Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

mRNA vaccine-elicited antibodies to SARS-CoV-2 and circulating variants

Abstract

Here we report on the antibody and memory B cell responses of a cohort of 20 volunteers who received the Moderna (mRNA-1273) or Pfizer–BioNTech (BNT162b2) vaccine against SARS-CoV-21,2,3,4. Eight weeks after the second injection of vaccine, volunteers showed high levels of IgM and IgG anti-SARS-CoV-2 spike protein (S) and receptor-binding-domain (RBD) binding titre. Moreover, the plasma neutralizing activity and relative numbers of RBD-specific memory B cells of vaccinated volunteers were equivalent to those of individuals who had recovered from natural infection5,6. However, activity against SARS-CoV-2 variants that encode E484K-, N501Y- or K417N/E484K/N501-mutant S was reduced by a small—but significant—margin. The monoclonal antibodies elicited by the vaccines potently neutralize SARS-CoV-2, and target a number of different RBD epitopes in common with monoclonal antibodies isolated from infected donors5,6,7,8. However, neutralization by 14 of the 17 most-potent monoclonal antibodies that we tested was reduced or abolished by the K417N, E484K or N501Y mutation. Notably, these mutations were selected when we cultured recombinant vesicular stomatitis virus expressing SARS-CoV-2 S in the presence of the monoclonal antibodies elicited by the vaccines. Together, these results suggest that the monoclonal antibodies in clinical use should be tested against newly arising variants, and that mRNA vaccines may need to be updated periodically to avoid a potential loss of clinical efficacy.

Main

Between 19 October 2020 and 15 January 2021, we recruited 20 volunteers who received 2 doses of the Moderna (n = 14) or Pfizer–BioNTech mRNA (n = 6) vaccines for blood donation. The volunteers ranged from 29 to 69 years old (median 43); 12 (60%) were men and 8 (40%) women. Sixteen participants identified as Caucasian, two identified as Hispanic, one identified as African American and one identified as Asian. The time from the second vaccination to sample collection varied between 3 and 14 weeks (average of 8 weeks). None of the volunteers had a history of previous infection with SARS-CoV-2, and none experienced serious adverse events after vaccination (Supplementary Table 1).

Plasma activity against SARS-CoV-2 after vaccination

Plasma IgM, IgG and IgA responses to SARS-CoV-2 S and RBD were measured by enzyme-linked immunosorbent assay (ELISA)5,6. All of the individuals we tested showed a reactivity to S and RBD that was significantly higher compared to historic controls (from before COVID-19) (Extended Data Fig. 1a–f). As might be expected, levels of anti-S and anti-RBD IgG were higher than those of IgM or IgA. Moreover, there was a strong positive correlation between the anti-RBD and anti-S response in all three of the immunoglobulin isotypes we measured (Extended Data Fig. 1g–i). Consistent with previous reports2,9,10, levels of IgG and IgM were significantly higher in the vaccinated group than in a cohort of patients who had recovered from COVID-19 (assayed at 1.3 and 6.2 months after infection), whereas IgA levels were similar between the two groups (Extended Data Fig. 1j–l).

We determined plasma neutralizing activity using human immunodeficiency virus-1 (HIV-1) pseudotyped with SARS-CoV-2 S5,6,11. Consistent with previous reports2,9,10, there was a broad range of plasma neutralizing activity at 3–14 weeks after the second dose of vaccine; this range was similar to that elicited by natural infection in the cohort of patients who had recovered from COVID-19 after 1.3 months, and greater than the activity at 6.2 months after infection (Fig. 1a, Supplementary Table 1). There was no significant difference in neutralizing activity between the Moderna and the Pfizer–BioNTech vaccines (Fig. 1b). Whereas antibody titres in the cohort of patients who had recovered from COVID-19 tended to correlate with severity and length of time of infection, additional sampling would be required to understand the correlates of the magnitude of the vaccine response. As expected, plasma neutralizing activity was directly correlated with anti-S and anti-RBD binding titres in ELISAs5,6 (Fig. 1c, d, Extended Data Fig. 2a–d). Finally, RBD and S binding, and neutralizing activities, were directly correlated with the time between the first dose of vaccine and blood donation: levels of all three measurements were significantly reduced with time12 (Fig. 1e–g, Extended Data Fig. 2e–h). However, this, and other small studies2,12, cannot accurately predict the half-life of the neutralizing response. Larger numbers of individuals in diverse cohorts will be needed to determine the precise half-life of the neutralizing response elicited by the vaccine.

Fig. 1: Plasma neutralizing activity.
figure1

a, SARS-CoV-2 pseudovirus neutralization assay. Half-maximal neutralizing titre (NT50) values for plasma obtained from patients who had recovered from COVID-19, measured at 1.3 months6 and 6.2 months5 after infection, as well as plasma from vaccinated individuals. NT50 values lower than 10 were plotted as 10. Mean of two independent experiments. Red bars and indicated values represent geometric mean NT50 values. Statistical significance was determined using the two-tailed Mann–Whitney U test. Pre-COVID-19 historical-control plasma was analysed as a negative control and showed no detectable neutralization (NT50 < 10, not included in panel). NS, not significant. b, NT50 values for recipients of the Moderna mRNA-1273 (black) and Pfizer–BioNTech BNT162b2 (red) vaccines. Red bars and indicated values represent geometric mean NT50 values. Statistical significance was determined using the two-tailed Mann–Whitney U test. c, Anti-RBD IgG area under the curve (AUC) values (y axis) plotted against NT50 (x axis). d, Anti-S IgG AUC (y axis) plotted against NT50 (x axis). e, Anti-RBD IgG AUC (y axis) plotted against time between first dose and blood draw (x axis). f, Anti-S IgG AUC (y axis) plotted against time between first dose and blood draw (x axis). g, NT50 (y axis) plotted against time between first dose and blood draw (x axis). The r and P values for correlations in cg were determined by two-tailed Spearman correlation. Individuals who received the Moderna vaccine are shown in black and individuals who received the Pfizer–BioNTech vaccine are in red. h, Examples of neutralization assays, comparing the sensitivity of pseudotyped viruses with wild-type (WT) and RBD-mutant (mut) SARS-CoV-2 S to plasma from vaccinated individuals. MOD1 and PFZ10 indicate two representative individuals who received the Moderna and Pfizer–BioNTech vaccine, respectively (Supplementary Table 1). RLU, relative luminescence units. i, NT50 values for neutralization by plasma from vaccinated individuals (n = 15) of pseudotyped viruses with wild-type and the indicated RBD-mutant SARS-CoV-2 S. Individuals vaccinated with the Pfizer–BioNTech vaccine are in red. j, NT50 values for neutralization by plasma from individuals who had recovered from COVID-19 (n = 45) of pseudotyped viruses with wild-type and K417N/E484K/N501Y SARS-CoV-2 S. Statistical significance in i, j was determined using a one-tailed t-test. All experiments were performed a minimum of two times. Pseudotyped viruses containing the E484K mutation and corresponding wild-type controls contain the R683G mutation (Methods).

To determine whether plasma from vaccinated individuals can neutralize circulating SARS-CoV-2 variants of concern and mutants that arise in vitro under antibody pressure13,14, we tested plasma from vaccinated individuals against a panel of ten mutant pseudotype viruses, including the recently reported N501Y (B1.1.7 variant), K417N, E484K and K417N/E484K/N501Y (501Y.V2 variant) mutations of the RBD15,16,17,18,19,20. Plasma from vaccinated individuals was significantly less effective in neutralizing the HIV-1 virus pseudotyped with particular mutants of the SARS-CoV-2 S (Fig. 1h, i, Extended Data Fig. 2j). In the plasma we tested, there was a one- to threefold decrease in neutralizing activity against E484K, N501Y and K417N/E484K/N501Y (P = 0.0033, P = 0.0002 and P < 0.0001, respectively) (Fig. 1h, i). Similarly, plasma obtained from the cohort of patients who had recovered from COVID-19 at 1.3 and 6.2 months after infection was 0.5- to 29-fold and 0.5- to 20.2-fold less effective in neutralizing the K417N/E484K/N501Y mutant (P = 0.001 and P < 0.0001, respectively) (Fig. 1j, Supplementary Table 2). We conclude that the plasma neutralizing activity elicited by mRNA vaccination or natural infection is variably, but significantly, less effective against pseudoviruses that carry RBD mutations that are found in emerging variants of SARS-CoV-2.

