| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Articles |
Reproductive Medicine Unit, Department of Obstetrics and Gynecology, University of Adelaide, The Queen Elizabeth Hospital, Woodville SA 5011, Australia
Address all correspondence and requests for reprints to: Theresa Hickey, M.D., Adelaide University, Department of Obstetrica and Gynecology, The Queen Elizabeth Hospital, First Floor, Maternity Building, 28 Woodville Road, Woodville, SA 5011, Australia. E-mail: theresa.hickey{at}adelaide.edu.au
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Polycystic ovary syndrome (PCOS) is an endocrine disorder characterized by abnormal androgen production and/or activity that leads to changes in the control of follicle development and maturation (12). Evidence in nonhuman primates ( 13) and transsexual women (14) treated with high doses of androgen indicate that the characteristic ovarian morphology, comprised of an enlarged ovary with numerous small follicular cysts, may be the result of direct, receptor-mediated androgen activity. In women with PCOS, this disruption often leads to chronic anovulation and subsequent infertility. Recently, Mifsud et al. (15) reported that normoandrogenic Chinese women with anovulation and polycystic ovaries had a significantly greater frequency of short AR CAG repeat alleles than hyperandrogenic Chinese women with the same condition. They suggest that the higher AR activity potentially conferred by shorter CAG repeat alleles may amplify the androgenic response in ovarian tissues, resulting in a PCOS phenotype. Acceptance of this proposal assumes preferential expression of the allele of shorter size by skewed inactivation of the allele of larger size, but patterns of X-inactivation were not assessed. This study showed no differences in CAG allele distribution between infertile Chinese women with PCOS and fertile Chinese women.
We investigated a cohort of Australian Caucasian women with PCOS to compare their patterns of CAG repeat distribution and X-chromosome inactivation against those of normal women. Vottero et al. (3) reported that preferential expression of the shorter CAG repeat allele was the sole difference between women with and without hirsutism, an androgen-sensitive condition. Considering the results of these previous studies, we expected to find an association between PCOS and increased frequency or preferential expression of AR CAG repeat alleles in the lower region of the normal polymorphic spectrum, but surprisingly found the opposite to occur in our population of women.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Women were recruited from infertility and antenatal clinics at The Queen Elizabeth Hospital in Adelaide, South Australia, after approval by the ethics committee of North Western Adelaide Health Services. Due to ethnic differences in allele frequency for the human AR (hAR) (CAG)n polymorphism, women from distinct Asian, Indian, Middle Eastern, or Aboriginal heritage were excluded from the study. The resultant study group represents women of various European cultural backgrounds and generally can be classified as Caucasian. A total of 171 subjects in an infertile relationship were recruited. All of these women were subject to the same infertility assessment protocol, which included interviews about reproductive history, pelvic ultrasound for ovarian morphology, blood tests for hormonal profiles, tracking of menstrual cycles, and height and weight measurements. Women with abnormal androgen levels were further assessed for disorders that form exclusion criteria for the diagnosis of PCOS (16). PCOS was based on the criteria of hyperandrogenism and anovulation with exclusion of congenital adrenal hyperplasia and Cushing’s syndrome. We also recorded the presence of polycystic ovaries on ultrasound due to documented AR-mediated causal effects and to accord with criteria used in a previous study of CAG repeats in infertile women (15). Polycystic ovaries were defined as the presence of at least eight peripheral cysts less than 10 mm in diameter, with increased ovarian stroma on ultrasound, occurring bilaterally (17). Hyperandrogenism was diagnosed as the presence of at least two of the following: high serum androstendione (>7.0 nmol/liter), high serum T (>2.0 nmol/liter), and low SHBG (<20 nmol/liter). Menstrual cycles were considered normal if they occurred every 23–35 d, and anovulation was inferred either by repetitively abnormal (>35 d) menstrual cycle history or serial serum progesterone levels less than 5 nmol/liter in the luteal phase. A total of 122 (71%) of women fulfilled our criteria for PCOS. We recruited 83 women from an antenatal clinic to serve as controls, all of whom were Caucasian, had had a history of normal menstrual cycles, had no history or evidence of cutaneous hyperandrogenemia, and had never sought treatment for infertility.
