Reduced neutralization and Fc effector function to Omicron subvariants in sera from SARS‐CoV‐1 survivors after two doses of CoronaVac plus one dose subunit vaccine

Severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) Omicron harbors more than 30 mutations of the spike protein and exhibits substantial immune evasion. Although previous study indicated that BNT162b2 messenger RNA vaccine induces potent cross‐clade pan‐sarbecovirus neutralizing antibodies in survivors of the infection by SARS‐CoV‐1, the neutralization activity and Fc‐mediated effector functions of these cross‐reactive antibodies elicited in SARS‐CoV‐1 survivors to Omicron subvariants still remain largely unknown. In this study, the neutralization activity and Fc‐mediated effector functions of antibodies boosted by a third dose vaccination were characterized in SARS‐CoV‐1 convalescents and healthy individuals. Potent cross‐clade broadly neutralizing antibodies were observed in SARS‐CoV‐1 survivors who received a three‐dose vaccination regimen consisting of two priming doses of CoronaVac followed by one booster dose of the protein subunit vaccine ZF2001. However, the induced antibodies exhibited both reduced neutralization and impaired Fc effector functions targeting multiple Omicron subvariants. Importantly, the data also support the notion that immune imprints resulted from SARS‐CoV‐1 infection may exacerbate the impairment of neutralization activity and Fc‐mediated effector functions to Omicron subvariants and provided invaluable information to vaccination strategy in future.


| INTRODUCTION
The causative agent of coronavirus disease 2019 (COVID- 19), severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has spread rapidly around the world since the first case was identified in December 2019. 1 As SARS-CoV-2 continues to disseminate, it has evolved and mutated, producing variant strains with enhanced transmissibility and immune evasion, such as Alpha (B.1.1.7),Beta (B.1.351),Gamma (P.1), Delta (B.1.617.2), and Omicron (B.1.1.529). 2 Omicron, a variant identified in South Africa in November 2021, is designated as a variant of concern (VOC) by the World Health Organization.Although the pronounced transmissibility and immune evasion of Omicron have attracted significant attention worldwide, it continues to evolve and mutate, particularly in the spike (S) gene.In January 2022, the initial Omicron strain BA.1 was replaced with BA.2, followed by various subvariants, including BA.2.12.1, BA.3, BA.4/5, BQ.1.1,XBB, and most recently EG.5, a descendant of the XBB strain, which has rapidly spread and become predominant worldwide. 2BF.7 and BA.5.2 dominate the first wave of outbreak in China, accounting for 90% of local infections that occurred after China optimized its epidemic response.Despite following different evolutionary paths, Omicron variants share mutations in certain hotspot regions of their receptor-binding domain (RBD) and these convergent mutations enable the virus to effectively evade neutralizing antibody (nAb) and convalescent plasma, while maintaining sufficient ACE2-binding affinity, [3][4][5][6] which brings serious challenges to the effectiveness of humoral immunity developed in response to vaccination and infection.
In addition to mediating neutralization that directly block viral infection, antibodies drive effector functions by engaging cellular receptors via their constant (Fc) region to eliminate virus-infected cells.These functions include antibody-dependent cellular cyto- host cells is mediated by the viral S protein and ACE2 receptor. 11The SARS-CoV-1 S protein shares about 76% amino acid identity with the SARS-CoV-2 S protein, and the amino acid sequence of the potential RBD of the SARS-CoV-1 S protein is only about 74% homologous to that of prototype strain of SARS-CoV-2. 12A previous study indicated that the BNT162b2 messenger RNA (mRNA) vaccine induces potent cross-clade pan-sarbecovirus neutralizing antibodies in patients who have survived infection with SARS-CoV-1. 13However, recent studies observed that Omicron subvariants, such as BA.2, BA.2.12.1 and BA.4/5, display increased evasion of these cross-reactive neutralizing antibodies. 14though previous studies indicated that SARS-CoV-1 infection

| Ethical statement
The study protocol was approved by the Ethics Committee of the Institute of Beijing Ditan Hospital, Capital Medical University (IRB#2021-(024)−02).Written informed consent was obtained from all participants before enrollment in the study.This clinical trial was registered with the Chinese Clinical Trial Registry under registration number ChiCTR2100051998.

