New Perspectives on the Skeletal Role of Estrogen

Robert Marcus M.D.

Stanford University School of Medicine Veterans Affairs Medical Center Palo Alto, California 94304

Address correspondence and requests for reprints to: Robert Marcus, M.D., Department of Aging Study Unit, Veterans Affairs Medical Center, Stanford University School of Medicine, 3801 Miranda Avenue, 182-B, Palo Alto, California 94304.


    Introduction
 Top
 Introduction
 References
 
IN case you hadn’t noticed, there’s been a revolution in the understanding of the skeletal roles of estrogen. It used to be so simple: estradiol, acting through a single receptor, was the important bone-related hormone for women; testosterone, acting through its own receptor, performed similar duty for men. Now we are confronted with the reality of at least two separate estradiol receptors, each with its characteristic tissue distribution (1). We have also learned, via fascinating experiments of nature (what our Editor would call "prismatic cases"), that initiation of adolescent growth and bone acquisition requires estradiol as the critical reproductive hormone in both boys and girls. Boys carrying mutations in the {alpha}-estradiol receptor or in the aromatase that converts androgen to estrogen show profound deficits in bone mineral density (BMD) and, because in these patients epiphyseal closure is not stimulated by estrogen, maintain slow prepubertal growth into adult life (2, 3, 4). In patients with aromatase deficiency, estrogen treatment leads rapidly to epiphyseal closure and striking gains in BMD (4). Surprises have emerged later in life as well. In older men, evidence suggests that circulating estrogens predict the rate of bone loss more strongly than do androgen concentrations (5, 6).

Now another shoe has fallen. In a series of three papers, investigators from The Study of Osteoporotic Fractures (SOF), a large observational study of more than 9000 older women, have described a strong inverse relation between endogenous serum 17-ß estradiol concentrations and skeletal health (7, 8, 9). Cummings et al. (7) reported a relationship of circulating estradiol to vertebral and hip fracture incidence, while Stone et al. (8) described a relationship of serum estradiol concentrations to rates of bone loss. In this issue of JCEM (see page 2239), Ettinger and colleagues (9) round out this story by showing a significant relationship between estradiol concentrations and BMD at multiple sites in a separate randomly drawn cohort of SOF participants (here called the validation cohort). They also extend the analysis of the original cohort of Cummings et al. (7), to show that "high-quartile" estradiol women had fewer prevalent vertebral fractures on baseline radiographs than did women with "low-quartile" estradiol concentrations (for unclear reasons, this finding was not replicated in the validation cohort).

It should not be surprising that such relationships exist. Earlier cross-sectional studies in older women indicated significant relationships between endogenous estrogen and bone mass and also suggested that women who had sustained a hip fracture had lower blood concentrations of free estradiol. However, the SOF data are the first to show prospectively that a lower postmenopausal estradiol concentration translates into a higher fracture incidence. What may come as a surprise to many readers are the apparently minuscule gradients in estradiol concentration, all within the nominal postmenopausal range, that appear capable of influencing BMD and fracture. Using sensitive assays with detection limits of only 2 pg/mL, higher BMD was shown for women with estradiol concentrations above 10 pg/mL than for those with values below 5 pg/mL. Two separate issues have delayed appreciation of this relationship until now. The first is the dominant paradigm in which the medical community has operated, i.e. the indisputable finding that estradiol concentrations drop below 30 pg/mL in virtually all menopausal women, that accelerated bone loss is linked to that drop, and that protection against bone loss requires administration of exogenous estrogen sufficient to restore plasma estradiol concentrations to at least 50 pg/mL. Thus, in the care of older women we have been so focused on establishing the estrogen dose necessary to conserve bone, that little interest has surrounded the contribution of residual endogenous estrogen to rates of loss in nontreated women. The second limitation has been the availability of estradiol assays with adequate sensitivity to distinguish gradients in this very low estradiol concentration range. Most commercial and even research assays that are calibrated to measure hormone concentrations typical of premenopausal women are simply inadequate for this task.

