An ancestral SARS-CoV-2 vaccine induces anti-Omicron variants antibodies by hypermutation – Nature.com

Study participants

Peripheral blood sampling was approved by the Institutional Ethics Review Board of Seoul National University Hospital (IRB approval number, 2102-032-1193). Informed consent was obtained from all participants of this study. The vaccinees had a median age of 30 years (range 2362) and showed a nearly equal distribution of males and females (46% and 54%, respectively). Demographic data for the 41 vaccinees are summarized in Supplementary Data5. Peripheral blood mononuclear cells (PBMCs) and plasma were separated using Lymphoprep (STEMCELL) Ficoll (Cytiva) following the manufacturers instructions. Total RNA was isolated using TRIzol Reagent (Invitrogen) according to the manufacturers instructions.

Plasma and phage ELISAs were performed in 96-well microtiter plates (Corning, 3690) coated with 100ng of recombinant SARS-CoV-2 proteins (Sino Biological, Ancestral RBD, 40592-V08H; Alpha RBD, 40592-V08H82; Beta RBD, 40592-V08H85; Gamma RBD, 40592-V08H86; Delta RBD, 40592-V08H90; Omicron BA.1 RBD, 40592-V08H121; Omicron BA.2 RBD, 40592-V08H123, Omicron BA.4/5 RBD, 40592-V08H130; Omicron BQ.1.1 RBD, 40592-V08H143; Omicron XBB.1.5, 40592-V08H146; Omicron XBB.1.16, 40592-V08H136; RBD Ancestral N, 40588-V08B) in coating buffer (0.1M sodium bicarbonate, pH 8.6) as described previously23. Briefly, the plates were coated with the antigen by incubation at 4C overnight and blocked with 3% bovine serum albumin (BSA) in PBS for 1h at 37C. Then, serially diluted plasma of 41 vaccinees and one COVID-19 patient23 (100-, 500-, 2,500-fold) or phage supernatant (twofold) in blocking buffer was added to the wells of microtiter plates, followed by incubation for 1h at 37C. Then, the plates were washed three times with 0.05% PBST. Horseradish peroxidase (HRP)-conjugated goat anti-human IgM and IgA (Invitrogen, A18835 and A18781, 1:5,000), rabbit anti-human IgG antibody (Invitrogen, 31423, 1:20,000) and HRP-conjugated anti-M13 antibody (Sino Biological, 11973-MM05T-H, 1:4,000) were used to determine the amount of bound antibody or M13 bacteriophage. A 3,3,5,5-tetramethylbenzidine liquid substrate solution (Thermo Fisher Scientific Inc.) was used as an HRP substrate.

In ELISA for the recombinant scFv-hFc-HA fusion proteins, the wells of microtiter plates (Corning) were first coated with 100ng of mouse anti-His antibody (Invitrogen, MA1-21315) and blocked. Then, the recombinant SARS-CoV-2 RBD protein with a polyhistidine tag (100nM) was added to the wells. After brief washing, the scFv-hFC-HA fusion proteins were serially diluted fourfold from 1 M to 0.24 pM in blocking buffer and added to wells of a microtiter plate. HRP-conjugated rat anti-HA antibody (Roche, 12013819001, 1:1,000) was used to determine the amount of bound antibody. All assays were performed in duplicate. The absorbance was measured at either 450 or 650nm using a microplate spectrophotometer (Thermo Fisher Scientific Inc., Multiskan GO), depending on the use of 2M sulfuric acid as the stop solution. All ELISA data were analyzed using GraphPad Prism software v6 (GraphPad Software).

