You say estren, I say estrogen. Let’s call the whole replacement off!

Estrogens and androgens play a key role in regulating bone mass. However, their clinical use as bone anabolic agents is limited due to unwanted side effects, particularly in reproductive organs. In 2002, the synthetic ligand estren was described to reproduce the bone anabolic, nongenotropic effects of sex steroids while having no effect on the uterus or seminal vesicles. But in the current issue of the JCI, Windahl et al. provide data showing that estrens are not as suitable a replacement for estrogen as was initially reported (see the related article beginning on page 2500). Though not catabolic, estrens triggered only minor, nonsignificant increases in bone mass in gonadectomized mice, all the while inducing hypertrophy of reproductive organs. Does this mean estrens should not be pursued as a therapy for osteoporosis?

The estrogen hormone family plays an essential role in the regulation of skeletal growth and homeostasis. While osteoblasts, osteocytes, and osteoclasts can be indirect targets of hormone signaling, they are also direct targets of estrogen and express functional estrogen and androgen receptors (ER and AR, respectively) (1). As estrogen or androgen deficiency can lead to rapid decreases in bone mass, therapies designed to return these sex hormones to their original levels would seem logical. However, these strategies have been fraught with difficulty due to the complex nature of hormone signaling.

In the classical (genomic) model of estrogen signaling, estrogens bind to the ER in the nucleus (Figure 1). Over the course of several hours, the estrogen-ER complex then induces a direct response through estrogen response element sequences or an indirect response by triggering expression of other proteins such as transcription factors of the AP1 family, among others. This is viewed as the main mode of action of estrogen. However, there is also evidence for a rapid, nongenomic response to estrogen. Signaling through the nongenomic pathway can lead to Ca²⁺ and NO release and activation of various kinases. For a more in-depth review of estrogen signaling, see refs. 1 and 2.

Figure 1

Pathways of estrogen, SERM, and estren signaling. In the genomic pathway of estrogen action (i), estrogen or SERMs bind to the ER, regulating transcription of target genes in the nucleus by binding to estrogen response element (ERE) regulatory sequences and by recruiting coregulatory proteins (CoRegs). Estrens were previously thought only to signal through the rapid, nongenomic pathway mediated by the ER located in or adjacent to the plasma membrane (ii), which may require the presence of adaptor proteins, which target the ER to the membrane. Activation of the membrane ER leads to a rapid change in cellular signaling molecules and stimulation of kinase activity, which in turn may affect transcription. Figure and legend adapted from ref. 2.

Agonizing and antagonizing the genomic estrogen signaling pathway is complicated by the fact that ERs are expressed in multiple organs. The selective estrogen receptor modulator (SERM) tamoxifen, as an example, activates the ER in bone and uterus but is an antagonist in the breast. Unfortunately, chronic administration of tamoxifen can lead to uterine cancer, so alternative SERMs and other methods for regulating the ER in specific organs have been sought. So far, all described SERMs have been shown to prevent bone loss, but their effects pale in comparison to the anabolic results seen with estrogen and androgen treatment.

In 2002, Kousteni, Manolagas, and colleagues described a synthetic ligand, 4-estren-3α,17β-diol (estren), that reproduced the nongenomic effects of estrogen (3). In their hands, estren increased bone mass and strength in gonadectomized Swiss Webster mice but had no effects on uterine or seminal vesicle weight. Estrens also had no effect on the proliferation of MCF-7 human breast cancer cells. The effect of estren was attributed to activation of transcription factors by several kinase cascades (4).

The potent effects of estrens on bone strength suggested that they could be used as a bone anabolic agent in cases of estrogen deficiency, such as menopause, given that their anabolic effects were restricted to bone. Estrens are currently in preclinical testing; however, the results of their use in humans have not yet been reported.

Since the original description of estren, a few reports have appeared that question whether estren acts in a nongenomic manner (5) and whether it really has no effects on the uterus (6). In the current issue of the JCI, Windahl, Baron, and colleagues report a systematic comparative analysis of estrens in gonadectomized mice to discern whether estrens are in fact nongenomic, bone anabolic compounds with no effects on reproductive organs (7).

The current study (7) reports that while estrens were able to prevent gonadectomy-induced bone loss, they showed no bone anabolic effects when given at the same doses and in the same manner as originally reported (Table 1). Furthermore, and in direct contrast to the original estren study (3), estren was shown to increase uterine and seminal vesicle weight and enhanced the proliferation of human breast cancer cells — the harmful effects estrens were designed to avoid.

