help button home button Endocrine Society JCEM
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Khosla, S.
Right arrow Articles by Brandi, M. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Khosla, S.
Right arrow Articles by Brandi, M. L.
The Journal of Clinical Endocrinology & Metabolism Vol. 89, No. 4 1808-1816
Copyright © 2004 by The Endocrine Society

Relationship of Estrogen Receptor Genotypes to Bone Mineral Density and to Rates of Bone Loss in Men

Sundeep Khosla, B. Lawrence Riggs, Elizabeth J. Atkinson, Ann L. Oberg, Carmelo Mavilia, Francesca Del Monte, L. Joseph Melton, III and Maria Luisa Brandi

Endocrine Research Unit (S.K., B.L.R.), Division of Endocrinology, Metabolism, and Nutrition, Department of Internal Medicine and the Department of Health Sciences Research (E.J.A., A.L.O., L.J.M.), Mayo Clinic College of Medicine, Rochester, Minnesota 55905; and Metabolic Diseases Unit (C.M., F.D.M., M.L.B.), Department of Internal Medicine, University of Florence, 50139 Florence, Italy

Address all correspondence and requests for reprints to: Sundeep Khosla, M.D., Mayo Clinic, 200 First Street SW, 5-194 Joseph, Rochester, Minnesota 55905. E-mail: khosla.sundeep{at}mayo.edu.


   Abstract
Top
 Abstract
Introduction
 Subjects and Methods
 Results
 Discussion
 References
 
Estrogen is now known to play an important role in bone metabolism in men. Thus, we examined possible relationships between polymorphisms of the estrogen receptor (ER)-{alpha} and -ß genes, bone mineral density (BMD), and rates of bone loss in an age-stratified random sample of 283 Rochester, Minnesota, men aged 22–90 yr. DNA was analyzed for the XbaI and PvuII ER-{alpha} and AluI ER-ß polymorphisms. The X/P and x/p alleles of the ER-{alpha} gene were in strong linkage disequilibrium. BMD at multiple sites did not differ as a function of either the ER-{alpha} or -ß genotype. However, the ER-{alpha} (but not ER-ß) genotypes did modulate the previously described relationships between BMD or rates of bone loss and bioavailable estradiol (E2) levels in these men. At the femoral neck, BMD was associated with bioavailable E2 levels in men with the XX (R = 0.66) or PP (R = 0.51) genotypes (P < 0.001 for both) but not in men with the xx (R = 0.15; P = 0.188) or pp (R = 0.12; P = 0.356) genotypes. The interactions between bioavailable E2 levels and the XbaI and PvuII genotypes were significant at the P < 0.001 and P < 0.009 levels, respectively. Moreover, rates of bone loss at the midradius in men aged 60–90 yr were modestly correlated with serum bioavailable E2 levels in subjects with the X (R = 0.47) or P (R = 0.42) alleles (P < 0.001 for both) but not in those with the xx (R = 0.15; P = 0.430) or pp (R = 0.21; P = 0.372) genotypes. The overall effect of genotype on midradius rate of bone loss was clearly significant for the XbaI polymorphism (P = 0.009) when bioavailable E2 levels were low (<40 pmol/liter) but not for the PvuII polymorphism. These data thus indicate that the ER-{alpha} genotype may modulate the relationship between BMD or rates of bone loss and estrogen levels in men and that bone mass in men with the X or P alleles may be more susceptible to the consequences of estrogen deficiency (and conversely, benefit most from estrogen sufficiency) than in men with the xx or pp genotypes.


   Introduction
Top
 Abstract
 Introduction
Subjects and Methods
 Results
 Discussion
 References
 
ACCUMULATING EVIDENCE NOW indicates that estrogen plays a major role in bone metabolism in men (1). Thus, homozygous mutations in either the estrogen receptor (ER)-{alpha} (2) or aromatase (3, 4, 5, 6) genes result in unfused epiphyses, increased bone remodeling, and impaired acquisition of bone mass during adolescence. In population studies, serum estrogen levels are correlated with bone mineral density (BMD) (7, 8, 9, 10, 11, 12, 13) and with rates of bone loss (14, 15) in aging men. Moreover, direct interventional studies have established that estrogen plays a significant, and perhaps dominant, role in regulating bone resorption in men (16, 17, 18), with both estrogen and testosterone contributing to the maintenance of bone formation (16, 17). In addition, previous work from our group (14) has suggested that older men with non-SHBG bound (or bioavailable) estradiol (E2) levels less than 40 pmol/liter (11 pg/ml) have higher rates of bone loss (at least at the radius and ulna) than men with bioavailable E2 levels above this value. Collectively, these findings are consistent with the hypothesis that the observed decrease in bioavailable E2 levels with aging (due principally to an age-related increase in SHBG levels) may contribute substantially to age-related bone loss in men (1, 14, 15).

The skeletal effects of estrogen are mediated by ERs that are present in osteoblasts (19, 20) and in osteoclasts (21, 22). Both ER-{alpha} and ER-ß are expressed by bone cells in vitro (23) and in vivo (24), and studies using mouse knockout models indicate that, whereas ER-{alpha} appears to be the major receptor mediating estrogen action in bone, ER-ß (at least in females) can modulate ER-{alpha} action (25). After binding and activation of the ER, estrogen regulates the production of a number of growth factors and cytokines, has direct effects on osteoblast differentiation (including type I collagen production) and apoptosis, and also regulates osteoclast development, activity, and apoptosis (for review, see Ref. 26).

