Simulation-led redesign of polyketide synthase biocatalysts

酶是一种特殊的生物催化剂，具有反应条件温和、专一性强、选择性高等特点。如果酶可以被可预测地和准确地改造，它们可以提供一种有效的途径来获得新的分子，如药物。在制药（和许多其他应用）中，控制精确的立体化学结构至关重要，因为这可以区分药物或毒药。在这个项目中，目的是开发控制聚酮化合物脱氢酶的方法，聚酮化合物脱氢酶是一类重要的天然生物催化机器，它可以设置其产物聚酮化合物的立体化学。这将通过修饰或重新设计聚酮化合物合成酶系统中的一种关键酶来完成，该酶有助于在聚酮化合物产物链中设置立体化学，即酮还原酶。特别地，在所谓的聚酮合酶II型系统中，酰基载体蛋白（ACP）将使进化的聚酮链进入酮还原酶，其随后将设置立体化学中心。现有的结构信息有助于定义ACP和酮还原酶如何参与制造聚酮放线菌紫素（一种天然抗生素），这将指导计算预测协议的发展。这将是大部分的工作，涉及蛋白质-蛋白质对接，分子动力学模拟和QM/MM反应模拟。模拟将预测改变立体化学结果的新酮还原酶变体。为了测试和改进这些计算预测，将对有前景的酶变体（产物结果、动力学和结构生物学）进行实验表征。一旦成功，新变体的原子细节将通过结构生物学技术（NMR，X射线晶体学）进行确认。因此，这个跨学科项目结合了布里斯托酶计算模拟的专业知识和国际领先的学术团队的专业知识，该团队具有聚酮化合物系统和相关实验技术（酶学，分子生物学，化学和结构生物学）的多学科专业知识。以这种方式结合模拟和实验仍在发展，但将变得越来越重要。因此，在这个项目中开发的战略和协议，工作的关键模型系统，将一般用于类似的修改（聚酮化合物）生物合成活动。

Enzymes are remarkable biocatalysts that can work under mild conditions and with high specificity and selectivity. If enzymes could be engineered predictably and accurately, they could provide an efficient route to new molecules such as pharmaceuticals. In pharmaceuticals (and many other applications), it is crucially important to control the precise stereochemical structure, as this can make the difference between medicine or poison. In this project, the aim is to develop ways to control how polyketide synthases, an important class of natural biocatalytic machinery, set the stereochemistry of their products, polyketides. This will be done by modifying - or redesigning - a key enzyme in polyketide synthase systems that helps set stereochemistry in the polyketide product chain, a ketoreductase. In particular, in the so-called polyketide synthase type II systems, an acyl-carrier protein (ACP) will bring the evolving polyketide chain to a ketoreductase, which will subsequently set a stereochemical center. Existing structural information helps define how the ACP and the ketoreductase are involved in making the polyketide actinorhodin (a natural antibiotic), and this will guide the development of computational prediction protocols. This will be the bulk of the work, involving protein-protein docking, molecular dynamics simulation and QM/MM reaction simulations. The simulations will predict new ketoreductase variants that alter the stereochemical outcome. To test and improve these computational predictions, experimental characterisation of promising enzyme variants (product outcome, kinetics and structural biology) will be performed. Once successful, the atomic detail of new variants will be confirmed through structural biology techniques (NMR, X-ray crystallography). This interdisciplinary project is thus combining the expertise in computational simulation of enzymes in Bristol and the expertise from an internationally leading academic team with multidisciplinary expertise of polyketide systems and the relevant experimental techniques (enzymology, molecular biology, chemistry and structural biology). Combining simulation and experiment in this way is still developing but will become increasingly important. The strategies and protocols developed in this project, working on key model systems, will therefore be of general use for similar modification of (polyketide) biosynthetic activities.