Sputnik V-induced Antibodies against SARS-CoV-2 Variants during the Dissemination of the Gamma Variant in Venezuela

1. Introduction

By March 2024, SARS-CoV-2, the virus responsible for the COVID-19 pandemic, had caused more than 700 million cases and claimed over 7 million lives [1], despite estimations suggesting even higher figures [2,3]. Notably, this viral family possesses a unique non-structural protein with proof-reading capacity, setting it apart from other RNA viruses. The virus has exhibited many mutations, due to numerous replication cycles in millions of hosts, a high frequency of recombination, and the effect of host editing enzymes [4].

Consequently, variants with higher transmission capacity and immune evasion have been selected [5]. Five variants of concern (VOCs) of SARS-CoV-2 have emerged since the end of 2020, resulting in new waves of infections, deaths, and chronic sequelae [6]: Alpha (Original lineage B.1.1.7), which arose in the UK, Beta (B.1.351), which appeared in South Africa, Gamma (B.1.1.28.2, P.1), which originated in Brazil, Delta (B.1.617.2), from India and Omicron (B.1.1.529) first identified in South Africa and whose sub-lineages are the only ones circulating at present [7]. Furthermore, the antigenic evolution of the S protein has dampened the efficacy and effectiveness of many vaccines [8].

Non-replicating adenoviral vectors, mRNA, whole inactivated viruses, and protein subunit-based vaccines are powerful tools in the fight against emergent pathogens, including SARS-CoV-2. Prior to the introduction of bivalent adapted boosters, most vaccine-approved candidates targeted the S protein of the ancestral variant first identified in China [9]. The Gam-COVID-Vac (Sputnik V) vaccine has played a significant role in vaccination campaigns in Venezuela and other Latin American countries. Studies have consistently shown that it is as effective as mRNA vaccines [10]. Sputnik V is composed of two non-replicative recombinant human adenoviruses (serotypes 26 and 5), both carrying the coding sequence for the ancestral S protein. The vectors have deletions in the early expression genes E1 and E3, which prevents its replication and admit the introduction of the gene coding for SARS-CoV-2 S glycoprotein. Additionally, the heterologous design of Sputnik-V is advantageous as it reduces immunogenicity issues arising from immunity against the vector [11]. Upon delivery, each antigenic stimulus produces antibodies through plasmatic cells, diversifies the B cell receptor (BCR) repertoire, and generates specific B memory cells against the entire S glycoprotein [12-14].

Understanding how the recognition and neutralization of antibodies generated after Sputnik-V vaccination might be affected by variants of this virus is an issue that requires further research. In contrast to mRNA-based vaccines, the effect of Sputnik V vaccine has been significantly less studied. In fact, a PubMed query shows the scarce number of publications about the Sputnik V vaccine compared to other most common ones, even though its use was approved in more than 70 countries (Table 1). In Venezuela, a clinical trial was conducted using the Sputnik V vaccine, from December 2020 to June 2021, time which coincide with the emergence of the Gamma VOC in the country [15]. The aim of this study was to evaluate the antibody reactivity and neutralization ability to SARS-CoV-2 variants and to assess the correlation between these parameters in the vaccinees in Venezuela.

2. Materials and Methods

2.1. Experimental Design, population and sample

Sera samples from the prospective, double-blind, randomized, placebo-controlled clinical trial of Sputnik V in Venezuela from December 2020 to July 2021 (NCT04642339) were evaluated. The inclusion and exclusion criteria are detailed on the clinical trial website (clinicaltrials.gov/ct2/show/NCT04642339). Volunteers were residents of Caracas, aged 20-80 years, and all signed an informed consent form approved by the National Ethics Committee for Research in COVID-19. Samples were collected 42 days after vaccination (dpv), counting after the first dose and 21 days after the second dose. Matching samples from 46 vaccinees were also collected at 180 dpv. Of the 133 volunteers, 105 were vaccinated and 28 received placebo. Sera from 85/105 vaccinees without serological evidence of exposure, as determined by nucleoprotein (N) reactivity [18], were analyzed for reactivity to S and RBD antigens at 42 dpv. For comparisons of 42 versus 180 dpv reactivity, 20 sera from N-negative vaccinees were selected. 18 samples from this subgroup were used for neutralization assays at 42 and 180 dpv.