Given the importance of estrogen signaling in bone, there have been numerous studies examining the effects of polymorphisms in the ER-{alpha} (27, 28, 29, 30, 31, 32, 33) or ER-ß (34, 35) genes on BMD or rates of bone loss in women. For ER-{alpha}, these studies have generally used digestion of genomic DNA with the PvuII or XbaI restriction enzymes, which have polymorphic sites in the first intron of the ER-{alpha} gene (36). Absence of the restriction site results in a larger DNA fragment (the P or X alleles, respectively), whereas presence of the restriction site results in the appearance of a smaller DNA fragment (the p or x allele, respectively). In contrast to the studies in women (27, 28, 29, 30, 31, 32, 33, 34, 35), however, data on the relationships between ER polymorphisms and BMD in men are more limited, although Ongphiphadhanakul et al. (10) did find that the presence of the PvuII P allele (i.e. PP or Pp genotype) was associated with higher BMD at the anteroposterior (AP) spine and distal radius in 81 Thai men. In addition, Sapir-Koren et al. (37) used microdensitometry of hand x-rays to compare cancellous bone at the third finger in a group of 183 Caucasian men and also found that subjects with the PvuII P allele and the XbaI X allele (PXPX genotype) had higher BMD Z-scores compared with subjects lacking the PX haplotype. In the present study, we used a somewhat different approach to assess the potential role of ER genotypes in bone metabolism in men than has previously been attempted. We hypothesized that variations in ER ({alpha} or ß) genotype may alter the relationship (i.e. the correlation) between bone mass and circulating bioavailable E2 levels in men. Thus, rather than simply comparing BMD in men with differing ER-{alpha} or ER-ß genotypes, we tested whether the ER genotypes modulated the previously established correlations of BMD (7, 8, 9, 10, 11, 12, 13) or rates of bone loss (14, 15) with circulating E2 levels in men.


   Subjects and Methods
Top
 Abstract
 Introduction
 Subjects and Methods
Results
 Discussion
 References
 
Study subjects

Subjects were recruited from an age-stratified random sample of Rochester, Minnesota, men who were selected using the medical records linkage system of the Rochester Epidemiology Project (38). Over half of the Rochester population is identified annually in this system, and the majority are seen in any 3-yr period. Thus, the enumerated population approximates the underlying population of the community, including both free-living and institutionalized individuals. A total of 1138 men aged 20 yr and over were approached, but 239 men were ineligible (109 were debilitated due to dementia and could not give informed consent, 13 were radiation workers, 91 died before contact, 25 were debilitated due to terminal cancer, and one was unable to participate due to pending legal action). Of the 899 eligible men, 348 participated and provided full study data. For the present study, we analyzed data from the 286 men who returned for more than one visit (allowing us to assess rates of bone loss) and who consented to providing DNA samples for analysis. Of these 286 men, an additional three were excluded because of inexplicably high (into the range of premenopausal women) bioavailable E2 levels. All but nine of the 283 men were Caucasian, reflecting the ethnic composition of the population (96% white in 1990). The men ranged in age from 22–90 yr. For the analyses, we divided the men into young (age, 22–39 yr; mean ± SD, 31.1 ± 4.9 yr; n = 78), middle-aged (40–59 yr; 50.0 ± 5.5 yr; n = 86), and older (60–90 yr; 73.9 ± 8.6 yr; n = 119) groups.

Study protocol

BMD (grams per square centimeter) was determined for the lumbar spine (L2–L4), femoral neck, and middistal radius using dual-energy x-ray absorptiometry with the Hologic QDR2000 instrument (Hologic, Waltham, MA) using software version 5.40. Because we did not specifically exclude subjects with spinal osteoarthritis or aortic calcification, which can confound the BMD measurement (39), we assessed the midlateral instead of the AP spine, which largely excludes these confounders from the scanning field. The coefficients of variation (CVs) for the lateral spine, femoral neck, and radius were 2.1, 1.5, and 1.7%, respectively. BMD was measured at baseline, 2 yr, and 4 yr, and annualized rates of change were calculated.

Fasting-state serum samples were obtained between 0800 and 0900 h, and a 24-h urine collection was turned in. All samples were collected upon entry into the study and stored at –70 C until analyzed.

Laboratory methods

All of the assays were performed in the Mayo Immunochemical Core Laboratory. Fasting serum samples were assayed by RIA for total E2 (Diagnostic Products Corporation, Los Angeles, CA; interassay CV < 16%) as well as the non-SHBG bound (bioavailable) fraction of E2, which was measured using a modification of the technique of O’Connor et al. (40) and Tremblay and Dube (41), as previously described (9).

Bone formation was assessed by serum osteocalcin measured by RIA using antibody G12 (interassay CV < 6%) (42), by serum bone-specific alkaline phosphatase (BSAP) measured by ELISA (Metra Biosystems, San Diego, CA; interassay CV < 11%), and by the carboxyterminal propeptide of type I procollagen (PICP) measured by ELISA (Prolagen-C, Metra Biosystems; interassay CV < 7%). Bone resorption was evaluated by measurement of serum levels of the N-telopeptide of type I collagen (NTx) using an ELISA kit (Osteomark NTx Serum, Ostex Inc., Seattle, WA; interassay CV < 17%) as well as by measurement of 24-h urine levels of NTx, assessed as nanomoles/liter glomerular filtrate, also using an ELISA kit (Osteomark, Ostex Inc.; interassay CV = 10%). The glomerular filtration rate was assessed by creatinine clearance.

