B.1.526 SARS-CoV-2 Variants Identified in New York City are Neutralized by Vaccine-Elicited and Therapeutic Monoclonal Antibodies

ABSTRACT DNA sequence analysis recently identified the novel SARS-CoV-2 variant B.1.526 that is spreading at an alarming rate in the New York City area. Two versions of the variant were identified, both with the prevalent D614G mutation in the spike protein, together with four novel point mutations and with an E484K or S477N mutation in the receptor-binding domain, raising concerns of possible resistance to vaccine-elicited and therapeutic antibodies. We report that convalescent-phase sera and vaccine-elicited antibodies retain full neutralizing titer against the S477N B.1.526 variant and neutralize the E484K version with a modest 3.5-fold decrease in titer compared to D614G. The E484K version was neutralized with a 12-fold decrease in titer by the REGN10933 monoclonal antibody, but the combination cocktail with REGN10987 was fully active. The findings suggest that current vaccines and Regeneron therapeutic monoclonal antibodies will remain protective against the B.1.526 variants. The findings further support the value of widespread vaccination.

IMPORTANCE A novel SARS-CoV-2 variant termed B.1.526 was recently identified in New York City and has been found to be spreading at an alarming rate. The variant has mutations in its spike protein that might allow it to escape neutralization by vaccine-elicited antibodies and might cause monoclonal antibody therapy for COVID-19 to be less successful. We report here that these fears are not substantiated; convalescentphase sera and vaccine-elicited antibodies neutralized the B.1.526 variant. One of the Regeneron therapeutic monoclonal antibodies was less effective against the B.1.526 (E484K) variant but the two-antibody combination cocktail was fully active. The findings should assuage concerns that current vaccines will be ineffective against the B.1.526 (E484K) variant and suggest the importance of continued widespread vaccination. S evere acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is a highly transmissible and pathogenic coronavirus that became an ongoing pandemic late in 2019. Infection rates have begun to fall, at least in part, due to large-scale vaccination efforts. In addition, treatment of infected patients with monoclonal antibodies against the spike protein have been found to reduce hospitalization and mortality. The recent emergence of SARS-CoV-2 variants raises concerns with regard to vaccine efficacy and the effectiveness of monoclonal antibody therapy. The vast majority of sequenced SARS-CoV-2 isolates contain a D614G mutation in the spike protein (1) that increases viral infectivity and transmissibility (2)(3)(4) and, subsequently, variants with multiple mutations in the spike protein and enhanced transmissibility have emerged in the United Kingdom (5, 6), South Africa (7), Brazil (8), and the United States (9), raising concerns of diminished neutralization by immune sera-elicited antibodies and escape from therapeutic monoclonal antibodies.
Recent reports have identified a novel variant in New York City termed B.1.526 that was rapidly spreading (10)(11)(12). The variant was identified in November 2020; by January 2021, the variant accounted for 5% of genomes sequenced from individuals in New York and by mid-February was detected with a frequency of 12.3% (10)(11)(12). The variant contains several mutations in the spike protein, some of which have not been found in previous variants. Two versions of B.1.526 were identified, both having the D614G and A701V mutations and, in addition, the mutations L5F, T95I, and D253G, which are not present in previously reported variants. One version of B.1.526 also contains the E484K mutation, which is present in the B.1.351 and B.1.1.248 variant spike proteins and allows for partial escape from immune serum neutralization (13)(14)(15)(16); the other lacks the E484K mutation but has a nearby S477N mutation, which lies within the receptor binding domain (RBD) and thus may influence affinity for the entry receptor ACE2. The D253G mutation is located in the amino-terminal supersite that serves as a binding site for neutralizing antibodies, while A701V is located adjacent to the furin processing site. The combination of mutations raises concerns that the B.1.526 variant might evade vaccine-elicited and therapeutic antibodies.
Previous studies have shown that the E484K mutation in the B.1.351 spike protein leads to a degree of resistance to neutralization by both infection-and vaccine-elicited antibodies, as well as to the REGN10933 therapeutic monoclonal antibody (17)(18)(19). Moreover, the B.1.351 variant spike protein has been found to reduce the level of protection provided by vaccination in populations in which the variant has become prevalent (17,20).
In this study, we determined the susceptibility of the B.1.526 variants to neutralization by convalescent-phase sera and sera from individuals vaccinated with the Pfizer BNT162b2 and Moderna mRNA-1273 vaccines, and by the Regeneron therapeutic monoclonal antibodies (18,21). We found that the B.1.526 variant (S477N) was fully susceptible to neutralization, while the B.1.526 variant with the E484K mutation neutralized with a modest (3.5-fold) reduction in titer by convalescent and vaccine-elicited antibodies. The B.1.526 spike proteins were readily neutralized by the Regeneron antibody cocktail.
