Constructing gene-regulatory networks to reveal the metabolic basis of lifespan i

描述(由申请人提供)：酿酒酵母是有丝分裂后细胞衰老的有用模型。这种存活的不分裂、静止但代谢活跃的细胞称为 时序寿命(CLS)。近年来，几个基因组规模的CLS筛查报告了丰富的CLS决定基因；因此，有数百个CLS的遗传决定因素被认为。现在需要的是一种方法来了解这么多基因是如何共同发挥作用来调节衰老过程的。然而，由于对主要参与者缺乏共识，实现这种洞察力存在困难，因为不同的屏幕没有达到相同的结果。目前还不清楚为什么一些基因不能提供可重复的表型，而且通常情况下，即使在同一个实验室里也可能是这样。我们的研究小组集合了互补的技能和技术来解决CLS的困境，我们将其称为构建“数据驱动网络”。我们开发了一种用于数据驱动分析的使能技术：称为定量高通量细胞阵列表型分析(Q-HTCP)，它使直接测量克隆培养物CLS表型的能力比现有技术提高了数百倍。我们希望应用Q-HTCP来系统和定量地评估在单倍体酵母生长中所有6000个非必需基因的突变敲除和必需基因的敲除中的CLS，以及在CLS的近交系模型中，CLS由更多的遗传异质性菌株组成。我们将在已知会影响CLS的各种环境条件下执行CLS。这些“扰动”将包括可变的葡萄糖浓度，这是众所周知的CLS的影响因素，但也包括硫代谢途径(SMP)的输入。此外，我们发现曝气可以影响CLS，并怀疑它与其他环境因素和遗传因素相互作用。SMP是生产必需氨基酸蛋氨酸所必需的，对蛋氨酸的饮食限制已被证明即使在没有限制卡路里的情况下也能延长寿命。在第一次使用Q-HTCP进行全基因组CLS筛查时，我们在4750株基因缺失菌株中鉴定出363株重复性(2株培养物中的2株)比参照对照株存活时间更长。长期存活者与野生型分离的顺利趋势，以及复制培养物之间CLS存活曲线的高度相关性，为我们的结果提供了信心。在我们最自信的363次点击中，只有69次与其他至少三个基因组屏幕中的前300个重叠；14个与三个中的两个重叠，没有一个与所有三个重叠。我们的结果证实了对CLS决定基因缺乏共识，因为我们的筛查与其他三个筛查结果基本一致。CLS数据质量有力地验证了Q-HTCP在酵母老化研究中的有效性，因此我们组建了一个跨学科的团队，包括Q-HTCP、SMP、CLS、转录组、代谢组和从大规模基因组数据集成构建生物网络的专业知识。通过跨学科的合作，我们的目标是为CLS提供一个系统级的框架，以帮助构建能够揭示细胞衰老机制的基因调控网络。

DESCRIPTION (provided by applicant): S. cerevisiae has been a useful model of aging for post-mitotic cells. This survival non-dividing, quiescent but metabolically active cells, is termed chronological lifespan (CLS). Several genome-scale CLS screens have reported abundant CLS-determining genes in recent years; thus, there are thought to be hundreds of genetic determinants of CLS. What is needed now is a way to understand how so many genes function together to regulate the aging process. However, there are difficulties in achieving such insight due to lack of consensus about the primary players, since different screens have not reached the same results. It remains obscure why some genes do not give reproducible phenotypes, and often this can be the case even within the same laboratory. Our research group assembles complementary skills and technologies to address this CLS quandary by an approach we call constructing "data-driven networks". We have developed an enabling technology for data driven analysis: it is called quantitative high throughput cell array phenotyping (Q-HTCP), and it increases the capacity to directly measure CLS phenotypes of clonal cultures by several hundred fold over existing technologies. We wish to apply Q-HTCP to systematically and quantitatively assess CLS in all 6000 mutant knockouts of non-essential genes and knockdowns of essential genes in haploid yeast growing, as well as in an outbred model for CLS, consisting of much more genetically heterogeneous strains. We will perform CLS under variable environmental conditions know to influence CLS. These 'perturbations' will include variable glucose concentration, a well-known effecter of CLS, but also inputs to the sulfur metabolic pathways (SMP). Additionally, we have found aeration to influence CLS and suspect that it interacts with other environmental, as well as genetic factors. SMP are required for production of the essential amino acid methionine, for which dietary restriction has been shown to extend life span even in the absence of caloric restriction. In the first use of Q-HTCP for a genome-wide CLS screen, we identified 363 out of 4750 gene deletion strains to reproducibly (2 of 2 cultures) have longer survival than the reference control strain. Lending confidence to our result were the smooth trends of separation of long survivors from the wild type, and the high correlation in CLS survival curves between replicate cultures. Among our most confident 363 hits, only 69 overlapped with the top 300 of at least one of three other genome screens; 14 overlapped with two of the three, and none overlapped with all three. Our result confirms a lack of consensus about CLS determining genes, since our screen was in accord about equally with all of the other three screens. The CLS data quality strongly validated the utility of Q-HTCP for yeast aging research, and so we have assembled an interdisciplinary team including expertise in Q-HTCP, SMP, CLS, transcriptomics, metabolimics and construction of biological networks from large-scale omic data integration. Through interdisciplinary collaboration, we aim to deliver a systems level framework for CLS to help construct gene regulatory networks that can reveal mechanisms of cellular aging.