Fc-γR-dependent antibody effector functions are required for vaccine-mediated protection against antigen-shifted variants of SARS-CoV-2

Abstract

Emerging severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variants with antigenic changes in the spike protein are neutralized less efficiently by serum antibodies elicited by legacy vaccines against the ancestral Wuhan-1 virus. Nonetheless, these vaccines, including mRNA-1273 and BNT162b2, retained their ability to protect against severe disease and death, suggesting that other aspects of immunity control infection in the lung. Vaccine-elicited antibodies can bind Fc gamma receptors (FcγRs) and mediate effector functions against SARS-CoV-2 variants, and this property correlates with improved clinical coronavirus disease 2019 outcome. However, a causal relationship between Fc effector functions and vaccine-mediated protection against infection has not been established. Here, using passive and active immunization approaches in wild-type and FcγR-knockout mice, we determined the requirement for Fc effector functions to control SARS-CoV-2 infection. The antiviral activity of passively transferred immune serum was lost against multiple SARS-CoV-2 strains in mice lacking expression of activating FcγRs, especially murine FcγR III (CD16), or depleted of alveolar macrophages. After immunization with the pre-clinical mRNA-1273 vaccine, control of Omicron BA.5 infection in the respiratory tract also was lost in mice lacking FcγR III. Our passive and active immunization studies in mice suggest that Fc-FcγR engagement and alveolar macrophages are required for vaccine-induced antibody-mediated protection against infection by antigenically changed SARS-CoV-2 variants, including Omicron strains.

Extended Data Fig. 1. Gating strategy for Luminex-based and Fc effector function assays

(a) Gating for Luminex-bead based antibody binding to spike-coated beads. (b) Gating for ADNP assay showing CD66⁺ neutrophils with opsinophagocytosed beads. (c) Gating for ADCP assay showing THP-1 monocytes and opsinophagocytosed beads. (d) Gating for ADCD assay showing complement deposition on spike and antibody coated beads. (e) Gating for NK cell activation assay showing CD107a expression.

Extended Data Fig. 2. T cell responses in mRNA-1273 vaccinated wild-type and FcγR KO mice

(a) Representative flow cytometry plots show gating scheme for quantification of spike-specific CD4⁺ and CD8⁺ T cell responses in the spleen of wild-type, FcγR III KO, and FcγR I/III/IV KO mice at day 10 after boosting with control or mRNA-1273 vaccines. (b-c) At day 10 after boosting, the spleen of wild-type and FcγR KO mice were harvested, and T cell responses were measured after spike peptide re-stimulation. Splenocytes were incubated overnight with class I MHC (b) or class II MHC (c) immunodominant spike peptides, and the percentages and numbers of IFNγ and TNFα positive CD8⁺ (b) or CD4⁺ (c) T cells were quantified by intracellular staining and flow cytometry. Data are pooled from two experiments (in order left to right n = 10, 10, 8, 10, 10, 9 (b-c)). Comparisons were made between groups that received the mRNA 1273 vaccine (one-way ANOVA with Tukey’s post-test; all comparisons were not significant; column height indicates mean values).

Extended Data Fig. 3. Levels of anti-BA.5 antibody in mice passively transferred vaccine-elicited serum antibody

(a) Levels of anti BA.5 spike IgG in serum of mice that were passively transferred naïve or immune sera. Amounts are compared to those in vaccine-elicited immune serum before transfer (n = 5 mice per group, columns indicate mean values). (b) Neutralizing antibody response against SARS-CoV-2 BA.5 using sera from naïve (circles) or spike protein vaccinated (grey squares) mice. Also shown is serum neutralizing antibody activity from recipient wild-type mice one day after transfer of immune sera (black squares). Data are representative of results with n = 5 mice per group.

Extended Data Fig. 4. FcγR expression on myeloid cells in the lung

Lung cells from wild-type, FcγR I KO, FcγR III KO, and FcγR I/III/IV KO mice were stained with antibodies for FcγR I, FcγR III, or FcγR IV. After gating on live cells, alveolar macrophages, neutrophils, and monocytes were defined (see Extended Data Fig 5). The data are representative of results with n = 3 mice per group, and histograms are shown.

Extended Data Fig. 5. Gating scheme for analysis of cell populations in the blood and lung

(a) Immune cell populations in the blood of C57BL/6 mice were analyzed using the indicated gating scheme and conjugated antibodies. After gating on live single cells, monocytes (P5) were defined as CD45⁺ CD11b^hi Ly6C^hi; neutrophils (P6) were defined as CD45⁺ CD11b^hi Ly6G^hi; natural killer (NK) cells (P7) were defined as CD45⁺ NK1.1⁺; and B cells (P8) were defined as CD45⁺ B220⁺ cells. (b) Immune cell populations in the lungs of C57BL/6 mice were analyzed using the indicated gating scheme and conjugated antibodies. After gating on live single cells, alveolar macrophages (P1) were defined as CD45⁺ CD11c⁺ Siglec-F⁺; interstitial macrophages (P2) were defined as CD45⁺ CD64⁺, eosinophils (P3) were defined as CD45⁺ CD11b⁺ Siglec-F⁺; CD11b dendritic cells (P4) were defined CD45⁺ CD11b⁺ CD11c⁺, Siglec-F^-, MHC II⁺; monocytes (P5) were defined as CD45⁺ Ly6C^hi; neutrophils (P6) were identified as CD45⁺ Ly6G⁺; natural killer (NK) cells (P7) were defined as CD45⁺ NK1.1⁺; and B cells (P8) were defined as CD45⁺ B220⁺.

