What immune genetic associations underlie antibody Fc-profiles across twins?

227 Fc-profiles measured across 510 twins

The study asked whether the quality, not just the quantity, of antibody responses is shaped by inherited genetic variation. Antibodies have two working ends: the Fab region recognizes targets, while the Fc region determines subclass, isotype and post-translational modifications that set which Fc receptors are engaged and which effector functions follow. The authors profiled 227 pathogen-specific antibody Fc-profiles against 20 antigens from 16 viral and bacterial pathogens plus 2 allergens.

The cohort came from TwinsUK: 510 self-reported female twins of white European descent, aged 41 to 78 (mean 61), comprising 168 monozygotic and 342 dizygotic matched twins. Of these, 498 had whole-genome genotyping data. Each assay was run in two technical replicates and the average value reported. Because monozygotic twins share essentially all their genome while dizygotic twins share about half, comparing similarity within and between twin types lets the authors partition variance into additive genetic, shared environmental and unique environmental components.

Heritability estimates reach 67% for some effector functions

Heritability here means the share of total variance attributable to additive genetics, on a scale from 0% to 100%. The authors estimated it for every pathogen-specific Fc feature using linear mixed models, and used the difference in Akaike Information Criteria (ΔAIC) between models with and without a genetic parameter to judge whether the genetic term was justified.

Heritability appeared across both Fc characteristics and pathogen specificities. IgM levels had the largest number of heritable features with ΔAIC > 0, followed by IgG1 levels and Fc-receptor binding strength. Among individual features, Measles-IgG1, pertussis toxin (Ptox)-IgG1, VZV-IgG4 and Post-RSV FcγR2B binding showed the greatest heritability. Effector functions also carried genetic signal, with the highest estimates for PPD-ADNP at 60% and RSV-ADCP at 67%. Across all features, most were more than 20% heritable.

Two cautions belong with these numbers. Heritability is a variance share within this cohort, not a fixed property of a trait, and the cohort is entirely female, older and of white European descent, so the estimates describe this population rather than all humans.

Five genomic regions explain up to 33% of variance

The authors took the top 45 heritable phenotypes into a genome-wide association study (GWAS), which tests whether individual single-nucleotide polymorphisms (SNPs) track with a measured trait. Five genomic regions crossed genome-wide significance (p < 5×10⁻⁸ by likelihood ratio test).

The strongest signal was rs679574 in FUT2 on chromosome 19, associated with five Post-RSV traits: FcγR3A binding (33.42% of variance, p = 3.58×10⁻¹⁴), FcγR3B (32.77%), IgG1 (32.02%), FcγR2B (31.03%) and FcγR2A (30.36%). Four other lead variants each explained roughly 24–25% of variance in a single trait: rs62493093 near DLC1 with HBV-IgG2, rs10770123 near RNF141 with Polio-FcγR3A, rs1806595 near SPRY2 with Flu-ADNP, and rs10131070 in the TRA cluster with Mumps-IgG4.

The 33% figure is the variance explained by a single variant for a single trait, not the share of overall antibody variation explained by genetics. The paper's discussion separately states that genetics accounted for up to 67% of total variation, which is the heritability estimate rather than a GWAS variance-explained value; the two numbers answer different questions and should not be swapped.

FUT2 links secretor status to Post-RSV antibodies

FUT2 encodes an enzyme that attaches A and B histo-blood group antigens to proteins and lipids on mucosal surfaces or in secretions. The rs601338A variant, in complete linkage disequilibrium with the lead variant rs679574, is a nonsense variant that abolishes enzyme activity, producing the "non-secretor" status in which those antigens are absent from mucosal surfaces.

The authors note that non-secretor status has been associated with resistance to norovirus GII.4 and to several respiratory pathogens including RSV. Because RSV entry depends on attachment through the G glycoprotein followed by fusion via the F glycoprotein, they suggest that loss of histo-blood group antigen might alter fusion dynamics and prolong exposure of the post-fusion conformation, indirectly shaping the humoral response. This is offered as a possible indirect mechanism, not a demonstrated causal chain.

The same FUT2 variants also showed nominal associations with Ptox-IgG1 and Fc-receptor binding, Pneumo-ADCP, and Measles- and VZV-IgM, but with opposite directions across variants. The authors interpret this as evidence that FUT2 may affect humoral function broadly while acting in a pathogen-specific direction, possibly through susceptibility-linked indirect routes.