Vaccine-elicited monoclonal antibodies

Although circulating antibodies derived from plasma cells wane over time, long-lived immune memory can persist in expanded clones of memory B cells5,21. We used flow cytometry to enumerate the circulating SARS-CoV-2 RBD-specific memory B cells elicited by mRNA immunization5,6 (Fig. 2a, Extended Data Fig. 3a, b). We focused on the RBD because it is the target of the majority of the more-potent SARS-CoV-2 neutralizing antibodies discovered to date22,23,24,25,26,27. Notably, the percentage of RBD-binding memory B cells in vaccinated individuals was significantly greater than in naturally infected individuals assayed after 1.3 months, but similar to the same individuals assayed after 6.2 months (Fig. 2b). The percentage of RBD-binding memory B cells in vaccinated individuals was not correlated with the time after vaccination (Extended Data Fig. 3c). Thus, mRNA vaccination elicits a robust SARS-CoV-2 RBD-specific B cell memory response that resembles that following natural infection.

Fig. 2: Memory B cell antibodies.
figure2

a, Representative flow cytometry plots showing dual AlexaFluor 647–RBD- and PE–RBD-binding B cells for four vaccinated individuals. b, As in a, dot plot summarizes the percentage of RBD-binding B cells in 19 vaccinated individuals, in comparison to a cohort of naturally infected individuals assayed at 1.3 and 6.2 months after infection5,6. Individuals who received the Moderna vaccine are shown in black and individuals who received the Pfizer–BioNTech vaccine are in red. Red horizontal bars indicate mean values. Statistical significance was determined using two-tailed Mann–Whitney U tests. c, Pie charts show the distribution of antibody sequences from the four individuals in a. The number in the inner circle indicates the number of sequences analysed. Pie-slice size is proportional to the number of clonally related sequences. The black outline indicates the frequency of clonally expanded sequences. d, As in c, graph shows relative clonality among 14 vaccinated individuals. Individuals who received the Moderna vaccine are shown in black and individuals who received the Pfizer–BioNTech vaccine are in red. Red horizontal bars indicate mean values. Statistical significance was determined using two-tailed Mann–Whitney U tests. e, Graph shows relative abundance of human IGVH genes in data from Sequence Read Archive accession SRP010970 (database) (orange), and vaccinated individuals (blue). A two-sided binomial test was used to compare the frequency distributions. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. f, Clonal relationships between sequences from 14 vaccinated individuals (Moderna in black, Pfizer–BioNTech in red) (Supplementary Table 3) and naturally infected individuals5,6 (green). Interconnecting lines indicate the relationship between antibodies that share V and J gene-segment sequences at both IGH and IGL. Purple, green, and grey lines connect related clones, clones and singles, and singles to each other, respectively. g, Number of somatic nucleotide mutations in the IGVH (top) and IGVL (bottom) in antibodies from vaccinated individuals (Supplementary Table 3), compared to those from naturally infected individuals obtained at 1.3 or 6.2 months after infection5,6. Statistical significance was determined using two-tailed Mann–Whitney U tests. Red horizontal bars indicate mean values. h, As in g, but for CDR3 length in amino acids (aa).

To examine the nature of the antibodies produced by memory B cells in response to vaccination, we obtained 1,409 paired antibody heavy and light chains from RBD-binding single B cells from 14 individuals (10 who had received the Moderna vaccine, and 4 who had received the Pfizer–BioNTech vaccine) (Supplementary Table 3). Expanded clones of cells comprised 4–50% of the overall RBD-binding memory B cell compartment (Fig. 2c, d, Extended Data Fig. 3d). Similar to natural infection, IGHV3-53 and IGHV3-30, as well as some IGLV genes, were significantly over-represented in the RBD-binding memory B cell compartment of vaccinated individuals (Fig. 2e, Extended Data Fig. 4a). In addition, antibodies that share the same combination of IGHV and IGLV genes in vaccinated individuals comprised 39% of all the clonal sequences (Extended Data Fig. 4b), and 59% when combined with naturally infected individuals5,6 (Fig. 2f); some of these antibodies were nearly identical (Supplementary Tables 3, 4). The number of nucleotide mutations in the V gene in vaccinated individuals is greater than in naturally infected individuals assayed after 1.3 months, but lower than that in the same individuals assayed after 6.2 months (Fig. 2g, Extended Data Fig. 5a). The length of the IgH CDR3 was similar in both naturally infected and vaccinated individuals, and hydrophobicity was below average28 (Fig. 2h, Extended Data Fig. 5a, b). Thus, the IgG memory response is similar in individuals who received the Pfizer–BioNTech and Moderna vaccines, and both of these groups are rich in recurrent and clonally expanded antibody sequences.

One hundred and twenty-seven representative antibodies from eight vaccinated individuals were expressed and tested for reactivity to the RBD (Supplementary Table 5). The antibodies included: (1) 76 that were randomly selected from those that appeared only once, and (2) 51 representatives of expanded clones. Of the antibodies tested, 98% (124 out of 127) bound to the RBD, which indicates that single-cell sorting by flow cytometry efficiently identified B cells that produce anti-RBD antibodies (Extended Data Fig. 6a–c, Supplementary Table 5). In anti-RBD ELISAs, the mean half-maximal effective concentration was higher than that observed for antibodies obtained from naturally infected individuals after 6.2 months but not significantly different from antibodies obtained 1.3 months after infection5,6 (Extended Data Fig. 6a, Supplementary Table 5). To examine memory B cell antibodies for binding to circulating SARS-CoV-2 variants and antibody-resistant mutants, we performed ELISAs using mutant RBDs7,8,13,16,29,30. Twenty-two (26%) of the 84 antibodies we tested showed at least fivefold-decreased binding to at least one of the mutant RBDs (Extended Data Fig. 6d–n, Supplementary Table 5).

We used SARS-CoV-2 S pseudotyped viruses to measure the neutralizing activity of all 127 antibodies5,6,11 (Fig. 3a, Supplementary Table 5). Consistent with the plasma neutralization results, the geometric mean neutralization half-maximal inhibitory concentration (IC50) values of the antibodies of individuals vaccinated with the Pfizer–BioNTech or Moderna vaccine were not significantly different from each other, or to antibody collections obtained from naturally infected individuals at 1.3 or 6.2 months after infection5,6 (Fig. 3a).

Fig. 3: Neutralizing activity of anti-SARS-CoV-2 RBD monoclonal antibodies.
figure3

a, SARS-CoV-2 pseudovirus neutralization assay. IC50 values for antibodies cloned from patients who had recovered from COVID-19 measured at 1.3 and 6.2 months5,6 after infection, as well as antibodies cloned from recipients of the Moderna mRNA-1273 (black) and Pfizer–BioNTech BNT162b2 (red) vaccines. Antibodies with IC50 values above 1,000 ng ml−1 were plotted as 1,000 ng ml−1. Mean of two independent experiments. Red bars and indicated values represent geometric mean IC50 values in ng ml−1. Statistical significance was determined using the two-tailed Mann–Whitney U test. Isotype-control antibody was analysed in parallel and showed no detectable neutralization. b, IC50 values for 17 selected monoclonal antibodies for neutralization of wild-type and the indicated mutant SARS-CoV-2 pseudoviruses. Colour gradient indicates IC50 values ranging from 0 (white) to 1,000 ng ml−1 (red). c, Antibody selection pressure can drive the emergence of S variants in cell culture; the percentage of sequence reads encoding the indicated RBD mutations after a single passage of rVSV–SARS-CoV-2 in the presence of the indicated antibodies is tabulated. Colour gradient indicates the percentage of sequence reads that bear the indicated mutation, ranging from 0 (white) to 100 (red). Positions for which no sequence read was detected are shown in grey. The percentages calculated for a given position are based on all the reads, and not only the reads that include that position. Mutations affecting K417, E484 and N501 are highlighted in b, c, as they constitute important circulating variants.

To examine the neutralizing breadth of the monoclonal antibodies and to begin to map their target epitopes, we tested 17 of the most potent antibodies (Supplementary Table 6)—8 of which carried IgHV3-53—against a panel of 12 SARS-CoV-2 variants: A475V is resistant to class-1 antibodies (structurally defined as previously described7); E484K and Q493R are resistant to class-2 antibodies5,6,7,8,13,14,31,32; and R346S, N439K and N440K are resistant to class-3 antibodies5,6,7,13,14,32. In addition, K417N, Y453F, S477R, N501Y and D614G represent circulating variants, some of which have been associated with rapidly increasing case numbers15,16,20,32,33,34. On the basis of their neutralizing activity against the variants, all but three of the antibodies were provisionally assigned to a defined antibody class or a combination (Fig. 3b). As seen in natural infection, a majority of the antibodies tested (9 out of 17) were at least tenfold-less effective against pseudotyped viruses carrying the E484K mutation5,7,13. In addition, five of the antibodies were less potent against K417N, and four were less potent against N501Y, by tenfold or more (Fig. 3b). We obtained similar results with antibodies that are being developed for clinical use (REGN10987, REGN10933, COV2-2196, COV2-2130, C135 and C144) (Extended Data Fig. 7). However, combinations of antibodies remained effective against all of the variants we tested, which confirms the importance of using combinations of antibodies in the clinic (Extended Data Fig. 7). Whether less-potent antibodies show similar degrees of sensitivity to these mutations remains to be determined.

