Determinants ofde novoB cell responses to drifted epitopes in post-vaccination SARS-CoV-2 infections

60 Within a year of the discovery of SARS-CoV -2 as the etiological agent of COVID-61 191, highly effective vaccines were developed and administered. Leading this class 62 were the monovalent mRNA vaccines BNT162b2 and mRNA-1273 encoding the 63 ancestral Spike protein, both of which achieved ~95% efficacies in preventing 64 symptomatic illness2,3. Other vaccine platforms also achieved high efficacies, especially 65 against severe illness and hospitalization4–8. Since the initial results of these clinical 66 trials, however, the protective capacity of these vaccines has declined9–12. This drop in 67 vaccine effectiveness is due to both waning of antibodies and viral evolution and escape 68 from vaccine-induced neutralizing antibodies, which are the best-known correlates of 69 protection13,14. While the known genetic diversity of SARS-CoV-2 was quite modest 70 through most of 202015, new variants with enhanced transmissibility and/or neutralizing 71 antibody escape mutations have since emerged and sequentially swept to global 72 dominance12,16–24. As of this writing, the dominant circulating variant is Omicron, which 73 comprises sublineages that contain Spike protein mutations located within most known 74 neutralizing antibody epitopes25. A key issue that will define both protection against 75 infections and the strategy underlying updates to the vaccines is the extent to which 76 pre-existing vaccine-induced immunity protects against heterologous challenges like 77 Omicron. 78 B cell responses following mRNA COVID-19 vaccination are characterized by 79 exceptionally long-lived germinal center reactions that persist for months while 80 continuously improving the breadth and affinity of antibodies26–29. Cells exiting the 81 germinal center carry affinity-enhancing mutations and can become long-lived antibody-82 5 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint secreting plasma cells or memory B cells30. Depending on the subset of memory B cell, 83 re-exposures to antigen trigger differentiation to new plasma cells or germinal center 84 reactions31–34. After antigens from infection or vaccine antigens have been cleared, 85 long-lived plasma cells and memory B cells persist to maintain humoral immunity. 86 While these features protect against homologous SARS-CoV-2 infections, it is 87 more difficult to predict the nature of responses to subsequent heterologous infections 88 or vaccines. Due to their expanded pre-existing numbers and intrinsic signaling and 89 transcriptional differences relative to naive B cells, memory B cells rapidly mount 90 responses that are of greater magnitude than those of naïve primary responses35–40 to 91 either initial infection or vaccination. Because of these properties, memory B cells that 92 react to epitopes conserved between the original and secondary challenges could 93 dominate the response to heterologous infections or vaccines41–43. If antigen and T cell 94 help are limiting, memory B cells might then outcompete naive B cells and new primary 95 antibody responses aimed at the new variant-specific epitopes. This phenomenon, 96 known as antigenic imprinting or “original antigenic sin”44, can be beneficial if antibodies 97 against the conserved epitopes are protective. However, recall responses to 98 heterologous pathogens can potentially be neutral or even detrimental if antibodies 99 targeting these conserved epitopes are not protective and variant-specific primary 100 responses are competitively inhibited45. As an example of the phenomenon, pre-existing 101 common coronavirus-specific memory B cells compose a majority of the early response 102 to SARS-CoV-2, but primary responses to epitopes unique to SARS-CoV-2 are readily 103 observed later26,46,47. Whether common coronavirus immunity is helpful, harmful, or 104 neutral for de novo responses to SARS-CoV-2 is unknown. 105 6 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint In influenza infections, antigenic imprinting has been proposed to explain the 106 age-associated differential in morbidity and mortality based on influenza subtype 107 exposure history48–51. The various hemagglutinin (HA) subtypes of influenza A virus fall 108 into one or the other of two phylogenetically distinct HA “groups” (group 1 or group 2). 109 Individuals have the highest antibody titers against influenza strains encountered early 110 in life, and they experience enhanced protection against influenza strains that are within 111 the same HA group as their primary infection strain compared to heterosubtypic 112 infections from the group that is mismatched to their first childhood infection. Previous 113 work has shown that childhood exposure to H1N1 (group 1 hemagglutinin (HA)) affords 114 protection against other group 1 HAs, such as H5N1. The same is true for individuals 115 with group 2 HAs, whereby childhood H3N2 infection affords protection against H7N9. 116 Conversely, individuals with group 1 imprinting experience an increase in mortality when 117 faced with a group 2 influenza virus infection, such as that observed for H7N9 118 infections48,51. 119 Though pre-existing immunity can certainly impact primary responses to 120 heterologous antigens, other mechanisms can also limit antibody responses to drifted 121 epitopes. Epitopes that were previously immunodominant for antibody responses do not 122 necessarily remain so once mutated, irrespective of prior immunity52. There are several 123 possible mechanistic reasons why not all epitopes are equal for antibody responses. 124 Factors such as naïve antigen-specific B cell precursor frequency and avidity vary 125 greatly across epitopes, which in turn correlate with their relative contribution to the 126 subsequent response53–57. Some epitopes can also be biophysically challenging for 127 antibody binding, such as those sterically blocked by glycan shields or appearing as 128 7 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint non-complex ‘smooth’ surfaces to B cells58,59. Further, epitopes that mimic self-antigens 129 also elicit poor responses due to tolerance mechanisms that remove or hamper B cells 130 from the repertoire that could otherwise respond60–63. Finally, V gene usage during 131 V(D)J recombination that gives rise to B cell receptors is uneven, as some segments 132 are more heavily utilized than others64,65. In turn, this can create ‘holes’ in the repertoire, 133 rendering some epitopes poorly immunogenic66. As SARS-CoV-2 variants of concern 134 accumulate mutations in antigenic regions, immunodominance might change in ways 135 that limit responses to drifted epitopes, with or without prior immunity. Thus, it has 136 remained difficult to examine the degree to which infrequent de novo variant-specific 137 responses in post-vaccination infections and heterologous boosters are due to changes 138 in immunodominance, antigenic imprinting, or some combination of both67–75. 139 Antigenic imprinting has remained nearly impossible to quantify directly and 140 instead has predominantly relied on historical epidemiological data to make inferences 141 about biological mechanisms that produce the documented patterns48,76,77. The COVID-142 19 pandemic presents a unique opportunity to address these questions: it has 143 encompassed adults with known infection histories and monovalent vaccines that 144 induce well characterized B cell responses78–81 and the emergence of antigenically 145 distinct viral variants25,82. Yet, as immunological histories become more complex and 146 with very few immunologically naïve adults remaining83,84, the Omicron BA.1 (BA.1, for 147 short) wave likely represented the final opportunity to recruit robust cohorts of 148 individuals that meet the key experimental and control criteria. Through voluntary saline-149 gargle PCR testing of University of Arizona students, staff, and faculty as part of COVID 150 mitigation efforts on campus from August 2020 to July 2023; serological testing at 17 151 8 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint University of Arizona-managed sites across the state of Arizona; and two CDC-funded 152 cohorts of essential workers, Arizona Healthcare, Emergency Response, and Other 153 Essential Workers Surveillance (AZ HEROES)85 and Research on Epidemiology of 154 SARS-CoV-2 in Essential Response Personnel (RECOVER)86, we recruited 155 unvaccinated individuals who had recovered from primary Delta (B.1.617.2 or 156 B.1.617.2-like) or BA.1 (B.1.1.529 or B.1.1.529-like) infections. These cohorts allowed157 us to characterize the immunodominance hierarchies for both Delta and BA.1 variants 158 and directly compare the specificity of antibody responses in unvaccinated individuals 159 infected by the antigenically drifted viral variants to those generated by post-vaccination 160 infection by Delta or BA.1. In doing so, we were able to quantify the impact of antigenic 161 imprinting on de novo responses to drifted epitopes. 162 163 9 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint

164 From our voluntary on-campus testing program at the University of Arizona, we 165 recruited 37 participants who had tested positive for SARS-CoV-2 infections between 166 July 1, 2021 and December 1, 2021 despite completion of the primary vaccine series of 167 monovalent BNT162b2 or mRNA-1273 prior to infection (described in detail in Methods 168 section). We also recruited 12 individuals who tested positive during this period but had 169 not received any COVID-19 vaccines. Symptoms reported by participants following 170 infections were similar between primary and post-vaccination infections; none required 171 hospitalization. A slightly larger portion of post-vaccination infections were 172 asymptomatic relative to primary infections (Figure S1A), and in general, the duration of 173 symptoms was significantly shorter in this group relative to those who were 174 unvaccinated at the time of infection (Figure S1B). All recruited individuals who tested 175 positive by RT-qPCR and had sufficient sequence coverage to assign a lineage had 176 sequences confirmed to be Delta (Figure S2A). During this period, the Delta variant 177 represented 100% of PCR+ samples on campus that could be assigned a PANGO-178 lineage87, as determined through viral sequencing of all remnant samples below a Ct 179 value of 35 (Figure S2B). We also selected 71 serum samples as part of our statewide 180 antibody testing initiative88 from vaccinated participants who had no self-reported prior 181 SARS-CoV-2 infections. This cohort was chosen based on matching for age, sex, and 182 time post-vaccination with our post-vaccination infection group. Characteristics of the 183 cohorts are listed in Table 1. 184 Participants provided blood samples at an average of 75 days (IQR for primary 185 and post-vaccination infections = 45.8 days, 97.3 days; Table 1) after testing positive for 186 10 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint SARS-CoV-2 infections and at an average of nine months (IQR for vaccinated only and 187 post-vaccination infections = 56 days, 317 days; Table 1) after their last vaccine dose. 188 Using plasma from these samples, we first performed live virus neutralization assays on 189 both an early-pandemic virus representative, (WA-1, from January 2020) or on the 190 antigenically drifted Delta variant. Against both WA-1 and Delta, post-vaccination Delta 191 infections led to significantly higher titers of neutralizing antibodies than both primary 192 infections and vaccinated only controls (Figure 1A), indicating a robust recall response. 193 Elevated neutralizing antibody titers in post-vaccination Delta infections could 194 arise from both memory B cell responses to conserved neutralizing epitopes and 195 primary responses against new variant-specific epitopes. To begin to determine the 196 relative specificities of antibodies following Delta infections, we performed ELISAs to 197 measure the magnitude of the antibody response against Wuhan/Hu1/2019 (hereafter 198 WuHu1) and Delta Spike antigens. WuHu1 was sampled in December 2019 and is the 199 SARS-CoV-2 reference sequence; its Spike amino acid sequence is identical to that of 200 WA-1. We first measured antibodies that bound the receptor binding domain (RBD), as 201 most neutralizing antibodies target this region89,90. Post-vaccination Delta infections led 202 to elevated RBD-binding antibody titers, both against WuHu1 and Delta, relative to 203 vaccination only and primary Delta infection controls (Figure 1B), again confirming a 204 robust recall response. As expected, vaccination-only controls showed slightly elevated 205 titers of WuHu1 RBD-binding antibodies relative to Delta RBD antibodies (Figure 1B, 206 right panel). Reciprocally, primary Delta infections led to a skewing towards Delta 207 RBD-binding antibodies (Figure 1B, right panel). Post-vaccination-Delta infections led 208 to an even ratio of WuHu1:Delta RBD-binding antibodies (Figure 1B, right panel), 209 11 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint similar to prior studies91. Aside from the RBD, neutralizing antibodies can also bind 210 other regions of the S1 domain of Spike92–94. As with RBD, post-vaccination Delta 211 infections led to an even ratio of antibodies that bound WuHu1 and Delta S1 relative to 212 vaccination alone or primary Delta infections (Figure S3). 213 To more directly assess antibody specificities with single cell resolution in post-214 vaccination Delta infections, memory B cells using WuHu1 S1 and Delta S1 antigen 215 tetramers were quantified by flow cytometry. We focused our analysis on the isotype-216 switched CD27+ subset (Figure 2A and Figure S4), since few Spike-specific cells are 217 observed in other memory B cell subsets95. Memory B cells that bound Delta S1 only 218 were observed in both primary infections and in post-vaccination Delta infections, 219 suggesting that in both cases, de novo responses aimed at variant-unique epitopes 220 were mounted (Figure 2A-B). However, the proportions of these cells in PBMCs were 221 slightly reduced in post-vaccination Delta infections relative to primary Delta infections 222 (Figure 2B). Reciprocally, cross-reactive memory B cells that bound both WuHu1 S1 223 and Delta S1 were elevated in post-vaccination Delta infections relative to primary Delta 224 infections (Figure 2B), consistent with a robust recall response and antigenic imprinting, 225 though for a subset of individuals this appears to be more modest. In both primary and 226 post-vaccination Delta infections, memory B cells that bound Delta S1 uniquely were 227 rare relative to cross-reactive cells that bound both WuHu1 and Delta S1 (Figures 2A-228 B). Although these data suggest that pre-existing immunity limits new primary 229 responses, cross-reactive and Delta-specific memory B cells were positively correlated 230 in post-vaccination Delta infections (Figure 2C), arguing against a mechanism of 231 competitive inhibition between these two cellular compartments. 232 12 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint The RBD of Delta contains two non-synonymous point mutations that deviate 233 from the vaccine sequence: T478K and L452R. The L452R mutation in particular leads 234 to neutralizing antibody escape90,96–98. To estimate the epitope preferences of serum 235 antibodies further, we produced a Delta RBD protein in which R452 was reverted to 236 L452. Vaccination led to a response that was skewed toward the L452-containing RBD 237 (Figure 3A, compare to Figure 1B, middle panel), confirming the strong antibody bias 238 and immunodominance of this epitope reported previously99. Yet reciprocal skewing to 239 R452-containing RBD was not observed in primary Delta infections, suggesting that a 240 new immunogenic epitope is not created by this mutation (Figure 3A). Post-vaccination 241 Delta infections led to a relatively even ratio of antibodies that bound Delta-L452 to 242 those that reacted to Delta-R452 (Figure 3A), perhaps due to boosted levels of 243 antibodies that bound other conserved sites on RBD and the T478K epitope. We also 244 produced chimeric WuHu1 S1 proteins in which the Delta N-terminal domain (NTD) 245 supersite mutations (T19R, G142D, E156-, F157-, R158G) were introduced onto a 246 WuHu1 background92–94. Vaccination only controls showed a relatively even distribution 247 of antibodies that bound WuHu1 S1 and Delta NTD-WuHu1 S1 (Figure 3B compare to 248 Figure S3, left panel). Primary Delta infections, however, were subtly but significantly 249 skewed towards the Delta NTD (Figure 3B). Together, these data demonstrate a 250 shifting of immunodominance profiles, even in the absence of prior SARS-CoV-2 251 immunity. 252 To more precisely measure clonal shifts in antibody specificities and 253 immunodominance than can be achieved by serological assays, we performed LIBRA-254 seq using PBMC samples from primary and post-vaccination Delta infections100. 255 13 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Streptavidin-phycoerythrin (PE) tetramers were constructed using WuHu1 S1, Delta S1, 256 Delta RBD, Delta RBD-L452, and Delta NTD-WuHu1-S1, as described in Figures 1, 3 257 and S3, each carrying unique oligonucleotide barcodes. PE-binding memory cells were 258 then enriched and subjected to scRNA/V(D)J-seq. Consistent with our serological data 259 (Figure 3A-B), we observed few memory B cells that bound Delta RBD- and NTD-260 specific epitopes (Figure 3C) in primary Delta infections and post-vaccination Delta 261 infections (Figure S5A). A clear preference for Delta-unique epitopes in the NTD 262 relative to the RBD was observed within individuals that had experienced a primary 263 Delta infection (Figure 3D). Within each group, we did not observe any clear differences 264 in epitope-dependence of somatic mutation frequencies in memory B cells (Figure 265 S5B). We did, however, observe a greater frequency of somatic mutations in Spike-266 specific memory B cells in the post-vaccination Delta infection cohort relative to primary 267 Delta infections (Figure S5C). Together, these data suggest a marked shift in antibody 268 specificities in primary Delta variant infections relative to WuHu1 Spike. This explains in 269 part why responses to at least some drifted epitopes are not observed, irrespective of 270 prior vaccination. 271 During the course of this work, the heavily mutated Omicron (BA.1) variant 272 rapidly overtook Delta and swept to global dominance. To define post-vaccination BA.1 273 responses, we recruited individuals from our voluntary on-campus testing program who 274 had tested positive for SARS-CoV-2 between January 1 and March 31, 2022, with the 275 expectation that primary responses would be robust against this more antigenically 276 distant variant101. All individuals for this study who tested positive by PCR had 277 sequences confirmed to be BA.1 (Figure S2A). Individuals with a SARS-CoV-2 278 14 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint infection caused by a Delta variant or other Omicron sublineages were excluded from 279 the study. Viral genome sequencing of all remnant PCR+ samples on campus during 280 this period below a Ct value of 35 demonstrated that 93.7% of samples that could be 281 assigned a PANGO-lineage87 were caused by the BA.1 sublineage of Omicron (Figure 282 S2B). To obtain controls for this cohort, some of whom had received 3 doses of mRNA 283 vaccines, we also recruited a new group of vaccinated individuals who had never tested 284 positive in our voluntary university testing system and reported no known prior SARS-285 CoV-2 infections. After testing plasma for nucleocapsid antibodies as a marker of prior 286 infection, samples from 5 individuals with titers well above the mean values seen in 287 verified infections were excluded from further consideration (Figure S6). Relative to 288 both primary and post-vaccination Delta infections, post-vaccination BA.1 infections 289 generally led to fewer symptoms such as wet cough (Figure S1A) and shorter duration 290 of symptoms (Figure S1B). 291 We were unable to recruit any unvaccinated individuals on campus who had 292 experienced BA.1 infections. However, we were able to obtain serum and, for a subset, 293 PBMC samples from a separate study from the Centers for Disease Control and 294 Prevention HEROES and RECOVER projects85, in which 53 individuals met these 295 criteria (Table 1). Neutralizing antibody titers were skewed towards WA-1 in individuals 296 who had been vaccinated but not infected (Figure 4A). Post-vaccination BA.1 infections 297 led to significantly higher neutralizing antibody titers against BA.1 compared to both 298 vaccinated controls who had not been infected and primary infections (Figure 4A), 299 consistent with a memory B cell recall response. 300 15 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint We next examined binding antibody titers against WuHu1 or BA.1 RBD. Post-301 vaccination BA.1 infections led to increased levels of RBD-binding antibodies, both for 302 WuHu1 and BA.1, relative to the vaccinated only control cohort and primary BA.1 303 infections (Figure 4B, left and middle panels). Vaccination alone led to greater RBD 304 titers against BA.1 than did primary BA.1 infections, despite the many mismatches in 305 sequence (Figure 4B, middle panel). As expected, antibodies from vaccinated only 306 individuals were skewed towards WuHu1 relative to BA.1 RBD (Figure 4B, right 307 panel). Of the few antibodies induced by primary BA.1 infections, we observed a 308 skewing of specificities towards BA.1 RBD (Figure 4B, right panel). Ratios of WuHu1 309 and BA.1 RBD-binding antibodies in post-vaccination BA.1 infections more closely 310 resembled vaccinated controls than primary BA.1 infections (Figure 4B, right panel). 311 To further evaluate the specificities of antibody responses in post-vaccination 312 BA.1 infections, we again used antigen tetramers to identify RBD-specific memory B 313 cells (Figures S7 and 5A). As expected, primary BA.1 infections generated a lower 314 frequency of WuHu1 RBD-specific memory B cells compared to vaccinated controls 315 (Figure 5B, left panel). Unexpectedly, BA.1-specific RBD memory B cells were not 316 consistently detectable above background in any experimental group, even primary 317 BA.1 infections (Figure 5B, middle panel). These data seem to differ from the modest 318 skewing of the serological response seen above in primary BA.1 infections (Figure 4B, 319 right panel), but can potentially be explained by low overall responses and prior studies 320 that observed only partial overlap between memory B and antibody-secreting plasma 321 cell specificities and repertoires43,102,103. Instead, most RBD-specific memory B cells 322 from all cohorts were cross-reactive against WuHu1 and BA.1 RBD (Figure 5A, 5B, 323 16 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint right panel). Primary BA.1 infections produced numerically fewer cross-reactive RBD 324 memory B cells than did post-vaccination BA.1 infections (Figure 5B, right panel). 325 Given that the overall antibody and memory B cell response to BA.1 RBD was 326 quite modest (Figures 4B, 5A-B), we employed tetramers of full-length Spike trimers of 327 WuHu1 and BA.1 Spike to capture a greater breadth of memory B cell specificities than 328 could be observed with RBD tetramers (Figures 5C). WuHu1-specific memory B cells 329 were observed in vaccinated controls and post-vaccination BA.1 infections, but not after 330 primary BA.1 infections (Figures 5D, left panel). We again failed to consistently 331 observe BA.1-specific memory B cells in any of the groups, including primary BA.1 332 infections, though a subset of post-vaccination BA.1 infections did appear to generate 333 such cells well above background levels (Figure 5D, middle panel). As with RBD, 334 cross-reactive Spike-specific memory cells were significantly elevated in post-335 vaccination BA.1 infections relative to primary BA.1 infections, but not relative to 336 vaccinated only controls (Figure 5D, right panel). Cross-reactive memory B cells 337 composed by far the largest portion of SARS-CoV-2 specific responses within all 338 experimental groups (Figure S8). 339 For a subset of primary BA.1 and post-vaccination BA.1 cohorts, we obtained 340 samples which enabled us to quantify WuHu1, BA.1, and cross-reactive Spike- and 341 RBD- specific memory B cell frequencies before and after BA.1 infection. Irrespective of 342 vaccination status, memory B cells that were either WuHu1- or BA.1-RBD-specific 343 increased in frequency for only a subset of individuals after BA.1 infection (Figure 6A, 344 left and middle panels). However, cross-reactive RBD memory B cells consistently 345 and significantly increased after both primary and post-vaccination BA.1 infections 346 17 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint (Figure 6A, right panel). The frequency of cross-reactive Spike memory B cells also 347 significantly increased after primary BA.1 infections (Figures 6B, right panel). 348 To infer potential mechanisms of antigenic imprinting from these samples, we 349 first correlated pre-infection cross-reactive Spike-specific memory B cells and post-350 infection BA.1 Spike-specific memory B cells. A negative correlation could indicate 351 detrimental imprinting, whereby pre-existing memory B cells outcompete naïve B cells 352 and inhibit the generation of variant-specific responses. Instead, we observed a slight 353 positive but non-statistically significant correlation between pre-infection cross-reactive 354 Spike-specific memory B cells and post-infection BA.1 Spike-specific memory B cells 355 (Figure 7A). Similarly, we observed a non-significant positive correlation between post-356 infection cross-reactive Spike-specific memory B cells and post-infection BA.1 Spike-357 specific memory B cells (Figure 7B). 358 Given that these data do not support a mechanism of competitive inhibition of 359 naïve B cells by cross-reactive memory B cells, we explored other mechanisms by 360 which de novo responses to drifted epitopes are indirectly suppressed, such as 361 accelerated viral clearance by neutralizing antibodies and/or T cells. We found a 362 negative, but non-statistically significant correlation of de novo responses with pre-363 infection BA.1 neutralizing antibody titers (Figure 7C). Similarly, we observed a small 364 and non-significant negative correlation with pre-infection BA.1 Spike-specific T cell 365 numbers and post-infection BA.1 Spike memory B cells (Figure 7D). The small sample 366 sizes and variable times of blood sampling prior to infection preclude us from making 367 definitive conclusions about mechanisms driving antigenic imprinting. Nonetheless, the 368 data suggest that neutralizing antibody and/or memory T cell-mediated viral clearance 369 18 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint may indirectly underlie suppression of responses to drifted epitopes. This overall impact 370 is quite small relative to the marked changes in antibody immunodominance observed 371 in even primary BA.1 variant infections, irrespective of prior immunity. 372 19 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint

373 Antigenic imprinting is neither inherently beneficial nor detrimental; rather the 374 impact of prior immunity is context-dependent45. For example, pre-existing serum 375 antibodies can improve and focus de novo responses upon reinfection to only mutated 376 novel epitopes through epitope masking104–106. Similarly, de novo responses to drifted 377 epitopes can be improved by pre-existing CD4+ memory T cells in what is classically 378 known as the hapten-carrier effect107. Alternatively, high affinity memory B cells can 379 competitively inhibit naïve B cells by consuming limited amounts of antigen and T cell 380 help, leading to a suppression of de novo antibody responses108. If these memory B 381 cells target non-protective epitopes, this could in theory leave one worse off than if there 382 were no prior immunity whatsoever44,109. Finally, pre-existing immunity could indirectly 383 suppress new antibody responses to drifted epitopes simply by clearing away virus and 384 antigen before naïve B cells can robustly participate. 385 Though neutralizing antibody titers were robust following post-vaccination 386 infections, our results demonstrated a small negative impact of prior immunity on de 387 novo responses to drifted epitopes. Yet we found no evidence to support a mechanism 388 of competitive inhibition by cross-reactive memory B cells. Though not definitive, our 389 data instead hint at a role for pre-infection neutralizing antibodies and memory T cells, 390 suggesting that antigen clearance is the main mechanism by which de novo B cell 391 responses are indirectly suppressed by prior immunity. Indeed, pre-existing neutralizing 392 antibodies likely accelerate viral clearance110,111, and viral and vaccine antigens can 393 potentially also be cleared by T cells or non-neutralizing antibodies via Fc effector 394 functions112–114. Animal studies offer an attractive way to further test mechanisms of 395 20 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint antigenic imprinting on heterologous vaccine and viral infection responses. For 396 example, genetic tracking studies were used to show robust de novo responses to 397 Omicron boosters in mice previously vaccinated against the ancestral strain. Yet this 398 required two booster doses, and a small negative impact of prior immunity was 399 observed in inverse proportion to the antigenic distance between the two 400 immunizations101. Similar results have been reported in other mouse studies115. These 401 systems can thus potentially be used to manipulate specific immune parameters and 402 measure their contributions to antigenic imprinting in ways that are not possible in 403 human studies, especially since few immunologically naïve adults remain to serve as 404 controls. 405 Immunodominance hierarchies can also determine which epitopes are available 406 to be targeted by antibodies, irrespective of prior immunity. Prior studies, confirmed in 407 our experiments, showed that a large portion of COVID-19 vaccine-induced antibodies 408 are aimed at the L452 class 3 epitope116. Yet in the post-vaccination Delta cohort, we 409 observed few antibodies directed at the epitope containing the L452R mutation. Under 410 the assumption that one immunodominant epitope was being mutated to another, one 411 might have concluded that the absence of R452-specific antibodies could be explained 412 by antigenic imprinting. Yet by including a primary infection cohort, we observed that the 413 Delta variant intrinsically did not elicit detectable antibody responses against the R452 414 epitope, even with no prior SARS-CoV-2 exposures, consistent with an independent 415 study52. We can instead conclude that Delta shifts antibody immunodominance 416 hierarchies to instead focus more on epitopes located in the NTD. These types of shifts 417 in immunodominance preempt any considerations of the impact of antigenic imprinting. 418 21 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint The basis and mechanisms of these shifts for SARS-CoV-2 clearly needs more 419 investigation to determine whether and how best to overcome them. 420 This study spanned a period from the Delta wave through the more antigenically 421 distinct BA.1 Omicron wave. A central expectation of antigenic imprinting is that the 422 extent to which prior immunity interferes with de novo responses should decrease as 423 antigenic distance increases101. We used the Delta and BA.1 variants to test this 424 expectation in SARS-CoV-2 and to understand the impacts of antigenic distance on 425 antigenic imprinting. Despite our prediction, we observe even less of a variant specific 426 response in post-vaccination BA.1 infections compared to post-vaccination Delta 427 infections. Much of this can be explained by shifts in immunodominance in which even 428 primary BA.1 infections elicited few memory B cell responses to drifted epitopes. Yet 429 longitudinal sampling during BA.1 infections has also shown that viral titers do not reach 430 the peak levels observed in Delta infections117, suggesting that immune responses to 431 drifted epitopes occur in proportion to need and antigen availability. 432 22 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Disclosures: The findings and conclusions in this report are those of the authors and 433 do not necessarily represent the official position of the Centers for Disease Control and 434 Prevention. 435 436

This work was supported by NIH grants R01AI099108 and 437 R01AI129945 (D.B.) and a research grant from the Arizona Board of Regents (M.W. 438 and D.B). This project has been funded in part with Federal funds from the National 439 Institute of Allergy and Infectious Diseases, National Institutes of Health, Department of 440 Health and Human Services, under Contract No. 75N93021C00015 (M.W.) The 441 HEROES-RECOVER cohort is supported by the National Center for Immunization and 442 Respiratory Diseases and the Centers for Disease Control and Prevention (contracts 443 75D30120R68013 to Marshfield Clinic Research Institute, 75D30120C08379 to the 444 University of Arizona, and 75D30120C08150 to Abt Associates). 445 446 Declaration of Interests: Sana Biotechnology has licensed intellectual property of D.B. 447 and Washington University in St. Louis. Gilead Sciences has licensed intellectual 448 property of D.B. and Stanford University. Clade Therapeutics has licensed intellectual 449 property of D.B. and University of Arizona. D.B. is a co-founder of Clade Therapeutics. 450 D.B. served on an advisory panel for GlaxoSmithKline. B.J.L. has a financial interest in 451 Cofactor Genomics, Inc. and Iron Horse Dx. Geneticure Inc. has licensed intellectual 452 property of R.S. and R.S is a co-founder of Geneticure Inc. M.W. has received 453 consulting fees from GLG on SARS-CoV-2 and the COVID-19 pandemic. 454 455 23 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint

456 457 Participant selection 458 All human studies conducted at The University of Arizona were approved by the 459 Institutional Review Board for the Human Subjects Protection Program†. Individuals who 460 had participated in the voluntary on-campus saline gargle testing program and had 461 either never tested positive or had tested positive during the Delta or BA.1 waves were 462 contacted by email by the program administrators (not the authors on this study) about 463 willingness to participate in this research study. Participants were provided a link to an 464 eligibility questionnaire and, once eligibility (no immunosuppressive therapy in the last 5 465 years and HIV negative) was confirmed, additional demographic questions and a link to 466 schedule an appointment for blood draws. Written consent was obtained through an 467 electronic form. All blood draws were performed at the Clinical and Translational 468 Sciences Center at The University of Arizona. Additional primary and post-vaccination 469 BA.1 infection samples were acquired from the CDC HEROES-RECOVERS†† cohort85. 470 This study was reviewed by CDC and approved by the institutional review boards at 471 participating sites or under a reliance agreement with Abt Associates institutional review 472 board and was conducted consistent with applicable federal law and CDC policy under 473 45 C.F.R. part 46, 21 C.F.R. part 56, 42 U.S.C. Sect. 241(d), 5 U.S.C. Sect. 552a, 44 474 U.S.C. Sect. 3501 et seq. Methods for the HEROES-RECOVER Cohorts have 475 been published previously85,86. In summary, cohorts consisted of health care 476 personnel, first responders, and other essential and frontline workers in eight 477 U.S. locations across six states. Participants collected weekly nasal swabs which 478 24 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint were tested for SARS-CoV-2 viral material by RT-qPCR and additional swabs 479 were collected and screened upon the onset of any COVID-19–like illness 480 symptoms. In addition, blood draws were collected at enrollment, then 481 approximately every 3 months and after immune modifying events such as 482 vaccination or infection. Vaccination was documented by self-report and verified 483 by vaccine cards or electronic medical records or state immunization registries. 484 HEROES-RECOVER participants were selected based on testing positive for SARS-485 CoV-2 during Delta or BA.1 waves and having completed a blood draw after infection. 486 487 Saline Gargle PCR testing for SARS-CoV-2 488 As part of Test All, Test Smart, the University of Arizona’s voluntary campus-wide 489 testing program, University staff, faculty and students had access to SARS-CoV-2 rRT-490 PCR tests from August 2020 – July 2023. At testing and collection sites throughout 491 campus, individuals were given 5 mL of 0.9 % sterile saline (AddiPak 5 mL sterile saline 492 single use tubes, Teleflex, LLC) and guided to complete three rounds of a 5-second 493 swish followed by 10 seconds of gargling (adapted from Goldfarb et al.118). Samples 494 were deposited into collection tubes and then screened for SARS-CoV-2 by rRT-PCR. 495 496 PBMC and plasma preparation 497 Twenty milliliters of blood was collected by venipuncture in heparinized Vacutainer 498 tubes (BD). For PBMCs, 15ml of Ficoll-Paque PLUS (Thermo Fisher Scientific) was 499 added to 50-ml Leucosep tubes (Greiner) and spun for 1min at 1,000g to transfer the 500 density gradient below the filter. Twenty milliliters of blood from the heparinized tubes 501 25 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint was then poured into the top of the Leucosep tube and spun at 1,000g for 10min at 502 room temperature with the brake off. The top plasma layer was carefully collected and 503 frozen at −20 °C, and the remaining supernatant containing PBMCs above the filter was 504 poured into a new 50-ml conical tube containing 10mL of PBS and spun at 250g for 505 10min. Cell pellets were resuspended in RPMI media containing 10% FCS and counted 506 on a Vi-Cell XR (Beckman Coulter). Cells were diluted to a concentration of 2 × 10! 507 cells per mL in RPMI media containing 10% FCS. An equal volume of 80% FCS + 20% 508 dimethyl sulfoxide was added dropwise and inverted once to mix. Suspensions were 509 distributed at 1ml per cryovial and frozen overnight at −80 °C in Mr. Frosty freezing 510 chambers (Nalgene). Vials were then transferred to storage in liquid nitrogen. 511 512 513 ELISA and quantification of antibody titers 514 Serological assays were performed as previously described88. WuHu1 RBD (cat. no. 515 SPD-C52H3), WuHu1 S1subdomain of the SARS-CoV-2 S glycoprotein (cat. no. S1N-516 C52H3), WuHu1 Spike (cat. no. SPN-C52H9), Delta RBD (cat. no. SPD-C52Hh), Delta 517 S1 subdomain (cat. no. S1N-C52Hu), Omicron (BA.1) RBD (cat. no. SPD-C522e), 518 Omicron (BA.1) Spike (cat. no. SPN-C52Hz) and Nucleocapsid (cat. no. NUN-C5227) 519 were purchased from Acro Biosystems. Chimeric proteins (Delta RBD-L452 and Delta 520 NTD-WuHu1 S1) were custom synthesized by GenScript. To obtain titers and single-521 dilution OD450 values, antigens were immobilized on high-adsorbency 384-well plates 522 at 5 ng mL−1. Plates were blocked with 1% non-fat dehydrated milk extract (Santa Cruz 523 Biotechnology, sc-2325) in sterile PBS (Thermo Fisher Scientific HyClone PBS, 524 26 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint SH2035) for 1 h, washed with PBS containing 0.05% Tween-20 and overlaid for 60 min 525 with either a single 1:60 dilution or five serial 1:3 dilutions beginning at a 1:60 dilution of 526 serum. Plates were then washed and incubated for 1 h in 1% PBS and milk containing 527 anti-human Pan-Ig HRP-conjugated antibody (Jackson ImmunoResearch, 109-035-064) 528 at a concentration of 1:2,000 for 1 h. Plates were washed with PBS-Tween solution 529 followed by PBS wash. To develop, plates were incubated in tetramethylbenzidine 530 (Fisher Scientific) before quenching with 2 N H2SO4. Plates were read for 450-nm 531 absorbance on CLARIOstar Plus from BMG Labtech. All samples were also read at 532 630 nm to detect any incomplete quenching. Any samples above background 630-nm 533 values were re-run. Area under the curve (AUC) values were calculated in GraphPad 534 Prism (v9). 535 536 537 Virus neutralization assays 538 All live virus assays were performed at Biosafety Level 3 and were approved by the 539 University of Arizona Institutional Biosafety Committee. SARS-CoV-2, isolate USA-540 WA1/2020, was deposited by Dr Natalie J. Thornburg at the Centers for Disease 541 Control and Prevention and obtained from the World Reference Center for Emerging 542 Viruses and Arboviruses. Stocks of WA1/2020 SARS-CoV-2 were generated as a single 543 passage from received stock vial on mycoplasma-negative Vero cells (ATCC CCL-81). 544 B.1.617.2 (Delta) was received from WRCEVA, strain designation GNL-1205. B.1.1.529 545 (Omicron) originated from a nasopharyngeal swab collected at the University of Arizona. 546 It was passaged once on Calu-3 cells and then once on Vero cells to generate a master 547 27 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint stock. Viral PANGO-lineage, BA.1.187, was confirmed by Illumina sequencing 548 (EPI_ISL_17886211) of the master stock. 549 Supernatant and cell lysate were combined, subjected to a single freeze–thaw 550 and then centrifuged at 1,800g for 10min to remove cell debris. For PRNTs for SARS-551 CoV-2, Vero cells (ATCC, CCL-81) were plated in 96-well tissue culture plates and 552 grown overnight. Vero cells were confirmed by PCR to be free of mycoplasma using the 553 Universal Mycoplasma Detection Kit (ATCC). Serial dilutions of serum samples were 554 performed in duplicate and incubated with 100 plaque-forming units of SARS-CoV-2 for 555 1h at 37 °C. Plasma/serum dilutions plus virus were transferred to the cell plates and 556 incubated for 2h at 37 °C in 5% CO2 and then overlaid with 1% methylcellulose. After 557 72h, plates were fixed with 10% neutral buffered formalin for 30min and stained with 1% 558 crystal violet. Plaques were imaged using an ImmunoSpot Versa plate reader. The most 559 dilute serum concentration that led to ten or fewer plaques was designated as the 560 PRNT90 titer. Input PFU for each experiment was confirmed by plaque assay. 561 562 Flow cytometry 563 One milliliter of pre-warmed FBS was added to a frozen cryovial of 107 PBMCs, which 564 was rapidly thawed in a 37℃ water bath. Samples were poured into 15 mL conical tubes 565 containing 5 mL of pre-warmed RPMI with 5% FBS and 1% anti/anti. Tubes were spun 566 at 250g for 5 min at room temperature. 567 Delta 568 Supernatants were removed and cell pellets were resuspended in 200 µL of staining 569 buffer containing 1 µL each of anti-CD38-APC (BioLegend, clone HIT2), anti-CD13-PE-570 28 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Cy7 (BioLegend, clone WM15), anti-CD21-PE-Dazzle (BioLegend, clone Bu32), anti-571 CD19-APC-efluor-780 (Invitrogen, clone HIB19), anti-IgD-PerCP-Cy5.5 (Biolegend, 572 clone IA6-2), anti-IgM-FITC (Biolegend, clone MHM-88), anti-CD27-BV510 (Biolegend, 573 clone M-T271), anti-CD11c-Alexa700 (BioLegend, clone Bu15). Staining buffer also 574 contained Delta-S1-PE and S1-BV421 tetramers. 575 BA.1 576 Supernatants were removed and cell pellets were resuspended in 200 µL of staining 577 buffer containing 1 µL each of anti-CD38-BV421 (BioLegend, clone HB-7), anti-CD13-578 PE-Dazzle 594 (BioLegend, clone WM15), anti-CD21-PerCP Cy 5.5 (BioLegend, clone 579 Bu32), anti-CD19-APC-efluor-780 (Invitrogen, clone HIB19), anti-IgD-BV510 580 (Biolegend, clone 11-26c.2a), anti-IgM-FITC (Biolegend, clone MHM-88), anti-CD27-PE 581 Cy 7 (Biolegend, clone M-T271), anti-CD11c-Alexa700 (BioLegend, clone Bu15). Cells 582 were stained with live-dead marker, Zombie Yellow (BioLegend) according to 583 manufacturer’s recommendations. Staining buffer also contained BA.1-Spike-PE and 584 Spike-Alexa Fluor 647 tetramers. 585 586 Antibodies were validated by the manufacturer on human PBMCs. Tetramer reagents 587 were assembled by mixing 100µg ml-1 of C-terminal AviTagged S1, Delta S1, WuHu1 588 Spike, or BA.1 Spike (ACROBiosystems) with 100µg ml-1 of streptavidin-PE(BioLegend), 589 streptavidin-BV421 (BioLegend), or streptavidin-Alexa Fluor 647 (BioLegend), 590 respectively, at a 6:1 molar ratio for S1 or 4:1 molar ratio for Spike, in which ⅕ of the 591 final volume of streptavidin was added every 10 min. S1 and Spike tetramers were 592 validated by staining Lenti-X 293T cells(Takara Bio) as a negative control or 293T-593 29 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint hACE2-expressing cells (BEI Resources, NR-52511) as a positive control. Lenti-X 293T 594 cells were confirmed to be free of mycoplasma; 293T-hACE2 cells were maintained in 595 media containing 1% pen/strep to minimize chances of contamination. PBMC samples 596 were stained for at least 20 minutes, washed and filtered through 70-µm nylon mesh. 597 Data were analyzed on either a BD LSR2 (tetramer validation only), a Fortessa 598 cytometer (Delta), or BD Cytek Aurora (BA.1). Data were analyzed using FlowJo 599 software. 600 601 Flow cytometry and Fluorescence Activated Cell Sorting 602 One milliliter of pre-warmed FBS was added to a frozen cryovial of PBMCs and thawed 603 by pipetting. Samples were added to 15 mL conical tubes containing 10 mL of pre-604 warmed RPMI with 20% FBS and 1% anti/anti. Tubes were spun at 1200 RPM for 5 605 minutes at room temperature. Supernatants were removed and cell pellets were 606 resuspended in 200 µL of staining buffer contained 1 µL each of anti-CD19-607 BV421(Biolegend, clone HIB19), anti-CD27-FITC(Biolegend, clone O323), anti-CD13-608 PE-Cy-7(Biolegend, clone WM15), anti-IgD-APC-Cy-7(Biolegend, clone IA6-2). Staining 609 buffer also contained either 5 or 2 LIBRA-Seq tetramers: S1-PE(Biolegend, TotalSeq-610 C0951_PE), Delta S1-PE(Biolegend, TotalSeq-C0952_PE), Delta NTD/S1-611 PE(Biolegend, TotalSeq-C0953_PE), Delta RBD/L452(Biolegend, TotalSeq-612 C0954_PE), and Delta RBD-PE(Biolegend, TotalSeq-C0955_PE). Biotinylated tetramer 613 reagents were assembled by mixing 100µg ml-1 of C-terminal AviTagged S1 614 (ACROBiosystems), Delta S1 (ACROBiosystems), Delta NTD/S1 (GenScript), Delta 615 RBD/L452 (Genscript), or Delta RBD (ACROBiosystems) with 100µg ml-1 of 616 30 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint streptavidin-TotalSeq-C-PE (BioLegend) at a 6:1 molar ratio in which ⅕ of the final 617 volume of streptavadin was added every 10 min. Tetramers were validated by staining 618 Lenti-X 293T cells(Takara Bio) as a negative control or 293T-hACE2-expressing cells 619 (BEI Resources, NR-52511) as a positive control. Lenti-X 293T cells were confirmed to 620 be free of mycoplasma; 293T-hACE2 cells were maintained in media containing 1% 621 pen/strep to minimize chances of contamination. Additionally, TotalSeq-C anti-human 622 Hashtag antibodies (Biolegend, TotalSeq™-C0251-10) were added to individual 623 samples and pooled after staining and washing. PBMCs were stained in the dark for 30 624 minutes at 4˚C, washed, pooled and filtered through a 35 µm strainer (Fisher Scientific). 625 SARS-CoV-2 specific memory B cells (CD19+IgD-IgM-CD27+) as well as non-antigen 626 specific memory B cells were sorted using a FACSAria II. 627 628 Single-cell RNA sequencing and analysis 629 Cells were prepared and processed according to the 10X Genomics Single Cell 5’ Dual 630 Index protocol with Feature Barcoding Technology for Cell Surface Protein and Immune 631 Receptor Mapping kit (10X Genomics). Reads were processed and aligned using the 632 10X CellRanger multi pipeline to GRCh38 gex and vdj reference genomes (10X 633 Genomics). Each sample feature barcode matrix was loaded into R and analyzed 634 utilizing the Seurat package for gene expression, vdj and antibody capture analysis119. 635 Cell processing was conducted as previously described100. Data are available at NCBI 636 GEO accession number GSE242775. 637 638 ELISpot Assay 639 31 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Cryopreserved PBMC (5 × 106/sample) were thawed in prewarmed RPMI-1640 media 640 supplemented with L-glutamine + 10% FCS and 300ug DNAse. Thawed PBMCs were 641 rested overnight at 37 °C in X-VIVO™-15 Medium (Lonza) supplemented with 5% 642 human-AB serum. Cells were stimulated with ~1 nmol of peptide pool corresponding to 643 spike of Omicron (B.1.1.529) variant (16-mer peptide pools, overlapping by 10 amino 644 acids (21st century Biochemicals Inc.) on pre-coated human IFN-γ ELISpot plates 645 (Mabtech, Inc.) and developed after 18 hours according to manufacturer instructions. 646 Spots were imaged and counted using Iris FLUOROspot reader (Mabtech). 647 648 Statistical methods 649 All analyses are listed in the figure legends and were performed in GraphPad Prism 9 650 and/or the R programming language (v4.0.5). 651 652 Footnotes 653 † See 45 C.F.R. part 46; 21 C.F.R. part 56 654 †† This study was reviewed by CDC and approved by the institutional review boards at 655 participating sites or under a reliance agreement with Abt Associates institutional review 656 board and was conducted consistent with applicable federal law and CDC policy under 657 45 C.F.R. part 46, 21 C.F.R. part 56, 42 U.S.C. Sect. 241(d), 5 U.S.C. Sect. 552a, 44 658 U.S.C. Sect. 3501 et seq.659 660 661 662 32 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint

663 1. Novel 2019 coronavirus genome (2020). Virological. https://virological.org/t/novel-664 2019-coronavirus-genome/319. 665 2. Baden, L.R., El Sahly, H.M., Essink, B., Kotloff, K., Frey, S., Novak, R., Diemert, D., 666 Spector, S.A., Rouphael, N., Creech, C.B., et al. (2021). Efficacy and safety of the 667 mRNA-1273 SARS-CoV-2 vaccine. N. Engl. J. Med. 384, 403–416. 668 10.1056/NEJMoa2035389. 669 3. Polack, F.P., Thomas, S.J., Kitchin, N., Absalon, J., Gurtman, A., Lockhart, S., 670 Perez, J.L., Pérez Marc, G., Moreira, E.D., Zerbini, C., et al. (2020). Safety and 671 Efficacy of the BNT162b2 mRNA Covid-19 Vaccine. N. Engl. J. Med. 383, 2603–672 2615. 10.1056/NEJMoa2034577. 673 4. Sadoff, J., Gray, G., Vandebosch, A., Cárdenas, V., Shukarev, G., Grinsztejn, B., 674 Goepfert, P.A., Truyers, C., Fennema, H., Spiessens, B., et al. (2021). Safety and 675 efficacy of single-dose Ad26.COV2.S vaccine against Covid-19. N. Engl. J. Med. 676 384, 2187–2201. 10.1056/NEJMoa2101544. 677 5. Tanriover, M.D., Doğanay, H.L., Akova, M., Güner, H.R., Azap, A., Akhan, S., Köse, 678 Ş., Erdinç, F.Ş., Akalın, E.H., Tabak, Ö.F., et al. (2021). Efficacy and safety of an 679 inactivated whole-virion SARS-CoV-2 vaccine (CoronaVac): interim results of a 680 double-blind, randomised, placebo-controlled, phase 3 trial in Turkey. Lancet 398, 681 213–222. 10.1016/S0140-6736(21)01429-X. 682 33 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint 6. Heath, P.T., Galiza, E.P., Baxter, D.N., Boffito, M., Browne, D., Burns, F.,683 Chadwick, D.R., Clark, R., Cosgrove, C., Galloway, J., et al. (2021). Safety and 684 efficacy of NVX-CoV2373 covid-19 vaccine. N. Engl. J. Med. 385, 1172–1183. 685 10.1056/NEJMoa2107659. 686 7. Voysey, M., Clemens, S.A.C., Madhi, S.A., Weckx, L.Y., Folegatti, P.M., Aley, P.K.,687 Angus, B., Baillie, V.L., Barnabas, S.L., Bhorat, Q.E., et al. (2021). Safety and 688 efficacy of the ChAdOx1 nCoV-19 vaccine (AZD1222) against SARS-CoV-2: an 689 interim analysis of four randomised controlled trials in Brazil, South Africa, and the 690 UK. Lancet 397, 99–111. 10.1016/S0140-6736(20)32661-1. 691 8. Xia, S., Zhang, Y., Wang, Y., Wang, H., Yang, Y., Gao, G.F., Tan, W., Wu, G., Xu,692 M., Lou, Z., et al. (2021). Safety and immunogenicity of an inactivated SARS-CoV-2 693 vaccine, BBIBP-CorV: a randomised, double-blind, placebo-controlled, phase 1/2 694 trial. Lancet Infect. Dis. 21, 39–51. 10.1016/S1473-3099(20)30831-8. 695 9. Pouwels, K.B., Pritchard, E., Matthews, P.C., Stoesser, N., Eyre, D.W., Vihta, K.-D.,696 House, T., Hay, J., Bell, J.I., Newton, J.N., et al. (2021). Effect of Delta variant on 697 viral burden and vaccine effectiveness against new SARS-CoV-2 infections in the 698 UK. Nat. Med. 27, 2127–2135. 10.1038/s41591-021-01548-7. 699 10. Chen, J., Wang, R., Gilby, N.B., and Wei, G.-W. (2022). Omicron variant700 (B.1.1.529): Infectivity, vaccine breakthrough, and antibody resistance. J. Chem. Inf. 701 Model. 62, 412–422. 10.1021/acs.jcim.1c01451. 702 34 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint 11. Tao, K., Tzou, P.L., Nouhin, J., Gupta, R.K., de Oliveira, T., Kosakovsky Pond, S.L., 703 Fera, D., and Shafer, R.W. (2021). The biological and clinical significance of 704 emerging SARS-CoV-2 variants. Nat. Rev. Genet., 1–17. 10.1038/s41576-021-705 00408-x. 706 12. Dejnirattisai, W., Huo, J., Zhou, D., Zahradník, J., Supasa, P., Liu, C., Duyvesteyn, 707 H.M.E., Ginn, H.M., Mentzer, A.J., Tuekprakhon, A., et al. (2022). SARS-CoV-2 708 Omicron-B.1.1.529 leads to widespread escape from neutralizing antibody 709 responses. Cell. 10.1016/j.cell.2021.12.046. 710 13. Gilbert, P.B., Montefiori, D.C., McDermott, A.B., Fong, Y., Benkeser, D., Deng, W., 711 Zhou, H., Houchens, C.R., Martins, K., Jayashankar, L., et al. (2022). Immune 712 correlates analysis of the mRNA-1273 COVID-19 vaccine efficacy clinical trial. 713 Science 375, 43–50. 10.1126/science.abm3425. 714 14. Khoury, D.S., Cromer, D., Reynaldi, A., Schlub, T.E., Wheatley, A.K., Juno, J.A., 715 Subbarao, K., Kent, S.J., Triccas, J.A., and Davenport, M.P. (2021). Neutralizing 716 antibody levels are highly predictive of immune protection from symptomatic SARS-717 CoV-2 infection. Nat. Med. 27, 1205–1211. 10.1038/s41591-021-01377-8. 718 15. Rausch, J.W., Capoferri, A.A., Katusiime, M.G., Patro, S.C., and Kearney, M.F. 719 (2020). Low genetic diversity may be an Achilles heel of SARS-CoV-2. Proc. Natl. 720 Acad. Sci. U. S. A. 117, 24614–24616. 10.1073/pnas.2017726117. 721 16. Volz, E. (2023). Fitness, growth and transmissibility of SARS-CoV-2 genetic 722 variants. Nat. Rev. Genet., 1–11. 10.1038/s41576-023-00610-z. 723 35 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint 17. Wiegand, T., Nemudryi, A., Nemudraia, A., McVey, A., Little, A., Taylor, D.N., Walk,724 S.T., and Wiedenheft, B. (2022). The Rise and Fall of SARS-CoV-2 Variants and725 Ongoing Diversification of Omicron. Viruses 14, 2009. 10.3390/v14092009. 726 18. Hill, V., Du Plessis, L., Peacock, T.P., Aggarwal, D., Colquhoun, R., Carabelli, A.M.,727 Ellaby, N., Gallagher, E., Groves, N., Jackson, B., et al. (2022). The origins and 728 molecular evolution of SARS-CoV-2 lineage B.1.1.7 in the UK. Virus Evol. 8, 729 veac080. 10.1093/ve/veac080. 730 19. Cao, Y., Song, W., Wang, L., Liu, P., Yue, C., Jian, F., Yu, Y., Yisimayi, A., Wang,731 P., Wang, Y., et al. (2022). Characterization of the enhanced infectivity and 732 antibody evasion of Omicron BA.2.75. Cell Host Microbe 30, 1527-1539.e5. 733 10.1016/j.chom.2022.09.018. 734 20. Volz, E., Mishra, S., Chand, M., Barrett, J.C., Johnson, R., Geidelberg, L., Hinsley,735 W.R., Laydon, D.J., Dabrera, G., O’Toole, Á., et al. (2021). Assessing736 transmissibility of SARS-CoV-2 lineage B.1.1.7 in England. Nature 593, 266–269. 737 10.1038/s41586-021-03470-x. 738 21. Cherian, S., Potdar, V., Jadhav, S., Yadav, P., Gupta, N., Das, M., Rakshit, P.,739 Singh, S., Abraham, P., Panda, S., et al. (2021). Convergent evolution of SARS-740 CoV-2 spike mutations, L452R, E484Q and P681R, in the second wave of COVID-741 19 in Maharashtra, India. bioRxiv, 2021.04.22.440932. 10.1101/2021.04.22.440932. 742 22. Cheng, Y.-W., Chao, T. -L., Li, C.-L., Wang, S.-H., Kao, H.-C., Tsai, Y.-M., Wang,743 H.-Y., Hsieh, C.-L., Lin, Y.-Y., Chen, P.-J., et al. (2021). D614G Substitution of 744 36 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint SARS-CoV-2 Spike Protein Increases Syncytium Formation and Virus Titer via 745 Enhanced Furin-Mediated Spike Cleavage. MBio, e0058721. 10.1128/mBio.00587-746 21. 747 23. McCallum, M., Bassi, J., De Marco, A., Chen, A., Walls, A.C., Di Iulio, J., Tortorici, 748 M.A., Navarro, M.-J., Silacci-Fregni, C., Saliba, C., et al. (2021). SARS-CoV-2 749 immune evasion by the B.1.427/B.1.429 variant of concern. Science. 750 10.1126/science.abi7994. 751 24. Imai, M., Halfmann, P.J., Yamayoshi, S., Iwatsuki-Horimoto, K., Chiba, S., 752 Watanabe, T., Nakajima, N., Ito, M., Kuroda, M., Kiso, M., et al. (2021). 753 Characterization of a new SARS-CoV-2 variant that emerged in Brazil. PNAS. 754 10.1073/pnas.2106535118/-/DCSupplemental. 755 25. Viana, R., Moyo, S., Amoako, D.G., Tegally, H., Scheepers, C., Althaus, C.L., 756 Anyaneji, U.J., Bester, P.A., Boni, M.F., Chand, M., et al. (2022). Rapid epidemic 757 expansion of the SARS-CoV-2 Omicron variant in southern Africa. Nature 603, 758 679–686. 10.1038/s41586-022-04411-y. 759 26. Sokal, A., Chappert, P., Barba-Spaeth, G., Roeser, A., Fourati, S., Azzaoui, I., 760 Vandenberghe, A., Fernandez, I., Meola, A., Bouvier-Alias, M., et al. (2021). 761 Maturation and persistence of the anti-SARS-CoV-2 memory B cell response. Cell 762 184, 1201-1213.e14. 10.1016/j.cell.2021.01.050. 763 27. Muecksch, F., Weisblum, Y., Barnes, C.O., Schmidt, F., Schaefer-Babajew, D., 764 Wang, Z., C Lorenzi, J.C., Flyak, A.I., DeLaitsch, A.T., Huey-Tubman, K.E., et al. 765 37 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint (2021). Affinity maturation of SARS-CoV-2 neutralizing antibodies confers potency, 766 breadth, and resilience to viral escape mutations. Immunity. 767 10.1016/j.immuni.2021.07.008. 768 28. Turner, J.S., O’Halloran, J.A., Kalaidina, E., Kim, W., Schmitz, A.J., Zhou, J.Q., Lei,769 T., Thapa, M., Chen, R.E., Case, J.B., et al. (2021). SARS-CoV-2 mRNA vaccines 770 induce persistent human germinal centre responses. Nature. 10.1038/s41586-021-771 03738-2. 772 29. Kim, W., Zhou, J.Q., Horvath, S.C., Schmitz, A.J., Sturtz, A.J., Lei, T., Liu, Z.,773 Kalaidina, E., Thapa, M., Alsoussi, W.B., et al. (2022). Germinal centre-driven 774 maturation of B cell response to mRNA vaccination. Nature, 1–8. 10.1038/s41586-775 022-04527-1.776 30. Victora, G.D., and Nussenzweig, M.C. (2022). Germinal centers. Annu. Rev.777 Immunol. 40, 413–442. 10.1146/annurev-immunol-120419-022408. 778 31. Seifert, M., Przekopowitz, M., Taudien, S., Lollies, A., Ronge, V., Drees, B.,779 Lindemann, M., Hillen, U., Engler, H., Singer, B.B., et al. (2015). Functional 780 capacities of human IgM memory B cells in early inflammatory responses and 781 secondary germinal center reactions. Proc. Natl. Acad. Sci. U. S. A. 112, E546-55. 782 10.1073/pnas.1416276112. 783 32. Dogan, I., Bertocci, B., Vilmont, V., Delbos, F., Mégret, J., Storck, S., Reynaud, C.-784 A., and Weill, J.-C. (2009). Multiple layers of B cell memory with different effector 785 functions. Nat. Immunol. 10, 1292–1299. 10.1038/ni.1814. 786 38 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint 33. Pape, K.A., Taylor, J.J., Maul, R.W., Gearhart, P.J., and Jenkins, M.K. (2011). 787 Different B cell populations mediate early and late memory during an endogenous 788 immune response. Science 331, 1203–1207. 10.1126/science.1201730. 789 34. Zuccarino-Catania, G.V., Sadanand, S., Weisel, F.J., Tomayko, M.M., Meng, H., 790 Kleinstein, S.H., Good-Jacobson, K.L., and Shlomchik, M.J. (2014). CD80 and PD-791 L2 define functionally distinct memory B cell subsets that are independent of 792 antibody isotype. Nat. Immunol. 15, 631–637. 10.1038/ni.2914. 793 35. Horikawa, K., Martin, S.W., Pogue, S.L., Silver, K., Peng, K., Takatsu, K., and 794 Goodnow, C.C. (2007). Enhancement and suppression of signaling by the 795 conserved tail of IgG memory-type B cell antigen receptors. J. Exp. Med. 204, 759–796 769. 10.1084/jem.20061923. 797 36. Waisman, A., Kraus, M., Seagal, J., Ghosh, S., Melamed, D., Song, J., Sasaki, Y., 798 Classen, S., Lutz, C., Brombacher, F., et al. (2007). IgG1 B cell receptor signaling is 799 inhibited by CD22 and promotes the development of B cells whose survival is less 800 dependent on Ig alpha/beta. J. Exp. Med. 204, 747–758. 10.1084/jem.20062024. 801 37. Engels, N., König, L.M., Heemann, C., Lutz, J., Tsubata, T., Griep, S., Schrader, V., 802 and Wienands, J. (2009). Recruitment of the cytoplasmic adaptor Grb2 to surface 803 IgG and IgE provides antigen receptor-intrinsic costimulation to class-switched B 804 cells. Nat. Immunol. 10, 1018–1025. 10.1038/ni.1764. 805 38. Kometani, K., Nakagawa, R., Shinnakasu, R., Kaji, T., Rybouchkin, A., Moriyama, 806 S., Furukawa, K., Koseki, H., Takemori, T., and Kurosaki, T. (2013). Repression of 807 39 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint the transcription factor Bach2 contributes to predisposition of IgG1 memory B cells 808 toward plasma cell differentiation. Immunity 39, 136–147. 809 10.1016/j.immuni.2013.06.011. 810 39. Bhattacharya, D., Cheah, M.T., Franco, C.B., Hosen, N., Pin, C.L., Sha, W.C., and811 Weissman, I.L. (2007). Transcriptional profiling of antigen-dependent murine B cell 812 differentiation and memory formation. J. Immunol. 179, 6808–6819. 813 10.4049/jimmunol.179.10.6808. 814 40. Tomayko, M.M., Anderson, S.M., Brayton, C.E., Sadanand, S., Steinel, N.C.,815 Behrens, T.W., and Shlomchik, M.J. (2008). Systematic comparison of gene 816 expression between murine memory and naive B cells demonstrates that memory B 817 cells have unique signaling capabilities. J. Immunol. 181, 27–38. 818 10.4049/jimmunol.181.1.27. 819 41. Wong, R., Belk, J.A., Govero, J., Uhrlaub, J.L., Reinartz, D., Zhao, H., Errico, J.M.,820 D’Souza, L., Ripperger, T.J., Nikolich-Zugich, J., et al. (2020). Affinity-Restricted 821 Memory B Cells Dominate Recall Responses to Heterologous Flaviviruses. 822 Immunity 53, 1078-1094.e7. 10.1016/j.immuni.2020.09.001. 823 42. Mesin, L., Schiepers, A., Ersching, J., Barbulescu, A., Cavazzoni, C.B., Angelini, A.,824 Okada, T., Kurosaki, T., and Victora, G.D. (2020). Restricted clonality and limited 825 germinal center reentry characterize memory B cell reactivation by boosting. Cell 826 180, 92-106.e11. 10.1016/j.cell.2019.11.032. 827 40 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint 43. Purtha, W.E., Tedder, T.F., Johnson, S., Bhattacharya, D., and Diamond, M.S.828 (2011). Memory B cells, but not long-lived plasma cells, possess antigen 829 specificities for viral escape mutants. J. Exp. Med. 208, 2599–2606. 830 10.1084/jem.20110740. 831 44. Francis, T. (1960). On the Doctrine of Original Antigenic Sin. Proc. Am. Philos. Soc.832 104, 572–578. 833 45. Worobey, M., Plotkin, S., and Hensley, S.E. (2020). Influenza Vaccines Delivered in834 Early Childhood Could Turn Antigenic Sin into Antigenic Blessings. Cold Spring 835 Harb. Perspect. Med. 10. 10.1101/cshperspect.a038471. 836 46. Anderson, E.M., Li, S.H., Awofolaju, M., Eilola, T., Goodwin, E., Bolton, M.J.,837 Gouma, S., Manzoni, T.B., Hicks, P., Goel, R.R., et al. (2022). SARS-CoV-2 838 infections elicit higher levels of original antigenic sin antibodies compared with 839 SARS-CoV-2 mRNA vaccinations. Cell Rep. 41, 111496. 840 10.1016/j.celrep.2022.111496. 841 47. Amanat, F., Thapa, M., Lei, T., Sayed Ahmed, S.M., Adelsberg, D.C., Carreno,842 J.M., Strohmeier, S., Schmitz, A.J., Zafar, S., Zhou, J.Q., et al. (2021). SARS-CoV-843 2 mRNA vaccination induces functionally diverse antibodies to NTD, RBD and S2. 844 Cell. 10.1016/j.cell.2021.06.005. 845 48. Worobey, M., Han, G.-Z., and Rambaut, A. (2014). Genesis and pathogenesis of846 the 1918 pandemic H1N1 influenza A virus. Proc. Natl. Acad. Sci. U. S. A. 111, 847 8107–8112. 10.1073/pnas.1324197111. 848 41 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint 49. Taubenberger, J.K., and Morens, D.M. (2006). 1918 Influenza: the mother of all849 pandemics. Emerg. Infect. Dis. 12, 15–22. 10.3201/eid1201.050979. 850 50. Gostic, K.M., Bridge, R., Brady, S., Viboud, C., Worobey, M., and Lloyd-Smith, J.O.851 (2019). Childhood immune imprinting to influenza A shapes birth year-specific risk 852 during seasonal H1N1 and H3N2 epidemics. PLoS Pathog. 15, e1008109. 853 10.1371/journal.ppat.1008109. 854 51. Gostic, K.M., Ambrose, M., Worobey, M., and Lloyd-Smith, J.O. (2016). Potent855 protection against H5N1 and H7N9 influenza via childhood hemagglutinin 856 imprinting. Science 354, 722–726. 10.1126/science.aag1322. 857 52. Greaney, A.J., Eguia, R.T., Starr, T.N., Khan, K., Franko, N., Logue, J.K., Lord,858 S.M., Speake, C., Chu, H.Y., Sigal, A., et al. (2022). The SARS-CoV-2 Delta variant859 induces an antibody response largely focused on class 1 and 2 antibody epitopes. 860 PLoS Pathog. 18, e1010592. 10.1371/journal.ppat.1010592. 861 53. Abbott, R.K., Lee, J.H., Menis, S., Skog, P., Rossi, M., Ota, T., Kulp, D.W., Bhullar,862 D., Kalyuzhniy, O., Havenar-Daughton, C., et al. (2018). Precursor Frequency and 863 Affinity Determine B Cell Competitive Fitness in Germinal Centers, Tested with 864 Germline-Targeting HIV Vaccine Immunogens. Immunity 48, 133-146.e6. 865 10.1016/j.immuni.2017.11.023. 866 54. Chan, T.D., Gatto, D., Wood, K., Camidge, T., Basten, A., and Brink, R. (2009).867 Antigen affinity controls rapid T-dependent antibody production by driving the 868 42 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint expansion rather than the differentiation or extrafollicular migration of early 869 plasmablasts. J. Immunol. 183, 3139–3149. 10.4049/jimmunol.0901690. 870 55. Anderson, S.M., Khalil, A., Uduman, M., Hershberg, U., Louzoun, Y., Haberman,871 A.M., Kleinstein, S.H., and Shlomchik, M.J. (2009). Taking advantage: high-affinity872 B cells in the germinal center have lower death rates, but similar rates of division, 873 compared to low-affinity cells. J. Immunol. 183, 7314–7325. 874 10.4049/jimmunol.0902452. 875 56. Shih, T.-A.Y., Meffre, E., Roederer, M., and Nussenzweig, M.C. (2002). Role of876 BCR affinity in T cell dependent antibody responses in vivo. Nat. Immunol. 3, 570–877 575. 10.1038/ni803.878 57. Yeh, C.-H., Nojima, T., Kuraoka, M., and Kelsoe, G. (2018). Germinal center entry879 not selection of B cells is controlled by peptide-MHCII complex density. Nat. 880 Commun. 9, 1–11. 10.1038/s41467-018-03382-x. 881 58. Kwong, P.D., Wyatt, R., Robinson, J., Sweet, R.W., Sodroski, J., and Hendrickson,882 W.A. (1998). Structure of an HIV gp120 envelope glycoprotein in complex with the 883 CD4 receptor and a neutralizing human antibody. Nature 393, 648–659. 884 10.1038/31405. 885 59. Grant, O.C., Montgomery, D., Ito, K., and Woods, R.J. (2020). Analysis of the886 SARS-CoV-2 spike protein glycan shield reveals implications for immune 887 recognition. Sci. Rep. 10, 14991. 10.1038/s41598-020-71748-7. 888 43 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint 60. Chan, T.D., Wood, K., Hermes, J.R., Butt, D., Jolly, C.J., Basten, A., and Brink, R.889 (2012). Elimination of germinal-center-derived self-reactive B cells is governed by 890 the location and concentration of self-antigen. Immunity 37, 893–904. 891 10.1016/j.immuni.2012.07.017. 892 61. Reed, J.H., Jackson, J., Christ, D., and Goodnow, C.C. (2016). Clonal redemption893 of autoantibodies by somatic hypermutation away from self-reactivity during human 894 immunization. J. Exp. Med. 213, 1255–1265. 10.1084/jem.20151978. 895 62. Sabouri, Z., Schofield, P., Horikawa, K., Spierings, E., Kipling, D., Randall, K.L.,896 Langley, D., Roome, B., Vazquez-Lombardi, R., Rouet, R., et al. (2014). 897 Redemption of autoantibodies on anergic B cells by variable-region glycosylation 898 and mutation away from self-reactivity. Proc. Natl. Acad. Sci. U. S. A. 111, E2567-899 75. 10.1073/pnas.1406974111.900 63. Sangesland, M., Torrents de la Peña, A., Boyoglu-Barnum, S., Ronsard, L.,901 Mohamed, F.A.N., Moreno, T.B., Barnes, R.M., Rohrer, D., Lonberg, N., 902 Ghebremichael, M., et al. (2022). Allelic polymorphism controls autoreactivity and 903 vaccine elicitation of human broadly neutralizing antibodies against influenza virus. 904 Immunity 55, 1693-1709.e8. 10.1016/j.immuni.2022.07.006. 905 64. Berman, J.E., Nickerson, K.G., Pollock, R.R., Barth, J.E., Schuurman, R.K.,906 Knowles, D.M., Chess, L., and Alt, F.W. (1991). VH gene usage in humans: biased 907 usage of the VH6 gene in immature B lymphoid cells. Eur. J. Immunol. 21, 1311–908 1314. 10.1002/eji.1830210532. 909 44 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint 65. Willems van Dijk, K., Milner, L.A., Sasso, E.H., and Milner, E.C. (1992).910 Chromosomal organization of the heavy chain variable region gene segments 911 comprising the human fetal antibody repertoire. Proc. Natl. Acad. Sci. U. S. A. 89, 912 10430–10434. 10.1073/pnas.89.21.10430. 913 66. Nadel, B., Tang, A., Lugo, G., Love, V., Escuro, G., and Feeney, A.J. (1998).914 Decreased frequency of rearrangement due to the synergistic effect of nucleotide 915 changes in the heptamer and nonamer of the recombination signal sequence of the 916 V kappa gene A2b, which is associated with increased susceptibility of Navajos to 917 Haemophilus influenzae type b disease. J. Immunol. 161, 6068–6073. 918 67. Murray, S.M., Ansari, A.M., Frater, J., Klenerman, P., Dunachie, S., Barnes, E., and919 Ogbe, A. (2023). The impact of pre-existing cross-reactive immunity on SARS-CoV-920 2 infection and vaccine responses. Nat. Rev. Immunol. 23, 304–316. 921 10.1038/s41577-022-00809-x. 922 68. Chemaitelly, H., Ayoub, H.H., Tang, P., Hasan, M.R., Coyle, P., Yassine, H.M., Al-923 Khatib, H.A., Smatti, M.K., Al-Kanaani, Z., Al-Kuwari, E., et al. (2022). Immune 924 Imprinting and Protection against Repeat Reinfection with SARS-CoV-2. N. Engl. J. 925 Med. 10.1056/NEJMc2211055. 926 69. Koutsakos, M., and Ellebedy, A.H. (2023). Immunological imprinting: Understanding927 COVID-19. Immunity 56, 909–913. 10.1016/j.immuni.2023.04.012. 928 70. Alsoussi, W.B., Malladi, S.K., Zhou, J.Q., Liu, Z., Ying, B., Kim, W., Schmitz, A.J.,929 Lei, T., Horvath, S.C., Sturtz, A.J., et al. (2023). SARS-CoV-2 Omicron boosting 930 45 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint induces de novo B cell response in humans. Nature 617, 592–598. 931 10.1038/s41586-023-06025-4. 932 71. Cao, Y., Yisimayi, A., Jian, F., Song, W., Xiao, T., Wang, L., Du, S., Wang, J., Li, 933 Q., Chen, X., et al. (2022). BA.2.12.1, BA.4 and BA.5 escape antibodies elicited by 934 Omicron infection. Nature 608, 593–602. 10.1038/s41586-022-04980-y. 935 72. Cao, Y., Jian, F., Wang, J., Yu, Y., Song, W., Yisimayi, A., Wang, J., An, R., Chen, 936 X., Zhang, N., et al. (2023). Imprinted SARS-CoV-2 humoral immunity induces 937 convergent Omicron RBD evolution. Nature 614, 521–529. 10.1038/s41586-022-938 05644-7. 939 73. Park, Y.-J., Pinto, D., Walls, A.C., Liu, Z., De Marco, A., Benigni, F., Zatta, F., 940 Silacci-Fregni, C., Bassi, J., Sprouse, K.R., et al. (2022). Imprinted antibody 941 responses against SARS-CoV-2 Omicron sublineages. Science 378, 619–627. 942 10.1126/science.adc9127. 943 74. Reynolds, C.J., Gibbons, J.M., Pade, C., Lin, K.-M., Sandoval, D.M., Pieper, F., 944 Butler, D.K., Liu, S., Otter, A.D., Joy, G., et al. (2022). Heterologous infection and 945 vaccination shapes immunity against SARS-CoV-2 variants. Science 375, 183–192. 946 10.1126/science.abm0811. 947 75. Gagne, M., Moliva, J.I., Foulds, K.E., Andrew, S.F., Flynn, B.J., Werner, A.P., 948 Wagner, D.A., Teng, I.-T., Lin, B.C., Moore, C., et al. (2022). mRNA-1273 or 949 mRNA-Omicron boost in vaccinated macaques elicits similar B cell expansion, 950 46 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint neutralizing responses, and protection from Omicron. Cell 185, 1556-1571.e18. 951 10.1016/j.cell.2022.03.038. 952 76. Masurel, N. (1969). RELATION BETWEEN HONG KONG VIRUS AND FORMER953 HUMAN A2 ISOLATES AND THE A/EQUI2 VIRUS IN HUMAN SERA COLLECTED 954 BEFORE 1957. Lancet 293, 907–910. 10.1016/S0140-6736(69)92544-6. 955 77. Masurel, N., and Heijtink, R.A. (1983). Recycling of H1N1 influenza A virus in man--956 a haemagglutinin antibody study. J. Hyg. 90, 397–402. 957 10.1017/s0022172400029028. 958 78. Zhang, Z., Mateus, J., Coelho, C.H., Dan, J.M., Moderbacher, C.R., Gálvez, R.I.,959 Cortes, F.H., Grifoni, A., Tarke, A., Chang, J., et al. (2022). Humoral and cellular 960 immune memory to four COVID-19 vaccines. Cell 185, 2434-2451.e17. 961 10.1016/j.cell.2022.05.022. 962 79. Kaplonek, P., Fischinger, S., Cizmeci, D., Bartsch, Y.C., Kang, J., Burke, J.S., Shin,963 S.A., Dayal, D., Martin, P., Mann, C., et al. (2022). mRNA-1273 vaccine-induced964 antibodies maintain Fc-effector functions across SARS-CoV-2 Variants of Concern. 965 Immunity. 10.1016/j.immuni.2022.01.001. 966 80. Goel, R.R., Apostolidis, S.A., Painter, M.M., Mathew, D., Pattekar, A., Kuthuru, O.,967 Gouma, S., Hicks, P., Meng, W., Rosenfeld, A.M., et al. (2021). Distinct antibody 968 and memory B cell responses in SARS-CoV-2 naïve and recovered individuals 969 following mRNA vaccination. Sci Immunol 6. 10.1126/sciimmunol.abi6950. 970 47 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint 81. Greaney, A.J., Loes, A.N., Gentles, L.E., Crawford, K.H.D., Starr, T.N., Malone,971 K.D., Chu, H.Y., and Bloom, J.D. (2021). Antibodies elicited by mRNA-1273972 vaccination bind more broadly to the receptor binding domain than do those from 973 SARS-CoV-2 infection. Science Translational Medicine. 974 10.1126/scitranslmed.abi9915. 975 82. Cherian, S., Potdar, V., Jadhav, S., Yadav, P., Gupta, N., Das, M., Rakshit, P.,976 Singh, S., Abraham, P., Panda, S., et al. (2021). SARS-CoV-2 spike mutations, 977 L452R, T478K, E484Q and P681R, in the second wave of COVID-19 in 978 Maharashtra, India. Microorganisms 9, 1542. 10.3390/microorganisms9071542. 979 83. Akinbami, L.J., Kruszon-Moran, D., Wang, C.-Y., Storandt, R.J., Clark, J., Riddles,980 M.K., and Mohadjer, L.K. (2022). SARS-CoV-2 serology and self-reported infection981 among adults - National Health and Nutrition Examination Survey, United States, 982 august 2021-may 2022. MMWR Morb. Mortal. Wkly. Rep. 71, 1522–1525. 983 10.15585/mmwr.mm7148a4. 984 84. Jones, J.M., Manrique, I.M., Stone, M.S., Grebe, E., Saa, P., Germanio, C.D.,985 Spencer, B.R., Notari, E., Bravo, M., Lanteri, M.C., et al. (2023). Estimates of 986 SARS-CoV-2 seroprevalence and incidence of primary SARS-CoV-2 infections 987 among blood donors, by COVID-19 vaccination status - United States, April 2021-988 September 2022. MMWR Morb. Mortal. Wkly. Rep. 72, 601–605. 989 10.15585/mmwr.mm7222a3. 990 85. Lutrick, K., Ellingson, K.D., Baccam, Z., Rivers, P., Beitel, S., Parker, J., Hollister,991 J., Sun, X., Gerald, J.K., Komatsu, K., et al. (2021). COVID-19 infection, reinfection, 992 48 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint and vaccine effectiveness in a prospective cohort of Arizona frontline/essential 993 workers: The AZ HEROES research protocol. JMIR Res. Protoc. 10, e28925. 994 10.2196/28925. 995 86. Edwards, L.J., Fowlkes, A.L., Wesley, M.G., Kuntz, J.L., Odean, M.J., Caban-996 Martinez, A.J., Dunnigan, K., Phillips, A.L., Grant, L., Herring, M.K., et al. (2021). 997 “Research on the Epidemiology of SARS-CoV-2 in Essential Response Personnel 998 (RECOVER) Study: Protocol for a multi-site longitudinal cohort.” JMIR Res Protoc 999 7. 1000 87. Rambaut, A., Holmes, E.C., O’Toole, Á., Hill, V., McCrone, J.T., Ruis, C., du1001 Plessis, L., and Pybus, O.G. (2020). A dynamic nomenclature proposal for SARS-1002 CoV-2 lineages to assist genomic epidemiology. Nat. Microbiol. 5, 1403–1407. 1003 10.1038/s41564-020-0770-5. 1004 88. Ripperger, T.J., Uhrlaub, J.L., Watanabe, M., Wong, R., Castaneda, Y., Pizzato,1005 H.A., Thompson, M.R., Bradshaw, C., Weinkauf, C.C., Bime, C., et al. (2020).1006 Orthogonal SARS-CoV-2 Serological Assays Enable Surveillance of Low-1007 Prevalence Communities and Reveal Durable Humoral Immunity. Immunity 53, 1008 925-933.e4. 10.1016/j.immuni.2020.10.004.1009 89. Piccoli, L., Park, Y.-J., Tortorici, M.A., Czudnochowski, N., Walls, A.C., Beltramello,1010 M., Silacci-Fregni, C., Pinto, D., Rosen, L.E., Bowen, J.E., et al. (2020). Mapping 1011 Neutralizing and Immunodominant Sites on the SARS-CoV-2 Spike Receptor-1012 Binding Domain by Structure-Guided High-Resolution Serology. Cell 183, 1024-1013 1042.e21. 10.1016/j.cell.2020.09.037. 1014 49 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint 90. Greaney, A.J., Loes, A.N., Crawford, K.H.D., Starr, T.N., Malone, K.D., Chu, H.Y.,1015 and Bloom, J.D. (2021). Comprehensive mapping of mutations in the SARS-CoV-2 1016 receptor-binding domain that affect recognition by polyclonal human plasma 1017 antibodies. Cell Host Microbe 29, 463-476.e6. 10.1016/j.chom.2021.02.003. 1018 91. Röltgen, K., Nielsen, S.C.A., Silva, O., Younes, S.F., Zaslavsky, M., Costales, C.,1019 Yang, F., Wirz, O.F., Solis, D., Hoh, R.A., et al. (2022). Immune imprinting, breadth 1020 of variant recognition, and germinal center response in human SARS-CoV-2 1021 infection and vaccination. Cell 185, 1025-1040.e14. 10.1016/j.cell.2022.01.018. 1022 92. Suryadevara, N., Shrihari, S., Gilchuk, P., VanBlargan, L.A., Binshtein, E., Zost,1023 S.J., Nargi, R.S., Sutton, R.E., Winkler, E.S., Chen, E.C., et al. (2021). Neutralizing1024 and protective human monoclonal antibodies recognizing the N-terminal domain of 1025 the SARS-CoV-2 spike protein. Cell 184, 2316-2331.e15. 1026 10.1016/j.cell.2021.03.029. 1027 93. Cerutti, G., Guo, Y., Zhou, T., Gorman, J., Lee, M., Rapp, M., Reddem, E.R., Yu, J.,1028 Bahna, F., Bimela, J., et al. (2021). Potent SARS-CoV-2 neutralizing antibodies 1029 directed against spike N-terminal domain target a single supersite. Cell Host 1030 Microbe 29, 819-833.e7. 10.1016/j.chom.2021.03.005. 1031 94. McCallum, M., De Marco, A., Lempp, F.A., Tortorici, M.A., Pinto, D., Walls, A.C.,1032 Beltramello, M., Chen, A., Liu, Z., Zatta, F., et al. (2021). N-terminal domain 1033 antigenic mapping reveals a site of vulnerability for SARS-CoV-2. Cell 184, 2332-1034 2347.e16. 10.1016/j.cell.2021.03.028. 1035 50 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint 95. Shroff, R.T., Chalasani, P., Wei, R., Pennington, D., Quirk, G., Schoenle, M.V.,1036 Peyton, K.L., Uhrlaub, J.L., Ripperger, T.J., Jergović, M., et al. (2021). Immune 1037 responses to two and three doses of the BNT162b2 mRNA vaccine in adults with 1038 solid tumors. Nat. Med., 1–10. 10.1038/s41591-021-01542-z. 1039 96. Motozono, C., Toyoda, M., Zahradnik, J., Saito, A., Nasser, H., Tan, T.S., Ngare, I.,1040 Kimura, I., Uriu, K., Kosugi, Y., et al. (2021). SARS-CoV-2 spike L452R variant 1041 evades cellular immunity and increases infectivity. Cell Host Microbe. 1042 10.1016/j.chom.2021.06.006. 1043 97. He, P., Liu, B., Gao, X., Yan, Q., Pei, R., Sun, J., Chen, Q., Hou, R., Li, Z., Zhang,1044 Y., et al. (2022). SARS-CoV-2 Delta and Omicron variants evade population 1045 antibody response by mutations in a single spike epitope. Nat Microbiol. 1046 10.1038/s41564-022-01235-4. 1047 98. Tchesnokova, V., Kulasekara, H., Larson, L., Bowers, V., Rechkina, E., Kisiela, D.,1048 Sledneva, Y., Choudhury, D., Maslova, I., Deng, K., et al. (2021). Acquisition of the 1049 L452R mutation in the ACE2-binding interface of spike protein triggers recent 1050 massive expansion of SARS-CoV-2 variants. J. Clin. Microbiol. 59, e0092121. 1051 10.1128/JCM.00921-21. 1052 99. Greaney, A.J., Starr, T.N., Eguia, R.T., Loes, A.N., Khan, K., Karim, F., Cele, S.,1053 Bowen, J.E., Logue, J.K., Corti, D., et al. (2022). A SARS-CoV-2 variant elicits an 1054 antibody response with a shifted immunodominance hierarchy. PLoS Pathog. 18, 1055 e1010248. 10.1371/journal.ppat.1010248. 1056 51 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint 100. Setliff, I., Shiakolas, A.R., Pilewski, K.A., Murji, A.A., Mapengo, R.E., Janowska,1057 K., Richardson, S., Oosthuysen, C., Raju, N., Ronsard, L., et al. (2019). High-1058 Throughput Mapping of B Cell Receptor Sequences to Antigen Specificity. Cell 179, 1059 1636-1646.e15. 10.1016/j.cell.2019.11.003. 1060 101. Schiepers, A., van ’t Wout, M.F.L., Greaney, A.J., Zang, T., Muramatsu, H., Lin,1061 P.J.C., Tam, Y.K., Mesin, L., Starr, T.N., Bieniasz, P.D., et al. (2023). Molecular 1062 fate-mapping of serum antibody responses to repeat immunization. Nature. 1063 10.1038/s41586-023-05715-3. 1064 102. Smith, K.G., Light, A., Nossal, G.J., and Tarlinton, D.M. (1997). The extent of1065 affinity maturation differs between the memory and antibody-forming cell 1066 compartments in the primary immune response. EMBO J. 16, 2996–3006. 1067 10.1093/emboj/16.11.2996. 1068 103. Lavinder, J.J., Wine, Y., Giesecke, C., Ippolito, G.C., Horton, A.P., Lungu, O.I.,1069 Hoi, K.H., DeKosky, B.J., Murrin, E.M., Wirth, M.M., et al. (2014). Identification and 1070 characterization of the constituent human serum antibodies elicited by vaccination. 1071 Proc. Natl. Acad. Sci. U. S. A. 111, 2259–2264. 10.1073/pnas.1317793111. 1072 104. Pape, K.A., Maul, R.W., Dileepan, T., Paustian, A.S., Gearhart, P.J., and1073 Jenkins, M.K. (2018). Naive B cells with high-avidity germline-encoded antigen 1074 receptors produce persistent IgM+ and transient IgG+ memory B cells. Immunity 48, 1075 1135-1143.e4. 10.1016/j.immuni.2018.04.019. 1076 52 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint 105. Andrews, S.F., Kaur, K., Pauli, N.T., Huang, M., Huang, Y., and Wilson, P.C.1077 (2015). High preexisting serological antibody levels correlate with diversification of 1078 the influenza vaccine response. J. Virol. 89, 3308–3317. 10.1128/JVI.02871-14. 1079 106. Inoue, T., Shinnakasu, R., Kawai, C., Yamamoto, H., Sakakibara, S., Ono, C.,1080 Itoh, Y., Terooatea, T., Yamashita, K., Okamoto, T., et al. (2023). Antibody 1081 feedback contributes to facilitating the development of Omicron-reactive memory B 1082 cells in SARS-CoV-2 mRNA vaccinees. J. Exp. Med. 220. 10.1084/jem.20221786. 1083 107. Liu, Y.J., Zhang, J., Lane, P.J., Chan, E.Y., and MacLennan, I.C. (1991). Sites of1084 specific B cell activation in primary and secondary responses to T cell-dependent 1085 and T cell-independent antigens. Eur. J. Immunol. 21, 2951–2962. 1086 10.1002/eji.1830211209. 1087 108. Abbott, R.K., and Crotty, S. (2020). Factors in B cell competition and1088 immunodominance. Immunol. Rev. 296, 120–131. 10.1111/imr.12861. 1089 109. Krammer, F. (2019). The human antibody response to influenza A virus infection1090 and vaccination. Nat. Rev. Immunol. 19, 383–397. 10.1038/s41577-019-0143-6. 1091 110. Chia, P.Y., Ong, S.W.X., Chiew, C.J., Ang, L.W., Chavatte, J.-M., Mak, T.-M.,1092 Cui, L., Kalimuddin, S., Chia, W.N., Tan, C.W., et al. (2022). Virological and 1093 serological kinetics of SARS-CoV-2 Delta variant vaccine breakthrough infections: a 1094 multicentre cohort study. Clin. Microbiol. Infect. 28, 612.e1-612.e7. 1095 10.1016/j.cmi.2021.11.010. 1096 53 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint 111. Schaefer-Babajew, D., Wang, Z., Muecksch, F., Cho, A., Loewe, M., Cipolla, M., 1097 Raspe, R., Johnson, B., Canis, M., DaSilva, J., et al. (2023). Antibody feedback 1098 regulates immune memory after SARS-CoV-2 mRNA vaccination. Nature 613, 735–1099 742. 10.1038/s41586-022-05609-w. 1100 112. Gao, Y., Cai, C., Grifoni, A., Müller, T.R., Niessl, J., Olofsson, A., Humbert, M., 1101 Hansson, L., Österborg, A., Bergman, P., et al. (2022). Ancestral SARS-CoV-2-1102 specific T cells cross-recognize the Omicron variant. Nat. Med. 28, 472–476. 1103 10.1038/s41591-022-01700-x. 1104 113. Keeton, R., Tincho, M.B., Ngomti, A., Baguma, R., Benede, N., Suzuki, A., Khan, 1105 K., Cele, S., Bernstein, M., Karim, F., et al. (2022). T cell responses to SARS-CoV-2 1106 spike cross-recognize Omicron. Nature 603, 488–492. 10.1038/s41586-022-04460-1107 3. 1108 114. Bartsch, Y.C., Wang, C., Zohar, T., Fischinger, S., Atyeo, C., Burke, J.S., Kang, 1109 J., Edlow, A.G., Fasano, A., Baden, L.R., et al. (2021). Humoral signatures of 1110 protective and pathological SARS-CoV-2 infection in children. Nat. Med. 27, 454–1111 462. 10.1038/s41591-021-01263-3. 1112 115. Ying, B., Scheaffer, S.M., Whitener, B., Liang, C.-Y., Dmytrenko, O., Mackin, S., 1113 Wu, K., Lee, D., Avena, L.E., Chong, Z., et al. (2022). Boosting with Omicron-1114 matched or historical mRNA vaccines increases neutralizing antibody responses 1115 and protection against B.1.1.529 infection in mice. bioRxivorg, 2022.02.07.479419. 1116 10.1101/2022.02.07.479419. 1117 54 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint 116. Barnes, C.O., West, A.P., Jr, Huey-Tubman, K.E., Hoffmann, M.A.G., Sharaf,1118 N.G., Hoffman, P.R., Koranda, N., Gristick, H.B., Gaebler, C., Muecksch, F., et al.1119 (2020). Structures of Human Antibodies Bound to SARS-CoV-2 Spike Reveal 1120 Common Epitopes and Recurrent Features of Antibodies. Cell 182, 828-842.e16. 1121 10.1016/j.cell.2020.06.025. 1122 117. Hay, J.A., Kissler, S.M., Fauver, J.R., Mack, C., Tai, C.G., Samant, R.M.,1123 Connolly, S., Anderson, D.J., Khullar, G., MacKay, M., et al. (2022). Quantifying the 1124 impact of immune history and variant on SARS-CoV-2 viral kinetics and infection 1125 rebound: A retrospective cohort study. Elife 11. 10.7554/eLife.81849. 1126 118. Goldfarb, D.M., Tilley, P., Al-Rawahi, G.N., Srigley, J.A., Ford, G., Pedersen, H.,1127 Pabbi, A., Hannam-Clark, S., Charles, M., Dittrick, M., et al. (2020). Self-collected 1128 saline gargle samples as an alternative to healthcare worker collected 1129 nasopharyngeal swabs for COVID-19 diagnosis in outpatients. J. Clin. Microbiol., 1130 2020.09.13.20188334. 10.1101/2020.09.13.20188334. 1131 119. Hao, Y., Hao, S., Andersen-Nissen, E., Mauck, W.M., 3rd, Zheng, S., Butler, A.,1132 Lee, M.J., Wilk, A.J., Darby, C., Zager, M., et al. (2021). Integrated analysis of 1133 multimodal single-cell data. Cell 184, 3573-3587.e29. 10.1016/j.cell.2021.04.048. 1134 55 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Figure 1 56for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Figure 1. Primary and recall antibody responses to Wuhan and Delta strains of 1135 SARS-COV-2. (A) Virus neutralization assays were performed using the WA-1 and 1136 Delta isolates of SARS-CoV-2. Serial 1:3 dilutions of serums were performed and tested 1137 for the ability to prevent plaque formation on Vero cells. The lowest concentration 1138 capable of preventing more than 90% of plaques was considered to be the PRNT90 1139 value. Each symbol represents an individual. Two-sided P values from t-test statistics 1140 were calculated for pairwise differences using two-way ANOVA. Post hoc testing for 1141 multiple comparisons between draws was performed using Tukey’s multiple 1142 comparisons test. P values greater than 0.05 are not depicted. (B) Quantitative titers of 1143 WuHu1- and Delta RBD-specific antibodies. Serum was initially diluted 1:60, serially 1144 diluted 1:3, assessed by ELISA for binding to the listed antigens, and area under the 1145 curve (AUC) values were calculated. Each symbol represents an individual. WuHu1 1146 AUC values were divided by their Delta AUC titer in the same individual to calculate a 1147 WuHu1:Delta RBD ratio in the rightmost panel. Two-sided P values from t-test statistics 1148 were calculated for pairwise differences using one-way ANOVA. Post hoc testing for 1149 multiple comparisons between draws was performed using Tukey’s multiple 1150 comparisons test. P values greater than 0.05 are not depicted. 1151 1152 1153 57 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Figure 2 58for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Figure 2. WuHu1 and Delta Memory B cell flow cytometric analysis and 1154 quantification. (A) Representative flow cytometric plots of Wuhu1 and Delta S1-1155 specific memory B cells (full gating strategy shown in Figure S2) in naïve, primary Delta 1156 infection, and post-vaccination Delta infection cohorts. Cells that bind both WuHu1 S1 1157 and Delta S1 are annotated as cross-reactive S1+, whereas cells that bind only WuHu1 1158 S1 or Delta S1 are annotated as WuHu1 S1+ or Delta S1+, respectively. (B) 1159 Quantification of isotype-switched memory B cells as a percentage of total PBMCs for 1160 Wuhu1 S1+, Delta S1+ and cross-reactive S1+ specificities for each cohort of SARS-1161 CoV-2 immune histories. Each symbol represents an individual. Two-sided P values 1162 from t-test statistics were calculated for pairwise differences using one-way ANOVA. 1163 Post hoc testing for multiple comparisons between draws was performed using Tukey’s 1164 multiple comparisons test. P values greater than 0.05 are not depicted. (C) Correlation 1165 of post-infection cross-reactive S1 MBCs (calculated as in Figure 2B) plotted against the 1166 frequency of post-infection Delta S1-specific MBCs (calculated as in Figure 2B) in 1167 individuals that experienced a post-vaccination Delta infection. Pearson correlation 1168 analysis was performed. 1169 1170 1171 59 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Figure 3 60for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Figure 3. Epitope-specific quantification of Delta RBD- and Delta NTD-specific 1172 antibodies and memory B cells. (A) A chimeric protein (Delta RBD-L452) was 1173 generated in which R452 was reverted to the ancestral L452. ELISAs were used to 1174 quantify serum antibodies that bound to Delta RBD-L452 in each cohort. Delta RBD-1175 L452 AUC titers were divided by Delta RBD titers (Figure 1B) in the same individuals to 1176 calculate a L452:R452 titer ratio. Each symbol represents an individual. Two-sided P 1177 values from t-test statistics were calculated for pairwise differences using one-way 1178 ANOVA. Post hoc testing for multiple comparisons between draws was performed using 1179 Tukey’s multiple comparisons test. P values greater than 0.05 are not depicted. (B) A 1180 chimeric protein (Delta NTD-WuHu1 S1) was generated in which Delta NTD mutated 1181 epitopes (T19R, G142D, E156-, F157-, R158G) were incorporated into the otherwise 1182 WuHu1 S1 backbone. ELISAs were used to quantify serum antibodies that bound to 1183 Delta NTD-WuHu1 S1 in each cohort. Delta RBD-L452 AUC titers were divided by their 1184 Delta RBD (Supplemental Fig 1A) titer to calculate a WuHu1 NTD:Delta NTD titer ratio. 1185 Each symbol represents an individual. Two-sided P values from t-test statistics were 1186 calculated for pairwise differences using one-way ANOVA. Post hoc testing for multiple 1187 comparisons between draws was performed using Tukey’s multiple comparisons test. P 1188 values greater than 0.05 are not depicted. (C) LIBRA-seq plots of isotype-switched 1189 memory B cells enriched for Spike-binding specificities from primary Delta infections. 1190 Read count thresholds to determine positivity were set using samples in which cells 1191 lacking Spike-binding specificities were sorted and sequenced. Plots are concatenated 1192 from ten individuals. (D) Quantification of Delta RBD-specific and Delta NTD-specific 1193 memory B cells (MBCs) in individuals that experienced a primary Delta infection. Lines 1194 61 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint connect specificities within the same individual. Delta RBD-specific cells were classified 1195 by cells that had Delta RBD read counts of greater than 160 and WuHu1 S1 read 1196 counts of less than 35. Delta NTD-specific cells were classified by cells that had Delta 1197 NTD-WuHu1 S1 read counts of greater than 23 and WuHu1 S1 read counts of less than 1198 35. Two-sided P values were calculated for pairwise differences using paired t-tests.1199 1200 1201 1202 1203 1204 62 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Figure 4 63for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Figure 4. Primary and recall antibody responses to Wuhan and BA.1 strains of 1205 SARS-COV-2. (A) Virus neutralization assays were performed using the WA-1 and 1206 BA.1 isolates of SARS-CoV-2. Serial 1:3 dilutions of serums were performed and tested 1207 for the ability to prevent plaque formation on Vero cells. The lowest concentration 1208 capable of preventing more than 90% of plaques was considered to the PRNT90 value. 1209 Each symbol represents an individual. Two-sided P values from t-test statistics were 1210 calculated for pairwise differences using two-way ANOVA. Post hoc testing for multiple 1211 comparisons between draws was performed using Tukey’s multiple comparisons test. P 1212 values greater than 0.05 are not depicted. (B) Quantitative titers of Wuhu1 and BA.1 1213 RBD antibodies. Serum was initially diluted 1:60, serially diluted 1:3, assessed by 1214 ELISA for binding to the listed antigens, and area under the curve (AUC) values were 1215 calculated. Each symbol represents an individual. WuHu1 AUC values were divided by 1216 their BA.1 RBD AUC titer in the same individual to calculate a ratio in the rightmost 1217 panel. Two-sided P values from t-test statistics were calculated for pairwise differences 1218 using one-way ANOVA. Post hoc testing for multiple comparisons between draws was 1219 performed using Tukey’s multiple comparisons test. P values greater than 0.05 are not 1220 depicted. 