Human population based genetic studies to elucidate the biology of NAFLD

 描述（由申请人提供）：非酒精性脂肪性肝病（NAFLD）是由肝脏脂肪变性（脂质蓄积）引起的，在美国影响多达2900万成年人， 到2020年全球肝病的主要原因。有一些有效的方法来预防或治疗NAFLD。需要更好地了解NAFLD的病因，以提高其诊断和治疗。我们先前确定NAFLD是可遗传的（遗传影响），并在与肝脂肪变性相关的5个基因座上鉴定出常见的（次要等位基因频率（MAF）>5%）单核苷酸多态性（SNP），首先在约7，000名欧洲血统的个体中，然后在不同血统中。5个位点的变异解释了~20%的遗传力，表明存在更有影响力的因果变异。我们现在的目标是确定1。因果遗传变异2.基因（S），他们通过发挥其作用，和3。导致疾病的机制。我们将发现样本的样本量和祖先多样性增加了两倍，从而提高了我们识别和精细绘制因果变异的能力。我们将全基因组基因型归因于新的更密集和祖先多样性的归因小组，提高我们识别跨祖先的因果常见和现在的低频（MAF 1-5%）变体的能力。我们将使用新的分析方法，不仅进行单一变异，而且进行条件和基于基因的分析，使我们能够有效地识别致病的单一变异和基因。我们将使用新的变异和基因注释算法来优先考虑功能随访的变异和基因。使用这种方法，我们确定了新的假定的因果NAFLD变异和基因错过了以前的分析。最后，我们开发了一种NAFLD的人类细胞系模型，我们使用该模型来表明，改变上述方法优先考虑的三个基因的功能会导致肝脏甘油三酯积累增加，这表明这是这些基因导致NAFLD的常见机制。我们假设，因果NAFLD促进人类遗传变异具有跨祖先的影响，并在肝细胞中以细胞自主的方式发挥其作用，以增加甘油三酯的积累。为了检验这一假设，我们将协调来自10个发现队列（n> 21 K）的计算机断层扫描肝脂肪变性表型和基因型，将基因型插补到1000个基因组世界性参考组，并在各组之间进行单变异、基于基因和条件Meta分析，以识别、精细映射和注释推定的因果变异和基因。我们将在超过4,000例组织学证实的NAFLD病例和约3,000例对照中随访最相关的变异，以复制我们的发现。我们将在人肝细胞系中敲低和过表达假定的致病基因（野生型和变体），并确定它们对甘油三酯积累的影响和遗传作用机制。我们将开始从我们以前的NAFLD荟萃分析中优先考虑的三个基因开始，并继续从1000个基因组荟萃分析中鉴定的新基因。这项工作将有助于确定导致NAFLD的遗传和代谢机制，并为开发新的生物标志物以及这种疾病的潜在治疗方法提供信息。

 DESCRIPTION (provided by applicant): Nonalcoholic fatty liver disease (NAFLD) is caused by hepatic steatosis (lipid accumulation), affects up to 29 million adults in the U.S. and will become the leading cause of liver disease worldwide by 2020. There are few effective ways to prevent or treat NAFLD. A better understanding of NAFLD etiology is needed to improve its diagnosis and treatment. We previously determined that NAFLD is heritable (genetically influenced) and identified common (minor allele frequency (MAF) >5%) single nucleotide polymorphisms (SNPs) at 5 loci that associate with hepatic steatosis, first in ~7,000 individuals of European ancestry and then across ancestries. Variants at the 5 loci explain ~20% of heritability, suggesting the existence of more influential causal variants. We now aim to identify 1. causal genetic variants 2. gene(s) they work through to exert their effects, and 3. mechanism(s) by which they lead to disease. We tripled the sample size and ancestral diversity of our discovery sample, thus improving our ability to identify and fine-map causal variants. We will impute genome- wide genotypes to new denser and ancestrally diverse imputation panels, increasing our ability to identify causal common and now low-frequency (MAF 1-5%) variants across ancestries. We will use new analytical methods to perform not only single variant but also conditional and gene-based analyses across ancestries allowing us to efficiently identify causal single variants and genes. We will use new variant and gene annotations algorithms to prioritize variants and genes for functional follow up. Using this approach we identify new putative causal NAFLD variants and genes missed by previous analyses. Finally, we developed a human cell line model of NAFLD that we use to show that altering the function of three genes prioritized by the above approaches results in increased liver triglyceride accumulation, suggesting this as a common mechanism by which these genes cause NAFLD. We hypothesize that causal NAFLD-promoting human genetic variants have effects across ancestries and exert their effects in a cell-autonomous manner in hepatocytes to increase triglyceride accumulation. To test this hypothesis, we will harmonize the computed tomography hepatic steatosis phenotype and genotypes from 10 discovery cohorts (n>21K), impute genotypes to the 1000 Genomes cosmopolitan reference panel, and perform single variant, gene-based, and conditional meta- analyses across groups to identify, fine-map, and annotate putative causal variants and genes. We will follow- up top associating variants in >4,000 histologically confirmed NAFLD cases and ~3,000 controls to replicate our findings. We will knockdown and overexpress putative causal genes (wild-type and variant) in human liver cell lines and determine their effect on triglyceride accumulation and genetic mechanism of action. We will begin with three genes prioritized from our previous NAFLD meta-analysis and proceed to new genes identified from the 1000 Genomes meta-analysis. This work will help define the genetic and metabolic mechanisms that cause NAFLD and inform development of new biomarkers as well as potential therapeutics for this condition.