To determine whether antibody-imposed selection pressure could also drive the emergence of resistance mutations in vitro, we cultured a recombinant vesicular stomatitis virus (rVSV)–SARS-CoV-2 virus in the presence of each of 18 neutralizing monoclonal antibodies. All of the tested antibodies selected for RBD mutations. Moreover, in all cases, the selected mutations corresponded to residues in the binding sites of their presumptive antibody class (Fig. 3b, c). For example, antibody C627—which was assigned to class 2 on the basis of sensitivity to the E484K mutation—selected for the emergence of the E484K mutation in vitro (Fig. 3c). Notably, six of the antibodies selected for K417N, K417E or K417T, five selected for E484K and three selected for N501Y, N501T or N501H, which coincide with mutations that are present in the circulating B.1.1.17/501Y.V1, B.1.351/501Y.V2 and B1.1.28/501.V3 (P.1) variants that have been associated with rapidly increasing case numbers in particular locales15,18,19,35.

Cryo-electron microscopy of antibody epitopes

To further characterize antibody epitopes and mechanisms of neutralization, we characterized seven complexes between Fab fragments of monoclonal antibodies and the prefusion, stabilized ectodomain trimer of SARS-CoV-2 S36 using single-particle cryo-electron microscopy (cryo-EM) (Fig. 4, Supplementary Table 7). Overall resolutions ranged from 5 to 8 Å (Extended Data Fig. 8), and we fit coordinates from S trimer and representative Fab crystal structures by rigid-body docking into the cryo-EM density maps to provide a general assessment of antibody footprints and RBD epitopes. Fab–S complexes exhibited multiple RBD-binding orientations that recognized either the ‘up’ and ‘down’ RBD conformations (Fig. 4a–j) or solely the ‘up’ RBD conformation (Fig. 4k–n), consistent with structurally defined antibody classes from natural infection7 (Fig. 4o). The majority of the monoclonal antibodies that we characterized (6 out of 7) recognized epitopes that included RBD residues involved in ACE2 recognition, which suggests a neutralization mechanism that directly blocks interactions between ACE2 and the RBD. In addition, structurally defined antibody epitopes were consistent with RBD positions that were selected in rVSV–SARS-CoV-2 recombinant virus outgrowth experiments, including residues K417, N439 and N440 (which are in close proximity to one another), E484, and N501 (Fig. 3c, Fig. 4f–j, m, n). Together, these data suggest that functionally similar antibodies are raised during vaccination and natural infection, and that the RBDs of S trimers translated from the mRNA delivered by vaccination adopt both up and down conformations, as observed on structures of trimer ectodomains7 and trimers on the surface of SARS-CoV-2 virions37.

Fig. 4: Cryo-EM reconstructions of Fab–S complexes.
figure4

an, Cryo-EM densities for Fab–S complexes (aek, l) and close-up views of antibody footprints on RBDs (fj, m, n) are shown for neutralizing monoclonal antibodies. As expected, owing to Fab interdomain flexibility, cryo-EM densities (ae, k, l) were weak for the Fab CH–CL domains. Models of antibody footprints on RBDs (fj, m, n) are presented as Fab VH–VL domains (cartoon) complexed with the RBD (surface). To generate models, coordinates of stabilized S trimer (Protein Data Bank code (PDB) 6XKL) and representative Fab fragments (PDB 6XCA or 7K8P) with CDR3 loops removed were fit by rigid-body docking into the cryo-EM density maps. a, f, C669. b, g, C643. c, h, C603. d, i, C601. e, j, C670. k, m, C666. l, n, C663. RBD residues K417, N439, N440, E484 and N501 are highlighted as red surfaces. The N343 glycan is shown as a teal sphere. HC, heavy chain; LC, light chain. o, Composite model illustrating targeted epitopes of RBD-specific neutralizing monoclonal antibodies (shown as VH–VL domains in colours from al) elicited from mRNA vaccines.

Discussion

The mRNA-based SARS-CoV-2 vaccines are safe and effective, and are being deployed globally to prevent infection and disease. The vaccines elicit antibody responses against the RBD (the major target of neutralizing antibodies22,23,24,25,26,27) in a manner that resembles natural infection. Notably, the neutralizing antibodies produced by mRNA vaccination target the same epitopes as those produced by natural infection. Data are consistent with SARS-CoV-2 S trimers translated from the injected RNA adopting a range of different conformations. Moreover, different individuals immunized with the Moderna (mRNA-1273) or Pfizer–BioNTech (BNT162b2) vaccines produce closely related, and nearly identical, antibodies. Whether or not neutralizing antibodies to epitopes other than those involving the RBD are elicited by vaccination remains to be determined.

Human neutralizing monoclonal antibodies to the SARS-CoV-2 RBD can be categorized as belonging to four classes on the basis of their target regions on the RBD7. Class-1 and -2 antibodies are among the most potent and also the most abundant antibodies5,6,22,23,26,38. These antibodies target epitopes that overlap or are closely associated with RBD residues K417, E484 and N501. They are frequently sensitive to mutation in these residues and select for K417N, E484K and N501Y mutations in SARS-CoV-2 S expression libraries in yeast and VSV13,16,32. To avert selection and escape, antibody therapies should be composed of combinations of antibodies that target non-overlapping or highly conserved epitopes6,13,14,32,39,40,41,42,43.

A number of circulating SARS-CoV-2 variants that have been associated with rapidly increasing case numbers and have particular prevalence in the UK (B.1.1.7/501Y.V1), South Africa (B.1.351/501Y.V2) and Brazil (P.1)15,18,19,35. Our experiments indicate that the RBD mutations found in these variants—and potentially others that carry K417N, K417T, E484K and/or N501Y mutations—can reduce the neutralization potency of plasma from vaccinated or naturally infected individuals against SARS-CoV-2 pseudotyped viruses. Although our assays are limited to pseudotyped viruses, there is an excellent correlation between pseudotyped and authentic SARS-CoV-2 neutralization assays11. In addition, similar results have previously been reported using plasma from vaccinated and naturally infected individuals, and a variety of different pseudotype and authentic virus assays16,30,44,45,46,47,48.

The comparatively modest effects of the mutations on viral sensitivity to plasma reflects the polyclonal nature of the neutralizing antibodies in plasma from vaccinated individuals. Nevertheless, emergence of these particular variants is consistent with the dominance of the class-1 and -2 antibody response in infected or vaccinated individuals. We speculate that these mutations emerged in response to immune selection in individuals with nonsterilizing immunity. What the long-term effect of the accumulation of mutations on the SARS-CoV-2 pandemic will be is not known, but the common-cold coronavirus HCoV-229E evolves antigenic variants that are comparatively resistant to the older sera but remain sensitive to contemporaneous sera49. Thus, it is possible that these mutations and others that emerge in individuals with suboptimal or waning immunity will erode the effectiveness of natural and vaccine-elicited immunity. The data suggest that SARS-CoV-2 vaccines and antibody therapies may need to be updated and immunity monitored to compensate for viral evolution.

Methods

Data reporting

No statistical methods were used to predetermine sample size. The experiments were not randomized and the investigators were not blinded to allocation during experiments and outcome assessment.

Study participants

To isolate and characterize anti-SARS-CoV-2 RBD antibodies from vaccinated individuals, a cohort of 20 individuals that participated in the Moderna or Pfizer–BioNTech phase-3 vaccine clinical trials and did not report previous history of SARS-CoV-2 infection was recruited at the NIH Blood Center and the Rockefeller University Hospital for blood donation. Eligible participants included adults who were at least 18 years of age, with no known heart, lung, kidney disease or bleeding disorders, and no history of HIV-1 or malaria infection. All participants were asymptomatic at the time of the study visit and had received a complete two-dose regimen of one of the mRNA vaccines. Informed consent was obtained from all participants and the study was conducted in accordance with Good Clinical Practice. The study visits and blood draws were reviewed and approved under the National Institutes of Health’s Federalwide Assurance (FWA00005897), in accordance with Federal regulations 45 CFR 46 and 21 CFR 5 by the NIH Intramural Research Program IRB committee (IRB no. 99CC0168, ‘Collection and Distribution of Blood Components from Healthy Donors for In vitro Research Use’) and by the Institutional Review Board of the Rockefeller University (IRB no. DRO-1006, ‘Peripheral Blood of Coronavirus Survivors to Identify Virus-Neutralizing Antibodies’). For detailed participant characteristics, see Supplementary Table 1.

