Computational Investigations of the Mechanisms Behind Microtubule Catastrophe

项目摘要 微管（MT）构成真核细胞骨架的最大组成部分，并促进过多的 多种功能，包括细胞内运输、细胞运动和细胞分裂。在有丝分裂期间，MT 聚集形成有丝分裂纺锤体，使其成为许多成功的化疗药物的有效药物靶点。 药物，包括紫杉醇和长春碱，被称为纺锤体毒药。MT靶向药物通过干扰 具有动态不稳定性（DI）：MT从聚合快速切换到解聚的能力（参见 灾难），反之亦然。Paclitaxel通过降低灾难率而长春碱则鼓励 灾难和抑制聚合。对MT灾变的充分理解将极大地帮助MT灾变的设计， 纺锤体毒物具有更少脱靶效应，以及极大地推进对DI的一般理解。 每个MT由αβ-微管蛋白异源二聚体组成，在排列的原丝（PF）中首尾堆叠， 横向地形成中空管。α-微管蛋白和β-微管蛋白都与鸟苷三磷酸（GTP）结合， 假设在β-微管蛋白结合位点处与GDP（鸟苷二磷酸）的结合诱导MT晶格上的应力。 这种压力逐渐建立，直到MT末端的亚基经历GTP水解，此时PF开始 分崩离析，灾难就发生了GTP水解和聚合之间的滞后产生了一个构建体 称为GTP帽：MT末端的一组亚单位，它们尚未水解GTP，释放出 产物无机磷酸盐（Pi），或经历结构转变。最近的研究引起了人们的怀疑， 关于这种转变的性质和对潜在机制的原子论理解的领域将导致一个 对灾难的充分理解。我建议通过计算解决灾难的三个关键方面： GTP水解的机制，Pi的释放，以及PF之间的结构耦合导致灾难。 首先，我将使用增强的采样方法来揭示GTP水解的酶机制， 重点放在属于α-微管蛋白的潜在催化残基上，α-微管蛋白位于β-微管蛋白的顶部， 聚合以形成活性位点。随后，我将开发新的计算技术来确定 水解后Pi释放的途径，并检查释放后结构变化的可能性。最后我 将开发一个粗粒度（CG）的完整MT模型，使用从以前的研究确定的速率，能够 经历灾难，以研究相邻亚基中的水解和Pi释放如何影响 这些反应发生在一个特定的亚单位。这将给一个前所未有的详细的看法的损失， GTP上限和导致灾难的步骤。此外，我将与两位领先的实验学家合作， 在MT社区开发突变体，专门测试我的假设，并获得晶格参数 MTs掺杂了纺锤体毒素。这将使我能够将药物的影响整合到CG模型中， 检查它们的影响如何沿着MT沿着。这些结果和发展的模型将大大推进 对DI的理解，并有望在未来导致更温和的MT靶向治疗的发展。

PROJECT SUMMARY Microtubules (MTs) constitute the largest components of the eukaryotic cytoskeleton and facilitate a plethora of diverse functions including intracellular transport, cellular motility, and, cell division. During mitosis, MTs aggregate to form the mitotic spindle, making them a potent drug target for many successful chemotherapeutic agents, including paclitaxel and vinblastine, known as spindle poisons. MT-targeting drugs operate by interfering with dynamic instability (DI): the ability of MTs to rapidly switch from polymerizing to depolymerizing (referred to as catastrophe) and vice-versa. Paclitaxel operates by decreasing catastrophe rate while vinblastine encourages catastrophe and inhibits polymerization. A full understanding of MT catastrophe will greatly aid in the design of spindle poisons with fewer off-target effects, as well as greatly advance general understanding of DI. Each MT is composed of αβ-tubulin heterodimers, stacked head-to-tail in protofilaments (PFs) which are aligned laterally to form a hollow tube. Both α- and β-tubulin bind guanosine triphosphate (GTP) and hydrolysis of GTP to GDP (guanosine diphosphate) at the β-tubulin binding site is hypothesized to induce stress on the MT lattice. This stress gradually builds until the subunits at the MT end undergo GTP hydrolysis, at which point PFs begin to peel apart and catastrophe has occurred. Lag between GTP hydrolysis and polymerization creates a construct referred to as the GTP cap: a group of subunits at the MT end that have yet to hydrolyze GTP, release the product inorganic phosphate (Pi), or undergo a structural transition. Recent studies have caused doubt in the field on the nature of this transition and an atomistic understanding of the underlying mechanisms will lead to a full understanding of catastrophe. I propose to computationally resolve three key aspects of catastrophe: the mechanism of GTP hydrolysis, the release of Pi, and the structural coupling between PFs leading to catastrophe. First, I will use enhanced sampling methodology to uncover the enzymatic mechanism of GTP hydrolysis, with emphasis placed on potential catalytic residues belonging to α-tubulin, which sits atop β-tubulin upon polymerization to form the active site. Subsequently, I will develop novel computational techniques to determine the pathway of Pi release post-hydrolysis and examine the potential for structural change upon release. Lastly, I will develop a coarse-grained (CG) model of a full MT, using rates determined from the previous studies, able to undergo catastrophe to examine how hydrolysis and Pi release in neighboring subunits affects the potential for these reactions to occur in a particular subunit. This will give an unprecedentedly detailed view of the loss of the GTP cap and the steps leading to catastrophe. Additionally, I will collaborate with two leading experimentalists in the MT community to develop mutants that specifically test my hypotheses and to obtain lattice parameters of MTs doped with spindle poisons. This will allow me to integrate the effects of drugs into the CG model and examine how their effects propagate along an MT. These results and the developed models will greatly advance the understanding of DI and hopefully lead to the development of gentler MT-targeting therapies in the future.