Increased Resistance of SARS-CoV-2 Variants B.1.351 and B.1.1.7 to Antibody Neutralization

The Covid-19 pandemic has ravaged the globe, and its causative agent, SARS-CoV-2, continues to rage. Prospects of ending this pandemic rest on the development of effective interventions. Two monoclonal antibody (mAb) therapeutics have received emergency use authorization1,2, and more are in the pipeline3–6. Furthermore, multiple vaccine constructs have shown promise7, including two with ~95% protective efficacy against Covid-198,9. However, these interventions were directed toward the initial SARS-CoV-2 that emerged in 2019. Considerable viral evolution has occurred since, including variants with a D614G mutation10 that have become dominant. Viruses with this mutation alone do not appear to be antigenically distinct, however11. Recent emergence of new SARS-CoV-2 variants B.1.1.7 in the UK12 and B.1.351 in South Africa13 is of concern because of their purported ease of transmission and extensive mutations in the spike protein. We now report that B.1.1.7 is refractory to neutralization by most mAbs to the N-terminal domain (NTD) of spike and relatively resistant to a number of mAbs to the receptor-binding domain (RBD). It is modestly more resistant to convalescent plasma (~3 fold) and vaccinee sera (~2 fold). Findings on B.1.351 are more worrisome in that this variant is not only refractory to neutralization by most NTD mAbs but also by multiple potent mAbs to the receptor-binding motif on RBD, largely due to an E484K mutation. Moreover, B.1.351 is markedly more resistant to neutralization by convalescent plasma (~11–33 fold) and vaccinee sera (~6.5–8.6 fold). B.1.351 and emergent variants14,15 with similar spike mutations present new challenges for mAb therapy and threaten the protective efficacy of current vaccines.

SARS-CoV-2 B.1.1.7, also known as 501Y.V1 in the GR clade (Fig. 1a), emerged in September 2020 in South East England and rapidly became the dominant variant in the UK, possibly due to its enhanced transmissibility 12 . This strain has now spread to over 50 countries. B.1.1.7 contains 8 spike mutations in addition to D614G, including two deletions (69-70del & 144del) in NTD, one mutation (N501Y) in RBD, and one mutation (P681H) near the furin cleavage site (Fig. 1b). SARS-CoV-2 B.1.351, also known as 501Y.V2 in the GH clade (Fig. 1a), emerged in late 2020 in Eastern Cape, South Africa (SA) 13 . This variant has since become dominant locally, raising the specter that it too has enhanced transmissibility. B.1.351 contains 9 spike mutations in addition to D614G, including a cluster of mutations (e.g., 242-244del & R246I) in NTD, three mutations (K417N, E484K, & N501Y) in RBD, and one mutation (A701V) near the furin cleavage site (Fig. 1b). There is a growing concern that these new variants could impair the efficacy of current mAb therapies or vaccines, because many of the mutations reside in the antigenic supersite in NTD 16,17 or in the ACE2-binding site (also known as the receptorbinding motif-RBM) that is a major target of potent virus-neutralizing antibodies. We therefore addressed this concern by creating VSV-based SARS-CoV-2 pseudoviruses that contain each of the individual mutations as well as one with all 8 mutations of the B.1.1.7 variant (UK∆8) and another with all 9 mutations of the B.1.351 variant (SA∆9). A total of 18 mutant pseudoviruses were made as previously described 18,19 , and each was found to have a robust titer (Extended Data Fig. 1) adequate to measure its susceptibility to neutralization by 30 mAbs, 20 convalescent plasma, and 22 vaccinee sera.

Monoclonal antibodies
We first assayed the neutralizing activity of 12 RBD mAbs against UK∆8, SA∆9, and WT (D614G) pseudoviruses in Vero E6 cells as previously described 18,19 . Three mAbs target the "inner side", four target RBM, and five target the "outer side". The footprints of these mAbs on RBD are shown in Fig. 2a, and their neutralization profiles are shown in Fig. 2b. and SA∆9 by these 12 mAbs are summarized in Fig. 2c as fold changes in IC50 neutralization titers relative to the WT. To understand the specific spike mutations responsible for the observed changes, we also tested the same panel of mAbs against pseudoviruses containing only a single mutation found in B. We also assessed the neutralizing activity of six NTD mAbs against UK∆8, SA∆9, and WT pseudoviruses. Both UK∆8 and SA∆9 are profoundly resistant to neutralization by our antibodies 5-24, 4-8, 2-17, and 4-19 18 , as well as by 4A8 25 (Fig. 2d). Note that 5-24, 4A8, and 4-8 are known to target the antigenic supersite in NTD 16 (Insert in Fig. 2d). The activity of 5-7 18 remains intact, however. To understand the specific mutations responsible for the observed changes, we then tested these mAbs against pseudoviruses containing only a single mutation found in B.1.1.7 or B.1.351 (Extended Data Fig. 2). The results are summarized in Fig. 2c as fold change relative to the WT. It is evident that the resistance of UK∆8 to most NTD mAbs is largely conferred by 144del, whereas the resistance of SA∆9 is largely conferred by 242-244del and/or R246I. Amino-acid residues 144, 242-244, and 246 all fall within the NTD supersite 16,17 (Insert in Fig. 2d; details in Extended Data Fig. 3b). The obvious exception is 5-7, whose neutralizing activity is actually enhanced. Needless to say, a detailed structural understanding of how 5-7 binds NTD will be important.
We next tested the neutralizing activity of 12 additional RBD mAbs, including ones from our own collection (1-20, 4-