Genes encoding the variable domain of the heavy chain (VH) and part of the first constant domain of the heavy chain (CH1) domain (VH-CH1) were amplified using specific primers, as described previously23. All primers used are listed in Supplementary Data6. These six IGHV-specific primers are expected to amplify a total of 269 out of 270 IGHV alleles (99.63%) available in the IMGT database. IGHV4-39*08 is expected to encounter difficulty in amplification, due to a mismatch with the germline sequence at the third nucleotide from the 3 end of the primer32 (Supplementary Data7). Briefly, total RNA was used as a template to synthesize complementary DNA (cDNA) using the SuperScript IV First-Strand Synthesis System (Invitrogen) with specific primers (primer No. 18) targeting the CH1 domain of each isotype (IgM, IgD, IgG, IgA, and IgE) according to the manufacturers protocol. After cDNA synthesis, 1.8 volumes of SPRI beads (Beckman Coulter, AMPure XP) were used to purify cDNA, which was eluted in 35l of water. The purified cDNA (15l) was subjected to second-strand synthesis in a 25l reaction volume using IGHV genespecific primers (primer No.914) and a KAPA Biosystems kit (Roche, KAPA HiFi HotStart). The PCR conditions were as follows: 95C for 3min, 98C for 30s, 60C for 45s, and 72C for 6min. After second-strand synthesis, dsDNA was purified using 1 volume of SPRI beads, as described above. VH-CH1 genes were amplified using purified dsDNA (15l) in a 25l reaction volume using primers containing indexing sequences (primers 15 and 16) and a KAPA Biosystems kit. The PCR conditions were as follows: 95C for 3min; 25 cycles of 98C for 30s, 60C for 30s, and 72C for 1min; and 72C for 5min. PCR products were subjected to electrophoresis on a 1.5% agarose gel and purified using a QIAquick gel extraction kit (QIAGEN Inc.) according to the manufacturers instructions. The gel-purified PCR products were purified again using 1 volume of SPRI beads and eluted in 20l water. The SPRI-purified sequencing libraries were quantified with a 4200 TapeStation System (Agilent Technologies) using a D1000 ScreenTape assay and subjected to next-generation sequencing on the Illumina NovaSeq 6000 250PE (SP Chip) platform using Novaseq 6000 reagent kits (Illumina Inc.). The total read counts of the chronological repertoires after NGS data processing is summarized in Supplementary Data8.

Raw sequencing reads obtained from sequencing the VH-CH1 region of B cells in peripheral blood were processed using a custom pipeline. The pipeline included adapter trimming and quality filtering; unique molecular identifier (UMI) processing; V(D)J gene annotation; clustering; quality control; and diversity analysis.

The forward reads (R1) and reverse reads (R2) of the raw NGS data were merged using paired-end read merger (PEAR) v0.9.10 with the default settings33. The merged reads were q-filtered under the q20p95 condition, resulting in 95% of base pairs in the reads having a Phred score greater than 20. Primer positions were identified in the quality-filtered reads, and primer regions were trimmed to remove the effects of primer synthesis errors while allowing one substitution or deletion. We identified 187 out of 270 IGHV alleles (69.3%) from the IMGT database across six time points in 41 vaccinees (246 BCR HC libraries) of Korean nationality. This ratio highly resembles the average allele distribution observed among Eastern Asian populations (62.7%)34 (Supplementary Data9).

Based on the primer recognition results, UMI sequences were extracted, and reads were clustered according to the UMI sequences. To eliminate index misassignment, we subclustered the clustered reads based on the similarity of the reads (allowing 5 mismatches in each subcluster) and matched the majority subcluster to the UMI. The subclustered reads were aligned using the multiple sequence alignment tool Clustal Omega v1.2.4 with the default settings35,36. Consensus calling was performed by selecting major frequency bases at every position of the aligned sequences. The number of reads in the consensus sequence was redefined as the number of UMI subclusters belonging to the consensus sequence.

Sequence annotation consisted of isotype annotation and V(D)J annotation. The consensus sequence was divided into a V(D)J region and a constant region. The isotype of the consensus sequence was annotated by aligning the extracted constant region with the constant gene of the International Immunogenetics Information System (IMGT)37. Then, the V(D)J region of the consensus sequence was annotated using an updated version of IgBLAST (v1.17.1)38. Among the annotation results, the IGHV genes, IGHJ genes, HCDR3 sequences, and number of SHMs were extracted for further analysis. The number of SHMs cannot be determined for sequences outside the primer binding sites; therefore, SHMs were calculated by comparing the amplified regions from primer-mediated amplification with the IGHV gene germline sequence. The nonfunctional consensus reads were defined and filtered using the following criteria: (i) sequence length shorter than 250 base pairs from each R1 and R2, (ii) presence of a stop codon or frameshift in the entire amino acid sequence, (iii) failure to annotate one or more HCDR1, HCDR2 and HCDR3 regions, and (iv) failure to annotate the isotype.

