| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Center for Health Sciences (H.Z.R., C.N.L., G.E.S., D.C.), SRI International, Menlo Park, California 94025; Department of Medical and Molecular Genetics (T.R.), Indiana University School of Medicine, Indianapolis, Indiana 46202; and Department of Medicine (R.M., L.H.), Stanford University School of Medicine, The Geriatrics Research, Education and Clinical Center, Veterans Affairs Palo Alto Health Care System, Palo Alto, California 94304
Address all correspondence and requests for reprints to: Huijun Z. Ring, Ph.D., Center for Health Sciences, SRI International, 333 Ravenswood Avenue, Menlo Park, California 94025. E-mail: huijun.ring{at}sri.com.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The age-related changes in sex hormone levels and their effect on health in elderly men is also a topic of intense study. As men age, testosterone concentrations decline and SHBG concentrations rise (14, 15, 16). Hypogonadal men share a variety of signs and symptoms such as decreased muscle mass, osteoporosis, increased fat mass, fatigue, and reduced sexual functioning (17, 18, 19). Recent results from the Baltimore Longitudinal Study on Aging demonstrated that high serum testosterone is associated with an increased risk of prostate cancer in older men (20).
Relatively few studies have addressed the role of genetic factors underlying individual differences in sex hormone concentrations in general and in men, in particular. An important approach to quantify the magnitude of genetic and environmental effects on physiological functions is the classical twin study (21, 22). In its simplest application, this involves quantifying the degree of similarity between monozygotic (MZ) and dizygotic (DZ) twins for a trait of interest, on the premise that differences between genetically identical MZ twins are attributable to environmental factors, whereas the closer similarity of genetically identical MZ twins compared with nonidentical DZ twins is due in part to their different degrees of genetic identity. One assumption in twin studies is that the environmental correlation between MZ twin pairs is equal to the environmental correlation between DZ twin pairs, which then allows the estimation of genetic and environmental components of variance. These estimates allow a calculation of heritability that measures the proportion of phenotypic variance attributable to additive genetic factors.
Twin studies examining the androgenic steroids in men have reported a range of estimates of genetic influence. Of four published studies in twins, the reported heritability for testosterone in men ranged from 17–28% in adults [age 20–60; (23, 24)], 66% in adolescents [age 14–21; (25)], and close to zero for newborn boys (26). The heritability estimates of serum testosterone levels in 801 members participating in the HERITAGE Family Study were 69% for white and 70% for black individuals, respectively (27). The different magnitude of heritability in different age groups as well as reports of absence of a correlation in testosterone concentrations between fathers and sons (25) suggest that different genetic mechanisms may influence plasma testosterone concentrations during the life span.
In the current study, a large sample of 530 healthy male twins ages 59–70 yr were assayed for serum sex steroid contents, including testosterone, estradiol, estrone, and SHBG. Our objective was to evaluate the relative contribution of genetics and environment to these sex steroid hormone concentrations in late adulthood.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This heritability analysis drew upon hormone level data collected from the National Heart, Lung, and Blood Institute (NHLBI) twin study. Participants were aged 59–70 yr when the blood samples were collected between 1981 and 1987. Participants in the present study are a subgroup drawn from a population-based registry of almost 16,000 pairs of white, male, veteran twin pairs maintained by the National Academy of Sciences-National Research Council and funded by the NHLBI. Baseline examinations were conducted during 1969–1973 on 514 intact twin pairs, or 1028 individuals at five research facilities in the United States. Details of the study design, methods, and analyses of baseline and follow-up risk factors and cardiovascular events have been published (28). Four follow-up examinations have been completed. Zygosity assessment was based on eight red cell blood groups (ABO, MNS, Rh, Kell, Lewis, Duffy, Gm, and Kidd) comprising 22 antigens. Zygosity assignment for some participants was later confirmed by DNA testing using variable number tandem repeat DNA markers. Proper informed consent was obtained from the subjects. The protocols for the described study were approved by the Institutional Review Boards of the research facilities.