Figure 1.. Systems serology analysis of vaccine-induced immune sera.

(a-c) Levels of IgG1 (a), IgG2b (b), IgG2c (c) that bind to SARS-CoV-2 spike [Wuhan-1, B.1.617.2, BA.1, and BA.4/5], or influenza hemagglutinin (HA) in naïve and vaccine-induced immune sera. (d-f) Levels of spike- or HA-binding IgG antibodies that engage FcγR IIb (d), FcγR III (e), or FcγR IV (f) in naïve and vaccine-induced immune sera. (g-j) Antibody effector functions. Antibody-mediated cellular phagocytosis with monocytes (ADCP, g) or neutrophils (ADNP, h) activity using vaccine-induced immune (squares) or naïve (circles) sera and beads coated with SARS-CoV-2 Wuhan-1 and BA.4/5 spike proteins and murine monocytes (g-h) or human donors (i) (bars indicate mean ± SEM; n = 4 (g); n = 3 (h) mice per group, one experiment (g), two experiments (h); one-way ANOVA with Tukey’s post-test; ns, not significant; in order left to right **P = 0.0042, **P = 0.0083 (g); **P = 0.0012, *P = 0.0106 (h)). (i) CD107a surface expression on natural killer cells (ADNKA) after incubation with beads encoded with Wuhan-1 or BA4/5 spike proteins and immune sera (bars indicate median values; n = 2 donors per group, one experiment; one-way ANOVA with Tukey’s post-test; ns, not significant). (j) Deposition of complement (ADCD) on beads coated with indicated SARS-CoV-2 spike or influenza HA proteins after treatment with naïve or immune sera.

Figure 2.. Vaccine-derived immune serum control of SARS-CoV-2 MA-10 infection in wild-type C1q KO, and FcγR I/III/IV KO mice.

(a) Scheme of passive transfer, virus challenge and tissue harvest. (b) Neutralizing antibody responses against SARS-CoV-2 MA-10 using sera from naïve (circles) or Wuhan-1 spike protein vaccinated mice (pooled from animals immunized and boosted with mRNA-1273 or ChAd-SARS-CoV-2-S) (squares). Also shown is serum neutralizing antibody activity from recipient wild-type and FcγR I/III/IV KO mice one day after transfer of immune sera. (c-g) Twelve-week-old male wild-type, C1q KO, and FcγR I/III/IV KO C57BL/6 mice were passively transferred by intraperitoneal injection 60 μL of naïve or vaccine-induced immune sera one day before intranasal challenge with 10³ FFU of SARS-CoV-2 MA-10. At 4 dpi, viral RNA in the nasal wash (c), nasal turbinates (d), and lungs (f) was quantified, and infectious virus in the nasal turbinates (e) and lungs (g) was determined (bars indicate mean ± SEM; in order left to right n = 5 (b); n = 6, 6, 7, 7, 6, 7 (c); n = 6, 6, 7, 7, 6, 7 (d); n = 6, 6, 7, 7, 6, 7 (e); n = 6, 6, 7, 7, 6, 7 (f); n = 6, 6, 7, 7, 6, 7 (g) mice per group, one experiment (b), two experiments (c-g), dotted lines show limit of detection [LOD]). One-way ANOVA with Tukey’s post-test; ns, not significant; *P = 0.0188 (f); ****P < 0.0001 (g)); additional statistical comparisons are shown in Supplementary Table 3. In c-g, the LOD are weight and volume based, and vary based on the amount of material collected for RNA extraction.

Figure 3.. Vaccine-elicited immune serum control of SARS-CoV-2 WA1/2020 N501/D614G infection in wild-type, FcγR I KO, FcγR II KO, FcγR III KO, and FcγR I/III/IV KO mice.