| Serum samples
Serum samples were provided by Beijing Ditan Hospital.The samples were taken from two groups of participants (Supporting Information:  S1.

| Cell transfection and pseudotyped virus production
The pSectag2 vector was used to construct recombinant plasmids encoding codon-optimized spike proteins from the SARS-CoV-1(Tor2), WIV-1, and SARS-CoV-2 prototype strain (Wuhan Hu-1 reference strain containing D614G mutation) and variants, with a 19-amino acid truncation at the C-terminus of the spike protein (the mutations are shown in Supporting Information: Figure S2). 15K-293T cells were transfected with the plasmids encoding different S proteins.After 6 h of transfection, VSV-ΔG-G*-Luc pseudovirus (Kerafast) was added and after 24 h of transfection the supernatant was replaced with fresh complete Dulbecco's modified Eagle medium.Supernatants were collected at 48 and 72 h after transfection, passed through a 0.45 μm filter, aliquoted and stored at −80°C.

| Pseudotyped virus titration
For titration, TREx-293/hACE2 cells were seeded into 96-well plates with 2 µg/mL Tet (~2 × 10 4 cells per well) and incubated at 37°C overnight.The following day, the pseudovirus stocks were thawed and diluted into the 96-well plates containing cell growth medium (100 µL per well), starting with a 10-fold dilution into the first set of wells and followed by eight serial threefold dilutions; six replicates were performed for each dilution.Another six wells containing growth medium only were used as blank controls.After incubation for 18 h, the luciferase substrate was added for chemiluminescence detection.The 50% tissue culture infectious dose (TCID 50 ) of the pseudovirus was calculated using the Reed-Muench method and the cutoff value was set at 10 times the value of blank control.

| Neutralization assay
The neutralization assay was performed as previously described. 16,17iefly, serum samples were inactivated at 56°C for 30 min and serially diluted threefold, commencing with a 30-fold dilution; two replicates were performed for each dilution.Control wells containing virus and cells only (with no sera) were included on each plate.
Equivalent pseudovirus (13 000 TCID 50 /mL) was incubated with the sera at 37°C for 1.5 h and the mixture was then added to the 96-well plate containing TREx-293/hACE2 cells.After 18 h, the neutralization assay was developed with a luciferase assay system (Promega) and the relative light units (RLUs) were read on a Promega GloMax Luminometer.The neutralization rate (%) was calculated as follows: Neutralizing antibody titers are expressed as the serum folddilution required to achieve 50% pseudovirus neutralization (pVNT 50 ).The pVNT 50 value was interpolated from the neutralization curves and determined using the log(inhibitor) versus normalized response-variable slope fit using the automatic outlier detection function of the GraphPad Prism software program.

| ADCC and ADCP assay
Jurkat-NFAT-hCD16 or Jurkat-NFAT-hCD32 effector cells, kindly provided by Professor Ningshao Xia, were used to assess ADCC and ADCP activities.In this assay, target antigens produced on the surface of 293T cells are used to determine the capacity of sera to activate the nuclear factor of activated T cells (NFAT) pathway via FcRIIIa (CD16a) or FcRIIa (CD32a) (the route that activates ADCC in NK cells and ADCP in macrophages). 18,19Briefly, 293T targeting    vaccines (Figure S4).Taken together, these findings imply that an immune imprint from prior SARS-CoV-1 infection results in a broad neutralizing response to both SARS-CoV-1, WIV-1, and SARS-CoV-2 VOCs, except for Omicron (BA.1) in response to booster vaccination.