2.2. Antigens

The recombinant antigens for the enzyme-linked immunosorbent assays (ELISA) were acquired from MyBioSource Inc. (San Diego, CA, USA). The ancestral S protein (MBS8574721) and RBD (MBS8574741), Alpha S (MBS184025), Beta S (MBS184021), Gamma S (MBS184022) and RBD (MBS434292), Delta S (MBS184024) and Omicron S (MBS553745) antigens, were employed. The concentration of antigens was determined using Qubit™ Protein Assay according to manufacturer’s specifications (catalog number Q32866, Thermo Fisher, Waltham, MA, USA).

2.3. Reactivity towards SARS-CoV-2 antigens

To assess the differences between reactivity against S or RBD from SARS-CoV-2 variants, an adapted version of the protocol proposed by Stadlbauer et al., 2020 was used [19] and thoroughly described by Cornejo et al., 2024 [18]. Briefly, 96-wells microtiter plates were coated using 1 µg/mL of each capture antigen. To prevent unspecific antibody binding commercial blocking solution was employed (Abcam, Cambridge, United Kingdom, ab126587). All samples were diluted 1/100 and incubated for 2 hours at 37 °C. Horseradish peroxidase (HRP) IgG-linked (Jackson ImmunoResearch Inc, West Grove, PA, USA,) was used as a detection antibody (1/70,000) and incubated for 1 hour. TMB (3,3′,5,5′-tetramethylbenzidine) was used as a chromogenic substrate of HRP. All washes were done with 0.01% PBS-Tween. HCl [3M] was used to stop the reaction. Data gathering was achieved with a spectrophotometer (SpectraMax 250, Hampton, NA, USA) at 450 nm. Reactivity against SARS-CoV-2 antigens was assessed within this group of 134 individuals. The optical densities (O.D.) of the blank were subtracted from the O.D of the samples. O.D exceeding the cut-off established with negative control mean plus 3 standard deviations were considered as responders to each antigen. The negative control was established with 18 prepandemic sera from apparently healthy individuals. A mixture of two sera from individuals with hybrid immunity (Sputnik V-vaccinated and two confirmed infections by PCR) was used as a positive control. Both controls were used during normalization to estimate the relative IgG levels using the sample-to-positive ratio (S/P) previously described [20]:

$$\frac{S}{P}=\frac{\left(O.D.sample-O.D.negativecontrol\right)}{(O.D.positivecontrol-O.D.negativecontrol)}\mathrm{*}100$$

2.4. Plaque Reduction Neutralization Test of Sputnik V vaccinees sera

A plaque reduction neutralization test (PRNT) was conducted on 18 samples of serum collected at 42 days post-vaccination (dpv) and 180 dpv. PRNT was performed according to a previously reported procedure [18,21]. VERO C1008 cells (Vero 76, clone E6, vero. ATCC, Manassas, VA, USA), were maintained at 37 °C and 5% CO₂ in RPMI medium (Thermo Fisher Scientific, 11875093) supplemented with 10% fetal bovine serum (FBS, Thermo Fisher Scientific, 16000044) and 1% penicillin–streptomycin (Thermo Fisher Scientific, Waltham, MA, USA). All procedures were performed in a biosafety cabinet class II (ESCO^®, Airstream^®, Singapore) within a level 3 biosecurity laboratory. Viral seeds of the different SARS-CoV-2 variant strains: ancestral (B.1.1.33), P.1 (Gamma VOC), AY122 (Delta VOC), B.1.621 (Mu VOI) and BA.1.1 (Omicron VOC) [18], were incubated for 30 min at 37 °C with different dilutions of vaccinated and control sera. The infection process was performed for one hour at 37 °C in a 5% CO₂ atmosphere, and the cells were washed twice with PBS to remove the non-internalized viruses. The cells were overlayed with 0.5% carboxymethyl cellulose in a culture medium and incubated for 72 h. Then, the cells were fixed with 4% formaldehyde and stained with crystal violet to count the number of lytic plaques under each condition. Highly responsive mixed of hybrid sera was used as a positive control whereas culture medium without serum or viral seeds was used as the mock control.