Cytokine signaling and myeloid antigen presentation emerge

To move from SNPs to plausible biology, the authors ran gene set analysis on SNPs nominally associated (p < 0.05) with the 45 heritable traits, querying immune system processes in the Gene Ontology database with MAGMA. Multiple gene sets were significantly associated with 16 heritable features, with the top processes involving B cell, T cell and innate immune function.

Specific links included PPD- and EBV-specific ADNP with lymph node development and germinal center formation, and Measles-specific IgG1 and FcγR2A/2B binding with T cell proliferation and differentiation. Notably, PPD-ADNP and Measles-IgG1 had no SNP reaching genome-wide significance, yet genes mapped to weakly associated SNPs were still enriched in significant immune processes.

Single-cell transcriptomic analysis added a cellular layer. DLC1, RNF141 and SPRY2 were robustly expressed in myeloid cells — DLC1 mainly in common myeloid progenitors, RNF141 across the myeloid lineage, and SPRY2 restricted to common myeloid progenitors and myeloid cells. TRAV8-2 in the TRA cluster appeared in a subset of regulatory T cells, though only in four donors, and FUT2 showed no expression in any immune cell type. Co-expression networks around DLC1 in common myeloid progenitors pointed to antigen processing and presentation via MHC class II, while SPRY2 networks involved cotranslational protein targeting to membrane.

Antibody features coordinate by antigen and by class

Before turning to genetics, the authors examined how antibody features move together. Pairwise Spearman correlations showed strong relationships within subclasses and isotypes, within Fc-receptor binding, and between IgG1 levels and Fc-receptor binding, and these cross-antigen relationships appeared in both monozygotic and dizygotic twins.

Coordination strength differed by feature: IgM responses were the most coordinated and FcγR3B, FcγR2A and FcγR3A binding the least. From the antigen side, Mumps-specific responses were the most coordinated and Polio-specific responses the least. Hierarchical clustering produced 20 distinct clusters falling into two larger groups: one organized largely by antigen specificity, spanning RSV, Betv1, Arah2, VZV, Noro, Polio, Measles and Mumps responses, and another organized by Fc characteristic, with separate IgM, IgA, IgG2, IgG3 and IgG4 clusters.

This structure matters for interpretation because it shows that an individual with strong IgG to a given target also tends to show strong Fc-receptor binding and function against that target. The heritability and GWAS results therefore describe features that are not statistically independent of one another.

Genetic association is not yet causal mechanism

The five regions are statistical associations between SNPs and humoral traits in one cohort. The pathway and co-expression analyses identify plausible immune processes and cell types, but the paper presents them as potential regulators rather than demonstrated causal pathways. The authors' own framing is that these data highlight potential genetic mechanisms governing Fc biology.

Several design features bound the conclusions. The cohort is 510 female twins, all self-reported white European descent, aged 41 to 78, so the allele frequencies and heritability estimates are population-specific; principal component analysis was used to check for stratification. The GWAS was restricted to the top 45 heritable phenotypes rather than all 227 profiles, so traits outside that set were not tested for genome-wide association. The transcriptomic co-expression work draws on an external single-cell dataset, and the TRAV8-2 observation rests on four donors.

FUT2 and Post-RSV is the signal to watch

If one result deserves priority, it is the FUT2–Post-RSV axis. It is the only region in the study associated with five traits at once, it carries the largest variance explained for any single trait (33.42% for FcγR3A binding), and it connects to a known functional variant, rs601338A, that abolishes enzyme activity and defines non-secretor status. That combination of statistical strength and a mechanistic candidate distinguishes it from the four single-trait signals near DLC1, RNF141, SPRY2 and the TRA cluster.

The practical reading is a split by trait type. Fc-receptor binding and effector functions such as RSV-ADCP (67%) and PPD-ADNP (60%) show the highest heritability, while the GWAS hits concentrate on subclass, isotype and Fc-receptor binding measures for specific pathogens. Readers comparing antibody traits should therefore ask which layer a number describes — variance explained by one SNP, heritability of a trait, or coordination between features — because the paper's headline figures of 33% and 67% come from different analyses and support different claims.

Sources used in this answer