請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/63212
標題: | 電腦輔助藥物設計模擬策略 A Simulation Strategy in Computer Aided Drug Design |
作者: | Shih-Jen Lu 呂適任 |
指導教授: | 張璞曾 |
關鍵字: | 分子接合,分子動態模擬,類黃酮衍生物,分子力場,H1N1, Molecular docking,molecular dynamics simulation,molecular force field,H1N1, |
出版年 : | 2012 |
學位: | 博士 |
摘要: | 電腦輔助藥物設計已經慢慢的成為藥物設計與開發中不可或缺的一環。然而,傳統的電腦輔助藥物設計,往往受限於既有的資訊不足,而使得空有許多分析工具與方法,卻英雄無用武之地。例如接受體-藥物三維結構影像常被用來作為結構為基礎藥物設計的重要資訊,透過許多分子模擬工具或方法,可以獲得許多藥物設計與開發的重要資訊,然而此三維結構的獲得卻常會受限於該結構是否已經為科學界所發現並公開,短則數月長則甚至數年才可能被揭露,更甚者該結構也可能永遠不會被揭露,如此一來輕則延宕藥物開發前期時程,重則可能錯失一可行藥物的開發。有鑑於此,我們提出一個完整、可行且有別於傳統以結構為基礎的藥物設計方法,可以用來做為當三維結構資訊不足時,透過我們所提出的方法獲得一個預測而來的分子三維結構,並經由分子接合得到一個最佳的三維結構,然後進行一系列的分子動態模擬分析,獲得結合自由能藉以確認最佳結構是穩定且可行的,並進行相關分析。
首先,我們採取依照現有存在的接受體-藥物的三維結構影像,輔以藥物的既有結構,使用比對排列的方式將此接受體-藥物三維結構影像中的藥物移除,並將與藥物作用相似且已經被證實具有一定程度療效的某一類化合物放入原先三維結構影像的接合部位,接著使用分子接合的方式找尋到能量最低之最佳接受體-化合物之預測結合構型;接著使用NAMD等分子動態模擬軟體搭配一些理想的溶劑參數、分子力場選擇以及模擬環境的設定與處理,將分子模擬所獲得到的最佳結合構型進行自由能的計算,以確定這樣的構型連接是穩定且可行的。這些最佳構型的三維結構影像將會在被用來做為功能性關鍵胺基酸殘基分析,以及鍵結力等相關分析。此外,為了確定此接受體-化合物之預測結合構型之穩定性在溶劑效應中也能保持穩定,採用溶劑影響能量法用來作為溶劑的自由能計算確定其穩定性。 在這個研究中,我們選定了H1N1神經醯胺酵素以及二十個已經被證實有療效之類黃酮衍生物做為我們實現此一藥物模擬策略的範例,利用這一系列的方法,我們成功的獲得了此二十個類黃酮素以及H1N1神經醯胺酵素的接受體-分子三維結構影像,並透過分子接合的RMSD、模擬過程中的結合自由能以及溶劑影響能量法的結合自由能計算(自由能<0),我們可以確認這些預測三維結構是穩定的,加上使用臨床上IC50 轉換連接自由能的比對,這兩個不同自由能相關係數為0.75,更證明我們這樣的自由能計算是可信的。在獲得穩定可信的二十個接受體-藥物三維結構影像,我們透過LIGPLOT作這二十個類黃酮素以及H1N1神經醯胺酵素連接的胺基酸殘基與連接力之分析,獲得發展此類藥物的重要模擬資訊。我們證實了使用我們提出的這一個模擬策略,將可以在三維結構資訊不足時,透過預測一個可能且可信的藥物模擬策略,獲得一個可能的三維結構,省卻掉被動等待真實三維結構,而主動的預測可能的三維結構,並使用自由能以及臨床上IC50轉換自由能來比對確認預測之三維結構為穩定且可信的結構,並進一步的分析得到藥物開發時的重要資訊。我們相信這一個藥物設計的方法不僅僅可節省藥物開發前期的時間,更提供更為主動的電腦輔助藥物設計的方法。 Receptor-drug three-dimensional binding complex structures are used as necessary input for structure-based drug design. Many important information for drug design and development can be obtained with the three-dimensional binding complex structures through molecular simulation means. However, obtaining three-dimensional binding complex structures is limited by whether the structure has been disclosed by academia. These structures usually are delayed disclose for either several months or several years, or even never. Therefore, it will result in the delay for early-stage drug design, and even the loss of the opportunity of drug discovery. In view of this, we propose a complete and feasible drug design method, which is different from the traditional structure-based drug design method. As long as we have enough information to predict or conjecture the possible binding site between receptor and ligand, we can obtain the reasonable and credible three-dimension structure for further simulations. Through this method, a predicted three-dimensional binding complex structure can be obtained and its conformation will be optimized by molecular docking method. The best conformation will be proved to be stable and viable by binding free energy with a series of analyses form molecular dynamics simulation. Much information will also be obtained. First, we adapted existing receptor-drug three-dimensional binding complex structures and the structure of the drug. Then we aligned the Receptor-drug structures and used the template structure drug to generate the binding site of the structures. Selected derivatives efficacy will be used to dock into the binding site of the receptor-drug structures. Though molecular docking, the best conformation of the predicted structure will be obtained. Then binding free energy calculation is used to verify whether the conformation is stable and feasible. By using LIGPLOT, functionally key residues and the affinity of the structures will be obtained. Besides, the SIE binding free energy could be calculated by SIE methods in order to confirm the stability in solvated interactions. A case study using this simulation strategy was performed. We selected the H1N1 neuraminidase and 20 flavonoid derivatives inhibitors. Then We aligned the H1N1 neuraminidase structure (PBD ID: 3NSS) and the template (PDB ID: 3B7E) [43], and used the template structure drug (zanamivir) to generate the active site of the H1N1 neuraminidase structure. The 20 flavonoid derivatives inhibitors will be used to dock into the binding site of the Receptor-drug three-dimensional binding structures. The best conformations will be obtained though molecular docking. RMSD and the binding free energy will be used to confirm the best conformations and to verify results from molecular docking are stable and feasible. From our simulations results, the correlation coefficient between the predicted binding free energies and experimental values of the 20 inhibitors is equal to 0.75. The results from our molecular simulation and clinical experiments are very relevant. Therefore, these results prove that our simulation strategy was similar to the actual situation. In our opinion, the simulation strategy will successfully compensate the delay and save funds in early drug development. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/63212 |
全文授權: | 有償授權 |
顯示於系所單位: | 電機工程學系 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
ntu-101-1.pdf 目前未授權公開取用 | 4.4 MB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。