利用分子動力學電腦模擬研究聚穀氨醯胺及胰島素的現象

Studying Polyglutamine and Insulin by Molecular Dynamics Simulations in Silico

江欣霖(Hsin-Lin Chiang)

指導教授 ： 陳俊榮 ； 胡進錕

國立清華大學/理學院/物理學系/博士(2014年)

https://doi.org/10.6843/NTHU.2014.00113

摘要

近年來，神經退化性疾病 (neurodegeneration disease) 的高罹病率使其在學界各領域受到相當高的關注。有許多的研究指出如阿茲海默症 (Alzheimer’s
disease)、帕金森氏症 (Parkinson’s disease)、亨丁頓跳舞症(Huntington’s disease) 等神經退化性疾病的病因皆來自不正常的蛋白質聚集 (abnormal protein aggregation)。然而，始終沒有研究確切指出人體內的蛋白質為何會聚集而造成神經退化性疾病的產生。在此篇論文中我將說明我如何使用分子動力學 (molecular dynamics) 模擬聚穀氨醯胺 (polyglutamine) 及胰島素 (insulin) 聚集的現象。聚穀氨醯胺的聚集已經被證實會造成漢廷頓舞蹈症，而胰島素的聚集則會造成糖尿病，此兩種疾病皆與神經退化性疾病有關。糖尿病二型 (type II diabetes) 雖然非神經退化性疾病，但其異常的胰島素聚集現象亦提高了神經退化性疾病的產生。利用電腦作分子動力學相較於實驗中的化學實驗是一種截然不同的方法，此方法使得科學家們有機
會觀察蛋白質在奈秒時間尺度下如何運動。為了瞭解聚穀氨醯胺如何聚集且形成 β-摺板 (β-sheet)，我們利用複製交換分子動力學 (replica-exchange
molecular dynamics) 對一條及兩條聚穀氨醯胺胜肽做模擬，我們連結了十
個穀氨醯胺來作成聚穀氨醯胺胜肽。這是在已知的學術發表論文中第一次使用全原子複製交換分子動力學及具體水分子 (explcit water molecules)模擬聚穀氨醯胺胜肽。由我們的結果得知，兩條聚穀氨醯胺胜肽的結構會隨著分子間距離而改變。當其距離較長時，此兩條聚穀氨醯胺胜肽會形成螺旋 (helix) 或無規捲曲 (coil)。當其距離較短時，此兩條聚穀氨醯胺胜肽會偶發地形成胜肽內的 β 摺板 (intra-peptide β-sheet)，最後形成胜肽間的 β-摺板 (inter-peptide β-sheet) 結構。我們也發現，聚穀氨醯胺二聚體偏好以反平行 β-摺板 (anti-parallel β-sheet) 的形式存在，此與目前已知的實驗結果相符。在第二個工作中，我使用統一原子 (united-atom) 分子動力學及顯性水分子研究體積較聚穀氨醯胺大的胰島素單體。目前已知研究指出胰島素可藉由疊加在 LVEALYL 胜肽上形成類澱粉纖維 (amyloid fibril)。此與胰島素穩定結合 (binding) 的 LVEALYL 胜肽來自胰島素的 B11-17 片段。我們發現，兩條 LVEALYL 胜肽會聚集而形成 β-摺板結構。由 molecular mechanic Poisson-Boltzman surface area(MM/PBSA) 方法我們亦可求得 LVEALYL 對胰島素的結合自由能相當強，此與已知結果相符。此外我們亦發現另一取自胰島素 B22-27 段的 RGFFYT 胜肽也有自我聚集現象且能與胰島素結合。

關鍵字

分子動力學 ； 蛋白質聚集

並列摘要

Study of neurodegenerative diseases is an important issue because of their high occurring rates all these years. Many reports indicated that neurodegenerative diseases such as Alzheimer’s disease, Parkinson’s disease, and Huntington’s disease are all caused by abnormal protein aggregation. Type II diabetes is not a neurodegenerative disease, but its abnormal insulin aggregation can increase the probability of neurodegenerative diseases. It is still not clear how do these proteins assemble and lose their function in vivo.
In this thesis I’ll report our results about polyglutamine and insulin aggregation from molecular dynamics simulation. Aggregation of polyglutamine and insulin have been proved as the cause of several human diseases, in particular to Huntington’s disease and diabetes. Molecular dynamics simulation in silico let scientists able to reach the experimental environment which protein can aggregate immediately and observe protein aggregation
process under nanosecond time scale. To study how polyglutamine aggregate and form β-sheets, we employed replica-exchange molecular dynamics to simulate one and two polyglutamine peptides with ten glutamine residues.
It is the first time that polyglutamine peptides are simulated by all-atom force field replica-exchange molecular dynamics accompanied with explicit water molecules. Our results show that the structures of two polyglutamine
peptides are changed depending on their inter-peptide distance. When the inter-peptide distance between two polyglutamine peptides is large, two peptides formed helix or coil structures as in the case of one chain. While the
inter-peptide distance decreases, the intra-peptdie β-sheet structures appear as an intermediate state occasionally, and become the inter-peptide β-sheets in the end. We also found that the polyglutamine dimer tends to form the anti-parallel β-sheet conformations rather than the parallel one, which is
consistent with previous experiments. In my second project, I employed united atom force field molecular dynamics with explicit water molecules to simulate insulin monomer, which is larger than polyglutamine peptide. Previous research indicated that insulin formed amyloid fibril via stacking on the peptide LVEALYL. LVEALYL is the fragment B11-17 of insulin, it is binding well with insulin. Our results shows that two LVEALYL peptides could aggregate as a β-sheet. By molecular mechanic Poisson-Boltzmann surface area (MM/PBSA) method we estimated the binding free energy of LVEALYL to
insulin, the result of strong binding affinity is consistent with the previous research. We also found a peptide RGFFYT, the fragment B22-27 of insulin, can aggregate and bind to insulin monomer, too. In our results we showed that RGFFYT has comparable binding affinity to insulin as LVEALYL.

並列關鍵字

molecular dynamics ； protein aggregation

參考文獻

[2] Andrej S˘ali, Eugene Shakhnovich, and Martin Karplus. How does a protein fold? Nature, 369:248–251, 1994.

[3] Nobuhiro Go. Theoretical studies of protein folding. Annu. Rev. Biophys. and Bioeng., 12(1):183–210, 1983.

[4] Peter E. Leopold, M. Mauricio Jos´e N, and Onuchic. Protein folding funnels - a kinetic approach to the sequence structure relationship. Proc. Natl. Acad. Sci., 18:8721–8725, 1992.

[5] Ken A. Dill, S. Banu Ozkan, M. Scott Shell, and Thomas R. Weikl. The protein folding problem. Annu. Rev. Biophys. Biomol. Struct., 37(1):289–316, 2008.

[7] Michel Goedert and Maria Grazia Spillantini. A century of Alzheimer’s disease. Science, 314(5800):777–781, 2006.