key: cord-0335922-znz3zfky authors: Brewer, R. C.; Ramadoss, N. S.; Lahey, L. J.; Robinson, W. H.; Lanz, T. V. title: BNT162b2 Vaccine Induces Divergent B cell responses to SARS-CoV-2 S1 and S2 date: 2021-07-22 journal: nan DOI: 10.1101/2021.07.20.21260822 sha: 1fda4d708ab2bf9dbafd42d518aaf266aa4da61f doc_id: 335922 cord_uid: znz3zfky The first ever messenger RNA (mRNA) vaccines received emergency approvals in December 2020 and are highly protective against SARS-CoV-2. However, the contribution of each dose to the generation of antibodies against SARS-CoV-2 spike (S) protein and the degree of protection against novel variants, including delta, warrant further study. Here, we investigated the B cell response to the BNT162b2 vaccine by integrating repertoire analysis with single-cell transcriptomics of B cells from serial blood collections pre- and post-vaccination. The first vaccine dose elicits highly mutated IgA+ plasmablasts against the S protein subunit S2 at day 7, suggestive of recall of a memory B cell response generated by prior infections with heterologous coronaviruses. On day 21, we observed minimally-mutated IgG+ activated switched memory B cells targeting the receptor binding domain (RBD) of the S protein, likely representing a primary response derived from naive B cells. The B cell response against RBD is specifically boosted by the second vaccine dose, and encodes antibodies that potently neutralize SARS-CoV-2 pseudovirus and partially neutralize novel variants, including delta. These results demonstrate that the first vaccine dose activates a non-neutralizing recall response predominantly targeting S2, while the second vaccine dose is vital to boosting neutralizing anti-S1 RBD B cell responses. The early S2 PB response is echoed by the early development of anti-S2 plasma 164 antibody titers, which start leveling off at D21. Strikingly, RBD and S1 IgG and IgA titers 165 remain low until day 28, one week after the second vaccination (Fig. 3i, Extended Data Fig. 7a-166 d). Accordingly, plasma neutralization of SARS-CoV-2 Wuhan-Hu-1 pseudovirus is 167 significantly boosted after the second vaccination (Fig. 3j) . As primarily anti-RBD antibodies 168 have the potential to block viral entry into the cell, this differentiation is critical and highlights 169 the importance of the second injection for a protective anti-RBD response. Notably, it is 170 overlooked in studies that focus solely on the anti-S antibody response. response to a novel antigen that has little structural overlap with previously encountered 184 pathogens 20 . In our study, D7 B cells specific for any of the three antigens showed mutation 185 frequencies similar to the antigen-negative sorted B cells, suggesting that this initial response 186 stems directly from the memory B cell pool (Fig. 3k) . At this early timepoint, S2-specific B cells 187 predominate over RBD and S1n-specific B cells, providing further evidence that the rapid 188 memory recall response is more effective for S2 than S1 (Fig. 3f, k) . Consistent with this 189 observation, S2 is more conserved among human pathogenic coronaviruses than S1 21, 22 . 190 Strikingly, at timepoints D21 and D28, with the influx of S1-specific B cells, mutation 191 frequencies decrease (Fig. 3k) . Together, these results indicate new recruitment from the naïve B 192 cell pool and short GC maturation in response to the vaccine. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2021. ; https://doi.org/10.1101/2021.07.20.21260822 doi: medRxiv preprint 194 To test if these newly-recruited, minimally mutated anti-S antibodies are better at binding 195 the S subunits than the early, highly mutated anti-S antibodies recruited from the memory B cell 196 pool, we expressed 50 representative BCR sequences as recombinant monoclonal antibodies 197 (rmAbs) -14 from S2-specific, 30 from S1/RBD-specific, and 6 from S1n-specific B cells 198 (Extended Data Table 2 ). Of those selected antibodies, 8 rmAbs bound S2, 15 rmAbs bound 199 RBD, and 3 rmAbs bound S1 but not RBD, as measured by ELISA and bio-layer interferometry 200 ( Fig. 4a and Extended Data Fig. 9 ). Interestingly, when we compared low and high binding anti-201 S1 rmAbs (encompassing RBD and S1n) we found that high binding rmAbs show significantly 202 lower mutation frequencies than low binding anti-S1 rmABs (Fig. 4b) . In contrast, mutation 203 frequencies of rmAbs against S2 did not differ significantly between low and high binders ( Fig. 204 4c). Additionally, the S2 binders with the highest affinities were derived from D7 PB, while 205 highest affinity antibodies against S1 and RBD stemmed from D21 and D28 (Fig. 4a) . S2 is the 206 more highly conserved subunit of S 22, 23 To determine if the rmAbs were cross-reactive to other 207 beta-coronaviruses, we tested