Increased memory B cell potency and breadth after a SARS-CoV-2 mRNA boost

The Omicron variant of SARS-CoV-2 infected many vaccinated and convalescent individuals1–3. Despite the reduced protection from infection, individuals who received three doses of an mRNA vaccine were highly protected from more serious consequences of infection4. Here we examine the memory B cell repertoire in a longitudinal cohort of individuals receiving three mRNA vaccine doses5,6. We find that the third dose is accompanied by an increase in, and evolution of, receptor-binding domain (RBD)-specific memory B cells. The increase is due to expansion of memory B cell clones that were present after the second dose as well as the emergence of new clones. The antibodies encoded by these cells showed significantly increased potency and breadth when compared with antibodies obtained after the second dose. Notably, the increase in potency was especially evident among newly developing clones of memory cells, which differed from persisting clones in targeting more conserved regions of the RBD. Overall, more than 50% of the analysed neutralizing antibodies in the memory compartment after the third mRNA vaccine dose neutralized the Omicron variant. Thus, individuals receiving three doses of an mRNA vaccine have a diverse memory B cell repertoire that can respond rapidly and produce antibodies capable of clearing even diversified variants such as Omicron. These data help to explain why a third dose of a vaccine that was not specifically designed to protect against variants is effective against variant-induced serious disease.

The Omicron variant of SARS-CoV-2 infected many vaccinated and convalescent individuals [1][2][3] . Despite the reduced protection from infection, individuals who received three doses of an mRNA vaccine were highly protected from more serious consequences of infection 4 . Here we examine the memory B cell repertoire in a longitudinal cohort of individuals receiving three mRNA vaccine doses 5,6 . We find that the third dose is accompanied by an increase in, and evolution of, receptor-binding domain (RBD)-specific memory B cells. The increase is due to expansion of memory B cell clones that were present after the second dose as well as the emergence of new clones. The antibodies encoded by these cells showed significantly increased potency and breadth when compared with antibodies obtained after the second dose. Notably, the increase in potency was especially evident among newly developing clones of memory cells, which differed from persisting clones in targeting more conserved regions of the RBD. Overall, more than 50% of the analysed neutralizing antibodies in the memory compartment after the third mRNA vaccine dose neutralized the Omicron variant. Thus, individuals receiving three doses of an mRNA vaccine have a diverse memory B cell repertoire that can respond rapidly and produce antibodies capable of clearing even diversified variants such as Omicron. These data help to explain why a third dose of a vaccine that was not specifically designed to protect against variants is effective against variant-induced serious disease.
We studied the immune responses to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) mRNA vaccination in a longitudinal cohort of 42 volunteers with no prior history of SARS-CoV-2 infection 5,6 who were recruited between 21 January 2021 and 14 December 2021 for sequential blood donation. The volunteers received either the Moderna (mRNA-1273; n = 8) or Pfizer-BioNTech (BNT162b2; n = 34) mRNA vaccine. Volunteers ranged in age from 23 to 78 years old; 48% were male and 52% were female (Methods and Supplementary Table 1). Samples were obtained at the following time points: 2.5 weeks after the prime, 1.3 and 5 months after the second vaccine dose and 1 month after the third dose.

Plasma binding and neutralization
Plasma IgM, IgG and IgA responses to SARS-CoV-2 Wuhan-Hu-1 receptor-binding domain (RBD) were measured by enzyme-linked immunosorbent assay (ELISA) 5,6 . While a significant decrease was observed in antibody reactivity at 5 months after the second vaccine dose, anti-RBD IgG titres were significantly increased after a third dose of an mRNA vaccine (P < 0.0001) ( Fig. 1a and Supplementary Table 1). The resulting titres were similar to those observed 1.3 months after the second dose (P > 0.99) (Fig. 1a). IgM and IgA titres were lower than IgG titres, and, although IgM titres were unchanged during the observation period, IgA titres were also significantly increased after the third vaccine dose (Extended Data Fig. 1 and Supplementary Table 1).
The plasma neutralizing activity in 42 participants was measured using HIV-1 pseudotyped with Wuhan-Hu-1 SARS-CoV-2 spike protein 5,6 ( Fig. 1b and Supplementary Table 1). While a 7.3-fold decrease in neutralizing titres occurred between 1.3 and 5 months after the second vaccine dose, administration of a third vaccine dose boosted neutralizing titres 11.9-fold, resulting in a geometric mean half-maximal neutralizing titre (NT 50 ) of 3,244 against Wuhan-Hu-1 (Fig. 1b). Plasma neutralizing antibodies elicited by mRNA vaccination are more potent against Wuhan-Hu-1 than variants 5,6 . Consistent with prior reports 3,7-10 , the third vaccine dose significantly boosted geometric mean NT 50 values by 16-fold, 12-fold and 37-fold for the Beta, Delta and Omicron BA.1 variants, respectively. The level of activity against the Beta and Delta variants was not significantly different from that against Wuhan-Hu-1, whereas the activity against Omicron BA.1 was 16-fold lower than that against Wuhan-Hu-1 (P = 0.58, P = 0.24 and P = 0.0013, respectively) (Fig. 1c). The correlates of protective neutralizing titres against Omicron BA.1 are not defined. Nevertheless, given the correlation between neutralizing antibody levels and protection from Wuhan-Hu-1 infection [11][12][13] , the reduced activity against Omicron BA.1 in recipients of a third dose of vaccine probably explains why vaccinated individuals remained particularly susceptible to infection by this variant.

