Rules of gene expression modeled on human dendritic cell response to pathogens

描述(申请人提供)：当细胞对环境刺激作出反应时发生的发育变化在很大程度上由涉及数千个基因的基因表达程序控制。转录因子(TF)、染色质修饰酶和顺式作用的DNA元件构成了构成这些程序的网络。将这些变量联系在一起的密码还有待破译，这种密码可以根据特定成分的存在来预测给定基因的表达。这里将基于对人类树突状细胞(DC)成熟时对病原体的反应进行全基因组分析，构建这样一个密码的模型。DC是抗原提呈细胞，启动并决定宿主免疫反应的质量和大小。干细胞生物学的最新技术进展、人类原代细胞的反向遗传工具以及对转录产物、局部染色质特征和远程染色质相互作用的全基因组评估，将在这里被用来构建一个转录调控网络的模型，该网络是人类树突状细胞病原体检测和成熟的基础。在目标1中，DC将在内毒素刺激后按一定时间段取样。刺激反应基因将通过RNA-Seq进行鉴定。染色质特征将使用芯片序列、高分辨率全球DNA足印和Hi-C绘制。所有响应基因及其相关调控区域的系统图谱将与已知的DNA结合基序相结合，以生成基因调控网络的初始模型。在Aim 2中，将从Aim 1中的模型中确定在内毒素刺激的前2小时内驱动转录的60个转录因子。这些因子将在hDC中被敲除，并将使用RNA-Seq来评估其对内毒素反应的影响。每个TF的这些功能数据将通过使用CHIP-SEQ测量蛋白质-DNA相互作用来补充；后者将利用CRISPR技术将公共表位标签融合到这些TF的内源编码序列中。通过精确定位关键转录因子、它们的结合部位和转录反应基因，这种分析将被用于 完善基因调控网络模型。在目标3中，将使用在以前的目标中收集的特征来确定对控制HDC对脂多糖的转录反应至关重要的顺式作用调节区的子集。选定的基因座将使用CRISPR技术进行干扰，并将检测对目标基因表达的影响。将使用已公布的数据以及将使用Pac-Bio进行基因分型的300个人类样本来评估相关基因座上的等位基因变异对基因表达的影响。同时，将通过5C评估五个个体上50个关键调控基因座的高阶染色质构象的差异。这些数据将被用来进一步完善我们的监管模式。这一分析的结果将是一个突出关键转录因子和顺式作用成分的模型，这些转录因子和顺式作用成分推动了内毒素刺激的人类DC中的基因表达。我们的发现有望为鉴定原代哺乳动物细胞中基因表达的遗传决定因素提供一种更普遍的方法。

DESCRIPTION (provided by applicant): The developmental shifts that occur when cells respond to environmental stimuli are controlled in large part by gene expression programs involving thousands of genes. Transcription factors (TFs), chromatin modifying enzymes, and cis-acting DNA elements contribute to the networks that underlie such programs. The code that links these variables in such a way that the expression of a given gene can be predicted based on the presence of specific components has yet to be deciphered. A model for such a code will be constructed here based on genome-wide analysis of human dendritic cells (DCs) as they mature in response to pathogens. DCs are antigen-presenting cells that initiate and determine the quality and magnitude of the host immune response. Recent technical advances in stem cell biology, reverse-genetic tools for primary human cells, and genome-wide assessment of transcripts, local chromatin features and long-range chromatin interactions, will be exploited here to construct a model for the transcriptional regulatory network that underlies pathogen detection and maturation in human DCs. In Aim 1, DCs will be sampled in a time-course following stimulation with LPS. Stimulation-responsive genes will be identified by RNA-Seq. Chromatin features will be mapped using ChIP-Seq, high-resolution global DNA foot printing, and Hi-C. This systematic map of all responsive genes and the regulatory regions associated with them will be combined with known DNA-binding motifs to generate an initial model for the gene regulatory network. In Aim 2 sixty transcription factors that drive transcription within the first 2 hrs of LPS stimulation will be identified from the model in Aim 1. These factors will be knocked-down in hDCs and the effect on the LPS response will be assessed using RNA-Seq. This functional data for each TF will be complemented by measurement of protein-DNA interactions using ChIP- Seq; the later will exploit CRISPR technology to fuse a common epitope tag to endogenous coding sequences for these TFs. By pinpointing key transcription factors, their binding sites, and transcriptionally-responsive genes, this analysis will be used to refine the model for the gene regulatory network. In Aim 3, a subset of cis- acting regulatory regions important for controlling the hDC transcriptional response to LPS will be identified using features collected in the previous aims. Selected loci will be perturbed using CRISPR technology and effects on target gene expression will be examined. The effect on gene expression of allelic variants in relevant loci will be assessed, using published data as well as 300 human samples that will be genotyped using Pac- Bio. In conjunction, the variance in the higher-order chromatin conformation for 50 key regulatory loci across five individuals will be assessed by 5C. These data will be used to further refine our regulatory model. The result of this analysis will be a model that highlights the key transcription factors and cis-acting components that drive gene expression in LPS-stimulated human DCs. Our findings are expected to provide a more general method for identifying the genetic determinants of gene expression in primary mammalian cells.