Analysis of ER-{alpha} and -ß genotypes

DNA was extracted from EDTA blood samples with the Qiagen DNA mini Kit (Qiagen, Hilden, Germany), and 0.1 µg of DNA was amplified in 50 µl of buffer solution [10 mM Tris-HCl (pH 9), 50 mM KCl, 5 mM MgCl2, 1% Triton X-100, and 200 µM each of the four deoxyribonucleotides]. To analyze ER-{alpha} gene polymorphisms, 1 U of Taq polymerase (Promega Corp., Madison, WI) and 0.4 µM of oligonucleotide primers (forward, 5'-CTGCCACCCTATCTGTATCTTTTCCTATTCTCC-3'; reverse, 5'TCTTTCTCTGCCACCCTGGCGTCGATTATCTGA-3') were used for each reaction (29). PCR was performed through 30 cycles by the following steps: denaturation at 94 C for 30 sec, annealing at 55 C for 1 min, and extension at 72 C for 90 sec. The PCR product was a 1.3-kb fragment of intron 1 and exon 2 in the ER-{alpha} gene. After amplification, the PCR product was digested with 10 U of either PvuII or XbaI restriction endonucleases (Roche Molecular Biochemicals, Indianapolis, IN) and was electrophoresed in a 2.0% agarose gel. For the 1730 (A-G) polymorphism of the 3'-untranslated region in the ER-ß gene (43), we used the following oligonucleotide primers: forward, 5'-CAAGTCCATCACGGGGT-3'; reverse, 5'-AGATGAACCCAGGCTGGTG-3', also at 0.4 µM. PCR was performed using 0.1 µg of DNA in 50 µl of buffer solution (10 mM Tris-HCl, 50 mM KCl, 2.5 mM MgCl2, 1% Triton X-100, and 200 µM deoxyribonucleotides) and 1 U of Taq polymerase through 35 cycles by the following steps: denaturation at 94 C for 30 sec, annealing at 60 C for 30 sec, and extension at 72 C for 90 sec. The PCR product was a 168-bp fragment, which was then digested with 20 U of AluI endonuclease enzyme (Promega Corp.). If 1730A was present, it resulted in an AluI restriction site, and PCR digestion generated two bands of 124 and 44 bp, respectively, on 3.5% agarose gel electrophoresis. The presence of the restriction site for each endonuclease was indicated with a lowercase letter (p, x, or a, respectively, for PvuII, XbaI, and AluI endonucleases), whereas an uppercase letter (P, X, or A) indicated the absence of the restriction site. The subjects were classified as pp, xx, and aa homozygotes; Pp, Xx, and Aa heterozygotes; or PP, XX, and AA homozygotes, according to the digestion pattern.

Statistical analyses

The relationship between bioavailable E2 levels and BMD was studied using linear regression analysis and the Pearson correlation coefficient. BMD at the femoral neck, midradius, and lateral spine was modeled as the dependent variable, whereas bioavailable E2, genotype, and the interaction between the two were treated as the independent variables. In those models using all the men, genotype was modeled using an additive assumption. In those models using only men more than 60 yr of age, the genotypes were modeled as carrier status (no copy of X or P vs. at least one copy). Linear regression models were used to help determine the best split points between rates of bone loss and bioavailable E2 levels, where the split point was determined based on the maximum model R2 value. In these models, where slopes were allowed to vary before and after the split point, we examined the interaction between the slopes and carrier status. The overall interactions were tested with a two-degree-of-freedom F test. Haplotypes were estimated using an expectation-maximization algorithm. Linkage disequilibrium was tested using the likelihood test. P < 0.05 was considered significant for all the analyses.


   Results
Top
 Abstract
 Introduction
 Subjects and Methods
 Results
Discussion
 References
 
Table 1Go shows the distribution of subjects based on the PvuII and XbaI ER-{alpha} and AluI ER-ß genotypes. The XbaI and PvuII genotypes were in strong linkage disequilibrium (D' = 0.87), with the majority of the haplotypes being XP (41%) or xp (48%). Also shown in Table 1Go are key anthropometric, BMD, and sex steroid values by genotype. As is evident, in the overall group of 283 men, neither BMD at any site nor total or bioavailable E2 levels differed by genotype. The only significant difference between groups was for body mass index (BMI) based on the XbaI polymorphism, with the XX individuals having a slightly higher BMI than the Xx or xx subjects.


View this table:
[in this window]
[in a new window]
  		TABLE 1. ER genotype and baseline characteristics of 283 Rochester, Minnesota, men according to genotype [XbaI (XX, Xx, and xx) and PvuII (PP, Pp, and pp) for ER-{alpha} and AluI (AA, Aa, and aa) for ER-ß]

 
Although BMD did not differ by genotype, we reasoned that the ER genotypes could be modulating the relationship between E2 levels and BMD in these men. This was, indeed, the case, as demonstrated for femoral neck BMD as a function of bioavailable E2 levels segregated by the XbaI (Fig. 1Go) and PvuII (Fig. 2Go) genotypes. As is evident in Fig. 1Go, there was a gene dosage effect for the XbaI polymorphism: the subjects with the XX genotype showed the strongest correlations between femoral neck BMD and bioavailable E2 levels; the ones with the Xx genotype demonstrated a weaker relationship; and those with the xx genotype showed no relationship (P < 0.001 for the interaction effect of genotype and bioavailable E2 levels). For the PvuII polymorphism, correlations between femoral neck BMD and bioavailable E2 levels were similar in the subjects with the PP and Pp genotypes, whereas there was no association between these variables in those with the pp genotype (P = 0.009 for the interaction effect of genotype and bioavailable E2 levels).



View larger version (21K):
[in this window]
[in a new window]
  		FIG. 1. Femoral neck BMD as a function of bioavailable E2 levels by XbaI ER-{alpha} genotype (A, XX; B, Xx; and C, xx) in 283 Rochester, Minnesota, men aged 22–90 yr. P < 0.001 for the interaction effect of genotype and bioavailable E2 levels.

 


View larger version (21K):
[in this window]
[in a new window]
  		FIG. 2. Femoral neck BMD as a function of bioavailable E2 levels by PvuII ER-{alpha} genotype (A, PP; B, Pp; and C, pp) in 283 Rochester, Minnesota, men aged 22–90 yr. P = 0.009 for the interaction effect of genotype and bioavailable E2 levels.

 
The comparable correlation coefficients relating midradius and lateral spine BMD to bioavailable E2 levels by PvuII and XbaI genotypes are shown in Table 2Go. As is evident, a similar pattern, although not statistically significant for the interaction effect of genotype and bioavailable E2 levels, was seen at the midradius, with the XX or PP and the heterozygote subjects showing the strongest relationships between midradius BMD and bioavailable E2 levels, and the xx or pp individuals showing weaker relationships. By contrast, no consistent pattern across genotypes was evident for the lateral spine. Moreover, there was no effect of the ER-ß genotype on any of these relationships in the men (data not shown).