Memory B cells
Under physiological conditions, memory B cells produce little if any secreted antibody. However, when challenged with antigen as in a breakthrough infection, these cells undergo clonal expansion and produce antibody-secreting plasma cells and memory and germinal centre B cells 14 . To examine the effects of the third vaccine dose on the memory compartment in our longitudinal cohort, we performed flow cytometry experiments using phycoerythrin (PE)-and Alexa-Fluor-647 (AF647)-labelled Wuhan-Hu-1 RBDs (Fig. 2a and Extended Data Fig. 2). Individuals who received a third vaccine dose developed significantly increased numbers of Wuhan-Hu-1 RBD-binding memory cells compared with those who received only two doses or who were naturally infected 5,6,15 (Fig. 2a, b). The number of memory cells produced after the third dose was not significantly higher than for vaccinated convalescent individuals (P = 0.08) (Fig. 2b). An increased proportion of memory B cells circulating after the third dose expressed IgG and lower levels of CD71 (Extended Data Fig. 2c).
We obtained 1,370 paired antibody sequences from five individuals who were sampled 5 months after the second vaccine dose and 1 month after the third vaccine dose. Two and three of these participants were additionally sampled 2.5 weeks after the first dose and 1.3 months after the second dose, respectively 5,6 ( Fig. 2c and Supplementary Table 2). After the third vaccine dose, all individuals examined showed expanded clones of memory B cells that were either persisting clones of memory B cells of the same clonal family detected at two or more time points after the initial vaccination or unique clones of memory B cells detected at only a single time point (Fig. 2c). Clones observed uniquely after the third vaccine dose could represent entirely new lineages elicited by the boost or memory cells that were present below the limit of detection at earlier time points. As at earlier time points, the IGHV3-30, IGHV3-53 and IGHV4-31 genes were over-represented 1 month after the third dose 5,6 (Extended Data Fig. 3). Thus, there is a persistent bias in IGHV gene representation in memory that is common to most individuals.
Expanded clones of memory cells accounted for 33% and 47% of the repertoire 5 months after the second dose and 1 month after the third dose, respectively ( Fig. 2c and Extended Data Fig. 4a). The relative increase in clonality was due in part to an average 3.1-fold expansion of persisting Wuhan-Hu-1 RBD-specific memory B cells (P < 0.0001) (Fig. 2d). Consistent with the relatively modest number of additional cell divisions by persisting clones, these clones accumulated on average only two additional somatic mutations, suggesting a germinal centre-independent process 14 ( Fig. 2e and Extended Data Fig. 4b).
In comparison with the increase observed in the number of newly emerging unique clones of memory cells at 5 months after the second dose, the increase in these clones after the third dose was more modest (1.7-fold) and did not reach statistical significance (P = 0.086) (Fig. 2d). Antibodies from these cells were more mutated than those from the unique clones present 5 months after the second vaccine dose, as were antibodies that were represented only once (singlets). In both cases, the numbers of somatic mutations were significantly greater than those observed 5 months after the second dose, indicating persisting evolution and cell division (P = 0.0009 and P < 0.0001, respectively) ( Fig. 2e and Extended Data Fig. 4). Therefore, a third mRNA vaccine dose is associated with expansion and further evolution of the memory B cell compartment.