Data collection
Blood samples were collected between 1981 and 1987 in the morning between 0800 and 1000 h after an overnight fast. Attempts were made to see both members of each twin pair on the same day. There was no exclusion from blood sampling because of health reasons. EDTA-plasma was aliquoted and frozen at –80 C, and repeated freeze/thaw cycles have been avoided. Between 2001 and 2002, estradiol, estrone, testosterone, and SHBG concentrations were measured in stored plasma samples. Specimens were from the same examination for both members of each twin pair. There was not enough plasma to measure all four hormones on all participants. The sample sizes for estradiol, estrone, testosterone, and SHBG measurements are 255, 260, 243, and 258 twin pairs, respectively. The present analysis determined the heritability of estradiol, estrone, testosterone, and SHBG concentrations. A separate analysis explored the association between sex hormone concentrations and coronary heart disease (12).
Laboratory methods
Hormone assays were carried out at the Palo Alto Veterans Affairs Health Care System as previously reported (29). Reagent kits were generously provided by Diagnostics Systems Laboratories, Inc. (Webster, TX). Attempts were made to run samples from both members of each twin pair in the same assay batch. Laboratory personnel were blinded to zygosity. Estradiol was measured by RIA, and the sensitivity and the intra- and interassay coefficients of variation were 0.6 pg/ml, 11.1 and 14.1%, respectively. Estrone was measured by RIA, and the sensitivity and the intra and interassay coefficients of variation were 1.2 pg/ml, 12.8 and 11.3%, respectively. Total testosterone was measured by RIA, and the sensitivity and the intra- and interassay coefficients of variation were 0.08 ng/ml, 6.9 and 10.8%, respectively. SHBG was measured by immunoradiometric assay, and the sensitivity and the intra- and interassay variation were 3 nmol/liter, 6.9 and 7.0%, respectively. The values we report for estradiol, testosterone, and SHBG fall within expected ranges for normal men as provided by the kit manufacturer and are typical of men in this age group as we have previously reported (12, 29).
Statistical methods
For each sex hormone and SHBG concentration, we compared means in all study participants (treated as individuals) between zygosity groups using ANOVA. We compared the correlations among sex hormones using Pearson correlation analysis. The effects of age and body mass index (BMI) on hormone concentrations were analyzed by linear regression.
For genetic analysis, we first computed intraclass twin pair correlations for each variable and each zygosity group. The comparison of the MZ with DZ intraclass correlation provides initial information on the presence and extent of genetic effects. If additive genetic effects (designated A) are present, the MZ correlation is expected to be twice the DZ correlation. An MZ correlation that is less than twice the DZ correlation suggests that both additive genetic and shared environmental effects (designated C) influence the phenotype of interest. Nonshared environmental effects (designated E) are present when the MZ intraclass correlation is less than unity. Common environment refers to experiences shared by cotwins, including the uterine environment, the rearing environment, the frequency of contact between twins, and their joint exposure to the same social and cultural environment. Nonshared environmental influences are all the factors (e.g. accidents, illnesses) that make members of a twin pair different from one another, and also include measurement error.
Heritability analysis was performed by using two different statistical approaches: 1) classical heritability analysis calculated as twice the difference between the MZ and DZ intraclass correlation (Falconer’s estimate) by using the program TWINAN90 (30); and 2) genetic models fitted by the method of maximum likelihood, using the Mx software (31). TWINAN90 performs ANOVA-based twin analyses and tests the significance of the difference in intraclass correlation coefficients between MZ and DZ pairs, from which the classical heritability estimate 2*(rMZ – rDZ) was calculated and tested for significance. Model fitting to twin data using the Mx software allows for the magnitude of the relative contribution of genetic and environmental effects to be quantified. Full models that allow estimation of all three parameters, that is genetic, shared environmental, and nonshared environmental effects (i.e. ACE models), were fitted to MZ and DZ twin pair covariance matrices. The significance of the A or C parameters was tested by equating each parameter to zero and testing the fit of the reduced model (AE or CE) against the full model, using the 
2 likelihood difference test. The best model fit was evaluated according to the principle of parsimony, in which models with fewer parameters are considered preferable if they do not show a significant worsening of fit compared with the full model. The parsimony principle is further operationalized by the Akaike Information Criterion, calculated as the model 2 minus two times the degrees of freedom (32).
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
summarizes the mean and SD of age, BMI, plasma sex hormone, and SHBG concentrations in MZ and DZ twins. There were no significant differences in age, BMI, hormone, or SHBG concentrations across zygosity groups.
|
. Testosterone concentration was significantly and positively correlated with concentrations of estradiol, estrone, and SHBG. Estradiol was significantly positively correlated to a similar degree with testosterone (r = 0.32) and with estrone (r = 0.39). Testosterone concentrations were more strongly associated with concentrations of estradiol than with estrone concentrations (r = 0.19) or with SHBG (r = 0.15). The correlations listed in Table 2 are modest (r <0.5), even for those that are statistically significant. The positive correlations between testosterone, estradiol, and estrone were expected, and those correlations were similar to those reported in the literature (26, 33). Such interhormone correlations could be understood in part by the biochemical processes involved in hormone biosynthesis, because both testosterone and estrone serve as substrates for estradiol. Furthermore, SHBG production is regulated by sex steroid concentrations, and liver SHBG secretion is inhibited by androgens. There is a positive correlation between testosterone and SHBG, which possibly reflects that total testosterone levels are largely influenced by the binding protein, SHBG.
|
. Age and BMI do not contribute to twin similarity in hormone concentrations because the amount of total variation (r2) explained by either age or BMI accounted for less than 5% of the total variance. All intraclass twin pair correlations deviated significantly from zero (all P < 0.01). The intraclass correlation was higher in MZ twin pairs than DZ pairs for all measurements, except for estrone, in which case it was close to equal in MZ (0.40) and DZ (0.39) twins.
|
, and the genetic and environmental parameter estimates for the best fitting model of each sex hormone and SHBG concentration are presented in Table 4. Using the maximum likelihood method, genetic factors accounted for about half of the variation of plasma testosterone (57%), about one fourth (25%) of the variation of estradiol, and 68% of the variation in SHBG concentration. The best fitting model for estrone indicated that twin similarity was due to shared environmental (37%) and not additive genetic factors. Shared environmental factors were important (44%) in twin pair similarity for estradiol as well. After accounting for genetic and/or shared environmental variance, the remaining phenotypic variances were attributable to nonshared environmental effects, which were high for estradiol (75%) and for estrone (63%) and lower for testosterone (43%) and SHBG (32%). Overall, the pattern of MZ and DZ twin pair correlations shown in Table 3 is consistent with the estimates of genetic and environmental influences shown in Table 4. For estradiol and estrone, the best fitting models derived using maximum likelihood methods and shown in Table 4 closely parallel the heritability estimates using the classical twin analysis method shown in Table 3, that is 20–25% heritability for estradiol and no significant heritability for estrone. For testosterone and SHBG, the intraclass correlation-based heritability estimates were lower than those estimated by the maximum likelihood method, and such deviations arise as a function of the established and accepted standard of model parsimony.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In this study, the overall pattern of MZ and DZ twin pair correlations presented in Table 3 is consistent with the estimates of genetic and environmental influences shown in Table 4. For testosterone and SHBG, the heritability estimates using the maximum likelihood methods are higher than those derived from the calculated MZ and DZ intraclass correlation coefficients. In general, estimating heritability using model fitting by maximum likelihood method is the preferred methodology, because heritability estimates using the intraclass correlations involve standardization by variances which lead to the loss of crucial information such as the actual magnitude of individual intrapair similarity, whereas model fitting by the maximum likelihood method uses all raw data directly and the SE of estimates of genetic variance are much smaller than those obtained from ANOVA. Using simulation, Christian et al. (34) compared the power of these two methods to detect twin covariance and to partition covariance into genetic and environmental effects. The simulation results on 50 MZ and 50 DZ pairs revealed that, with maximum likelihood methods, the comparison of the ACE vs. the AE model is fairly weak with sufficient statistical power only to detect large shared environmental effects, leading to the conclusion that estimates of heritability using the maximum likelihood method could be inflated in small twin studies. Therefore, it is important to present the heritability estimates derived from multiple methods as well as from both the full ACE and the best fitting AE models in this report, although our twin sample size is considerably larger than published simulation data set. For the testosterone result, comparing the associated 95% confidence intervals (CI) for estimates for the full ACE model and for the best fitting AE model, it is conceivable that the small magnitude of shared environmental effects (12%; 95% CI, 0.02–0.42) estimated in the full ACE model is indeed important. This also applies to the SHBG results where shared environmental effects may account for as much as 15% (95% CI, 0.01–0.43) of the total phenotypic variance. Nonetheless, substantial genetic influences on variation in both testosterone and SHBG concentration remain even in the full ACE model (44 and 52%, respectively).
