生物质可以在很多方面替代石油产品。基于生物质的生物转化平台可以生产出成本低廉，结构新颖的新型化工产品。由2-吡喃酮香豆酸与乙烯通过Diels-Alder反应合成的香豆酸双环内酯，就是一种非常具有潜力的生物转化平台，它可以被转化为各种多官能团化的环己二烯和苯，在医药，农药等多方面具有潜在的应用价值。在一系列的双环内酯转化反应中，其无催化剂的醇解反应引起我们的关注。香豆酸双环内酯的醇解反应属于溶液参与的反应，研究甲醇溶剂对反应的影响对更深刻的理解反应机理具有重要意义。在此醇解反应中，甲醇溶剂具有双重作用。一方面甲醇分子本身作为反应物直接参与醇解过程，另一方面甲醇也构成了溶液环境，间接协助反应进行，这导致反应具有更加复杂的机理。实际上，已有的研究香豆酸双环内酯醇解反应机制的工作考虑并不够全面。本论文对香豆酸双环内酯的醇解反应进行更细致的探讨。我们从协同机制和分步机制，以及不同数量的溶剂分子参与对反应能量带来的影响等方面深入地研究了该醇解反应的微观过程。分别基于反应物分离和反应复合物的两种不同模型对反应机理进行的研究，结果表明对于此反应体系，香豆酸双环内酯在甲醇中的醇解反应以三分子甲醇参与的分步机制进行，计算得到的动力学数据与实验一致。基于对反应的能量计算过程进行分析，基于氢键复合物的反应物模型计算得到的能量比基于分离反应物所得能量更为准确，原因在于后者在能量计算过程中引入了较大的溶剂化效应和熵效应计算误差，而前者所依据的反应复合物则避免了以上误差。同时，进一步的巨动力学计算表明，在反应条件下，香豆酸双环内酯和参与反应的甲醇分子能够形成稳定的氢键复合物，这也从物理机制上肯定了使用反应复合物作为反应的能量起点的合理性。

在攻读硕士研究生期间，本人还与学院其他研究生合作进行了酚酸类化合物与溶菌酶的非共价相互作用的研究工作，本人的工作是利用分子对接方法探讨酚酸类药物分子与溶菌酶蛋白质之间结合位点和结合方式，阐明药物分子和蛋白质之间非共价相互作用的识别机制。该工作对于理解酚酸类药物在体内的传递，以及进一步快速筛选临床候选药物和生物标志物具有重要意义。因为传统的实验测量手段对溶液中非共价相互作用的真实表征有局限性，所以本工作在质谱和光谱等实验方法的基础上，采用分子对接的方法，研究绿原酸、迷迭香酸, 1,3-O-二咖啡酰基奎宁酸和4,5-二咖啡酰奎宁酸这四种酚酸类化合物与溶菌酶的非共价相互作用。对接结果表明配体的结合位置为溶菌酶的裂隙处。在对接复合物中，酚酸类化合物的苯环与对接位点周围的色氨酸、异亮氨酸等非极性残基存在不同程度的疏水作用，这与实验结果一致，表明疏水作用力在复合物形成过程中起主要作用。此外配体与溶菌酶残基之间的氢键和静电作用也促进了它们的结合。分子对接方法作为一种初步探索大分子之间的几何构型对接状况和表征非共价相互作用的研究手段，对指导深入研究酚酸与蛋白质相互作用提供了有益的启示。

Bio-based platform molecule 2-pyrone coumalic acid (CMA) and its derivatives have received special interest and are the subject of extensive research due to their potential to be transformed into the bicyclic intermediate through Diels-Alder chemistry. Based on bicyclic lactone, Brent et al. recently developed the chemical conversion pathways that bridge biomass-derived bulk chemicals to higher-value products, such as dihydrobenzenes and 1,3-diacid six-membered rings. They particularly noted that bicyclic lactone could be converted to a novel species 6-hydroxy-3-(methoxycarbonyl)cyclohex-1-ene-1-carboxylic acid through a catalyst-free ring-opening reaction in the methanol solvent. In this unexpected methanolysis process, the solvent methanol gets involved in the reaction as the reactant. The chemical reactions proceed in the polar solvent usually experience very complex mechanisms. In the methanolysis reaction, the methanol solvent has a dual role. On the one hand, methanol molecules participate in the methanolysis process as a reactant directly. On the other hand, methanol also constitutes a solution environment, which assists the reaction indirectly. There are multiple aspects of the inter-relationship of solvent effect and mechanism to consider. However, Brent’s DFT calculations were less comprehensive into the methanolysis mechanism, causing their results to contradict with the experimental value. They explored only one path and did not fully consider the impact of the number and configurations. The alternatively new methanolysis mechanism with a lower energy barrier must exist to be responsible for the methanolysis of the bicyclic lactone molecule. According to their results, the methanolysis of bicyclic lactones intermediate follows the concerted methanolysis mechanism, the concerted methanolysis encounters a high-energy barrier which thus contradicts experimental value. Here we will revisit the methanolysis process of the bicyclic lactones intermediate using metadynamics simulation and the classical transition state theory. We deem that bicyclic lactones prefer to follow the stepwise methanolysis mechanism rather than the concerted mechanism. The former has a lower and more consistent with the experiment. Therefore, the stepwise mechanism is much more completive and should be considered as the dominant mechanism responsible for the methanolysis of the bicyclic lactones intermediate when studying the kinetics of the relevant reactions in the future. At the same time, when studying the mechanism of the solvent reaction, the entropy effect of the conformational rearrangement of the solvent must be investigated.

Elucidating the recognition mechanisms of non-covalent interaction between pharmaceutical molecules and proteins is important for understanding the drug delivery in vivo, and for further rapid screening clinical drug candidates and biomarkers. Hence, the study of the interaction between phenolic acids and protein is of great significance to understand its transport in vivo, improve bioavailability, and predict the toxicity risk. There are limitations on the real characterization of the non-covalent interactions in a solution using traditional experimental techniques. The molecular docking method can better analyze the non-covalent interaction of ligand and lysozyme from a microscopic perspective.

Our work uses molecular docking methods to study the non-covalent interactions of four phenolic compounds (Caffeoylquinic acid (chlorogenic acid, CQA), rosmarinic acid (RA), 1,3-O-dicaffeoylquinic acid (1,3-CQA), 4,5-dicaffeoylquinic acid (4,5-CQA)) and lysozyme. Molecular docking was performed to investigate the binging sites and binging modes of phenolic acid on Lysozyme. It shows that four ligands are all located at the deep cleft of lysozyme in our work. In the docking complexes, the aromatic rings from these four ligands showed a different extent of hydrophobic interaction with the nonpolar residues such as Trp, Leu and Ile around the cleft suggesting that hydrophobic force play a major role during the complexes formation, which support mass spectrometry (MS) and isothermal titration calorimetry (ITC) results. Besides the hydrophobic interactions, there are also hydrogen bonds formed between the hydroxyl groups of phenolic acid ligands and lysozyme. Besides, the docking shows that there is approximately one binding site for ligands binding to lysozyme via electrostatic forces. Electrostatic interaction between carboxyl anion of phenolic acids and positively charged residues of lysozyme also facilitates their binding. The molecular docking can more comprehensively and truly characterize the non-covalent interaction and can guide further research on the interaction of phenolic acid with other proteins. The information obtained in this study could provide a better understanding of the bioactivities of phenolic acids. The developed strategy could be applied to study noncovalent interactions between natural products and other proteins.