Monoclonal antibodies
We cloned the sequences for and expressed 472 monoclonal antibodies, including 1 representative monoclonal antibody from each clonally expanded family and at least 9 randomly selected monoclonal antibodies from individual memory B cells detected only once in each participant (Supplementary Table 3). When tested by ELISA, 459 of these antibodies bound to Wuhan-Hu-1 RBD, indicating the high efficiency of the RBD-specific memory B cell isolation method used here (Extended Data Fig. 5 and Supplementary Table 3). Moreover, 191 antibodies obtained after the third vaccine dose were compared with 34 antibodies isolated after the prime as well as with 79 and 168 antibodies isolated 1.3 and 5 months after the second vaccine dose, respectively. Overall, there was no significant change in binding over time or with the number of vaccine doses (Extended Data Fig. 5a and Supplementary Table 3). This was true for all antibodies combined as a b  16,33 in the plasma samples shown in a. c, Plasma neutralizing activity against the indicated SARS-CoV-2 variants of interest or concern for n = 15 randomly selected samples assayed in HT1080Ace2cl.14 cells.
Wuhan-Hu-1 and Omicron BA.1 NT 50 values were derived from ref. 7 . Pseudoviruses in c were based on a spike protein that also included the R683G substitution, which disrupts the furin cleavage site and increases particle infectivity. All experiments were performed at least in duplicate. Red bars and values in a-c represent geometric mean values. Statistical significance in a and b was determined by two-tailed Kruskal-Wallis test with subsequent Dunn's multiple-comparisons test. Statistical significance in c was determined by two-tailed Friedman test with subsequent Dunn's multiple-comparisons test. P values are as indicated.

Article
well as for persisting clones detected at multiple time points, unique clones detected at only a single time point and single antibodies (Extended Data Fig. 5a-c).
All 459 Wuhan-Hu-1 RBD-binding antibodies were subjected to a SARS-CoV-2 pseudotype neutralization assay based on the Wuhan-Hu-1 SARS-CoV-2 spike protein 5,6 . There was not a significant change in antibody potency against Wuhan-Hu-1 between 1.3 and 5 months after the second vaccine dose (half-maximal inhibitory concentration (IC 50 ) of 290 versus 182 ng ml −1 , P = 0.60) (Fig. 3a); however, antibody potency was greater after the third vaccine dose than 5 months after the second dose (IC 50 of 111 versus 182 ng ml −1 , P = 0.049) (Fig. 3a). The overall improvement between equivalent time points after the second and third doses, from an IC 50 of 290 ng ml −1 to 111 ng ml −1 was highly significant (P = 0.0023) ( Fig. 3a    (P = 0.69) [15][16][17] (Fig. 3a). The improved neutralizing activity was most evident among unique clones, with a marked change in IC 50 from 323 ng ml −1 1.3 months after the second vaccine dose to 67 ng ml −1 at 1 month after the third dose (P = 0.034) ( Fig. 3b and Supplementary Table 3). Persisting clones also showed improved neutralizing activity after the third dose (P = 0.043); although a trend towards improved neutralizing activity was evident among single antibodies, the difference did not reach statistical significance (Fig. 3b, Extended Data Fig. 5d and Supplementary Tables 3 and 4). In all cases, the relative potency of the antibodies isolated 1 month after the third dose was similar to that of antibodies isolated from convalescent vaccinated individuals 12 months after infection (Fig. 3a, b). Taken together, these results indicate significant improvement in the neutralizing potency against Wuhan-Hu-1 of the antibodies expressed in the memory B cell compartment 1 month after administration of a third mRNA vaccine dose compared with that observed 1.3 months after the second dose. Newly detected singlets and unique clones of expanded memory B cells detected only after the third vaccine dose accounted for most of the improvement in neutralizing activity between 5 months after the second dose and 1 month after the third dose.

Epitopes and neutralization breadth
The majority of the anti-RBD neutralizing antibodies obtained from vaccinated individuals after the second vaccine dose belong to class 1 and class 2, which target a region overlapping the ACE2-binding site 18,19 ( Fig. 4a). These antibodies are generally more potent than class 3 and class 4 antibodies, which target the more conserved base of the RBD and do not directly interfere with ACE2 binding 17 ( Fig. 4a and Extended Data Fig. 6). Whereas class 1 and class 2 antibodies that develop early are susceptible to mutations in and around the ACE2-binding site found in many of the variants of concern, evolved versions of the same antibodies can be resistant 15,20 . According to structural information and sequence conservation among betacoronaviruses, antibodies that RBD Class 1    Table 4) and 134 previously reported 6 . Red bars and numbers indicate geometric mean values. Statistical significance was determined by two-tailed Kruskal-Wallis test with subsequent Dunn's multiple-comparisons test. All experiments were performed at least twice.

