Combining experiments of man and nature to target human insulin resistance

摘要 胰岛素抵抗是慢性疾病的主要原因，包括2型糖尿病(T2D)、心脏病发作、中风 和癌症。只有一类药物，噻唑烷二酮(TZD)，专门针对胰岛素抵抗 主要是通过激活脂肪细胞中的转录因子PPARG。虽然TZD已经在临床上得到了证明 在预防T2D、心脏病发作和中风方面有效，但严重的副作用限制了它们的临床应用。新的 对抗胰岛素抵抗的治疗目标是必要的。从理论上讲，基因表达的变化导致 通过TZD治疗可以发现新的胰岛素增敏效应，但TZD改变了 数以百计的基因。必须通过对因果效应的功能调查来筛选这些因素，而不仅仅是 相关或有毒的旁观者。标准的基于实验室的功能检查需要侵入性检查 难以衡量的模型系统中的生理测量。即使当潜在的胰岛素 致敏效应基因在实验室中得到验证，证明其与人类胰岛素敏感性有关 需要药物开发和人体试验，这是另一个可扩展性差的过程，通常会导致失败 因为缺乏疗效。然而，最近积累的基因组序列很大，具有临床特征 人类已经揭示，大自然已经进行了无数次人类试验，其形式是数百万的自然 发生的、改变蛋白质的遗传变异几乎分散了基因组中的每一个基因。解决问题的关键 释放这两个机会：1)从TZD治疗中识别新的候选基因；2)利用 《自然》用于评估治疗潜力的临床试验是高通量的功能分析。在这 应用，我们建议利用新开发的大规模平行脂肪细胞分化/脂质 综合基因组方法中的蓄积试验： 目标1：快速对TZD改变的基因进行排序，以了解其可能是胰岛素增敏效应和 目标2：使用以下数据确定的人类治疗潜力的凭证和新的候选对象 10万个被测序的个体。 这项工作将对TZDS的治疗效果进行系统的剖析，以确定新的胰岛素 并直接评估人类对四种疾病的调节功能的治疗潜力 这些基因。

Abstract Insulin resistance is a major cause of chronic diseases including type 2 diabetes (T2D), heart attacks, strokes and cancer. Only one class of medications, thiazolidinediones (TZDs), specifically targets insulin resistance primarily by activating the transcription factor PPARG in adipocytes. While TZDs have proven clinically effective in preventing T2D, heart attacks and strokes, serious side effects have limited their clinical use. New therapeutic targets to combat insulin resistance are needed. In theory, the gene expression changes caused by TZD-treatment could be mined for novel insulin-sensitizing effectors, but TZDs alter the expression of hundreds of genes. These must be sifted by functional investigation for causal effectors versus merely correlated or toxic bystanders. Standard laboratory-based functional investigation requires invasive physiological measurements in model systems that are difficult to scale. Even when a potential insulin sensitizing effector gene is validated in the lab, credentialing its relevance to human insulin sensitivity necessitates drug development and human trials, another poorly scalable process that usually results in failure for lack of efficacy. However, the recent accumulation of genome sequences in large, clinically characterized populations has revealed that nature has performed countless human trials in the form of millions of naturally occurring, protein-altering genetic variants scattered throughput almost every gene in the genome. The key to unlocking both these opportunities: 1) identifying novel candidate genes from TZD treatment and 2) leveraging nature’s clinical trials for assessing therapeutic potential, are high-throughput functional assays. In this application, we propose to utilize a newly developed massively parallel adipocyte differentiation/ lipid accumulation assay in an integrative genomic approach to: Aim 1: Rapidly sort TZD-altered genes for likelihood of being insulin sensitizing effectors and Aim 2: Credential already identified and novel candidates for therapeutic potential in humans using data from 100,000 sequenced individuals. This work will produce a systematic dissection of the therapeutic effect of TZDs to identify novel insulin sensitivity genes and directly assess the therapeutic potential of modulating function in humans for four of these genes.