Misfoldome-centered multiOMICS approach to unravel preeclampsia subphenotypes.

 描述(由申请人提供)：先兆子痫是一种特定于人类妊娠的异质性疾病，是全球孕产妇和新生儿发病率和死亡率的重要因素。到目前为止，除了分娩之外，还没有治愈先兆子痫的方法。在我们之前的蛋白质组学研究的带头下，我们发现子痫前期具有包括阿尔茨海默氏症在内的既定构象障碍的蛋白质错误折叠的特征。这些特征包括尿液同源性(对淀粉样嗜性染料刚果红的亲和力)，对构象状态依赖的抗体的亲和力，以及胎盘和蜕膜中淀粉样蛋白分解途径的异常调节。我们的主要假设是，子痫前期错误折叠蛋白的过度形成是由于缺陷清除和/或潜在的代谢障碍导致的大分子拥挤增加，从而导致错误的蛋白质折叠。因此，与正确折叠的蛋白质组相比，错误折叠的蛋白质(MisFoldome)可能是与疾病病因学更密切相关的生物标记物的丰富来源。我们建议使用现有的生物信息库来了解不同子痫前期亚型的基础。具体地说，我们的目标是发现与每一种先兆子痫亚型相关的特定标记物和可用药靶点，这些标记物和药物靶点可以在显性疾病发作之前纠正。为了实现这些目标，我们将研究4个“组学”层面：蛋白质组学、转录组学、代谢组学和表观组学，它们通过机器学习生物信息学方法独特地整合在一起，旨在解决复杂和相互关联的系统生物学数据。这些算法包括：线性判别分析(LDA)、条件随机场(CRF)和用于集成多组学数据层的“模糊”软聚类算法。特定目标1计划 应用鸟枪式自下而上的蛋白质组学方法，对不同临床亚型子痫前期患者尿液中反映的错折叠穹顶蛋白成分进行分类。将分析错折叠穹顶蛋白序列中的蛋白质和生物物理特性，并与总尿蛋白和血清蛋白质组进行比较。具体目标2计划通过蛋白质组学与转录组学(胎盘、绒毛和蜕膜组织的RNAseq)和代谢组学(血清和尿液)数据的三角测量来阐明人们高度感兴趣的生物途径。最后，特殊目标3将验证尿液嗜多染红细胞和新发现的在妇女第一次怀孕期间纵向跟踪的大型生物储存库中的分子特征(nuMoM2B队列)。总之，这项提案的三个目标提供了一个独特的机会，在临床明显疾病发作之前，为先兆子痫的个性化治疗选择提供了一个独特的机会。

 DESCRIPTION (provided by applicant): Preeclampsia is a heterogeneous disorder specific to human pregnancy and an important contributor to maternal and neonatal morbidity and mortality worldwide. To date, there is no cure for preeclampsia except delivery. Spearheaded by our prior proteomics research, we discovered that preeclampsia shares characteristics of protein misfolding with established conformational disorders including Alzheimer's. These features involve urine congophilia (affinity for the amyloidophilic dye Congo red), affinity for conformational state-dependent antibodies, and dysregulation in the amyloid proteolytic pathway in the placenta and decidua. Our overarching hypothesis is that the excessive formation of misfolded proteins in preeclampsia is driven by increased macromolecular crowding due to defective clearance and/or underlying metabolic disorders leading to faulty protein folding. As a result, the universe of misfolded proteins (misfoldome) could be a rich source of biomarkers more closely related to disease etiology than the properly folded proteome. We propose to use existing biorepositories to understand the underpinnings of different subtypes of preeclampsia. Specifically, we aim to discover specific markers and druggable targets relevant to each preeclampsia subtype that can be corrected before the onset of manifest disease. To achieve these goals we will investigate 4 "omics" layers: proteomics, transcriptomics, metabolomics and phenomics uniquely integrated through machine learning bioinformatics approaches aimed to solve complex and interconnected systems biology data. These include: Linear Discriminant Analysis (LDA), Conditional Random Fields (CRFs) and "fuzzy" soft clustering algorithms for integration of multi-omics data layers. Specific Aim 1 plans to apply shotgun bottom-up proteomics methods to catalogue the protein components of the misfoldome as reflected in urine congophilic aggregates of women with various clinical subphenotypes of preeclampsia. The proteins and biophysical characteristics of peptide sequences in the misfoldome will be analyzed and compared with those of total urine and serum proteomes. Specific Aim 2 plans to illuminate biological pathways of high interest by triangulating proteomics with transcriptomics (RNAseq on placental villous and decidual tissues) and metabolomics (serum and urine) data. Lastly, Specific Aim 3 will validate urine congophila and the newly discovered molecular signatures in a large biorepository of women followed longitudinally during their first pregnancy (nuMoM2b cohort). Together, the three aims of this proposal offer a unique opportunity toward personalized therapeutic options for preeclampsia before the onset of clinically manifest disease.