Analysis of the hAR (CAG)n gene locus
DNA extraction. Peripheral blood lymphocytes were purified from whole blood using Lymphoprep (Nycomed Pharma, Oslo, Norway) and kept frozen in saline at -20 C until DNA extraction. DNA was extracted from this tissue using a QIAmp Blood Kit (QIAGEN, Chatsworth, CA) protocol, quantified by spectrophotometry, and stored at -20 C.
CAG allele characterization. Alleles were characterized by the number of CAG repeat units in the hAR microsatellite. Our technique is similar to those used in other published studies (8, 18). Genomic DNA (100 ng) was amplified by PCR using primers that flank the hAR (CAG)n polymorphism and include two HpaII sites that are methylated on the inactive X-chromosome: 165-hAR-CAG-S, 5'-gtgcgcgaagtgatccagaa-3' and 392-hAR-CAG-AS, 5'-tagcctgtggggcctctacg-3'. The sense primer was 5'-labeled with the florescent dye tetrachloro fluorescein. PCR was performed with the DNA polymerase HotStarTaq (QIAGEN) using reagents and protocol provided with the enzyme and a primer annealing temperature of 65 C. PCR products were run on Genescan (ABI, Foster City, CA) under standard conditions and analyzed by Genescan 3.1 software. Sequenced standards were run on each gel to correlate PCR fragment size with CAG repeat number.
X-inactivation analysis. Subjects homozygous at the hAR(CAG)n gene locus were excluded from X-inactivation analysis due to an inability to distinguish between the two alleles: 13% (n = 16) PCOS and 17% (n = 14) fertile controls. The remaining heterozygous samples, including 87% (n = 106) PCOS and 83% (n = 69) fertile controls, were analyzed for X-inactivation ratio by assessment of methylation status using the methylation-sensitive restriction enzyme, HpaII (Roche Diagnostic Systems, Mannheim, Germany). Nonmethylated (active X) DNA segments digest with enzyme and are thereby unavailable for PCR amplification. Methylated (inactive X) HpaII sites do not digest with enzyme and remain intact for amplification. Postdigestion PCR products therefore represent methylated (inactive X) DNA sequences only.
Equivalent 2 µg DNA aliquots were either digested with HpaII (30 U) or mock-digested in digestion buffer with no enzyme. Samples were digested overnight at 37 C in a 20 µl reaction volume, with a final enzyme denaturation step of 95 C for 5 min. Aliquots of 5 µl were amplified by PCR and analyzed on Genescan as described above. Total florescent peak areas for both alleles and the ratio to which each allele contributed to the total were calculated for digested and undigested samples using Genescan 3.1 software. Differences in the allele ratio between undigested and digested samples represent the degree to which one allele is more or less methylated than the other in a DNA sample. We used calculations similar to those reported by Naumova et al. (19), expressing final ratios in terms of degree of inactivation of the longer allele (which directly correlates to the degree of expression of the shorter allele).
All samples were analyzed in duplicate in nondigested and digested conditions. In calculating the degree of X-inactivation for individuals, we used the average value of the duplicate samples for both digested and undigested DNA samples in the calculations. Although total florescent peak areas for replicates varied, the ratio that each amplified allele contributed to total florescence remained remarkably constant, with a mean difference of 1.4% and SD of 1.5% between replicates.
