rs5934505 — FAM9B

FAM9B Xp22 Variant — A Testosterone Locus Hidden on the X Chromosome

The X chromosome contains far fewer genes than autosomes, and its contribution to testosterone biology in men was largely overlooked until large genome-wide association studies began scanning the full genome including sex chromosomes. rs5934505 sits in the Xp22 region in a structural variant-rich zone between two genes, FAM9B and FAM9A¹¹ FAM9B and FAM9A
family with sequence similarity 9, members B and A — both encoded on the X chromosome and expressed exclusively in the testis, approximately 79 kb downstream of FAM9B and 145 kb upstream of FAM9A. Despite being in a gene-sparse intergenic region, this SNP has reached genome-wide significance for serum testosterone concentrations in men across multiple independent GWAS cohorts. It represents one of only a handful of replicated genetic loci influencing circulating testosterone levels.

The Mechanism

The precise molecular mechanism by which rs5934505 influences testosterone has not been established. Both FAM9B and FAM9A are expressed almost exclusively in the testis — specifically in spermatogenic cells — making them prime candidates for regulatory elements that govern Leydig cell androgen synthesis or spermatogenic feedback on testosterone production. As an intergenic variant in a known copy-number variable (CNV) region on Xp22, rs5934505 may tag a regulatory element affecting the transcription of one or both flanking genes, with downstream effects on steroidogenesis in testicular tissue. Because the locus is on the X chromosome, the effect is hemizygous in men — each male carries exactly one copy (T or C), with no heterozygous intermediate. Women carry two copies and can be TT, TC, or CC.

The T allele is the reference/major allele, present in approximately 73% of chromosomes globally. Despite being common, the T allele is associated with lower serum testosterone. This is an example of the common-variant, common-phenotype pattern where the major allele at a locus defines a disadvantageous baseline — the C allele (minor allele, ~27%) appears to be the ancestrally-derived testosterone-promoting variant retained at moderate population frequency by either selection or drift.

The Evidence

The primary discovery study, Ohlsson et al. 2011 (PLoS Genetics), conducted a GWAS meta-analysis across 10 independent cohorts totaling 14,429 Caucasian men²² Ohlsson et al. 2011 (PLoS Genetics), conducted a GWAS meta-analysis across 10 independent cohorts totaling 14,429 Caucasian men
Genetic determinants of serum testosterone concentrations in men, finding rs5934505 at a combined p=5.6×10⁻¹⁶ for total testosterone and calculated free testosterone — one of only two loci reaching genome-wide significance. Men with the T allele had lower mean serum testosterone than those with the C allele. The variant explained 0.6% of the variance in serum testosterone concentrations in the MrOS Sweden replication cohort, making it a modest but reliably detected effect at the population level.

Jin et al. 2012 (Human Molecular Genetics), examining 3,225 men in an Australian cohort, independently confirmed the FAM9B locus at genome-wide significance (p=1.61×10⁻⁸)³³ Jin et al. 2012 (Human Molecular Genetics), examining 3,225 men in an Australian cohort, independently confirmed the FAM9B locus at genome-wide significance (p=1.61×10⁻⁸)
Genome-wide association study identifies a new locus JMJD1C at 10q21 that may influence serum androgen levels in men. The same study also found a nominally significant association with dihydrotestosterone (DHT; p=1.10×10⁻⁵), suggesting the variant may influence the full androgen cascade rather than testosterone alone. The minor allele (C, ~28% MAF in this Australian cohort) was associated with higher levels.

A 2018 meta-analysis extending the search to circulating estrogen levels in men, Eriksson et al. 2018 (J Clin Endocrinol Metab)⁴⁴ Eriksson et al. 2018 (J Clin Endocrinol Metab)
Genetic determinants of circulating estrogen levels and evidence of a causal effect of estradiol on bone density in men, also identified rs5934505 at the FAM9B locus at genome-wide significance (p=3.4×10⁻⁸) for estradiol levels, suggesting this Xp22 locus may influence both androgen and estrogen concentrations — consistent with the biological link between testosterone and its aromatization product estradiol.

Practical Actions

For men carrying the T allele (the majority), awareness of this testosterone- lowering predisposition is most relevant when interpreting lab results in the context of symptoms of low testosterone — fatigue, reduced libido, loss of muscle mass, or mood changes. While the T allele's effect size (explaining ~0.6% of variance) is modest at the population level, it compounds with other testosterone-relevant loci (SHBG at rs727428, JMJD1C at rs10822184) and lifestyle factors (adiposity, sleep quality, zinc status, vitamin D) that all influence the same hormone. Men with borderline testosterone levels near the lower reference range may benefit from targeted testing and consideration of whether their genetic predisposition is compounding the effect.

For women, the X-linked nature means heterozygous TC carriers and homozygous CC carriers are expected to have different androgenic tone than TT homozygotes, though specific GWAS data for women at this locus are limited since the primary studies focused on male cohorts. The testis-exclusive expression of FAM9B and FAM9A suggests the primary mechanism is testicular in origin, with the X-linkage creating a dose asymmetry in women that may have milder hormonal consequences.

Interactions

Rs5934505 is one of a small set of confirmed testosterone-associated loci. The SHBG locus (rs727428, chromosome 17) is the strongest autosomal testosterone determinant through its regulation of sex hormone-binding globulin — which controls the ratio of free to total testosterone. The JMJD1C locus (rs10822184, chromosome 10) affects androgen levels through a different pathway. Men carrying the T allele at rs5934505 alongside testosterone-lowering variants at SHBG or JMJD1C may have compounded reductions in circulating androgen that approach the threshold for clinical concern.

CYP17A1 rs743572, which encodes a promoter variant of the rate-limiting steroidogenic enzyme, is in a separate category as it affects androgen precursor synthesis in the adrenal and gonadal steroidogenesis cascade. The FAM9B Xp22 locus is most likely acting through a distinct, testis-local mechanism rather than a shared pathway.