High-affinity CD16A polymorphism associated with reduced risk ofsevere COVID-19

Abstract

CD16A is an activating Fc receptor on NK cells that mediates antibody-dependent cellular cytotoxicity (ADCC), a key mechanism in antiviral immunity. However, the role of NK cell-mediated ADCC in SARS-CoV-2 infection remains unclear, particularly whether it limits viral spread and disease severity or contributes to the immunopathogenesis of COVID-19. We hypothesized that the high-affinity CD16AV176 polymorphism influences these outcomes. Using an in vitro reporter system, we demonstrated that CD16AV176 is a more potent and sensitive activator than the common CD16AF176 allele. To assess its clinical relevance, we analyzed 1,027 patients hospitalized with COVID-19 from the Immunophenotyping Assessment in a COVID-19 cohort (IMPACC), a comprehensive longitudinal dataset with extensive transcriptomic, proteomic, and clinical data. The high-affinity CD16AV176 allele was associated with a significantly reduced risk of ICU admission, mechanical ventilation, and severe disease trajectories. Lower anti-SARS-CoV-2 IgG titers were correlated to CD16AV176; however, there was no difference in viral load across CD16A genotypes. Proteomic analysis revealed that participants homozygous for CD16AV176 had lower levels of inflammatory mediators. These findings suggest that CD16AV176 enhances early NK cell-mediated immune responses, limiting severe respiratory complications in COVID-19. This study identifies a protective genetic factor against severe COVID-19, informing future host-directed therapeutic strategies.

Keywords: COVID-19; Cellular immune response; Immunology; Innate immunity; NK cells.

Figure 1. Development of CELLISA assay to measure CD16A responses to antibodies from plasma.

(A) Schematic of CELLISA assay showing the steps involved to perform functional assays to measure anti-Spike RBD antibodies using BWZ.CD16A reporters or NK cells. (B and C) Functional assays using BWZ.CD16A^F176 stimulated using plasma from vaccinated healthy donors (B) or acute patients with COVID-19 (C) incubated with Ancestral or Omicron Spike RBD. (D–F) Stimulation of NK cells from PBMCs cultured overnight in IL-2 with plasma from vaccinated donor plasma incubated with Ancestral or Omicron Spike RBD. Representative flow plots (D) and quantification of degranulation as measured by CD107A⁺ (E) and IFN-γ production (F). Data in E and F are mean ± SEM, each dot representative of replicates (n = 6–8). Data were analyzed using a 2-tailed Student’s test (*P < 0.05).

Figure 2. The high affinity CD16A allele is more sensitive at recognizing antibodies generated through SARS-CoV-2 vaccination or infection.

(A–C) Functional assays using BWZ.CD16A^F176 and BWZ.CD16A^V176 stimulated by plasma from vaccinated healthy donors at different time points. Data shown are results using Ancestral spike RBD (A), Delta spike RBD (B), and Omicron spike RBD (C). (D–F) Functional assays using BWZ.CD16A^F176 and BWZ.CD16A^V176 stimulated by plasma from patients with COVID-19 at acute (10–14 days following clinical diagnosis) and convalescent (2–3 months after initial onset of symptoms) time points. Data shown are results using ancestral spike RBD (D), Delta spike RBD (E), and Omicron spike RBD (F). Data are shown as mean ± SEM with each dot representing biological replicates (n = 8). Data were analyzed using 2-way ANOVA. *P < 0.0332; **P < 0.0021; ****P < 0.0001.

Figure 3. High-affinity CD16A^V176 is associated with protection from severe COVID-19.

(A) Breakdown of IMPACC participants by CD16A genotype, determined by Infinium Global Diversity Array sequencing of DNA. CC, homozygous for high affinity allele (n = 101); AC, heterozygous (n = 420); AA, homozygous for low affinity allele (n = 506). (B) Trajectory group 4 and 5 (severe COVID-19) status stratified by the number of high-affinity CD16A alleles present (odds ratio [OR] = 0.796, 95% CI = 0.641–0.985, adjusted P = 0.038). (C) ICU status among CD16A genotypes (OR = 0.767, 95% CI = 0.626–0.936, P = 0.009). (D) Percentage of IMPACC participants that were ever mechanically ventilated split by CD16A genotype (OR = 0.662, 95% CI = 0.515-0.845, P = 0.001). (E) Respiratory status breakdown by CD16A genotype at visit 1 (<48 hours from hospital admission), P = 0.004. 3 = hospitalized + no O₂, 4 = hospitalized + required O₂, 5 = noninvasive ventilation or high flow O₂, and 6 = mechanical ventilation or ECMO. For all statistical tests unless otherwise stated, the high-affinity CD16A^V176 variant allele count was treated as the independent variable and adjusted for age and sex. Analyses were modeled as clinical outcome-Y on CD16A^V176 variant allele count-X by binomial regression for B–D and ordinal logistic regression for E.

Figure 4. High-affinity CD16A^V176 is associated with lower anti–SARS-CoV-2 antibody titers and globally reduced soluble mediators of inflammation.

(A) Antibody levels against SARS-CoV-2 RBD were measured via ELISA as AUC and stratified by CD16A genotype (P = 5.6 × 10^–4). (B) Antibody levels against SARS-CoV-2 spike were measured via ELISA and stratified by CD16A genotype (P = 0.001). (C) SARS-CoV-2 viral load reported as viral reads per million (rpM) was determined from nasal metagenomics and compared across CD16A genotypes (P = 0.148). Data from A–C are from cisit 1 (<48 hours from hospital admission). (D) Oligonucleotide-linked antibody detection (Olink) was performed on n = 948 participants to quantify cytokines, chemokines, and soluble receptors in serum. Displayed are the 12 that were significantly different across CD16A genotypes (P ≤ 0.05) after correcting for multiple testing. CC, homozygous high affinity. (E and F) Cytometry by time of flight (CyTOF) was performed on whole blood of n = 788 participants and immune cell counts were determined. NK cell counts (P = 0.136) in E and CD39^loCD4⁺ Treg counts (P = 3.0 × 10^–6) in F were recorded split by CD16A genotype. Analyses were modeled as clinical outcome-Y on CD16A^V176 variant allele count-X by linear regression.