Blood sample processing and storage

Samples collected at NIH were drawn from participants at the study visit and processed within 24 h. In brief, whole-blood samples were subjected to Ficoll gradient centrifugation after 1:1 dilution in PBS. Plasma and peripheral blood mononuclear cell (PBMC) samples were obtained through phase separation of plasma layer and Buffy coat phase, respectively. PBMCs were further prepared through centrifugation, red blood cell lysis and washing steps, and stored in CellBanker cell freezing medium (Amsbio). All samples were aliquoted and stored at −80 °C and shipped on dry ice. Before experiments, aliquots of plasma samples were heat-inactivated (56 °C for 1 h) and then stored at 4 °C. PBMCs obtained from samples collected at Rockefeller University were purified as previously reported5,6 by gradient centrifugation and stored in liquid nitrogen in the presence of FCS and DMSO. Heparinized plasma samples were aliquoted and stored at −20 °C or below. Before experiments, aliquots of plasma samples were heat-inactivated (56 °C for 1 h) and then stored at 4 °C.

ELISAs

ELISAs50,51 to evaluate antibodies binding to SARS-CoV-2 S (BioHub), RBD and additional mutated RBDs were performed by coating high-binding 96-half-well plates (Corning 3690) with 50 μl per well of a 1 μg ml−1 protein solution in PBS overnight at 4 °C. Plates were washed 6 times with washing buffer (1× PBS with 0.05% Tween-20 (Sigma-Aldrich)) and incubated with 170 μl per well blocking buffer (1× PBS with 2% BSA and 0.05% Tween-20 (Sigma)) for 1 h at room temperature. Immediately after blocking, monoclonal antibodies or plasma samples were added in PBS and incubated for 1 h at room temperature. Plasma samples were assayed at a 1:66.6 (Rockefeller University samples) or a 1:33.3 (NIH samples) starting dilution. and 7 additional 3-fold serial dilutions. Monoclonal antibodies were tested at 10 μg ml−1 starting concentration, and 10 additional 4-fold serial dilutions. Plates were washed 6 times with washing buffer and then incubated with anti-human IgG, IgM or IgA secondary antibody conjugated to horseradish peroxidase (HRP) (Jackson Immuno Research 109-036-088 109-035-129 and Sigma A0295) in blocking buffer at a 1:5,000 dilution (IgM and IgG) or 1:3,000 dilution (IgA). Plates were developed by addition of the HRP substrate, TMB (Thermo Fisher Scientific) for 10 min (plasma samples) or 4 min (monoclonal antibodies); the developing reaction was then stopped by adding 50 μl 1 M H2SO4, and absorbance was measured at 450 nm with an ELISA microplate reader (FluoStar Omega, BMG Labtech) with Omega and Omega MARS software for analysis. For plasma samples, a positive control (plasma from participant COV725,6, diluted 66.6-fold and with 7 additional 3-fold serial dilutions in PBS) was added to every assay plate for validation. The average of its signal was used for normalization of all of the other values on the same plate with Excel software before calculating the AUC using Prism v.8.4 (GraphPad). For monoclonal antibodies, the half-maximal effective concentration (EC50) was determined using four-parameter nonlinear regression (GraphPad Prism v.8.4).

Expression of RBD proteins

Mammalian expression vectors encoding the RBDs of SARS-CoV-2 (GenBank MN985325.1; S residues 319–539) and eight additional mutant RBD proteins (E484K, Q493R, R346S, N493K, N440K, V367F, A475V, S477N and V483A) with an N-terminal human IL-2 or Mu phosphatase signal peptide have previously been described8.

Cells and viruses

We cultured 293TAce26, 293T/ACE2.cl22 and HT1080/ACE2.cl14 cells11 in Dulbecco’s modified Eagle medium supplemented with 10% fetal bovine serum at 37 °C and 5% CO2. Cells were periodically tested for contamination with mycoplasma or retroviruses.

rVSV–SARS-CoV-2–GFP chimeric virus stocks were generated by infecting 293T/ACE2.cl22 cells. Supernatant was collected at 1 day after infection, cleared of cellular debris and stored at −80 °C. A plaque purified variant designated rVSV/SARS-CoV-2/GFP2E1 that encodes D215G/R683G substitutions was used in these studies11.

Selection and analysis of antibody-escape mutations

For the selection of monoclonal-antibody-resistant S variants, an rVSV/SARS-CoV-2/GFP2E111 population containing 106 infectious units was incubated with monoclonal antibodies at 10–40 μg ml−1 for 1 h at 37 °C. The virus–antibody mixtures were subsequently incubated with 5 × 105 293T/ACE2cl.22 cells in 6-well plates. One day after infection, the medium was replaced with fresh medium containing the equivalent concentration of antibodies. Supernatant was collected 2 days after infection and 150 μl of the cleared supernatant was used to infect cells for passage 2, and 150 μl was subjected to RNA extraction and sequencing.

For identification of putative antibody-resistance mutations, RNA was extracted using NucleoSpin 96 Virus Core Kit (Macherey-Nagel). The RNA was reversed-transcribed using the SuperScript VILO cDNA Synthesis Kit (Thermo Fisher Scientific). KOD Xtreme Hot Start DNA Polymerase (Millipore Sigma) was used for amplification of cDNA using primers flanking the S-encoding sequence. The PCR products were purified and sequenced as previously described5,13. In brief, tagmentation reactions were performed using 1 μl diluted cDNA, 0.25 μl Nextera TDE1 Tagment DNA enzyme (catalogue no. 15027865), and 1.25 μl TD Tagment DNA buffer (catalogue no. 15027866; Illumina). Next, the DNA was ligated to unique i5/i7 barcoded primer combinations using the Illumina Nextera XT Index Kit v.2 and KAPA HiFi HotStart ReadyMix (2×; KAPA Biosystems) and purified using AmPure Beads XP (Agencourt), after which the samples were pooled into one library and subjected to paired-end sequencing using Illumina MiSeq Nano 300 V2 cycle kits (Illumina) at a concentration of 12 pM.

For analysis of the sequencing data, the raw paired-end reads were preprocessed to remove trim adaptor sequences and to remove low-quality reads (Phred quality score < 20) using BBDuk. Reads were mapped to the rVSV–SARS-CoV-2–GFP sequence using Geneious Prime (v.2020.1.2). Mutations were annotated using Geneious Prime, with a P value cutoff of 10−6. Because reads have randomly generated ends and different lengths, the mutations do not necessarily have to occur on the same read: for example, K417N and N501Y might occur on the same read or on different reads. The percentages calculated for position X are calculated on the basis of all the reads, and not only the reads that include position X.

SARS-CoV-2 pseudotyped reporter virus

A panel of plasmids expressing RBD-mutant SARS-CoV-2 S proteins in the context of pSARS-CoV-2-SΔ19 (based on NC_045512) have previously been described13. Additional substitutions were introduced using PCR primer-mediated mutagenesis or with synthetic gene fragments (IDT) followed by Gibson assembly. The mutants E484K and K417N/E484K/N501Y were constructed in the context of a pSARS-CoV-2-SΔ19 variant with a mutation in the furin cleavage site (R683G). The NT50 and IC50 values of these pseudotypes were compared to a wild-type SARS-CoV-2 S sequence carrying R683G in the subsequent analyses, as appropriate.

The generation of SARS-CoV-2 pseudotyped HIV-1 particles was performed as previously described6. In brief, 293T cells were transfected with pNL4-3ΔEnv-nanoluc and pSARS-CoV-2-SΔ19 and pseudotyped virus stocks were collected 48 h after transfection, filtered and stored at −80 °C.

SARS-CoV-2 pseudotype neutralization assays

Plasma or monoclonal antibodies from vaccine recipients were fourfold or fivefold serially diluted and then incubated with SARS-CoV-2 pseudotyped HIV-1 reporter virus for 1 h at 37 °C. The antibody and pseudotyped virus mixture was added to 293TAce2 cells6 (for comparisons of plasma or monoclonal antibodies from individuals who had recovered from COVID-19 and vaccine recipients) or HT1080ACE2.cl14 cells11 (for analysis of S mutants with plasma or monoclonal antibodies from recipients of vaccines). After 48 h, cells were washed with PBS and lysed with Luciferase Cell Culture Lysis 5× reagent (Promega) and nanoluc luciferase activity in lysates was measured using the Nano-Glo Luciferase Assay System (Promega) with the Glomax Navigator (Promega). Relative luminescence units were normalized to those derived from cells infected with SARS-CoV-2 pseudotyped virus in the absence of plasma or monoclonal antibodies. The NT50 and 80% or 90% neutralization titres (NT80 or NT90) for plasma, or IC50 and 90% inhibitory concentrations (IC90) for monoclonal antibodies, were determined using 4-parameter nonlinear regression (least-squares regression method without weighting; constraints: top = 1, bottom = 0) (GraphPad Prism).

Biotinylation of viral protein for use in flow cytometry

Purified and Avi-tagged SARS-CoV-2 RBD was biotinylated using the Biotin-Protein Ligase-BIRA kit according to manufacturer’s instructions (Avidity), as previously described6. Ovalbumin (Sigma, A5503-1G) was biotinylated using the EZ-Link Sulfo-NHS-LC-Biotinylation kit according to the manufacturer’s instructions (Thermo Fisher Scientific). Biotinylated ovalbumin was conjugated to streptavidin–BV711 (BD Biosciences, 563262) and RBD to streptavidin–PE (BD Biosciences, 554061) and streptavidin–AlexaFluor647 (Biolegend, 405237)6.