Six human scFv phage display libraries were constructed using the total RNA prepared from the sixth blood sample for vaccine Nos. 22, 27, 32, 35, 39 and 43 as described previously23. Briefly, for the VH and V/V genes, total RNA was employed to synthesize cDNA using the SuperScript IV First-Strand Synthesis System (Invitrogen) with gene-specific primers targeting the JH and C/ genes (primer No.1722), respectively. After cDNA synthesis, 1.8 volumes of SPRI beads (Beckman Coulter) were used to purify cDNA, which was eluted in 20l of water. The purified cDNA (11.25l) of the VH gene was subjected to second-strand synthesis in a 25l reaction volume using IGHV gene-specific primers (primer No. 2332, 7.5l) and a KAPA Biosystems kit (Roche). The reaction conditions were as follows: 95C for 3min, 98C for 1min, 60C for 1min, and 72C for 5min. In the case of the V/V gene, the eluted cDNA (17.25l) was used for the first round of PCR synthesis in a 25l reaction volume using V/V and J /J gene-specific primers (primer No. 3369, 0.75l). The PCR conditions were as follows: 95C for 3min; 4 cycles of 98C for 1min, 60C for 1min, and 72C for 1min; and 72C for 10min. After second-strand synthesis for the VH gene or PCR amplification of the V/V gene, double-stranded DNA (dsDNA) was purified with 1 volume of SPRI beads and eluted in 40l of water. VH and V/V genes were amplified using 10l of purified dsDNA, 2.5 pmol of the primers (primer No. 7073), and KAPA Biosystems kit components in a 50l total reaction volume with the following thermal cycling program: 95C for 3min; 30 cycles of 98C for 30s, 60C for 30s, and 72C for 1min; and 72C for 10min. Then, the amplified VH and V/V genes were subjected to electrophoresis on a 1.5% agarose gel and purified using a QIAquick gel extraction kit (QIAGEN Inc.) according to the manufacturers instructions. The purified VH and V/V gene fragments (100ng) were mixed and subjected to overlap extension PCR to generate scFv genes using 2.5 pmol of the overlap extension primers (primer No. 74 and 75) using the KAPA Biosystems kit. The PCR conditions were as follows: 95C for 3min; 25 cycles of 98C for 20s, 65C for 15s, and 72C for 1min; and 72C for 10min. The amplified scFv gene was purified and cloned into a phagemid vector39.

Human scFv phage display libraries with 7.1108, 6.1108, 8.1108, 8.3108, 7.9108, and 7.4108 colony-forming units were generated using cDNA prepared from vaccinee Nos. 22, 27, 32, 35, 39 and 43, respectively. The libraries were subjected to five rounds of biopanning against the recombinant SARS-CoV-2 BA.1 RBD protein (Sino Biological Inc.) as described previously40. Immune tubes (SPL, 43015) coated with 17g of the BA.1 RBD were used for the first round, and 5106 magnetic beads (Invitrogen, Dynabeads M-270 epoxy) conjugated with 1.4g of the BA.1 RBD protein were used for the other rounds. After each round of biopanning, the bound phages were eluted and amplified for the next round of biopanning. For the selection of BA.1 RBD-reactive scFv phage clones, individual phage clones were amplified from the titration plate of the last round and subjected to phage ELISA41. The genes encoding BA.1 RBD-reactive scFv clones were identified using phagemid DNA prepared from phage clones and Sanger nucleotide sequencing40. A recombinant scFv protein fused with human IgG1 FC and the HA peptide (scFv-hFc-HA) was expressed using a mammalian expression system with Expi293F cells (Gibco, A14527) and purified as described previously41.