Article
span class 3 or class 4 and either class 1 or class 2 are potentially broader than those spanning class 1 or class 2 alone because their epitopes include conserved sequences, and they might be more potent than antibodies from class 3 or class 4 alone because they could interfere directly with the interaction between ACE2 and the RBD (Fig. 4b and Extended Data Fig. 6).
To examine the epitopes targeted by RBD-binding antibodies after the third vaccine dose, we performed biolayer interferometry (BLI) experiments in which a preformed antibody-RBD (Wuhan-Hu-1) complex was exposed to a second antibody targeting one of the four classes of structurally defined epitopes (C105 for class 1, C144 for class 2, C135 for class 3 and C118 for class 1/4; refs. 16,18 ) (Fig. 4a). Among the 168 random RBD-binding antibodies tested, 20, 29 and 36 exhibited neutralizing activity with an IC 50 of lower than 1,000 ng ml −1 from 1.3 and 5 months after the second vaccine dose and 1 month after the third dose, respectively. As might be expected, the largest group of RBD-binding antibodies obtained after the second vaccine dose belonged to class 1/2 (Fig. 4c). Although the overall distribution of classes for antibodies binding RBD did not change significantly between 1.3 and 5 months after the second dose, the relative representation of class 1 and class 2 antibodies decreased (Fig. 4c). This trend continued after the third vaccine dose, with increased representation of RBD-binding antibodies in class 1/4 and class 3, resulting in a significant difference in the epitope distribution for RBD-binding antibodies between the early time points after the second and third dose (P = 0.005) (Fig. 4c). As expected, these differences could primarily be accounted for by the emergence of unique clones and singlets after the third vaccine dose (Fig. 4d) C3145  17  7  8  17  10  9  9  29  12  336  5  159  C3095  18  4  9  8  8  7  5  18  10  3  3  13   C3011  14  2  8  3  6  7  3  20  11  5  2  105  C3119  17  2  7  7  8  6  4  19  9  2 20  2  1 ,000   C3034  12  6  7  151  8  3 4  7  14  9  9 90  6  6 67  C3057  10  3  7  4  6  5  3  11  6  1 2  3  47   C2523  12  4  8  95  7  6  5  14  12  1,000  3  1,000  C2524  15  3  6  3  7  5  4  16  10  128  2  120   C2485  7  2  3  174  4  3  2  7   antibodies, with an initial dominance of antibodies in class 1/2 followed by increasing representation of class 1/4 and class 3 over time (Fig. 4e). The neutralizing breadth of the antibodies elicited by infection increases significantly after 6 months 15,17,20 , whereas no statistically significant increase in breadth has been observed 5 months after the second dose of an mRNA vaccine 6 . To determine whether neutralizing antibodies in clones that persisted from 5 months after the second dose to 1 month after the third dose develop increased breadth, we compared 18 antibody pairs that were randomly selected across participants and had measurable neutralizing activity against Wuhan-Hu-1. Neutralizing activity was measured against a panel of SARS-CoV-2 pseudoviruses harbouring RBD amino acid substitutions representative of variants, including Delta and Omicron BA.1 (Fig. 5a and Supplementary Table 5). The clonal pairs were dominated by antibodies belonging to classes 1/2, 2/3 and 3, as determined by BLI (Fig. 5a). Of the 18 antibody pairs, 15 neutralized pseudovirus carrying the Delta RBD amino acid substitutions at low antibody concentrations at both time points, with IC 50 values of <1-154 ng ml −1 (Fig. 5a and Supplementary Table 5). Although the Omicron BA.1 pseudovirus showed the highest degree of neutralization resistance, 11 of the 18 antibodies isolated 1 month after the third dose neutralized this virus, 9 at an IC 50 of below 120 ng ml −1 (Fig. 5a and Supplementary Table 5). Most pairs of antibodies obtained from clones persisting between 5 months after the second vaccine dose and 1 month after the third dose showed exceptionally broad neutralization, and there was little change in antibody breadth within the analysed pairs (Fig. 5a).
We extended the analysis to compare the activity of antibodies present in memory cells at 1.3 months after the second dose 6 to that of antibodies detected uniquely 1 month after the third vaccine dose. A representative group of antibodies was selected across participants and was tested against viruses pseudotyped with spike proteins containing the RBD of Wuhan-Hu-1, Delta and Omicron BA.1 (Fig. 5b and Supplementary Table 6). We found that the proportion of Omicron BA.1-neutralizing antibodies increased from 15% after the second dose to 50% among the unique antibodies found after the third dose (P = 0.035, Fisher's exact test; Fig. 5b). This increase in frequency of Omicron-neutralizing antibodies is consistent with the observation that a third vaccine dose increased the frequency of Omicron RBD-binding B cells compared with samples collected shortly before the boost 21 . Among all antibodies evaluated, the increase in breadth between the second and third vaccine doses was reflected by an increase in potency from 689 to 124 ng ml −1 for the IC 50 against Omicron BA.1 (P = 0.0004; Fig. 5c and Supplementary Tables 5 and 6). Similar results were observed for Delta neutralization (Fig. 5c). Thus, memory B cell clones emerging after the third vaccine dose showed increasing breadth and potency against pseudoviruses representing variants that were not present in the vaccine.
Finally, we compared the neutralization breadth of antibodies elicited by the third vaccine dose, measured approximately 9 months after the prime, with that of antibodies obtained from a cohort of convalescent unvaccinated individuals 12 months after infection. The latter showed a significant increase in neutralizing breadth over time after infection [15][16][17] (Fig. 5d). The two groups of antibodies were equally and remarkably broad. Moreover, 92% and 94% of the convalescent and third-dose antibodies neutralized pseudoviruses carrying the Beta RBD, and 27% and 56% neutralized Omicron BA.1, respectively. Thus, the antibodies elicited by the third dose of vaccine were at least as broad as those elicited by infection (Fig. 5d).

