Systematic dissection of an aberrant splicing program driving cancer metastasis

 描述（由申请人提供）：细胞依赖于蛋白质在空间和时间上的精确表达来驱动关键的生物过程。因此，解码潜在的调控基因组是开发现实和准确的细胞行为模型的关键一步。虽然转录网络已被广泛研究，转录后调控程序仍然在很大程度上未被表征。RNA介导的相互作用受到局部二级结构的高度影响，并且转录后调控程序的完整表征需要捕获由二级结构及其基础序列提供的信息。这个K99/R 00建议的主要目标是开发一个原则性的综合方法，在正常和病理状态下的转录后调控的系统解剖。我以前描述了一种计算方法（TEISER），它已成功地揭示了调节程序，调节转录稳定性的哺乳动物细胞。在这些发现的指导下，这项拟议的研究旨在为研究各种转录后调控机制提供一个综合框架，并进一步阐明它们在疾病进展中的作用。理解和定量建模复杂细胞行为背后的调控网络对于成功开发有前景的治疗干预至关重要。 最近，我发现了一个结构RNA元素显着丰富的转录不稳定的转移性乳腺癌细胞。我确定了双链RNA结合蛋白质TARBP 2作为结合该元件并调节其靶点稳定性的因子。进一步表征thi调节子导致发现新的TARBP 2介导的促转移调节程序。我的第一个目标，作为这个建议（K99）的指导阶段的一部分，是从功能上表征TARBP 2影响转录稳定性的分子机制。基于广泛的TARBP 2结合其目标的内含子区域，我假设TARBP 2结合导致内含子保留在结合的转录本中，这反过来又增加RNA降解。虽然我的初步结果和最近发表的研究结果支持这一假设，但需要有针对性的实验和分析来建立TARBP 2调节其靶点的调控机制。虽然本质上是多学科的，但我在整个职业生涯中的研究都植根于计算生物学。我相信，我将从这个项目中学到的科学和实践经验，以及我的导师（Tavazoie博士）和共同导师（Bieniasz博士）以及他们各自在洛克菲勒大学的实验室接受的培训，将进一步扩展我作为实验生物学家的专业知识，并为我未来的研究奠定坚实的基础。 我的建议的一部分，跨越辅导和独立的阶段，重点是实验和计算组件，是必不可少的，使TEISER成为一个真正的综合框架，此后被称为iTEISER。计算能力和实验方法的最新进展，使一个更系统的方法发现的结构元素。在实验方面，我将采用最近开发的基于差异核酸酶消化（或基于DMS的修饰）和高通量测序的全转录组二级结构作图方法，以获得体外和体内细胞RNA二级结构的信息。我还将包括一个在silico折叠步骤，以分配一个概率的局部形成一个gven结构。然后可以采用贝叶斯模型来评估基于体内、体外和计算机数据形成结构元件的可能性。将依赖于简单RNA折叠算法的iTEISER的初步版本与体外差异核酸酶消化模式相结合以评估给定结构元件的存在，我重新分析了乳腺癌细胞系中的差异转录稳定性测量。值得注意的是，我发现了一个富含A的茎环元件在转移性细胞中不稳定的转录本中过度表达，这在以前低于TEISER的敏感性。我的初步结果，在培养滴定实验的基础上，强烈支持这种新的元素的功能。此外，作为R 00阶段的一部分，我将专注于扩展上述方法，以调节乳腺癌模型中的选择性剪接。剪接的调控是一个复杂的过程，涉及许多RNA-蛋白质相互作用，可以通过本提案中概述的方法有效地剖析。我将利用来自转移性和非转移性细胞的深度RNA测序数据来识别在多个细胞系中差异剪接的外显子。然后，我将扫描这些外显子和它们的侧翼内含子序列，寻找解释观察到的失调的常见结构/线性剪接元件。我的初步研究结果表明，不仅可以发现这些剪接元件，而且其中一些与多种癌症类型有关。iTEISER将能够以系统和公正的方式识别这些元素。确定选择性剪接的调控网络并研究其在疾病进展中的作用将是特别有意义的。这里概述的框架有可能为研究转录后调控的不同方面（例如，除了转录稳定性和剪接之外，RNA定位和翻译）提供实质性的动力。这些，反过来，将作为一个01提案的基础上准备完成本研究计划的主要阶段。

 DESCRIPTION (provided by applicant): Cells rely on spatially and temporally precise expression of proteins to drive key biological processes. Thus, decoding the underlying regulatory genome is a crucial step towards developing realistic and accurate models of cellular behavior. While transcriptional networks have been widely studied, post-transcriptional regulatory programs remain largely uncharacterized. RNA-mediated interactions are highly influenced by local secondary structures and a full characterization of post-transcriptional regulatory programs requires capturing information provided by both the secondary structure and its underlying sequence. The main objective of this K99/R00 proposal is to develop a principled integrated approach for a systematic dissection of post-transcriptional regulation in both normal and pathologic states. I have previously described a computational approach (TEISER), which has been successfully employed to reveal regulatory programs that modulate transcript stability in mammalian cells. Guided by these findings, this proposed research aims to contribute an integrated framework for studying various post-transcriptional regulatory mechanisms and to further elucidate their role in disease progression. Understanding and quantitatively modeling the regulatory networks underlying complex cellular behaviors is crucial for the successful development of promising therapeutic interventions. Recently, I discovered a structural RNA element significantly enriched among the transcripts destabilized in metastatic breast cancer cells. I identified the double-stranded RNA-binding protein TARBP2 as the factor that binds this element and modulates the stability of its targets. Further characterization of thi regulon led to the discovery of a novel TARBP2-mediated pro-metastatic regulatory program. My first goal, as part of the mentored phase of this proposal (K99), is to functionally characteriz the molecular mechanisms through which TARBP2 affects transcript stability. Based on extensive TARBP2 binding to intronic regions of its targets, I have hypothesized that TARBP2 binding results in intron-retention in bound transcripts, which in turn increases RNA degradation. While my preliminary results and recently published findings support this hypothesis, focused experiments and analyses are required to establish the regulatory mechanisms through which TARBP2 modulates its targets. While multidisciplinary in nature, my research throughout my career has been rooted in computational biology. I believe that the scientific and practical lessons that I will learn from this project, in conjunction with training received from my mentor (Dr. Tavazoie) and co-mentor (Dr. Bieniasz) and their respective labs here at Rockefeller University, will further expand my expertise as an experimental biologist and form a strong foundation for my future research. A part of my proposal, spanning both the mentored and independent phases, focuses on the experimental and computational components that are essential for enabling TEISER to become a truly integrated framework, henceforth referred to as iTEISER. Recent advances in computational power and experimental methodologies enable a more systematic approach for discovery of structural elements. On the experimental side, I will adopt the recently developed transcriptome-wide secondary structure mapping approaches based on differential nuclease digestion (or DMS-based modifications) and high-throughput sequencing to gain information from the secondary structure of cellular RNA in vitro and in vivo. I will also include an in silico folding step to assign a probability to the local formation of a gven structure. A Bayesian model can then be employed to assess the likelihood of structural elements forming based on both in vivo, in vitro, and in silico data. Combining a preliminary version of iTEISER, which relies on a simple RNA folding algorithm, with in vitro differential nuclease digestion patterns to assess the presence of a given structural element, I re-analyzed the differential transcript stability measurements in breast cancer lines. Remarkably, I discovered an A-rich stem-loop element over-represented in the transcripts destabilized in metastatic cells, which previously fell below the sensitivity of TEISER. My preliminary results, based on in-culture titration experiments, strongly support the functionality of this novel element Additionally, as part of the R00 phase, I will focus on expanding the approach outlined above to regulation of alternative splicing in breast cancer models. Regulation of splicing is a complex process involving many RNA-protein interactions that can be effectively dissected through the approach outlined in this proposal. I will take advantage of deep RNA sequencing data from metastatic and non-metastatic cells to identify the exons that are differentially spliced across multiple cell lines. I will then scan these exons and their flanking intronic sequences for common structural/linear splicing elements that explain the observed deregulations. My preliminary findings indicate that not only can these splicing elements be discovered, but also that a number of them are implicated in multiple cancer types. iTEISER will enable the identification of such elements in a systematic and unbiased manner. It would be of particular interest to identify regulatory networks of alternative splicing and study their role in disease progression. The framework outlined here has the potential to provide substantial momentum towards studying different aspects of post-transcriptional regulation (e.g. RNA localization and translation in addition to transcript stability and splicing). These, in turn, will serve as the foundation of an 01 proposal to be prepared upon the completion of the main stages of this research plan.