Flow cytometry and single-cell sorting

Single-cell sorting by flow cytometry was performed as previously described6. In brief, PBMCs were enriched for B cells by negative selection using a pan-B cell isolation kit according to the manufacturer’s instructions (Miltenyi Biotec, 130-101-638). The enriched B cells were incubated in FACS buffer (1× PBS, 2% FCS, 1 mM EDTA) with the following anti-human antibodies (all at 1:200 dilution): anti-CD20–PECy7 (BD Biosciences, 335793), anti-CD3–APC–eFluro 780 (Invitrogen, 47-0037-41), anti-CD8–APC–eFluor 780 (Invitrogen, 47-0086-42), anti-CD16–APC–eFluor 780 (Invitrogen, 47-0168-41), anti-CD14–APC–eFluor 780 (Invitrogen, 47-0149-42), as well as Zombie NIR (BioLegend, 423105) and fluorophore-labelled RBD and ovalbumin (Ova) for 30 min on ice. Single CD3CD8CD14CD16CD20+OvaRBD–PE+RBD–AlexaFluor647+ B cells were sorted into individual wells of 96-well plates containing 4 μl of lysis buffer (0.5× PBS, 10 mM DTT, 3,000 units per ml RNasin Ribonuclease Inhibitors (Promega, N2615)) per well using a FACS Aria III and FACSDiva software (Becton Dickinson) for acquisition and FlowJo for analysis. The sorted cells were frozen on dry ice, and then stored at −80 °C or immediately used for subsequent RNA reverse transcription.

Antibody sequencing, cloning and expression

Antibodies were identified and sequenced as previously described6. In brief, RNA from single cells was reverse-transcribed (SuperScript III Reverse Transcriptase, Invitrogen, 18080-044) and the cDNA stored at −20 °C or used for subsequent amplification of the variable IGH, IGL and IGK genes by nested PCR and Sanger sequencing. Sequence analysis was performed using MacVector. Amplicons from the first PCR reaction were used as templates for sequence- and ligation-independent cloning into antibody expression vectors. Recombinant monoclonal antibodies and Fabs were produced and purified as previously described6.

Cryo-EM sample preparation

Expression and purification of SARS-CoV-2 6P stabilized S trimers36 was conducted as previously described52. Purified Fab and S 6P trimer were incubated at a 1.1:1 molar ratio per protomer on ice for 30 min before deposition on a freshly glow-discharged 300 mesh, 1.2/1.3 Quantifoil copper grid. Immediately before 3 μl of complex was applied to the grid, fluorinated octyl-malotiside was added to the Fab–S complex to a final detergent concentration of 0.02% w/v, resulting in a final complex concentration of 3 mg ml−1. Samples were vitrified in 100% liquid ethane using a Mark IV Vitrobot after blotting for 3 s with Whatman no. 1 filter paper at 22 °C and 100% humidity.

Cryo-EM data collection and processing

Data collection and processing followed a similar workflow to that previously described7. In brief, micrographs were collected on a Talos Arctica transmission electron microscope (Thermo Fisher Scientific) operating at 200 kV for all Fab–S complexes. Data were collected using SerialEM automated data collection software53 and movies were recorded with a K3 camera (Gatan). For all datasets, cryo-EM movies were patch-motion-corrected for beam-induced motion including dose-weighting within cryoSPARC v.2.1554 after binning super-resolution movies. The non-dose-weighted images were used to estimate CTF parameters using cryoSPARC implementation of the Patch CTF job. Particles were picked using Blob picker and extracted, 4× binned and 2D classified. Class averages corresponding to distinct views with secondary structure features were chosen and ab initio models were generated. Three-dimensional classes that showed features of a Fab–S complex were re-extracted, unbinned (0.869 Å per pixel) and homogenously refined with C1 symmetry. Overall resolutions were reported on the basis of gold-standard Fourier shell correlation calculations.

Cryo-EM structure modelling and refinement

Coordinates for initial complexes were generated by docking individual chains from reference structures into cryo-EM density using UCSF Chimera55 (S trimer, PDB 6KXL; Fab, PDB 6XCA or 7K8P, after trimming CDR3 loops and converting to a polyalanine model). Models were then refined into cryo-EM maps by rigid-body and real-space refinement in Phenix56. If the resolution allowed, partial CDR3 loops were built manually in Coot57 and then refined using real-space refinement in Phenix.

Computational analyses of antibody sequences

Antibody sequences were trimmed based on quality and annotated using Igblastn v.1.14. with IMGT domain delineation system. Annotation was performed systematically using Change-O toolkit v.0.4.54058. Heavy and light chains derived from the same cell were paired, and clonotypes were assigned based on their V and J genes using in-house R and Perl scripts (Fig. 2c, f, Extended Data Figs. 3d, 4b). All scripts and the data used to process antibody sequences are publicly available at https://github.com/stratust/igpipeline.

The frequency distributions of human V genes in anti-SARS-CoV-2 antibodies from this study were compared to 131,284,220 previously generated IgH and IgL sequences59 and downloaded from cAb-Rep60, a database of human shared B cell receptor clonotypes (https://cab-rep.c2b2.columbia.edu/). On the basis of the 97 distinct V genes that make up the 4,186 analysed sequences from immunoglobulin repertoire of the 14 participants present in this study, we selected the IgH and IgL sequences from the database that are partially coded by the same V genes and counted them according to the constant region. The frequencies shown in Fig. 2e, Extended Data Fig. 4a are relative to the source and isotype analysed. We used the two-sided binomial test to check whether the number of sequences belonging to a specific IGHV or IGLV gene in the repertoire is different according to the frequency of the same IGV gene in the database. Adjusted P values were calculated using the false discovery rate correction. Significant differences are denoted with asterisks.

Nucleotide somatic hypermutation and CDR3 length were determined using in-house R and Perl scripts. For somatic hypermutations, IGHV and IGLV nucleotide sequences were aligned against their closest germlines using Igblastn and the number of differences were considered nucleotide mutations. The average mutations for V genes were calculated by dividing the sum of all nucleotide mutations across all participants by the number of sequences used for the analysis. To calculate the GRAVY scores of hydrophobicity61 we used Guy H. R. Hydrophobicity scale based on free energy of transfer (kcal mol−1)62 implemented by the R package Peptides (https://journal.r-project.org/archive/2015/RJ-2015-001/RJ-2015-001.pdf). We used 1,405 heavy-chain CDR3 amino acid sequences from this study and 22,654,256 IgH CDR3 sequences from the public database of memory B cell receptor sequences63. The two-tailed Wilcoxon nonparametric test was used to test whether there was a difference in the hydrophobicity distribution.

Data presentation

Figures were arranged in Adobe Illustrator 2020.

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this paper.

Data availability

Data are provided in Supplementary Tables 17. The raw sequencing data and computer scripts associated with Fig. 2 have been deposited at https://github.com/stratust/igpipeline. This study also uses data from refs. 59,63, and from PDB 6VYB and6NB6. Cryo-EM maps associated with data reported in this Article have been deposited in the Electron Microscopy Data Bank (EMDB) (https://www.ebi.ac.uk/pdbe/emdb/) under accession codes EMD-23393 (C601–S), EMD-23394 (C603–S), EMD-23395 (C643–S), EMD-23396 (C663–S), EMD-23397 (C666–S), EMD-23398 (C669–S) and EMD-23399 (C670–S).

Code availability

The computer code used to process the antibody sequences is available at https://github.com/stratust/igpipeline.

References

  1. 1.

    Gaebler, C. & Nussenzweig, M. C. All eyes on a hurdle race for a SARS-CoV-2 vaccine. Nature 586, 501–502 (2020).

    ADS  CAS  PubMed  Article  PubMed Central  Google Scholar 

  2. 2.

    Jackson, L. A. et al. An mRNA vaccine against SARS-CoV-2 – preliminary report. N. Engl. J. Med. 383, 1920–1931 (2020).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  3. 3.

    Krammer, F. SARS-CoV-2 vaccines in development. Nature 586, 516–527 (2020).

    ADS  CAS  PubMed  Article  PubMed Central  Google Scholar 

  4. 4.

    Polack, F. P. et al. Safety and efficacy of the BNT162b2 mRNA Covid-19 vaccine. N. Engl. J. Med. 383, 2603–2615 (2020).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  5. 5.

    Gaebler, C. et al. Evolution of antibody immunity to SARS-CoV-2. Nature, https://doi.org/10.1038/s41586-021-03207-w (2021).

  6. 6.

