COVID-19 mRNA vaccine induced antibody responses and neutralizing antibodies against three SARS-CoV-2 variants

22 As SARS-CoV-2 has been circulating for over a year, dozens of vaccine candidates are under 23 development or in clinical use. The BNT162b2 mRNA COVID-19 vaccine induces spike protein- 24 specific neutralizing antibodies associated with protective immunity. The emergence of the B.1.1.7 and 25 B.1.351 variants has raised concerns of reduced vaccine efficacy and increased re-infection rates. Here 26 we show, that after the second dose, the sera of BNT162b2-vaccinated health care workers (n = 180) 27 effectively neutralize the SARS-CoV-2 variant with the D614G substitution and the B.1.1.7 variant, 28 whereas the neutralization of the B.1.351 variant is five-fold reduced. Despite the reduction, 92% of the 29 vaccinees have a neutralization titre of >20 for the B.1.351 variant indicating some protection. The 30 vaccinees’ neu tralization titres exceeded those of recovered non-hospitalized COVID-19 patients. Our 31 work provides strong evidence that the second dose of the BNT162b2 vaccine induces efficient cross- 32 neutralization of SARS-CoV-2 variants currently circulating in the world. we characterize the BNT162b2 vaccine-induced antibody responses among a sequential serum sample cohort of 180 who, two doses of COVID-19 vaccine with three weeks interval. SARS-CoV-2 S1-specific IgG, 62 IgA, and IgM antibody responses and neutralization titres for three SARS-CoV-2 variants were 63 determined. We show that two-dose immunization yields high levels of anti-S1 IgG antibodies in 100% 64 of vaccinees. The second vaccine dose induces antibodies for efficient neutralization of D614G and 65 B.1.1.7. variants, whereas the neutralization titres for B.1.351 are lower.

Neutralizing antibody titres against SARS-CoV-2 variants 143 To measure the neutralizing potential of the vaccinees' sera against all four SARS-CoV-2 isolates, 144 neutralizing antibody titres elicited by the BNT162b2 vaccine were analyzed with microneutralization 145 test (MNT). The neutralizing titres with two D614G isolates FIN-25 and SR121 were almost identical 146 9 both three weeks (p=0.02) and six weeks after the first dose (p=0.11) (Fig. 3A), indicating that the 147 mutations in FIN-25 spike protein due to initial propagation in VeroE6 cells did not affect the 148 neutralizing titres. 149 Before vaccination (0 day sampling) those 11/180 with a likely SARS-CoV-2 infection based on EIA 150 results, showed increased geometric mean titres (GMT) of 35, 31 (Fig. 3B). 171 The MNT titres for two D614G-containing isolates, FIN-25 and SR121, correlated very well, as also 172 did FIN-25 and 85HEL (B.1.1.7) (r>0.8, p<0.0001) (Fig. 4). MNT titres for FIN-25 and HEL-12-102 173 (B.1.351) correlated relatively well and highly significantly (r=0.74, p<0.0001), as did the two variants 174 of concern, 85HEL (B.1.1.7) and HEL-12-102 (B.1.351) (r=0.75, p<0.0001). 175 To analyze the effect of age and gender to the antibody responses, the vaccinees were divided into age 176 and gender groups and the S1 IgG EIA and MNT results were compared between the groups (Fig. 5A  177 and B). After the first vaccine dose, anti-S1 IgG antibody levels and neutralization titres decreased 178 significantly in the older age group (55-65 years) compared to younger age groups (20-34 and 35-44 179 years) (Fig. 5A). However after the second vaccine dose, the neutralization titres were similar between 180 the age groups (GMT 257, 268, 200 and 206 in age groups of 20-34, 35-44, 45-54 and 55-65 years, 181 respectively) (Fig. 5A). We also compared gender-related antibody responses even though male 182 vaccinees were underrepresented, comprising only 17% (29/169) of the vaccinees. After the second 183 dose, female vaccinees had slightly higher neutralization titres than male (p=0.0412), although the anti-184 S1 IgG antibody levels remained at the same level ( anti-S1 IgG and total anti-S1 Ig were compared to neutralization titres against FIN-25 (Fig. 6, 190 Supplementary Fig. 3). Both anti-S1 IgG and total anti-S1 Ig EIA measurements correlated very well 191 11 with MNT titres (r>0.9, p<0.0001) suggesting that EIA, especially IgG EIA, using spike protein as an 192 antigen could be a useful method to determine COVID-19 immunity. Humoural immune response to vaccinations has been shown to decline with age 39,40 . Consistently, we 256 observed a trend of declining immune response to the COVID-19 mRNA vaccine by age. This trend 257 was not very strong, presumably because the ages of our vaccinees ranged from 20 to 65 years, while 258 age-dependent immunosenescence should be more pronounced in the age group >65 years 39 . Another 259 explanation might be that the BNT162b2 mRNA vaccine is exceptionally immunogenic and therefore, 260 especially when given two doses, it enables practically all individuals regardless of gender and age, to 261 develop high antibody levels and neutralization titres. 262 In summary, in the present study we show that the Pfizer-BioNTech BNT162b2 COVID-19 mRNA 263 vaccine is highly immunogenic, and particularly after two vaccine doses all vaccinees showed very 264 high humoral immune response to D614G variant viruses. Immunity to a recent B. Expression and purification of SARS-CoV-2 nucleoprotein and S1 antigens 298 SARS-CoV-2 protein expression was done as described previously (Jalkanen et al. 2020, submitted). 299 Briefly, SARS-CoV-2 N and S sequences were obtained from GenBank (NC_045512.2 and 300 MN908947.3, respectively). N protein was expressed as a fusion protein with glutathione S-transferase 301 16 (N-GST) in Spodoptera frugiperda (Sf-9) cells. GST alone was produced to be used as a control 302 protein. S1 domain of the spike protein (amino acid residues 16-541) was expressed as a fusion protein 303 with mouse IgG2a Fc and 8xhistidine tag (S1-mFc-8xhis) in human embryonic kidney (HEK293F) 304 cells. Mouse promyostatin (ProMstn)-mFc(IgG2a)-6xhis (later referred as Mstn-mFc) was produced to 305 be used as a control protein. Proteins were purified and buffer was exchanged to PBS. Concentrations 306 of produced proteins were measured with BCA protein assay kit (Thermo Fisher Scientific). 307 IgG, IgA, and IgM EIA and total Ig EIA for SARS-CoV-2 S1 and N protein

378
Geometric means with geometric standard deviations (SD) were calculated with GraphPad Prism 8 379 software. Statistical significance of differences between variants were analyzed with Wilcoxon 380 matched pairs signed-rank test, and two-tailed p-values <0.05 were considered significant. Differences 381 between age and gender groups were tested with two-tailed Mann-Whitney U test. 382 A. BNT162b2 vaccinated health care workers were divided into four age groups. Age specific 541 differences of anti-S1 IgG antibody levels and neutralization titres against FIN-25 virus isolate were 542 analyzed. Sera was collected three weeks (3w) and six weeks (6wk) after the first vaccine dose. B. 543 Gender-specific differences in antibody responses and neutralization titres. IgG antibody levels are 544 represented as EIA units. Differences between age and gender groups were tested with two-tailed 545 Mann-Whitney U test. Two-tailed p-values *<0.05, **<0.01, ****<0.0001 were considered significant. 546 Fig. 6 Correlation of anti-S1 antibody levels with SARS-CoV-2 neutralization titres. 547 Table 1 Antibody responses in BNT162b2 vaccinated health care workers (HCW) and non-552 hospitalized convalescent phase COVID-19 patients. HCW samples were collected before 553 vaccination (0d), and three (3wk) and six (6wk; three weeks after the second vaccine dose) weeks after 554 the first vaccine dose. Geometric mean (GM) and number of positive samples for anti-S1 IgG and total 555 Ig, and anti-N IgG antibodies and neutralizing antibodies is indicated. In microneutralization test 556 (MNT) neutralization titre >20 was considered positive and for calculation of geometric means a value 557  Antibody responses against SARS-CoV-2 S1 and N proteins in BNT162b2 vaccinated health care workers and non-hospitalized recovered COVID-19 patients. A. Anti-S1 and B. anti-N IgG, IgA, IgM, and total Ig antibody levels were measured with EIA. Serum samples from BNT162b2 vaccinated participants (n=169) were collected before vaccination (0d), and three (3wk) and six (6wk) weeks after the rst dose of the vaccine. All vaccinees received the second dose of the vaccine three weeks after the rst dose.   (B.1.351) variants before (0d), three (3wk) and six weeks (6wk) after the rst dose of BNT162b2 vaccine and neutralization titres of convalescent sera of non-hospitalized patients. Values above the groups indicate geometric mean titres (GMTs) and data is shown as geometric means and geometric SDs.

Figure 5
Antibody responses against SARS-CoV-2 S1 protein and neutralization of B.1.1.7 and B.1.351 variants by age and gender. A. BNT162b2 vaccinated health care workers were divided into four age groups. Age speci c differences of anti-S1 IgG antibody levels and neutralization titres against FIN-25 virus isolate were analyzed. Sera was collected three weeks (3w) and six weeks (6wk) after the rst vaccine dose. B.
Gender-speci c differences in antibody responses and neutralization titres. IgG antibody levels are represented as EIA units. Differences between age and gender groups were tested with two-tailed Mann-Whitney U test. Two-tailed p-values *<0.05, **<0.01, ****<0.0001 were considered signi cant.

Figure 6
Correlation of anti-S1 antibody levels with SARS-CoV-2 neutralization titres. Anti-S1 IgG and total Ig antibody levels were determined with EIA and neutralization titres of BNT162b2 vaccinated health care workers (n=169) against FIN-25 virus isolate were obtained with microneutralization test (MNT). All sequential serum samples (0d, 3wk and 6wk) were included in the analysis. Spearman's rank correlation coe cient (r) is indicated.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. SupplementaryInformation.pdf