WNT4

WNT4 rs7521902 — When a Signaling Gene Shapes Endometriosis Risk

WNT4 encodes one of the Wnt family of secreted signaling proteins, a group
essential for embryonic development of the female reproductive tract and for
the monthly remodeling of the uterine lining. The protein coordinates the
formation of Müllerian duct structures — the precursors to the uterus,
fallopian tubes, and cervix — and continues to regulate endometrial stromal
cell behavior throughout reproductive life. A common variant approximately
21 kilobases downstream of the WNT4 gene, rs7521902, has emerged from
multiple large genome-wide association studies as one of the most replicated
genetic risk factors for endometriosis, particularly moderate-to-severe disease.

The Mechanism

rs7521902 sits within an intronic region of an uncharacterized neighboring
locus (LOC105376850) in the 1p36.12 region and does not directly alter the
WNT4 protein sequence. Its functional effect is regulatory: variants in tight
linkage disequilibrium | LD: the tendency for nearby genetic variants to be
inherited together with rs7521902
— particularly rs3820282, located in WNT4 intron 1 — have been shown to
introduce a high-affinity estrogen receptor alpha binding site | ERE: an
estrogen response element, a DNA sequence where the estrogen receptor attaches
to regulate gene transcription.
The result is upregulated WNT4 transcription in endometrial stromal cells
following the preovulatory estrogen peak, with a 1.5–3.3 fold increase in
mouse transgenic models.

This elevated WNT4 expression in stromal cells activates
non-canonical Wnt signaling | Wnt pathways that do not proceed through
β-catenin; involved in cell polarity and invasive behavior
and produces a uterine environment that is more permissive to cellular
invasion. The same mechanism that may improve embryo implantation — by
increasing stromal receptivity — appears to simultaneously increase the
permissiveness of the endometrium to invasion by ectopic endometriotic tissue.
This antagonistic pleiotropy may explain why the risk allele has been maintained
at appreciable frequency despite its disease association.

WNT4 also acts downstream of BMP2 to regulate
decidualization | the monthly transformation of endometrial stromal cells
into specialized secretory cells in preparation for embryo
implantation, a process frequently
disrupted in endometriosis. The IHH–COUPTFII–WNT4 pathway coordinates
progesterone response in the endometrium; its disruption contributes to the
progesterone resistance characteristic of endometriotic tissue. In uterine
fibroids, MED12 mutations — present in the majority of fibroids — directly
upregulate WNT4 expression, driving cell proliferation through β-catenin
signaling and mTOR activation.

The Evidence

The first genome-wide significant association between rs7521902 and
endometriosis was reported in a
combined Japanese–European meta-analysis | Nyholt et al. 2012, Nature
Genetics (P=4.2×10⁻⁸,
OR=1.19, 95% CI 1.12–1.27). A subsequent
meta-analysis of eight GWAS datasets | Rahmioglu et al. 2014, Human
Reproduction Update; PMC4132588
in 11,506 cases and 32,678 controls confirmed the association at
P=1.8×10⁻¹⁵ (OR=1.18, 95% CI 1.13–1.23). Critically, the effect size
strengthened for stage III/IV disease (OR=1.23, 95% CI 1.17–1.28,
P=2.7×10⁻¹⁷), indicating the variant is particularly relevant to moderate
and severe endometriosis, the phenotypes most likely to cause chronic pain
and fertility impairment. Eight of nine genome-wide significant loci in
that meta-analysis showed consistently stronger effects in stage III/IV
subgroup analyses.

An Italian replication study | Pagliardini et al. J Med Genet
2013 of 305 surgically confirmed
cases and 2,710 controls confirmed the association (P=5.6×10⁻³) and
identified an epistatic interaction between rs7521902 and rs1250248
(OR=1.56, P=0.012). A
2024 systematic review and meta-analysis | PMID 38354602
of 10 case-control studies found the CC (homozygous reference) genotype
protective: pooled OR=0.86 (95% CI 0.76–0.99). Not all populations
replicate the association: a Brazilian cohort of infertile women | Mafra et al.
J Assist Reprod Genet 2015 found
no significant association for rs7521902 (p=0.18), though rs16826658 was
significant in that cohort (OR=1.44). A Chinese Han cohort similarly did not
replicate rs7521902 but found rs2235529 significant for advanced disease. These
population-specific results likely reflect differences in linkage disequilibrium
structure around the WNT4 locus rather than absence of effect.

The WNT4 locus also shows pleiotropic association with
uterine leiomyomas | fibroids; benign smooth muscle tumors of the
uterus (OR=1.12–1.19), bone
mineral density, and pelvic organ prolapse — consistent with WNT4's
broad role in gynecological tissue maintenance.

Practical Actions

For women carrying one or two copies of the A allele, the most actionable
implication is heightened awareness of endometriosis symptoms: cyclic pelvic
pain, deep dyspareunia, dysmenorrhea, and unexplained infertility. Earlier
investigation via gynecological ultrasound (for ovarian endometriomas and
fibroids) or laparoscopy is appropriate when symptoms are present rather
than waiting for symptoms to become severe. Symptom severity does not
reliably correlate with disease stage, so evaluation should not depend on
pain intensity alone.

No evidence supports a supplement or dietary intervention that specifically
modifies WNT4 signaling in the endometrium. Progestin-based hormonal
therapies are the mainstay of endometriosis management and address the
progesterone-resistance pathway through which WNT4 dysregulation is thought
to act — this is a medical decision to be made with a gynecologist.

Interactions

The WNT4 locus displays an epistatic interaction with rs1250248 (OR=1.56
for the interaction term; Pagliardini et al. 2013, PMID 23142796). The
functional variant rs3820282, in strong LD with rs7521902 in European
populations, appears to be a more direct molecular mediator of WNT4
expression change, and warrants inclusion in any compound analysis.
rs16826658, also in the WNT4 region, was independently associated with
endometriosis in some populations (OR=1.44 in the Brazilian cohort). Given
that these variants all tag the same ~100–150 kb haplotype block on 1p36.12,
users carrying risk alleles at multiple WNT4-region SNPs may reflect
greater cumulative haplotype risk, though formal compound action data for
this specific combination are not yet established in the literature.