INTRODUCTION: MECHANICAL LOADING AND REGULATION OF BONE ARCHITECTURE
Bone mass and architecture are determined by genetic, nutritional, hormonal, and physical factors (11). Of these, mechanical loading provides the functional influence, which ensures that the skeleton is sufficiently robust to withstand the forces of everyday activities without fracture (9). Loading exerts its influence on bone cells through the strains it engenders within bone tissue. The bone cell population seems to adjust bone mass and architecture to regulate these strains (Fig. 1). The presumed objective of this homeostatic mechanism is to ensure that bones are strong enough to resist fracture from all the loads habitually encountered and from a safe proportion of those resulting from accidents.
The influence of mechanical loading on bone mass can be observed most easily in people at the extremes of the mechanical loading spectrum: athletes at one extreme and astronauts or paraplegics at the other (9). Whenever physical activity is substantially reduced, bone tissue is lost and low levels of physical activity during growth result in a low bone mass. Conversely, substantial increases in bone mass occur in response to physical exercise. However, not all physical activity is equally effective. Human exercise studies support the results of animal experiments, which suggest that fast, high impact, high strain, diverse activities (such as squash and triple jump) are much more osteogenic than less error-rich activities (such as swimming and cycling), which engender a bone mass no higher than in physically active controls.
CHANGES IN BONE MASS THROUGHOUT LIFE
Individuals normally achieve their peak bone mass by the age of 20 to 30 yrs after a period of rapid growth during the peripubertal years (Fig. 2) (14). After the achievement of peak bone mass, there is a gradual decline in bone mass with age. At menopause, this decline accelerates dramatically and is associated with deterioration in bone architecture. In at least 30% of women (and 13% of men) more than 50 yrs of age, bone loss results in osteoporosis (defined as a bone mass more than 2.5 standard deviations below the relevant young adult mean (11)). This low bone mass is commonly associated with one or more nontraumatic vertebral fractures, or fractures of the wrist or hip sustained after a fall from no more than the standing position. The pattern of bone loss in these older women is remarkably similar to that which results from disuse.
BONE’S ADAPTIVE RESPONSE TO MECHANICAL LOADING AND REPRODUCTIVE STATUS
The responsiveness of bone to exercise varies with different stages of life. The skeleton clearly is more responsive during the peripubertal years than afterward (2). The disparity between the higher bone mass in the dominant arm of professional tennis players relative to their nondominant arm is much greater in players who started playing in the prepubertal or peripubertal years than in those who started playing after puberty.
The loss of bone that follows menopause is itself an indication of a failure to maintain the appropriate balance between bone mass and bone loading. Moreover, postmenopausally, only high intensity resistance programs involving high load and low repetition strength exercises seem to be capable of preventing bone loss and stimulating increases in bone mineral density (BMD) of 1% to 4% per year (6). Aerobic, endurance training involving walking and jogging type exercises, which are not associated with high strains and strain rates, have only a marginal positive effect in preventing postmenopausal bone loss (6).
If one assumes that the objective of mechanically adaptive bone modeling and remodeling is to achieve and maintain strain targets, then either these targets change throughout life or the response to them changes in its effectiveness. Peripubertally, target strains are relatively low, or the response to them is more effective, compared with the prepubertal and postmenopausal states when either the strain targets are higher or the response to them is less effective.
ESTROGEN CONCENTRATIONS ALTER THE SET POINT FOR BONE MASS HOMEOSTASIS
Changes in the target strain or the effectiveness of the adaptive response seem to coincide with changes in serum sex hormone concentrations. The increase in the potential of mechanical loading to stimulate bone gain in the peripubertal period is associated with marked increases in serum estrogen. However, in amenorrheic gymnasts, dancers, and runners of the same age, mechanical loading does not enhance bone gain. As a consequence, these young sports players have a lower BMD than age-matched controls (15). The onset of postmenopausal osteoporosis is associated with the 80% to 85% decline in estrogen, which occurs at menopause. Male age-related osteoporosis also is correlated closely with a steady decline in bioavailable estrogen (11). It follows that in postmenopausal women, estrogen replacement therapy should enhance the osteogenic response to an appropriate exercise regimen. This seems to be the case; Kohrt et al. (7) report an annual increase in BMD of up to 8.3% in response to such exercise in the presence of estrogen replacement therapy.
