Identifying secreted protein networks affecting human pancreatic islet function in type 2 diabetes using public omic databases

糖尿病的流行影响了大约10%的美国成年人。血糖水平升高是糖尿病的标志 糖尿病和极其重要的胰岛的内分泌激素的协调分泌 朗格汉斯是适当控制全身葡萄糖代谢所必需的。代谢应激增加 由于肥胖导致每种类型的胰岛细胞(a，b，d)通过改变其激素分泌来适应。但是，在 某些肥胖个体，这种适应的失败，破坏了胰岛微环境，导致血液升高。 血糖水平与2型糖尿病(T2D)发病的关系。胰岛细胞分化的潜在机制 影响彼此的功能尚不清楚。分泌蛋白是细胞内和细胞间的关键代谢类型 监管机构提高了我们对肥胖导致T2D的潜在机制的理解。因此， 这个项目的前提是胰岛中分泌的蛋白质介导的串扰是正常功能所必需的。 以及a、b、d细胞在瘦、肥胖和T2D状态下的适应。分泌蛋白约占总蛋白的11%。 人类转录组，我们的初步数据已经确定了大约850个差异表达的转录本 肥胖小鼠胰岛分泌蛋白的编码。然而，它们中只有一小部分的功能是 刻画得很好。我们的长期目标是确定能改善胰岛功能的分泌性蛋白质，用于治疗。 人类的T2D。实现这一目标的一个主要障碍是识别和成本高昂的技术限制 然而，使用常规生化方法对胰岛分泌蛋白进行功能鉴定非常耗时 接近了。在对一个高度严格的数据集的测试分析中，44个来自胰岛的分泌蛋白调节因子 被证实在肥胖时会影响小鼠的胰岛功能。有趣的是，顶部的功能表征 候选分泌蛋白导致了一种抑制b细胞胰岛素分泌的新途径的发现。 令人兴奋的是，验证我们的定量生物信息学框架的使用是朝着有效数据的飞跃 挖掘在加快鉴定与胰岛功能相关的新的分泌型蛋白调节剂中的作用 病情(S)。本研究的目的是确定影响人类胰岛功能的分泌蛋白调节因子。 T2D使用网络分析组合的公共可用的整个胰岛转录数据集。我们建议 以下目标旨在实现：1)确定候选的分泌型蛋白调节器；2)超簇 与人类肥胖和T2D相关的胰岛候选分泌蛋白的功能预测； 对影响胰岛功能的候选分泌蛋白进行生物学验证。成功完成后， 确定人类肥胖和T2D中胰岛功能的新调节因素，提高对机制的认识 潜在的人类T2D风险，并可能确定改善T2D胰岛功能的治疗靶点。 此外，通过整合多个数据集获得的洞察力说明了样品制备中的差异 以及测序(平台偏差)、测序深度和网络/关联架构(因样本而定 处理)将为阐明跨越不同胰岛细胞类型的分泌蛋白质网络奠定基础。

The diabetes epidemic affects ~10% of the US adult population. An elevated blood sugar level is the hallmark of diabetes, and the coordinated secretion of endocrine hormones from critically important pancreatic islets of Langerhans is required for the proper control of whole-body glucose metabolism. Increased metabolic stress due to obesity causes each islet cell type (a, b, d) to adapt by altering their hormone secretion. However, in certain obese individuals, failure of this adaptation, disrupts the islet microenvironment, leading to elevated blood glucose levels and the onset of type 2 diabetes (T2D). The underlying mechanisms of how distinct islet cells affect each other’s functions are not known. Secreted proteins are critical intra- and inter- cell type metabolic regulators that have improved our understanding of mechanisms underlying obesity-induced T2D. Thus, the premise of this project is that secreted proteins-mediated crosstalk in islets is essential for proper functioning and adaptation of a, b, d-cells in lean, obese, and T2D states. Secreted proteins comprise ~11% of the total human transcriptome, and our preliminary data have identified ~850 differentially expressed transcripts that encode for secreted proteins in mouse islets with obesity. Yet, the function for only a handful of them has been well-characterized. Our long-term goal is to identify secreted proteins that improve islet function for the treatment of human T2D. A major roadblock towards achieving this goal is the technical limitations in identifying and costly yet time-consuming functional characterization of secreted proteins in islets using conventional biochemical approaches. In a test analysis of one data set at high stringency, 44 islet-derived secreted protein regulators were identified to affect mouse islet function in obesity. Interestingly, the functional characterization of the top candidate secreted protein led to the discovery of a novel pathway inhibiting insulin secretion from b-cells. Excitingly, validation of the use of our quantitative bioinformatics framework is a leap towards effective data mining in expediting the identification of novel secreted protein regulators of islet function associated with the disease state (s). The objective here is to identify secreted protein regulators that affect islet function in human T2D using network analysis on combined publicly available whole islet transcriptomics datasets. We propose the following aims to achieve the objective: 1) Identify candidate secreted protein regulators; 2) superclusters for functional prediction of candidate secreted proteins in islets associated with human obesity and T2D; and 3) biological validation of the candidate secreted proteins to affect islet function. The successful completion will identify novel regulators of islet function in human obesity and T2D, improving knowledge of mechanisms underlying human T2D risks, and possibly identifying therapeutic targets to improve islet function in T2D. Additionally, insights obtained by integrating multiple data sets accounting for variations in sample preparation and sequencing (platform bias), sequencing depths, and networks/correlation architecture (due to sample handling) will form the basis for elucidating the secreted protein network across distinct islet cell-types.