The PRNT50 of each serum was determined through a nonlinear regression test, defining this value as the reverse of the dilution at which 50% of the virus is neutralized. After lytic plaque counting the percentage of neutralization for each serum dilution was calculated using the following formula:

$$Neutralization=\frac{\left(numberoflyticplaquesperserum\right)}{\left(averagenumberoflyticplaquespresentintheinfectioncontrol\right)}\mathrm{*}100$$

Afterwards, Log2 transformed reciprocal of each serum dilution (X axis) vs the percentage of neutralization (Y axis) was plotted and fitted to a non-linear regression variable slope dose-response inhibition model to determine the dilution needed to reduce 50% of lytic plaque formation observed in the infection control (IC).

2.5. Statistical analysis

To evaluate differences between reactivities against the different antigens at 42 dpv the Kruskal Wallis test (p < 0.05) was used with Dunn’s post-hoc analysis. For comparisons between 42 and 180 dpv a Wilcoxon matched-pairs signed rank test was carried out to estimate significant differences of reactivity between both times (p < 0.05). To establish differences among neutralization titers towards the different SARS-CoV-2 isolates within 42 and 180 dpv a Wilcoxon matched-pairs signed rank test was conducted (p < 0.05, PRISM GraphPad 9.0 ©, California, USA). Statistical difference was assessed with Chi square test (Epi Info™, Centers for Disease Control and Prevention, Atlanta, USA) for frequency comparisons among groups. Association between vaccinees ages, IgG antibody reactivity and PRNT50 was determined through a hierarchical clustering correlation matrix heatmap by the Spearman method with a confidence interval (CI) of 95% employing the “visdat” and “ggcorrplot” packages of R version 4.3.3. To explore patterns of antibody response an ascending hierarchical classification of the individuals was performed using the “shiny” package of R version 4.3.3.

3. Results

3.1. Reactivity to SARS-CoV-2 variants S protein

The reactivity to S from the 5 VOCs was assessed by ELISA in 85 vaccinated volunteers, without evidence of exposure to the virus. Statistically significant differences were observed between the mean reactivities of the S protein of the different variants. Reactivity to the S protein of the Alpha and Delta variants was higher than that observed for the ancestral variant (p <00001). No significant differences were observed between the reactivity to the S protein of the ancestral variant and that of the beta and gamma variants. The mean reactivity to the ancestral variant was higher (3-fold) than the one to the Omicron variant (p <00001). Almost all the sera reacted to the S protein of the different variants, except for the Omicron one, which was only recognized by 67% of the sera, a frequency statistically lower (p<0.0001) (Figure 1).

The reactivity to S VOCs was also assessed in follow up sera. The S/P values were lower at 180 dpv, with a reduction of 3-fold (S ancestral), 2-fold (S alpha, beta, gamma and delta) and 3-fold (S omicron). However, no reduction was observed in the number of sera recognizing each variant: an even higher frequency of sera recognizing the S protein was observed. The exception was the Omicron variant, for which even less sera recognized the protein, although this difference was not significant (Figure 2).

3.2. Neutralizing activity of Sputnik-V-induced antibodies against SARS-CoV-2 variants at 42 and 180 dpv.

Plaque reduction neutralization titer was assessed on 18 sera taken at 42 and 180 dpv from vaccinees with no signs of infection by the virus (Figure 3A). A modest reduction in the neutralization titer was observed with the Delta variants at 42 and 180 dpv, compared to the ancestral one, while the mean titer of neutralization against the Gamma variant was higher compared to the ancestral one, although this difference was not significant. In the case of Omicron, a significant reduction in neutralization titer was observed at 42 and 180 dpv, of 6 and 7 times, respectively, in addition to a significant reduction in the number of sera neutralizing the variant.

When dissecting the neutralizing activity against the Gamma variant, a group of sera (n = 10) exhibited a modest decrease in neutralizing titers against it at 42 and 180 dpv (Figure 3B), while a 3.7-fold increase in neutralization titer was observed against this variant in the sera of the other group (n = 8), at 42 and 180 dpv compared to the ancestral variant (Figure 3C).