Allele distribution profiles
Alleles were analyzed following the conventional method of representing data of this type in women. First, total alleles were plotted whereby each woman contributed two independent values that represented both CAG repeat alleles. Three further modes of allele representation were employed, using 1) the mean value of the two alleles (biallelic mean), 2) the shorter allele alone, and 3) the longer allele alone. In the latter two instances, the use of the terms shorter and longer is relative per individual and does not represent an absolute CAG repeat number or range of numbers. We also formulated a new method of calculating biallelic means whereby the results of X-inactivation analysis were used to create a mean value that represents differences in the expression of constituent alleles. This was achieved by multiplying each allele in a genotypic pair by its percentage of total expression (100 minus % inactivity) and summing the two adjusted repeat values to achieve a new mean value that we call the X_weighted_biallelic_mean. Individuals homozygous at the hAR (CAG)n locus were included in the distribution comparison of X_weighted_biallelic_means because variation of allele expression would not alter the mean value of alleles of equivalent repeat number.
Statistical analysis
All statistics were generated using SigmaStat software (version 2, SPSS, Inc., Chicago, IL). Parameters for statistical analysis among groups lacked normal distribution profiles, requiring nonparametric statistical tests. Allele distributions between groups were analyzed by dividing control profiles into two sections of approximately equal size and then using the median CAG repeat value as a cut-point for division of PCOS alleles. Then, differences in distribution were determined by Fisher’s exact test. Simple linear regressions and Spearman rank order correlations were used to compare serum T and body mass index values with CAG repeat number in the PCOS group. If a correlation was found, t tests comparing alleles divided by specific CAG repeat number cut-points were used to further characterize the correlation. Frequency of skewed (
80%) inactivation of one allele, as defined by Naumova et al. (19), was analyzed using Fisher’s exact test. For all tests, significance was set at 5%.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Androgen receptor CAG repeats ranged from 14–30 in control women and 8–32 in PCOS women, which reflects the normal polymorphic range reported at this genetic locus for a large cohort (n = 831) of predominantly fertile (74%) Australian Caucasian women (20). Control allele distribution profiles corresponding to total alleles, biallelic means, and X_weighted_biallelic_means were divided into two approximately equal sections at a cut-point of no more than 22 CAG repeats. Short allele distributions had a cut-point of no more than 21, and long allele distributions had a cut-point of no more than 23 to ensure equal partition. Women with PCOS had a significantly greater frequency of alleles in the upper half of the polymorphic spectrum when comparing distributions of biallelic means and long alleles with control women (Table 1
). Interestingly, when biallelic means were adjusted for preferential allele expression via X-inactivation analysis, this difference between groups became highly significant (P = 0.04 for biallelic means vs. P = 0.009 for X_weighted_biallelic_means) (Fig. 1). We also compared the Australian Caucasian data with our group data and found a very significant difference (P < 0.001) between this large cohort and the PCOS group for the distribution of biallelic means and long alleles (Table 1). In contrast, the allele distribution patterns of fertile controls were concordant with those of the larger population profile.
|
|
The PCOS group appeared to have a greater incidence of nonrandom X-inactivation (alleles, 60% inactive) than controls, but this was not significant (63 vs. 49%; P = 0.08) (Fig. 2). We found no evidence of abnormal incidence of skewed X-inactivation (alleles, 80% inactive) in either group (14% controls, 15% PCOS), which is comparable to previously published studies (18, 19).
|
). This difference explains the significant shift in frequency of X_weighted_biallelic_means toward the upper region of the polymorphic range in the PCOS group compared with controls (Fig. 1
).
|
Fertile control and PCOS women were well matched for age (median ± SD, 28.5 ± 7.3 vs. 29.4 ± 4.8 yr). All women in this study are less than 50-yr-old and fall within a two decade range; therefore age should have minimal effect on the results of X-inactivation analysis (21, 22).
A common clinical feature of PCOS is obesity, and its effects may influence ovulation (23, 24). We did not find any correlation between body mass index and CAG repeat number in this study, which concurs with previously published results (25).
A positive association was found between serum T and the longer CAG allele (r = 0.21; P = 0.03) in PCOS women. However, T levels in these women were significantly different only when sorted by X_weighted_biallelic_means, not by longer alleles. PCOS women with X_weighted_biallelic_means of at least 23 had significantly higher levels of T (5.42 nmol/liter ± 2.7) than PCOS women with means less than 23 (4.1 nmol/liter ± 2.6) (P = 0.01).
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Preferential expression of longer CAG alleles augments the disparity found between healthy fertile women and infertile women with PCOS. It also correlates with higher levels of serum T within the PCOS group. These findings endorse the use of X-inactivation analysis when comparing groups of women at the hAR (CAG)n gene locus. Following convention, we used DNA from peripheral blood samples to assess X-inactivation patterns, which may not reflect patterns found in other tissues. There is no data that compares peripheral blood X-inactivation patterns to those in ovarian tissues, however Van Deerlin et al. (27) have reported that eight of nine normal cycling women have one or more ovarian follicles that express only one AR allele. Theoretically, these follicles could behave quite differently in terms of AR-mediated activity to follicles that solely express the other allele, or to follicles that express different degrees of both alleles. This is an intriguing concept that has yet to be explored in PCOS and could potentially explain variations in ovarian morphology and follicle development within the disorder.
In contrast to the current study, Mifsud et al. (15) found no differences between infertile and fertile women for AR CAG repeat distribution patterns in a cohort of predominantly Chinese women resident in Singapore. Genetically, both studies reflected previously reported ethnic differences at the hAR (CAG)n gene locus (9). Clinically, anovulation and polycystic ovaries were common inclusion criteria for infertile women, but our study also included hyperandrogenemia, a primary diagnostic criteria for PCOS. Both studies suffered from lack of a control group that had ovarian ultrasound scan and serum hormone measurements. Such control groups are difficult to attain because of the intrusive nature and cost of these procedures. In our experience, women seeking infertility treatment who definitely lacked a PCOS phenotype did not comprise an adequate control group due to diverse infertility diagnoses. A long-term prospective study involving large numbers of women is needed to confirm the preliminary findings presented by these two studies.
Despite differences, an interesting commonality emerges between the Mifsud et al. (15) and current studies: higher serum T was associated with increased frequency or greater expression of longer CAG alleles. In both studies, the relative term "long" refers to the same sector of the polymorphic spectrum and thereby assumes the same potential physiological significance. Collectively, these data may indicate that hyperandrogenism as a characteristic of PCOS correlates to long CAG repeat alleles. This would be consistent with reports that hyperandrogenism per se is the strongest genetically inherited characteristic in familial cases of PCOS (28, 29).
Two previous studies have investigated X-inactivation patterns of AR CAG repeat alleles in women with hirsutism, an androgen-sensitive disorder (3, 8). Increased methylation of longer CAG alleles in hirsute women was found in the former but not the latter study. We also found increased methylation of longer CAG alleles, but this was not exclusive to women with PCOS, and the longer alleles were considerably larger in terms of absolute repeat numbers than the alleles in these former studies. To our knowledge, we are the first to report preferential methylation of CAG alleles of more than 22 repeats within the normal polymorphic range.
It is difficult to explain how an increased frequency and/or expression of long CAG repeat alleles relates to the condition of PCOS. Our data partially contradict data from other studies as well as our initial hypothesis, which emphasizes the need for future comparative studies. It is possible that the association we found has no physiological effect and therefore is of no significance. However, incremental differences in CAG repeat number do appear to have clinical relevance, presumably due to cumulative pathological effects (30). Also, the differential effect of CAG repeat number on AR transactivation capability in vitro is enhanced with elevated levels of androgen (31), so these effects may be amplified in a hyperandrogenic environment.
If we adhere to the theory of an inverse relationship between CAG repeat number and receptor activity, our results indicate that infertile women with PCOS tend to have and/or preferentially express receptors of relatively low androgenic activity compared with fertile controls. Also, those women with the lowest potential androgenic activity had higher levels of serum T. This latter data has no correlate in the literature because it depends on the evaluation of X-inactivation patterns and has sufficient study numbers to span the entire normal polymorphic spectrum. Considering the theory that PCOS may arise from an early developmental event that evolves into a self-perpetuating disease process, we are compelled to ask whether low androgenic activity could trigger a mechanism that ultimately results in increased androgen secretion. This could involve hypothalamic AR activity (32) as opposed or in addition to ovarian AR activity.
In summary, we conclude that the AR CAG repeat polymorphism and its differential methylation have some influence in the manifestation of PCOS and warrant continued investigation. Although evidence suggests that the hAR (CAG)n gene locus does not have the strongest linkage to PCOS in a study of 37 candidate gene loci (33), it ranked among the top 10 in associative strength. Furthermore, this genetic locus has recently been linked to male pattern baldness (34), a putative male PCOS phenotype ( 35), providing further support to this view.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received March 14, 2001.
Accepted September 24, 2001.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  M. Simoni, C.B. Tempfer, B. Destenaves, and B.C.J.M. Fauser Functional genetic polymorphisms and female reproductive disorders: Part I: polycystic ovary syndrome and ovarian response Hum. Reprod. Update, July 4, 2008; (2008) dmn024v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. A. Shah, H. J. Antoine, M. Pall, K. D. Taylor, R. Azziz, and M. O. Goodarzi Association of Androgen Receptor CAG Repeat Polymorphism and Polycystic Ovary Syndrome J. Clin. Endocrinol. Metab., May 1, 2008; 93(5): 1939 - 1945. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M.O. Goodarzi, N. Xu, J. Cui, X. Guo, Y.I. Chen, and R. Azziz Small glutamine-rich tetratricopeptide repeat-containing protein alpha (SGTA), a candidate gene for polycystic ovary syndrome Hum. Reprod., May 1, 2008; 23(5): 1214 - 1219. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Lappalainen, P. Utriainen, T. Kuulasmaa, R. Voutilainen, and J. Jaaskelainen Androgen Receptor Gene CAG Repeat Polymorphism and X-Chromosome Inactivation in Children with Premature Adrenarche J. Clin. Endocrinol. Metab., April 1, 2008; 93(4): 1304 - 1309. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Xita, I. Georgiou, L. Lazaros, V. Psofaki, G Kolios, and A. Tsatsoulis The role of sex hormone-binding globulin and androgen receptor gene variants in the development of polycystic ovary syndrome Hum. Reprod., March 1, 2008; 23(3): 693 - 698. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K.A. Walters, C.M. Allan, and D.J. Handelsman Androgen Actions and the Ovary Biol Reprod, March 1, 2008; 78(3): 380 - 389. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. K Ong, F. de Zegher, A. Lopez-Bermejo, D. B Dunger, and L. Ibanez Flutamide metformin for post-menarcheal girls with preclinical ovarian androgen excess: evidence for differential response by androgen receptor genotype Eur. J. Endocrinol., November 1, 2007; 157(5): 661 - 668. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. A. Walters, C. M. Allan, M. Jimenez, P. R. Lim, R. A. Davey, J. D. Zajac, P. Illingworth, and D. J. Handelsman Female Mice Haploinsufficient for an Inactivated Androgen Receptor (AR) Exhibit Age-Dependent Defects That Resemble the AR Null Phenotype of Dysfunctional Late Follicle Development, Ovulation, and Fertility Endocrinology, August 1, 2007; 148(8): 3674 - 3684. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Rajender, V. Rajani, N. J. Gupta, B. Chakravarty, L. Singh, and K. Thangaraj No Association of Androgen Receptor GGN Repeat Length Polymorphism With Infertility in Indian Men J Androl, November 1, 2006; 27(6): 785 - 789. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Hughes, M. Elgasim, R. Layfield, and W. Atiomo Genomic and post-genomic approaches to polycystic ovary syndrome--progress so far: Mini Review Hum. Reprod., November 1, 2006; 21(11): 2766 - 2775. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Mohlig, A. Jurgens, J. Spranger, K. Hoffmann, M. O Weickert, H. W Schlosser, T. Schill, G. Brabant, A. Schuring, A. F H Pfeiffer, et al. The androgen receptor CAG repeat modifies the impact of testosterone on insulin resistance in women with polycystic ovary syndrome. Eur. J. Endocrinol., July 1, 2006; 155(1): 127 - 130. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. E. Hickey, R. S. Legro, and R. J. Norman Epigenetic Modification of the X Chromosome Influences Susceptibility to Polycystic Ovary Syndrome J. Clin. Endocrinol. Metab., July 1, 2006; 91(7): 2789 - 2791. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D.H. Abbott, D.K. Barnett, C.M. Bruns, and D.A. Dumesic Androgen excess fetal programming of female reproduction: a developmental aetiology for polycystic ovary syndrome? Hum. Reprod. Update, July 1, 2005; 11(4): 357 - 374. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. F. Escobar-Morreale, M. Luque-Ramirez, and J. L. San Millan The Molecular-Genetic Basis of Functional Hyperandrogenism and the Polycystic Ovary Syndrome Endocr. Rev., April 1, 2005; 26(2): 251 - 282. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. A. Ehrmann Polycystic Ovary Syndrome N. Engl. J. Med., March 24, 2005; 352(12): 1223 - 1236. [Full Text] [PDF]
|
|
|
|
 |
|
 I. S. Brum, P. M. Spritzer, F. Paris, M. A. Maturana, F. Audran, and C. Sultan Association Between Androgen Receptor Gene CAG Repeat Polymorphism and Plasma Testosterone Levels in Postmenopausal Women Reproductive Sciences, February 1, 2005; 12(2): 135 - 141. [Abstract] [PDF]
|
|
|
|
|
|
M. Zitzmann, M. Depenbusch, J. Gromoll, and E. Nieschlag X-Chromosome Inactivation Patterns and Androgen Receptor Functionality Influence Phenotype and Social Characteristics as Well as Pharmacogenetics of Testosterone Therapy in Klinefelter Patients J. Clin. Endocrinol. Metab., December 1, 2004; 89(12): 6208 - 6217. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Lattuada, P. Vigano, E. Somigliana, M. P. Odorizzi, M. Vignali, and A. M. Di Blasio Androgen Receptor Gene Cytosine, Adenine, and Guanine Trinucleotide Repeats in Patients With Endometriosis Reproductive Sciences, May 1, 2004; 11(4): 237 - 240. [Abstract] [PDF]
|
|
|
|
 |
|
 L Pinson, J Auge, S Audollent, G Mattei, H Etchevers, N Gigarel, F Razavi, D Lacombe, S Odent, M Le Merrer, et al. Embryonic expression of the human MID1 gene and its mutations in Opitz syndrome J. Med. Genet., May 1, 2004; 41(5): 381 - 386. [Full Text] [PDF]
|
|
|
|
|
|
L. Ibanez, K. K. Ong, N. Mongan, J. Jaaskelainen, M. V. Marcos, I. A. Hughes, F. de Zegher, and D. B. Dunger Androgen Receptor Gene CAG Repeat Polymorphism in the Development of Ovarian Hyperandrogenism J. Clin. Endocrinol. Metab., July 1, 2003; 88(7): 3333 - 3338. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Dejager, H. Bry-Gauillard, E. Bruckert, B. Eymard, F. Salachas, E. LeGuern, S. Tardieu, R. Chadarevian, P. Giral, and G. Turpin A Comprehensive Endocrine Description of Kennedy's Disease Revealing Androgen Insensitivity Linked to CAG Repeat Length J. Clin. Endocrinol. Metab., August 1, 2002; 87(8): 3893 - 3901. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. B. Seminara and W. F. Crowley Jr. Genetic Approaches to Unraveling Reproductive Disorders: Examples of Bedside to Bench Research in the Genomic Era Endocr. Rev., June 1, 2002; 23(3): 382 - 392. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
