Sergio Andreu-Sánchez, Arno R. Bourgonje, Thomas Vogl, Alexander Kurilshikov, Sigal Leviatan, Angel J. Ruiz Moreno, Shixian Hu, Trishla Sinha, Arnau Vich Vila, Shelley Klompus, Iris N. Kalka, Karina de Leeuw, Suzanne Arends, Iris Jonkers, Sebo Withoff, Elisabeth Brouwer, Adina Weinberger, Cisca Wijmenga, Eran Segal, Rinse K. Weersma, Jingyuan Fu, Alexandra Zhernakova
10 min
How do genetic, environmental, and intrinsic factors influence the human adaptive immune system's antibody epitope repertoire? While the adaptive immune system is known for its versatility, the specific drivers of antibody responses against diverse microbial and environmental antigens in the general population remain poorly understood.
Researchers performed high-throughput profiling of antibody repertoires in 1,443 individuals from the Lifelines-DEEP population cohort using Phage-displayed Immunoprecipitation Sequencing (PhIP-seq). This method allowed the screening of 344,000 peptides from various sources, including human gut microbiota, pathogens, environmental allergens, and bacteriophages. The team integrated this immune data with genome-wide genotype data, gut microbiome sequencing, and extensive metadata, including clinical blood parameters, family information, and self-reported lifestyle and disease questionnaires. They also performed computational structural modeling to validate the functional role of identified HLA genetic associations.
The researchers found that human antibody repertoires are highly personalized and show temporal consistency over time. The study identified three major genetic hotspots affecting peptide reactivity: the HLA region (particularly class II alleles), the IGHV immunoglobulin heavy chain variable region, and the FUT2 gene. The HLA associations were linked to specific peptide presentation mechanisms, supported by structural modeling showing how particular HLA alleles bind peptide cores. Additionally, the study discovered hundreds of associations between antibody responses and phenotypic factors, such as age, smoking, and blood cell counts. For instance, smoking and age were linked to distinct profiles of viral and bacterial antibody reactivity. Conversely, the study found no significant direct association between fecal microbial taxonomic abundance and the detected antibody-bound peptide repertoires, suggesting that these antibody responses might reflect past microbial exposure or local mucosal interactions rather than current fecal microbial composition.
Phage-displayed immunoprecipitation sequencing (PhIP-seq) has enabled high-throughput profiling of human antibody repertoires. However, a comprehensive overview of environmental and genetic determinants shaping human adaptive immunity is lacking. In this study, we investigated the effects of genetic, environmental, and intrinsic factors on the variation in human antibody repertoires. We characterized serological antibody repertoires against 344,000 peptides using PhIP-seq libraries from a wide range of microbial and environmental antigens in 1,443 participants from a population cohort. We detected individual-specificity, temporal consistency, and co-housing similarities in antibody repertoires. Genetic analyses showed the involvement of the HLA, IGHV, and FUT2 gene regions in antibody-bound peptide reactivity. Furthermore, we uncovered associations between phenotypic factors (including age, cell counts, sex, smoking behavior, and allergies, among others) and particular antibody-bound peptides. Our results indicate that human antibody epitope repertoires are shaped by both genetics and environmental exposures and highlight specific signatures of distinct phenotypes and genotypes.
Alex: The main analyses used about 1,400 participants. The cohort also supplied genetic data, clinical blood measurements, lifestyle questionnaires, family information, and stool microbiome sequencing. That combination let the authors compare antibody patterns with many possible influences.
Sam: What made these patterns personal? Could a few common infections explain most of the variation?
Alex: Most bound peptides were uncommon across participants. The variation was spread widely across the profiles, rather than concentrated in a few features. Cytomegalovirus, a herpesvirus, did produce a clear division between groups, but it didn't account for the broader diversity.
Sam: A personal pattern at one visit could still be transient. Did they test whether people remained similar to themselves?
Alex: They compared repeat samples from 322 participants. The visits were four years apart. On average, a person's follow-up profile was closer to their own earlier profile than to unrelated people's profiles.
Sam: Stable doesn't mean frozen, though. What changed between visits?
Alex: Most peptide-binding results stayed consistent, while a subset changed. Loss of detectable binding was more common than gain. The authors suggest time since antigen exposure could affect detection, but these data don't establish that explanation.
Sam: Families share genes and experiences. Could the family comparisons separate those contributions?