View this table:
[in this window]
[in a new window]
  		TABLE 2. Pearson correlation coefficients between midradius and lateral spine BMD and bioavailable E2 levels as a function of XbaI and PvuII ER-{alpha} genotypes in 283 Rochester, Minnesota, men aged 22–90 yr

 
We next tested whether the effects of the XbaI and PvuII genotypes on cross-sectional BMD that we found in the entire cohort of men were also present for rates of bone loss in the older men (n = 119). In the overall older group, mean annualized rates of bone loss at the midradius, lateral spine, and femoral neck (P values are for difference from zero) were –0.67%/yr (P < 0.001), –0.34%/yr (P = 0.182), and 0.09%/yr (P = 0.639), respectively. Table 3Go shows the baseline characteristics, bone turnover markers, and rates of bone loss at the various sites in the older men by genotype. As is evident, there were no significant differences in these parameters in this group of older men as a function of the XbaI genotype; however, for the PvuII genotype, serum PICP as well as serum and urine NTx did show a significant trend across genotypes, with subjects with the pp genotype having the highest values. Because, in this relatively small group of older men, the rate of bone loss was significantly different from zero for the overall group only at the midradius, we focused on this site for the longitudinal analysis.


View this table:
[in this window]
[in a new window]
  		TABLE 3. ER-{alpha} genotype, baseline characteristics, and rates of bone loss in 119 Rochester, Minnesota, men aged 60–90 yr, according to genotype [XbaI (XX, Xx, and xx) and PvuII (PP, Pp, and pp)]

 
As shown in Figs. 3Go and 4Go for the XbaI and PvuII genotypes, respectively, the rate of change in midradius BMD in the older men was clearly related to bioavailable E2 levels for men with the XX (Fig. 3Go) or PP (Fig. 4Go) genotypes and for the heterozygote subjects, whereas there was no significant relationship between midradius BMD change and bioavailable E2 levels for men with the xx (Fig. 3Go) or pp (Fig. 4Go) genotypes. However, probably due to the relatively small number of subjects, the overall effect of genotype on the relationship between the midradius rates of bone loss and bioavailable E2 levels was not significant (P = 0.136 and 0.326 for the XbaI and PvuII polymorphisms, respectively).



View larger version (16K):
[in this window]
[in a new window]
  		FIG. 3. Rate of change in midradius BMD as a function of bioavailable E2 levels by XbaI genotype (A, XX; B, Xx; and C, xx) in 119 older Rochester, Minnesota, men aged 60–90 yr.

 


View larger version (17K):
[in this window]
[in a new window]
  		FIG. 4. Rate of change in midradius BMD as a function of bioavailable E2 levels by PvuII genotype (A, PP; B, Pp; and C, pp) in 119 older Rochester, Minnesota, men aged 60–90 yr.

 
The possible interaction between ER-{alpha} genotype and rates of bone loss in the older men was explored further by comparing the subjects with the X or P alleles (i.e. XX, Xx, PP, or Pp) with those without these alleles (xx or pp). In this analysis, we also used the split point model we have used previously (14), allowing for the slope to change at 40 pmol/liter, and tested for an interaction between the presence of the X or the P allele and bioavailable E2 levels for effects on midradius rates of bone loss. Overall, the interaction between the X allele and bioavailable E2 was significant (P = 0.03), driven mainly by the different slopes of the relationship between the rate of bone loss and bioavailable E2 levels at low bioavailable E2 levels in the +X (i.e. XX or Xx) vs. the xx subjects (P = 0.009 for the interaction effect of genotype and bioavailable E2 levels for bioavailable E2 < 40 pmol/liter). By contrast, the overall interaction between the P allele and bioavailable E2 was not significant (P = 0.613).

To examine this issue using a slightly different approach, in Fig. 5Go, we show rates of midradius bone loss in men with the X or P alleles vs. those with the xx or pp genotypes, stratified by bioavailable E2 levels above or below 40 pmol/liter. As is evident, for older men with the X or P alleles, the rate of bone loss was significantly lower when bioavailable E2 levels were above 40 pmol/liter compared with the men with these alleles and bioavailable E2 levels below 40 pmol/liter; by contrast, rates of bone loss did not differ in men with high or low bioavailable E2 levels if they had the xx or pp genotypes. The more rigorous statistical analysis testing for an interaction between the ER-{alpha} genotype and bioavailable E2 levels above or below 40 pmol/liter was significant for the XbaI genotype (P < 0.05) but not for the PvuII genotype (P = 0.554).



View larger version (17K):
[in this window]
[in a new window]
  		FIG. 5. Rate of change in midradius BMD stratified by bioavailable E2 levels greater than 40 pmol/liter (open bars) or less than 40 pmol/liter (solid bars) in 119 older Rochester, Minnesota, men aged 60–90 yr as a function of XbaI (A) or PvuII (B) genotypes. *, P < 0.05; and **, P < 0.01 for comparison of rates vs. bioavailable E2 levels less than 40 pmol/liter.

 
Collectively, the above data indicated that both the BMD at femoral neck and midradius and the rates of bone loss depended on E2 levels in men with the X or P alleles to a greater extent than in men with the xx or pp genotypes. To explore the possible mechanism for this, we tested whether there was a difference between genotypes in the relationship of bone turnover markers and bioavailable E2 levels in the older men. As is evident from Table 4Go, in the overall group of older men and in the group segregated by genotype, there were either weak inverse associations or no relationship between the bone turnover markers and serum bioavailable E2 levels. The exception to this was the relationship between serum PICP and bioavailable E2 levels in men with the XX or PP genotypes: in contrast to all of the other associations, serum PICP levels were, in fact, positively associated with bioavailable E2 levels in men with these genotypes.


View this table:
[in this window]
[in a new window]
  		TABLE 4. Pearson correlation coefficients between bone turnover markers and bioavailable E2 levels in the overall group of 119 Rochester, Minnesota, men aged 60–90 yr and segregated by genotype

 

   Discussion
Top
 Abstract
 Introduction
 Subjects and Methods
 Results
 Discussion
References
 
Based on the accumulating evidence for an important role for estrogen in regulating bone metabolism in men, we sought in the present study to examine the effect of polymorphisms in the ER-{alpha} and ER-ß genes on BMD in men. Our findings indicate that the ER-{alpha} genotype may modulate the relationship between BMD or rates of bone loss and bioavailable E2 levels in men. Thus, BMD at the femoral neck and midradius and rates of bone loss at the midradius depended much more on circulating E2 levels in men with the X or P alleles than they did in men with the xx or pp genotypes. As such, bone mass in men with the X or P alleles may be more susceptible to the consequences of estrogen deficiency (and conversely, benefit most from estrogen sufficiency) than bone mass in men with the xx or pp genotypes.