The B.1.526 variant spike proteins contain the D614G mutation, a shared set of novel mutations (L5F, T95I, D253G, and A701V), and either E484K or S477N, both of which lie within the RBD ( Fig. 1A and B). To study the B.1.526 spike proteins, we constructed spike protein expression vectors for both B.1.526 versions and used these to produce lentiviral pseudotype reporter viruses as previously described. Immunoblot analysis showed that both B.1.526 spike proteins were expressed and processed in transfected human 293T cells and that both were incorporated into virions at a level comparable to that of the wild-type (D614G) spike protein (Fig. 1C). The infectivity of B.1.526 variant pseudotypes on ACE2.293T was similar to that of wild type (Fig. 1D).
To test the ability of convalescent-phase sera to neutralize the B.1.526 viruses, we determined the neutralizing antibody titers of sera from individuals who had been infected prior to April 2020 on viruses with D614G, B.1.526 and E484K spike proteins ( Fig. 2A). The results showed that neutralizing titers against the S477N B.1.526 variant were similar to that of D614G, while the neutralizing titers against the E484K B.1.526 variant decreased by 3.8-fold, a modest decrease that was attributed to the E484K mutation.
To determine the ability of vaccine-elicited antibodies to neutralize the B.1.526 viruses, we determined neutralizing titers of serum specimens from individuals vaccinated with Pfizer BNT162b2 or Moderna mRNA-1273 vaccines. The results showed that BNT162b2 vaccine serum-elicited antibodies neutralized the D614G and B.1.1.7 viruses with similarly high titers, while titer for neutralization of B.1.351 was decreased by 3.4fold (Fig. 2B, left). Analysis of the B.1.526 titers showed the S477N version was neutralized with a titer similar to D614G; neutralization titers of the E484K version were decreased by 3.6-fold, a titer similar to that of B.1.351. Neutralization by sera elicited by the Moderna vaccine showed a very similar pattern (Fig. 2B, right).
Analysis of the Regeneron monoclonal antibodies showed that REGN10987 neutralized both B.1.526 variants with no loss of titer (Fig. 2C, Table 1). REGN10933 neutralized virus with the S477N B.1.526 spike protein with a high titer but was 12-fold less active  (Fig. 2C, Table 1). The REGN-COV2 cocktail potently neutralized the B.1.526 spike variants despite the partial loss of neutralizing activity against the E484K version of B.1.526. While the two monoclonal antibodies do not have overlapping binding sites, it appeared they may have some synergistic effect when combined. To test this possibility, we held one antibody constant at its 50% inhibitory concentration (IC 50 ) and then titrated in the other antibody. The results showed that when REGN10987 was titrated in, it neutralized the virus efficiently but that, conversely, when REGN10933 was titrated in, the virus could not be completely neutralized (Fig. 2D). This result suggests that REGN10933 binding may act to make the REGN10987 epitope more accessible.
The recently identified B.1.526 variant SARS-CoV-2 appears to be increasing in prevalence in New York City, raising concerns about reinfection and immunoevasion (10)(11)(12). We report here that both S477N and E484K versions of B.1.526 were neutralized well by convalescent and vaccine-elicited antibodies. The E484K version of B.1.526 did show a significant, nearly 4-fold, decrease in neutralization by vaccine-elicited antibodies, but this represents a modest decrease in titer that is not expected to result in a significant decrease in the protection provided by vaccination and is not expected to result in an increased susceptibility to reinfection. The S477N version was neutralized with no decrease in titer. As we reported, neutralizing antibody titers determined by pseudotyped virus assay closely reflect those measured by live SARS-CoV-2 assay (22).
Our results showed that REGN10987, which binds to the side of the RBD (18,21), maintains potent neutralizing activity against both versions of B.1.526 but that REGN10933, which binds to the top face of the RBD, that interacts with ACE2, loses 12-fold potency against the E484K version. The decrease in neutralizing titer was caused by the E484K mutation and is similar to the previously reported loss of titer against B.1.351, which also bears the mutation (17)(18)(19)(20)(23)(24)(25). Despite the partial loss of activity by REGN10933 against B.1.526, the neutralizing activity of the combined antibody cocktail remained high.
Our findings should assuage concerns that the B.1.526 variant will evade protection provided by vaccine-elicited antibodies and suggest that Regeneron therapeutic antibody therapy will retain its effectiveness against the variant. Nevertheless, B.1.526 appears to be spreading at an alarming rate, demonstrating the value of widespread vaccination efforts.
Human sera and monoclonal antibodies. Convalescent sera and sera from BNT162b2-or Moderna-vaccinated individuals were collected on day 7 following the second immunization at the NYU vaccine center with written consent under IRB approval (IRB 20-00595 and IRB 18-02037). Donor age and gender were not reported. Regeneron monoclonal antibodies (REGN10933 and REGN10987) were prepared as previously described (26).
Quantification and statistical analysis. All experiments were performed in technical duplicates or triplicates and data were analyzed using GraphPad Prism 8. Statistical significance was determined by the two-tailed unpaired t test. Significance was based on two-sided testing and attributed at P , 0.05. Confidence intervals are shown as the mean 6 standard deviation (SD) or standard error of the mean (SEM). (*, P # 0.05; **, P # 0.01; ***, P # 0.001; ****, P # 0.0001). The PDB file of D614G SARS-CoV-2 spike protein (7BNM) was downloaded from the Protein Data Bank. A 3D view of the protein was obtained using PyMOL.

ACKNOWLEDGMENTS
The work was funded by grants from the NIH to N. We declare no competing interests.