| Omicron subvariants exhibit stronger neutralizing antibody evasion
To assess the ability of the recently emerged Omicron subvariants to evade neutralizing antibodies, neutralization assays against SARS-CoV-1, WIV-1, D614G, Delta, and Omicron subvariants (including BA.1, BA.2, BA.2.12.1, BA.4/5, and BF.7) were performed using serum samples from SARS-CoV-1 survivors who had received two doses of CoronaVac vaccine and one dose of ZF2001 vaccine.As expected, among the individuals in both groups, neutralization titers for the Delta variant were lower than for the D614G variant.
Although receipt of a booster of the ZF2001 vaccine may increase the level of cross-neutralizing antibodies to the Omicron subvariants, the serum samples from vaccinated persons in both panels neutralized the Omicron variants to a much lesser extent than D614G and Delta.In healthy individuals who had received the ZF2001 vaccine booster, BA.1 and BA.2 showed no significant difference in resistance to neutralization by sera (Figure 2A) concordant with a previous report. 22However, it was found that the newly emerged Omicron subvariants BA.2.12.1, BA.4/5, and BF.7 showed increased immune evasion capability over BA.1 in both cohorts of healthy controls and SARS-CoV-1 survivors (Figure 2A,B), which agrees with a previous report. 14There was no significant difference in the neutralization titers between SARS-CoV-1 survivors and healthy controls, although the GMTs of neutralizing antibodies against Omicron variants in SARS-CoV-1 survivors are lower than that of healthy controls (Figure 2C).To further evaluate the difference in neutralizing activity between these two groups, we compared neutralization rate (at a dilution of 1:20) and a significant difference between the two panels was observed.Interestingly, the neutralization rate of the human serum specimens against BA.2.12.1, BA.4/5, and BF.7 tended to be lower in the SARS-CoV-1 survivor group (Figure 2D).The frequency of serum samples with a neutralization rate over 50% (indicated as positive) was also compared between healthy controls and SARS- CoV-1 survivors.The frequency of positive serum against BA.2.12.1, BA.4/5, and BF.7 in SARS-CoV-1 survivors statistically lower than that in healthy controls (Supporting Information: Figure S5).Taken together, the aforementioned data indicated that multiple Omicron subvariants exhibit a stronger ability to evade neutralizing antibodies, particularly in the group of SARS-CoV-1 survivors.

| Neutralizing antibody titers decay over time
To evaluate the decline of immunity against SARS-CoV-2 over time in both cohorts, neutralizing antibody titers against SARS-CoV-1, WIV-1, and three variants of SARS-CoV-2 were compared at Day 14 and Day 90 following administration of the third vaccine dose. 20,23At Day 90 post the third vaccination, the neutralization GMTs towards tested SARS-CoV-2 variants were dramatically lower than those measured at Day 14 in both two groups (Figure 3).In the healthy control group, the neutralization GMTs decreased from Day

| Newly emerging Omicron subvariants evades Fc-mediated effector function
In addition to neutralizing activity, spike-binding antibodies also have antibody Fc effector functions that are known to contribute to reduced disease severity and increased vaccine efficacy. 24 (Supporting Information: Figure S6).In similarity to nAb response, significantly higher ADCC activity against SARS-CoV-1 and WIV-1 compared with D614G was detected in SARS-CoV-1 survivors, but not in healthy individuals (Figure 4), indicating that the cross-reactive antibodies induced in SARS-CoV-1 survivors mediated ADCC function towards both SARS-CoV-1 and SARS-CoV-2 D614G.

| DISCUSSION
This study investigated the antibody responses against Omicron subvariants in healthy individuals and SARS-CoV-1 convalescents who received two doses of the heterologous CoronaVac vaccine followed by the ZF2001 RBD recombinant protein subunit booster vaccine.A previous study showed that the production of potent cross-clade pan-sarbecovirus neutralizing antibodies is induced in SARS-CoV-1 survivors who have been immunized with the BNT162b2 mRNA vaccine. 13In the study, it was observed that SARS-CoV-1 survivors exhibited reduced neutralization and Fc  Interestingly, a recent study reported that administering a SARS-CoV-1-based booster after two priming doses of the BNT162b2 vaccine results in enhanced breadth and durability of neutralizing response.These findings suggest immune imprinting from prior SARS-CoV-1 infection 18 years ago and employing SARS-CoV-1 as a booster yields divergent outcomes of antibody response. 25One limitation in this study is that the two groups were not matched for toxicity (ADCC), cellular phagocytosis (ADCP), antibody-dependent cellular trogocytosis, and complement deposition.ADCC and ADCP are mediated by the Fab segment of the antibody targeting the epitopes of virus-infected cells, and meanwhile, the Fc segment binds to the Fc receptors (FcRs) on the surface of killer cells (e.g., natural killer [NK] cells and macrophages), inducing them to directly kill or phagocytose target cells.Neutralizing antibody responses have been demonstrated to preferentially target the RBD of the SARS-CoV-2 S protein or certain sites of S proteins; however, theoretically, antibodies can fulfill Fc effector function by binding to the entire surface of the spike.Several studies indicate that antibodies generated in response to SARS-CoV-2 vaccination and/or infection may promote Fc effector functions, which could contribute to preventing infection with SARS-CoV-2. 7-10SARS-CoV (here designated as SARS-CoV-1, to distinguish it from SARS-CoV-2) is a coronavirus that caused the global SARS outbreak in 2003.Similar to SARS-CoV-2, SARS-CoV-1 entry into affected the antibodies response to SARS-CoV-2 vaccination and/or infection, it is still largely unknown how the imprinted immunity resulted from SARS-CoV-1 infection impact the antibody response induced by SARS-CoV-2 vaccination, including neutralization and Fc-mediated effector functions, to SARS-CoV-2 Omicron subvariants.In this study, features of neutralization and Fc-mediated effector function were characterized in serum from patients recovering from SARS-CoV-1 and from healthy individuals, before and after receiving the third vaccine dose against SARS-CoV-2.Reduced neutralization and impaired Fc effector function targeting multiple Omicron subvariants were observed in sera from the above two panels of participants after three doses of vaccination, which may have been exacerbated by immune imprinting from SARS-CoV-1 infection.