Similar to our findings in men, Salmen et al. (27) have reported that in postmenopausal women, the reductions in lumbar spine rates of bone loss and in overall fracture risk after estrogen treatment (28) were greater in women with the P allele than in women with the pp genotype. Moreover, Ongphiphadhanakul et al. (32) found that low-dose estrogen treatment (0.3 mg conjugated equine estrogens) increased lumbar spine BMD to a greater extent in women with the P allele compared with those with the pp genotype. Collectively, our data in men and these previous findings in women are consistent with the hypothesis that the pp or xx genotypes may be relatively hormone-insensitive genotypes and that subjects with the P or X alleles may benefit more from the protective effects of estrogen on bone than subjects with the pp or xx genotypes. However, there may be significant interactions between the ER genotype and other genes [such as the vitamin D receptor genotype (31, 44)], and the above formulation may well be somewhat of an oversimplification, albeit a useful one on which to base further studies.

Of note, our finding that bone mass in men with the PvuII P or XbaI X alleles is more dependent on circulating estrogen levels than it is in men with the pp or xx genotypes is also consistent with the work of Herrington et al. (36), who found that postmenopausal women who have the ER-{alpha} IVS-401 C/C (or IVS-354 G/G) genotypes have an augmented response of high-density lipoprotein cholesterol or serum SHBG levels to hormone replacement therapy when compared with women with the ER-{alpha} IVS-401 C/T or T/T genotypes (or IVS-354 G/A or A/A genotypes). The IVS-401 C/C and IVS-354 G/G genotypes in the study of Herrington et al. (36) correspond, in fact, to the PvuII PP and XbaI XX genotypes, respectively. Thus, the PvuII P and XbaI X alleles appear to confer greater sensitivity to estrogen not only in bone but also in the liver (36), raising the possibility that these alleles are associated with increased sensitivity to estrogen in multiple tissues.

The polymorphisms in the ER-{alpha} gene that we and others (10, 27, 28, 29, 30, 31, 32, 33, 36) have examined are in the first intron of the gene, and how these variations may alter the response to estrogen remains to be determined. Herrington et al. (45) have noted that the IVS-401 T->C transition associated with loss of the PvuII site (and the production of the P allele) results in a potential binding site for myb transcription factors. Moreover, in vitro coexpression of an ER-{alpha} IVS-401 C allele/luciferase construct with a myb expression vector produced a more than 10-fold increase in luciferase activity compared with an only 2.5-fold increase observed in cells transfected with the IVS-401 T allele/luciferase construct (corresponding to the PvuII p allele) (45). Thus, as suggested by Herrington et al. (45), because expression of the B-myb gene is also induced by estrogen (46), it is possible that in cell types that commonly express B-myb or related transcription factors, estrogen-induced increases in myb expression lead, in turn, to increased ER-{alpha} levels in individuals with the IVS-401 C (PvuII P) allele. However, alternate explanations are also possible. Thus, in addition to directly altering the transcription of the ER-{alpha} gene (i.e. through altered myb binding), the polymorphisms in the ER-{alpha} gene being examined could be linked to other sequence variations that 1) affect the level of expression of the ER-{alpha} gene through other transcriptional regulators; 2) lie in the coding region of the receptor leading to an alteration in ER-{alpha} protein function; or, less likely, 3) are in linkage disequilibrium with another, unidentified gene adjacent to the ER-{alpha} gene that modulates the relationship between circulating estrogen levels and the skeletal response to estrogen.

In contrast to the findings at the femoral neck and forearm, we did not note any consistent effect of genotype on the relationship between serum bioavailable E2 levels and lateral lumbar spine BMD. The reasons for this are unclear. We used the lateral instead of AP spine BMD due to concerns about osteoarthritis of the spine in the middle-aged and older men (39). This measurement does have somewhat greater variability than the femoral neck and forearm measurements (CV of 2.1% vs. 1.5 and 1.7% for the femoral neck and forearm, respectively), perhaps making it more difficult to see possible relationships. Alternatively, there may be biological reasons for the lack of an effect of ER-{alpha} genotype at the lumbar spine. For example, it appears that cancellous bone (which predominates in the spine) contains both ER-{alpha} and -ß, whereas cortical bone (which predominates at the femoral neck and forearm) contains largely ER-{alpha} (23, 24). Thus, effects of ER-{alpha} genotype may be most evident at sites rich in cortical as opposed to cancellous bone. However, if that were the case, the findings should have been even more robust for midradius BMD compared with femoral neck BMD (because there is even more cortical bone at the midradius vs. the femoral neck), and this was not the case (Table 2Go). Clearly, further work is needed to address this issue.

Although we found that the ER-{alpha} gene polymorphisms appeared to modulate the relationship between BMD or forearm rates of bone loss and serum bioavailable E2 levels in men, we did not detect any clear differences in BMD across genotypes. This is in contrast to the two previous studies in men noted earlier that did find that men with the P allele had higher AP spine and distal radius BMD (10) or hand x-ray microdensitometry Z-scores (37) than men with the pp genotype. The reasons for this discrepancy are unclear but may relate to possible differences in serum bioavailable E2 levels in the particular populations studied. Thus, for example, based on the data in Figs. 1Go and 2Go, at relatively high serum bioavailable E2 levels, subjects with the XX or PP genotypes could well have higher femoral neck BMD values than subjects with the xx or pp genotypes. By contrast, this trend would be reversed at low bioavailable E2 levels. In our study, which included a relatively broad range of bioavailable E2 levels, these opposing trends may have essentially canceled each other, resulting in the lack of overall difference in BMD values across genotypes.

Although our study had limited power to do so, we also attempted to examine the possible mechanism for the greater dependence of BMD in men with the X or P alleles on circulating estrogen levels. The only clue to this came from the observation that serum PICP levels were positively associated with circulating bioavailable E2 levels in older men with the XX or PP genotypes but not in those with the other genotypes. Although this association could be due to chance, it does suggest the possibility that the X or P alleles may confer increased dependence of type I collagen production by osteoblastic cells on circulating estrogen levels. Although estrogen has been shown to stimulate type I collagen production by osteoblastic cells in vitro (47, 48), further studies are needed to test whether this effect is modulated by the ER-{alpha} genotype.

