Gene Variants Tied to Vitamin D Reshape Gut Microbes in Postmenopausal Women

Common variants in genes that govern vitamin D metabolism are linked to measurable differences in gut microbial diversity among postmenopausal women, according to a study published in Genes & Nutrition. Researchers analyzed 466 postmenopausal women from the Mexican Health Workers Cohort Study and found that three variants—one in the vitamin D receptor gene VDR and two in NADSYN1, a gene involved in NAD biosynthesis—were associated with lower alpha diversity and shifts in the abundance of specific bacterial genera. The work, led by Rafael Velázquez-Cruz and colleagues, adds to evidence that host genetics help shape the microbial communities living in the human intestine.

The team genotyped nine single-nucleotide variants spanning genes involved in vitamin D precursor biosynthesis (NADSYN1), transport (GC), hydroxylation and catabolism (CYP2R1, CYP27B1, and CYP24A1), and signaling through the vitamin D receptor (VDR). Gut microbiota were profiled by sequencing the V4 hypervariable region of the 16S rRNA gene, with samples rarefied to 10,589 reads to account for uneven sequencing depth. Diversity was quantified using observed amplicon sequence variant counts, Chao1 richness, the Shannon index, and the Simpson index, and associations were tested under a dominant inheritance model with adjustment for age, energy intake, physical activity, alcohol consumption, hormone replacement therapy, vitamin D intake, blood collection season, and sequencing kit.

Three variants stood out. Carriers of the C allele of rs4516035 in VDR showed lower observed species, Chao1, and Shannon indices than TT homozygotes, although these associations were suggestive after false discovery rate correction (PFDR = 0.082). The A allele of rs4944957 in NADSYN1 was suggestively linked to lower Shannon diversity (PFDR = 0.0527) and significantly associated with Simpson index values (PFDR = 0.0489). Carriers of the C allele of rs3794060, also in NADSYN1, showed significantly lower Shannon (PFDR = 0.0131) and Simpson (PFDR = 0.0038) diversity than TT homozygotes. Quantile regression analyses across the 25th, 50th, and 75th percentiles of each diversity index yielded no significant results after correction, indicating that the genotype effects were not confined to particular parts of the diversity distribution.

Beyond diversity metrics, differential abundance analysis identified taxonomic disparities tied to genotype. Carriers of at least one risk allele in NADSYN1 exhibited significantly higher abundance of the genera Negativibacillus and DTU014 than those with the ancestral allele. Fusobacterium, Erysipelatoclostridium, Acidaminococcus, and the [Clostridium] innocuum group were significantly enriched in carriers of the VDR rs4516035 C-risk allele. When alleles were grouped by genotype, increased abundance of Akkermansia, Adlercreutzia, Pseudomonas, and Anaerotruncus appeared in the CC genotype of rs4516035, while the AA genotype of rs4944957 showed enrichment for DTU014 and the CC genotype of rs3794060 was associated with increased Peptococcus and the Treponema group. These analyses used linear discriminant analysis effect size with a threshold LDA score of at least 1.5 and Benjamini–Hochberg correction at FDR below 0.1.

The researchers also probed predicted microbial function using PICRUSt2, which infers metabolic pathways from 16S rRNA data through KEGG orthology. Carriers of the rs4516035-C allele showed significant enrichment in pathways including galactose metabolism and the phosphotransferase system, while pathways related to thiamine metabolism, terpenoid backbone biosynthesis, and histidine and phenylalanine/tyrosine/tryptophan biosynthesis were overrepresented in TT homozygotes. The rs4588-T allele in the GC gene was associated with enrichment in histidine metabolism and secondary bile acid biosynthesis, alongside depletion in D-alanine, glutathione, D-glutamine, D-glutamate, and pyrimidine metabolism. The NADSYN1 risk alleles were enriched in the citrate cycle, oxidative phosphorylation, folate biosynthesis, pentose and glucuronate interconversions, and biotin metabolism. The authors emphasize that these predictions are exploratory and require validation through metatranscriptomics, metabolomics, or targeted biochemical assays.

