A direct in vitro effect of 17p-estradiol Although gonadal steroids have a profound effect on skeletal tissues (1-3), and estrogen deficiency has been established as a major etiologic factor in postmenopausal osteoporosis (1, 2), their influence has been considered to be indirect. This prevailing opinion was due to a lack of evidence for either specific estradiol receptors in bone (4) or direct biologic effects of sex steroids on bone cells (5, 6). The situation has dramatically reversed since recent studies have demonstrated significant, albeit low, concentrations of 17/3-estradiol (E2) receptors (refs. 7-9; 1) as well as androgen receptors ( ¶) in bone cells. The brain type (BB) isoenzyme of creatine kinase (CK; ATP:creatine N-phosphotransferase, EC 2.7.3.2; ref. 10) involved in the "energy buffer" system, which regulates cellular concentrations of ATP and ADP, is the major component (11) of the E2-induced protein of the rat uterus (12). E2-induced protein synthesis, and more recently, modulation of CK activity, has been a useful marker for studies on the mechanism ofaction of E2 in uterus and in other tissues that contain E2 receptors (13), because of its rapid response to E2 in vivo and in vitro (14). Moreover, it is a convenient marker for estrogen-modulated gene expression, since E2 treatment increases the steady-state level of mRNA for CK BB in the rat uterus (15). The speed and sensitivity of the assay for CK activity makes CK stimulation an efficient response marker to detect the action of E2 and other hormones (16) in skeletal tissues. In this report, we present evidence that E2 acts directly on cultured osteoblasts and epiphyseal cartilage cells, leading to increased CK activity as well as increased [3H]thymidine incorporation into DNA. Moreover, we report a rapid and sex-specific action of E2 and testosterone (T) on these markers in bones of prepubertal rats. Rat epiphyseal cells were obtained from vitamin Ddeficient 16-to 18-day-old rats, which have a wider epiphyseal cartilage zone than normally fed rats. Epiphyseal cartilage plates were isolated under a binocular dissecting microscope. Cells released by digestion with 0.25% collagenase (Worthington) in phosphate-buffered saline for 60 min at 370C were cultured as described above for calvaria cells (in 2 mM Ca2+). tTo whom reprint requests should be addressed. ¶Spelsberg, T. MATERIALS AND METHODS

We previously reported a non-enzymatic method for isolation of human bone cells in culture that display osteoblastic features and respond to 1,25 dihydroxy vitamin D (1,25) and to parathyroid hormone (PTH). The present study was undertaken to analyze the response of cultured human bone cells to 17beta-estradiol (E2) and to dihydrotestosterone (DHT) as a function of gender and age. Cultured human bone cells, obtained from biopsies during orthopedic surgery, were divided into four groups defined by gender and age: pre- and post-menopausal healthy non-osteoporotic women that were not under hormone replacement therapy (HRT) and mature (<55-year-old) and older (>60-year-old) men. We found gender specific responses to gonadal steroids using the specific activity of the brain type (BB) isozyme of creatine kinase (CK) as a response marker. Constitutive levels of CK activity did not change with age or gender and the enzyme extracted from cells from the different sexes and ages did not respond to either progesterone (P) or to 1,25. CK from the different cells responded to gonadal steroids in a gender specific manner, i.e. CK from female derived cells responded to E2 only and the enzyme from male derived cells responded to DHT only. In female derived cells the response to E2 declined significantly with age, while the response to DHT in CK from male derived cells did not vary with age. This may be due to either decreased proportion of mature osteoblasts and/or their differentiation state and/or changes in the levels of estrogen receptor(s), coactivators or corepressors in these cells. These results extend our knowledge of human osteoblast biology (beyond murine cells) and are therefore more relevant for developing models for treatment of human metabolic bone diseases such as post-menopausal osteoporosis.

No abstract

We have previously demonstrated sex-specific stimulation of creatine kinase specific activity (CK) in bone cells both in vivo and in vitro, in primary culture cells derived from rat and human bone and in established human bone-derived cell lines. We found that the female-derived cell line, SaOS-2, responded to 17 beta-estradiol (E2) by increased CK specific activity. The effects of E2 on the CK activity in SaOS-2 cells was inhibited by 100-fold excess of 4-hydroxytamoxifen (Tam) as well as by the other antiestrogen, ICI 164,384. Tam by itself had some stimulatory effect whereas ICI 164,384 showed no estrogenic activity. We also demonstrated the estrogenic-like effect of another anti-estrogen, raloxifene (Ral), which is agonist only in the SaOS-2 osteoblast-like cells but not in the human endometrial, Ishikawa cell line. Ishikawa cells respond to E2 and to Tam by increased CK activity. In both osteoblasts and endometrial cell lines, Ral and Tam were inhibitory in the presence of E2. The effects of E2 on SaOS-2 cells are at least partially mediated by the estrogen receptor (ER) at the level of transcription as demonstrated by transient transfection experiments using the human creatine kinase promoter chloramphenicol acetyltransferase in these cells. Pretreatment of SaOS-2 with calcitropic hormones, either 1,25 dihydroxyvitamin D3 (1,25(OH)2D3) or human parathyroid hormone (1-34) (hPTH(1-34)) increased the stimulation of CK by E2 by 40-60% relative to E2 alone and significantly increased the sensitivity of the cells to E2 by lowering the effective hormonal dose needed for stimulation of CK by E2 by 100-fold. This stimulatory effect of pretreatment of the cells with 1,25(OH)2D3 was due to a 2.5-fold increase in the level of ER expression as measured directly by enzyme immunoassay in the SaOS-2/1 subline. The increase in the responsiveness to E2 by hPTH(1-34) was not due to an increase in ER level in the cells. We can conclude that in cell cultures as in vivo, Ral shows different effects depending on the cell type, namely estrogenic-like activity in skeletal cells but not in uterine cells. We can also conclude that as with rat-derived cells, in bone cells derived from human bone 1,25(OH)2D3 increased the sensitivity to E2 due to an increase in the number of ER in the cells, whereas PTH(1-34) augmented the response to E2 without increasing ER, by another, as yet unknown, mechanism. These studies suggest that the treatment of pathological bone disorders may be improved by combined hormone therapy.

No abstract