In contrast to the effects of ER-{alpha} genotype, we did not find any associations between the ER-ß genotype and BMD in men, and there were also no significant interactions between the ER-ß genotype and serum bioavailable E2 levels for effects on BMD or rates of bone loss in men. Thus, at least in men, the AluI ER-ß polymorphism did not to have a measurable impact on bone metabolism in this study.

In summary, our findings demonstrate that the ER-{alpha} genotype modulates the relationship between BMD or rates of bone loss and bioavailable E2 levels in men. Both parameters are more dependent on circulating estrogen levels in men carrying the ER-{alpha} XbaI X or PvuII P alleles compared with those with the xx or pp genotypes. These data are consistent with data from previous studies indicating that the reductions in rates of bone loss or in fracture risk after estrogen therapy are greater in postmenopausal women with the ER-{alpha} P allele compared with those with the pp genotype (27, 28, 32). In addition, our findings are also consistent with the data of Herrington et al. (36) demonstrating augmented responses of circulating high-density lipoprotein cholesterol and SHBG levels to estrogen in women with the PP or XX genotypes. Collectively, these data raise the possibility, which warrants further study, that specific variants of the ER-{alpha} gene result in enhanced sensitivity to estrogen in multiple tissues, including bone, in both women and in men.


   Acknowledgments
 
We thank Ms. Vickie Gathje and Ms. Joan Muhs for their help in recruiting and studying the subjects, Ms. Kelley Hoey for sampling handling, and Ms. Sara Achenbach for help with statistical analyses.


   Footnotes
 
This work was supported by Grants AR27065 and M01-RR00585 from the National Institutes of Health and by the GENOMOS Project of the European Community.

Abbreviations: AP, Anteroposterior; BMD, bone mineral density; BMI, body mass index; BSAP, bone-specific alkaline phosphatase; CV, coefficient of variation; E2, estradiol; ER, estrogen receptor; NTx, N-telopeptide of type I collagen; PICP, carboxy-terminal propeptide of type I procollagen.

Received August 19, 2003.

Accepted December 17, 2003.


   References
Top
 Abstract
 Introduction
 Subjects and Methods
 Results
 Discussion
 References
 

