were determined using serial serum dilution on antigen-coated wells next to Ig-capturing antibody standard wells on the same ELISA plate.

The binding affinity of obtained antibodies to RBD was examined by the BLItz system (Sartorius Japan) using protein A-coated biosensors. 10g/ml of antibody was captured by the biosensor and equilibrated, followed by sequential binding of each concentration of RBD. For dissociation, biosensors were dipped in PBST for 900sec. Results were analyzed on BLItz Pro (v1.3.1.3, Molecular Devices).

psPAX2 (Addgene, no.12260) was a gift from Didier Trono. pCDNA3.3_CoV2_B.1.1.7 (Addgene, no.170451) for Alpha-S and pcDNA3.3-SARS2-B.1.617.2 (Addgene, no.172320) for Delta-S proteins, were gifts from David Nemazee36. pTwist-SARS-CoV-2 18 B.1.351v1 (Addgene, no.169462) for Beta-S protein was a gift from Alejandro Balazs37. Lentiviral vector, pWPI-ffLuc-P2A-EGFP for luciferase reporter assay and pTRC2puro-ACE2-P2A-TMPRSS2 for the generation of 293T cell line susceptible to SARS-CoV-2 infection was created from pWPI-IRES-Puro-Ak-ACE2-TMPRSS2, a gift from Sonja Best (Addgene, no.154987) by In-Fusion technology (Takara Bio). pcDNA3.4 expression plasmids encoding SARS-CoV-2 S proteins with human codon optimization and 19 a.a deletion of C-terminus (C-del19) from Wuhan, D614G, and Omicron were generated by assembly of PCR products, annealed oligonucleotides, or artificial synthetic gene fragments (Integrated DNA Technologies, IDT) using In-Fusion technology. For Delta plus, Kappa and Lambda variants, S proteins with only RBD, D614, and P681 mutations were created from pcDNA3.4 encoding human codon-optimized Wuhan S protein (C-del19). LentiX-293T cells (Takara Bio) and 293T cells were maintained in culture with Dulbeccos Modified Eagles Medium (DMEM) containing 10% fetal bovine serum (FBS), penicillin-streptomycin (Nacalai tesque), and 25mM HEPES (Nacalai tesque).

To generate stable 293T-ACE2.TMPRSS2 cells (293T/TRCAT), lentiviral vector VSV-G-pseudotyped lentivirus carrying ACE2 and TMPRSS2 genes were produced in LentiX-293T cells (Clontech) by transfecting with pTRC2puro-ACE2-P2A-TMPRSS2, psPAX2 (gag-pol), and pMD2G-VSV-G (envelope) using PEI-MAX (Polysciences). Packaged lentivirus was used to transduce 293T cells (Applied Biological Materials) in the presence of 5g/mL polybrene. At 72h post-infection, the resulting bulk transduced population positive for Human ACE2 expression stained by FITC-anti-ACE2 Antibody (Sinobiological) was sorted by Special Order System BD FACSAria II (BD Biosciences) and maintained in the culture medium in the presence of 2g/ml of puromycin.

Pseudoviruses bearing SARS-Cov2 S-glycoprotein and carrying a firefly luciferase (ffLuc) reporter gene were produced in LentiX-293T cells by transfecting with pWPI-ffLuc-P2A-EGFP, psPAX2, and either of S variant from Wuhan, D614G, Alpha, Beta, Delta, Delta plus, Kappa, Lambda, or Omicron using PEI-MAX (Polyscience). Pseudovirus supernatants were collected approximately 72h post-transfection and used immediately or stored at 80C. Pseudovirus titers were measured by infecting 293T/TRCAT cells for 72h before measuring luciferase activity (ONE-Glo Luciferase Assay System, Promega, Madison, WI). Pseudovirus titers were expressed as relative luminescence units per milliliter of pseudovirus supernatants (RLU/ml). For neutralization assay, pseudoviruses with titers of 14106RLU/ml were incubated with antibodies or sera for 0.5h at 37C. Pseudovirus and antibody mixtures (50l) were then inoculated with 5g/ml of polybrene onto 96-well plates that were seeded with 50l of 1104 293T/TRCAT cells/well one day before infection. Pseudovirus infectivity was scored 72h later for luciferase activity measured on ARVO X13 and 2030 Workstation (Perkin Elmer). The serum dilution or antibody concentration causing a 50% reduction of RLU compared to control (ED50 or IC50, respectively) were reported as the neutralizing antibody titers. ED50 or IC50 were calculated using a nonlinear regression curve fit on Prism (v9.0, GraphPad).

