Identification of differences in the magnitude and specificity of SARS-CoV-2 nucleocapsid antibody responses in naturally infected and vaccinated individuals

23

As there are limited data on B cell epitopes for the nucleocapsid protein in 24 SARS-CoV-2, we sought to identify the immunodominant regions within the N protein, 25 recognized by patients with varying severity of natural infection with the Wuhan strain (WT), 26 delta, omicron and in those who received the Sinopharm vaccines, which is an inactivated, 27 whole virus vaccine. 28

Using overlapping peptides representing the N protein, with an in-house ELISA, 29 we mapped the immunodominant regions within the N protein, in seronegative (n=30), WT 30 infected (n=30), delta infected (n=30), omicron infected+vaccinated (n=20) and Sinopharm 31 (BBIBP-CorV) vaccinees (n=30). We then investigated the sensitivity and specificity of these 32 immunodominant regions and analysed their conservation with other SARS-CoV-2 variants 33 of concern, seasonal human coronaviruses and bat Sarbecoviruses. We then investigated the 34 kinetics of responses to these regions in those with varying severity of acute COVID-19. 35

We identified four immunodominant regions aa 29-52, aa 155-178, aa 274 to 297 36 and aa 365 to 388, were highly conserved within SARS-CoV-2 and the bat coronaviruses. 37 The magnitude of responses to these regions varied based on the infecting SARS-CoV-2 38 variants, with WT infected individuals predominantly recognizing aa155 to 178 regions, delta 39 infected individuals and vaccinated+omicron infected individuals predominantly recognizing 40 regions aa 29 to 52 and aa 274 to 294 regions. Sinopharm vaccinees recognized all four 41 regions, with the magnitude of responses significantly lower than other groups. >80% of 42 individuals gave responses above the positive cut-off threshold to many of the four regions, 43 with some differences with individuals who were infected with different VoCs. These regions 44 were found to be 100% specific, as none of the seronegative individuals gave any responses. 45 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 3

N-protein specific responses appear to be detectable in over 90% of those who 46 were naturally infected or vaccinated with a whole virus inactivated vaccine, with responses 47 mainly directed against four regions of the protein, which were highly conserved. As these 48 regions were highly specific with high sensitivity, they have a potential to be used to develop 49 diagnostic assays and to be used in development of vaccines. 50 51

SARS-COv-2; nucleocapsid protein; ELISA; overlapping peptides; 52 immunodominant; conservation; sarbecoviruses; variants; Sinopharm 53 54 55 56 57 58 59 60 61 62 63 64 65 66 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 4

67 The SARS-CoV-2 virus continues to evolve, giving rise to more immune evasive and more 68 transmissible variants, which continue to drive outbreaks globally [21]. Although variants 69 such as omicron (BA.1) were thought to initially cause milder illness, the sub-lineages that 70 subsequently emerged such as BA.2, were associated with more severe disease in certain 71 populations [30]. In fact, BA.2 outbreaks in the United States and in Hong Kong resulted in 72 several fold higher mortality rates than seen during the delta outbreaks in many countries 73 [13]. Many factors could contribute to the differences in mortality rates and hospitalization 74 rates during different outbreaks in different countries such as co-morbidities, age, vaccination 75 rates of a population, the proportion of individuals naturally infected, COVID-19 control 76 measures, better treatment modalities, infra-structure to manage hospitalized patients and 77 seasonal changes [2,5,18]. Among all the factors that have contributed to a reduction in 78 mortality rates, COVID-19 vaccines, are likely to be one of the single most important factors 79 that were responsible for this reduction [2,29]. 80 81 While neutralization antibodies (Nabs) have shown to associate with protection against 82 severe disease when infected with the SARS-CoV-2 [1,12], the mRNA COVID-19 vaccines 83 appear to induce higher levels of Nabs compared to other vaccines [15]. However, there is 84 emerging evidence that nucleocapsid (N) protein specific antibody responses may be 85 protective based on data in animal models [8]. Indeed, a high frequency of polyfunctional T 86 cell responses specific for certain epitopes within the N protein was found to associate with 87 milder illness [22] and N protein specific antibody responses were detected earlier in 88 infection and were present at detectable levels in a larger proportion of individuals compared 89 to spike protein specific antibody responses [6]. 90 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 5 The N protein is one of the most abundant, highly conserved RNA-binding proteins, which 91 plays an important role in the packing of the SARS-CoV-2 genome [3]. It plays an important 92 role in the regulation of the virus replication cycle, inhibits interferon response and induced 93 apoptosis [3]. The N protein, which spans 419 amino acids, consists of five domains and all 94 five have shown to bind to RNA [6]. The region starting from the 388 amino acid position 95 was found to induce a high frequency of immune responses in patients with acute COVID-19 96 (from the Wuhan strain) and was found to be 100% specific to detect infection with SARS-97 CoV-2. Although the N protein is an important T cell and antibody target, the main 98 immunodominant regions within this protein, targeted by antibodies has not been extensively 99 studied. For instance, although the N protein is highly conserved, as it is an important 100 antibody target, certain mutations in SARS-CoV-2 variants of concern (VoC), can give rise to 101 differences in the magnitude of antibody responses to certain regions. Therefore, we sought 102 to identify the immunodominant regions within the N protein, recognized by patients with 103 varying severity of natural infection with the Wuhan strain (WT), delta, omicron and in those 104 who received the Sinopharm vaccines, which is an inactivated, whole virus vaccine. 105 106 107 108 109 110 111 112 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 6