  1. Khosla S, Melton III LJ, Riggs BL 2002 Clinical review 144: estrogen and the male skeleton. J Clin Endocrinol Metab 87:1443–1450[Abstract/Free Full Text]
  2. Smith EP, Boyd J, Frank GR, Takahashi H, Cohen RM, Specker B, Williams TC, Lubahn DB, Korach KS 1994 Estrogen resistance caused by a mutation in the estrogen-receptor gene in a man. N Engl J Med 331:1056–1061[Abstract/Free Full Text]
  3. Morishima A, Grumbach MM, Simpson ER, Fisher C, Qin K 1995 Aromatase deficiency in male and female siblings caused by a novel mutation and the physiological role of estrogens. J Clin Endocrinol Metab 80:3689–3698[Abstract]
  4. Carani C, Simoni M, Faustini-Fustini M, Serpente S, Korach KS, Simpson ER 1997 Effect of testosterone and estradiol in a man with aromatase deficiency. N Engl J Med 337:91–95[Free Full Text]
  5. Bilezikian JP, Morishima A, Bell J, Grumbach MM 1998 Increased bone mass as a result of estrogen therapy in a man with aromatase deficiency. N Engl J Med 339:599–603[Free Full Text]
  6. Rochira V, Faustini-Fustini M, Balestrieri A, Carani C 2000 Estrogen replacement therapy in a man with congenital aromatase deficiency: effects of different doses of transdermal estradiol on bone mineral density and hormonal parameters. J Clin Endocrinol Metab 85:1841–1845[Abstract/Free Full Text]
  7. 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[Medline]
  8. Greendale GA, Edelstein S, Barrett-Connor E 1997 Endogenous sex steroids and bone mineral density in older women and men: the Rancho Bernardo Study. J Bone Miner Res 12:1833–1843[CrossRef][Medline]
  9. Khosla S, Melton III LJ, Atkinson EJ, O’Fallon WM, Klee GG, Riggs BL 1998 Relationship of serum sex steroid levels and bone turnover markers with bone mineral density in men and women: a key role for bioavailable estrogen. J Clin Endocrinol Metab 83:2266–2274[Abstract/Free Full Text]
  10. Ongphiphadhanakul B, Rajatanavin R, Chanprasertyothin S, Piaseu N, Chailurkit L 1998 Serum oestradiol and oestrogen-receptor gene polymorphism are associated with bone mineral density independently of serum testosterone in normal males. Clin Endocrinol (Oxf) 49:803–809[CrossRef][Medline]
  11. Center JR, Nguyen TV, Sambrook PN, Eisman JA 1999 Hormonal and biochemical parameters in the determination of osteoporosis in elderly men. J Clin Endocrinol Metab 84:3626–3635[Abstract/Free Full Text]
  12. Amin S, Zhang Y, Sawin CT, Evans SR, Hannan MT, Kiel DP, Wilson PWF, Felson DT 2000 Association of hypogonadism and estradiol levels with bone mineral density in elderly men from the Framingham study. Ann Intern Med 133:951–963[Abstract/Free Full Text]
  13. Szulc P, Munoz F, Claustrat B, Garnero P, Marchand F, Duboeuf F, Delmas PD 2001 Bioavailable estradiol may be an important determinant of osteoporosis in men: the MINOS study. J Clin Endocrinol Metab 86:192–199[Abstract/Free Full Text]
  14. Khosla S, Melton III LJ, Atkinson EJ, O’Fallon WM 2001 Relationship of serum sex steroid levels to longitudinal changes in bone density in young versus elderly men. J Clin Endocrinol Metab 86:3555–3561[Abstract/Free Full Text]
  15. Van Pottelbergh IV, Goemaere S, Kaufman JM 2003 Bioavailable estradiol and an aromatase gene polymorphism are determinants of bone mineral density changes in men over 70 years of age. J Clin Endocrinol Metab 88:3075–3081[Abstract/Free Full Text]
  16. Falahati-Nini A, Riggs BL, Atkinson EJ, O’Fallon WM, Eastell R, Khosla S 2000 Relative contributions of testosterone and estrogen in regulating bone resorption and formation in normal elderly men. J Clin Invest 106:1553–1560[Medline]
  17. Taxel P, Kennedy DG, Fall PM, Willard AK, Clive JM, Raisz LG 2001 The effect of aromatase inhibition on sex steroids, gonadotropins, and markers of bone turnover in older men. J Clin Endocrinol Metab 86:2869–2874[Abstract/Free Full Text]
  18. Leder BZ, LeBlanc KM, Schoenfeld DA, Eastell R, Finkelstein JS 2003 Differential effects of androgens and estrogens on bone turnover in normal men. J Clin Endocrinol Metab 88:204–210[Abstract/Free Full Text]
  19. Eriksen EF, Colvard DS, Berg NJ, Graham ML, Mann KG, Spelsberg TC, Riggs BL 1988 Evidence of estrogen receptors in normal human osteoblast-like cells. Science 241:84–86[Abstract/Free Full Text]
  20. Komm BS, Terpening CM, Benz DJ, Graeme KA, O’Malley BW, Haussler MR 1988 Estrogen binding receptor mRNA, and biologic response in osteoblast-like osteosarcoma cells. Science 241:81–84[Abstract/Free Full Text]
  21. Oursler MJ, Osdoby P, Pyfferoen J, Riggs BL, Spelsberg TC 1991 Avian osteoclasts as estrogen target cells. Proc Natl Acad Sci USA 88:6613–6617[Abstract/Free Full Text]
  22. Oursler MJ, Pederson L, Fitzpatrick L, Riggs BL 1994 Human giant cell tumors of the bone (osteoclastomas) are estrogen target cells. Proc Natl Acad Sci USA 91:5227–5231[Abstract/Free Full Text]
  23. Onoe Y, Miyaura C, Ohta H, Nozawa S, Suda T 1997 Expression of estrogen receptor ß in rat bone. Endocrinology 138:4509–4512[Abstract/Free Full Text]
  24. Bord S, Horner A, Beavan S, Compston J 2001 Estrogen receptors {alpha} and ß are differentially expressed in developing human bone. J Clin Endocrinol Metab 86:2309–2314[Abstract/Free Full Text]
  25. Sims NA, Clement-Lacroic P, Minet D, Fraslon-Vanhulle C, Gaillard-Kelly M, Resche-Rigon M, Baron R 2003 A functional androgen receptor is not sufficient to allow estradiol to protect bone after gonadectomy in estradiol receptor-deficient mice. J Clin Invest 111:1319–1327[CrossRef][Medline]
  26. Spelsberg TC, Subramaniam M, Riggs BL, Khosla S 1999 The actions and interactions of sex steroids and growth factors/cytokines on the skeleton. Mol Endocrinol 13:819–828[Free Full Text]
  27. Salmen T, Heikkinen AM, Mahonen A, Kroger H, Komulainen M, Saarikoski S, Honkanen R, Maenpaa PH 2000 Early postmenopausal bone loss is associated with PvuII estrogen receptor gene polymorphism in Finnish women: effect of hormone replacement therapy. J Bone Miner Res 15:315–321[CrossRef][Medline]
  28. Salmen T, Heikkinen AM, Mahonen A, Kroger H, Komulainen M, Saarikoski S, Honkanen R, Maenpaa PH 2000 The protective effect of hormone-replacement therapy on fracture risk is modulated by estrogen receptor {alpha} genotype in early postmenopausal women. J Bone Miner Res 15:2479–2486[CrossRef][Medline]
  29. Becherini L, Gennari L, Masi L, Mansani R, Massart F, Morelli A, Falchetti A, Gonnelli S, Fiorelli G, Tanini A, Brandi ML 2000 Evidence of linkage disequilibrium between polymorphisms in the human estrogen receptor {alpha} gene and their relationship to bone mass variation in postmenopausal Italian women. Hum Mol Genet 9:2043–2050[Abstract/Free Full Text]
  30. Gennari L, Becherini L, Masi L, Mansani R, Gonnelli S, Cepollaro C, Martini S, Montagnani A, Lentini G, Becorpi AM, Brandi ML 1998 Vitamin D and estrogen receptor allelic variants in Italian postmenopausal women: evidence of multiple gene contribution to bone mineral density. J Clin Endocrinol Metab 83:939–944[Abstract/Free Full Text]
  31. Deng HW, Li J, Li JL, Johnson M, Gong G, Davis KM, Recker RR 1998 Change of bone mass in postmenopausal Caucasian women with and without hormone replacement therapy is associated with vitamin D receptor and estrogen receptor genotypes. Hum Genet 103:576–585[CrossRef][Medline]
  32. Ongphiphadhanakul B, Chanprasertyothin S, Payatikul P, Tung SS, Piaseu N, Chailurkit L, Chansirikarn S, Puavilai G, Rajatanavin R 2000 Oestrogen-receptor-{alpha} gene polymorphism affects response in bone mineral density to oestrogen in post-menopausal women. Clin Endocrinol (Oxf) 52:581–585[CrossRef][Medline]
  33. Han KO, Moon IG, Kang YS, Chung HY, Min HK, Han IK 1997 Nonassociation of estrogen receptor genotypes with bone mineral density and estrogen responsiveness to hormone replacement therapy in Korean postmenopausal women. J Clin Endocrinol Metab 82:991–995[Abstract/Free Full Text]
  34. Arko B, Prezelj J, Komel R, Kocijancic A, Marc J 2002 No major effect of estrogen receptor ß gene RsaI polymorphism on bone mineral density and response to alendronate therapy in postmenopausal osteoporosis. J Steroid Biochem Mol Biol 81:147–152[CrossRef][Medline]
  35. Lau HH, Ho AY, Luk KD, Kung AW 2002 Estrogen receptor ß gene polymorphisms are associated with higher bone mineral density in premenopausal, but not postmenopausal southern Chinese women. Bone 31:276–281[Medline]
  36. Herrington DM, Howard TD, Hawkins GA, Reboussin DM, Xu J, Zheng SL, Broshihan B, Meyers DA, Bleecker ER 2002 Estrogen-receptor polymorphisms and effects of estrogen replacement on high-density lipoprotein cholesterol in women with coronary disease. N Engl J Med 346:967–974[Abstract/Free Full Text]
  37. Sapir-Koren R, Livshits G, Landsman T, Kobyliansky E 2001 Bone mineral density is associated with estrogen receptor gene polymorphism in men. Anthropol Anz 4:343–353
  38. Melton III LJ 1996 History of the Rochester Epidemiology Project. Mayo Clin Proc 71:266–274[Abstract]
  39. Reid IR, Evans, Ames R, Wattie DJ 1991 The influence of osteophytes and aortic calcification on spinal mineral density in postmenopausal women. J Clin Endocrinol Metab 72:1372–1374[Abstract/Free Full Text]
  40. O’Connor S, Baker HWG, Dulmanis A, Hudson B 1973 The measurement of sex steroid binding globulin by differential ammonium sulphate precipitation. J Steroid Biochem 4:331–339[CrossRef][Medline]
  41. Tremblay RR, Dube JY 1974 Plasma concentrations of free and non-TeBG bound testosterone in women on oral contraceptives. Contraception 10:599–605[CrossRef][Medline]
  42. Delmas PD, Stenner D, Wahner HW, Mann KG, Riggs BL 1983 Increase in serum bone {gamma}-carboxyglutamic acid protein with aging in women. Implications for the mechanism of age-related bone loss. J Clin Invest 71:1316–1321
  43. Rosenkranz K, Hinney A, Ziegler A, Hermann H, Fichter M, Mayer H, Siegfried W, Young JK, Remschmidt H, Hebebrand J 1998 Systematic mutation screening of the estrogen receptor ß gene in probands of different weight extremes: identification of several genetic variants. J Clin Endocrinol Metab 83:4524–4527[Abstract/Free Full Text]
  44. Colin EM, Uitterlinden AG, Meurs JB, Bergink AP, van de Klift M, Fang Y, Arp PP, Hofman A, van Leeuwen JP, Pols HA 2003 Interaction between vitamin D receptor genotype and estrogen receptor {alpha} genotype influences vertebral fracture risk. J Clin Endocrinol Metab 88:3777–3784[Abstract/Free Full Text]
  45. Herrington DM, Howard TD, Brosnihan B, McDonnell DP, Li X, Hawkins GA, Reboussin DM, Xu J, Zheng SL, Meyers DA, Bleecker ER 2002 Common estrogen receptor polymorphism augments effects of hormone replacement therapy on E-selectin but not C-reactive protein. Circulation 105:1879–1882[Abstract/Free Full Text]
  46. Santen RJ 1998 Estrogen receptor expression and function in long-term estrogen-deprived human breast cancer cells. Endocrinology 139:4164–4174[Abstract/Free Full Text]
  47. Majeska RJ, Ryaby JT, Einhorn TA 1994 Direct modulation of osteoblastic activity with estrogen. J Bone Joint Surg 76:713–721[Abstract/Free Full Text]
  48. Ireland DC, Bord S, Beavan SR, Compston JE 2002 Effects of estrogen on collagen synthesis by cultured human osteoblasts depend on the rate of cellular differentiation. J Cell Biochem 86:251–257[CrossRef][Medline]