Table 1

Comparison of estren studies

Windahl et al. (7) show that estrens bind more strongly to the AR than the ER and suggest that they act more as androgens than estrogens. This was confirmed when the authors treated the gonadectomized animals with estrens and anti-androgens or anti-estrogens, as well as in ERα-KO mice: the addition of anti-androgen completely blocked the response to estren, and removal of estrogen only partially blocked estren’s effects. In agreement with earlier reports (5), estren was shown to have transcriptional activity — suggesting that estrens could potentially exert their effects through the genomic pathway in addition to the nongenomic one.

There are a few differences in the studies that may lead us to believe that one or the other is more reliable (Table 1). First, the earlier estren study used Swiss Webster mice, while the current authors used C57BL/6 mice (but used Swiss Webster mice when comparing the effects of estrens on uterine weight). Could strain effects account for the differences seen? The real question is whether estrens are bone specific in humans, which remains to be seen.

Also, the age of the animals studied was different — the current study used mice 3–5 months younger than those in the original study, but the authors note that Moverare et al. reported uterotrophic effects of estren in 11-month-old mice at the same doses used here (5). Despite the age difference, the fact that estrens could have an effect on reproductive organs at any stage of life raises serious concerns about their use as a SERM.

Data in the current article (7) show that when the dose of estren was reduced, the adverse effects on reproductive organs disappeared, but, unfortunately, so did the associated bone preservation capacity. Together, the data from the current study and others in the literature make a compelling case that estrens are not suitable for treatment of osteoporosis.

Given that estren may not be the ideal SERM for treating osteoporosis, the search continues for what could become a blockbuster drug. Some of the authors of the current study have also attempted to enter the fray by testing a new SERM, PSK3471 (7).

The miracle SERM for osteoporosis may be out there somewhere, but it has not been found yet. Perhaps it is PSK3471; perhaps it may still turn out to be estren — results from clinical testing in humans will provide the definitive proof. But until then, the search must continue.

The author wishes to thank members of the bone research community for guidance in preparing this commentary.

Address correspondence to: Ushma S. Neill, Journal of Clinical Investigation, 630 West 168th Street, Box 57A, New York, New York 10032, USA. Phone: (212) 342-0497; Fax: (212) 342-0499; E-mail: editors@the-jci.org .

Nonstandard abbreviations used: AR, androgen receptor; ER, estrogen receptor; SERM, selective estrogen receptor modulator.

Conflict of interest: The author has declared that no conflict of interest exists.

Reference information: J. Clin. Invest.116:2327–2329 (2006). doi:10.1172/JCI29733.

See the related article beginning on page 2500.

1. Weitzmann, M.N., Pacifici, R. 2006. Estrogen deficiency and bone loss: an inflammatory tale. J. Clin. Invest. 116:1186-1194.
   View this article via: JCI CrossRef PubMed Google Scholar
2. Deroo, B.J., Korach, K.S. 2006. Estrogen receptors and human disease. J. Clin. Invest. 116:561-570.
   View this article via: JCI CrossRef PubMed Google Scholar
3. Kousteni, S., et al. 2002. Reversal of bone loss in mice by nongenotropic signaling of sex steroids. Science. 298:843-846.
   View this article via: PubMed CrossRef Google Scholar
4. Kousteni, S., et al. 2003. Kinase-mediated regulation of common transcription factors accounts for the bone-protective effects of sex steroids. J. Clin. Invest. 111:1651-1664.
   View this article via: JCI CrossRef PubMed Google Scholar
5. Moverare, S., et al. 2003. Estren is a selective estrogen receptor modulator with transcriptional activity. Mol. Pharmacol. 64:1428-1433.
   View this article via: PubMed CrossRef Google Scholar
6. Hewitt, S.C., Collins, J., Grissom, S., Hamilton, K., Korach, K.S. 2006. Estren behaves as a weak estrogen rather than a nongenomic selective activator in the mouse uterus. Endocrinology. 147:2203-2214.
   View this article via: PubMed CrossRef Google Scholar
7. Windahl, S.H., et al. 2006. Bone protection by estrens occurs through non–tissue-selective activation of the androgen receptor. J. Clin. Invest. 116:2500-2509.
   View this article via: JCI CrossRef PubMed Google Scholar