A Network-based Approach to Associate HDL Subspeciation with Function

描述(申请人提供)：高密度脂蛋白(Hdl)是血液中蛋白质和脂质的复合体，在预防心血管疾病中起着关键作用。 尽管高密度脂蛋白的组成异质性和功能多样性，但在临床环境中，高密度脂蛋白仍然被普遍认为是一个主要作用于脂质运输的单一实体。最近，包括我们的研究在内的越来越多的证据表明，有许多独立的功能由不同的稳定亚种介导，这些亚种恰好与经典定义的“高密度脂蛋白”协同作用。不幸的是，人们对基础状态或患病状态下的高密度脂蛋白亚种形成知之甚少。我们实验室的长期目标是了解高密度脂蛋白预防心血管疾病的分子基础。这项应用的总体目标是开发、验证和标准化一种新的方法，该方法结合先进的蛋白质组分析、功能分析和基于网络的计算框架来识别正常人体血浆中的新的高密度脂蛋白物种，并将它们与已知的高密度脂蛋白功能相关联。我们的假设是，高密度脂蛋白由许多不同的颗粒亚群组成，每个亚群都含有独特的蛋白质组成，发挥着从胆固醇运输到血管信号转导到先天免疫功能等不同的生理功能。我们将追求以下三个具体目标：1.高密度脂蛋白亚组分的蛋白质组学特征和功能图谱。我们假设，高密度脂蛋白颗粒的蛋白质组组成的变化导致每个亚种的功能能力不同。在我们的 初步研究，我们开发了四种正交分离技术。我们将使用质谱学来描述每种分离技术产生的10-20个组分的蛋白质丰度水平，并检查蛋白质对之间潜在的共迁移模式。这些组分将同时接受四项功能分析：(A)防止低密度脂蛋白(LDL)颗粒氧化的能力；(B)促进巨噬细胞胆固醇外流的能力；(C)对血管功能的影响，以内皮型一氧化氮合酶(ENOS)的激活衡量；以及(D)抑制激动剂诱导的血小板聚集。2.用综合方法预测高密度脂蛋白相互作用组网络。相互作用的蛋白质经常被发现具有共同的性质，例如相似的系统发育图谱和共表达模式。这些共同的特征已被证明是蛋白质相互作用的预测特征。我们假设，通过整合高密度脂蛋白的共迁移模式和四个最相关的特征，可以准确地预测高密度脂蛋白之间的相互作用。这些相互作用的蛋白质对可能共存于功能协同的高密度脂蛋白亚颗粒中。潜在的相互作用蛋白将通过免疫沉淀实验进行验证，测试结果将通过调整参数来进一步提高预测的准确性。3.从高密度脂蛋白交互组网络中识别负责已知高密度脂蛋白功能的功能模块。我们假设，使用基于网络的分类，与功能活动图谱最佳相关的功能模块可以从高密度脂蛋白相互作用组网络中识别出来。由一组高密度脂蛋白组成的模块可以对应于执行给定高密度脂蛋白功能的高密度脂蛋白颗粒的全部或部分。我们预计这些基于网络的模块将在重复性和准确性方面超过单个蛋白质作为高密度脂蛋白功能的标记。这将为未来的研究奠定基础，在该研究中，将使用遗传基因敲除小鼠模型来验证这种粒子-功能关系。这一应用具有很高的创新性，因为计算和实验方法的集成将以一种以前从未尝试过的方式揭示高密度脂蛋白亚种形成和功能之间的关系。因此，它将填补我们对高密度脂蛋白颗粒组成和功能异质性的理解的一个主要空白。这项工作将在几个方面产生重大影响：首先，该项目将通过同时识别新的高密度脂蛋白亚种并将它们与已知功能联系起来，促进我们对高密度脂蛋白功能的分子理解。第二，研究高密度脂蛋白相互作用网络可能揭示新的高密度脂蛋白功能。第三，经过验证的基于网络的方法也可以应用于将高密度脂蛋白亚型与心血管疾病状态相关联，从而产生有效的疾病生物标记物。最后，从长远来看，治疗策略可以是 旨在修饰某些高密度脂蛋白亚粒子或模仿其影响，以减少心血管疾病。

DESCRIPTION (provided by applicant): High density lipoproteins (HDL) are blood-borne complexes of protein and lipid that play critical roles in the prevention of cardiovascular disease (CVD), the major cause of mortality in the U.S. Despite its compositional heterogeneity and functional diversity, in a clinical setting, HDL is still commonly thought of as a single entity tht primarily functions in lipid transport. Recently, a growing body of evidence, including our research, has suggested there are numerous separate functions mediated by distinct stable subspecies which happen to cofractionate with classically defined "HDL". Unfortunately, little is understood about the HDL subspeciation in either basal or diseased states. The long-term goal of our laboratories is to understand the molecular basis of HDL's protection against CVD. The overall objective of this application is to develop, validate and standardize a novel approach which combines advanced proteomic analysis, functional assays and a network-based computational framework to identify new HDL species in normal human plasma and associate them with known HDL functions. Our hypothesis is that HDL is composed of numerous distinct particle subpopulations, each containing a unique protein make-up, which plays distinct physiological roles ranging from cholesterol transport to vascular signaling to innate immune function. We will pursue the following three specific aims: 1. Proteomic characterization and functional profiling of HDL sub-fractions. We hypothesize that variation in the proteomic composition of HDL particles results in different