Understanding causal mechanisms in preeclampsia through genetic instrumental variables

项目概要/摘要 先兆子痫发生在3% - 6%的美国妇女中，并且是妊娠期间母体和胎儿发病的主要来源。 怀孕，怀孕后立即，对母亲和孩子的心血管健康有长期的影响。 先兆子痫母亲和婴儿在分娩后一年内的医疗保健管理费用平均超过28亿美元 美元每年。在美国，先兆子痫的患病率随着时间的推移而增加。非洲裔美国女性的 先兆子痫的患病率，更有可能患有严重的先兆子痫，死亡率是白色人的三倍 美国人因怀孕引起的并发症。没有有效的预防策略，先兆子痫存在部分日期 由于缺乏对因果关系的理解-无论是临床风险因素、遗传倾向还是社会介导的因素。 流行病学研究确定了先兆子痫的多种临床危险因素和预测因子，如肥胖，糖尿病， 既存高血压、慢性肾病和血栓形成倾向。研究还表明，在没有这些的女性中， 预先存在的条件，那些先兆子痫是在发展高血压，慢性肾脏疾病， 静脉血栓栓塞，中风和糖尿病5至10年后。推断这些关联的因果关系是困难的， 由于潜在的混杂和反向因果关系，仅流行病学数据。先兆子痫，它的许多危险因素， 它的后果是遗传的，数百种遗传变异被确定为一些特征，如血压，糖尿病和 肾功能由于遗传变异在一生中不会改变，也不会受到反向因果关系的影响， 由于孟德尔的遗传定律，它们不太容易混淆，孟德尔的遗传定律规定了随机分配和分离， 基因，我们可以设计孟德尔随机化（MR）实验，使用遗传变异作为工具变量， 在某些假设条件下，风险暴露可以有力地评估风险暴露与结果之间的因果关系。合并遗传数据 从多个来源和利用现有的EHR相关生物库（BioVU）在范德比尔特，我们形成了 先兆子痫遗传学网络（PEGNet）研究先兆子痫的遗传结构超过28，000先兆子痫 病例和290，000多例对照。使用PEGNet的先兆子痫数据和MR方法的最新进展，我们 我建议评估临床危险因素和先兆子痫的预测因素之间的因果关系，包括血液 血压、肾功能、肝功能、肥胖和代谢特征。我们将评估先兆子痫是否是 与未来心血管并发症相关，或者是否是由于反向因果关系。随着新兴的MR方法， 作为药物靶点MR，我们建议筛选与先兆子痫相关的基因和蛋白质靶点，并且也是可药物化的。我们 提出MR实验来验证现有药物的基因靶点，用于预防先兆子痫，包括 阿司匹林、二甲双胍、他汀类药物和PDE 5阻滞剂。我们提出了一个新的MR框架来阐明角色色彩主义，一个社会 构建基于肤色的区分，通过使用皮肤色素沉着的遗传变异作为 皮肤色调的仪器变量。我们提出混合映射，以了解遗传祖先如何影响 非洲血统的女性患上先兆子痫以创新的方式使用MR方法，我们的研究告知因果关系 子痫前期的发病机制，设计有效干预策略的第一步。

PROJECT SUMMARY/ABSTRACT Preeclampsia occurs in 3% – 6% of women in the US and is a leading source of maternal and fetal morbidity during pregnancy, immediately after pregnancy and has long term cardiovascular health implications for both mother and child. Cost of healthcare management of preeclamptic mothers and infants within one year of delivery averages over $2.8 billion dollars annually. Prevalence of preeclampsia is increasing overtime in the US. African American women have higher prevalence of preeclampsia, are more likely to have severe preeclampsia and are three times as likely to die as white Americans due to pregnancy related complications. No effective preventative strategy for preeclampsia exists to-date in part due to a lack of understanding of causality – be it clinical risk factors, genetic predisposition or socially mediated factors. Epidemiological studies identify multiple clinical risk factors anad predictors for preeclampsia such as obesity, diabetes, preexisting hypertension, chronic kidney disease, and thrombophilia. Studies also show that in women without these preexisting conditions, those with preeclampsia are at higher risk of developing hypertension, chronic kidney disease, venous thromboembolism, stroke and diabetes 5 to 10 years later. Inferring causality for these associations is difficult with epidemiologic data alone due to potential for confounding and reverse causation. Preeclampsia, many of its risk factors and its consequences are heritable with hundreds of genetic variants identified for some traits like blood pressure, diabetes and kidney function. Since genetic variants do not change during a lifetime and cannot be influenced by reverse causation, and are less prone to confounding due to Mendel's laws of inheritance which dictate random assortment and segregation of genes, we can design Mendelian randomization (MR) experiments to use genetic variants as instrumental variables for exposures to robustly evaluate causality between exposure and outcome under certain assumptions. Coalescing genetic data on preeclampsia from multiple sources and leveraging existing EHR-linked biobank (BioVU) at Vanderbilt, we form the PreEclampsia Genetics Network (PEGNet) to study the genetic architecture of preeclampsia in over 28,000 preeclampsia cases and over 290,000 controls. Using preeclampsia data from PEGNet and recent developments in MR methods, we propose to evaluate causal relationships between clinical risk factors and predictors of preeclampsia including blood pressure, kidney function, liver function, obesity and metabolic traits. We will evaluate whether preeclampsia is causally associated with future cardiovascular complications or if this is due to reverse causality. With emerging MR methods such as drug target MR, we propose to screen gene and protein targets associated with preeclampsia and also druggable. We propose MR experiments to validate gene targets for existing drugs in the pipeline for preeclampsia prevention including aspirin, metformin, statins, and PDE5 blockers. We propose a novel MR framework to elucidate the role colorism, a social construct of discrimination based on skin color, on preeclampsia risk by using genetic variants of skin pigmentation as instrumental variables for skin tone. We propose admixture mapping to understand how genetic ancestry influences preeclampsia in women of African ancestry. Using MR methods in innovative ways, our research informs causal mechanisms in preeclampsia, the first step necessary to design effective intervention strategies.