This article has been cited by other articles:


Home page
 J. Clin. Endocrinol. Metab.Home page
J. H. Tobias, C. D. Steer, C. Vilarino-Guell, and M. A. Brown
Estrogen Receptor {alpha} Regulates Area-Adjusted Bone Mineral Content in Late Pubertal Girls
J. Clin. Endocrinol. Metab., February 1, 2007; 92(2): 641 - 647.
[Abstract] [Full Text] [PDF]

Home page
 Eur J EndocrinolHome page
V Rochira, A Balestrieri, B Madeo, L Zirilli, A R M Granata, and C Carani
Osteoporosis and male age-related hypogonadism: role of sex steroids on bone (patho)physiology
Eur. J. Endocrinol., February 1, 2006; 154(2): 175 - 185.
[Abstract] [Full Text] [PDF]

Home page
 J. Clin. Endocrinol. Metab.Home page
S. Ichikawa, D. L. Koller, M. Peacock, M. L. Johnson, D. Lai, S. L. Hui, C. C. Johnston, T. M. Foroud, and M. J. Econs
Polymorphisms in the Estrogen Receptor {beta} (ESR2) Gene Are Associated with Bone Mineral Density in Caucasian Men and Women
J. Clin. Endocrinol. Metab., November 1, 2005; 90(11): 5921 - 5927.
[Abstract] [Full Text] [PDF]

Home page
 Ann Rheum DisHome page
M Johansson, L Arlestig, B Moller, T Smedby, and S Rantapaa-Dahlqvist
Oestrogen receptor {alpha} gene polymorphisms in systemic lupus erythematosus
Ann Rheum Dis, November 1, 2005; 64(11): 1611 - 1617.
[Abstract] [Full Text] [PDF]

Home page
 Hum Mol GenetHome page
A. M. Shearman, S. Demissie, L. A. Cupples, I. Peter, C. H. Schmid, J. M. Ordovas, M. E. Mendelsohn, and D. E. Housman
Tobacco smoking, estrogen receptor {alpha} gene variation and small low density lipoprotein level
Hum. Mol. Genet., August 15, 2005; 14(16): 2405 - 2413.
[Abstract] [Full Text] [PDF]

Home page
 Am J EpidemiolHome page
L. Gennari, D. Merlotti, V. De Paola, A. Calabro, L. Becherini, G. Martini, and R. Nuti
Estrogen Receptor Gene Polymorphisms and the Genetics of Osteoporosis: A HuGE Review
Am. J. Epidemiol., February 15, 2005; 161(4): 307 - 320.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Khosla, S.
Right arrow Articles by Brandi, M. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Khosla, S.
Right arrow Articles by Brandi, M. L.

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