细胞膜环境下自禀无序蛋白质构象变化的分子动力学研究

Intrinsically disordered proteins (IDP) are abundant in the human proteome. Due to their high structural flexibility, IDP play an important role in many biological processes, including signal transduction and post-transcriptional modification. As an important component of life, the cellular membrane regulates the spatial distribution and structural changes of proteins. Meanwhile, the amyloid fibrils formed by self-aggregation of IDP have been discovered in more than 30 human diseases. The presence of cellular membrane accelerates the fibril formation, while the membrane integrity is also disrupted during the fibrillation process. The mechanism of membrane disruption induced by IDP has been controversial so far, and there is still a lack of explanation for microscopic mechanisms that can unify the different experimental results. This project focuses on studying the structural formation and initial aggregation processes of IDP in cellular membrane environment, by utilizing all-atom molecular dynamics simulations. To reach the slow time-scales of these complex conformational changes in the simulations, this project intends to use non-linear dimensionality reduction and reinforcement learning methods to improve the efficiency and accuracy of the existing enhanced sampling methods. In this project, the conformational changes of both IDP and cellular membrane will be combined to construct the Markov State Models, in order to analyze the thermodynamic and kinetic properties of the interactions between IDP and cellular membrane. The results of this research project will have great biophysical significance.

自禀无序蛋白质(IDP)在人体蛋白质组中广泛存在，其结构的高度可塑性使得其在信号传导、转录后修饰等生命过程中起着重要作用。细胞膜作为生命的重要组成成分，参与调控着蛋白质的空间分布及结构变化。与此同时，由IDP自聚集所形成的淀粉样纤维在30余种人类疾病中被发现。细胞膜的存在加速了淀粉样纤维的形成，而细胞膜本身也在纤维形成的过程中被破坏。目前对IDP破坏细胞膜的机制仍有争论，且依然缺乏能统一不同实验结果的微观解释。本项目拟采用全原子分子动力学模拟，研究IDP在细胞膜环境下形成二级结构及其初始聚集的动力学过程。为了使模拟达到这一复杂构象变化的时间尺度，本项目拟运用非线性降维与强化学习方法，以提高现有增强采样方法的效率及准确性。本项目拟将IDP与细胞膜的构象变化结合共同建立马尔科夫态模型，从而分析IDP与细胞膜相互作用的热力学与动力学性质。本项目的研究成果将具有重要的生物物理意义。

作为结构高度可塑的蛋白质，自禀无序蛋白质（IDP）在信号传导、转录后修饰等诸多生命过程中起着重要作用，但由IDP自聚集所形成的淀粉样纤维在30余种人类疾病中被发现，并且IDP聚集的过程伴随着细胞的死亡。细胞膜的存在加速了淀粉样纤维的形成，而细胞膜本身也在纤维形成的过程中被破坏。目前，对于IDP破坏细胞膜的机制尚有争议。本项目通过全原子分子动力学模拟，研究了胰岛淀粉样多肽(IAPP)在细胞膜环境下的结构变化与自发聚集过程。从水溶液中具有无序结构出发，IAPP单体自发吸附到细胞膜，并在膜表面形成α螺旋与β片层结构。α螺旋结构进一步插入细胞膜内部并造成了膜的拉伸，而在表面的β片层导致了膜的弯曲。动力学上，β发夹结构的吸附与α螺旋结构的插入都造成了细胞膜局部动力学的减缓。相较于插入细胞膜的α螺旋结构，在表面的β发夹结构的扩散速率更慢，因而更易与水溶液中的单体相结合并发生聚集。更进一步，我们通过模拟观察到IAPP在细胞膜环境下的自发聚集，所形成寡聚体中的β片层结构进一步增加，而α螺旋略有减少，且单体构象进一步伸展，伸展后的IAPP在C端形成了类似于淀粉样纤维的cross-β片层结构。与此同时，β发夹结构中的疏水氨基酸进一步插入细胞膜内部。伴随着β片层结构的增加，IAPP寡聚化对细胞膜的破坏程度显著增加，且局部细胞膜的动力学进一步变慢。我们的模拟揭示了两种不同聚集机制的共存，从微观上解释了之前不同实验结果所带来的分歧。