    Robbiani, D. F. et al. Convergent antibody responses to SARS-CoV-2 in convalescent individuals. Nature 584, 437–442 (2020).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  7. 7.

    Barnes, C. O. et al. SARS-CoV-2 neutralizing antibody structures inform therapeutic strategies. Nature 588, 682–687 (2020).

    ADS  CAS  PubMed  Article  PubMed Central  Google Scholar 

  8. 8.

    Barnes, C. O. et al. Structures of human antibodies bound to SARS-CoV-2 spike reveal common epitopes and recurrent features of antibodies. Cell 182, 828–842 (2020).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  9. 9.

    Corbett, K. S. et al. Evaluation of the mRNA-1273 vaccine against SARS-CoV-2 in nonhuman primates. N. Engl. J. Med. 383, 1544–1555 (2020).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  10. 10.

    Sahin, U. et al. COVID-19 vaccine BNT162b1 elicits human antibody and TH1 T cell responses. Nature 586, 594–599 (2020).

    ADS  CAS  PubMed  Article  PubMed Central  Google Scholar 

  11. 11.

    Schmidt, F. et al. Measuring SARS-CoV-2 neutralizing antibody activity using pseudotyped and chimeric viruses. J. Exp. Med. 217, e20201181 (2020).

    PubMed  PubMed Central  Article  CAS  Google Scholar 

  12. 12.

    Widge, A. T. et al. Durability of responses after SARS-CoV-2 mRNA-1273 vaccination. N. Engl. J. Med. 384, 80–82 (2021).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  13. 13.

    Weisblum, Y. et al. Escape from neutralizing antibodies by SARS-CoV-2 spike protein variants. eLife 9, e61312 (2020).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  14. 14.

    Baum, A. et al. Antibody cocktail to SARS-CoV-2 spike protein prevents rapid mutational escape seen with individual antibodies. Science 369, 1014–1018 (2020).

    ADS  CAS  PubMed  Article  PubMed Central  Google Scholar 

  15. 15.

    Davies, N. G. et al. Estimated transmissibility and severity of novel SARS-CoV-2 variant of concern 202012/01 in England. Preprint at https://doi.org/10.1101/2020.12.24.20248822 (2020).

  16. 16.

    Greaney, A. J. et al. Comprehensive mapping of mutations to the SARS-CoV-2 receptor-binding domain that affect recognition by polyclonal human serum antibodies. Preprint at https://doi.org/10.1101/2020.12.31.425021 (2021).

  17. 17.

    Greaney, A. J. et al. Complete mapping of mutations to the SARS-CoV-2 spike receptor-binding domain that escape antibody recognition. Cell Host Microbe 29, 44–57 (2021).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  18. 18.

    National Institute of Infectious Diseases. Brief report: new variant strain of SARS-CoV-2 identified in travelers from Brazil. https://www.niid.go.jp/niid/en/2019-ncov-e/10108-covid19-33-en.html (2021).

  19. 19.

    Tegally, H. et al. Emergence and rapid spread of a new severe acute respiratory syndrome-related coronavirus 2 (SARS-CoV-2) lineage with multiple spike mutations in South Africa. Preprint at https://doi.org/10.1101/2020.12.21.20248640 (2020).

  20. 20.

    Lauring, A. S. & Hodcroft, E. B. Genetic variants of SARS-CoV-2–what do they mean? J. Am. Med. Assoc. 325, 529–531 (2021).

    CAS  Article  Google Scholar 

  21. 21.

    Amanna, I. J., Carlson, N. E. & Slifka, M. K. Duration of humoral immunity to common viral and vaccine antigens. N. Engl. J. Med. 357, 1903–1915 (2007).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  22. 22.

    Brouwer, P. J. M. et al. Potent neutralizing antibodies from COVID-19 patients define multiple targets of vulnerability. Science 369, 643–650 (2020).

    ADS  CAS  PubMed  PubMed Central  Article  Google Scholar 

  23. 23.

    Cao, Y. et al. Potent neutralizing antibodies against SARS-CoV-2 identified by high-throughput single-cell sequencing of convalescent patients’ B cells. Cell 182, 73–84 (2020).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  24. 24.

    Ju, B. et al. Human neutralizing antibodies elicited by SARS-CoV-2 infection. Nature 584, 115–119 (2020).

    ADS  CAS  PubMed  Article  PubMed Central  Google Scholar 

  25. 25.

    Liu, L. et al. Potent neutralizing antibodies against multiple epitopes on SARS-CoV-2 spike. Nature 584, 450–456 (2020).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  26. 26.

    Rogers, T. F. et al. Isolation of potent SARS-CoV-2 neutralizing antibodies and protection from disease in a small animal model. Science 369, 956–963 (2020).

    ADS  CAS  PubMed  PubMed Central  Article  Google Scholar 

  27. 27.

    Zost, S. J. et al. Potently neutralizing and protective human antibodies against SARS-CoV-2. Nature 584, 443–449 (2020).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  28. 28.

    Briney, B., Inderbitzin, A., Joyce, C. & Burton, D. R. Commonality despite exceptional diversity in the baseline human antibody repertoire. Nature 566, 393–397 (2019).

    ADS  PubMed  PubMed Central  Article  CAS  Google Scholar 

  29. 29.

    Andreano, E. et al. SARS-CoV-2 escape in vitro from a highly neutralizing COVID-19 convalescent plasma. Preprint at https://doi.org/10.1101/2020.12.28.424451 (2020).

  30. 30.

    Thomson, E. C. et al. Circulating SARS-CoV-2 spike N439K variants maintain fitness while evading antibody-mediated immunity. Cell, https://doi.org/10.1016/j.cell.2021.01.037 (2021).

  31. 31.

    Tortorici, M. A. et al. Ultrapotent human antibodies protect against SARS-CoV-2 challenge via multiple mechanisms. Science 370, 950–957 (2020).

    ADS  CAS  PubMed  PubMed Central  Article  Google Scholar 

  32. 32.

    Starr, T. N. et al. Prospective mapping of viral mutations that escape antibodies used to treat COVID-19. Science 371, 850–854 (2021).

  33. 33.

    Singer, J., Gifford, R., Cotten, M. & Robertson, D. CoV-GLUE: a web application for tracking SARS-CoV-2 genomic variation. Preprint at https://doi.org/10.20944/preprints202006.0225.v1 (2020).

  34. 34.

    Elbe, S. & Buckland-Merrett, G. Data, disease and diplomacy: GISAID’s innovative contribution to global health. Glob Chall. 1, 33–46 (2017).

    PubMed  PubMed Central  Article  Google Scholar 

  35. 35.

    Voloch, C. M. et al. Genomic characterization of a novel SARS-CoV-2 lineage from Rio de Janeiro, Brazil. Preprint at https://doi.org/10.1101/2020.12.23.20248598 (2020).

  36. 36.

    Hsieh, C. L. et al. Structure-based design of prefusion-stabilized SARS-CoV-2 spikes. Science 369, 1501–1505 (2020).

    ADS  CAS  PubMed  Article  PubMed Central  Google Scholar 

  37. 37.

    Ke, Z. et al. Structures and distributions of SARS-CoV-2 spike proteins on intact virions. Nature 588, 498–502 (2020).

    ADS  CAS  PubMed  Article  PubMed Central  Google Scholar 

  38. 38.

    Piccoli, L. et al. Mapping neutralizing and immunodominant sites on the SARS-CoV-2 spike receptor-binding domain by structure-guided high-resolution serology. Cell 183, 1024–1042 (2020).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  39. 39.

    Schäfer, A. et al. Antibody potency, effector function, and combinations in protection and therapy for SARS-CoV-2 infection in vivo. J. Exp. Med. 218, e20201993 (2021).

    PubMed  Article  CAS  Google Scholar 

  40. 40.

    Mendoza, P. et al. Combination therapy with anti-HIV-1 antibodies maintains viral suppression. Nature 561, 479–484 (2018).

    ADS  CAS  PubMed  PubMed Central  Article  Google Scholar 

  41. 41.

    Bar-On, Y. et al. Safety and antiviral activity of combination HIV-1 broadly neutralizing antibodies in viremic individuals. Nat. Med. 24, 1701–1707 (2018).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  42. 42.

    Pinto, D. et al. Cross-neutralization of SARS-CoV-2 by a human monoclonal SARS-CoV antibody. Nature 583, 290–295 (2020).

    ADS  CAS  PubMed  Article  PubMed Central  Google Scholar 

  43. 43.

    Rappazzo, C. G. et al. An engineered antibody with broad protective efficacy in murine models of SARS and COVID-19. Preprint at https://doi.org/10.1101/2020.11.17.385500 (2020).

  44. 44.

    Cele, S. et al. Escape of SARS-CoV-2 501Y.V2 variants from neutralization by convalescent plasma. Preprint at https://doi.org/10.1101/2021.01.26.21250224 (2021).

  45. 45.