That variation in estrone (37% shared environmental and 63% individual specific environmental effects) and estradiol concentrations (25% genetic effect, 44% shared environmental effects, 31% individual specific environmental effects) were largely influenced by environmental factors should be explored through further epidemiological research. Many environmental factors are known to affect plasma sex steroid concentrations, including stress (34), marked obesity (8), cigarette smoking (35), alcohol consumption (36), and aging (15). Epidemiological studies on the possible effects of environmental estrogens on human health could also be affected by these findings. Regional differences in health, such as breast cancer rates, have been observed with exogenous endocrine disruptors proposed as one possible contributing factor. Endogenous sex hormone concentrations could affect the proposed toxicity of exogenous endocrine modulators (38). Therefore, the results of this study lend further support to a possible environmental etiology to some of these health problems.
Variation in testosterone concentration was significantly influenced by genetic factors (57%; 95% CI, 0.46–0.68). Genetic modeling did not detect a significant influence of shared environmental effect, however the ratio of the MZ to DZ twin pair correlation is less than two, suggesting some influence of shared environment. The enzymatic steps of testosterone biosynthesis and metabolism have been well characterized. Serum testosterone concentration is subjected to regulation at multiple levels, and could be affected by the activities of many enzymes involved in biosynthesis and degradation, such as 17ß-hydroxysteroid dehydrogenase, 5
-reductase, P450 aromatase, and catechol-o-methyltransferase. The genes encoding those enzymes have been cloned, and genetic polymorphisms in these genes have been identified (38, 39). Molecular epidemiological studies will need to be conducted in human populations to evaluate the effects of DNA polymorphisms in the genes involved in steroid hormone production and metabolism, and such studies would help to understand the biological mechanisms underlying individual differences in plasma sex steroid hormone concentrations.
Additive genetic factors also accounted for the majority of the total phenotypic variation in serum concentrations of SHBG (68%; 95% CI, 0.60–0.76). SHBG is a plasma glycoprotein that binds estrogens and androgens with high affinity, and it regulates the bioavailability and biological activities of male and female sex steroids (41). The gene encoding human SHBG is located on the short arm region of chromosome 17 and contains eight exons (42). Genetic polymorphisms in the SHBG gene have been found to influence SHBG levels, at least in women (43, 44). Several studies in humans have also shown that plasma SHBG concentrations are regulated by different physiological and pathological conditions as well as by sex hormone concentrations, nutritional status and body mass (44, 45, 46, 47).