The gene fragment encoding the VH region with a randomized HCDR3 sequence and another gene fragment encoding the rest of the scFv were amplified using phagemid DNA of clone 2760, primers for randomization (Primer No. 7679) and the KAPA Biosystems kit. The PCR conditions were as follows: 95C for 3min; 25 cycles of 98C for 20s, 65C for 15s, and 72C for 1min; and 72C for 10min. Then, the amplified genes were subjected to electrophoresis on a 1.5% agarose gel and purified using a QIAquick gel extraction kit (QIAGEN Inc.) according to the manufacturers instructions. The purified gene fragments (200ng) were mixed and subjected to overlap extension PCR to generate scFv genes using 2.5 pmol of the overlap extension primers (primer No.76 and 79) and the KAPA Biosystems kit. The PCR conditions were as follows: 95C for 3min; 25 cycles of 98C for 20s, 65C for 15s, and 72C for 1min; and 72C for 5min. The amplified scFv genes were purified and cloned into a phagemid vector38. For phage ELISA, individual phage clones were amplified from the titration plate and subjected to phage ELISA42.

The ancestral SARS-CoV-2 (CoV/Korea/KCDC03/2020 NCCP43326), Alpha B.1.1.7 (hCoV-19/Korea/KDCA51463/2021 (NCCP 43381), Beta B.1.351 (hCoV-19/Korea/KDCA55905/2021 (NCCP 43382), Gamma P.1 (hCoV-19/Korea/KDCA95637/2021 (NCCP 43388), Delta B.1.617.2 (hCoV-19/Korea/KDCA119861/2021 (NCCP 43390), and Omicron B.1.1.529 (hCoV-19/Korea/KDCA447321/2021 NCCP43408) viruses were obtained from the Korea Disease Control and Prevention Agency. The viruses were propagated in Vero cells (ATCC, CCL-81) in Dulbeccos Modified Eagles Medium (DMEM, Welgene) in the presence of 2% fetal bovine serum (Gibco, Thermo Fisher Scientific Inc.)23,43. Neutralization assays were performed as described previously23,44. Briefly, Vero cells were seeded in 96-well plates (1.5104 or 0.5104 cells/well) in Opti-PRO SFM (Thermo Fisher Scientific Inc.) supplemented with 4 mM L-glutamine and 1 Antibiotics-Antimycotic (Thermo Fisher Scientific Inc/) and grown for 24h at 37C in a 5% CO2 environment. Recombinant scFv-hFc proteins were diluted from 100 to 0.1953 g/ml (twofold) in phosphate-buffered saline (PBS, Welgene) and mixed with 100 or 500 TCID50 of SARS-CoV-2. Then, the mixture was incubated for 30min at 37C and added to the cells in tetrads, followed by incubation for 4 or 6 days at 37C in a 5% CO2 environment. The cytopathic effect (CPE) in each well was visualized following crystal violet staining 4 or 6 days post infection, and measured using the EVOS Digital Inverted Imaging System with 40X lens (AMG). The IC50 values were calculated using the doseresponse inhibition equation of GraphPad Prism v6 (GraphPad Software).

The most frequent BCR HC sequences of the 2760, 35-15, 3546, 43-09, and 43-34 clonotypes in the third BCR repertoires were mapped to the chronological BCR repertoires following the same definition of BCR HC clonotypes. All the mapped nucleotide sequences were aligned using the multiple sequence alignment tool Clustal Omega v1.2.4 with the default settings35,36 and processed using Ugene software v1.16.2. The aligned mapped nucleotide sequences were interpreted using a phylogenetic tree generated by IgPhyML v1.1.3 052020 using the HLP model option45. The phylogenetic trees were plotted using Interactive Tree of Life (iTOL) online tool v646.

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

More here:

An ancestral SARS-CoV-2 vaccine induces anti-Omicron variants antibodies by hypermutation - Nature.com