Patterns and functional consequences of antibody speciation in maternal-fetal transfer of coronavirus-specific humoral immunity

Abstract

Maternal antibodies serve as a temporary form of inherited immunity, providing humoral protection to vulnerable neonates. Whereas IgG is actively transferred up a concentration gradient via the neonatal Fc Receptor (FcRn), maternal IgA and IgM are typically excluded from fetal circulation. Further, not all IgG molecules exhibit the same transfer efficiency, being influenced by subclass, Fab and Fc domain glycosylation, antigen-specificity, and the temporal dynamics of maternal antibody responses. Here, we investigate the phenotypes and functions of maternal and cord blood antibodies induced by SARS-CoV-2 infection and compare them to those induced by mRNA vaccination, focusing on breadth of antigen recognition and antiviral functions including neutralization and effector function. While cord blood coronavirus-specific antibody functional breadth and potency appeared to be more compromised than binding breadth and potency in both groups, vaccination induced substantially greater function and breadth in cord blood than did natural infection. These functional phenotypes were associated with speciation of the maternal serum repertoires, as some IgG subpopulations were enriched while others were relatively depleted from cord blood. Relevant to the continued protection of vulnerable infants in the context of a diversifying pathogen, key observations included the greater breadth of antibody effector functions as compared to neutralization, which was associated with greater affinity for antigen and the more efficient placental transfer of IgG subclasses with better affinity to Fc receptors. This work provides new insights into the binding and functional breadth of inherited antibody responses that are likely responsible for the protection of infants born to seropositive mothers from severe SARS-CoV-2 infection despite continued viral diversification.

Fig 1. Antibody isotype, magnitude, and breadth across SARS-CoV-2 VOC in maternal and cord blood samples following vaccination or natural infection.

A. Coronavirus-specific antibody response features after dimensional reduction in maternal (filled) and cord (open) samples among convalescent (n = 38) (blue) or vaccinated (n = 50) (green) individuals. B. Volcano plot presenting the fold-change (x-axis) and statistical significance (Mann Whitney test, y-axis) of differences between convalescent and vaccinated participants. Antibody isotype is indicated by shape, with RBD and whole spike indicated in hollow and filled symbols, respectively. SARS-CoV-2 variant is indicated by color. C. Median Fluorescent Intensity (MFI) of IgG responses to spike of SARS CoV-2 VOCs as defined by multiplex assay for IgM (top), IgA (middle), and IgG (bottom). Responses among SARS-CoV-2 naïve subjects (n = 38) are shown in black. Bar indicates the median response. D. Breadth-potency (fraction of whole versus MFI) curves represent the fraction of subjects with a response exceeding a given level for IgG antibody responses across the panel of VOC. Population means are shown with a thick line, and individual subjects illustrated in thin lines. E. IgG breadth scores, as defined by geometric mean MFI across antigens, for each subject. Bar indicates the median. Statistical significance was defined by ANOVA Kruskal–Wallis test with Dunn’s correction and α = 0.05 (*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001).

Fig 2. Vaccination and infection elicit distinct IgG subclass and Fc receptor breadth in cord blood.

A. Breadth–potency curves representing the fraction of subjects with a response exceeding a given level for each IgG subclass (left) and for binding to FcR (right) across the panel of VOC in cord blood samples following maternal vaccination or infection. Population mean is shown with a thick line, and individual subjects are illustrated in thin lines. B. IgG subclass (top) and FcR binding (bottom) breadth scores for each cord blood samples following maternal vaccination (filled) or infection (hollow). Statistical significance was defined by ANOVA Kruskal–Wallis test with Dunn’s correction and α = 0.05 (**p < 0.01, ***p < 0.001, ****p < 0.0001). C. Transfer ratio (cord/maternal levels) of antigen-specific (SARS: Wuhan and Omicron Spike; Control: Pertussis and Tetanus) IgG subclasses in vaccinated (green) and convalescent (blue) dyads. D. Median transfer ratios of each SARS-CoV-2 spike VOC as compared to control antigens by subclass. Bars indicate median.

Fig 3. mRNA vaccination results in superior breadth of SARS-CoV-2- specific Ab effector function and neutralization in cord blood.

A. Ab effector functions in cord blood from vaccinated (top, green shading) and convalescent (bottom, blue shading) cord blood for SARS spike variants or pertussis and tetanus control (red box) antigens. Phagocytosis, ADCC, and Complement deposition activities were assessed at each of three serum dilutions (1:50, 1:100, 1:250). Functional activity is reported in arbitrary units (AU), relative light units (RLU), and median fluorescent intensity (MFI). Inset. Functional breadth scores across variants in maternal, cord, and naïve subject samples. Statistical significance was defined by ANOVA Kruskal–Wallis test with Dunn’s correction and α = 0.05. B-C. Transfer ratios (cord/maternal) of Fc effector functions to indicated antigens in vaccinated (green) or convalescent (blue) dyads at the 1:50 dilution for individual antigens (B) and for the set of SARS and control antigens (C). D. Neutralization titers (NT₅₀) observed for a subset of vaccinated (n = 23) (green) and convalescent (n = 26) (blue) maternal and cord samples against Wuhan (black) and Omicron (orange) strains. The limit of detection (LOD) is indicated by the horizontal dotted line. E. Transfer ratios of neutralization activity. Bars indicates median. Unless otherwise noted, data presented includes samples from 37 naive subjects, and 50 vaccinated and 38 convalescent dyads.

Fig 4. Contributions of each isotype to antibody effector functions in cord and maternal blood.

Maternal (M) and cord (C) blood samples from vaccinated (green, n = 15) and convalescent (n = 15) dyads depleted of IgM (top), IgA (center), and IgG (bottom) antibodies. A-D. Binding levels (A, left) and effector functions (B-D) (right) to Wuhan spike protein were measured on mock and depleted (+) samples for each isotype to measure efficiency of each depletion and the impact on antibody activity against Wuhan spike antigen. Mock and depleted (+) samples were compared using a paired mixed effect model corrected for multiple hypothesis testing using the Benjamini, Krieger, and Yekutieli to control the false discovery rate (*q < Q, where Q = 0.05). Functional activity is reported in arbitrary units (AU), relative light units (RLU), and median fluorescent intensity (MFI).

Fig 5. Cross reactivity to emergent and endemic coronavirus.

A. Coronavirus-specific antibody response features after dimensional reduction in maternal (filled) and cord (open) samples among convalescent (n = 38) (blue) or vaccinated (n = 50) (green) individuals. Naive subjects (n = 37) are shown in black. IgG binding experiments were performed at 1:2500 and 1:5000 dilutions. IgM, IgA, and IgG binding responses to SARS-CoV-1 S and MERS S. B. IgM, IgA, and IgG binding responses to endemic coronaviruses HKU1, OC43, NL63, and 229E. All antigens are full length spike. C. ADCP, ADCC, and ADCD functional responses to SARS-CoV-1 spike. D. ADCP, ADCC, and ADCD responses against OC43 S. E. IgM, IgA, and IgG antibody binding profiles to endemic coronavirus antigens in other conformations (HKU1 S1, OC43S2, OC43S2P, NL62 S1, 229E S1). F. ADCP, ADCC, and ADCD responses against OC43 S2 antigen. Functional activity is reported in arbitrary units (AU), relative light units (RLU), and median fluorescent intensity (MFI). G. Antibody association and dissociation traces for affinity analysis of binding to OC43 S2 for a subset of samples from each group (n = 5) at 10 mM. H. Binding of each subject to the panel of antigens measured. I. Heatmap antibody affinities (K_D M) to each antigen tested. Darker red denotes higher affinity binding interaction.