Figure
Figure S1): (1) SARS-CoV-1 survivors who had received three doses of heterologous vaccines (two priming doses of CoronaVac 28 days apart 4-8 months earlier, followed by one booster dose of the cells were transfected with plasmids encoding the S protein with N-terminal Myc tagged from the SARS-CoV-1 (Tor2), WIV-1, SARS-CoV-2 D614G prototypical strain, Delta variant, Omicron subvariants using Lipofectamine 3000 (Invitrogen).After 48 h of transfection, the cells were collected for immunostaining with anti-Myc mouse mAb (Cell Signaling Technology).An Alexa-488-labeled goat antimouse IgG (H + L) secondary antibody (Abcam) was used for detection.The fluorescent signal was examined using flow cytometer.For all ADCC or ADCP assays, target cells (2 × 10 5 per well) were incubated with 1:100 dilutions of serum samples at 37°C for 1 h.ADCC or ADCP effector cells (2 × 10 5 per well) were then added to each well.After incubation at 37°C for 18 h, the RLUs were detected in accordance with the instruction manual provided by QuantiLuc (InvivoGen).The luciferase (Luc) fold induction (ADCC or ADCP Luc induction fold) was calculated as follows: RLU (induced-background)/ RLU (no serum control-background).

3 | RESULTS 3 . 1 |
Administration of a third COVID-19 vaccine dose in SARS-CoV-1 survivors increased neutralizing breadth covering SARS-CoV-1 and SARS-CoV-2 VOCs except OmicronTo determine whether booster vaccination induces the production of broadly neutralizing antibodies in SARS-CoV-1 survivors, neutralization assays were conducted using pseudoviruses that displayed the spike protein from SARS-CoV-1, WIV-1, D614G, Delta, or Omicron BA.1.The administration of a third dose of the ZF2001 vaccine was observed to elicit a rapid and significant increase in humoral immune response.The GMTs of neutralizing antibodies against the three SARS-CoV-2 variants were significantly increased at Day 14 after administration of the third dose in both SARS-CoV-1 survivor panel and healthy control panel, in consistence with third dose booster studies (Figure1).20,21A boosting of anti-SARS-CoV-1 and anti-WIV-1 neutralizing antibodies were also observed in SARS-CoV-1 survivors (Figure1B) but not in healthy controls (Figure1A).Importantly, the Omicron neutralization titer was dramatically lower than that to D614G, indicating that the third dose of RBD subunit vaccine of SARS-CoV-2 administrated in SARS-CoV-1 survivors induced cross-reactive antibodies that exhibited increased neutralizing breadth covering both SARS-CoV-1 and SARS-CoV-2 early variants, but not Omicron variants (Figure1and Supporting Information: FigureS3).

F I G U R E 1
Boosting of cross-clade neutralizing antibodies against severe acute respiratory syndrome coronavirus 1 (SARS-CoV-1) and SARS-CoV-2 variants of concern (VOCs) in SARS-CoV-1 survivors.Results of pseudovirus neutralization assays using participants' sera against SARS-CoV-1, SARS-like CoV WIV-1, SARS-CoV-2 D614G, Delta, and Omicron (BA.1) variants the day before and 14 days after the third vaccination in healthy controls (A) and SARS-CoV-1 survivors (B).Neutralizing antibody titers are expressed as sera fold dilution required to achieve 50% pseudovirus neutralization (pVNT50).Dots indicate individual serum samples, dark horizontal lines for each group denotes (GMTs).pVNT50 below the quantitative range but still within the qualitative range (i.e., partial inhibition is observed but a dose-response curve cannot be fit because it does not sufficiently span the pVNT50) was counted half (15) of limit of detection (LOD) and no inhibition at all was counted as 1 in statistical analysis.The error bars indicate the 95% confidence intervals (95% CIs) and the dashed lines indicate the lower LOD (30).The fold increase in GMT after the third vaccination is represented as a number with the "x" symbol.There was no significant difference in GMTs against SARS-CoV-2 VOCs at Day 14 after the third vaccine dose between SARS-CoV-1 survivors and healthy controls (Figure1).However, at Day 0 of the experiment, which corresponded to 4-8 months after receiving two doses of CoronaVac and just before receiving the third vaccination dose, more SARS-CoV-1 survivors had detectable levels of neutralizing antibodies (ID50 > 30) against D614G and Delta (D614G 66.7% and Delta 16.7%) than healthy controls (D614G 13.3% and Delta none).Meanwhile, certain amounts of neutralizing antibodies towards SARS-CoV-1 (Tor2) and WIV-1 were also observed in SARS-CoV-1 survivors but not in healthy controls, indicating that cross-reactive antibodies were maintained in SARS-CoV-1 survivors 4-8 months after two doses of CoronaVac.Of note, SARS-CoV-1 or WIV-1 neutralization was still detectable 18 years after the initial SARS-CoV-1-infection before the administration of any SARS-CoV-2 14 to Day 90 by 1.8-fold (D614G), 1.5-fold (Delta), and 2.5-fold (Omicron BA.1).In SARS-CoV-1 survivor group, the GMTs decreased from Day 14 to Day 90 by 2.2-fold (D614G), 1.9-fold (Delta), and 2.4-fold (Omicron BA.1).These data demonstrated a waning of neutralizing antibody response towards SARS-CoV-2 variants over time in both groups.Interestingly, it was observed that the neutralization GMTs towards SARS-CoV-1 has minimal decease from Day 14 to Day 90 in SARS-CoV-1 survivors, indicating that the antibodies induced by the ZF2001 vaccine in SARS-CoV-1 survivors maintained a steady and higher level of neutralizing activity towards SARS-CoV-1 than SARS-CoV-2 within 90 days.
However, the impact of prior SARS-CoV-1 infection on Fc effector responses is unknown.To assess ADCC and ADCP activities against various sarbecoviruses, including SARS-CoV-1, WIV-1, and SARS-CoV-2 VOCs (D614G, Delta, and Omicron subvariants), the capacity of serum antibodies to induce cross-linking between effector cells expressing FcγRIIIa or FcγRIIa and target cells expressing the indicated spikes on the surface was detected in both groups, as a surrogate for ADCC or ADCP assay.S proteins used in this study are expressed equally at the cell surface, with percentages of S proteinpositive cells ranging from 32.1% to 44.0% and mean fluorescence intensity values ranging from 15969 to 21030 among target cells
effector function against SARS-CoV-2 Omicron subvariants in their sera compared to healthy controls after receiving three vaccine doses.The effect of previous exposures to human coronaviruses on immune responses to SARS-CoV-2 vaccination provide important implication for vaccine design and viral epidemiology.This immunological imprint resulted from the previous exposure encompass two underlying mechanisms: immunological memory and cross-reactivity.Previous studies revealed that BNT162b2 mRNA vaccine applied in SARS-CoV-1 survivors recalled cross-reactive memory B cells response and potent cross-clade pan-sarbecovirus neutralizing antibodies that efficiently block the infection driven by the S protein from both SARS-CoV-1-like and multiple SARS-CoV-2 VOCs, including B.1.1.7 (Alpha), B.1.351(Beta), and B.1.617.2 (Delta), indicating that imprint from previous SARS-CoV-1 infection extended the neutralization breadth.In line with their finding, it was found in this study that a third dose of booster vaccination of the protein subunit vaccine ZF2001 in this SARS-CoV-1 survivor cohort (people recovered from SARS-CoV-1 infection 18 years ago) induced a high level of neutralizing antibodies against both SARS-CoV-1 and SARS-CoV-2 including D614G and Delta.The rise in SARS-CoV-1 antibody levels following SARS-CoV-2 vaccination may be attributed to the activation of cross-reactive memory B cells that were generated in response to prior SARS-CoV-1 infection.Importantly, this study revealed that Omicron and its sub-lineages, such as BA.

