Inferring Gene Regulatory Networks Governing Definitive Endoderm Differentiation from Single Cell RNA Velocity Measurements

项目摘要 世界卫生组织估计，超过6500万人患有中度至重度慢性 阻塞性肺疾病，一种以气流不畅和呼吸受限为特征的疾病1。的能力 再生受损的肺组织将大大改善这些人的生活质量， 降低全球肺部疾病的患病率和负担。解决这个问题的一个有希望的方法 是利用人类多能干细胞来生产肺和气道祖细胞。事实上， 已经开发了将干细胞转化为定形内胚层（一种肺前体细胞类型10-19）的方案。 这些方案使用小分子来调节肺发育的关键调节因子的表达 包括WNT、TGFβ、BMP和FGF;然而，这些方案受限于不能产生 定形内胚层细胞的同质群体11，15。这个问题需要一个更好的机制 了解单个细胞如何从多能细胞状态转变为定形内胚层。 具体地说，迫切需要了解给定细胞中的基因调控网络如何控制其表达。 形态发生、增殖和分化决定。因此，随着长期目标的增加， 肺前体细胞的同质性，这项研究的目的是确定如何 人胚胎干细胞的转录异质性影响其对决定性分化的承诺 内胚层我假设，细胞初始群体的异质性产生了交替的轨迹， 定形内胚层（或其他细胞类型），并且由于相互抑制，这些差异随时间增加 在特定的转录因子对之间（例如，OCT 4/SOX 17、NANOG/GATA 6）。为了验证这个假设，我 将首先使用单细胞RNA测序26来定义人类多能干细胞中的转录异质性， 细胞分化为定形内胚层。然后我将量化基因的时间依赖性变化， 使用RNA速度计算每个细胞的表达，这是一种使用剪接和未剪接转录本的计算方法， 计数以估计未来基因表达状态28-29。利用这些单细胞测量， 开发一个基因调控网络的机制模型，管理分化为DE，并验证 使用已知的基因-基因相互作用模型。模型模拟将：（1）确认主要的基因调控，驱动 （2）鉴定在分化之前和分化期间控制异质性的新基因网络;以及 (3)揭示了控制分化的基因调控网络和其他正在进行的细胞 如增殖和代谢过程。建议的实验和计算研究 提供了一个总体框架，以系统地鉴定控制分化的基因调控机制， 确定内胚层和援助的发展更有效和均匀 用于再生细胞疗法的分化/转分化方案。

Project Summary The World Health Organization estimates that over 65 million people suffer from moderate to severe chronic obstructive pulmonary disease, a condition characterized by poor airflow and restricted breathing 1. The ability to regenerate damaged lung tissue would dramatically improve the quality of life for these individuals while reducing the prevalence and burden of pulmonary diseases worldwide. A promising approach to this problem is to use human pluripotent stem cells to produce lung and airway progenitor cells. Indeed, specialized protocols have been developed to convert stem cells into definitive endoderm, a lung precursor cell type 10-19. These protocols use small molecules to modulate the expression of key regulators of lung development including WNT, TGFβ, BMP, and FGF; however, these protocols are limited by the inability to generate a homogeneous population of definitive endoderm cells 11,15. This problem necessitates a better mechanistic understanding of how individual cells transition from their pluripotent cell state into definitive endoderm. Specifically, there is a critical need to understand how the gene regulatory networks in a given cell control its morphogenesis, proliferation, and differentiation decisions. Therefore, with the long-term goal of increasing homogeneity in lung precursor cells, the research objective of this fellowship is to determine how transcriptional heterogeneity in human embryonic stem cells influences their commitment to definitive endoderm. I hypothesize that heterogeneity in the starting population of cells generates alternate trajectories to definitive endoderm (or other cell types) and that these differences increase over time due to mutual inhibition between specific pairs of transcription factors (e.g., OCT4/SOX17, NANOG/GATA6). To test this hypothesis, I will first use single-cell RNA sequencing26 to define the transcriptional heterogeneity in human pluripotent stem cells during differentiation to definitive endoderm. I will then quantify the time-dependent changes in gene expression for each cell using RNA velocity, a computational method that uses spliced and unspliced transcript counts to estimate future gene expression states 28-29. Using these single-cell measurements, I will then develop a mechanistic model of the gene regulatory networks governing differentiation to DE and validate the model using known gene-gene interactions. Model simulations will: (1) confirm major gene regulators that drive differentiation; (2) identify novel gene networks that control heterogeneity before and during differentiation; and (3) reveal crosstalk among gene regulatory networks governing differentiation and other ongoing cellular processes such as proliferation and metabolism. The proposed experimental and computational studies provide a general framework to systematically identify gene regulatory mechanisms controlling differentiation to definitive endoderm and aid in the development of more efficient and homogeneous differentiation/transdifferentiation protocols for regenerative cellular therapies.