113 Participants for identification of immunodominant regions within the N protein 114 Blood samples from healthy adult volunteers who were either vaccinated or naturally infected 115 with SARS-CoV-2 were obtained following informed written consent. Serum separated from 116 blood samples were used to assess antibody responses to the overlapping N peptides in the 117 following groups of individuals. 118 A. SARS-CoV-2 seronegative negative individuals (n=30) prior to COVID-19 119 vaccination (negative) 120 B. Unvaccinated individuals who were naturally infected (n=30) with the SARS-CoV-2 121 wild type/Wuhan strain (WT) from day 14 to 21 from day of onset of symptoms. 5/30 122 of them had severe illness and 25/30 had mild. Clinical disease severity was classified 123 according to the WHO COVID-19 disease severity classification [31]. (WT) 124 C. Unvaccinated individuals who were naturally infected with the SARS-CoV-2 delta 125 variant (n=30), 7 to 21 days from the onset of symptoms. All individuals had mild 126 infection. (delta) 127 D. Those who were vaccinated or who possibly had prior infection (infection status 128 unknown) in those who were subsequently infected with omicron (n=20) days 14 to 129 21 since onset of illness. (Omicron+vaccinated) 130 E. Sinopharm (BBIBP-CorV) vaccine recipients 2 weeks post second dose (n=30) 131 (Sinopharm) 132 133 Ethics statement 134 Blood samples were obtained following informed written consent. Ethics approval was 135 obtained from the Ethics Review Committee of University of Sri Jayewardenepura. 136 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 7 Participants for assessing the kinetics of antibody responses to immunodominant 137 regions of N protein 138 After identification of immunodominant regions within the N protein, the responses to these 139 regions were further assessed in the groups of individuals described above. However, smaller 140 numbers were included in the analysis due to limitations in the sample volume available. 141 A. SARS-CoV-2 seronegative negative individuals (n=15) prior to COVID-19 vaccination 142 B. Unvaccinated individuals who were naturally infected (n=12) with the SARS-CoV-2 wild 143 type/Wuhan strain (WT) from day 14 to 21 since onset of illness 144 C. Unvaccinated individuals who were naturally infected with the SARS-CoV-2 delta variant 145 (n=12), with mild illness, from day 7 to 14 since onset of illness. 146 D. Those who were vaccinated (different vaccines) or who possibly had prior infection 147 (infection status unknown) in those who were infected with omicron (n=22), 14 to 21 days 148 since onset of illness. All participated individuals had mild infection. 149 E. Sinopharm (BBIBP-CorV) vaccine recipients 2 weeks post second dose (n=12) 150 F. Uninfected individuals who received COVID-19 vaccines, which only contain the SARS-151 CoV-2 spike protein, 3 months since obtaining the second dose. AZD1222 (ChAdOx1) 152 (n=10), Moderna (mRNA-1273) (n=10) and Sputnik V (Gam-COVID) (n=10). This was to 153 assess the specificity of the responses to the immunodominant N peptides. 154 155 Participants with acute infection due to SARS-CoV-2 WT virus 156 Adult patients who were acute infected with the SARS-CoV-2 virus and had mild illness 157 (n=16) or severe illness (n=9) during acute stage (<7 days since onset of symptoms) and 158 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 8 during late infection (21 to 28 days since onset of symptoms) were recruited following 159 informed written consent, to compare antibody responses against four immunodominant 160 regions between individuals with mild and severe disease during early and late stages of the 161 illness. Clinical disease severity was classified according to the WHO COVID-19 disease 162 severity classification [31]. 163 164 N protein peptide array 165 Overlapping peptides representing the N protein of SARS-CoV-2 virus (USA-WA1/2020 166 strain of SARS-CoV-2; QHO60601) was obtained through BEI Resources, NIAID, NIH: 167 Peptide Array, SARS-Related Coronavirus 2 Nucleocapsid (N) Protein, NR-52404. The 168 whole peptide array consists of 59 overlapping peptides, which overlap by 10aa to 17aa and 169 10aa with the adjacent peptide. All peptides were dissolved in appropriate solvent mentioned 170 by the manufacturer. Initially, all 59 peptides were pooled in to 4 pools. Namely, pool 1 171 (peptide 1 to 15), pool 2 (peptide 16 to 30), pool 3 (peptide 31-45) and pool 4 (peptide 46-172 59). Antibody responses to the peptides which gave the highest responses were further 173 assessed. 174 175 Identification of SARS-CoV-2 serostatus of the participants 176 The Wantai SARS-CoV-2 total antibody ELISA (Beijing Wantai Biological Pharmacy 177 Enterprise, China) was used to identify the presence of antibodies (IgM, IgG and IgA) to the 178 receptor binding domain (RBD) of the virus. The specificity of this assay in the Sri Lankan 179 population was found to be 100% [17] SARS-CoV-2. Those who tested negative for the 180 presence of total antibodies to the RBD by this assay, were considered to be seronegative. 181 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 9 182 In those who had received the spike protein contained vaccines, the presence of 183 asymptomatic infection with the virus was assessed by the presence of N protein specific 184 antibodies. This was done by using the Elecsys® Anti-SARS-CoV-2 185 electrochemiluminescence immunoassay (Cat: 09 203 095 190, Roche Diagnostics, 186 Germany) using the Cobas e 411 analyzer (Roche Diagnostics, Germany). A Cutoff index 187 (COI) ≥ 1.0 was interpreted as reactive and COI <1.00 was considered non-reactive as 188 indicated by the manufacturer. 189 190 Measuring ACE2 blocking antibodies by the surrogate virus neutralizing test (sVNT) 191 ACE2 blocking antibodies were measured using the sVNT assay which measures the 192 percentage of inhibition of binding of the RBD of the S protein to recombinant ACE2 193 (Genscript Biotech, USA). Inhibition percentage ≥ 25% in a sample was considered as 194 positive for Nabs in the Sri Lankan population as previously described [26]. 195 196 Identification of SARS-CoV-2 variants in individuals who were naturally infected with 197 SARS-CoV-2 198 In this study we recruited individuals infected with the WT, delta and omicron. All those who 199 were considered to be infected with the WT had a confirmed SARS-CoV-2 infection (PCR 200 positive) between in March to May 2020, when other VoC were not detected. Infection with 201 either delta or omicron was identified by carrying out genomic sequencing using either the 202 Oxford Nanopore (ONT) or the Illumina platforms as previously [23]. 