Development of multi-scale models for enzyme catalysis in complex environments

威斯康星大学的崔强在化学系化学理论、模型和计算方法计划的支持下，开发了新的计算方法，可以用理论技术有效地分析复杂环境中的酶催化。第一个开发的目标是涉及酶中复杂金属簇配体的化学反应。Cui和他的同事正在开发一种新的量子力学/分子力学(QM/MM)框架，在该框架中，只有反应配体和其他明确参与化学的基团被QM处理，而其余的金属基序则在MM水平上用先进的价键力场处理。在反应过程中，QM配体和MM金属基序之间的相互作用的变化是用化学势均衡的方法来处理的，这避免了显式地将QM区域视为一个开放系统所带来的技术复杂性。第二个主题涉及混合分辨率模拟的发展。这一开发利用了Cui小组建立的针对水、脂肪和氨基酸的粗粒度(CG)模型，该模型具有仔细的静电处理功能。对于原子/CG相互作用的校准，除了使用全原子模拟作为参考的常见策略外，这种方法的独特之处在于还利用了多组分溶液的实验热力学数据。这些发展满足了与理解酶如何催化化学反应相关的计算和理论挑战。通过避免显式处理金属团簇的复杂电子结构，新的QM/MM框架允许计算化学家和生物学家研究具有生物相关性的复杂系统中的化学反应。它特别适用于模拟有趣和重要的复杂过程，例如光合作用中发生的质子耦合电子转移。在许多应用中，需要对蛋白质的高分辨率描述和对复杂环境的真实描述，例如具有特定曲率的多组分脂质双层，需要混合原子/CG方法。这些方法在超越物理化学的广泛领域中得到应用，包括酶学、生物物理/生物化学和生物无机化学。混合分辨率模型的发展极大地扩展了系统的规模和复杂性，这些系统可以通过计算进行有意义的处理，从而影响材料科学和细胞生物化学。该项目中开发的计算方法已在流行的分子模拟程序包CHARMM中实现，并可用于研究化学和生物中的广泛问题。该项目的外展部分有助于强调理论和计算化学在理解自然方面的作用，并鼓励年轻学生追求科学事业。

Qiang Cui of the University of Wisconsin is supported by the Chemical Theory, Models and Computational Methods program in the Chemistry Division to develop novel computational methods such that enzyme catalysis in complex environments can be effectively analyzed with theoretical techniques. The first development targets chemical reactions that involve ligands of complex metal clusters in enzymes. Cui and coworkers are developing a novel quantum mechanical/molecular mechanical (QM/MM) framework in which only the reactive ligand and other groups that explicitly participate the chemistry are treated with QM, while the rest of the metal motif is treated at a MM level with an advanced valence-bond force field. Variations in the interaction between the QM ligand and the MM metal motif during the reaction are treated with a chemical potential equalization approach, which avoids technical complications associated with explicitly treating the QM region as an open system. The second subject concerns the development of mixed-resolution simulations. This development takes advantage of a coarse-grained (CG) model established in the Cui group for water, lipids and amino acids that features a careful treatment of electrostatics. For the calibration of atomistic/CG interactions, in addition to the common strategy of using all-atom simulations as a reference, the unique aspect of this approach is to also take advantage of experimental thermodynamic data for multi-component solutions.These developments meet the computational and theoretical challenges associated with understanding how enzymes catalyze chemical reactions. By avoiding the explicit treatment of the complex electronic structure of metal clusters, the novel QM/MM framework allows computational chemists and biologists to study chemical reactions in complex systems of biological relevance. It is particularly powerful for modeling interesting and important complex processes such as the proton-coupled-electron-transfers that occur in photosynthesis. The mixed atomistic/CG approach is demanded in many applications that require a high-resolution description of the protein and a realistic description of the complex environment, such as a multi-component lipid bilayer with specific curvature. These methods find applications in broad areas that go beyond physical chemistry to include enzymology, biophysics/biochemistry and bio-inorganic chemistry. Development of mixed-resolution models greatly expand the scale and complexity of systems that can be meaningfully treated with computations, and thus impact materials science and cellular biochemistry. The computational methods developed in this project are implemented into the popular molecular simulation package, CHARMM, and can be used to study a broad range of problems in chemistry and biology. The outreach component of the project helps emphasize the role of theoretical and computational chemistry in understanding nature and stimulate young students to pursue careers in science.