Type 2 diabetes (T2D) is a common and complex metabolic disease that is associated with serious complications and premature mortality. Several prospective studies and clinical experiments support the role of sex hormones in the development of T2D.

This dissertation examines the roles of sex hormones and sex hormone-binding globulin (SHBG) in the development of T2D as well as the determinants of SHBG levels. Chapter 1 introduces general background information regarding sex hormones and SHBG in relation to the pathogenesis of T2D as well as determinants of plasma SHBG levels. Chapter 2 explores whether the levels of sex hormones and SHBG may play a role in the inverse relation between coffee intake and the risk of T2D in the Women's Health Study. Chapter 3 examines the potential determinants of circulating levels of SHBG in a pooled analysis of studies included in the Women's Health Initiative Randomized Trial and Observational Study. Chapter 4 investigates roles of sex hormone pathway genes in the development of T2D and examines consistency across African-American women, Hispanic women, and European-American based on data from the Women's Health Initiative SNP Health Association Resource (WHI-SHARe) and the WHI Genomics and Randomized Trials Network (WHI-GARNET).

Our findings indicated that the observed inverse associations of caffeinated-coffee and caffeine intakes with the risk of T2D were substantially attenuated after taking into account of SHBG levels. Findings from a pooled analysis of studies included in the Women's Health Initiative suggested that age, usage of exogenous estrogen, BMI, and lifestyle factors may be associated with plasma levels of SHBG. In a genetic study examining associations between sex hormone pathway genes and T2D risk, possible associations between genes involved in sex hormones synthesis and metabolism, estrogen receptors, and androgen receptor and the risk of T2D were seen in African Americans, and possible relations between genes involved in the metabolism of androgen and progesterone receptor and the risk of T2D risk were seen in Hispanics. The directions of effect estimates were generally consistent across ethnicity.

These findings support the notion that sex hormones and SHBG may play roles in the pathogenesis of T2D. Also, the findings suggest that age, usage of exogenous estrogen, physical activity, regular coffee intake, and adiposity may play roles in the regulation of SHBG metabolism. Further studies in replicative and experimental settings are warranted to confirm the findings.

Type 2 diabetes (T2D) is a metabolic disorder that is characterized by high blood glucose in the context of insulin resistance and β-cell dysfunction. Excessive body weight is associated with an increased risk of T2D and other major chronic diseases including cardiovascular disease and certain cancers. The current work focuses on the identification of systemic biomarkers in various biological samples, including serum, urine, and DNA, as indicators for environmental risk factors and the development of T2D and obesity in different human populations. Environmental endocrine-disrupting chemicals (EDCs) have been shown to affect the biosynthesis, secretion, transport, binding action, and metabolism of endogenous sex hormones, and were suggested to be associated with increased risk of developing T2D and obesity. In Chapter 1, we conducted a meta-analysis study to comprehensively assess the role of EDCs in affecting risk of T2D and related metabolic traits. We found that both persistent (including dioxin, polychlorinated biphenyls, and chlorinated pesticides) and non-persistent EDCs (including bisphenol A and phthalates) may affect the risk of developing T2D. To evaluate urinary concentrations of BPA and major phthalate metabolites in relation to prospective weight change, we conducted a prospective analysis in the Nurses' Health Study (NHS) and NHSII in Chapter 2. Our data suggest urinary concentrations of BPA and certain individual phthalate metabolites were associated with modestly greater weight gain in a dose-response fashion. These data are consistent with a potential role of BPA and phthalates in weight gain, although more prospective data are needed to corroborate these observations.

In previous studies, T2D was associated with biomarkers on different molecular pathways. Earlier prospective studies have shown that low birth weight (LBW), an indicator of intrauterine growth restriction, was predictive of higher risk of T2D in adulthood, but the pathways that lead LBW individuals to the onset of T2D are still unclear. In Chapter 3, we assessed the effect of LBW on T2D risk that is explained by potential mediators on different biological pathways using mediation modeling in a case-control study of T2D in the Women's Health Initiative (WHI). We found that the effect of LBW on T2D risk seems mainly mediated by insulin resistance, which is further explained by circulating levels of sex hormone-binding globulin (SHBG), E-selectin, and systolic blood pressure. These prospective data provide quantifiable mechanistic evidence linking LBW to increased risk of T2D whilst presenting risk stratification and intervention in a population at greater risk of developing T2D later in life.