To test the combined effect of the three diversity-associated variants, the team constructed an unweighted genetic risk score summing risk alleles at rs4516035, rs4944957, and rs3794060. Among women aged 47 years or older, carriers of more than three risk alleles showed lower diversity than those carrying three or fewer, with significant differences after correction for the Observed (PFDR = 0.042), Shannon (PFDR = 0.0384), and Simpson (PFDR = 0.0384) indices and a suggestive association for Chao1 (PFDR = 0.064). Five genera—Weissella, Negativibacillus, Victivallaceae, Peptococcus, and Paeniclostridium—were significantly enriched in high-score carriers. Exploratory functional analyses identified significant enrichment of the insulin signaling pathway (q = 0.031) and suggestive enrichment of pentose and glucuronate interconversions (q = 0.069) and oxidative phosphorylation (q = 0.075). MaAsLin3 modeling linked the score to the prevalence of Victivallaceae (OR = 1.07, PFDR = 0.044). The authors caution that overfitting risk from variant selection within the same dataset makes these findings hypothesis-generating rather than conclusive.

Notably, serum 25-hydroxyvitamin D concentrations themselves did not differ by microbial diversity status. No significant differences in alpha diversity were observed between women with vitamin D deficiency, defined as serum 25(OH)D below 20 ng/mL, and non-deficient participants in the overall population. The sex-stratified analysis also failed to replicate previously reported associations between these variants and circulating vitamin D, possibly reflecting reduced statistical power after stratification. Two variants, rs6013897 in CYP24A1 and rs4588 in GC, showed only suggestive associations with increased odds of vitamin D deficiency (OR = 1.82 and OR = 1.70, both PFDR = 0.077). The authors suggest that the relationship between vitamin D and gut microbiota may involve distinct or indirect mechanisms, with host genetic variation acting on microbial ecology independently of measured serum levels.

The study population consisted of employees of the Instituto Mexicano del Seguro Social and their relatives in urban central Mexico, recruited between 2017 and 2019. Individuals with malnutrition, renal damage, or vitamin D supplementation were excluded because these conditions can compromise vitamin D metabolism. The postmenopausal subsample had a median age of 60.0 years and a median BMI of 27.2 kg/m², with 42.3% classified as overweight and 28.5% as living with obesity. Among ten factors significantly associated with inter-individual microbial distance, BMI, triglycerides, and the variants rs4516035 and rs4588 showed suggestive associations with microbiota composition at the genus level, together explaining 4.86% of the variance in microbial composition.

The authors frame the findings within a growing literature on host genetics and the gut microbiome. A genome-wide association study in two northern German cohorts identified significant associations between microbial variation and the VDR gene, and analyses in Caucasian Spanish individuals found differences in alpha diversity across VDR genotypes. Prior work has also shown that VDR knockout mice are more susceptible to Clostridium rodentium infection, implicating VDR expression in innate lymphoid cell regulation and interleukin-22 production. In the present cohort, however, no correlation was observed between the vitamin D-related variants and circulating vitamin D concentrations, and the functional consequences of the lead variants remain unknown.

Several limitations temper the conclusions. The cross-sectional design cannot establish causality or the direction of the associations, 16S rRNA profiling provides taxonomic resolution coarser than shotgun metagenomic sequencing, and PICRUSt2 predictions are inherently exploratory. Different DNA extraction kits may introduce technical variability, the BMI–vitamin D connection was not directly examined, and residual false-positive risk cannot be excluded given the large number of comparisons. The authors note that postmenopausal women with specific gene variants could potentially benefit from targeted probiotic or dietary interventions, but direct functional profiling studies are needed to determine whether the inferred pathway differences translate into altered microbial metabolic activity or host physiological outcomes. Further research in larger independent cohorts is essential to clarify the mechanisms by which VDR and NADSYN1 variants influence gut microbial composition and function.

Subject of Research: Associations between vitamin D metabolism gene variants and gut microbiota diversity in postmenopausal women

Article Title: Common genetic variants in vitamin D metabolism-related genes affect gut microbiota diversity in postmenopausal women

Article References: Becerra-Cervera, A., Jiménez-Ortega, R. F., Aparicio-Bautista, D. I., López-Pérez, T. V., Denova-Gutiérrez, E., Palacios-González, B., Hidalgo-Bravo, A., Canizales-Quinteros, S., Salmerón, J., Rivera-Paredez, B., & Velázquez-Cruz, R. (2026). Common genetic variants in vitamin D metabolism-related genes affect gut microbiota diversity in postmenopausal women. Genes & Nutrition, 21(1), Article 17. https://doi.org/10.1186/s12263-026-00806-y

Image Credits: AI Generated

DOI: 10.1186/s12263-026-00806-y

Keywords: vitamin D, gut microbiota, postmenopausal women, VDR, NADSYN1, genetic variants, alpha diversity, 16S rRNA sequencing, PICRUSt2, genetic risk score, Health Workers Cohort Study, microbiome