1221 1222 1223 64 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Figure 5 65for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Figure 5. WuHu1 and BA.1 Memory B cell flow cytometric analysis and 1224 quantification. (A) Representative flow cytometric plots of Wuhu1 and BA.1 RBD-1225 specific memory B cells (full gating strategy shown in Figure S3) in naïve, vaccinated 1226 only, primary BA.1 infection, and post-vaccination BA.1 infection cohorts. Cells that bind 1227 both WuHu1 RBD and BA.1 RBD are annotated as cross-reactive RBD+, whereas cells 1228 that bind only WuHu1 RBD or BA.1 RBD are annotated as WuHu1 RBD+ or BA.1 1229 RBD+, respectively. (B) Quantification of isotype-switched memory B cells for Wuhu1 1230 RBD+, BA.1 RBD+ and cross-reactive RBD+ specificities for each cohort of SARS-CoV-1231 2 immune histories. Each symbol represents an individual. Two-sided P values from t-1232 test statistics were calculated for pairwise differences using one-way ANOVA. Post hoc 1233 testing for multiple comparisons between draws was performed using Tukey’s multiple 1234 comparisons test. P values greater than 0.05 are not depicted. (C) Representative flow 1235 cytometric plots of Wuhu1 and BA.1 Spike-specific memory B cells (full gating strategy 1236 shown in Figure S3) in naïve, vaccinated only, primary BA.1 infection, and post-1237 vaccination BA.1 infection cohorts. Cells that bind both WuHu1 RBD and BA.1 Spike 1238 are annotated as cross-reactive Spike+, whereas cells that bind only WuHu1 Spike or 1239 BA.1 Spike are annotated as WuHu1 Spike+ or BA.1 Spike+, respectively. (D) 1240 Quantification of isotype-switched memory B cells for Wuhu1 Spike+, BA.1 Spike+ and 1241 cross-reactive Spike+ specificities for each cohort of SARS-CoV-2 immune histories. 1242 Each symbol represents an individual. Two-sided P values from t-test statistics were 1243 calculated for pairwise differences using one-way ANOVA. Post hoc testing for multiple 1244 comparisons between draws was performed using Tukey’s multiple comparisons test. P 1245 values greater than 0.05 are not depicted. 1246 66 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Figure 6 67for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Figure 6. Frequency of WuHu1- and BA.1-specific memory B cells before and after 1247 BA.1 infection. (A) Frequencies of isotype-switched memory B cells with Wuhu1 1248 RBD+, BA.1 RBD+ and cross-reactive RBD+ specificities in both unvaccinated and 1249 vaccinated individuals before and after BA.1 infection. Lines connect the same 1250 individual from pre-infection frequency to post-infection frequency. In primary infections, 1251 pre-infection blood draws were taken on average 75.6 days before infection and post-1252 infection blood draws occurred on 37.8 days after infection. In post-vaccination 1253 infections, pre-infection blood draws were taken on average 87.6 days before infection 1254 and post-infection draws were taken an average of 38.3 days after infection. Individuals 1255 that received a vaccine after the pre-infection draw were excluded from analysis. P 1256 values were calculated using Wilcoxon matched-pairs signed rank test on each row and 1257 post hoc testing for multiple comparisons between draws was performed using two-1258 stage linear step-up procedure of Benjamini, Krieger and Yekutieli. P values greater 1259 than 0.05 are not depicted. (B) Frequencies of isotype-switched memory B cells with 1260 Wuhu1 Spike+, BA.1 Spike+ and cross-reactive Spike+ specificities in both 1261 unvaccinated and vaccinated individuals before and after BA.1 infection. Lines connect 1262 the same individual from pre-infection frequency to post-infection frequency. P values 1263 were calculated using Wilcoxon matched-pairs signed rank test on each row and post 1264 hoc testing for multiple comparisons between draws was performed using two-stage 1265 linear step-up procedure of Benjamini, Krieger and Yekutieli. P values greater than 0.05 1266 are not depicted. 1267 1268 1269 68 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Figure 7 69for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Figure 7. Correlations of pre-infection and post-infection BA.1-specific antibody, 1270 T and B cell responses. (A) Correlation of pre-infection cross-reactive Spike MBCs 1271 (calculated as in Figure 5C) plotted against the frequency of post-infection BA.1 Spike 1272 MBCs (calculated as in Figure 5C) in individuals that experienced a post-vaccination 1273 BA.1 infection. Pearson correlation analysis was performed. Pre-infection blood draws 1274 were taken on average 87.6 days before infection and post-infection draws were taken 1275 an average of 38.3 days after infection. Individuals that received a vaccine after the pre-1276 infection draw were excluded from analysis. (B) Correlation of post-infection cross-1277 reactive Spike MBCs (calculated as in Figure 6B) plotted against the frequency of post-1278 infection BA.1 Spike MBCs (calculated as in Figure 5C) in individuals that experienced a 1279 post-vaccination BA.1 infection. Pearson correlation analysis was performed. (C) 1280 Correlation of pre-infection BA.1 neutralizing antibody titer (calculated as in Figure 4a) 1281 plotted against post infection BA.1 Spike MBCs (calculated as in Figure 5c) in 1282 individuals that experienced a post-vaccination BA.1 infection. Pearson correlation 1283 analysis was performed. (D) Correlation of pre-infection BA.1 Spike-specific T cells as 1284 measured by IFNg ELISPOTs plotted against post-infection BA.1 Spike MBCs in 1285 individuals that experienced a post-vaccination BA.1 infection. Pearson correlation 1286 analysis was performed. 1287 1288 1289 70 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Figure S1 71for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Figure S1. Test All, Test Smart (TATS) symptom report. (A) Percentage of 1290 individuals from each TATS cohort that reported experiencing various respiratory/cold 1291 symptoms in study entry survey. (B) Reported days until symptoms resolved for each 1292 TATS cohort. Two-sided P values from t-test statistics were calculated for pairwise 1293 differences using one-way ANOVA. Post hoc testing for multiple comparisons between 1294 draws was performed using Tukey’s multiple comparisons test. P values greater than 1295 0.05 are not depicted. 1296 1297 72 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Figure S2 73for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Figure S2. PANGO-lineage assignments from TATS PCR positive individuals. (A) 1298 Delta or BA.1 PANGO-lineage assignments after SARS-CoV-2 viral amplicon 1299 sequencing (Integrated DNA Technologies). Unassigned sequences could not be 1300 assigned to a PANGO-lineage due to insufficient viral RNA recovery and low sequence 1301 coverage. (B) PANGO-lineage assignments of all TATS samples submitted during the 1302 period of Delta cohort recruitment, July 1, 2021-December 1, 2021 (left panel) or during 1303 the period of BA.1 cohort recruitment, January 1, 2022-March 31, 2022 (right panel). 1304 Unassigned sequences could not be assigned a lineage due to insufficient viral RNA 1305 recovery and low sequence coverage. 1306 74 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Figure S3 75for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Figure S3. Primary and recall antibody responses to Wuhan and Delta strains of 1307 SARS-COV-2. Quantitative titers of WuHu1- and Delta S1-specific antibodies. Serum 1308 was initially diluted 1:60, serially diluted 1:3, assessed by ELISA for binding to the listed 1309 antigens, and area under the curve (AUC) values were calculated. Each symbol 1310 represents an individual. WuHu1 AUC values were divided by their Delta AUC titer in 1311 the same individual to calculate a WuHu1:Delta S1 ratio in the rightmost panel. Two-1312 sided P values from t-test statistics were calculated for pairwise differences using one-1313 way ANOVA. Post hoc testing for multiple comparisons between draws was performed 1314 using Tukey’s multiple comparisons test. P values greater than 0.05 are not depicted. 1315 1316 1317 1318 1319 76 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Figure S4 primary Delta post-vaccination Delta 77 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Figure S4. Flow cytometric gating strategy with Delta S1 and WuHu1 S1 1320 tetramers. Examples of a sample from a primary Delta infection (top) and post-1321 vaccination Delta infection (bottom) are shown. 1322 1323 78 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Figure S5 79for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Figures S5. LIBRA-seq analysis in primary and post-vaccination Delta infections 1324 and quantification of somatic mutations. (A) A chimeric protein (Delta NTD-WuHu1 1325 S1) was generated in which Delta NTD mutated epitopes (T19R, G142D, E156-, F157-, 1326 R158G) were incorporated into the otherwise WuHu1 S1 backbone. Quantification of 1327 Delta RBD-specific (left) and Delta NTD-specific memory B cells (right) in individuals 1328 that experienced a post-vaccination Delta infection. Delta RBD-specific cells were 1329 classified by cells that had Delta RBD read counts of greater than 300 and WuHu1 S1 1330 read counts of less than 35. Delta NTD-specific cells were classified by cells that had 1331 Delta NTD-WuHu1 S1 read counts of greater than 23 and WuHu1 S1 read counts of 1332 less than 35. Read count thresholds to determine positivity were set using samples in 1333 which cells lacking Spike-binding specificities were sorted and sequenced. Plots are 1334 concatenated from ten individuals. (B) Somatic mutations were calculated using the 1335 observedMutations command in the Shazam Immcantation package in R. Specificities 1336 of cells are determined using the same cutoffs described in Figure S3A and 3D. (C) 1337 Quantification of somatic mutations of all Spike specific cells subjected to scRNAseq 1338 from either ten primary or post-vaccination Delta infections. 1339 1340 1341 1342 1343 1344 80 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Figure S6 uninfected post-infection 0.0 0.5 1.0 1.5OD450 ⍺-Nucleocapsid 1:60 Titer 81 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Figure S6. anti-Nucleocapsid titers in uninfected individuals. Individuals with ⍺-1345 Nucleocapsid titers of greater than 0.6 at a1:60 serum dilution were considered 1346 previously infected and excluded from the study. 1347 1348 1349 82 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Figure S7 83for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Figure S7. Flow cytometric gating strategy with BA.1 RBD, BA.1 Spike, WuHu1 1350 RBD and WuHu1 Spike tetramers. An example of a sample from a post-vaccination 1351 BA.1 infection is shown. 1352 1353 1354 84 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Figure S8 85for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Figure S8. WuHu1 and BA.1 Memory B cell flow cytometric quantification. (A) 1355 Cells that bind both WuHu1 RBD and BA.1 RBD are annotated as cross-reactive RBD+, 1356 whereas cells that bind only WuHu1 RBD or BA.1 RBD are annotated as WuHu1 RBD+ 1357 or BA.1 RBD+, respectively. Quantification of isotype-switched memory B cells as a 1358 percentage of total PBMCs for Wuhu1 RBD+, BA.1 RBD+ and cross-reactive RBD+ 1359 specificities for each cohort of SARS-CoV-2 immune histories. Each symbol represents 1360 an individual. Two-sided P values from t-test statistics were calculated for pairwise 1361 differences using two-way ANOVA. Post hoc testing for multiple comparisons between 1362 draws was performed using Tukey’s multiple comparisons test. P values greater than 1363 0.05 are not depicted. (B) Cells that bind both WuHu1 RBD and BA.1 Spike are 1364 annotated as cross-reactive Spike+, whereas cells that bind only WuHu1 Spike or BA.1 1365 Spike are annotated as WuHu1 Spike+ or BA.1 Spike+, respectively. Quantification of 1366 isotype-switched memory B cells for Wuhu1 Spike+, BA.1 Spike+ and cross-reactive 1367 Spike+ specificities for each cohort of SARS-CoV-2 immune histories. Each symbol 1368 represents an individual. Two-sided P values from t-test statistics were calculated for 1369 pairwise differences using two-way ANOVA. Post hoc testing for multiple comparisons 1370 between draws was performed using Tukey’s multiple comparisons test. P values 1371 greater than 0.05 are not depicted. 1372 1373 86 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Table 1 vaccinated only (statewide antibody testing initiative) (n=74) primary Delta (n=12) post-vaccination Delta (n=37) vaccinated only (TATS) (n=62) primary BA.1 (n=69) post- vaccination BA.1 (n=62) Age 38.0 (32.0, 54.0) 21.9 (20.2, 40.7) 23.3 (18.6, 65.8) 31.9 (18.6, 65.0) 44 (25, 62) 40 (19, 71.5) Mean (s.d.) Sex Male 22 (32%) 5 (42%) 8 (22%) 24 (39%) 24 (39%) Female 52 (68%) 7 (58%) 29 (78%) 37 (60%) 34 (55%) Prior COVID infection Yes 12 (100%) 37 (100%) 69 (100%) 64 (100%) No 74 (100%) Time since COVID infection 67.5 days (32, 99.3) 71.5 days (48.5, 89.5) 40 days (31, 44.5) 54.4 days (34.5, 71) paired pre- and post- infection samples 10 21 Time from vaccination to pre- infection draw 138 (32.5, 217) Time from pre- infection draw to infection 73.3 days (30, 99) 112 days (33, 187) Time from infection to post- infection draw 42.7 days (29.5, 47.5) 44.8 days (34, 49.3) COVID Vaccination 37 0 37 62 0 62 # of shots 2 37 (100%) 37 (100%) 16 (26%) 13 (20%) 3 46 (74%) 15 (23%) 4 3 (5%) time since vaccination 135.6 days (126, 270) 273.9 days (56, 317) 176.0 days (113.5, 188) 192.2 days (107, 302.3) 87 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint Table 1. Characteristics of cohorts 1374 Interquartile range (IQR) is listed in parentheses unless otherwise stated in the table. 1375 1376 88 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for thisthis version posted September 14, 2023. ; https://doi.org/10.1101/2023.09.12.23295384doi: medRxiv preprint