rs244715 — ZNF346

ZNF346 rs244715 — The DNA-Repair Locus and Your Reproductive Clock

The chromosome 5q35.2 region was one of four genome-wide significant loci first identified in the landmark 2009 GWAS of 17,438 women¹¹ GWAS of 17,438 women
He et al. 2009, Nature Genetics — the first genome-wide scan for menopause-timing variants. At the center of this locus sits UIMC1 (also called RAP80) — a BRCA1 adaptor protein — and nearby, approximately 71 kb downstream, ZNF346 (zinc finger protein 346, also known as JAZ). The intronic variant rs244715 in ZNF346 is a proxy SNP for this locus, in partial linkage disequilibrium with the index UIMC1 coding variant rs365132 (r²=0.677 in European populations). Each G allele at rs244715 may be associated with a modest shift toward earlier menopause onset.

The Mechanism

The 5q35.2 locus is anchored biologically by UIMC1 (ubiquitin interaction motif containing 1), a protein that physically interacts with BRCA1 and is required to recruit the BRCA1-A complex to sites of DNA double-strand breaks. This recruitment initiates G2/M checkpoint control²² G2/M checkpoint control
The G2/M checkpoint prevents cells with damaged DNA from dividing; failure of this checkpoint accelerates DNA damage accumulation in dividing cells and homologous recombination repair. Because oocytes are some of the most DNA-damage-sensitive cells in the body — they can arrest in meiosis for decades and are exposed to oxidative stress throughout a woman's reproductive life — efficient DNA repair is critical for maintaining follicular integrity. Impaired repair capacity at this locus may accelerate the rate at which follicles accumulate irreparable DNA damage, triggering apoptosis and depleting the ovarian reserve prematurely.

ZNF346 itself adds a plausible second layer. As a nucleolar zinc finger protein (also called JAZ — "just another zinc finger protein"), it preferentially binds double-stranded RNA rather than DNA, and has been shown to positively regulate p53 transcriptional activity³³ positively regulate p53 transcriptional activity
p53 is a master regulator of cellular stress responses; when p53 is activated, it can trigger cell-cycle arrest or apoptosis depending on context, mediating G1 cell-cycle arrest and apoptosis. ZNF346 is expressed in ovarian tissue (11.9 nTPM, Human Protein Atlas) and at notably higher levels in granulosa cells (21.1 nCPM), the somatic cells that nurse follicles and regulate their fate. Altered ZNF346 expression — potentially driven by the eQTL effects of the 5q35.2 haplotype — may modulate granulosa cell survival and follicular apoptosis thresholds. Whether rs244715 itself directly alters ZNF346 expression or function in the ovary remains under investigation.

Separately, a 2021 study found that rs244715 was associated with anti-Hsp27 antibody titers⁴⁴ anti-Hsp27 antibody titers
Heat shock protein 27 (Hsp27) is a stress-response chaperone; elevated anti-Hsp27 antibodies are a marker of immune activation and oxidative stress in premature menopause cases, suggesting a possible inflammatory or oxidative-stress component to this variant's effect on ovarian aging.

The Evidence

The 5q35.2 locus has been replicated in multiple large-scale studies. The He et al. 2009 discovery GWAS⁵⁵ He et al. 2009 discovery GWAS
n=17,438 women, European ancestry; four menopause loci identified at p<1×10⁻⁷ was confirmed by the Stolk et al. 2012 ReproGen meta-analysis⁶⁶ Stolk et al. 2012 ReproGen meta-analysis
n=38,968 discovery + 14,435 replication, 22 independent European-ancestry cohorts, which identified the 5q35.2 locus as one of 13 independent menopause-timing signals. The index variant rs365132 in UIMC1 had a beta of approximately 0.29 years (~15 weeks) per minor allele in the original discovery cohort.

The Breakthrough Generations Study⁷⁷ Breakthrough Generations Study
Murray et al. 2011, Hum Mol Genet, n=2,007 women including 694 early menopause cases; UK prospective cohort directly evaluated rs244715 and found the G allele yields an OR of 1.20 (95% CI 1.09–1.32, p=1.7×10⁻⁴) for early menopause (defined as menopause before age 46). Women homozygous for risk alleles across all four 2009 GWAS loci (including rs244715 GG) had approximately 4-fold higher odds of early menopause than women with three or fewer risk alleles. The per-allele effect on age at menopause in this study was 0.059 years, though this did not reach significance in the smaller quantitative trait analysis.