functional capacities for each subspecies. In our preliminary study, we have developed four orthogonal separation techniques. We will use mass spectrometry to profile the protein abundance level in 10-20 fractions derived from each separation technique, and examine potential co-migration patterns among protein pairs. The fractions will simultaneously be subjected to a panel of four functional assays: (a) ability to prevent oxidation of low density lipoprotein (LDL) particles; (b) ability to promote cholesterol efflux from macrophages; (c) effects on vascular function, measured as activation of endothelial nitric oxide synthase (eNOS); and (d) inhibition of agonist induced platelet aggregation. 2. Prediction of HDL interactome network using an integrative approach. Interacting proteins are often found to share common properties, e.g., similar phylogenetic profiles and co-expression patterns. These common characteristics have been shown to be predictive of protein interactions as features. We hypothesize that the interactions among HDL proteins can be accurately predicted by integrating their co-migration patterns and four most relevant features. These interacting protein pairs may co-exist in functionally synergistic HDL subparticles. Potential interacting proteins will be verified by immunoprecipitation experiments and the testing results will be used to further improve the accuracy of predictions by adjusting parameters. 3. Identification of functional modules responsible for known HDL functions from HDL interactome network. We hypothesize that, using a network-based classification, functional modules that optimally correlate with functional activity profiles can be identified from the HDL interactome network. A module, consisting of a group of HDL proteins, may correspond to the entirety or part of an HDL particle that carries out a given HDL function. We expect these network-based modules will outperform individual proteins as markers for HDL functions in both reproducibility and accuracy. This will set the stage for a future study where genetic knockout mouse models will be used to verify this particle-function relationship. This application is highly innovative because the integration of computational and experimental approaches will uncover the relationship between HDL subspeciation and function in a way that has not been attempted previously. As such, it will fill a major gap in our understanding of the compositional and functional heterogeneity of HDL particles. This work will have significant impacts on several fronts: First, the project will facilitate our molecular understanding of HDL functions by simultaneously identifying new HDL subspecies and linking them with known functions. Second, studying the HDL interactome network may reveal novel HDL functions. Third, the validated network-based approach can also be applicable to correlate HDL subspecies with CVD status, resulting in effective disease biomarkers. Finally, in the long term, therapeutic strategies can be designed to modify certain HDL subparticles or mimic their effects with the goal of reducing CVD.