Developing and applying large-scale simulation approach to understand the mechanisms of kinesins' motilities along microtubules

摘要： 抗有丝分裂药物是癌症治疗中非常理想的化疗药物。传统抗有丝分裂 药物通过解聚或稳定微管来破坏微管动力学以杀死过度活跃的癌症 细胞尽管这些抗有丝分裂药物取得了巨大的成功，但它们面临两个重大问题：1） 严重的副作用;和2）对某些类型的癌症有很强的耐药性。为了解决这两个问题， 最近发现驱动蛋白是理想的替代药物靶标。虽然微管为有丝分裂提供了支架， 正是驱动蛋白与微管的相互作用负责有丝分裂分离。此外，不同 不同类型的驱动蛋白负责不同的微管功能，允许药物的可能设计 特异于有丝分裂，副作用更少。最近的实验工作已经完成，以揭示 驱动蛋白运动机制的研究。然而，许多驱动蛋白在原子水平上的机制仍然是未知的。 由于分辨率在时间和长度上的限制，目前的实验方法中缺少这种方法。 计算工作可以弥补原子细节与当前实验分辨率之间的差距 接近。然而，由于驱动蛋白的大尺寸和高密度，驱动蛋白的模拟极具挑战性。 微管系统基于近年来算法的快速改进，PI将发展成大规模的 模拟软件包，能够准确模拟大型驱动蛋白-微管复合物。这个包裹 将用于揭示驱动蛋白结合和运动特性的重要机制， 以驱动蛋白为靶点的抗有丝分裂药物设计。PI在软件开发方面具有丰富的经验， 蛋白质-蛋白质相互作用，静电计算，结合能计算，pKa计算， 和大规模的模拟。此外，PI还在驱动蛋白等方面积累了丰富的研究经验， 分子发动机。PI最近的计算工作揭示了驱动蛋白运动 结构域和微管是驱动蛋白运动特性的重要因素。和疾病突变 驱动蛋白与微管之间的静电力变化趋势较强。因此，我们认为， 利用精确和全面的计算方法研究驱动蛋白是一个非常有前途的方向 了解驱动蛋白的作用机制，发现新的驱动蛋白靶向抗有丝分裂药物。除了有丝分裂 驱动蛋白、其他驱动蛋白上的突变和缺陷也是神经障碍和严重的神经系统疾病的原因。 例如阿尔茨海默病、亨廷顿病、帕金森病和许多其他疾病。大规模模拟 这项工作中开发的软件包也将有助于发现这些疾病的新疗法。而且这 该软件包解决了传统仿真软件包的规模限制问题，具有广泛的应用前景 研究复杂的生物系统，如动力蛋白-微管复合物，病毒衣壳组装，G蛋白 膜上的系统，以及许多其他系统。

Abstract: Anti-mitotic drugs are highly desirable chemotherapy drugs for cancer treatment. Traditional anti-mitotic drugs destroy microtubule dynamics by depolymerizing or stabilizing microtubules to kill the overactive cancer cells. Even though these anti-mitotic drugs have achieved great success, they face two significant issues: 1) Serious side effects; and 2) Strong drug resistance for some types of cancers. To overcome these two issues, kinesins are recently found to be ideal alternative drug targets. While microtubules provide the scaffold for mitosis, it is the interaction of kinesins with microtubule that is responsible for mitotic separation. Moreover, different types of kinesins are responsible for different microtubule functions, allowing for the possible design of drugs specific to mitosis with fewer side effects. Recent experimental works have been performed to reveal mechanisms of kinesin motility successfully. However, many kinesins’ mechanisms at the atomic level are still missing in current experimental approaches due to the limitations of resolutions, both in time and in length. Computational works can bridge the gap between atomic details and the resolutions of current experimental approaches. However, simulations for kinesins are extremely challenging due to the large size of kinesin and microtubule system. Based on fast improvements of algorithms in recent years, the PI will develop a large-scale simulation package that is capable of simulating large kinesin-microtubule complexes accurately. This package will be applied to reveal the important mechanisms for kinesins’ binding and motility features, which will shed light on kinesin targeting anti-mitotic drug design. The PI has extensive experience of software developments in the areas of protein-protein interactions, electrostatic calculations, binding energy calculations, pKa calculations, and large-scale simulations. Besides, the PI also has gained rich experience of studying kinesins and other molecular motors. The PI’s recent computational woks have revealed that the interaction between kinesin motor domains and the microtubule is an important factor for kinesin’s motility features. And disease mutations on kinesins show strong tendency of electrostatic force changes between kinesins and microtubules. Therefore, investigating kinesins using accurate and comprehensive computational approaches is a very promising direction to understand the mechanisms of kinesins and discover new kinesin targeting anti-mitotic drugs. Besides mitotic kinesins, mutations and defects on other kinesins are also responsible for neurological disorders and serious diseases such as Alzheimer, Huntington, Parkinson disease and many others. The large-scale simulation package developed in this work will also help to discover novel treatments of those diseases. Furthermore, this package will solve the scale limitation issue of traditional simulation packages and therefore can be widely used to study complex biological systems, such as the dynein-microtubule complex, viral capsid assembly, G-proteins systems on the membrane, and many others.