Association of SRD5A2 Variants and Serum Androstane-3α,17β-Diol Glucuronide Concentration in Chinese Elderly Men

BACKGROUND: Results of recent studies have demonstrated that genetic variants of the enzyme steroid 5 alpha reductase type II (SRD5A2) are associated with serum concentrations of major androgen metabolites such as conjugates of androstane-3 alpha,17 beta-diol-glucuronide (3 alpha-diol-G). However, this association was not consistently found among different ethnic groups. Thus, we aimed to determine whether the association with SRD5A2 genetic variations exists in a cohort of healthy Chinese elderly men, by examining 2 metabolite conjugates: androstane-3 alpha,17 beta-diol-3-glucuronide (3 alpha-diol-3G) and androstane-3 alpha,17 beta-diol-17-glucuronide (3 alpha-diol-17G). METHODS: We used GC-MS and LC-MS to measure serum sex steroid concentrations, including testosterone and dihydrotestosterone, and 3 alpha-diol-3G and 3 alpha-diol-17G in 1182 Chinese elderly men age 65 and older. Genotyping of the 3 SRD5A2 tagSNPs [ rs3731586, rs12470143, and rs523349 (V89L)] was performed by using melting-temperature-shift allele-specific PCR. RESULTS: The well-described SRD5A2 missense variant rs523349 (V89L) was modestly associated with the 3 alpha-diol-17G concentration (P = 0.040). On the other hand, SNP rs12470143 was found to be significantly correlated with 3 alpha-diol-3G concentration (P = 0.021). Results of haplotype analysis suggested that the presence of an A-C-G haplotype leads to an increased 3 alpha-diol-3G concentration, a finding consistent with results of single SNP analysis. CONCLUSIONS: The genetic variation of SRD5A2 is associated with circulating 3 alpha-diol-3G and 3 alpha-diol-17G concentrations in Chinese elderly men. In addition, we showed that SRD5A2 haplotypic association, rather than a single SNP alone, might be a better predictor of the 3 alpha-diol-G concentration. Thus, the effect of either the haplotype itself or of other ungenotyped SNPs in linkage disequilibrium with the haplotype is responsible for the interindividual variation of 3 alpha-diol-G. (C) 2010 American Association for Clinical Chemistry