Research

rs12870438 — EPSTI1

Intronic variant in the immune-response gene EPSTI1, associated in recessive models with reduced sperm concentration, total sperm count, sperm motility, and azoospermia/oligospermia risk in men.

Moderate Risk Factor Share

Details

Gene
EPSTI1
Chromosome
13
Risk allele
A
Consequence
Intronic
Inheritance
Autosomal Recessive
Clinical
Risk Factor
Evidence
Moderate
Chip coverage
v3 v4 v5

Population Frequency

GG
53%
AG
39%
AA
7%

Ancestry Frequencies

european
38%
latino
22%
south_asian
16%
east_asian
15%
african
8%

See your personal result for EPSTI1

Upload your DNA data to find out which genotype you carry and what it means for you.

Upload your DNA data

Works with 23andMe, AncestryDNA, and other DNA test exports. Results in under 60 seconds.

EPSTI1 rs12870438 — When Testicular Immunity Affects Sperm Production

The testes occupy a paradoxical position in the immune system: they must shield developing sperm — which display "foreign" surface antigens not seen by the immune system before puberty — from attack, while still defending against microbial infection. EPSTI1 (epithelial stromal interaction 1) is an interferon-response gene11 interferon-response gene
a gene whose expression is rapidly induced when the body detects viral or bacterial pathogens
that may help maintain this delicate testicular immune balance. An intronic variant, rs12870438, has emerged as a candidate genetic influence on sperm production in multiple studies.

The Mechanism

EPSTI1 protein is expressed broadly across immune tissues — spleen, lymph nodes, macrophages — and is also found in the testes, with particularly elevated expression in late primary spermatocytes and early spermatids22 late primary spermatocytes and early spermatids
the germ cells undergoing meiosis and initial sperm formation
(111 and 77 nCPM respectively, versus a testis-wide average of 6.9 nTPM). The protein interacts with valosin-containing protein to promote nuclear translocation of NF-κB and STAT1 — key inflammatory signaling molecules. In macrophages, EPSTI1 amplifies responses to IFN-γ and bacterial lipopolysaccharide.

The variant rs12870438 is intronic and does not change the protein sequence. Expression quantitative trait locus (eQTL) analyses of testicular tissue have not identified a significant effect of this SNP on EPSTI1 transcript levels in the testes, meaning the precise molecular mechanism by which the A allele may impair spermatogenesis remains an open research question33 an open research question
the original Hutterite GWAS authors noted EPSTI1's role in testicular immune privilege as a hypothesis
. One candidate explanation is that altered NF-κB or interferon signaling in spermatocytes disrupts the immune-tolerant microenvironment that developing sperm require.

The Evidence

The original discovery came from a two-stage GWAS44 two-stage GWAS
genome-wide association study
in 269 Hutterite men (a religious founder population that proscribes contraception, providing a natural fertility endpoint) followed by validation in 123 ethnically diverse men from Chicago with documented semen analyses. Under a recessive genetic model, homozygous A-allele carriers showed associations with reduced sperm concentration, total sperm count, total motile sperm count, average forward velocity, and mean amplitude of lateral head displacement (ALH) — five independent measures of sperm quantity and quality, all reaching p<0.05 after permutation testing.

A subsequent Japanese case-control study of 917 subjects55 Japanese case-control study of 917 subjects
Sato et al. Human Reproduction 2015
(76 azoospermic men, 50 oligospermic men, 791 fertile controls) found striking associations under the same recessive model: AA homozygotes had an odds ratio of 10.90 (95% CI 2.67–44.60) for azoospermia and 8.54 (95% CI 1.52–47.90) for oligospermia — both surviving correction for multiple testing. A separate larger Japanese replication study66 separate larger Japanese replication study
Sato et al. Human Reproduction 2015 — replication arm
in 2,015 men did not confirm the association with continuous semen parameters in meta-analysis, which the authors attributed to differing linkage disequilibrium structures around the locus between ethnic groups. A European cohort study of severe spermatogenic failure77 European cohort study of severe spermatogenic failure
Cerván-Martín et al. J Pers Med 2021
found EPSTI1-rs12870438 associated with severe oligospermia (additive model OR ~0.75, minor allele appearing protective in that direction).

Taken together, the evidence supports a genuine but modest and population-context-dependent association, which warrants an evidence level of moderate — replicated across multiple studies and populations but not yet mechanistically resolved, and with inconsistent replication in large continuous-parameter analyses.