ESTROGEN RECEPTOR α
A number of studies suggest that estrogen’s ability to alter the strain set point for bone’s adaptive response to mechanical loading is the result of an indirect effect on the number or functional competence of estrogen receptor α (ERα) in bone cells (9), or both. Estrogen receptor α belongs to a family of transcription factors that are activated to regulate gene expression by binding to specific ligands (8). As its name suggests, ERα is activated by estrogen and is responsible for mediating many of the actions of estrogen. In the classical model of ERα function (Fig. 3), the binding of estrogen to ERα results in its phosphorylation and dimerization. This in turn facilitates binding to specific DNA sequences, known as estrogen response elements (ERE). This is followed by gene transcription. However, a number of nonclassical pathways for ERα function have been identified, which include the activation of ERα by phosphorylation in the absence of estrogen and an increase in the expression of genes, which do not contain EREs, by ERα (5,8).
Estrogen receptor α has been detected in osteoblasts that form bone, osteoclasts that remove bone, and osteocytes that are former osteoblasts now embedded within bone tissue (11). A major function of osteocytes is believed to be the initial response to strain (i.e., strain measurement). Osteocytes also may be involved in the control of bone (re)modeling by their communication of strain-related information from the depths of the bone to osteoblasts on the surface and possibly osteoclasts in the circulation (9).
ESTROGEN RECEPTOR α IN BONE’S ADAPTIVE RESPONSE TO MECHANICAL LOADING
It is generally accepted that ERα mediates most of the functions of estrogen in bone, including the regulation of osteoblast proliferation and differentiation, the synthesis of bone matrix proteins, the promotion of osteocyte survival, the suppression of bone resorption, and the promotion of osteoclast death (11). We believe that there is now convincing data to suggest that ERα in osteoblasts (and by inference in osteocytes) is also required for bone’s adaptive response to mechanical loading.
One of the early osteogenic responses to mechanical loading in vivo is the proliferation of cells of the osteoblast lineage before their activity in laying down new bone. This response can be replicated in vitro by straining cultures of osteoblast-like cells in the absence of estrogen. Furthermore, in rodent cells at least the strain-related proliferation of these osteoblast-like cells can be enhanced by increasing ERα number per osteoblast and blocked by selective estrogen receptor modulators such as Tamoxifen (Sigma in Poole, Dorset, U.K.) and ICI182780 (9).
Bone’s adaptive response to mechanical strain in vivo is dependent on an increase in intracellular calcium, nitric oxide production, and prostaglandin synthesis. In vitro it can be shown that activation of these three processes leads to ERα phosphorylation and ERE activation (4). Strain-induced ERα phosphorylation does not require the presence of estrogen, but is dependent on extracellular regulated kinase, a member of the mitogen activated protein kinase family (4). Therefore, from our experiments, it seems that the early stages of the pathway that leads from stimulation by strain (or its direct consequences such as fluid flow) to adaptive (re) modeling include: an increase in intracellular calcium; nitric oxide and prostaglandin production; extracellular regulated kinase activation; and ERα phosphorylation. Moreover, despite the requirement for ERα, estrogen per se is not an integral component of this pathway (Fig. 4).
Perhaps the most persuasive evidence that ERα is necessary for bone’s in vivo adaptive response to loading is a study that shows that this response is deficient in animals lacking a competent ERα gene (ERα−/-;Fig. 5) (10). Ten-min periods of mechanical loading engendering high physiological strains and strain rates activates bone formation in the ulnae of adult female wild-type mice (ERα+/+), resulting in an increase in bone area. The same loading stimulus also activates bone formation in the ulnae of adult female ERα−/- mice, but the increase in bone area is 70% lower than in wild types (Fig. 5). These in vivo data are supported by in vitro experiments that show that whereas osteoblasts derived from ERα+/+ mice proliferate in response to strain, those from their ERα−/- littermates do not. Moreover, insertion of the ERα gene into ERα−/- osteoblasts rectifies this deficiency. These studies indicate that ERα is required for at least those parts of bone’s adaptive response to mechanical loading that involve osteoblast proliferation and bone formation.
