細胞外調節激酶、纖維母細胞生長受體激酶與其抑制劑結合自由能之計算： 熱力學積分分子動態及分子嵌合模擬研究

Abstract

癌症一直為十大死因之一。因此與癌症有關的標靶藥物研發是目前國際間藥物化學家研究的主要方向之一。由於在實驗上合成藥物需要大量的人力、物力。因此若可以使用電腦來輔助藥物設計、藥物探索，不僅可以縮短研發的時程，更可以大量減少有機合成研發的成本。蛋白質激酶目前被認為與許多癌症有關。在本篇論文中，我們使用電腦輔助的方法來探討化學分子與蛋白質激酶間的作用力型態以及如何提升篩選具有生物活性的小分子。我們使用兩種理論方法，分別是熱力學積分及分子嵌合。

在熱力學積分中，我們所針對的蛋白質標的為ERK2。我們以原本已知的活性分子開始。預測了一系列新的類似物，這些類似物是以苯環上的極性或非極性官能基做為區別。在我們的結果中，羥基(OH) 取代後發現其與蛋白質間的結合能力約比之前研究中的抑制劑分子好2 kcal/mol，而其他官能基的取代約增進1 kcal/mol。我們的預測結果可為實驗學家提供有機合成、蛋白質與細胞實驗的理論依據。

除了分子動力學之外，我們也以分子嵌合計算探討FGFR1蛋白質幾個效應對豐富指數(enrichment factor)的影響。豐富指數為評價虛擬篩選優劣的指標之一。在豐富指數的研究中，我們採用DUD資料庫中活性及非活性的分子及設定不同的蛋白質條件。在設定條件時發現擺動514號胺基酸(賴氨酸)能夠大幅度增加正確挑選出活性分子的數量。並且設定絞鍊片段上564號胺基酸(丙胺酸)的氫鍵限制以進一步探究篩選效率。從本論文的研究中，我們尋找到最佳的條件，也希望這些最佳化條件可以進一步成為此類蛋白激酶大量虛擬篩選中的基礎。進而幫助先導化合物的找尋。

Cancer has been one of the top ten leading causes of death for several decades. The target drugs research for cancer therapy is now a popular field among international medicinal chemists. Because of the significant amount of money and human resources spent in the drug development process, computer-aided drug design method is an attractive tool to reduce cost and assist drug discovery. Protein kinases are one of the protein families which are drug targets for cancer therapy. Here, we selected two kinases, which are ERK2 and FGFR1 kinases, and used computer modeling to investigate binding energy of inhibitor-protein complexes for these two kinases. In the part of ERK2, we used thermodynamic integration MD method to compute relative binding free energy of several ERK2-inhibitor complexes of interest. We carried out computations to predict G for new analogs, focusing on placing polar and nonpolar functional groups at the meta site of benzene ring, to see if these ligands have better binding affinity than the above ligands. The computations resulted that a ligand with polar –OH group has better binding affinity than the previous examined ligand by ~2.0 kcal/mol and two other ligands have better affinity by ~1.0 kcal/mol. The predicted better inhibitors of this kind should be of interest to experimentalists for future experimental enzyme and/or cell assays. In addition to TI-MD simulation, we also worked on interactions of FGFR1 kinase-inhibitor complexes using docking computation, focusing on how enrichment factor (EF) enhances in virtual screening by including side chain movement and applying hydrogen bond constraint for this kinase. To this end, active and decoy compounds from the Directory of Useful Decoys 1 database was obtained and benchmarked with GOLD program. Interestingly, among combinations of side chains which were allowed to move, EF is significantly higher with movement of Lys514 compared with others. In addition, the effect of adding hydrogen bond constraint at a residue located in the hinge segment, Ala564, was also examined. The results were analyzed and discussed. The present results should be useful for virtual screening of large databases against this kinase.