Practical Actions

For men carrying the AA genotype (approximately 7% globally, up to 14% in European populations using Hardy-Weinberg estimates), awareness of potential spermatogenic vulnerability may be relevant to family planning timelines and warrants earlier rather than later fertility evaluation if conception is not occurring as expected. No specific supplement or nutritional intervention is known to modify the effect of this variant. Clinical semen analysis is the appropriate diagnostic next step for AA-homozygous men who are investigating fertility.

Interactions

The variant rs12870438 was originally identified alongside three other candidate male fertility loci in the same GWAS: rs7867029 (PSAT1), rs7174015 (USP8), and rs724078. No compound effects between these four loci have been formally studied, but men carrying multiple risk alleles across these GWAS-identified loci may have compounding effects on spermatogenesis given their distinct biological pathways (amino acid biosynthesis via PSAT1, ubiquitin-mediated acrosome assembly via USP8). Compound action data are not available for this variant combination.

Genotype Interpretations

What each possible genotype means for this variant:

GG “Common Genotype” Normal

Common genotype — typical sperm production potential

The GG genotype carries neither copy of the minor A allele and was the reference against which sperm parameters and azoospermia risk were measured in both the Hutterite/Chicago GWAS (Kosova et al. 2012) and the Japanese case-control studies (Sato et al. 2015). No specific action is indicated based on this variant alone. Other genetic and non-genetic factors (lifestyle, infections, hormonal status, other fertility-related variants such as USP8 rs7174015) remain relevant to overall fertility outcomes.

AA “Homozygous Risk” High Risk Warning

Two copies of the risk allele — elevated risk of reduced sperm production

The AA genotype was the risk group in the Hutterite GWAS (Kosova et al. 2012, PMID 22633400), where it associated with five sperm parameters simultaneously under a recessive model. In the Japanese case-control study (Sato et al. 2015, PMID 25908656), AA homozygotes showed an OR of 10.90 for azoospermia (95% CI 2.67–44.60, p=0.00087) and 8.54 for oligospermia (95% CI 1.52–47.90, p=0.015), both surviving multiple-testing correction. A replication study in a larger Japanese cohort (Sato et al. 2015, PMID 25908655) did not replicate the association with continuous semen parameters in meta-analysis, suggesting the recessive model signal may be sensitive to population structure and LD differences. The AA genotype frequency is ~7% globally but roughly 14% in European populations given the higher A-allele frequency (~38%) there.

EPSTI1's highest single-cell expression in the testis is found in late primary spermatocytes and early spermatids — precisely the cells most critical to meiotic completion and post-meiotic sperm maturation — consistent with a biological role in spermatogenesis, though no testicular eQTL has been confirmed for this variant.

The association is best understood as a genetic risk factor that increases susceptibility, not a deterministic cause of infertility. Men with this genotype who have not had formal semen analysis and are experiencing delayed conception should prioritize evaluation.

AG “Carrier” Carrier Caution

One copy of the risk allele — carrier status, typical fertility expected

The recessive model for rs12870438 means that the original Hutterite GWAS detected its strongest signal when comparing AA homozygotes to both GG and AG carriers combined. The Japanese case-control study (Sato et al. 2015, PMID 25908656) tested the recessive model specifically. Heterozygous carriers were not separated as a risk group in these analyses. The European severe-spermatogenesis cohort (Cerván- Martín et al. 2021, PMID 33383876) did find a signal under additive and dominant models, so a minor contribution cannot be entirely excluded — but current evidence does not support treating AG carriers as a clinically meaningful risk group for infertility.

Key References

PMID: 22633400

Kosova et al. 2012 — GWAS of 269 Hutterite men + 123 Chicago men; rs12870438 (EPSTI1) associated with sperm concentration, total count, total motile count, velocity, and ALH (all p<0.05 after permutation)

PMID: 25908656

Sato et al. 2015 — 917 Japanese subjects; rs12870438 associated with azoospermia (OR=10.90, 95% CI 2.67–44.60, p=0.00087, recessive) and oligozoospermia (OR=8.54, p=0.015)

PMID: 25908655

Sato et al. 2015 — Replication in 2,015 Japanese subjects; rs12870438 did not reach significance for continuous semen parameters in meta-analysis of two cohorts, suggesting population-specific LD structure

PMID: 33383876

Cerván-Martín et al. 2021 — European cohort with severe spermatogenic failure; EPSTI1-rs12870438 associated with severe oligospermia under additive and dominant models (OR=0.75 protective direction for minor allele)