Volume 13, Issue 3

Testosterone (T) restores bone mass loss in postmenopausal women and osteoporotic men mainly through its bioconversion to estradiol (E 2 ). In target tissues, T is also biotransformed to the A-ring-reduced metabolites 3α,5α-androstanediol (3α,5α-diol) and 3β,5α-androstanediol (3β,5α-diol), which are potent estrogen receptor (ER) agonists; however, their biological role in bone has not been completely elucidated. To assess if osteoblasts bioconvert T to 3α,5α-diol and to 3β,5α-diol, we studied in cultured neonatal rat osteoblasts the metabolism of [ 14 C]-labeled T. In addition, the intrinsic estrogenic potency of diols on cell proliferation and differentiation in neonatal calvarial rat osteoblasts was also investigated. Osteoblast function was assessed by determining cell DNA, cell-associated osteocalcin, and calcium content, as well as alkaline phosphatase activity and Alp1 gene expression. The results demonstrated that diols were the major bioconversion products of T, with dihydrotestosterone being an obligatory intermediary, thus demonstrating in the rat osteoblasts the activities of 5α-steroid reductase and 3α- and 3β-hydroxysteroid dehydrogenases. The most important finding was that 3β,5α- and 3α,5α-diols induced osteoblast proliferation and differentiation, mimicking the effect of E 2 . The observation that osteoblast differentiation induced by diols was abolished by the presence of the antiestrogen ICI 182,780, but not by the antiandrogen 2-hydroxyflutamide, suggests that diols effects are mediated through an ER mechanism. The osteoblast capability to bioconvert T into diols with intrinsic estrogen-like potency offers new insights to understand the mechanism of action of T on bone cells and provides new avenues for hormone replacement therapy to maintain bone mass density.

Background: In women with breast cancer who were treated with either continuous tamoxifen alone or sequential tamoxifen followed by megestrol acetate (MA), we demonstrated significant positive associations between the breast tumor estrogen receptor (ER) and an increase in serum sex hormone-binding globulin (SHBG) during tamoxifen treatment. We interpreted this as “ER uniformity” in different tissues, e.g., breast, liver. No other associations with ER were found. In the same study, the breast tumor progesterone receptor (PR) was determined. Our aim was to see if there were any associations between PR and endocrine changes during MA treatment. Methods: The breast tumor PR before treatment and serum insulin-like growth factor I (∂IGF-1), steroids, steroid-binding proteins, and insulin before and during treatment were measured in 17 postmenopausal women with breast cancer who were treated sequentially with tamoxifen 40 mg/day followed by MA 160 mg/day in alternating 3-month periods. Results: During MA treatment periods, the levels of IGF-1 and insulin increased significantly, whereas the levels of androgens, SHBG, corticosteroid-binding globulin, and cortisol decreased significantly. Significant positive correlations were found between the PR content and increments in ∂IGF-1 but not between PR and any other endocrine change. Conclusions: PR expression in human liver is very weak, but malignant and normal breast tissues secrete considerable amounts of growth hormone and IGF-1 in vitro and in vivo. This activity is stimulated by progestogens. The association between PR and ∂IGF-1 may therefore reflect a direct PR-mediated action of MA on malignant and normal human breast tissues in vivo.