203 204 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 10 205 206 In-house ELISA to determine IgG antibody responses to the SARS-CoV-2 overlapping 207 peptides of the N protein 208 Ninety-six-well microtitre plates (Thermofisher, USA, Pierce™ Cat: 15031) were coated 209 with the overlapping peptide representing the different pools and incubated overnight 210 at4/i2 °C. The peptides were diluted in bicarbonate/carbonate coating buffer (pH 9.6) and the 211 final concentration of each peptide was 1µg/100 μ l. The plates were blocked with PBS with 212 2% (w/v) bovine serum albumin (Sigma Aldrich, Germany, Cat: A7030) and incubated for 2 213 hours at room temperature and washed before incubation with serum samples diluted 1:500 in 214 1% BSA. After an incubation of 30 /i2 min at room temperature, the plates were washed, and 215 incubated with biotinylated goat anti-human IgG antibody (Mabtech, Sweden, Cat: 3820-4-216 250) diluted 1:1000 in 1% BSA. After a 30-minute incubation at room temperature, the plates 217 were washed and further incubated with Streptavidin–HRP (Mabtech (Sweden) Cat: 3310-9) 218 diluted 1:1000 in 1% BSA solution for 30 minutes. After washing the plates, the TMB 219 ELISA substrate solution (Mabtech, Sweden, Cat: 3652-F10) was added at 100 μ l/well and 220 the plates were incubated in the dark for 10 /i2 minutes at room temperature. The reaction was 221 stopped by adding 2M H2SO4 (Sigma Aldrich, Germany, Cat: 339741) and absorbance values 222 were read at 450nm. Optic density (OD) values above the mean± 3SD of the OD values of 223 the sera from SARS-CoV-2 seronegative individuals was considered as a positive response 224 for a particular peptide or a pool of peptides. 225 226 Statistical Analysis 227 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 11 GraphPad Prism version 9 was used for statistical analysis. As the data were not normally 228 distributed, differences in means were compared using the Mann-Whitney U test (two tailed). 229 The descriptive statistics including the mean and frequencies were used to compare antibody 230 responses of individual peptides. Kruskal-Wallis test was used to determine the differences 231 between the antibody levels (indicated by the OD value) in the four peptide pools (pool 1, 2, 232 3, and 4). and four immunodominant regions (P5/6, P23/24, P40/41, and P53/54). If Kruskal-233 Wallis test was significant, a post hoc test (Dunn test) was done to identify which group or 234 groups different from others. Spearman’s correlation coefficient was used to determine the 235 correlation between antibody responses against immunodominant regions of N protein and 236 neutralizing antibodies. 237 238 239 240 241 242 243 244 245 246 247 248 249 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 12 250 251

252 Identification immunodominant regions within the N protein of SARS-CoV-2 253 We initially tested the four pools of the SARS CoV-2 overlapping peptides of the N protein 254 in the cohorts A to E. WT, delta, omicron infected and sinopharm vaccinated individual’s 255 antibody responses were significantly different in the four overlapping pools of peptides 256 (Figure 1A to 1D). The number of individuals included in each of the cohorts that tested 257 positive for the different pools is shown in table 1. Omicron infected + vaccinated individuals 258 (who were vaccinated) had the highest positivity rates (>75%) for all four peptide pools. 259 260 Of those in who were infected prior to being vaccinated, the WT infected individuals had the 261 highest positivity rates for pool 2, while delta infected individuals gave highest antibody 262 responses to pool 3 and 4. Sinopharm vaccinees had the highest positivity rates for pool 1. 263 However, Sinopharm vaccinees had overall lower positivity rates and magnitude of responses 264 for all four peptide pools compared to naturally infected individuals. 265 266 Table 1: The number of individuals in different cohorts who gave a positive response to 267 different overlapping peptide pools of the N protein 268 SARS-CoV-2 Variant Number of individuals who gave a positive antibody response Pool 1 Pool 2 Pool 3 Pool 4 At least one pool . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 13 WT (n=30) 9 (30%) 15 (50%) 2 (6.7%) 3 (10%) 18/30 (60%) Delta (n=30) 10 (33.3%) 9 (30%) 17 (56.7%) 16 (53.3%) 22/30 (73.3%) Omicron + vaccinated (n=20) 16 (80%) 15 (75%) 18 (90%) 16 (80%) 18/20 (90%) Sinopharm (n=30) 11 (36.7%) 7 (23.3%) 5 (16.7%) 4 (13.3%) 14/30 (46.7%) 269 Mapping of antibody responses in the different cohorts to identify immunodominant 270 regions of N protein 271 As the WT infected individuals (cohort B) had the highest responses to pool 2, Sinopharm 272 vaccinees (cohort E) to pool 1, delta infected individuals (cohort C) to pool 3 and 4, we 273 proceeded to map the immunodominant regions within these different pools of overlapping 274 peptides, by testing antibody responses to these individual peptides separately. 275 276 In cohort D and E, the highest responses were observed for the two overlapping peptides 5 277 and 6 of pool 1 (peptides 1 to 15) (Supplementary Figure 1A and 1B). Cohort B (WT infected 278 individuals) and D (omicron infected+ vaccinated) had the highest responses to overlapping 279 peptides 23 and 24 of pool 2 (Peptide 16 to 30) (Supplementary Figure 1C and 1D). 280 Individuals from cohort C (delta infected) and D, had the highest responses to overlapping 281 peptide 40 and 41 of pool 3 (Peptide 31 to 45) (Supplementary Figure 1E 1F). In cohort C 282 and D, the highest responses were observed for peptide 53 and peptide 54 of pool 4 (Peptide 283 46 to 59) (Supplementary Figure 1G and 1H). Based on these results, overlapping peptides 5 284 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 14 and 6 of pool 1, overlapping peptides 23 and 24 of pool 2, overlapping peptides 40 and 41 of 285 pool 3 and overlapping peptides 53 and 54 of pool 4, were the immunodominant regions 286 within the N protein. 287 288 Characterizing antibody responses to the immunodominant regions identified within N 289 protein 290 In order to further characterize the antibody responses to the above immunodominant regions 291 within the N protein, the overlapping peptides 5 and 6, 23 and 24, 40 and 41 and 53 and 54 292 were pooled together from four different pools. The antibody responses for these regions 293 were assessed in all cohort (cohort A to E), to identify responses in all individuals for these 294 pools. Although the WT infected individuals gave the highest antibody responses to P23/24 295 (Figure 2A and B), there was no significant difference (p=0.057) in the magnitude of 296 responses for the four immunodominant peptide pools in this group. Similarly, delta infected 297 individuals had a similar magnitude of responses for the four pools (p=0.53). Omicron 298 infected+vaccinated (cohort D) had the highest responses to P40/41 and P53/54, while 299 Sinopharm vaccinees had the highest responses to P5/6 (Figure 2A and B). 