Discussion
Memory B cells can develop from the germinal centre or directly from a germinal centre-independent activated B cell compartment 14 . B cells residing in germinal centres undergo multiple rounds of division, mutation and selection, whereas those in the activated compartment undergo only a limited number of divisions and carry fewer mutations 14 . Both pathways remain active throughout the immune response 22,23 . Our data indicate that the third dose of mRNA vaccines against SARS-CoV-2 expands persisting clones of memory B cells and a cohort of previously undetected clones that carry mutations indicative of germinal centre residence. The latter differ from persistent clones in that they appear to target more conserved regions of the RBD. Several different mechanisms could account for the antigenic shift, including epitope masking by high-affinity antibodies elicited by earlier vaccine doses that primarily target the less conserved RBD 18,19,24 .
Passively administered antibodies are protective against SARS-CoV-2 infection and can prevent serious disease if provided early [25][26][27][28][29] . The third dose of mRNA vaccines boosts plasma antibody responses to multiple SARS-CoV-2 variants, including Omicron, although the levels are insufficient to prevent breakthrough infection in many individuals 2,3 . The third dose also elicits increased numbers of memory B cells that express more potent and broader antibodies 10,21 . Although our data do not exclude the possibility that Omicron-specific memory was present before and was unaffected by the boost, others have demonstrated that the boost increases the frequency of Omicron RBD-binding memory B cells 21 . Our antibody cloning data provide a mechanistic explanation for the observed increase in Omicron-specific memory B cells after the boost. Although the memory B cells expressing more potent and broader antibodies do not appear to contribute to circulating plasma antibody levels, upon challenge with antigen in the form of a vaccine or infection, they produce large amounts of antibodies within 3-5 days 30 . Passive administration of antibodies within this same time window prevents the most serious consequences of infection 25,28,31 . Thus, rapid recall by memory T cells and a diversified and expanded memory B cell compartment are likely to be key mechanisms that contribute to the enhanced protection against severe disease by a third mRNA vaccine dose.

Online content
Any methods, additional references, Nature Research reporting summaries, source data, extended data, supplementary information, acknowledgements, peer review information; details of author contributions and competing interests; and statements of data and code availability are available at https://doi.org/10.1038/s41586-022-04778-y.