Because of the unexpected higher neutralization activity in some sera of the Gamma variant respect the ancestral strain, the reactivity to the S and RBD antigens of the SARS-CoV-2 variants was evaluated for the sera analyzed for neutralizing activity. A statistically significant higher reactivity to the S Gamma variant, compared to the ancestral protein, was observed in the sera exhibiting higher neutralizing activity against this variant. This difference was not observed in the reactivity to RBD (Figure 4B and D).

3.3. Correlation between age, reactivity and PRNT50.

The Spearman test was performed to correlate all the parameters analyzed in this study. There was no significant correlation between antibody levels, neutralization titers and the age of the vaccinees. Regarding associations between ELISA reactivity and PRNT50, strong (S gamma vs PRNT50 gamma) to moderate positive correlations were observed at 42 dpv. Conversely, moderate (S omicron vs PRNT50 omicron) to low correlations were observed at 180 dpv (Figure 5).

3.5. Signatures of antibody immune response in Sputnik V vaccinees.

An ascending hierarchical classification of the vaccinees was performed to compare their pattern of reactivity and neutralization. The classification made on individuals revealed two main patterns of response. One main cluster grouped most of the vaccinees (low responders, n=14), and the remaining (high responders, n=4) with a high reactivity and PRNT50 titer to Gamma, Delta or Omicron variants (Figure 6).

4. Discussion

As vaccines were based on ancestral antigens and because of the evolution of SARS-CoV-2, it was critical to determine whether the binding, neutralization and life span of antibodies induced after vaccination were affected by the new variants. Sputnik-V played a pivotal role in the vaccination initiatives throughout Venezuela and Latin America, so it was essential to assess the antibody response elicited by it in the face of the sequential emergence of variants of concern, considering moreover the lesser amount of information available for this vaccine compared to mRNA ones. In order to perform comparisons of reactivity among antigens and to assess how these changed over time, we used an in-house ELISA. On the other hand, we used the PRNT to estimate the neutralization titers of these sera. We also evaluated the possible correlation between these variables.

No differences in reactivity to Beta and Gamma compared to ancestral antigens were reported. Other authors have described a reduced reactivity and neutralization of Sputnik-V-induced antibodies against these variants [13,22]. Interestingly, the S antigen from the Alpha and Delta variants were recognized with a higher reactivity compared to the ancestral S protein at 42 dpv, and for Delta also at 180 dpv, consistent with the presence of high affinity and neutralizing antibodies generated by this vaccine against SARS-CoV-2 variant antigens [23]. With the exception of Omicron, where a significant decrease in reactivity to S and RBD antigens (RBD, data not shown) was also observed at both time points, seroconversion to all other antigens ranged from 88 to 99% and correlated with the efficacy of this vaccine as previously reported [24,25]. Although reactivity tended to decline over time, the vaccine maintained an effective response at six months. The durable antibody response induced by this vaccine has also been described by others [18,26,27]. These observations support the notion that the antibody response generated by immunization with Sputnik-V persists for at least 6 months after administration of the first vaccine component.

Higher neutralization titers were reported against the ancestral variant compared to the other variants, and Omicron was the variant for which the highest reduction in neutralization titer was observed. Omicron and its sub-lineages have exhibited greater antigenic variation than previous variants, significantly altering the patterns of recognition and neutralization of vaccine-induced antibodies [28]. This change in viral behavior threatens potential outbreaks from other emerging SARS-CoV-2 variants or new coronaviruses. Although the neutralization titer against Delta and Omicron is reduced, we observed that some sera were able to recognize and neutralize these variants. A possible explanation is that after Sputnik-V injection, naïve B cells undergo germinal center (GC) selection by somatic hypermutation (SHM) of variable heavy (VH) and variable light (VL) chain genes, thus stimulating B cell receptor (BCR) repertoire diversification, increasing binding affinity and neutralizing activity, which are considered indicators of antibody affinity maturation [29]. Indeed, mRNA vaccines induce a robust B cell, germinal center and follicular helper T cell (Tfh) response, which increases after the second dose [30]. Furthermore, Tfhs have been detected in blood and lymph nodes for at least 200 days after vaccination [31]. In addition, memory B cell clones that interact with the RBD of the alpha, beta and delta variants have higher levels of SHM than those that bind only to the RBD of the ancestral variant after mRNA vaccination [32]. In addition, the persistence of GC activity after immunization, and therefore SHM, appears to promote a protective antibody response against variants of this virus [33] by increasing the likelihood of cross-neutralization of other variants, a response already reported during the convalescent phase [34,35,36].