    Wu, K. et al. mRNA-1273 vaccine induces neutralizing antibodies against spike mutants from global SARS-CoV-2 variants. Preprint at https://doi.org/10.1101/2021.01.25.427948 (2021).

  46. 46.

    Xie, X. et al. Neutralization of N501Y mutant SARS-CoV-2 by BNT162b2 vaccine-elicited sera. Preprint at https://doi.org/10.1101/2021.01.07.425740 (2021).

  47. 47.

    Wibmer, C. K. et al. SARS-CoV-2 501Y.V2 escapes neutralization by South African COVID-19 donor plasma. Preprint at https://doi.org/10.1101/2021.01.18.427166 (2021).

  48. 48.

    Li, Q. et al. The impact of mutations in SARS-CoV-2 spike on viral infectivity and antigenicity. Cell 182, 1284–1294 (2020).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  49. 49.

    Eguia, R. et al. A human coronavirus evolves antigenically to escape antibody immunity. Preprint at https://doi.org/10.1101/2020.12.17.423313 (2020).

  50. 50.

    Amanat, F. et al. A serological assay to detect SARS-CoV-2 seroconversion in humans. Nat. Med. 26, 1033–1036 (2020).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  51. 51.

    Grifoni, A. et al. Targets of T cell responses to SARS-CoV-2 coronavirus in humans with COVID-19 disease and unexposed individuals. Cell 181, 1489–1501 (2020).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  52. 52.

    Cohen, A. A. et al. Mosaic nanoparticles elicit cross-reactive immune responses to zoonotic coronaviruses in mice. Science 371, 735–741 (2021).

  53. 53.

    Mastronarde, D. N. Automated electron microscope tomography using robust prediction of specimen movements. J. Struct. Biol. 152, 36–51 (2005).

    PubMed  Article  PubMed Central  Google Scholar 

  54. 54.

    Punjani, A., Rubinstein, J. L., Fleet, D. J. & Brubaker, M. A. cryoSPARC: algorithms for rapid unsupervised cryo-EM structure determination. Nat. Methods 14, 290–296 (2017).

    CAS  Article  Google Scholar 

  55. 55.

    Goddard, T. D. et al. UCSF ChimeraX: meeting modern challenges in visualization and analysis. Protein Sci. 27, 14–25 (2018).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  56. 56.

    Terwilliger, T. C., Adams, P. D., Afonine, P. V. & Sobolev, O. V. A fully automatic method yielding initial models from high-resolution cryo-electron microscopy maps. Nat. Methods 15, 905–908 (2018).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  57. 57.

    Emsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. Features and development of Coot. Acta Crystallogr. D 66, 486–501 (2010).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  58. 58.

    Gupta, N. T. et al. Change-O: a toolkit for analyzing large-scale B cell immunoglobulin repertoire sequencing data. Bioinformatics 31, 3356–3358 (2015).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  59. 59.

    Soto, C. et al. High frequency of shared clonotypes in human B cell receptor repertoires. Nature 566, 398–402 (2019).

    ADS  CAS  PubMed  PubMed Central  Article  Google Scholar 

  60. 60.

    Guo, Y., Chen, K., Kwong, P. D., Shapiro, L. & Sheng, Z. cAb-Rep: a database of curated antibody repertoires for exploring antibody diversity and predicting antibody prevalence. Front. Immunol. 10, 2365 (2019).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  61. 61.

    Kyte, J. & Doolittle, R. F. A simple method for displaying the hydropathic character of a protein. J. Mol. Biol. 157, 105–132 (1982).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  62. 62.

    Guy, H. R. Amino acid side-chain partition energies and distribution of residues in soluble proteins. Biophys. J. 47, 61–70 (1985).

    ADS  CAS  PubMed  PubMed Central  Article  Google Scholar 

  63. 63.

    DeWitt, W. S. et al. A public database of memory and naive B-cell receptor sequences. PLoS ONE 11, e0160853 (2016).

    PubMed  PubMed Central  Article  CAS  Google Scholar 

  64. 64.

    Rubelt, F. et al. Onset of immune senescence defined by unbiased pyrosequencing of human immunoglobulin mRNA repertoires. PLoS ONE 7, e49774 (2012).

    ADS  CAS  PubMed  PubMed Central  Article  Google Scholar 

  65. 65.

    Baum, A. et al. REGN-COV2 antibodies prevent and treat SARS-CoV-2 infection in rhesus macaques and hamsters. Science 370, 1110–1115 (2020).

    ADS  CAS  PubMed  PubMed Central  Article  Google Scholar 

  66. 66.

    Weinreich, D. M. et al. REGN-COV2, a neutralizing antibody cocktail, in outpatients with Covid-19. N. Engl. J. Med. 384, 238–251 (2021).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

Download references

Acknowledgements

We thank all study participants who devoted time to our research; B. Coller, S. Schlesinger and the Rockefeller University Hospital Clinical Research Support Office and nursing staff; C. M. Rice and all members of the M.C.N. laboratory for helpful discussions and M. Jankovic for laboratory support; and J. Vielmetter and the Protein Expression Center in the Beckman Institute at Caltech for expression assistance. Electron microscopy was performed in the Caltech Beckman Institute Resource Center for Transmission Electron Microscopy, and we thank S. Chen and A. Malyutin for technical assistance. This work was supported by NIH grants P01-AI138398-S1 (M.C.N. and P.J.B.) and 2U19AI111825 (M.C.N.); the Caltech Merkin Institute for Translational Research and P50 AI150464-13 (P.J.B.); a George Mason University Fast Grant (P.J.B.); R37-AI64003 to P.D.B.; and R01AI78788 to T.H. C.O.B. is supported by the HHMI Hanna Gray and Burroughs Wellcome PDEP fellowships. C.G. was supported by the Robert S. Wennett Post-Doctoral Fellowship, in part by the National Center for Advancing Translational Sciences (NIH Clinical and Translational Science Award programme, grant UL1 TR001866), and by the Shapiro–Silverberg Fund for the Advancement of Translational Research. P.D.B. and M.C.N. are Howard Hughes Medical Institute Investigators.

Author information

Affiliations

Authors

Contributions

P.D.B., P.J.B., R.C., T.H., M.C.N., Z.W., F.S., Y.W., F.M., C.O.B., S.F., D.S.-B. and M. Cipolla conceived, designed and analysed the experiments. M. Caskey, C.G., J.A.L., K.A.W. and D.S.-B. designed clinical protocols. Z.W., F.S., Y.W., F.M., C.O.B., S.F., D.S.-B., M. Cipolla, J.D.S., A.G., Z.Y., M.E.A. K.G. and K.E.H.-T. carried out experiments. C.G., M. Caskey, R.A.C., A.H., M.T. and K.G.M. recruited participants and executed clinical protocols. I.S., R.P., J.D., J.X. and C.U.-O. processed clinical samples. T.Y.O. and V.R. performed bioinformatic analysis. R.C., P.D.B., P.J.B., T.H. and M.C.N. wrote the manuscript, with input from all co-authors.

Corresponding authors

Correspondence to Pamela J. Bjorkman or Rafael Casellas or Theodora Hatziioannou or Paul D. Bieniasz or Michel C. Nussenzweig.

Ethics declarations

Competing interests

The Rockefeller University has filed a provisional patent application in connection with this work, on which Z.W. and M.C.N. are inventors (US patent 63/199, 676).

Additional information

Peer review information Nature thanks the anonymous reviewers for their contribution to the peer review of this work.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data figures and tables

Extended Data Fig. 1 Plasma antibodies against SARS-CoV-2.

af, Results of ELISAs measuring plasma reactivity to S (a, c, e) and RBD protein (b, d, f) of 20 vaccinated individuals (grey curves) and 8 control individuals (black curves). a, Anti-S IgG. b, Anti-RBD IgG. c, Anti-S IgM. d, Anti-RBD IgM. e, Anti-S IgA. f, Anti-RBD IgA. Left, optical density at 450 nm (OD 450 nm) for the indicated reciprocal plasma dilutions. Right, normalized AUC values for the 8 control and 20 vaccinated individuals. Horizontal bars indicate geometric mean. Statistical significance was determined using the two-tailed Mann–Whitney U test. Average of two or more experiments. gi, Correlations of plasma antibody measurements. g, Normalized AUC for IgG anti-S (x axis) plotted against normalized AUC for IgG anti-RBD (y axis). h, Normalized AUC for IgM anti-S (x axis) plotted against normalized AUC for IgM anti-RBD (y axis). i, Normalized AUC for IgA anti-S (x axis) plotted against normalized AUC for IgA anti-RBD (y axis). The r and P values in gi were determined with the two-tailed Spearman’s correlation test. Individuals vaccinated with the Moderna vaccine are in black; individuals vaccinated with the Pfizer–BioNTech vaccine are in red. jl, Results of ELISAs measuring plasma reactivity to RBD in volunteers who had recovered from COVID-19 at 1.3 and 6.2 months after infection5,6 and in 20 vaccinated individuals, who received the Moderna vaccine (black dots) or Pfizer–BioNTech vaccine (red dots). j, Anti-RBD IgG. k, Anti-RBD IgM. l, Anti-RBD IgA. The normalized AUC values are shown. Positive and negative controls were included for validation. Red horizontal bars and indicated values represent geometric mean. Statistical significance was determined using two-tailed Mann–Whitney U test or Wilcoxon matched-pairs signed rank test.

