| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Molecular Endocrinology Group, Division of Medicine and, Medical Research Council Clinical Sciences Centre, Imperial College London, Hammersmith Hospital, London W12 0NN, United Kingdom
Address all correspondence and requests for reprints to: Graham R. Williams, Molecular Endocrinology Group, 5th Clinical Research Building, Medical Research Council Clinical Sciences Centre, Hammersmith Hospital, Du Cane Road, London W12 0NN, United Kingdom. E-mail: graham.williams{at}imperial.ac.uk.
For many years, postmenopausal osteoporosis has been attributed to estrogen deficiency (1). This established view has, nevertheless, been challenged by the recent proposal that FSH is required for hypogonadal bone loss and that high levels of FSH cause the condition (2). The conclusion arose from studies of FSHR knockout (FORKO) and FSHß knockout (FSHß–/–) mice, which were found to be resistant to bone loss despite severe estrogen deficiency. Surprisingly, hypogonadal FSHß–/– mice even exhibited increased bone mass at the femur, whereas eugonadal FSHß+/– heterozygotes displayed a generalized increase in bone mass with evidence of reduced bone resorption (2). FSHR protein was detected in osteoclasts, and FSH increased osteoclastogenesis and dentine resorption in vitro but did not influence bone formation or osteoblast activity directly. The novel idea that high circulating FSH causes hypogonadal bone loss sparked a series of editorials and correspondence debating the issue (3, 4, 5, 6, 7). Important criticisms emerged and several key questions have now been investigated experimentally. The findings are reported by Gao et al. in this issue of Endocrinology (8).
It has been argued that women with hypothalamic or hypergonadotrophic hypogonadism have similar rates of trabecular bone loss even though hypothalamic hypogonadism results in low FSH and estrogens, whereas hypogonadism after oophorectomy or the menopause results in high FSH and low estrogens. Thus, bone loss in women occurs equally in response to estrogen deficiency in the presence of either low or high FSH levels (4). Furthermore, hypogonadal hpg mice harbor a partial deletion of the GnRH gene (9) and develop severe FSH, LH, and sex hormone deficiencies (10), thus providing a model of hypothalamic hypogonadism. Both male and female hpg mice exhibit severe bone loss even though FSH is undetectable (11, 12), and this is reversed after testosterone or estrogen replacement. Nevertheless, bone loss in hpg mice is accompanied by predominant effects on osteoblastic bone formation (11), whereas FSH acts directly in osteoclasts (2) and may not necessarily be expected to influence defects of osteoblast function (7). Despite this, in hpg mice, FSH is not required for the low bone mass evident in hypothalamic hypogonadism and a lack of FSH does not prevent amelioration of the phenotype in response to sex steroids. How might these observations be reconciled with the findings in FORKO and FSHß–/– mice, which were interpreted to indicate an estrogen-independent and direct effect of FSH on the skeleton resulting in bone loss and a requirement for FSH in the manifestation of hypogonadal bone loss (2)?
Unfortunately, circulating levels of estrogen, testosterone and LH in FORKO and FSHß–/– mice were not reported (2). Nevertheless, in previous studies homozygous FORKO mice were shown to have elevated LH concentrations and a 10-fold increase in testosterone (13, 14, 15), whereas early ovarian failure and age-related elevations of FSH, LH, and testosterone were reported in heterozygotes (16). Raised LH has also been documented in FSHß–/– mice (15). In addition to high levels of androgens, and in contrast to the findings of Sun et al. (2), FORKO mice displayed an age-related kyphosis due to vertebral compression associated with radiographic and histological evidence of bone loss (13). Thus, it is argued that the FORKO mouse per se is a poor model of hypogonadism because of elevated androgens, and that FSHß–/– mice do not appropriately represent the perimenopause (4, 17).
Are elevated levels of LH-dependent testosterone actually responsible for the lack of bone loss observed in FORKO mice (2), as suggested by others (5, 6)? Estrogens are undoubtedly important in females. Estrogen replacement alone is sufficient to reverse or prevent low bone mass after acute estrogen withdrawal (1). Moreover, female estrogen receptor (ER) 
and ß double knockout mice exhibit bone loss that cannot be prevented by estrogen replacement (18) despite normal levels of FSH (19). Estrogens and the peripheral conversion of androgens to estrogens by the aromatase enzyme are also important in males because ER deletion (20) or aromatase deficiency (21) in men both result in bone loss. Furthermore, high levels of testosterone in aromatase knockout mice do not prevent bone loss accompanied by high bone turnover, whereas the phenotype is reversed by estrogen replacement in both sexes (22, 23), suggesting that aromatization of androgens is important for maintenance of bone. On the other hand, the increased osteoclastic bone resorption and high levels of FSH observed in aromatase knockout mice (24) have been proposed to indicate that FSH increases bone resorption despite high levels of testosterone (7). It was suggested further that conversion of testosterone to estrogen in FORKO and FSHß–/– hypogonadal mice is unlikely to be important because of low levels of aromatase expression and negligible serum concentrations of estrogen in these animals (7). In this issue of Endocrinology, Gao et al. address this debate experimentally for the first time using female FORKO mice (8).
In these studies (8), FORKO mice were hypogonadal with reduced levels of estradiol but elevated LH accompanied by a marked increase in testosterone. After ovarian transplantation, circulating testosterone levels in FORKO mice were restored to normal, estrogen levels increased and LH levels fell. In contrast to observations reported by Sun et al. (2), FORKO mice exhibited age-related bone loss accompanied by reduced osteoblastic bone formation, together with evidence of increased osteoclastic bone resorption. Ovarian transplantation of FORKO mice resulted in amelioration of the bone mineral density and trabecular bone losses observed in sham operated mice. These findings demonstrate that ovarian function is important for age-related bone loss in FORKO mice.
To investigate bone maintenance, Gao et al. ovariectomized 3-wk-old wild-type and FORKO mice and examined the effect at 3 months of age. Even though sham-operated FORKO mice displayed greater age-related bone loss than sham-operated wild-type littermates, ovariectomy resulted in further bone loss such that levels of reduced bone mineral density and trabecular bone volume were the same in ovariectomized wild-type and FORKO mice. Bone loss in ovariectomized wild-type mice was accompanied by increased osteoblastic bone formation and increased osteoclastic resorption, whereas in FORKO mice bone formation was unaffected by ovariectomy. These findings demonstrate that ovarian function is required for bone maintenance in FORKO mice, perhaps acting primarily by regulating bone resorption.
To examine whether elevated ovarian androgens were involved in the skeletal responses to ovariectomy, Gao et al. treated FORKO mice with the androgen receptor antagonist flutamide and the aromatase inhibitor letrozole. Flutamide resulted in a small nonsignificant diminution of bone mineral density compared with the significant reduction observed after ovariectomy, and a small but significant loss of trabecular bone volume, indicating that loss of direct actions of ovarian androgens at the androgen receptor do not account fully for the skeletal effects of ovariectomy in FORKO mice. By contrast, treatment with letrozole reduced bone mineral density by an amount equivalent to the reduction seen after ovariectomy and resulted in a 2-fold greater diminution of trabecular bone volume compared with treatment with flutamide. These findings indicate that peripheral conversion of elevated androgens to estrogens by aromatase has a predominant role in regulating bone maintenance in FORKO mice. The skeletal response to androgen receptor blockade was accompanied by evidence of increased osteoclastic bone resorption, but no effect on bone formation parameters, whereas the response to aromatase inhibition included effects on both bone formation and resorption. Together, these findings indicate that bone loss in FORKO mice is not as severe as the bone loss observed in ovariectomized wild-type mice. This is because bone loss in FORKO mice is mitigated by elevated ovarian androgens, which act predominantly after peripheral conversion to estrogens to inhibit bone loss. Because FORKO mice are resistant to the actions of FSH these studies reveal skeletal actions of ovarian androgens and estrogens that are independent of the bone resorptive actions of FSH.
FORKO mice in the current study exhibited age-dependent declines in bone mineral density and trabecular bone volume (8), whereas bone loss was not observed in the FORKO or FSHß–/– mice studied by Sun et al. (2). Although the reasons for this discrepancy have not been defined, the findings of Gao et al. are important but they do not tell the whole story. The evidence for direct actions of FSH in osteoclasts is strong (2), and recent studies have also revealed indirect actions of FSH in bone marrow macrophages and granulocytes that stimulate osteoblast and osteoclast formation (25). Furthermore, androgen levels remain to be reported in FSHß–/– mice, the effects of gonadectomy and the use of androgen receptor blockade and aromatase inhibition remain to be seen in male FORKO mice and in FSHß–/– mice of both sexes. Finally, inhibins, in addition to sex steroids, exert negative feedback regulation of the HPG axis (26, 27) and are inhibitors of osteoblast and osteoclast differentiation in vitro (28, 29) that also regulate bone mass in vivo (30).
Regulation of bone by the hypothalamic-pituitary-gonadal axis is now a complex and controversial subject that involves not only simple estrogen deficiency. It will be important to establish more definitively the relative contributions of FSH excess, inhibins and elevated ovarian androgens in response to hypogonadism because such analyses may open novel and unexpected therapeutic avenues. There is excitement in the field but further progress will require sophisticated approaches to dissociate the physiological inverse relationship between estrogen (and inhibins) and FSH to resolve the relative importance of the pituitary and the ovary. The studies of Gao et al. in this issue (8) start this journey by presenting new evidence that tips the balance back in favor of the ovary.
 |
Footnotes |
|---|
Disclosure Summary: The author has nothing to declare.
Received March 12, 2007.
Accepted for publication March 13, 2007.
|
References |
|---|
 Top References |
|---|
but not ERß. Mol Endocrinol 17:1039–1053
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