In the paper by Ettinger et al. (9) the emergence of total estradiol as the most robust predictor of BMD does not preclude other circulating factors from contributing to the overall estrogenic milieu of these women. One such factor is the concentration of biologically active unbound, or free, estradiol. Although that component of total estrogens was not specifically measured, women in the high estradiol quartile had average concentrations of SHBG, the primary sex-steroid binding protein, that were 18% lower than those of women in the low quartile. Consequently, the high quartile women should have an amplification of their free estradiol concentrations. The explanation for lower SHBG concentrations is not clear, but could result from their higher concentrations of testosterone, a known inhibitor of SHBG production.

The most abundant circulating estrogen in postmenopausal women is not 17-ß estradiol, but estrone. Indeed, women in the high estradiol quartile also had estrone concentrations that were more than double those of women in the lowest quartile, and concentrations of both estrogens were highly correlated. Although estradiol showed a stronger independent relationship to BMD, it is premature to exclude estrone from playing a contributory role. First, by reduction of the ketone group in carbon 17, estrone is the proximate source of most circulating estradiol in postmenopausal women. Second, although estradiol has considerably greater potency in vitro, estrone does possess intrinsic estrogenic activity. The fact that estrone is not 17-hydroxylated means that it cannot bind to SHBG, and its free concentrations exceed by a considerable margin those of estradiol. Thus, in the postmenopausal woman, relative contributions of these two hormones to total estrogenic activity may be closer than is generally supposed. In this regard, a companion paper by Stone et al.(8) published elsewhere reported significant inverse relationships of both estradiol and estrone to bone loss at the hip.

One fly in the ointment of this construction comes from a companion paper (also published in another journal) of Cummings et al. (7), which reported the counterintuitive observation that women with estrone concentrations above 1.5 ng/dL had a significantly higher risk for fracture (adjusted odds ratio = 2.3) than those with estrone concentrations below that amount! Try as I might, no biological underpinning for such a result comes to mind. However, two points are worthy of speculation. First, the cutoff of 1.5 ng/mL may have isolated a group of women whose extraordinarily low estrone concentrations were due to some particular characteristic, such as a high transfer constant for converting estrone to estradiol. Alternatively, extremely low estrone concentrations could also be a marker for unusual patterns of estrogen metabolism, such as higher than usual conversion to active 16-hydroxylated metabolites as opposed to relatively inactive 2-hydroxylated compounds (10). At present there is no basis for selecting any of these possibilities.

Postmenopausal osteoporosis has been considered the province of slender women. Indeed, significant positive relationships between body mass and BMD have been frequently reported (11, 12), and among postmenopausal women not taking HRT, obesity appears to minimize the rate of bone loss (13). Whether skeletal protection of heavier women reflects their increased mechanical loading or the fact that adipose tissue aromatizes androgens to estrogens has been a matter of frequent speculation. In the report of Ettinger et al. (9), a progressive increase in body weight was observed across estradiol quartiles. However, the relationship between estradiol and BMD persisted even when adjustments for weight were made.

Administered in conventional doses, estrogen actions on bone result from inhibition of bone remodeling. It is difficult to accept the argument that this mechanism was of paramount importance for the results described in the SOF papers. Although serum concentrations of the bone turnover marker, osteocalcin, were slightly lower in the high estrogen women, their substantial elevation in all estradiol quartiles indicated a globally high degree of bone turnover. Therefore, it is intriguing to ponder alternative mechanisms for these results. One possibility concerns the preservation of osteocyte viability. Tomkinson et al. (14) showed that obliteration of estrogen production by aromatase inhibition increased apoptotic osteocyte death, and gross depletion of viable osteocytes has been demonstrated histologically in femoral bone from hip fracture patients. The basis for linkage between osteocyte number and bone integrity is not certain, but osteocytes are normally considered to be the primary cellular transducer of mechanical signals in bone.

Several possibilities also exist for estrogen actions on nonskeletal tissues that could indirectly promote skeletal benefit. Estrogen stimulates renal calcium conservation, increases production of 1,25-dihydroxyvitamin D, and promotes intestinal calcium absorption, any of which actions might conserve bone. It is possible that lower estrogen concentrations suffice for these nonskeletal actions than are required for suppression of bone remodeling.

