基于功能模块计算辅助设计的新功能酶定制及其构效关系研究

Glutamate decarboxylase (GAD) can catalyze α-decarboxylation of L-glutamate into γ-aminobutyric acid (GABA). As a major inhibitory neurotransmitter in mammalian nervous systems, GABA has several physiological functions. The strategy “customizing and optimizing enzyme structures and properties in accordance to specific demands” is applied to GAD de novo computational design. Firstly, the catalytic function module of GAD is reprogrammed by computational enzyme design, and the GAD theozyme model as the minimum active centre is constructed according to crystal structures of two mammalian GADs with higher enzyme activity toward near-neutral pH and the geometry of transition-state conformation. In the next step, “inverse rotamer tree” algorithm is applied to scan a large set of thermostable protein scaffolds from PDB database with known X-ray structures to find out the backbone positions which could closely combine the transition-state and harbor GAD theozyme. And then Monte Carlo energy optimization method is applied to confirm the structure of customized GAD. The corresponding genes are synthesized according to the customized GAD amino acid sequences, and those enzyme variants with catalytic activity and thermostability are further improved by site-directed mutagenesis and site-saturation mutagenesis in order to fine-tune the micro-environment of the enzyme active center, and finally the the desired novel GAD will be created that features higher enzyme activity toward near-neutral pH and higher thermal stability. According to data analysis of enzyme reaction kinetics, the spectroscopic data (X-ray diffraction, fluorescence spectra and circular dichroism spectrum) and bioinformatics (molecular docking and molecular dynamics simulation), molecular mechanism and structure foundation of the customized GAD with higher thermal stability and neutral optimal pH will be illuminated. The research not only builds and improves theoretical system about “customizing and optimizing enzyme structures and properties in accordance to specific demands”, but also develops theory and technology of enzyme molecular modification.

根据“按照特定需要，定制和优化酶的结构和性质”的研究思路，以谷氨酸脱羧酶为模式体系，通过对酶催化功能模块的计算设计，建立酶催化反应过渡态模型和“理论酶”结构模型，再利用“inverse rotamer tree”算法搜索PDB数据库，识别可“容纳和支撑”理论酶结构的嗜热蛋白质分子骨架并进行“定制酶”的组装，采用Monte Carlo方法进行“定制酶”的结构－能量优化，然后合成对应的基因并进行“定制酶”的重组表达，最后通过定向进化方法进一步优化“定制酶”的结构和性质，最终获得催化活力高、热稳定性好、最适pH值为中性的全新谷氨酸脱羧酶。在此基础上，根据酶催化反应动力学、波谱学以及生物信息学数据分析，阐明全新谷氨酸脱羧酶的热稳定性和最适催化pH发生“质变”的分子机理和结构基础。该研究不仅能建立和完善“按需定制酶分子结构和性质”的理论体系和系统方法，而且为酶分子改造提供了新的理论体系和方法指导。

为了探索“根据特定的应用需要，对酶的结构与性质进行定制和优化”的技术路线和实现方法，研究以短乳杆菌谷氨酸脱羧酶的晶体结构模型为基础，运用Rosetta Ligand对接算法构建了L-谷氨酸与GAD反应过渡态最小活性中心模型，再利用"inverse rotamer tree"几何散列算法在PDB中搜索可以容纳理论酶空间构象的蛋白质分子框架，并将该蛋白质分子的框架目标区中原有序列改变为理论酶的氨基酸序列，得到与理论酶结构相匹配的蛋白质结构模型。采用Monte Carlo方法对得到的蛋白质结构模型的序列进行了能量优化, 根据理论酶在该蛋白质分子框架内的相容性和稳定性，筛选出了合理的结构模型，再根据大肠杆菌的密码子偏好性，将定制GAD酶的蛋白序列转化基因编码序列，然后通过全序列合成得到相应的基因，然后克隆到表达型载体pET28a(+)，并在大肠杆菌BL21(DE3)中进行定制酶的重组表达，但定制酶的催化活性较低。其原因可能是包装GAD活性中心的嗜热蛋白质骨架形成了不利于谷氨酸进入酶底物口袋、并与活性中心有效作用的空间位阻。为了探索更有效的酶热稳定性改造方法，研究进一步综合考察了进行脯氨酸替换、删除Loop区域表面不稳定氨基酸、序列一致性突变、引入二硫键、利用B-factor值结合FoldX能量优化等多种理性手段对GAD及(R)-ω-转氨酶进行了分子改造，不仅获得了多个热稳定性和催化活性都得到提高的突变酶，而且较为系统和全面的比较了不同方法的优劣，不仅为其他酶的热稳定性改造研究提供了案例参考和理论指导，也促进了酶分子高效改造理论和方法探索。