The Mashhad premature ovarian insufficiency cohort⁸⁸ Mashhad premature ovarian insufficiency cohort
Ziaee et al. 2021, 117 POI cases vs. 183 controls; Iranian women found allelic association of rs244715 G with POI (OR 1.71, 95% CI 1.17–2.50, p=0.005). The homozygote contrast (GG vs. AA) had an OR of 3.93 (95% CI 1.40–11.00, p=0.009) — a large effect consistent with an additive architecture — though none of these associations survived Bonferroni correction for the eight SNPs studied. The 5q35.2 locus was further extended in the Day et al. 2015 expanded GWAS⁹⁹ Day et al. 2015 expanded GWAS
n~70,000 European women; 54 independent signals in 44 genomic regions identified, confirming this as a durable, replicated signal.

Multi-ethnic replication is incomplete. The PAGE study¹⁰¹⁰ PAGE study
Carty et al. 2013, multi-ethnic US cohorts found rs365132 (the UIMC1 index SNP) significantly replicated in non-European populations, though the LD structure at the locus varies substantially across ancestries, meaning rs244715 may be a less reliable proxy in non-European groups.

Practical Implications

With an estimated per-allele shift of roughly 2–3 weeks in menopause timing, rs244715 has a modest individual effect. Its clinical utility lies in contributing to a polygenic burden score for ovarian aging — when combined with other replicated loci (MCM8 rs16991615, PRRC2A rs1046089, FNDC4 rs2303369), the cumulative genetic signal becomes more predictive of early reproductive aging than any single variant.

For women with the GG genotype who are considering when to start their family or whether to investigate fertility preservation, this variant is most informative when interpreted alongside AMH (anti-Müllerian hormone) levels — the most sensitive biomarker of remaining ovarian reserve — and antral follicle count on ultrasound.

The DNA repair context of this locus also raises a plausible lifestyle modulator: oxidative stress is a key driver of follicular DNA damage, and the 5q35.2 locus mechanism suggests that factors increasing cellular oxidative load (smoking, chronic inflammation) may interact with reduced UIMC1/ZNF346 pathway efficiency. However, there is currently no published evidence directly testing this gene-environment interaction.

Interactions

UIMC1 rs365132: The biological index SNP at this locus. rs244715 and rs365132 are in partial LD (r²=0.677 in Europeans), meaning they are highly correlated but not perfectly interchangeable. rs365132 (a synonymous UIMC1 coding variant) is the more functionally annotated variant and may be a better proxy for the biological effect at this locus. If a user has both, interpret the more significant association; they should not be treated as independent signals.

MCM8 rs16991615 (E341K): The strongest-effect DNA repair locus for menopause timing (~1 year per allele). Both MCM8 and the UIMC1/ZNF346 locus operate through overlapping DNA repair pathways (MCM8: replication fork restart; UIMC1: BRCA1-mediated DSB repair). A woman carrying risk alleles at both loci has two independent DNA-repair hits on reproductive lifespan. The combined signal may warrant earlier AMH baseline assessment.

A proposed compound action: Women who carry GG at rs244715 and carry the risk genotype at MCM8 rs16991615 (GG, associated with earlier menopause/lower AMH) represent an additive polygenic burden from two independent DNA-repair loci. The combined recommendation would be to obtain a baseline AMH panel before age 30, with repeat testing at 2-year intervals, and to discuss reproductive timeline planning with a reproductive endocrinologist if AMH is trending below age-expected norms.

PRRC2A rs1046089: A distinct menopause-timing locus on chromosome 6 operating through an immune/HLA pathway rather than DNA repair. rs244715 and rs1046089 are on different chromosomes and have no LD relationship — they are independent signals. Carrying risk alleles at both would represent additive contributions from two biologically distinct pathways (DNA repair and immune-mediated follicle depletion) to earlier ovarian aging.