Alex: Only partly. In 26 family trios, parent–offspring pairs and mother–father pairs were more similar than unrelated pairs. Similarity between the parents supports a shared-environment contribution, while parent–offspring similarity could reflect genetics, early-life environment, or shared lifestyle.
Sam: So family resemblance points toward several explanations, not a clean genetic estimate. What did the direct genetic analysis add?
Alex: Across the selected peptides, common genetic variation explained about ten percent of binding variability on average. That average concealed wide differences between peptides. The authors also caution that their genetic sample was too small for accurate estimates of heritability and genetic correlations.
Sam: An average contribution doesn't tell us where the biology sits. Did particular genetic regions stand out?
Alex: The genetic association analysis highlighted HLA, IGHV, and FUT2. HLA means human leukocyte antigen: proteins that display peptides to immune cells. That region gave the strongest signal, especially variants involved in presenting peptides to helper T cells.
Sam: What made that more than a familiar association between immune genes and immune responses?
Alex: More detailed HLA analysis linked variants to over 500 peptides. Earlier work had largely examined viruses, toxins, and virulence factors. This study extended the genetic map to allergens and proteins from gut microbes, alongside those other targets.
Sam: Can you give one example where the genetic association connects to something physically plausible?
Alex: A peptide from streptococcal C5a peptidase, a bacterial enzyme, was associated with particular HLA variants. The authors modeled a short core of that peptide inside the associated HLA complexes. Their calculations predicted more favorable binding than for a non-associated HLA complex used as a negative comparison.
Sam: That supports peptide presentation as an explanation, but the binding happened in a computer model. Did they experimentally validate that mechanism here?
Alex: The evidence presented here is computational, not experimental binding validation. It supports the hypothesis that different HLA proteins preferentially display particular peptides. It doesn't prove the complete pathway from inherited variant to the measured antibody response.
Sam: And the other genetic regions could act at different steps. What did those associations suggest?
Alex: IGHV is the immunoglobulin heavy-chain variable region, which helps encode antibody recognition. Its association with antibody profiles was new according to the authors, but difficult to interpret in this complex region. FUT2 controls secretion of a sugar structure called the H-antigen.
Sam: Why would secretion of a sugar structure show up in an antibody-binding study?
Alex: The FUT2 associations involved Norwalk virus peptides. That virus requires a particular H-antigen structure to attach successfully to cells. The finding supports a previously reported connection between secretor status and susceptibility, rather than establishing a new mechanism.
Sam: There's another possible route: current gut microbes might determine the antibodies against bacterial peptides. Did the stool data support that?
Alex: The authors found no strongly supported associations between fecal bacterial abundance and peptide-binding profiles, even after combining cohorts. That doesn't establish that gut microbes are irrelevant. Antibodies may reflect past encounters, and stool samples may miss microbes attached to the intestinal lining.
Sam: So a lasting blood response needn't match a microbial snapshot. What did lifestyle and clinical measurements contribute?
Alex: Age, blood-cell counts, smoking, sex, and allergies were associated with binding patterns. Older participants more often recognized several herpesvirus peptides, while some other viral responses were more frequent in younger people. Smokers showed higher prevalence of several respiratory-virus peptide responses.
Sam: Could related peptides make that look like exposure to several viruses when antibodies recognize shared sequences?
Alex: The authors raise that possibility for smoking-associated enterovirus responses. They also found groups of related peptides that tended to be recognized together, consistent with cross-reactivity. Some unrelated peptides clustered too, but those groupings were harder to interpret.
Sam: Allergy associations sound especially tempting to interpret clinically. Is this measuring the antibody pathway that usually triggers allergy?
Alex: Not directly. Allergy usually involves IgE, a different antibody class, while this assay mainly captured IgG. Dust allergy was associated with house-dust-mite peptides, but the authors explicitly caution that these associations may differ from the classical allergy pathway.
Sam: Given those boundaries, who should read the full document, and where should they start?
Alex: Researchers studying antibody profiling, immune genetics, microbiome immunity, or allergy should read it. Start with the genetic-results section and its HLA modeling comparison, then read Study limitations. The mostly cross-sectional design makes causal interpretation difficult, and the assay misses three-dimensional recognition sites and protein modifications.
Sam: And what should everyone else carry away, without turning these associations into explanations?
Alex: Your antibody profile is a lasting but selective record of recognition, shaped by both inheritance and experience. It isn't a census of today's microbes, or proof of what caused disease.
Sam: Keep that distinction in mind when reading an immune fingerprint.