This persistent GC activity could be explained by antigenic persistence of adenovirus vectors or mRNA after immunization in different tissues [37,38]. In addition, persistence of S and S1 antigens has been reported in some mRNA vaccinated individuals for up to 60 days after immunization [39,40]. The persistence, distribution in primary and secondary lymphoid organs and prolonged activation of GC due to the capture and prolonged presentation of S, S1, S2, RBD and peptides could support that GC activity persists for periods of time yet to be defined after vaccination, thus promoting the generation of antibodies with greater diversity and affinity for epitopes of the S protein. Indeed, there is evidence that structurally undamaged antigens are preserved in B cell follicles due to low protease activity in dendritic cells located in this compartment, which may explain the long-term availability of antigens after immunization, thus facilitating antigen presentation for a prolonged period [41,42]. Furthermore, the presence of B cells in germinal centers that recognize protein S for at least 30 weeks after immunization with mRNA has been described [30,33].

To our knowledge, GC activity in tissues from Sputnik-V vaccinated individuals remains to be elucidated. However, the same mechanisms observed in mRNA vaccines may be operating to counteract SARS-CoV-2n variants. GC activity has been described in other vaccines using both adenoviruses used in Sputnik-V [43,44,45]. Furthermore, an increase in memory, antibody-secreting, FcRL4+ and Tfh cells has been reported after immunization with BNT162b2, AZD1222 and Sputnik-V [46]. In addition, an increase in cytokines that mediate B- and Tfh cell proliferation, differentiation, localization, interaction, and antibody class switching [47] such as IL-6, IL-10, IFN-γ, IL-2, IL-17A, and sCD40L, has been identified [48], suggesting GC activity after administration of this adenoviral vaccine.

A subset of sera exhibited higher neutralization titers to the Gamma VOC compared to the ancestral variant. This observation suggests that these individuals might have been exposed to this variant during the study period, as it was disseminating in Venezuela during March-August 2021 [15,49]. None of these volunteers reported an infectious event during the clinical trial. These sera showed no reactivity to the N protein [18], which would have been expected in the event of infection, and a modest increase in reactivity to S but not to RBD Gamma VOC. We suggest that they might had an asymptomatic or subclinical infection, related to the protective effect of the vaccine received, manifested only at the presence of neutralizing antibodies against the infecting variant, and a modest increase in reactivity to the Gamma S protein variant. Other authors have found neutralizing antibodies in asymptomatic or mildly affected COVID-19 individuals with no detectable IgG antibodies [50]. Similar results (higher titer against the Gamma variant without apparent previous infection) were obtained by another group who studied the neutralization of Sputnik V-induced antibodies in a cohort of volunteers in Argentina [51]. On the other hand, the fact the sera from our study exhibited just a modest increase in reactivity towards the S protein, and not the RBD region, might be due to the fact that other regions of the S protein, like the N-terminal region, might be contributing in inducing neutralizing antibodies against the virus, as suggested by our previous study [18].

Taken together, our results suggest that the Sputnik-V vaccine induces antibodies that are similarly able to interact with S and neutralizing SARS-CoV-2 variants, with the Omicron VOC being the most resistant to antibody immunity, as previously reported [14,25]. Some volunteers may have been exposed to the Gamma variant during the clinical trial, without showing symptoms. This reinforces the significant beneficial effect of the COVID-19 vaccines in protecting, if not completely against infection, against symptomatic infection due to variant emergence, and then against severe disease manifestation, death and long-term sequelae [52,53]. The antigenic distance between different variants enhances our understanding of how immunized populations may be susceptible to the emergence of new variants, and has been and should be continued to be considered in the design of the next generation of vaccines [8,54]. Therefore, continuous monitoring of SARS-CoV-2 evolutionary trajectories should be a priority in the design of new vaccines that include immunogens that promote GC responses to provide broader protection against infection and disease caused by emerging variants. We provide further evidence on the magnitude, breadth and persistence of antibodies induced by this adenoviral vaccine in the context of SARS-CoV-2 evolution, providing insights for the design of future vaccines against new potential coronaviruses.