300 301 The positivity rates for each of the four immunodominant regions in these cohorts is shown in 302 table 2. Overall, all cohorts had >80% positivity rates for all 4 regions, except lower 303 positivity rates in delta infected and omicron infected+ vaccinated individuals for P23/24 304 region. In contrast, all WT infected individuals gave a positive response for this region, while 305 they had low positivity rates (66.7%) for P53/54. 306 307 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 15 We also assessed specificity of the antibody responses to the four immunodominant regions 308 of the N protein (P5/6, P23/24, P40/41, and P53/54), by investigating responses in those who 309 received COVID-19 vaccines with only the spike protein. Accordingly, responses were 310 assessed in those who received AZD1222 (ChAdOx1) (n=10), Moderna (mRNA-1273) 311 (n=10), and Sputnik V (Gam-COVID-Vac) (n=10). None of the SARS-CoV-2 seronegative 312 individuals (cohort A) or those who received ChAdOx1, Gam-COVID-Vac or mRNA-1273 313 responded to the peptides P53/54, while one individual (1/30) had a positive response to the 314 peptides P5/6. Two individuals (2/30) responded to the peptides P23/24 and P40/41 (Table 2, 315 Figure 2C). Therefore, while the specificity of peptide P53/54 was 100% in detecting SARS-316 CoV-2 N protein specific responses following or vaccination (whole virus vaccine) or 317 infection, the specificity of peptides P5/6 was 96.7% while for peptides P23/24 and P40/41 it 318 was 93.3%. 319 320 Table 2: The number of individuals in different cohorts who gave a positive response to 321 immunodominant regions within the N protein 322 SARS-CoV-2 Variant Number of individuals who gave a positive antibody response P5/6 P23/24 P40/41 P53/54 At least for one pair WT (n=12) 10 (83.3%) 12 (100%) 11 (91.7%) 8 (66.7%) 12/12 (100%) Delta (n=12) 11 (91.7%) 9 (75%) 11 (91.7%) 10 (83.3%) 12/12 (100%) Omicron+ vaccinated (n=22) 20 (90.9%) 16 (72.7%) 20 (90.9%) 18 (81.8%) 20/22 (90.9%) . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 16 Sinopharm (n=12) 12 (100%) 11 (91.7%) 11 (91.7%) 11 (91.7%) 12/12(100%) Total (n=58) 53 (91.4%) 48 (82.8%) 53 (91.4%) 47 (81.0%) 56/58 (96.5%) AstraZeneca (n=10) 0 1 (10%) 1 (10%) 0 1/10 (10%) Sputnik V (n=10) 0 0 1 (10%) 0 1/10 (10%) Moderna (n=10) 1 (10%) 1 (10%) 0 0 2/10 (20%) Total (n=30) 1 (3.3%) 2 (6.7%) 2 (6.7%) 0 4/30 (13.3%) Conservation of immunodominant regions of the N protein of SARS-CoV-2 with 323 seasonal human coronavirus and SARS-CoV-2 variants of concern (VoC) 324 As the consensus peptide sequence may not be representative of the infecting subtype, we 325 determined the conservation within these four immunodominant regions within the different 326 SARS-CoV-2 variants (Alpha, QVX37034.1; Beta, QWW93444.1; Gamma, QXF23757.1; 327 Delta, UKA47847.1; Omicron, (BA.1 (SriLanka/aicbu4450/2022), BA.2 328 (SriLanka/aicbu4463/2022) and BA.5 (USA/CA-CDPH-FS27225444/2022) and also the 329 cross reactivity with other seasonal human coronaviruses (OC43, QBP84763.1; HKU1, 330 ABG77571.1; NL63, YP_003771.1). We used Jalview software [28] and tools available at 331 European Bioinformatics Institute (EBI) (www.ebi.sc.uk , 22 March 2022) to determine 332 conservation between identified four immunodominant regions of the wild type SARS-CoV-2 333 (YP_009724397.2) The four regions in which the conservation and cross reactivity were 334 assessed are as follows; 335 P5/P6: 29NGERSGARSKQRRPQGLPNNTASW52 336 P23/P24:155AAIVLQLPQGTTLPKGFYAEGSRG178 337 P40/P41:274FGRRGPEQTQGNFGDQELIRQGTD297) 338 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 17 P53/54: 365PTEPKKDKKKKADETQALPQRQKK388 339 All regions showed <50% sequence identity with the three seasonal human coronaviruses 340 (Supplementary table 1 Figure 3A to 3D). P5/6 showed <20% sequence identity with OC43, 341 HKU1, and NL63, with P53/54 showing 45% of 343 sequence identity with OC43 and HKU1 but not with NL63. All four immunodominant 344 regions were found to be conserved in both alpha and beta variants (Supplementary table 1 345 and Figure 3E to H), while there was 95.8% sequence identity with delta (single amino acid 346 replacement) in the P53/54 region and 95.8% in the P40/41 region with the gamma variant 347 (single amino acid replacement). All three omicron sub-lineages (BA.1, BA.2 and BA.5) P5/6 348 have a 3 amino acid deletion within the regions represented by P5/6 and therefore, a sequence 349 identity of 87.5% (Figure 3E). P23/24 regions was 100% conserved in all five VoC (Figure 350 3F). 351 352 As the SARS-CoV-2 virus continues to further evolve and due to the future threat of other bat 353 coronaviruses spilling over and causing future pandemics, many Pan-Sarbecovirus vaccines 354 are currently under development [9,19]. Therefore, we proceeded to find out the conservation 355 of these four regions with 4 bat coronaviruses, RS4081, WIV1, RatG13 and Rf1 (Figure 4) 356 and Supplementary table 1. These four regions showed >91% sequence identity with all the 357 four bat coronaviruses. 358 359 Antibody responses to the four immunodominant regions in patients with varying 360 severity of COVID-19 due to the WT virus 361 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 18 We then sought to compare antibody responses between mild and severe disease during early 362 and late stage of the infection in individuals who had mild illness (n=16) or severe illness 363 (n=9) during acute stage (<7 days since onset of symptoms) and during late infection (21 to 364 28 days since onset of symptoms). Those with severe illness had significantly higher antibody 365 responses to P23/24, P40/41 and P53/54 during the first week of illness compared to those 366 with mild illness (Figure 5A). During late infection, those with severe disease had 367 significantly higher antibody responses to all four regions than individuals with mild illness 368 (Figure 5B). 369 370 We then sought to explore if antibody responses to these immunodominant regions, correlate 371 with neutralizing antibody (Nab) responses, by comparing ACE2 blocking antibodies (which 372 were shown to correlate with Nabs), by using a surrogate SARS-CoV-2-neutralizing antibody 373 assay. The ACE2 blocking antibodies did not correlate with antibody response against all 374 four regions (Spearman’s r=0.17, p=0.02) (Figure 5C). 375 376 377 378 379 380 381 382 383 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 19 384 385 386