F I G U R E 3
The neutralizing antibody titers decay over times.Pie charts show the proportion of vaccinees within each group with detectable neutralization against the indicated severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pseudovirus at 14 and 90 days following the third dose of vaccination.Fold-decrease in geometric mean titer (GMT) at Day 14 of D614G, Delta, and Omicron BA.1 relative to Day 90 within healthy controls (A) and SARS-CoV-1 survivors (B) (shown as a number with the "x" symbol).Statistical significance was analyzed by Wilcoxon matched-pairs signed-rank test.50% pseudovirus neutralization (pVNT50) below the lower limit of detection (<30) was recorded as 15 and no inhibition at all was counted as 1 in the geometric mean calculation.*p < 0.05, **p < 0.01, ***p < 0.001 and ns = nonsignificant.age and the median age of individuals in SARS-CoV-1 survivor group (62 years) is higher than healthy control group (39 years), which might contribute to the compromised antibody responses in the former group.However, there was no difference between healthy controls and SARS-CoV-1 survivors in neutralizing antibodies against D614G and Delta, suggesting that the age mismatch may not play an obvious effect.Collectively, SARS-CoV-1 imprint broadens the neutralization breadth against pan-sarbecoviruses such as SARS-CoV-1, WIV-1, and SARS-CoV-2 (D614G and Delta).On the other hand, it exacerbates the immune evasion of newly emerging Omicron subvariants.This study observed a waning of neutralizing antibody activities over time in the 3rd-dose immunized subjects toward all SARS-CoV-2 VOCs tested in both healthy control and SARS-CoV-1 survivor cohorts.Surprisingly, the decay of neutralizing activity toward SARS-CoV-1 was much slower than the one to SARS-CoV-2 variants tested in SARS-CoV-1 survivors, although they have a similar level of neutralizing activity toward SARS-CoV-1 and SARS-CoV-2 D614G at 14 days after third dose of boosting, which indicated that the preexposure to SARS-CoV-1 may induce an antigenic bias toward SARS-CoV-1, an original strain encountered 18 years ago.Consistent with this finding, sera from 4 to 8 months post second-dosevaccinated SARS-CoV-1 survivors showed a marked higher neutralization titer toward SARS-CoV-1 compared with that toward SARS-CoV-2 D614G.Collectively, although SARS-CoV-2 vaccination in SARS-CoV-1 survivors extend the neutralization spectrum to encompass both original strains (SARS-CoV-1) and newer strains (SARS-CoV-2 D614G), reactivity toward the original strain (SARS-CoV-1) is maintained at higher levels than reactivity toward newer F I G U R E 4 Antibody-dependent cellular cytotoxicity (ADCC) and cellular phagocytosis (ADCP) activities against severe acute respiratory syndrome coronavirus 1 (SARS-CoV-1), WIV-1, SARS-CoV-2 variants of concern (VOCs) and Omicron subvariants in healthy controls and SARS survivors after the third vaccination (Day 14).ADCC activity in healthy controls (A) and SARS-CoV-1 survivors (B) is shown as the induction fold through FcγRIIIa-expressing cells.(C) Bars show geometric mean of ADCC Luc-induction fold with 95% confidence interval (95% CI) for healthy controls (blue) and SARS-CoV-1 survivors (red) individuals against VOCs.ADCP activity in healthy controls (D) and SARS-CoV-1 survivors (E) is shown as the Luc-induction fold through FcγRIIa-expressing cells.(F) Bars show geometric mean of ADCP Luc-induction fold with 95% CI for healthy controls (blue) and SARS-CoV-1 survivors (red) individuals against VOCs.Dotted lines indicate the limit of detection of the particular assay.Statistical significance across variants is shown by Wilcoxon signed-rank test and statistical significance between samples of healthy controls and SARS-CoV-1 survivors is shown by Mann-Whitney test.*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 and ns = nonsignificant.