ESTROGEN RECEPTOR β CANNOT COMPENSATE FOR ERα
Estrogen receptor β (ERβ) is the second known ER. It has been detected in osteoblasts, but its function in bone is currently uncertain. Studies of ERβ knockout mice (ERβ−/-) suggest that ERβ is not required for the protective effect of estrogen on bone mass, but may mediate these effects in the absence of ERα (12). Additionally, ERβ may have a repressive effect on ERα expression in bone and therefore may counteract the effects of estrogen on bone (16). We are not aware of any published studies documenting the role of any ERβ in the adaptation of bone to mechanical loading. Our own unpublished data suggest that ERβ may be involved in bone’s adaptive response to mechanical loading, but that ERβ cannot compensate for the absence of ERα.
ERα AND BONE MASS HOMEOSTASIS IN HUMANS
The puzzle as to why bone’s adaptive response to mechanical loading fails to maintain an appropriate bone mass and architecture in states of relative estrogen deficiency could be explained by decreases in the number or function of ERα in bone cells of the osteoblast lineage, or both. Fewer ERα-positive osteocytes have been detected in bone biopsies from estrogen-deficient women compared with estrogen-replete women, and fewer ERα-positive osteoblasts and osteocytes have been identified in bone biopsies from men with idiopathic osteoporosis compared with age-matched controls (3). In addition, osteoblasts cultured from postmenopausal women not on estrogen replacement therapy have been found to be less responsive to estrogen in terms of estrogen-induced collagen synthesis and ERE activity compared with osteoblasts from younger women (1). Additional clinical studies are required to confirm the association between: decreases in ERα number and activity; changes in bone’s adaptive response to mechanical loading; and bone loss in humans.
Further investigations are needed to determine whether the enhanced adaptive response to mechanical loading during the peripubertal years is associated with increased ERα function in bone. However, a role for ERα in the achievement of a normal peak bone mass in humans has been demonstrated clearly by the finding of severely low bone mass in a young man with a null mutation in the ERα gene. At 28 yrs of age, this man had a bone age of 15 yrs, which was associated with a BMD in the lumbar spine of more than two standard deviations below the mean for 15-yr-old boys (13).
Mechanical loading is required for the achievement and maintenance of an appropriate bone mass and architecture sufficient to withstand the forces of everyday activities without fracture. Experimental studies using animals in vivo and bone-derived cells in vitro suggest that functional ERα is necessary for bone’s adaptive response to mechanical loading. Human clinical data support the hypothesis that the functional ERα number in resident bone cells is a limiting factor for bone’s ability to respond to mechanical loading. Moreover, ERα function and number in bone cells is dependent on estrogen. It follows that the central reason for bone loss after menopause may be a decline in ERα number or function, or both, resulting from estrogen deficiency. If this is the case, therapy designed to combine sensible levels of exercise with increased ERα number (or enhanced ERα function) in bone cells would be an effective means of increasing peak bone mass in the young population or preventing the development of postmenopausal and male age-related osteoporosis, or both.
1. Ankrom, M. A. Patterson, J. A. d’Avis, P. Y. Vetter, U. K. Blackman, M. R. Sponseller, P. D. Tayback, M. Robey, P. G. Shapiro, J. R. Fedarko. N. S. Age-related changes in human oestrogen receptor alpha function and levels in osteoblasts. Biochem. J. 333: 787–794, 1998.
2. Bass, S. L. Saxon, L. Daly, R. M. Turner, C. H. Robling, A. G. Seeman, E. Stuckey. S. The effect of mechanical loading on the size and shape of bone in pre-, peri-, and postpubertal girls: a study in tennis players. J. Bone Miner. Res. 17: 2274–2280, 2002.