387 In this study we have identified four immunodominant regions within the N protein, which 388 gave a high frequency of responses in those infected with the WT virus, delta, omicron and 389 those vaccinated with Sinopharm. Overall, >80% of individuals gave responses above the 390 positive cut-off threshold to many of the four regions, with some differences with individuals 391 who were infected with different VoCs. These regions were found to be 100% specific, as 392 none of the seronegative individuals gave any responses. However, 6.7% to 10% of 393 individuals who had received the spike proteins vaccines and therefore, should not have 394 responses to these regions also responded. Although they had tested negative for infection 395 with SARS-CoV-2 by the commercial N protein-specific antibody assay, it is possible that 396 they could still have been naturally infected. Due to the high sensitivity and specificity of 397 these regions, they have a potential to be used to identify N protein specific antibody 398 responses, especially regions aa 29 to 52 (P5/6) and aa 365 to 388 (P40/41), for which >90% 399 of individuals responded to. However, as we used overlapping peptides to identify potential 400 epitopes, we are likely to have missed conformational epitopes which could be key antibody 401 recognition sites. 402 403 The five domains of the N protein have shown to bind to RNA and carry out multiple 404 functions including RNA interference, regulating virus replication and host immune evasion 405 [6]. Of the four regions identified here, one is within the N terminal domain (aa 29-52, P5/6), 406 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 20 one in the RNA binding domain (aa 155-178, P23/24), one within the dimerization domain 407 (aa 274 to 297, P40/41) and one in the C-terminal domain (aa365 to 388/ P53/54). Some 408 previous studies had identified immunodominant regions of the N protein, in llamas 409 (domesticated South American camel), and had identified antibodies that bind to highly 410 conserved regions within the N protein [32]. One of these antibodies had shown to bind to the 411 C-terminal domain (aa 49-174) and two antibodies to the N-terminal domain (aa 247-364 and 412 aa 365-419)[32]. The four immunodominant regions that were identified here, also fall within 413 these three regions. Another study, which screened for B cell epitopes within the N protein 414 using mouse models, identified aa 401 to 408 as the main antibody target. Our data show that, 415 the predominant B cell epitopes identified within the N protein differs based on the infecting 416 SARS-CoV-2 variant. For instance, those who were infected with the WT virus, 417 predominantly recognized the aa155 to 178 regions (P23/24), whereas those who were 418 infected with delta had the highest responses to the aa 29 to 52 and aa 274 to 294 regions. 419 However, the responses to these overlapping peptides were assessed using the sequence of a 420 Wuhan virus strain isolated from USA in 2020 and the epitope recognition could be different, 421 based on the sequence of peptides used. In those who were infected with different omicron 422 sub-lineages (BA.1 and BA.2) also had the highest responses to the aa 29 to 52 and aa 274 to 423 294 regions. Therefore, the predominant B cell epitope recognition, appears to differ based on 424 the variant of infection. Although Sinopharm vaccinees responded to all four regions, the 425 magnitude of the responses was significantly lower than following natural infection. This is 426 possibly due to natural infection inducing more robust responses to the N protein than 427 following inactivated vaccines containing the whole protein. 428 429 The mortality rates and hospitalization rates have varied widely throughout the COVID-19 430 pandemic in different countries, with many countries in Europe, and United States reporting 431 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 21 higher mortality rates and hospitalization rates than some countries in Africa and Asia, 432 despite higher rates of vaccination [13,14]. These differences could be attributed to reporting 433 of COVID-19 deaths and limitations in testing, as many countries in sub-Saharan Africa have 434 reported high excess mortality rates [10]. Sri Lanka experienced high mortality rates during 435 the delta outbreak during the months from June to October 2021 prior to vaccination [24]. 436 However, mortality rates have been significantly less (0.85/ million individuals in Sri Lanka) 437 during the massive omicron wave (BA.2), than many European countries and the United 438 States (mortality rates 4.03/million individuals in Europe and 5.4/million individuals in 439 United States) [13]. Only 18% of Sri Lankans had received an mRNA booster dose, when the 440 omicron variant was rapidly spreading in Sri Lanka [14]. The lower mortality rates seen in Sri 441 Lanka during the omicron outbreak were unlikely to be due to under reporting or limited 442 testing as the excess mortality rates in Sri Lanka were found to be less than the excess 443 mortality rates reported in Europe and North America [10]. Sinopharm/BBIBP-CorV was the 444 most widely used vaccine, in Sri Lanka with 12 million (70.6%) individuals receiving this 445 vaccine by end of December 2021 [9]. Sinopharm/BBIBP-CorV vaccine was found to be less 446 immunogenic than the mRNA-1273, AZD1222 and Sputnik V, 3 months post second dose, in 447 a head-to-head comparison in the Sri Lankan population, based on ACE2 blocking antibodies 448 and antibodies to the receptor binding domain of the spike protein [15]. However, as 449 Sinopharm/BBIBP-CorV is an inactivated vaccine, it did induce T cell and antibody 450 responses to the N protein [16]. In addition, although the N protein was thought to be 451 localized to the cytosol, it was recently shown that this protein was expressed on the surface 452 of infected cells [20]. As the N protein has shown to bind to several different types of 453 chemokines, antibodies against the N protein could also inhibit chemotaxis of leucocytes 454 [20]. Furthermore, antibodies bound to N protein were shown to activate FcR expressing 455 innate immune cells, further contributing to the phagocytosis and apoptosis of infected 456 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 22 cells[20]. Although there could be many reasons for the differences in mortality rates for 457 different variants, it is possible that antibody responses to the N protein, offered additional 458 protection in Sinopharm vaccinees, which should be further investigated. 459 460 Due to the rapidly evolving nature of SARS-CoV-2 and emergence of more immune evasive 461 omicron sub-lineages, there is a global effort to develop a pan-Sarbecovirus vaccine [7,9,11]. 462 However, many pan-Sarbecovirus vaccines only use the spike protein as the immunogen and 463 explore the immune responses to the spike protein [19,25], while only a few vaccines also 464 include the N protein [9]. In this study we show that the four immunodominant regions 465 identified here, were highly conserved regions within SARS-CoV-2 and the bat 466 coronaviruses. Although the neutralizing antibody responses for many bat Sarbecoviruses has 467 been investigated [27], there are no data if antibodies targeting the main B cell epitopes 468 within the N protein, also cross-neutralize the most frequent bat Sarbecoviruses, which would 469 be important. While vaccination, especially with an mRNA booster dose induced high levels 470 of Nabs and protected individuals from severe disease [4], natural infection and vaccination 471 induced a high magnitude of durable immune responses [2]. In fact, it was shown that two 472 doses of an mRNA vaccine and natural infection gave similar immune responses as three 473 doses of a mRNA vaccine, while the immune responses induced by natural infection were 474 longer lasting [2]. Therefore, in order to induce persistent and broad immune responses, 475 antibody and T cell responses to the N protein may play an important role in addition to 476 Nabs. 477 478