VeroE6/TMPRSS2 cells (African green monkey kidney-derived cells expressing human TMPRSS2, purchased from the Japanese Collection of Research Bioresources (JCRB) Cell Bank, JCRB1819, were maintained in DMEM containing 10% FBS and 1mg/ml G418 at 37C in 5% CO2. The virus was propagated in VeroE6/TMPRSS2 cells and the virus titer was determined by the 50% tissue culture infectious dose (TCID50) method and expressed as TCID50/ml38. The viral strains used are SARS-CoV-2/JP/Hiroshima-46059T/2020 (B.1.1, D614G, EPI_ISL_628993239), SARS-CoV-2/JP/HiroC77/2021, (AY.29, Delta, EPI_ISL_6316561), and SARS-CoV-2/JP/FH-229/2021 (BA.1.1, Omicron, EPI_ISL_11505197).

The serially diluted antibody (50l) was mixed with 100 TCID50/50l of the virus and reacted at 37C for 1h, then inoculated into VeroE6/TMPRSS2 cells to determine the minimum inhibitory concentration (MIC). Alternatively, the infectivity of the reacted antibody-virus mixture was measured by inoculating to 8 wells of a 96-well plate and observing cytopathic effects (CPE) or by the plaque assay using 10% methylcellulose to determine 50% effective dose (ED50) of the antibody. SARS-CoV-2 infection was performed in the BSL3 facility of Hiroshima University.

SARS-CoV-2 was incubated with mAb at twice the concentration of the EC50 corresponding to that viral load at 37C for 60min. After the incubation, 100l of the mixture was added to one well of a 24-well plate with confluent VeroE6/TMPRSS2 cells and incubated for 72h at 37C with 5% CO2. The supernatants were collected as an escape-mutant virus when CPE was manifested. A no-antibody-control was included to confirm the amount of test virus required.

Viral RNA was extracted from virus-infected culture medium by using Maxwell RSC Instrument (Promega, AS4500). cDNA preparation and amplification were done in accordance with protocols published by the ARTIC network (https://artic.network/ncov-2019) using V4 version of the ARTIC primer set from Integrated DNA Technologies to create tiled amplicons across the virus genome. The sequencing library was prepared using the NEB Next Ultra II DNA Library Prep Kit for Illumina (New England Biolabs, E7645). Paired-end, 300bp sequencing was performed using MiSeq (Illumina) with the MiSeq reagent kit v3 (Illumina, MS-102-3003). Consensus sequences were obtained by using the DRAGEN COVID lineage software (Illumina, ver. 3.5.6). Variant calling and annotation were performed using the Nextclade website (https://clades.nextstrain.org).

RBD from Wuhan-Hu-1, Delta, and Omicron variant and Fab fragment from NCV2SG48 with 6xHis-tag expressed in Expi293F cells (Thermo Fisher Scientific) were purified using Ni-NTA Agarose resin (QIAGEN). Fab fragment of NCV2SG53 was isolated from papain digests of the monoclonal antibody expressed in Expi293F cells using HP Protein G column (Cytiva). Purified each Fab fragments and RBD were mixed in the molar ration of 1:1.2 and incubated on ice for 1h. The mixture was loaded onto a Superdex 200 increase 10/300 GL column (Cytiva) equilibrated in 20mM Tris-HCl pH7.5, 150mM NaCl for removing the excess RBD. Fractions containing RBD and each Fab were collected and concentrated for crystallization. Chromatography was performed using NGC Chromatography Systems (BIO-RAD) and ChromLab v6 (BIO-RAD).