Convalescent plasma
We obtained convalescent plasma from 20 patients more than one month after documented SARS-CoV-2 infection in the Spring of 2020. Ten had severe disease and 10 had non-severe disease, as previously defined 19 . Their ages ranged from 34 to 79, with a mean of 54. Six were female, and 14 were male.
Each plasma sample was then assayed for neutralization against UK∆8, SA∆9, and WT pseudoviruses. Fig. 3a shows that most plasma samples lost >2-fold neutralizing activity against the new variants relative to the WT. The loss in potency is more frequent against SA∆9 (16 of 20) than against UK∆8 (11 of 20). Only plasma from P7, P10, P18, and P20 retain neutralizing activities identical or similar to those against the WT. These results are summarized as fold change in plasma neutralization IC50 titers in Fig. 3b.
Furthermore, the magnitude of the drop in plasma neutralization is better seen in Fig. 3c, with the overall mean loss of activity being modest against UK∆8 (2.7 to 3.8 fold), but more substantial against SA∆9 (11.0 to 33.1 fold).
Every plasma sample was also tested against each single-mutation pseudovirus, and those findings are shown in Extended Data Fig. 6 and summarized in Fig. 3b. Unlike the data for mAbs (Fig. 2c), no single mutation could predictably account for the loss of plasma neutralizing activity against UK∆8, indicating that the mutations in this variant do not perturb an immunodominant epitope on the spike that is shared by many infected persons. S982A seems to have a discernible negative impact on the plasma neutralizing activity of 9 samples (Fig. 3b), perhaps due to its interaction with the bottom of RBD (Extended Data Fig. 3c). On the other hand, the loss of plasma neutralizing activity against SA∆9 could be largely attributed to E484K (Fig. 3b), suggesting that this RBM mutation to be situated in an immunodominant epitope for most infected persons. It is also interesting to note that cases such as P7 and P10 have neutralizing antibodies that are essentially unperturbed by the multitude of spike mutations found in these two new variants (Fig. 3b). A detailed analysis of their antibody repertoire against the viral spike could be informative.

Vaccinee Sera
Sera were obtained from 12 participants of a Phase 1 clinical trial of Moderna SARS-Co-2 mRNA-1273 Vaccine 8 conducted at the NIH. These volunteers received two immunizations with the vaccine (100 µg) on days 0 and 28, and blood was collected on day 43. Additional vaccinee sera were obtained at Columbia University Irving Medical Center from 10 health care workers who received the Pfizer BNT162b2 Covid-19 Vaccine 9 at the clinical dose on days 0 and 21. Blood was collected on day 28 or later.
Each vaccinee serum sample was assayed for neutralization against UK∆8, SA∆9, and WT pseudoviruses. Fig. 4a shows only a minority of sera to have lost >2-fold neutralizing activity against UK∆8, whereas every sample lost activity against SA∆9, ranging from slight to substantial. These results are quantified and tabulated as fold change in neutralization IC50 titers in Fig. 4b, and the extent of the decline in neutralization activity is more evident in Fig. 4c. Overall, the mean loss of neutralizing activity against UK∆8 appears to be small (1.8 fold, Moderna; 2.0 fold, Pfizer), but quite significant against SA∆9 (8.6 fold, Moderna; 6.5 fold, Pfizer).
Every vaccinee serum was also tested against each single-mutation pseudovirus, and the results are presented in Extended Data Fig. 7 and summarized in Fig. 4b. As was the case for convalescent plasma (Fig. 3b), no single mutation could predictably account for the small loss of serum neutralizing activity against UK∆8. Again, S982A seems to have a minor negative impact on the plasma neutralizing activity of every serum sample (Fig.   4b), possibly due to distal effects on the RBD (Extended Data Fig. 3c). The loss of neutralizing activity against SA∆9 in vaccinee sera could be principally attributed to E484K (Fig. 4b), indicating that this RBM mutation to be situated in an immunodominant epitope recognized by all vaccinees studied. Our findings do not reveal any significant differences between the two different vaccines.

Discussion
Both SARS-CoV-2 variants B.1.1.7 and B.1.351 are raising concerns not only because of their increase transmissibility but also because of their extensive mutations in spike that could lead to antigenic changes detrimental to mAb therapies and vaccine protection. It is of equal concern that another variant, B.1.1.28 or 501Y.V3, is increasing rapidly in Brazil and spreading far beyond 14,15 Fig. 3b). More importantly, SA∆9 is resistant to a major group of potent mAbs that target the RBM, including two authorized for emergency use (Fig. 2c). LY-CoV555 is inactive against SA∆9, and the ~60-fold loss in potency of Convalescent plasma from patients infected with SARS-CoV-2 from early in the pandemic show slightly decreased neutralizing activity against UK∆8, but the diminution against SA∆9 is remarkable (Figs. 3b &3c). This relative resistance is largely due to E484K, a mutation shared by B.1.351 and B.1.1.28 [13][14][15] . Again, in areas where such viruses are common, one would have heightened concerns about re-infection, which has already been well documented even in the absence of antigenic changes 27,28 .
As for the ramifications of our findings for the protective efficacy of current SARS-CoV-2 vaccines, the ~2-fold loss of neutralizing activity of vaccinee sera against UK∆8 is unlikely to have an adverse impact due to the large "cushion" of residual neutralizing antibody titer (Fig. 4c). On the other hand, the loss of ~6.5-8.6 fold in activity against SA∆9 is more worrisome, although the clinical implication for vaccine efficacy remains to be determined.
The results from ongoing trials in South Africa using these or similar vaccine constructs should be informative.