479 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 23 We have identified four immunodominant regions within the N protein of SARS-CoV-2, 480 which are highly conserved in the SARS-CoV-2 variants and also show high conservation in 481 bat Sarbecoviruses. Responses to these regions were highly specific and elicited responses in 482 > 90% of naturally infected individuals or those who received a whole virus inactivated 483 vaccine to at least two of the regions. As these regions were highly specific with high 484 sensitivity, they have a potential to be used to develop diagnostic assays and to be used in 485 development of vaccines. 486 487 List of abbreviations 488 Nabs: Neutralization antibodies 489 N protein: Nucleocapsid protein 490 VoC: variants of concern 491 sVNT: surrogate virus neutralizing test 492 RBD: receptor binding domain 493 WT: Wuhan strain of SARS-CoV-2 494 Ethics approval and consent to participate 495 Ethics approval was obtained by the Ethics Review Committee of the University of Sri 496 Jayewardenepura. All individuals gave informed, written consent. 497 Consent for publication 498 Not applicable 499 Availability of data and materials 500 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 24 All data is available in the manuscript and figures. 501 Competing interests 502 Authors have no competing interests. 503 504 Funding 505 We are grateful to Allergy Immunology and Cell Biology Unit, University of Sri 506 Jayewardenepura, the NIH, USA (grant number 5U01AI151788-02), World bank, Sri Lanka 507 Covid 19 Emergency Response and Health Systems Preparedness Project (ERHSP) 508 of Ministry of Health Sri Lanka funded by World Bank and the UK Medical Research 509 Council for funding. 510 Authors' contributions 511 Conceptualization of the study: PDP, CJ, GNM 512 Data curation: ISA, TN, JJ, TR, HK, SD 513 Project administration: CJ, AW, GNM 514 Experiments: PDP, FB, DM, LP 515 Data analysis: PDP 516 Funding: CJ, GSO, GNM 517 Writing the manuscript: PDP, GNM 518 Reviewing the manuscript: CJ, GSO 519 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 25 520

521 None 522 523

524 1. Addetia A., Crawford K.H.D., Dingens A., Zhu H., Roychoudhury P., Huang M.L., 525 Jerome K.R., Bloom J.D. and Greninger A.L. Neutralizing Antibodies Correlate with 526 Protection from SARS-CoV-2 in Humans during a Fishery Vessel Outbreak with a 527 High Attack Rate. Journal of clinical microbiology 58(11), 2020. 528 2. Altarawneh H.N., Chemaitelly H., Ayoub H.H., Tang P., Hasan M.R., Yassine H.M., 529 Al-Khatib H.A., Smatti M.K., Coyle P., Al-Kanaani Z., Al-Kuwari E., Jeremijenko 530 A., Kaleeckal A.H., Latif A.N., Shaik R.M., Abdul-Rahim H.F., Nasrallah G.K., Al-531 Kuwari M.G., Butt A.A., Al-Romaihi H.E., Al-Thani M.H., Al-Khal A., Bertollini R. 532 and Abu-Raddad L.J. Effects of Previous Infection and Vaccination on Symptomatic 533 Omicron Infections. The New England journal of medicine, 2022. 534 3. Bai Z., Cao Y., Liu W. and Li J. The SARS-CoV-2 Nucleocapsid Protein and Its Role 535 in Viral Structure, Biological Functions, and a Potential Target for Drug or Vaccine 536 Mitigation. Viruses 13(6), 2021. 537 4. Barda N., Dagan N., Cohen C., Hernan M.A., Lipsitch M., Kohane I.S., Reis B.Y. and 538 Balicer R.D. Effectiveness of a third dose of the BNT162b2 mRNA COVID-19 539 vaccine for preventing severe outcomes in Israel: an observational study. Lancet 540 398(10316):2093-2100, 2021. 541 5. Bechman K., Yates M., Mann K., Nagra D., Smith L.J., Rutherford A.I., Patel A., 542 Periselneris J., Walder D., Dobson R.J.B., Kraljevic Z., Teo J.H.T., Bernal W., Barker 543 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 26 R., Galloway J.B. and Norton S. Inpatient COVID-19 mortality has reduced over 544 time: Results from an observational cohort. PloS one 17(1):e0261142, 2022. 545 6. Burbelo P.D., Riedo F.X., Morishima C., Rawlings S., Smith D., Das S., Strich J.R., 546 Chertow D.S., Davey R.T. and Cohen J.I. Sensitivity in Detection of Antibodies to 547 Nucleocapsid and Spike Proteins of Severe Acute Respiratory Syndrome Coronavirus 548 2 in Patients With Coronavirus Disease 2019. The Journal of infectious diseases 549 222(2):206-213, 2020. 550 7. Cao Y., Yisimayi A., Jian F., Song W., Xiao T., Wang L., Du S., Wang J., Li Q., 551 Chen X., Yu Y., Wang P., Zhang Z., Liu P., An R., Hao X., Wang Y., Wang J., Feng 552 R., Sun H., Zhao L., Zhang W., Zhao D., Zheng J., Yu L., Li C., Zhang N., Wang R., 553 Niu X., Yang S., Song X., Chai Y., Hu Y., Shi Y., Zheng L., Li Z., Gu Q., Shao F., 554 Huang W., Jin R., Shen Z., Wang Y., Wang X., Xiao J. and Xie X.S. BA.2.12.1, BA.4 555 and BA.5 escape antibodies elicited by Omicron infection. Nature, 2022. 556 8. Chansaenroj J., Yorsaeng R., Posuwan N., Puenpa J., Wanlapakorn N., Sudhinaraset 557 N., Sripramote M., Chalongviriyalert P., Jirajariyavej S., Kiatpanabhikul P., Saiyarin 558 J., Soudon C., Thienfaidee O., Palakawong Na Ayuthaya T., Brukesawan C., 559 Chirathaworn C., Intharasongkroh D., Chaiwanichsiri D., Issarasongkhram M., 560 Kitphati R., Mungaomklang A., Nagavajara P. and Poovorawan Y. Long-term 561 specific IgG response to SARS-CoV-2 nucleocapsid protein in recovered COVID-19 562 patients. Sci Rep 11(1):23216, 2021. 563 9. Coleon S., Wiedemann A., Surenaud M., Lacabaratz C., Hue S., Prague M., 564 Cervantes-Gonzalez M., Wang Z., Ellis J., Sansoni A., Pierini C., Bardin Q., Fabregue 565 M., Sharkaoui S., Hoest P., Dupaty L., Picard F., El Hajj M., Centlivre M., Ghosn J., 566 French C.C.S.G., Thiebaut R., Cardinaud S., Malissen B., Zurawski G., Zarubica A., 567 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 27 Zurawski S.M., Godot V. and Levy Y. Design, immunogenicity, and efficacy of a 568 pan-sarbecovirus dendritic-cell targeting vaccine. EBioMedicine 80:104062, 2022. 