3. Braidman, I. P. Baris, C. Selby, P. L. Adams, J. E. Freemont, A. J. Hoyland. J. A. Preliminary report of impaired oestrogen receptor-alpha expression in bone, but no involvement of androgen receptor, in male idiopathic osteoporosis. J. Pathol. 192: 90–96, 2000.
4. Jessop, H. L. Rawlinson, S. C. Pitsillides, A. A. Lanyon. L. E. Mechanical strain and fluid movement both activate extracellular regulated kinase (ERK) in osteoblast-like cells but via different signaling pathways. Bone 31: 186–194, 2002.
5. Kato, S. Masuhiro, Y. Watanabe, M. Kobayashi, Y. Takeyama, K. I. Endoh, H. Yanagisawa. J. Molecular mechanism of a cross-talk between oestrogen and growth factor signaling pathways. Genes Cells 5: 593–601, 2000.
6. Kerr, D. Morton, A. Dick, I. Prince. R. Exercise
effects on bone mass in postmenopausal women are site specific and load-dependent. J. Bone Miner. Res. 11: 218–225, 1996.
7. Kohrt, W. M. Snead, D. B. Slatopolsky, E. Birge. S. J. Additive effects of weight-bearing exercise
and estrogen on bone mineral density in older women. J. Bone Miner. Res. 10: 1303–1311, 1995.
8. Kushner, P. J. Agard, D. Feng, W. J. Lopez, G. Schiau, A. Uht, R. Webb, P. Greene. G. Oestrogen receptor function at classical and alternative response elements. Novartis. Found. Symp. 230: 20–26, 2000.
9. Lanyon, L. Skerry. T. Postmenopausal osteoporosis as a failure of bone’s adaptation
to functional loading: a hypothesis. J. Bone Miner. Res. 16: 1937–1947, 2001.
10. Lee, K. Jessop, H. Suswillo, R. Zaman, G. Lanyon. L. Bone adaptation
requires oestrogen receptor-alpha. Nature 424: 389, 2003.
11. Riggs, B. L. Khosla, S. Melton L. J. III. Sex steroids and the construction and conservation of the adult skeleton. Endocr. Rev. 23: 279–302, 2002.
12. Sims, N. A. Dupont, S. Krust, A. Clement-Lacroix, P. Minet, D. Resche-Rigon, M. Gaillard-Kelly, M. Baron. R. Deletion of estrogen receptors reveals a regulatory role for estrogen receptors-beta in bone remodeling in females but not in males. Bone 30: 18–25, 2002.
13. Smith, E. P. Boyd, J. Frank, G. R. Takahashi, H. Cohen, R. M. Specker, B. Williams, T. C. Lubahn, D. B. Korach. K. S. Estrogen resistance caused by a mutation in the estrogen-receptor gene in a man. N. Engl. J. Med. 331: 1056–1061, 1994.
14. Theintz, G. Buchs, B. Rizzoli, R. Slosman, D. Clavien, H. Sizonenko, P. C. Bonjour. J.-P. Longitudinal monitoring of bone mass accumulation in healthy adolescents: evidence for a marked reduction after 16 years of age at the levels of lumbar spine and femoral neck in female subjects. J. Clin. Endocrinol. Metab. 75: 1060–1065, 1992.
15. Warren, M. P. Brooks-Gunn, J. Fox, R. P. Holderness, C. C. Hyle, E. P. Hamilton. W. G. Osteopenia in exercise
-associated amenorrhea using ballet dancers as a model: a longitudinal study. J. Clin. Endocrinol. Metab 87: 3162–3168, 2002.
16. Windahl, S. H. Hollberg, K. Vidal, O. Gustafsson, J. A. Ohlsson, C. Andersson. G. Female estrogen receptor beta-/- mice are partially protected against age-related trabecular bone loss. J. Bone Miner. Res. 16: 1388–1398, 2001.