(a) Scheme of passive transfer, virus challenge, and tissue harvest. (b) Neutralizing antibody response against SARS-CoV-2 WA1/2020 N501Y/D614G using sera from naïve (circles) or Wuhan-1 spike protein vaccinated (squares) mice. Also shown is serum neutralizing antibody activity from recipient wild-type and FcγR I/III/IV KO mice one day after transfer of immune sera. (c-g) Twelve-week-old male wild-type, FcγR I KO, FcγR II KO, FcγR III KO, and FcγR I/III/IV KO mice were passively transferred by intraperitoneal injection 60 μL of naïve or vaccine-immune sera one day before intranasal challenge with 10⁴ FFU of WA1/2020 N501Y/D614G. At 4 dpi, viral RNA and infectious virus were measured in the upper respiratory tract (nasal wash, c; nasal turbinates, d-e; or lungs, f-g). Panels c-e: wild-type mice only; panels f-g: wild-type, FcγR I KO, FcγR II KO, FcγR III KO, and FcγR I/III/IV KO mice (bars indicate mean ± SEM; in order left to right n =5 (b); n = 18, 11 (c); n = 18, 11 (d); n = 18, 11 (e); n = 18, 9, 9, 12, 10, 11, 8, 10, 11, 11 (f); n = 18, 9, 9, 12, 10, 11, 8, 10, 11, 11 (g) mice per group, one experiment (b), three experiments (c-g), dotted lines show LOD). One-way ANOVA with Tukey’s post-test (ns, not significant; **P = 0.0068, ****P < 0.0001 (f); ***P = 0.0002, ****P < 0.0001 (g)); additional statistical comparisons are shown in Supplementary Table 3.

Figure 4.. Control of SARS-CoV-2 BA.5 infection after mRNA-1273 vaccination of wild-type and FcγR-deficient C57BL/6 mice.

(a) Scheme of immunization, serum sampling, virus challenge, and tissue harvest. (b-c) Anti-Wuhan-1 (b) and BA.4/5 (c) RBD IgG responses from serum of mice immunized with control or mRNA-1273 vaccines. (d-e) Neutralizing antibody responses against WA1/2020 N501Y/D614G (d) and BA.5 (e) from serum collected from wild-type, FcγR I KO, FcγR III KO, and FcγR I/III/IV KO mice 25 days after completion of a two-dose primary vaccination series with control (closed circles) or mRNA-1273 (open circles). (f-g) Nine-week-old male wild-type, FcγR I KO, FcγR III KO, and FcγR I/III/IV KO mice were immunized twice at four-week intervals with control or mRNA-1273 vaccine via intramuscular route. Four weeks after the primary vaccination series, mice were challenged via intranasal route with 10⁴ FFU of BA.5. At 3 dpi, infectious virus in the nasal turbinates (f) and lungs (g) was determined (boxes illustrate geometric mean titers [GMT], dotted lines show LOD, bars indicate mean ± SEM; in order left to right n = 10, 10, 10, 10, 25, 10, 10, 14 (b); n = 12, 10, 10, 10, 22, 10, 10, 15 (c); n = 22, 9, 9, 12, 30, 15, 15, 15 (d); n = 22, 9, 9, 12, 25, 10, 10, 15 (e); n = 19, 10, 8, 8, 19, 9, 10, 10 (f); n = 19, 10, 8, 8, 19, 9, 10, 10 (g) mice per group, one experiment (b-e), two experiments (f-g), dotted lines show LOD, one-way ANOVA with Dunnett’s test (ns, not significant, ***P = 0.002, ****P < 0.0001 (f); **P = 0.0044, ****P < 0.0001 (g)); additional statistical comparisons are shown in Supplementary Table 3.

Figure 5.. Depletion of alveolar macrophages impairs the protective activity of passively-transferred immune sera against BA.5 infection.

(a) Scheme for depletion of neutrophils and monocytes, passive transfer of immune sera, and BA.5 challenge. (b-d) Analysis of depletion of immune cell subsets in blood of mice receiving anti-Gr-1 (Ly6C/Ly6G) or isotype antibody at 3 dpi. Summary of different cell types (b) based on the gating strategy (see Extended Data Fig 5a). Representative flow cytometry dot plots showing depletion of neutrophils (c) and monocytes (d) with numbers indicating the cell population as a percentage of CD45⁺ cells. (e-f) Analysis of infectious viral burden at 3 dpi in nasal turbinates (e) and lungs (f) after BA.5 challenge. (g) Scheme for depletion of alveolar macrophages, passive transfer of immune sera, and BA.5 challenge. (h) Analysis of depletion of immune cell subsets in lungs of mice receiving liposomes at 3 dpi. Summary of different cell types (h) based on gating strategy (see Extended Data Fig 5b). (i) Representative flow cytometry dot plots showing depletion of alveolar macrophages after liposome administration with numbers indicating the cell population as a percentage of CD45⁺ cells. (j-k) Analysis of infectious viral burden at 3 dpi in nasal turbinates (j) and lungs (k) after BA.5 challenge (boxes illustrate GMT, bars indicate mean; in order left to right n = 10, 10, 10, 6 (b); n = 10, 9, 9, 6 (e); n = 10, 9, 9, 6 (f); n = 9, 10, 7, 12 (h); n = 9, 10, 7, 12 (j); n = 9, 10, 7, 12 (k) mice per group, two experiments (b, e-f, h, j-k), dotted lines show LOD, Mann-Whitney test with Bonferroni post-test correction (b, h), Mann-Whitney test (e-f, j-k), ns, not significant, **P = 0.0077, ****P < 0.0001, ***P = 0.0003 (b); **P = 0.003, ***P = 0.0007, **P = 0.0034, **P = 0.0011 (h); ****P < 0.0001 (k).