the highest binders to RBD and S2 against the spike protein of four 208 pathogenic beta-coronaviruses -HCoV-229E, HCoV-HKU1, HCoV-NL63, HCoV-OC43. Of The emergence of novel SARS-CoV-2 variants poses additional challenges to the management 219 of the pandemic, and could jeopardize vaccine efficacy and the prospects of an expeditious 220 return to normalcy. RBD mutations contribute significantly to immune escape. To evaluate the 221 effects of variants on antibody neutralization, we selected ten high-affinity Wuhan-Hu-1-222 neutralizing anti-RBD rmAbs from three individuals (P1, P2, P6) with different levels of plasma 223 neutralization against delta (Fig. 4d-f ). Five different SARS-CoV-2 variants of concern 224 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2021 CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2021. were included (Extended Data Table 3 Our study provides a detailed characterization of the B cell response to the BNT162b2 258 mRNA vaccine on a single-cell level. Parsing the anti-S1 and S2 responses provides important 259 insights into why the second vaccine dose is vital for protection. Our results demonstrate that the 260 first vaccine dose activates a non-neutralizing recall response predominantly targeting epitopes 261 in the S2 protein subunit, which is conserved across human-pathogenic coronaviruses 22, 23 , while 262 the second vaccine dose is vital to boosting neutralizing B cell responses to S1 and RBD. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. In contrast, the antigen-specific PB and ASM response is likely derived from GC-272 dependent processes. The first vaccination induces an IgA-dominant PB response against S2 273 with high mutation frequencies, which is cross-reactive to the human-pathogenic beta-274 coronaviruses OC43 and HKU1. Our data are consistent with a recall response of mucosal 275 memory B cells that matured during prior pulmonary coronavirus infections. After this initial response, we observed an influx of minimally mutated ASM on D21 and 277 D28, which target S1 and RBD. We show that low mutation frequency corresponds to high 278 affinity against RBD. mRNA vaccines have been shown to induce robust and prolonged GC 279 reactions, with PB and SM persisting in GC for over three months 29,30 . Despite their low 280 mutation frequency, the delayed development and the switched phenotype of S1/RBD-specific 281 PB and ASM indicate that they underwent GC maturation. High BCR affinity and a naïve 282 phenotype foster preferential recruitment into GC 31-33 and high affinity also promotes release 283 from the GC as PB, plasma cells, or memory B cells 34 . High affinity of minimally mutated BCRs 284 could therefore limit GC maturation to a relatively short time frame. 285 We utilized plasma samples and our low mutation RBD-binding rmAbs to characterize 286 neutralization against several SARS-CoV-2 variants of concern. While we found a significant 287 degree of immune escape, we also identified antibodies with neutralization potency against each 288 variant. Importantly, we observed potent neutralization activity against the highly infectious 289 delta variant, which has quickly become the dominant strain world-wide. Delta-neutralizing B 290 cells were even found in an individual with very limited plasma neutralization potency, fueling 291 hope that even individuals with low neutralizing titers can raise a recall memory response upon 292 infection with delta. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 22, 2021. ; https://doi.org/10.1101/2021.07.20.21260822 doi: medRxiv preprint Together, our study provides a detailed characterization of the blood B cell response to 294 the BNT162b2 mRNA vaccine. Our data emphasize the importance of the second vaccine dose 295 in inducing generation of anti-S1 RBD antibodies that contribute to neutralization of SARS-296 CoV-2 variants, including delta. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 22, 2021. ; https://doi.org/10.1101/2021.07.20.21260822 doi: medRxiv preprint variants. ****P <0.0001 according to unpaired two-tailed t-test. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (Fig 2a) . Cells corresponding to the clonal expansions shown in (Fig. 2d) . Number of cells in each clonal 369 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (Fig. 3d) . c,e, n=9 396 individual values, means, and standard deviations are shown. *P <0.05, **P < 0.01, ***P < 0.001 397 according to two-tailed one-way ANOVA test. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 22, 2021. ; https://doi.org/10.1101/2021.07.20.21260822 doi: medRxiv preprint . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 22, 2021. Plasma reactivities to RBD for IgG (left) and IgA (right), on D28 compared to day 120 (D120), 420 two weeks after COVID-19 diagnosis. P1 is highlighted in red, the other eight individuals are 421 grey. *P <0.05, **P < 0.01, ***P < 0.001 according to a-c, two-tailed one-way ANOVA test and 422 d, paired t-test. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 22, 2021. ; https://doi.org/10.1101/2021.07.20.21260822 doi: medRxiv preprint . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 22, 2021. Wuhan-Hu-1, Alpha+E484K, Beta, Gamma, Epsilon, and Delta. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 22, 2021 TotalSeq-C hashtag 1-9 antibodies (Biolegend) for demultiplexing individual samples in cells) were removed and the data was normalized, integrated, and clustered with only the B cells. Specific B cell clusters were identified using canonical B cell markers 11 (Extended Data Fig. 1 ). Identification of differentially expressed genes (DEG) and functional enrichment 503 We performed differential gene expression testing using the FindMarkers function in Seurat with is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2021. ; https://doi.org/10.1101/2021.07.20.21260822 doi: medRxiv preprint Biological, 40604-V08B) in carbonate-bicarbonate buffer at 4˚C overnight. Plates were washed 6 530 times with PBST (PBS + 0.1% Tween20) after each step. The plates were blocked with blocking plasmids (HDM-Hgpm2, HDM-tat1b, pRC-CMV-Rev1b), and wildtype or variant SARS-CoV-2 559 spike plasmids (parent plasmids publicly available from Jesse Bloom lab). After 48 to 60 hours, 560 viral supernatants were collected and spun at 1000xg for 10m to remove cell debris. The 561 lentiviral supernatants were concentrated using LentiX concentrator (Takara) according to the 562 manufacturer's instructions. The lentiviral pellets were resuspended at 20-fold viral increase in 563 EMEM media and stored at -80°C. Virus was titrated on HeLa-ACE2 cells. Acknowledgements: 576 We thank all study participants who devoted time to our research. We thank Dennis Burton for 577 providing the ACE2-HeLa cells. We thank Matthew Baker for key discussions. We thank 578 Shaghayegh Jahanbani for antibody production. This work was supported by T32 AI007290-35 The raw sequencing data will be uploaded in the GEO database before final publication. SARS-CoV-2 mRNA vaccination induces functionally diverse antibodies 604 to NTD, RBD, and S2 mRNA vaccination compared to infection elicits an IgG-predominant 606 response with greater SARS-CoV-2 specificity and similar decrease in variant spike Mapping Neutralizing and Immunodominant Sites on the SARS-CoV-2 Reduced sensitivity of SARS-CoV-2 variant Delta to antibody 621 neutralization Single-Cell Sequencing of Plasma Cells from COVID-19 Patients 623 Reveals Highly Expanded Clonal Lineages Produce Specific and Neutralizing Antibodies to 624 SARS-CoV-2 Dimensionality reduction for visualizing single-cell data using UMAP Challenges and Opportunities for Consistent Classification of Human B Cell 628 and Plasma Cell Populations CD83 Modulates B Cell Activation and Germinal Center Responses Human B cells induce dendritic cell maturation and favour Th2 632 polarization by inducing OX-40 ligand What is and what should always have been: long-lived plasma 634 cells induced by T cell-independent antigens Robust memory responses against influenza vaccination in pemphigus 636 patients previously treated with rituximab Origin and Function of Circulating Plasmablasts during Acute Viral Infections virus infection and vaccination in humans Epitope-resolved profiling of the SARS-CoV-2 antibody response 649 identifies cross-reactivity with endemic human coronaviruses The receptor binding domain of the viral spike protein is an 652 immunodominant and highly specific target of antibodies in SARS-CoV-2 patients Epitope-resolved profiling of the SARS-CoV-2 antibody response 655 identifies cross-reactivity with an endemic human CoV Longitudinal Isolation of Potent Near-Germline SARS-CoV-2-Neutralizing 658 Antibodies from COVID-19 Patients Evolution of antibody immunity to SARS-CoV-2 SARS-CoV-2 501Y.V2 variants lack higher infectivity but do have immune 662 escape The Impact of Mutations in SARS-CoV-2 Spike on Viral Infectivity and 664 SARS-CoV-2 mRNA Vaccines Foster Potent Antigen-Specific Germinal 668 influenza vaccination Metascape provides a biologist-oriented resource for the analysis of systems-682 level datasets Integrated analysis of multimodal single-cell data IMGT(®) tools for the nucleotide 686 analysis of immunoglobulin (IG) and T cell receptor (TR) V-(D)-J repertoires Isolation of potent SARS-CoV-2 neutralizing antibodies and protection 690 from disease in a small animal model Protocol and Reagents for Pseudotyping Lentiviral Particles with 692 SARS-CoV-2 Spike Protein for Neutralization Assays International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity