Estrogen receptor alpha gene polymorphism associated with type 2 diabetes mellitus and the serum lipid concentration in Chinese women in Guangzhou

Type 2 diabetes mellitus is thought to be a multifactorial disease and both genetic and acquired factors contribute to its pathogenesis. Identification of the susceptibility genes for type 2 diabetes mellitus thus may lead to primary prevention of the disease. In most patients, type 2 diabetes mellitus results from genetic changes, each has a partial and additive effect. The inheritance pattern is thus complex, and environmental factors play an important role in favoring or delaying the expression of the disease.

Estrogens exert its physiological effects mainly through the estrogen receptors (ERs), ERメ and ERモ, and might prevent menopause syndrome, osteoporosis, arteriosclerosis, and diabetes, etc. ER is found in many tissues, including the pancreatic islets¹ and skeletal muscle,² adipose tissue,³ and cardiovascular system, which all participate in the pathogenesis of type 2 diabetes. Epidemiological findings have indicated that the prevalence of type 2 diabetes mellitus is significantly lower in patients on estrogen replacement therapy, and estrogen seems useful for the treatment of diabetes.^4,5 Type 2 diabetes mellitus is a heterogeneous disorder caused by a combination of genetic and acquired abnormalities that affect insulin sensitivity and insulin secretion.⁶ Estrogen has been implicated in adipose tissue biology,^1,7 stimulation of liver fatty acid metabolism,^4,8 suppression of hepatic glucose production,⁹ reduction of both hyperglycemia and plasma insulin levels, protection of pancreatic モ-cell function/survival and insulin secretion in conditions of oxidative stress, and the blunting of neuroendocrine and metabolic responses to hypoglycemia.¹⁰ Estrogen modulates insulin secretion, regulates K-ATP channel activity, and regulates calcium signals via plasma membrane receptors.^11,12 Thus, the gene encoding ERメ gene (ESR1, gene map locus 6q25.1) is a potential candidate gene for susceptibility to type 2 diabetes. A number of polymorphisms in ESR1 have been reported. The PvuII and XbaI restriction fragment length polymorphisms (RFLPs) are two commonly employed markers for genetic analyses of the ESR1; both are located in intron 1 of the ESR1. Association between ESR1 and disease has been demonstrated in several cases. An association has been found between the P allele (an allele with a T/C transition) and the absent expressed ESR1 gene in breast cancer cell and breast tumor tissue.¹³ PvuII polymorphism may affect the splicing of ER mRNA, resulting in the alteration of protein expression. XbaI and PvuII RFLPs of ESR1 may play a role in the pathogenesis of such diseases as breast and prostate cancer, osteoporosis, Alzheimer's disease, cardiovascular diseases, and stroke.

In the present work, XbaI and PvuII polymorphisms (rs9340799 and rs2234693) of the ESR1 gene were studied in Southern Chinese people of Guangzhou, which had higher type 2 diabetes mellitus incidences than the average incidence in Chinese cities. Genotype and allele frequencies in postmenopausal women and men of similar age were compared between type 2 diabetic patients and control subjects, these results would be supported by analysis of the relation between blood glucose levels and the genotype distribution among the healthy subjects whose blood glucose had not been controlled by the drugs.

The alarming increase in the prevalence of type 2 diabetes mellitus and associated traits such as obesity, dyslipidemias, and hypertension in the overall population has become a worldwide challenge for health care systems. Estrogen has beneficial effects on lipids and lipoproteins metabolism.^3,5,8 In order to find if the ESR1 variant relates to serum lipid level, which may contribute to the mechanism causing type 2 diabetes, the relationship between the gene polymorphism and serum lipid level was also investigated.

Subjects
From 2004 to 2006, participants were selected who were Guangzhou residents, whose ancestral homes were in Guangdong Province, aged 51－70 years, not consanguineous, and had no significant liver damage or renal dysfunction. Patients and control subjects had no history of a sex-hormone-dependent disease and never received hormone replacement therapy. All women were postmenopausal. Informed written consent was obtained from each individual. The Sun Yat-sen Medical College Institutional Human Research Committee approved this study.

Diagnosis of type 2 diabetes mellitus was based on WHO criteria (1999). Fasting blood glucose (FBG) and postprandial blood glucose (PBG) (2 hours) were collected and separated within 30 minutes of venipuncture. A standard oral glucose tolerance test was performed for diagnosis of type 2 diabetes mellitus (FBG∶7.0 mmol/L, or PBG∶11.1 mmol/L). Individuals whose blood relation of family were no diabetes mellitus, with FBG <5.6 mmol/L and 2-hour-PBG <7.8 mmol/L were put in the healthy group. Others with borderline values (7.0 mmol/L >FBG∶5.6 mmol/L, or 11.1 mmol/L >PBG∶7.8 mmol/L) were excluded. All biochemical measurements were performed in the Diabetes Center of Sun Yat-sen Memorial Hospital using standard enzymatic methods, with quality control in the procedure. ESR1 genotyping was performed on the 341 healthy controls and 299 type 2 diabetic patients that were finally selected. All the data were put in the excel tables. The clinical characteristics of subjects are shown in Table 1.

Genotyping
Peripheral venous blood was collected and stored in －70ⅹC, and defrosted to extract DNA. Genomic DNA was extracted from peripheral blood leukocytes using a blood kit (Takara, Tokyo, Japan). PCR-RFLP assays were used to determine the ESR1 genotypes. The oligonucleotide primers used to determine the PvuII polymorphisms within the ESR1 gene were the same as those used by Yaich et al¹⁴ (forward, 5'-CTGCCACCCTATCTGTATCTTTTCCTATTCTCC-3'; reverse, 5'-TCTTTCTCTGCCACCCTGGCGTC GATTATCTGA-3'). PCR was performed in an AB2700 (Applied Biosystems, Foster City, CA, USA) and a PTC100 (MJ Research, Waltham, MA, USA). The PCR product of approximately 1.3 kb was completely digested with restriction endonuclease PvuII or XbaI for 8 hours, and electrophoresed in a 1.5% agarose gel. The PvuII polymorphism (P/p), which is caused by a T/C transition in intron 1, is located approximately 0.4 kb upstream of exon 2. The XbaI polymorphism (X/x), which is caused by an A/G transition in intron 1, is approximately 50 bp away from the PvuII polymorphism site. Gels were stained with SYBR green I, visualized under ultraviolet light, and photographed. P and X alleles were identified by the absence of the PvuII and XbaI sites, respectively, and p and x alleles by the presence of these sites. The assay was repeated and the polymorphisms were considered specific and not the result of experimental variation, if identical results were obtained in 18 or more of 100 samples tested.

Statistical analysis
All quantitative variables are expressed as mean 【 standard deviation (SD). The characteristics in Table 1 were assessed using the Wilcoxon rank-sum test. The effects of the allelic frequency on quantitative variables were assessed using t test. Genotypes differences between type 2 diabetic patients and healthy subjects were analyzed (adjusting for age or sex by Logistic regression analysis) to identify those associated with risk of type 2 diabetes. Stepwise Logistic regression analysis was performed to estimate the type 2 diabetes mellitus risk factors, include the age, the sex, the PvuII polymorphisms, the XbaI polymorphisms and the possible relationships between these factors.

Only the genotype of healthy subjects was used in association analyses of clinical variables, because type 2 diabetes mellitus treatments may affect those variables. Stepwise Logistic regression analysis was performed to estimate the risk factors of the clinical characteristics [FBG, PBG, total cholesterol (TC), triglyceride (TG), low density lipoprotein (LDL)-c, or high density lipoprotein (HDL)-c, respectively], including the age, the sex, the PvuII polymorphisms, the XbaI polymorphisms and the possible relationships between these factors. The nominal threshold for significance was set at P<0.05 for comparison of baseline and association analyses. All calculations were performed using the SPSS v10.0 statistical package.

Controls accord with Hardy-Weinberg equilibrium (HWE)，and cases did not. When the two RFLPs of ESR1 genotypes were combined in all subjects, all 9 genotypes (PPXX, PPXx, PPxx, PpXx, Ppxx, ppXx, ppxx, PpXX and ppXX) were recognized (Table 2), both p allele and x allele were the majorities allele, P allele and X allele were the rare ones.

In case-control study, genotypes and allele frequencies of PvuII or XbaI polymorphisms in the control subjects and the patients were shown in Table 2. Allele frequencies of the PvuII polymorphism (P=0.001), but not the XbaI polymorphism (P=0.122), were significantly different between control subjects and patients. Analysis of genotype distribution (pp vs Pp+PP) of PvuII in different sex (Table 3) found a significant difference in women (P<0.001) but not in men (P=0.854) when the age factor was controlled by Logistic regression; Analysis of genotype distribution (pp vs Pp+PP) in PvuII found a significant difference in 51－60 years (OR=1.900, 95%CI=1.222-2.954, P=0.004) and 61－70 years age group (OR=2.176, 95%CI= 1.322-3.582, P=0.002) when the sex factor was controlled by Logistic regression. It was showed that the XbaI variant was not related to the type 2 diabetes mellitus (Table 4).

view in a new window

Table 3. Contributions of genotype in PvuII polymorphisms to the risk for type 2 diabetes mellitus (after controlling sex or age by Logistic regressions)

view in a new window

Table 4. Contributions of genotype in XbaI polymorphisms to the risk for type 2 diabetes mellitus after controlling sex or age by Logistic regressions

Stepwise Logistic regressions showed that the risk factors associated with type 2 diabetes mellitus were PvuII variant (OR=1.795, 95%CI=1.212-2.658, P=0.003), age (OR=2.823, 95%CI=1.908-4.117, P<0.001), sex (OR=1.621, 95%CI=1.073-2.448, P=0.022). That means age factor can significantly affect the relationship between the PvuII variant genotype and the female type 2 diabetes mellitus in case-control study (Table 5).

view in a new window

Table 5. Risk factors associated with type 2 diabetes mellitus in stepwise Logistic regressions of the PvuII and XbaI genotypes

In addition, most of the patients were in drugs treatment, so the glucose and lipid data had been declined to normal and were not analyzed in our study. Only in the healthy control group, analysis of the associations between ESR1 genotypes and clinical characteristics showed that PvuII genotype was significantly related to higher blood glucose concentration (FBG, OR=1.777, 95%CI=1.098-2.876, P=0.019; PBG, OR=2.112, 95%CI=1.296-3.444, P=0.003) in women with PP and Pp genotypes than in those with pp genotypes (Table 6), and stepwise Logistic regressions showed that the risk factors associated with FBG and PBG were PvuII variant (FBG: OR=1.484, 95%CI=1.009-2.183, P=0.045; PBG: OR=1.824, 95%CI=1.239-2.686, P=0.002), age (FBG: OR=2.603, 95%CI=1.767-3.834, P<0.001; PBG: OR=2.698, 95%CI=1.834-3.968, P<0.001), sex (FBG: OR=1.908, 95%CI=1.259-2.891, P=0.002; PBG: OR=1.617, 95%CI=1.071-2.441, P=0.022). That means age factor can significantly affect the relationship between the PvuII variant genotype and the female blood glucose in healthy control people.

view in a new window

Table 6. Association between haplotypes and the level of blood glucose or serum lipid after controlling sex and age by Logistic regressions in healthy control [OR (95%CI)]

In healthy control group, the analysis after controlling age or sex found PvuII genotype was significantly related to female higher serum lipid concentration (TC, OR=1.683, 95%CI=1.140-2.484, P=0.009; LDL-c, OR=1.570, 95%CI=1.067-2.316, P=0.022) in women with PP and Pp genotypes than in those with pp genotypes (Table 6). At the same time, stepwise Logistic regression analysis showed that there might be a relationship between the genotype and TC, LDL-c. So, all these results suggest that PvuII genotype was related to female TC and LDL-c. It was also showed that the XbaI variant was not related to the PBG, FBG, TC, LDL-c, HDL-c in healthy control, but affected the healthy control TG (OR=0.667, 95%CI=0.458-0.972, P=0.045) in women, which were confirmed by the results of stepwise Logistic regression analysis, indicating a protective effect of the XbaI variant in TG (OR=0.632, 95%CI=0.432-0.926, P=0.019), which should be validated by study of larger samples.

In total, the PvuII variant genotype was found significantly related to type 2 diabetes mellitus of the women in case-control study, and also significantly related to PBG, FBG, TC, and LDL-c of the healthy female controls; the XbaI variant genotype was only significantly related to the TG of the healthy control women.

DISCUSSION

ERメ is a member of the nuclear steroid receptor family and a ligand-inducible transcription factor. ERs are found in the cytosol, nucleus, and plasma membrane. ER isoforms are expressed in most tissues of the human body. The human ESR1 (located on chromosome 6q25.1) has eight exons separated by seven intronic regions and spans more than 140 kb.¹⁵ The genotype P and p, X and x distributing of health Guangzhou population in our study, was similar to other studies about Chinese healthy population,^16,17Japanese population¹⁸ and Caucasian population,¹⁹ had the majorities of p and x alleles respectively.

An association has been found between the P allele (an allele with a T/C transition) and the absent expressed ESR1 gene in breast cancer cell and breast tumor tissue.¹³ PvuII polymorphism may affect the splicing of ERメ mRNA, resulting in the alteration of protein expression. Herrington et al²⁰ noted that the P allele (which has a T/C transition associated with loss of the PvuII site) has a potential binding site for myeloblastosis (myb) transcription factors that, in the presence of B-myb, is capable of augmenting in vitro transcription of a downstream reporter construct 10-fold. Thus, in some settings, the presence of the P allele might amplify ERメ transcription. Because B-myb expression is itself responsive to estrogen activation, it may contribute to a signal-amplifying system producing augmented responses to estrogen in cell types that commonly express B-myb or related transcription factors. The X allele (XbaI restriction site) was also found more frequently in genomic DNA from breast cancer patients than from control subjects. PvuII polymorphism and XbaI polymorphism may be different in their effects of ESR1. A study of 49 Caucasians with type 2 diabetes mellitus showed a significant difference between type 2 diabetes mellitus group and control group of ESR1 in the allelic frequency of XbaI (significantly higher X allelic frequency in the type 2 diabetes mellitus group) but not PvuII polymorphisms, indicated that ESR1 may be associated with type 2 diabetes, but in this population, genotype distribution in the healthy is different from that of other Caucasians and Chinese. The genotype related to type 2 diabetes mellitus maybe just the reason why the cases groups did not accord with HWE.

In our case-control study, in the 51－70 years populations from Guangzhou, P allele was significantly more frequent in the type 2 diabetes mellitus group, and PvuII variant genotype of the ESR1 was significantly related to female type 2 diabetes mellitus, and improved by the increasing age, which never showed in XbaI genotype. At the same time, PvuII genotype of the ESR1 was significantly related to healthy female FBG and PBG in all age groups, and also improved by the increasing age, which also never showed in XbaI genotype, just like the tendency between the relationship of PvuII genotype and type 2 diabetes mellitus. FBG and PBG are the major factors to judge the type 2 diabetes mellitus, so these data from type 2 diabetes mellitus or health samples confirm that PvuII genotype of the ESR1 is significantly related to the occurring and development of female type 2 diabetes mellitus, and improved with the grown age.

In addition, ERメ can co-modulate with many important factors involved in the molecular mechanism of type 2 diabetes, such as Akt/PKB,²¹ insulin receptor substrate 1(IRS1),²² insulin-like growth factor receptor (IGFR),²³ islet amyloid polypeptide (IAPP),¹ peroxisome-proliferator- activated receptor (PPAR),²⁴ PPAR gamma- coactivator-1 (PGC-1),²⁵ PGC-1 related estrogen receptor coactivator (PERC),²⁶ caveolin-3 (Cln-3),²⁷ vitamin D receptor gene,⁷ etc. These factors affect insulin secretion, insulin resistance, and insulin signaling. ESR1 polymorphisms may influence many molecular mechanisms of type 2 diabetes mellitus through these factors indirectly.

New research data provided evidence that genes predisposing African-American individuals to type 2 diabetes mellitus were located in the 6q 24-q27 regions of the genome.²⁸ A Chinese research reported that type 2 diabetes mellitus was significantly linked to chromosome 6q21-q23, these regions on 6q were hot with type 2 diabetes.²⁹ ESR1 is just located in the 6q25.1 region. These data are very interesting to present study and need study in further.

Sex hormone may act on diabetogenic sensitivity in men and women in several ways. Induction of insulin-dependent diabetes by streptozotocin can be inhibited by estrogens and potentiated by androgens.^30,31 Estrogen appears to play a major role in the sexual dimorphism present in counter regulatory responses to hypoglycemia in healthy humans.³² The activity of glucose-6-phosphate dehydrogenase (G6PD) in rats is markedly enhanced by low doses of E2, which appears to be largely responsible for the sex differences in the activity of this enzyme in rats.⁹ Sex hormone mechanisms are obviously different between men and women. Our study found a significant difference in the genotype distribution and allele frequencies of PvuII polymorphism of women, but not of men; at the same time, genotype distribution of PvuII polymorphism also related to female FBG and PBG, not to the male FBG and PBG, suggesting ESR1 variant might be a risk factor for type 2 diabetes in women but not in men.

In this study, XbaI genotype was not related with the male type 2 diabetes mellitus and FBG，PBG of the healthy, but can affect the female TG. So, we guess that the role may be weak and complicated in the control people that would not be enough to develop type 2 diabetes mellitus. These results indicated that, for the two SNPs, difference in the locations might lead to the difference in the roles. However, the different roles must be studied further.

Type 2 diabetes is associated with obesity and dyslipidemias. Estrogen has beneficial effects on lipoprotein metabolism.^5,33,34 ERメ is the main estrogen receptor expressed in mouse white adipose tissue (WAT). Several genes involved in lipogenesis and lipolysis were identified as novel targets that could mediate estrogenic effects on adipose tissue.³ The endogenous and exogenous estrogens decrease LDL-cholesterol, increase HDL-cholesterol levels and lead to an increase in the mRNA for the LDL receptor.⁸ It is reported that PvuII polymorphism is related to LDL metabolism of Japanese school children. Furthermore, moderate fasting hypertriglyceridemia in noninsulin-dependent diabetes mellitus (NIDDM) is thought to be predictive of a constellation of postprandial changes in lipids and lipoproteins in NIDDM.³⁵ It was reported that the PP genotype(C/C allele), or any other closely related genotypes, have an augmented responds of HDL cholesterol to the hormone-replacement therapy. Our study suggests that ESR1 variant of healthy postmenopausal women might be a risk factor for higher serum TC and LDL-c content, which might be another ESR1-related mechanism causing type 2 diabetes in postmenopausal women.

These findings suggest that the PvuII polymorphism in ESR1 is associated with type 2 diabetes and related to the serum lipid levels in postmenopausal women. Further studies are needed: the molecular mechanism of type 2 diabetes pathogenesis mediated by the ESR1 PvuII polymorphism should be elucidated in experimental animals, while the distribution and effects of these polymorphisms are being investigated in other ethnic groups.

Acknowledgements: We are grateful to Prof. CHEN Hua, the Director of Diabetes Center of Sun Yat-sen Memorial Hospital for the support, and we also thank for the passionate help and responsible work of all the doctors, technicians and nurses in the Diabetes Center.

REFERENCES

1. Geisler JG, Zawalich W, Zawalich K, Lakey JR, Stukenbrok H, Milici AJ, et al. Estrogen can prevent or reverse obesity and diabetes in mice expressing human islet amyloid polypeptide. Diabetes 2002; 51: 2158-2169.

2. Capozza F, Combs TP, Cohen AW, Cho YR, Park SY, Schubert W, et al. Caveolin-3 knockout mice show increased adiposity and whole body insulin resistance, with ligand-induced insulin receptor instability in skeletal muscle. Am J Physiol Cell Physiol 2005; 88: C1317-1331.

3. Lundholm L, Moverare S, Steffensen KR, Nilsson M, Otsuki M, Ohlsson C, et al. Gene expression profiling identifies liver X receptor alpha as an estrogen-regulated gene in mouse adipose tissue. J Mol Endocrinol 2004; 32: 879-892.

4. Brussaard HE, Gevers Leuven JA, Frolich M, Kluft C, Krans HM. Short-term oestrogen replacement therapy improves insulin resistance, lipids and fibrinolysis in postmenopausal women with NIDDM. Diabetologia 1997; 40: 843-849.

5. Knopp RH, Zhu X. Multiple beneficial effects of estrogen on lipoprotein metabolism. J Clin Endocrinol Metab 1997; 82: 3952-3954.

6. Szoke E, Gerich JE. Role of impaired insulin secretion and insulin resistance in the pathogenesis of type 2 diabetes mellitus. Compr Ther 2005; 31: 106-112.

7. Speer G, Cseh K, Winkler G, Vargha P, Braun E, Takacs I, et al. Vitamin D and estrogen receptor gene polymorphisms in type 2 diabetes mellitus and in android type obesity. Eur J Endocrinol 2001; 144: 385-389.

8. Di Croce L, Bruscalupi G, Trentalance A. Independent behavior of rat liver LDL receptor and HMGCoA reductase under estrogen treatment. Biochem Biophys Res Commun 1996; 224: 345-350.

9. Ibim SE, Randall R, Han P, Musey PI. Modulation of hepatic glucose-6-phosphate dehydrogenase activity in male and female rats by estrogen. Life Sci 1989; 45: 1559-1565.

10. Sandoval DA, Ertl AC, Richardson MA, Tate DB, Davis SN. Estrogen blunts neuroendocrine and metabolic responses to hypoglycemia. Diabetes 2003; 52: 1749-1755.

11. Nadal A, Rovira JM, Laribi O, Leon-quinto T, Andreu E, Ripoll C, et al. Rapid insulinotropic effect of 17beta-estradiol via a plasma membrane receptor. FASEB J 1998; 12: 1341-1348.

12. Song RX, Barnes CJ, Zhang Z, Bao Y, Kumar R, Santen RJ. The role of Shc and insulin-like growth factor 1 receptor in mediating the translocation of estrogen receptor alpha to the plasma membrane. Proc Natl Acad Sci USA 2004; 101: 2076-2081.

13. Hill SM, Fuqua SA, Chamness GC, Greene GL, McGuire WL. Estrogen receptor expression in human breast cancer associated with an estrogen receptor gene restriction fragment length polymorphism. Cancer Res 1989; 49: 145-148.

14. Yaich L, Dupont WD, Cavener DR, Parl FF. Analysis of the PvuII restriction fragment-length polymorphism and exon structure of the estrogen receptor gene in breast cancer and peripheral blood. Cancer Res 1992; 52: 77-83.

15. Menasce LP, White GR, Harrison CJ, Boyle JM. Localization of the estrogen receptor locus (ESR) to chromosome 6q25.1 by FISH and a simple post-FISH banding technique. Genomics 1993; 17: 263-265.

16. Long J, Liu P, Zhang Y, Shen H, Liu Y, Dvornyk V, et al. Interaction effects between estrogen receptor alpha gene, vitamin D receptor gene, age, and sex on bone mineral density in Chinese. J Hum Genet 2003; 48: 514-519.

17. Liu JM, Zhu HM, Zhu XY, Dai M, Jiang L, Xu MY, et al. Estrogen receptor gene polymorphisms and bone mineral density in Chinese postmenopausal women. Chin Med J 2003; 116: 364-367.[PubMed]

18. Matsushita H, Kurabayashi T, Tomita M, Tanaka K. Effects of vitamin D and estrogen receptor gene polymorphisms on the changes in lumbar bone mineral density with multiple pregnancies in Japanese women. Hum Reprod 2004; 19: 59-63.

19. Deng HW, Li J, Li JL, Johnson M, Gong G, Recker RR. Association of VDR and estrogen receptor genotypes with bone mass in postmenopausal Caucasian women: different conclusions with different analyses and the implications. Osteoporos Int 1999; 9: 499-507.

20. Herrington DM, Howard TD, Brosnihan KB, McDonnell DP, Li X, Hawkins GA, et al. Common estrogen receptor polymorphism augments effects of hormone replacement therapy on E-selectin but not C-reactive protein. Circulation 2002; 105: 1879-1882.

21. Aquila S, Sisci D, Gentile M, Middea E, Catalano S, Carpino A, et al. Estrogen receptor (ER) alpha and ER beta are both expressed in human ejaculated spermatozoa: evidence of their direct interaction with phosphatidylinositol-3-OH kinase/Akt pathway. J Clin Endocrinol Metab 2004; 89: 1443-1451.

22. Ando' S, Panno ML, Salerno M, Sisci D, Mauro L, Lanzino M, et al. Role of IRS-1 signaling in insulin-induced modulation of estrogen receptors in breast cancer cells. Biochem Biophys Res Commun 1998; 253: 315-319.

23. Karl M, Potier M, Schulman IH, Rivera A, Werner H, Fornoni A, et al. Autocrine activation of the local insulin-like growth factor I system is up-regulated by estrogen receptor (ER)-independent estrogen actions and accounts for decreased ER expression in type 2 diabetic mesangial cells. Endocrinology 2005; 146: 889-900.

24. Sottile V, Seuwen K, Kneissel M. Enhanced marrow adipogenesis and bone resorption in estrogen-deprived rats treated with the PPARgamma agonist BRL49653 (rosiglitazone). Calcif Tissue Int 2004; 75: 329-337.

25. Bourdoncle A, Labesse G, Margueron R, Castet A, Cavailles V, Royer CA. The nuclear receptor coactivator PGC-1alpha exhibits modes of interaction with the estrogen receptor distinct from those of SRC-1. J Mol Biol 2005; 347: 921-934.

26. Hentschke M, Susens U, Borgmeyer U. PGC-1 and PERC, coactivators of the estrogen receptor-related receptor gamma. Biochem Biophys Res Commun 2002; 299: 872-879.

27. Wang X, Abdel-Rahman AA. Estrogen modulation of eNOS activity and its association with caveolin-3 and calmodulin in rat hearts. Am J Physiol Heart Circ Physiol 2002; 282: H2309-H2315.

28. Sale MM, Freedman BI, Langefeld CD, Williams AH, Hicks PJ, Colicigno CJ, et al. A genome-wide scan for type 2 diabetes in African-American families reveals evidence for a locus on chromosome 6q. Diabetes 2004; 53: 830-837,

29. Xiang K, Wang Y, Zheng T, Jia W, Li J, Chen L, et al. Genome-wide search for type 2 diabetes/Impaired glucose homeostasis susceptibility genes in the Chinese (significant linkage to chromosome 6q21-q23 and chromosome 1q21-q24). Diabetes 2004; 53: 228-234.

30. Paik SG, Michelis MA, Kim YT, Shin S. Induction of insulin-dependent diabetes by streptozotocin. Inhibition by estrogens and potentiation by androgens. Diabetes 1982; 31(8 Pt 1): 724-729.

31. Van Beek AP, Ruijter-Heijstek FC, Jansen H, Erkelens DW, de Bruin TW. Sex steroids and plasma lipoprotein levels in healthy women: The importance of androgens in the estrogen-deficient state. Metabolism 2004; 53: 187-192.

32. Arslanian SA, Heil BV, Becker DJ, Drash AL. Sexual dimorphism in insulin sensitivity in adolescents with insulin-dependent diabetes mellitus. J Clin Endocrinol Metab 1991; 72: 920-926.

33. Blum A, Cannon RO III. Selective estrogen receptor modulator effects on serum lipoproteins and vascular function in postmenopausal women and in hypercholesterolemic men. Ann N Y Acad Sci 2001; 949: 168-174.

34. Carr MC, Knopp RH, Brunzell JD, Wheeler BS, Zhu X, Lakshmanan M, et al. Effect of raloxifene on serum triglycerides in women with a history of hypertriglyceridemia while on oral estrogen therapy. Diabetes Care 2005; 28: 1555-1561.

35. Lewis GF, O'Meara NM, Soltys PA, Blackman JD, Iverius PH, Pugh WL, et al. Fasting hypertriglyceridemia in noninsulin-dependent diabetes mellitus is an important predictor of postprandial lipid and lipoprotein abnormalities. J Clin Endocrinol Metab 1991; 72: 934-944.