The process of biological aging involves in the pathogenesis process of metabolic disorders including T2D and obesity. It is also necessary to examine the potential predictors and determinants of biological aging. Despite the consistent observations that women outlive men, the mechanisms underlying the relation between sex and longevity have not been fully elucidated. In Chapter 4, we examined the associations of circulating estradiol and testosterone with leukocyte TL in WHI. We found that serum concentration of estradiol was not associated with leukocyte TL in this large sample of postmenopausal women. Total and free testosterone levels were inversely associated with TL in Asian/Pacific Islander women but not in black and Hispanic women. Future studies to replicate our observations are warranted to address potential ethnicity-specific relations.

Much work has targeted the detection of disease genes through genetic mapping for metabolic diseases such as type 2 diabetes (T2D), cardiovascular diseases (CVD), and other diabetes-related traits such as body mass index (BMI) and hemoglobin (HbA1C) levels. However, the etiology of metabolic diseases remains partially understood hampering the development of more personalized diagnosis, treatment and prevention strategies.

This dissertation examines the association between genetic variants with risk of metabolic diseases and diabetes-related quantitative traits in both candidate gene and genome-wide scan settings. In particular, we assessed the association of genetic loci related to adiposity, inflammation, and lipid storage, with the risk of diabetes using a candidate gene approach. We also investigated biological pathways that may give rise to the development of vascular disease (T2D and/or CVD) and also further investigated genetic variants related to BMI and HbA1C levels using a genome-wide approach. Chapter 1 introduces general background on the evolution of genetic research in the arena of metabolic diseases. Chapter 2 investigates common variants in the genomic region of FABP4, CRP, TNF, IL6 and PPARG in relation to diabetes risk among postmenopausal women enrolled in the Women's Health Initiative Observational Study (WHI-OS). Chapter 3 examines whether common variants involved in vascular disease risk are clustered in multiple pathways among African and Hispanic American participants in the WHI SNP Health Association Resource (SHARe) cohort. Chapter 4 examines the association between genetic variants with BMI and HbA1C levels using a family-based genome-wide association approach among participants in the Framingham Heart Study (FHS).

Our main findings are: 1) Candidate gene-based studies indicate that variation exists across even the candidate gene regions. FABP4 genotypes were associated with reduced VCAM-1 levels, though none of the common genetic variants in the FABP4 gene examined were associated with risk of T2D. We also observed modest associations between TNF genetic variants and circulating concentrations of TNF-α-R2, although common variants of CRP, TNF, and IL6 genes were not associated with T2D risk. Using the example of the PPARG gene with T2D risk, however, we replicated the association between the PPARG Pro12Ala genetic variant with diabetes risk and found that haplotype-based analysis is more powerful than single-SNP analysis for identifying genetic variants. 2) Using a pathway-based analytical approach and genome-wide scan data collected among African and Hispanic American postmenopausal women, we observed that genetic variants associated with vascular disease appeared to cluster into several biological pathways including the glycerolipid metabolism and PPAR signaling pathways. 3) We found strong associations between SNPs near the LOC100507205 locus and BMI in the family-based Framingham Heart Study with three generations. We also replicated five well-validated genes that have been previously reported to be significantly associated with the BMI trait. These findings contribute to the growing body of literature in identifying genetic variants in the development of metabolic disease, further future work (e.g. in the area of structure and functional variants) are warranted to improve understanding of the genetic architecture for metabolic outcomes. Increasing integration of cutting edge genomic science into population-based epidemiologic investigation will accelerate and improve our understanding of the genetic susceptibility of complex diseases. The work described in this dissertation represents a tip of our effort toward the ultimate improvement of the diagnosis, treatment and prevention of metabolic diseases in human populations.

Objective

Design

Setting

Participants

Main outcome measure

Results

Conclusions

Background

Methods and results

Conclusions