Decoding mechanisms of phenotypic memory in single cells

项目摘要 耐药性和转移都是癌症中的致命过程，人们对此知之甚少。在 在许多情况下，耐药或转移来自个体肿瘤内的一小部分细胞， 行为与其他人不同。因为这些细胞只是患者肿瘤细胞的一小部分， 不能使用传统方法对它们进行测序或分析。因此，单细胞分析提供了一个窗口， 这些有害过程背后的细胞之间的变异性;然而，诸如单细胞 测序是在固定或裂解细胞后进行的，这使得研究人员无法 进行进一步的下游分析，例如测试细胞的抗性或侵袭性表型。这些 这些努力在单细胞水平上对分子变异性进行了编目，但未能确定这些变量是如何产生的。 特征涉及存在于单细胞中的表型。 我的研究通过开发一种无偏见的高通量测序方法来解决这一障碍 用于鉴定在单细胞中足够根深蒂固以产生表型的变异性。这种方法 根据Luria和Delbrück 1943年的波动分析。它结合了他们聪明的实验设计， 高通量测序分析的现代转折。当与RNA测序相结合时，我们的方法 （MemorySeq）允许我们量化基因表达动态，以找到单细胞基因表达 这种状态是缓慢波动的，可以通过多次细胞分裂遗传。我们假设这些 波动的基因表达状态允许单细胞发生显著和根深蒂固的变化， 这是在癌症中产生抗性和转移的有害表型所必需的。我们的目标是利用我们的 一种新的MemorySeq方法：1）测试基因表达长期波动的假设， 癌症中的重要表型，特别是耐药性和侵袭，以及2）鉴定转录 因子，激酶和表观遗传调节蛋白负责产生和维持这些长寿命 基因表达的波动。这些目标将通过一个高度创新和 互补方法，结合高通量测序，CRISPR/Cas9基因筛选， 单细胞成像这项研究将确定罕见的单细胞基因表达特征， 耐药和侵袭性人群中的多种癌症类型，并将列举转录因子， 激酶和控制这些表达状态的表观遗传调节蛋白。 这项工作的结果将是重大的癌症研究界，因为他们将产生新的 治疗靶点以特异性抑制或破坏这些不期望的稀有细胞群。而且这 概念框架具有普遍性，科学研究界可以广泛使用。在 将来，这些波动分析方法可用于揭示缓慢波动基因的贡献 在其他生物学过程中的表达状态，如发育、伤口愈合和细胞命运决定。

PROJECT SUMMARY Drug resistance and metastasis are both deadly processes in cancer that remain poorly understood. In many instances, resistance or metastasis arise from a small subset of the cells within an individual's tumor that behave differently from the rest. Because these cells are only a small fraction of the cells in a patient's tumor, they cannot be sequenced or profiled using traditional methods. Thus, single-cell analysis provides a window into the variability between cells that underlie these harmful processes; however, methods such as single-cell sequencing are performed after fixing or lysing of the cells, which prevents researchers from being able to perform further downstream analysis such as testing the cells for resistant or invasive phenotypes. These efforts catalogue the molecular variability at the single-cell level, but fail to determine how these variable features relate to the phenotypes present in single cells. My research addresses this hurdle via development of a unbiased, high-throughput sequencing method for identifying variability that is sufficiently ingrained in single cells to generate phenotypes. This approach is based upon Luria and Delbrück's 1943 fluctuation analysis. It combines their clever experimental design with the modern twist of high-throughput sequencing assays. When combined with RNA sequencing, our method (MemorySeq) allows us to quantify gene expression dynamics in order to find single-cell gene expression states that are slowly fluctuating and heritable through multiple cell divisions. We hypothesize that these slowly fluctuating gene expression states allow for significant and ingrained changes in single-cells, which are necessary to generate the detrimental phenotypes of resistance and metastasis in cancer. We aim to use our new MemorySeq method to 1) test the hypothesis that long-lived fluctuations in gene expression underly important phenotypes in cancer, specifically drug resistance and invasion, and to 2) identify transcription factors, kinases, and epigenetic regulator proteins responsible for generating and maintaining these long-lived fluctuations in gene expression. These aims will be accomplished using a highly innovative and complementary approach that combines high-throughput sequencing, CRISPR/Cas9 genetic screening, and single-cell imaging. This line of research will determine the single-cell gene expression signatures of rare resistant and invasive populations in multiple cancer types, and will enumerate the transcription factors, kinases, and epigenetic regulator proteins that govern these expression states. The results of this work will be significant to the cancer research community as they will yield new therapeutic targets to specifically inhibit or destroy these undesirable rare cell populations. Furthermore, this conceptual framework is generalizable and broadly accessible to the scientific research community. In the future, these fluctuation analysis methods can be applied to unravel the contribution of slowly fluctuating gene expression states in other biological processes such as development, wound healing, and cell fate decisions.