Crystallization was carried out by the sitting-drop vapor diffusion method at 20C. Crystals of RBD (Wuhan-Hu-1)-Fab (NCV2SG48) were grown in 2l drops containing a 1:1 (v/v) mixture of 7.5mg/ml RBD solution and 0.1M Bis-Tris pH5.5, 0.5M ammonium sulfate and 19% PEG3350. Crystals of RBD (Delta)-Fab (NCV2SG48) were grown in 2l drops containing a 1:1 (v/v) mixture of 7.5mg/ml RBD solution and 0.1M Bis-Tris pH6.5, 17.5% PEG10000, 100mM Ammonium acetate and 5% Glycerol. Crystals of RBD (Omicron BA.1)-Fab (NCV2SG48) were grown in 2l drops containing a 1:1 (v/v) mixture of 7.5mg/ml RBD solution and 0.1M Bis-Tris pH 5.5, 0.5M ammonium sulfate, 19.5% PEG 3350, 1mM EDTA and 10% glycerol. Crystals of RBD (Wuhan-Hu-1)-Fab (NCV2SG53) were grown in 2l drops containing a 1:1 (v/v) mixture of 7.5mg/ml RBD solution and 0.1M MES pH 6.0, 0.25M ammonium sulfate and 22.5% PEG3350. Crystals of RBD (Delta)-Fab (NCV2SG53) were grown in 0.6l drops containing a 1:1 (v/v) mixture of 7.5mg/ml RBD solution and 25% PEG1500. The single crystals suitable for X-ray experiments were obtained in a few weeks. X-ray diffraction data collections were performed using synchrotron radiation at SPring-8 beamline BL44XU40 in a nitrogen vapor stream at 100K. The data sets were indexed and integrated using the XDS package41, scaled, and merged using the program Aimless42 in the CCP4 program package43. The scaling statistics were shown in Table1.

Phase determinations were carried out by the molecular replacement method using the program Phaser44 in the PHENIX package45 and the program Molrep46 in the CCP4 program package with the combination of RBD structure (PDB ID:7EAM) and Fab structures (PDB ID:7CHB and 7CHP) as search models. The structure refinement was performed using the program phenix.refine47 in the PHENIX package and the program coot48 in the CCP4 program package. The final refinement statistics were shown in Table1. Interactions between RBD and Fabs were analyzed using the program PISA49 in the CCP4 program package. All figures of structures were generated by the program pymol (The PyMOL Molecular Graphics System, Version 2.4.0., Schrdinger, LLC.). Class 2a/AZD8895 (PDB ID:7L7D), Class 3a/REGN10987 (PDB ID:6XDG), Class 3b/S309 (PDB ID:7JX3), Class 4a/CR3022 (PDB ID:6ZLR), Class 4b/S2X259 (PDB ID:7M7W), and Class 5/S2H97 (PDB ID:7M7W) open data were used in Fig.3a.

The statistical analysis was performed using Prism 9.0 (GraphPad, La Jolla, CA, USA). Ordinary One-way ANOVA, Two-way ANOVA, Kruskal-Wallis test, Wilcoxon rank test, and Friedman test were used to compare data. P-value 0.05 was considered statistically significant. Statistical tests are reported in figure legends and significance is reported at p0.05. To verify reproducibility, we repeated experiments more than two times as indicated in Figure legends. Detailed information on the sample is provided in Supplementary Tables.

Further information on research design is available in theNature Portfolio Reporting Summary linked to this article.

Go here to see the original:
Structural basis of spike RBM-specific human antibodies ... - Nature.com