569 10. Collaborators C.-E.M. Estimating excess mortality due to the COVID-19 pandemic: a 570 systematic analysis of COVID-19-related mortality, 2020-21. Lancet 571 399(10334):1513-1536, 2022. 572 11. Fountain-Jones N.M., Appaw R.C., Carver S., Didelot X., Volz E. and Charleston M. 573 Emerging phylogenetic structure of the SARS-CoV-2 pandemic. Virus Evol 574 6(2):veaa082, 2020. 575 12. Garcia-Beltran W.F., Lam E.C., Astudillo M.G., Yang D., Miller T.E., Feldman J., 576 Hauser B.M., Caradonna T.M., Clayton K.L., Nitido A.D., Murali M.R., Alter G., 577 Charles R.C., Dighe A., Branda J.A., Lennerz J.K., Lingwood D., Schmidt A.G., 578 Iafrate A.J. and Balazs A.B. COVID-19-neutralizing antibodies predict disease 579 severity and survival. Cell 184(2):476-488 e411, 2021. 580 13. Hannah Ritchie E.M., Lucas Rodés-Guirao, Cameron Appel, Charlie Giattino, 581 Esteban Ortiz-Ospina, Joe Hasell, Bobbie Macdonald, Diana Beltekian and Max 582 Roser (2020). Coronavirus Pandemic (COVID-19). 02/24/2022 ed, 2022. 583 14. Hannah Ritchie E.O.-O., Diana Beltekian, Edouard Mathieu, Joe Hasell, Bobbie 584 Macdonald, Charlie Giattino, Cameron Appel and Max Roser. 2021. Coronavirus 585 (COVID-19) Vaccinations. In COVID-19 vaccine doses administered per 100 people, 586 Apr 1, 2021 Our World in Data . 587 Accessed 588 15. Jeewandara C., Aberathna I.S., Danasekara S., Gomes L., Fernando S., Guruge D., 589 Ranasinghe T., Gunasekera B., Kamaladasa A., Kuruppu H., Somathilake G., 590 Jayamali J., Jayathilaka D., Wijayatilake H.D.K., Pushpakumara P.D., Harvie M., 591 Nimasha T., de Silva S.D.G., Wijayamuni R., Schimanski L., Rijal P., Tan J., 592 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 28 Townsend A., Ogg G.S. and Malavige G.N. Comparison of the Immunogenicity of 593 five COVID-19 vaccines in Sri Lanka. Immunology, 2022. 594 16. Jeewandara C., Aberathna I.S., Pushpakumara P.D., Kamaladasa A., Guruge D., 595 Wijesinghe A., Gunasekera B., Tanussiya S., Kuruppu H., Ranasinghe T., Dayarathne 596 S., Dissanayake O., Gamalath N., Ekanayake D., Mpdj J., Jayathilaka D., 597 Dissanayake M., Madusanka D., Jayadas T.T., Mudunkotuwa A., Somathilake G., 598 Harvie M., Nimasha T., Danasekara S., Wijayamuni R., Schimanski L., Rijal P., Tan 599 T.K., Dong T., Townsend A., Ogg G.S. and Malavige G.N. Immune responses to 600 Sinopharm/BBIBP-CorV in individuals in Sri Lanka. Immunology, 2022. 601 17. Jeewandara C., Guruge D., Jayathilaka D., Deshan Madhusanka P.A., Pushpakumara 602 P.D., Tanussiya Ramu S., Sepali Aberathna I., Saubhagya Rasikangani Danasekara 603 D.R., Pathmanathan T., Gunatilaka B., Malavige S., Dias Y., Wijayamuni R., Ogg 604 G.S. and Malavige G.N. Transmission dynamics, clinical characteristics and sero-605 surveillance in the COVID-19 outbreak in a population dense area of Colombo, Sri 606 Lanka April- May 2020. PloS one 16(11):e0257548, 2021. 607 18. Kifer D., Bugada D., Villar-Garcia J., Gudelj I., Menni C., Sudre C., Vuckovic F., 608 Ugrina I., Lorini L.F., Posso M., Bettinelli S., Ughi N., Maloberti A., Epis O., 609 Giannattasio C., Rossetti C., Kalogjera L., Persec J., Ollivere L., Ollivere B.J., Yan 610 H., Cai T., Aithal G.P., Steves C.J., Kantele A., Kajova M., Vapalahti O., Sajantila A., 611 Wojtowicz R., Wierzba W., Krol Z., Zaczynski A., Zycinska K., Postula M., Luksic 612 I., Civljak R., Markotic A., Brachmann J., Markl A., Mahnkopf C., Murray B., 613 Ourselin S., Valdes A.M., Horcajada J.P., Castells X., Pascual J., Allegri M., 614 Primorac D., Spector T.D., Barrios C. and Lauc G. Effects of Environmental Factors 615 on Severity and Mortality of COVID-19. Front Med (Lausanne) 7:607786, 2020. 616 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 29 19. Liu Z., Chan J.F., Zhou J., Wang M., Wang Q., Zhang G., Xu W., Chik K.K., Zhang 617 Y., Wang Y., Yuen K.Y., Lu L. and Jiang S. A pan-sarbecovirus vaccine induces 618 highly potent and durable neutralizing antibody responses in non-human primates 619 against SARS-CoV-2 Omicron variant. Cell research 32(5):495-497, 2022. 620 20. Lopez-Munoz A.D., Kosik I., Holly J. and Yewdell J.W. Cell surface SARS-CoV-2 621 nucleocapsid protein modulates innate and adaptive immunity. Sci Adv 622 8(31):eabp9770, 2022. 623 21. Markov P.V., Katzourakis A. and Stilianakis N.I. Antigenic evolution will lead to new 624 SARS-CoV-2 variants with unpredictable severity. Nature reviews, 2022. 625 22. Peng Y., Felce S.L., Dong D., Penkava F., Mentzer A.J., Yao X., Liu G., Yin Z., Chen 626 J.L., Lu Y., Wellington D., Wing P.A.C., Dominey-Foy D.C.C., Jin C., Wang W., 627 Hamid M.A., Fernandes R.A., Wang B., Fries A., Zhuang X., Ashley N., Rostron T., 628 Waugh C., Sopp P., Hublitz P., Beveridge R., Tan T.K., Dold C., Kwok A.J., Rich-629 Griffin C., Dejnirattisa W., Liu C., Kurupati P., Nassiri I., Watson R.A., Tong O., 630 Taylor C.A., Kumar Sharma P., Sun B., Curion F., Revale S., Garner L.C., Jansen K., 631 Ferreira R.C., Attar M., Fry J.W., Russell R.A., Consortium C., Stauss H.J., James 632 W., Townsend A., Ho L.P., Klenerman P., Mongkolsapaya J., Screaton G.R., 633 Dendrou C., Sansom S.N., Bashford-Rogers R., Chain B., Smith G.L., McKeating 634 J.A., Fairfax B.P., Bowness P., McMichael A.J., Ogg G., Knight J.C. and Dong T. An 635 immunodominant NP105-113-B*07:02 cytotoxic T cell response controls viral 636 replication and is associated with less severe COVID-19 disease. Nature immunology 637 23(1):50-61, 2022. 638 23. Ranasinghe D., Jayadas T.T.P., Jayathilaka D., Jeewandara C., Dissanayake O., 639 Guruge D., Ariyaratne D., Gunasinghe D., Gomes L., Wijesinghe A., Wijayamuni R. 640 and Malavige G.N. Comparison of different sequencing techniques for identification 641 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 30 of SARS-CoV-2 variants of concern with multiplex real-time PCR. PloS one 642 17(4):e0265220, 2022. 643 24. Ranasinghe D., Jayathilaka D., Jeewandara C., Gunasinghe D., Ariyaratne D., Jayadas 644 T.T.P., Kuruppu H., Wijesinghe A., Bary F.F., Madhusanka D., Pushpakumara P.D., 645 Guruge D., Wijayamuni R., Ogg G.S. and Malavige G.N. Molecular Epidemiology of 646 AY.28 and AY.104 Delta Sub-lineages in Sri Lanka. Front Public Health 10:873633, 647 2022. 648 25. Su S., Li W. and Jiang S. Developing pan-beta-coronavirus vaccines against emerging 649 SARS-CoV-2 variants of concern. Trends in immunology 43(3):170-172, 2022. 650 26. Tan C.W., Chia W.N., Qin X., Liu P., Chen M.I., Tiu C., Hu Z., Chen V.C., Young 651 B.E., Sia W.R., Tan Y.J., Foo R., Yi Y., Lye D.C., Anderson D.E. and Wang L.F. A 652 SARS-CoV-2 surrogate virus neutralization test based on antibody-mediated blockage 653 of ACE2-spike protein-protein interaction. Nature biotechnology 38(9):1073-1078, 654 2020. 655 27. Tan C.W., Chia W.N., Young B.E., Zhu F., Lim B.L., Sia W.R., Thein T.L., Chen 656 M.I., Leo Y.S., Lye D.C. and Wang L.F. Pan-Sarbecovirus Neutralizing Antibodies in 657 BNT162b2-Immunized SARS-CoV-1 Survivors. The New England journal of 658 medicine 385(15):1401-1406, 2021. 659 28. Waterhouse A.M., Procter J.B., Martin D.M., Clamp M. and Barton G.J. Jalview 660 Version 2--a multiple sequence alignment editor and analysis workbench. 661 Bioinformatics 25(9):1189-1191, 2009. 662 29. Watson O.J., Barnsley G., Toor J., Hogan A.B., Winskill P. and Ghani A.C. Global 663 impact of the first year of COVID-19 vaccination: a mathematical modelling study. 664 The Lancet infectious diseases, 2022. 665 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 31 30. Whitaker M., Elliott J., Bodinier B., Barclay W., Ward H., Cooke G., Donnelly C.A., 666 Chadeau-Hyam M. and Elliott P. Variant-specific symptoms of COVID-19 among 667 1,542,510 people in England. medRxiv:2022.2005.2021.22275368, 2022. 668 31. WHO. C li ni c al ma nag emen t of se vere ac ut e re spir atory inf e ctio n w he n novel 669 coronaviru s ( 20 19 -nC oV) infection is suspected: interim guidance: WHO; 2020 28th 670 January 2020. 671 32. Ye Q., Lu S. and Corbett K.D. Structural Basis for SARS-CoV-2 Nucleocapsid 672 Protein Recognition by Single-Domain Antibodies. Frontiers in immunology 673 12:719037, 2021. 674 675 676 677 678 679 680 681 682 683 684 685 686 687 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 32 688 689 Figure legends 690 Figure 1: Antibody responses for the four overlapping pools of peptide of the N protein 691 of SARS-CoV-2 692 IgG antibody responses were measured by an in-house ELISA for pool 1 (A), pool 2 (B), 693 pool 3 (C) and pool 4 (D) containing overlapping peptides of the N protein in SARS-CoV-2 694 seronegative (non-vaccinated) individuals (n=30), WT infected individuals (n=30), delta 695 infected individuals (n=30), omicron infected and vaccinated individuals (n=20), and 696 Sinopharm vaccinees (n=30). Kruskal-Wallis test was used to determine the differences 697 between the levels of antibody responses in four peptide pools (pool 1, 2, 3, and 4). Dotted 698 line shows the cutoff value (OD value) of a positive response for the each of the peptide 699 pools. The error bars indicate the median and the interquartile ranges. Black: seronegatives ; 700 Brown: WT infected individuals ; Blue: Delta infected individuals ; Green: Omicron + 701 vaccinated; Red: Sinopharm vaccinees. 702 703 Figure 2: Characterizing antibody responses to the immunodominant regions identified 704 within N protein 705 IgG antibody responses to the four immunodominant regions (P5/6, P23/24, P40/41, and 706 P53/54) in the N protein were measured by an in-house ELISA in SARS-CoV-2 in SARS-707 CoV-2 seronegative (non-vaccinated) individuals (n=15), WT infected individuals (n=12), 708 delta infected individuals (n=12), omicron infected and vaccinated individuals (n=22), and 709 Sinopharm vaccinees (n=12) (A). The magnitude of antibody responses to these regions in 710 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 33 the above cohorts were compared with each other (B). In order to determine specificity, 711 antibody responses were measured in individuals who were vaccinated (uninfected) with 712 AZD1222 (n=10), Moderna (n=10), and Sputnik V (n=10). Kruskal-Wallis test was used to 713 determine the differences between the levels of antibody responses in four peptide pools 714 (pool 1, 2, 3, and 4). The error bars indicate the median and the interquartile ranges. 715 716 Figure 3: Analysis of conservation of immunodominant regions of the N protein of 717 SARS-CoV-2 with seasonal human coronavirus and SARS-CoV-2 variants of concern 718 (VoC) 719 The cross reactivity of the four immunodominant regions P5/6 (A), P23/24 (B), P40/41 (C), 720 and P53/54 (D) with three seasonal human corona viruses (OC43, HKU1, and NL63) were 721 determined The conservation within these four immunodominant regions were also assessed 722 for the five VoCs (alpha, beta, gamma, delta, and omicron (BA.1, BA.2, and BA.5). P5/6 (E), 723 P23/24 (F), P40/41 (G), and P53/54 (H). Matching (sequence identity 100%) respective 724 immunodominant regions were highlighted purple color. 725 726 Figure 4: Analysis of conservation of immunodominant regions of the N protein of 727 SARS-CoV-2 with bat coronaviruses 728 The cross reactivity of the four immunodominant regions P5/6 (A), P23/24 (B), P40/41 (C), 729 and P53/54 (D) RS4081, WIV1, RatG13 and Rf1 were analyzed. Matching (sequence identity 730 100%) respective immunodominant regions were highlighted purple color. 731 732 Figure 5: Antibody responses to the four immunodominant regions in patients with 733 varying severity of COVID-19 due to the WT virus 734 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 34 Antibody responses to the four immunodominant regions were measured by an in-house 735 ELISA, in patients infected with the WT of SARS-CoV-2 with mild (n=16) and severe 736 disease (n=9) during early illness (<7 days since onset of symptoms) (A) and late illness (21 737 to 28 days since onset of symptoms (B). The antibody responses to the four regions were 738 correlated with ACE2 blocking antibodies measured by the surrogate virus neutralization test, 739 and the ACE2 blocking antibodies did not correlate with the levels of the four 740 immunodominant regions (C). The Mann-Whitney U test (two-tailed) was used to determine 741 the differences in antibody levels between those with mild and severe disease. All tests were 742 two sided. The error bars indicate the median and the interquartile ranges. 743 744 Supplementary figure legends 745 Supplementary Figure 1 : Mapping of antibody responses in the different cohorts to 746 identify immunodominant regions within the pools of overlapping peptides. 747 IgG antibody responses were measured by an in-house ELISA for individual overlapping 748 peptides of pool 1 (peptide 1 to 15) in omicron + vaccinated (A) and Sinopharm vaccinees 749 (B), pool 2 (peptide 16 to 30) in WT infected (C), and omicron+ vaccinated (D), pool 3 750 (peptide 31 to 45) delta infected (E) and omicron + vaccinated (F) and in pool 4 in delta 751 infected(G) and omicron+ vaccinated (H). 10 individuals were included in each cohort to 752 identify individual antibody responses. The error bars indicate the median and the 753 interquartile ranges. 754 755 756 757 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint 35 758 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted January 7, 2023. ; https://doi.org/10.1101/2023.01.05.23284247doi: medRxiv preprint