CD38 is associated with premenopausal and postmenopausal bone mineral density and postmenopausal bone loss.

One goal of osteoporosis research is to identify the genes and environmental factors that contribute to low bone mineral density (BMD) and fracture. Linkage analyses have identified quantitative trait loci (QTLs), however, the genes contributing to low BMD are largely unknown. We examined the potential association of an intronic polymorphism in CD38 with BMD and postmenopausal bone loss. CD38 resides in 4p15, where a QTL for BMD has been described. CD38-/- mice display an osteoporotic phenotype at 3 months, with normalization of BMD by 5 months. The CD38 polymorphism was identified by polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) analysis in 457 postmenopausal and 173 premenopausal Caucasian women whose spine and hip BMD was measured by dual energy X-ray absorptiometry (DXA). Influence of the CD38 polymorphism on bone loss was analyzed in 273 postmenopausal women over a follow-up of 2.94 +/- 1.50 years. The CD38-PvuII polymorphism was significantly associated with premenopausal and postmenopausal (P = 0.001) lumbar spine BMD. Women homozygous for the G allele had >14% lower spinal BMD than women with GC/CC genotypes. An allele dose effect was observed at the spine in premenopausal (P = 0.002) and postmenopausal (P < 0.001) cohorts. The CD38-PvuII polymorphism was significantly associated with femoral neck BMD in pre- and postmenopausal women (P = 0.002 and P = 0.011, respectively). However, significance was lost following adjustment of hip BMD for covariates in the postmenopausal cohort (P = 0.081). The CD38-PvuII polymorphism was weakly associated with bone loss at the spine (P = 0.024), in postmenopausal women not taking hormone replacement therapy. We suggest that the CD38-PvuII polymorphism may influence the attainment and maintenance of peak BMD and postmenopausal bone loss.