Study participants
Participants were healthy volunteers who had previously received the initial two-dose regimen of either the Moderna (mRNA-1273) or Pfizer-BioNTech (BNT162b2) mRNA vaccine against the WT (Wuhan-Hu-1) strain of SARS-CoV-2. For this study, participants were recruited for serial blood donations at The Rockefeller University Hospital in New York between 21 January and 14 December 2021. 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. The majority of participants (n = 32) were follow-ups from a longitudinal cohort that we previously reported on 5,6 , while a smaller subgroup of individuals (n = 10) was de novo recruited for this study (for details, see Supplementary Table 1). Eligible participants (n = 42) were healthy adults with no history of infection with SARS-CoV-2 during or before the observation period (as determined by clinical history and confirmed through serology testing) who had received two doses of one of the two currently approved SARS-CoV-2 mRNA vaccines, Moderna (mRNA-1273) and Pfizer-BioNTech (BNT162b2). Additionally, a subgroup of individuals (n = 34) who had received a third vaccine dose was included. The specifics of each participant's vaccination regimen were at the discretion of the individual and their healthcare provider, consistent with current dosing and interval guidelines, and, as such, not influenced by participation in our study. Exclusion criteria included incomplete vaccination status (defined as fewer than two doses), the presence of clinical signs and symptoms suggestive of acute infection with SARS-CoV-2, or positive reverse transcription PCR (RT-PCR) results for SARS-CoV-2 in saliva or positive COVID-19 serology. No other parameters that could lead to potential self-selection bias were used to exclude or include patients. After enrolment, participant plasma samples obtained after the third vaccine dose were tested for binding activity toward the nucleocapsid (N) protein (Sino Biological, 40588-V08B) of SARS-CoV-2. The absence of serological activity towards N protein was used to ensure a negative history for infection with SARS-CoV-2 for each participant. Participants presented to The Rockefeller University Hospital for blood sample collection and were asked to provide details of their vaccination regimen, possible side effects, comorbidities and possible COVID-19 history. Clinical data collection and management were conducted using the software iRIS by iMedRIS (version 11.02). All participants provided written informed consent before participation in the study, and the study was conducted in accordance with Good Clinical Practice. The study was performed in compliance with all relevant ethical regulations, and the protocol (DRO-1006) for studies with human participants was approved by the Institutional Review Board of The Rockefeller University. For detailed participant characteristics, see Supplementary Table 1.

Blood sample processing and storage
Peripheral blood mononuclear cells (PBMCs) obtained from samples collected at The Rockefeller University were purified as previously reported by gradient centrifugation and were stored in liquid nitrogen in the presence of FCS and dimethylsulfoxide (DMSO) 16,17 . Heparinized plasma and serum samples were aliquoted and stored at −20 °C or lower temperatures. Before experiments, aliquots of plasma samples were heat inactivated (56 °C for 1 h) and then stored at 4 °C.

ELISAs
ELISAs 34,35 used to evaluate antibody binding to SARS-CoV-2 Wuhan-Hu-1 RBD 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. The plates were washed six times with washing buffer (1× PBS with 0.05% Tween-20 (Sigma-Aldrich)) and were 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 to PBS and were incubated for 1 h at room temperature. Plasma samples were assayed at a 1:66 starting dilution with 10 additional threefold serial dilutions. Monoclonal antibodies were tested at a starting concentration of 10 μg ml −1 with 10 additional fourfold serial dilutions. Plates were washed six times with washing buffer and were then incubated with anti-human IgG, IgM or IgA secondary antibody conjugated to horseradish peroxidase (HRP) ( Jackson ImmunoResearch, 109-036-088 and 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 3,3′,5,5′-tetramethylbenzidine (TMB) (ThermoFisher) for 10 min (plasma samples) or 4 min (monoclonal antibodies). The developing reaction was stopped by adding 50 μl of 1 M H 2 SO 4 , 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 COV72, diluted 66.6-fold with 10 additional 3-fold serial dilutions in PBS) was added to every assay plate for normalization. The average of its signal was used for normalization of all other values on the same plate with Excel software before calculating the area under the curve using Prism version 9.3 (GraphPad). Negative controls of pre-pandemic plasma samples from healthy donors were used for validation; details of this process have previously been reported 16 . For monoclonal antibodies, the ELISA half-maximal effective concentration (EC 50 ) was determined using four-parameter nonlinear regression (GraphPad Prism version 9.3). EC 50 values above 1,000 ng ml −1 were considered to indicate non-binders.