Genes and Gene Networks Associated with Obesity and Diabetes

描述(由申请人提供)：本项目的目标是确定在肥胖诱导的2型糖尿病的发展中起作用的基因和基因网络。在之前的资助期间，我们进行了第一次研究，调查了交叉杂交中胰岛基因的表达。该杂交系来源于对Leptinob突变引起的糖尿病具有抗性(C57BL/6或B6)或易感(BTBR)的小鼠品系。除了胰岛，我们还调查了其他五个组织(肝脏、脂肪、肌肉、下丘脑和肾脏)的基因表达。所有小鼠都接受了B6和BTBR之间5000个单核苷酸多态的基因分型。我们调查了F2样本中的100个临床性状，以及肝脏中数百个微小RNA(MiRNAs)。我们的研究产生了前所未有的数据集，使我们能够开发因果调控网络，预测基因、中间表型(如miRNAs和mRNAs)和下游临床特征之间的相互作用。具体地说，这些研究使我们能够：1)确定表达数量性状基因座(EQTL)； 2)确定“热点”，即似乎受单个基因座共同调控的大组转录本； 3)以遗传学为锚，建立定向eQTL网络模型；4)将eQTL网络模型与肥胖和糖尿病相关的关键生理表型整合。综上所述，遗传结构(eQTL和热点)和重建的网络已经确定了基因组区域和基因，这些基因组区域和基因可能包含构成肥胖依赖生理学基础的调节元件 2型糖尿病。我们已经使用我们的网络构建方法来识别与临床表型密切相关的基因。这一建议的具体目的是：1)识别和测试控制胰岛细胞周期基因的调控基因组位点，这些基因对细胞复制至关重要。改进和实验测试模型，以预测调节基因座和胰岛细胞周期基因之间的因果关系，这些基因负责细胞的增殖能力；2)识别和测试控制肝脏脂肪变性的调节基因组基因座。构建和完善共同调节的肝脏基因与肝脏甘油三酯积累相关的网络模型；3)利用miRNA丰度的遗传力来预测mRNA靶标，并将miRNAs、mRNAs和临床表型与可检验的模型联系起来。具体目标1-3的成功完成关键取决于新的统计和计算方法的发展。这些方法将包括：a)解决多个生理和表达性状映射到的遗传区间；b)扩展我们的网络重建方法，以推断性状之间的相关性并纳入先验信息；以及c)识别差异相关的基因集。 公共卫生相关性：该项目旨在了解肥胖引起的2型糖尿病的遗传因素。我们使用遗传和基因组方法来确定与这种疾病背后的关键代谢和病理过程有因果关系的基因。我们的研究将有助于理解部分肥胖者(但不是所有肥胖者)是如何患上糖尿病的，并可能开发生物标记物来识别患糖尿病风险最高的个人。

DESCRIPTION (provided by applicant): The objectives of this project are to identify genes and gene networks that play a role in the development of obesity-induced type 2 diabetes. In the previous grant period, we carried out the first-ever study that surveyed pancreatic islet gene expression in an intercross. The intercross was derived from mouse strains resistant (C57BL/6 or B6) or susceptible (BTBR) to diabetes induced by the leptinob mutation. In addition to islets, we surveyed gene expression in five other tissues (liver, adipose, muscle, hypothalamus, and kidney). All mice were genotyped for 5,000 single nucleotide polymorphisms between B6 and BTBR. We surveyed >100 clinical traits in the F2 sample, as well as hundreds of micro-RNAs (miRNAs) in liver. Our studies have yielded an unprecedented dataset, allowing us to develop causal regulatory networks that predict the interaction among genes, intermediate phenotypes such as miRNAs and mRNAs, and downstream clinical traits. Specifically these studies have allowed us to: 1) identify expression quantitative trait loci (eQTLs); 2) identify "hotspots", or large sets of transcripts that appear to be co-regulated by a single locus; 3) use genetics as an anchor to develop directional eQTL network models; and 4) integrate eQTL network models with key physiological phenotypes related to obesity and diabetes. Taken together, the genetic architecture (eQTL and hotspots) and reconstructed networks have identified genomic regions and genes that may contain regulatory elements that underlie the physiology of obesity-dependent type 2 diabetes. We have used our methods for network construction to identify genes that are closely associated with clinical phenotypes. The specific aims of this proposal are to: 1) Identify and test regulatory genomic loci that govern pancreatic islet cell cycle genes essential for ¿-cell replication. Refine and experimentally test models that predict the causal relationship between the regulatory loci and islet cell cycle genes responsible for the proliferative capacity of ¿-cells; 2) Identify and test the regulatory genomic loci governing hepatic steatosis. Construct and refine network models that relate co-regulated hepatic genes with accumulation of liver triglycerides; 3) Utilize the heritability of miRNA abundance to predict mRNA targets and relate miRNAs, mRNAs and clinical phenotypes with testable models. Successful completion of Specific Aims 1 - 3 is critically dependent on the development of novel statistical and computational methods. These will include methods to: a) resolve genetic intervals to which multiple physiological and expression traits map; b) extend our network reconstruction approach to infer dependence among traits and incorporate prior information; and c) identify differentially correlated gene sets. PUBLIC HEALTH RELEVANCE: This project is aimed at understanding the genetic factors responsible for obesity- induced type 2 diabetes. We use genetic and genomic approaches to identify genes that are causally related to the key metabolic and pathological processes underlying this disease. Our research will help to understand how some, but not all obese people become diabetic and perhaps develop biomarkers to identify individuals at greatest risk for developing diabetes.