The implications for clinical practice of these several studies remain unclear. Extrapolation of the data would suggest that some health benefits might be achieved by administering estrogen in lower than traditional amounts. Indeed, one recent study demonstrated increased BMD when women were given a combination of calcium and esterified estrogens at a daily dose of 0.3 mg, which by itself is considered subtherapeutic for preventing bone loss (15). Studies may be warranted to determine whether this regimen might prove useful for patients with higher risks for breast cancer or venous thrombosis. However, one must remember that a driving principle supporting the widespread use of postmenopausal estrogen is the substantial evidence that HRT protects women from coronary heart disease, an experience that largely reflects estrogen use at doses equivalent to 0.625 mg of conjugated estrogens. Any treatment schedule that reduced the apparent degree of cardiovascular protection provided by conventional estrogen doses would severely alter the HRT risk-benefit equation, and considerable thought would be required before endorsing such a strategy. Regardless of potential therapeutic applications that may result, the SOF papers offer fascinating insights into the health consequences of estrogen deprivation.

Received March 23, 1998.

Accepted March 25, 1998.


    References
 Top
 Introduction
 References
 

  1. Kuiper GG, Carlsson B, Grandien K, et al. 1997 Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptors alpha and beta. Endocrinology. 138:863–870.[Abstract/Free Full Text]
  2. Smith EP, Boyd J, Frank GR, et al. 1994 Estrogen resistance caused by a mutation in the estrogen-receptor in a man. N Engl J Med. 331:1056–1061.[Abstract/Free Full Text]
  3. Carani C, Qin K, Simoni M, et al. 1997 Effect of testosterone and estradiol in a man with aromatase deficiency. N Engl J Med. 337:91–95.
  4. Morishima A, Grumbach MM, Bilezikian JP. 1997 Estrogen markedly increases bone mass in an estrogen deficient young man with aromatase deficiency. Abstract 96, J Bone Miner Res. 12S:S126.
  5. Slemenda CW, Longcope C, Zhou L, Hui SL, Peacock M, Johnston CC. 1997 Sex steroids and bone mass in older men. Positive associations with serum estrogens and negative associations with androgens. J Clin Invest. 100:1755–1759.[Abstract/Free Full Text]
  6. Khosla S, Melton LJ III, Atkinson EJ, Riggs BL. 1997 Serum estrogen levels, rather than serum testosterone levels, are the major determinants of decreases in bone mineral density in aging men. J Bone Miner Res. 12S:S127 (Abstract 97).
  7. Cummings SR, Browner WS, Bauer D, Stone K, Ensrud K, Jamal S. Endogenous sex and calciotropic hormones and the risk of hip and vertebral fractures in older women: The Study of Osteoporotic Fractures. N Engl J Med. In press.
  8. Stone K, Bauer DC, Black DM, Sklarin P, Ensrud KE, Cummings SR. Hormonal predictors of bone loss in elderly women: a prospective study. J Bone Miner Res. In press.
  9. Ettinger B, Pressman A, Sklarin P, Bauer DC, Cauley JA, Cummings SR. 1998 Associations between low levels of serum estradiol, bone density, and fractures among elderly women: :The Study of Osteoporotic Fractures. J Clin Endocrinol Metab. 83:2239–2243.[Abstract/Free Full Text]
  10. Klug TL, Bradlow HL, Sepkovic DW. 1994 Monoclonal antibody-based enzyme immunoassay for simultaneous quantitation of 2- and 16{alpha}-hydroxyestrone in urine. Steroids. 59:648–655.[CrossRef][Medline]
  11. Dawson-Hughes B, Shipp C, Sadowski L, Dallal G. 1987 Bone density of the radius, spine, and hip in relation to percent of ideal body weight in postmenopausal women. Calcif Tiss Intl. 40:310–314.[Medline]
  12. Marcus R, Greendale G, Blunt B, et al. 1994 Correlates of bone mineral density in the post-menopausal estrogen/progestin interventions trial (PEPI). J Bone Miner Res. 9:1467–1476.[Medline]
  13. Harris S, Dawson-Hughes B. 1992 Rates of change in bone mineral density of the spine, heel, femoral neck, and radius in healthy postmenopausal women. Bone Miner. 17:87–95.[Medline]
  14. Tomkinson A, Reeve J, Shaw RW, Noble BS. 1997 The death of osteocytes via apoptosis accompanies estrogen withdrawal in human bone. J Clin Endocrinol Metab. 82:3128–3135.[Abstract/Free Full Text]
  15. Genant HK, Lucas J, Weiss S, et al. 1997 Low-dose esterified estrogen therapy. Arch Intern Med. 157:2609–2615.[Abstract]