There are several important limitations of the current twin study. For each subject, hormone measurements were made on a single plasma sample in the morning. Because steroid hormone concentrations undergo substantial daily fluctuation, a single hormone measurement may not suffice to accurately describe an individual’s overall hormone status. Thus, multiple specimens may be needed to assess the overall hormone status of an individual. However, the impact of such fluctuation is minimized in this study by the large number of subjects. A second methodological issue in our study is potential sample deterioration resulting from the storage of plasma samples for 15–22 yr before hormone measurement. Although serum samples were aliquoted and did not go through freeze-thaw cycles, other studies suggest that estradiol and estrone values decrease with time in storage, even at –80 C (49, 50, 51). However, we note that the hormone values that we reported fall within expected ranges for normal men as provided by the kit manufacturer.
It is also important to note that in our study total estradiol and total testosterone were determined, but total measurements may not estimate hormone exposure as well as measurements of bioavailable or free hormone. Attempts were made to approximate bioavailable estradiol and testosterone by calculating the ratios between the total hormones and SHBG. Furthermore, there are two commonly used laboratory methods to quantify SHBG concentration in plasma samples: competitive protein binding assay and the radioimmunometric procedure. The heritability for SHBG concentrations is negligible (<5%) in studies in which SHBG concentrations were measured using a competitive protein binding technique (23, 24, 52). However, significant heritability was reported in a more recent twin study [62% (15)] and in two family-based studies [31–50% (53, 54)] in which SHBG concentrations were measured using a radioimmunometric procedure. Our study used the radioimmunometric procedure, and our heritability estimate for SHBG level (68%) is more in line with the studies that used the same procedure.
In this study we have defined genetic and environmental influences in a broad sense. Instead of identifying and evaluating specific gene or environmental influences, we estimated the proportion of total phenotypic variance due to genetic variation. It is important to note that high heritability does not imply that environmental factors are unimportant. Rather, it indicates that within the population, genetic factors are responsible for the majority of the variation. The effects of specific known regulators of sex hormone levels, including insulin and dietary factors, were not evaluated in this analysis, and future studies will be needed to ascertain specific gene and nongenetic influences. Examination of the extent of genetic influence on hormone values across time in the same twin subjects would also be very interesting. Unfortunately, plasma samples were not available from earlier or later exams of the NHLBI twin study. We are unaware of any longitudinal study on heritability using repeated assessment of hormones in twin or family study samples.
A number of molecular epidemiological studies have been conducted to evaluate associations between polymorphic genes involved in steroid hormone production and metabolism, including cytochrome p450 enzymes, sulfatase, sulfotransferases, catechol-o-methyltransferase, and uridine-5'-diphosphate glucuronosyltransferase (38, 39). Variability in the expression levels or activities of these proteins may underlie the interindividual difference in plasma sex steroid hormone content. As more data become available about the regulation of sex steroids and SHBG concentrations, it is evident that the system regulating the serum sex steroid content is very diverse. We consider our analyses as the initial step and plan to further investigate the phenotypes with significant heritability, i.e. serum testosterone and SHBG concentrations, in future molecular association studies. Additionally, multivariate genetic analysis will demonstrate the degree of genetic or environmental factors that may underlie the observed phenotypic correlations among sex hormone concentrations. Understanding the molecular basis of the variation and covariation in sex hormone concentrations may help in the search for genes underlying complex traits affected by plasma concentrations of sex hormones.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
R.M. is now affiliated with Eli Lilly Co. (Indianapolis, IN).
First Published Online March 8, 2005
Abbreviations: BMI, Body mass index; CI, confidence interval(s); DZ, dizygotic; MZ, monozygotic.
Received May 30, 2004.