Extended Data Fig. 2 Plasma neutralizing activity.

a, Anti-S IgM AUC (y axis) plotted against NT50 (x axis). r = 0.12, P < 0.62. b, Anti-S IgA AUC (y axis) plotted against NT50 (x axis). r = 0.79, P < 0.0001. c, Anti-RBD IgM AUC (y axis) plotted against NT50 (x axis). r = −0.079, P = 0.74. d, Anti-RBD IgA AUC (y axis) plotted against NT50 (x axis). r = 0.69, P = 0.0008. e, NT50 (y axis) plotted against time between last dose and blood draw (x axis). r = −0.63, P = 0.0032. f, NT50 (y axis) plotted against time between doses (x axis). r = 0.03, P = 0.89. g, Anti-RBD IgG AUC (y axis) plotted against time between last dose and blood draw (x axis). r = −0.57, P = 0.0087. h, Anti-S IgG AUC (y axis) plotted against time between last dose and blood draw (x axis). r = −0.59, P = 0.0064. i, Age (y axis) plotted against NT50 (x axis). r = −0.06, P = 0.82. The r and P values were determined by two-tailed Spearman’s. Individuals vaccinated with the Moderna vaccine are in black; individuals vaccinated with  the Pfizer–BioNTech vaccine are in red. j, NT50 values for neutralization with plasma from vaccinated individuals (n = 15) of pseudotyped viruses with wild-type and the indicated RBD-mutant SARS-CoV-2 S. P values determined using one-tailed t-test.

Extended Data Fig. 3 Flow cytometry.

a, Gating strategy used for cell sorting. Gating was on singlets that were CD20+ and CD3CD8CD14CD16Ova. Sorted cells were RBD–PE+ and RBD–AlexaFluor647+. b, Flow cytometry showing the percentage of RBD-double-positive memory B cells from a pre-COVID-19 control (HD) and 15 vaccinated individuals, who received the Moderna vaccine (shown in black) or Pfizer–BioNTech vaccine (shown in red). c, The percentage of RBD-binding memory B cells in vaccinated individuals (y axis) plotted against time between first dose and blood draw (x axis) (r = 0.40, P = 0.087) (left) and between last dose and blood draw (x axis) (r = 0.33, P = 0.17) (right). Individuals vaccinated with the Moderna vaccine are in black; individuals vaccinated with the Pfizer–BioNTech vaccine are in red. The r and P values for correlations were determined by two-tailed Spearman’s correlation test. d, Pie charts show the distribution of antibody sequences from ten individuals in b. The number in the inner circle indicates the number of sequences analysed. Pie-slice size is proportional to the number of clonally related sequences. The black outline indicates the frequency of clonally expanded sequences.

Extended Data Fig. 4 Frequency distributions of human VL genes.

Graph shows relative abundance of human IGVK (left) and IGVL (right) genes of Sequence Read Archive accession SRP010970 (database) (orange)64, and vaccinated individuals (blue). Two-sided binomial tests with unequal variance were used to compare the frequency distributions. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. b. Sequences from 14 individuals (Supplementary Table 3) with clonal relationships. Interconnecting lines indicate the relationship between antibodies that share V and J gene segment sequences at both IGH and IGL. Purple, green, and grey lines connect related clones, clones and singles, and singles to each other, respectively.

Extended Data Fig. 5 Antibody somatic hypermutation, and CDR3 length.

a, Number of somatic nucleotide mutations in both the IGVH and IGVL in 14 participants (left). Individuals who received the Moderna vaccine are shown in black, and individuals who received the Pfizer–BioNTech vaccine are shown in in red. For each individual, the number of amino acids in the CDR3 at the IGVH and IGVL is shown (right). The horizontal bars indicate the mean. The number of antibody sequences (IGVH and IGVL) evaluated for each participant are n = 68 (MOD1), n = 45 (MOD2), n = 117 (MOD3), n = 123 (MOD4), n = 110 (MOD6), n = 109 (MOD7), n = 144 (MOD8), n = 102 (MOD9), n = 132 (PFZ10), n = 109 (MOD11), n = 91 (PFZ12), n = 78 (C001), n = 66 (C003) and n = 115 (C004). b, Distribution of the hydrophobicity GRAVY scores at the IgH CDR3 compared to a public database (Methods). The box limits are the lower and upper quartiles, the centre line indicates the median, the whiskers are 1.5× interquartile range and the dots represent outliers. Statistical significance was determined using two-tailed Wilcoxon matched-pairs signed-rank test. ****P < 0.0001.

Extended Data Fig. 6 Monoclonal antibody ELISAs.

a, Graphs show anti-SARS-CoV-2 RBD antibody reactivity. ELISA EC50 values for all antibodies isolated from individuals who had recovered from COVID-19 assayed at 1.3 and 6.2 months after infection5,6 and 127 selected monoclonal antibodies isolated from 4 individuals who received the Moderna vaccine (black dots) and 4 who received the Pfizer–BioNTech vaccine (red dots), measured at 8 weeks after the boost. Red horizontal bars and indicated values represent geometric mean. Statistical significance was determined using two-tailed Mann–Whitney U test. b, c, Graphs show ELISA titration curves for 86 monoclonal antibodies isolated from individuals vaccinated with the Moderna vaccine (b) and 41 monoclonal antibodies isolated from individuals vaccinated with the Pfizer–BioNTech vaccine (c). dl, Graphs show ELISA titrations for 84 antibodies isolated from individuals vaccinated with the Moderna vaccine against the indicated RBD variants. Isotype control and low-binding antibodies are indicated in colours. C661 is a nonbinding antibody. Data are representative of two independent experiments. m, Relative change in EC50 values for the indicated RBD variants over wild-type RBD of 84 antibodies isolated from individuals vaccinated with the Moderna vaccine. Red horizontal bars represent geometric mean. n, Heat-map summary of EC50 values for binding to wild-type RBD and the indicated mutant RBDs for 17 top neutralizing antibodies.

Extended Data Fig. 7 Neutralizing activity of monoclonal antibodies in clinical development against SARS-CoV-2 variants.

Results of a SARS-CoV-2 pseudovirus neutralization assay. IC50 values for six different monoclonal antibodies, alone or in their clinically designated combinations, for neutralization of wild-type and the indicated mutant SARS-CoV-2 pseudotyped viruses. Antibodies with IC50 values above 1,000 ng ml−1 are plotted as 1,000 ng ml−1. Data are the mean of two independent experiments. Colour gradient indicates IC50 values, ranging from 0 (white) to 1,000 ng ml−1 (red). The combination of REGN10987 and REGN10933 (casirivimab and imdevimab, respectively)14,65,66 has been granted emergency use authorization by the US Food and Drug Administration; the combination of COV2-2196 and COV2-2130 (licensed to AstraZeneca as AZD7442)27 and the combination of C135 and C144 (Rockefeller University)6 are currently in clinical trials (NCT04507256 and NCT04700163, respectively).

Extended Data Fig. 8 Local resolution estimates of Fab–S cryo-EM reconstructions.

ag, Local-resolution maps calculated using cryoSPARC for C669–S (a), C643–S (b), C603-S (c), C601–S (d), C666–S (e), C663–S (f) and C670–S (g) complexes. Close-up views for Fab–RBD interfaces are highlighted for C669 (a) and C643 (b). h, Gold-standard Fourier shell correlation curves for Fab–S complexes. The 0.5 and 0.143 cut-offs are indicated by dashed lines.

Supplementary information

Reporting Summary

Supplementary Table 1

Individual vaccine characteristics.

Supplementary Table 2

Neutralization of WT/KEN pseudovirus by convalescent plasma.

Supplementary Table 3

Antibody sequences from vaccines.

Supplementary Table 4

CDR3 alignment of highly identical clonal sequences.

Supplementary Table 5

Sequences, half maximal effective concentrations (EC50s) and inhibitory concentrations (IC50s) of the cloned monoclonal antibodies.

Supplementary Table 6

Neutralization activity of mAbs against mutant SARS-CoV-2 pseudoviruses.

Supplementary Table 7

Cryo-EM data collection and processing statistics.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Wang, Z., Schmidt, F., Weisblum, Y. et al. mRNA vaccine-elicited antibodies to SARS-CoV-2 and circulating variants. Nature (2021). https://doi.org/10.1038/s41586-021-03324-6

Download citation

Further reading

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing