rs2303369 — FNDC4

FNDC4 and Ovarian Aging — A Follicle Signaling Variant

Your body's reproductive clock is partly set by genes long before you are born. Among the loci robustly linked to the timing of natural menopause and the risk of primary ovarian insufficiency (POI) is rs2303369, an intronic variant in the FNDC4¹¹ FNDC4
Fibronectin Type III Domain Containing 4 — a secreted protein structurally related to the exercise myokine irisin gene on chromosome 2. The T allele at this position has been associated with slightly earlier menopause onset across large population studies, and the homozygous TT genotype has shown elevated odds of premature ovarian insufficiency in clinical cohorts.

The Mechanism

FNDC4 encodes a protein belonging to the fibronectin type III domain-containing family — the same structural family as FNDC5/irisin, the well-studied exercise-released myokine. Like irisin, FNDC4's extracellular domain can be proteolytically cleaved and secreted as a circulating factor. Its primary receptor, ADGRF5 (also known as GPR116)²² ADGRF5 (also known as GPR116)
an adhesion G protein-coupled receptor expressed in adipose tissue and the ovary, is present in mouse ovarian tissue, pointing to a direct autocrine or paracrine role in follicular biology.

Recent in vitro and in vivo work has begun to clarify this role: Daudon et al. 2025³³ Daudon et al. 2025
Daudon M, et al. FNDC4 modulates in vitro bovine granulosa and theca cell metabolism and alters follicle development in vivo. Animal Reproduction Science, 2025 showed that FNDC4 increases glucose uptake in granulosa cells, decreases lipid content in theca cells, and — when applied directly to growing follicles in vivo — caused follicle regression, likely through reduced cellular metabolic output.

How exactly the intronic rs2303369 variant alters FNDC4 expression or splicing is not yet established — it is likely a regulatory or tagging variant in linkage disequilibrium with the functional change. The locus also shows pleiotropy⁴⁴ pleiotropy
when a single genetic variant influences multiple seemingly unrelated traits, with nearby variants correlated with kidney function, type 2 diabetes, serum triglycerides, and C-reactive protein, suggesting FNDC4 sits at a broader metabolic-reproductive intersection.

The Evidence

The strongest association evidence comes from Stolk et al. 2012⁵⁵ Stolk et al. 2012
Stolk L, et al. Meta-analyses identify 13 loci associated with age at menopause and highlight DNA repair and immune pathways. Nature Genetics, 2012, a meta-analysis of 22 GWAS in 38,968 European women with replication in up to 14,435 additional women. The FNDC4 locus reached genome-wide significance (p = 2 × 10⁻¹²), with each T allele associated with approximately 0.175 years (~2 months) earlier menopause onset. This effect was subsequently included in Day et al. 2015⁶⁶ Day et al. 2015
Day FR, et al. Large-scale genomic analyses link reproductive aging to hypothalamic signaling, breast cancer susceptibility and BRCA1-mediated DNA repair. Nature Genetics, 2015, which expanded the menopause GWAS to approximately 70,000 women of European ancestry and identified 54 independent menopause-timing signals.

Clinical cohort data from Mirinezhad et al. 2021⁷⁷ Mirinezhad et al. 2021
Mirinezhad MR, et al. Genetic Determinants of Premature Menopause in A Mashhad Population Cohort. Int J Fertil Steril, 2021 examined 117 women with premature menopause (before age 40) against 183 healthy controls. The TT genotype at rs2303369 was associated with an odds ratio of 2.40 (95% CI 1.13–5.10, p=0.020) for premature menopause under a recessive model. It is notable that this association did not survive Bonferroni correction in the small cohort, and population-level GWAS studies rather than case-control studies provide the more robust evidence base for this locus.

Effect sizes are modest: 0.175 years per allele in the largest meta-analysis. This locus accounts for a small fraction of the total genetic variance in menopause timing, which is estimated to be 50–70% heritable. The practical significance lies in accumulation across multiple loci — women carrying multiple menopause-timing risk variants at independent loci may have meaningfully earlier expected follicle depletion compared to the population average.

Practical Actions

For CC genotype carriers (no T alleles), this locus provides no signal toward earlier follicle depletion. This is the most common genotype globally (~37% of the population) and represents the baseline for this variant.

For CT and TT carriers, each T allele carries a modest signal toward somewhat earlier ovarian aging. The T allele is notably less common in East Asian populations (~13%) compared to European (~39%) and South Asian (~42%) populations, so the population-attributable risk differs substantially by ancestry.

Serum anti-Müllerian hormone (AMH)⁸⁸ anti-Müllerian hormone (AMH)
a hormone secreted by granulosa cells in small growing follicles; the most sensitive and cycle-independent measure of the remaining follicle pool testing provides the most clinically actionable measure of where ovarian reserve actually stands, regardless of genotype. This genetic signal is most useful as context for interpreting AMH results that fall at the lower end of age-specific reference ranges.

Interactions

FNDC4 rs2303369 + MCM8 rs16991615 (dual reproductive aging loci): MCM8 encodes a DNA repair helicase also strongly associated with menopause timing and AMH levels. Women lacking the protective MCM8 A allele (GG genotype) who also carry T alleles at rs2303369 carry independent risk signals from two separate biological pathways — DNA repair helicase insufficiency (MCM8) and follicular metabolic signaling (FNDC4 locus). Whether these effects are strictly additive or interact beyond additivity is not established in published literature, but the combined profile of multiple independently replicated menopause-timing risk alleles at unlinked loci represents a higher prior probability of earlier follicle depletion than either variant alone. A compound action for this combination — emphasizing proactive AMH baseline testing and early fertility timeline discussion — is warranted for women carrying risk alleles at both loci. See related SNP rs16991615.

Note on HELQ locus context: Early GWAS publications and some follow-up studies refer to this chromosomal region as the "HELQ locus" because HELQ (encoding a separate DNA repair helicase on chromosome 4) was among the DNA repair genes highlighted in the pathway analysis of the same 2012 study. The index SNP rs2303369 maps to FNDC4 on chromosome 2 in dbSNP, while the true HELQ index SNP is rs4693089 on chromosome 4. Both were identified in the same landmark GWAS.