Accepted February 28, 2005.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A. L. Eriksson, M. Lorentzon, L. Vandenput, F. Labrie, M. Lindersson, A.-C. Syvanen, E. S. Orwoll, S. R. Cummings, J. M. Zmuda, O. Ljunggren, et al. Genetic Variations in Sex Steroid-Related Genes as Predictors of Serum Estrogen Levels in Men J. Clin. Endocrinol. Metab., March 1, 2009; 94(3): 1033 - 1041. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. J. Thompson, C. S. Healey, C. Baynes, B. Kalmyrzaev, S. Ahmed, M. Dowsett, E. Folkerd, R. N. Luben, D. Cox, D. Ballinger, et al. Identification of Common Variants in the SHBG Gene Affecting Sex Hormone-Binding Globulin Levels and Breast Cancer Risk in Postmenopausal Women Cancer Epidemiol. Biomarkers Prev., December 1, 2008; 17(12): 3490 - 3498. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Peter, A. Kelley-Hedgepeth, C. S. Fox, L. Adrienne Cupples, G. S. Huggins, D. E. Housman, R. H. Karas, M. E. Mendelsohn, D. Levy, and J. M. Murabito Variation in Estrogen-Related Genes Associated with Cardiovascular Phenotypes and Circulating Estradiol, Testosterone, and Dehydroepiandrosterone Sulfate Levels J. Clin. Endocrinol. Metab., July 1, 2008; 93(7): 2779 - 2785. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. D Williams Individual variation in endocrine systems: moving beyond the 'tyranny of the Golden Mean' Phil Trans R Soc B, May 12, 2008; 363(1497): 1687 - 1698. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Kempenaers, A. Peters, and K. Foerster Sources of individual variation in plasma testosterone levels Phil Trans R Soc B, May 12, 2008; 363(1497): 1711 - 1723. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Crabbe, V. Bogaert, D. De Bacquer, S. Goemaere, H. Zmierczak, and J. M. Kaufman Part of the Interindividual Variation in Serum Testosterone Levels in Healthy Men Reflects Differences in Androgen Sensitivity and Feedback Set Point: Contribution of the Androgen Receptor Polyglutamine Tract Polymorphism J. Clin. Endocrinol. Metab., September 1, 2007; 92(9): 3604 - 3610. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E.A.M. Kuijper, C.B. Lambalk, D.I. Boomsma, S. van der Sluis, M.A. Blankenstein, E.J.C. de Geus, and D. Posthuma Heritability of reproductive hormones in adult male twins Hum. Reprod., August 1, 2007; 22(8): 2153 - 2159. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. I. Berndt, N. Chatterjee, W.-Y. Huang, S. J. Chanock, R. Welch, E. D. Crawford, and R. B. Hayes Variant in Sex Hormone-Binding Globulin Gene and the Risk of Prostate Cancer Cancer Epidemiol. Biomarkers Prev., January 1, 2007; 16(1): 165 - 168. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. L. Eriksson, M. Lorentzon, D. Mellstrom, L. Vandenput, C. Swanson, N. Andersson, G. L. Hammond, J. Jakobsson, A. Rane, E. S. Orwoll, et al. SHBG Gene Promoter Polymorphisms in Men Are Associated with Serum Sex Hormone-Binding Globulin, Androgen and Androgen Metabolite Levels, and Hip Bone Mineral Density J. Clin. Endocrinol. Metab., December 1, 2006; 91(12): 5029 - 5037. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Jasienska, M. Kapiszewska, P. T. Ellison, M. Kalemba-Drozdz, I. Nenko, I. Thune, and A. Ziomkiewicz CYP17 Genotypes Differ in Salivary 17-{beta} Estradiol Levels: A Study Based on Hormonal Profiles from Entire Menstrual Cycles. Cancer Epidemiol. Biomarkers Prev., November 1, 2006; 15(11): 2131 - 2135. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Raven, F. H. de Jong, J.-M. Kaufman, and W. de Ronde In Men, Peripheral Estradiol Levels Directly Reflect the Action of Estrogens at the Hypothalamo-Pituitary Level to Inhibit Gonadotropin Secretion J. Clin. Endocrinol. Metab., September 1, 2006; 91(9): 3324 - 3328. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. N. Lessov-Schlaggar, T. Reed, G. E. Swan, R. E. Krasnow, C. DeCarli, R. Marcus, L. Holloway, P. A. Wolf, and D. Carmelli Association of sex steroid hormones with brain morphology and cognition in healthy elderly men Neurology, November 22, 2005; 65(10): 1591 - 1596. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
