Virtual screening for β-secretase (BACE1) inhibitors reveals the importance of protonation states at Asp32 and Asp228
Virtual screening for β-secretase (BACE1) inhibitors reveals the importance of protonation states at Asp32 and Asp228
复制标题
DOI：
10.1021/jm049133b
复制
发表时间：
2005-06-02
影响因子：
7.3
通讯作者：
Keserü, GM
中科院分区：
文献类型：
--
作者：
Polgár, T;Keserü, GM
使用不同的对接方法（FlexX和FlexX-Pharm）、评分函数（Dock、Gold、Chem、PMF、FlexX）、质子化状态（默认和计算）和蛋白质构象（载脂蛋白和配体结合）对β-分泌酶（BACE 1）抑制剂进行了比较虚拟筛选。发现BACE 1的Apo和配体结合构象都适合于虚拟筛选。对催化Asp 32和Asp 228残基的质子化状态的重新计算导致富集因子的显著改善，如在排名数据库的1%处计算的。使用1FKN，我们通过FlexX/D-Score没有获得富集，当考虑计算的质子化状态时，其提高到36。我们还表明，使用FlexX-Pharm/D-Score将计算的质子化状态与药效团约束相结合，将富集进一步提高到41。本研究中报告的富集表明，我们的筛选方案将有效地虚拟筛选BACE 1抑制剂的大型化合物库。
A comparative virtual screen for beta-secretase (BACE1) inhibitors using different docking methods (FlexX and FlexX-Pharm), scoring functions (Dock, Gold, Chem, PMF, FlexX), protonation states (default and calculated), and protein conformations (apo and ligand bound) has been performed. Apo and ligand bound conformations of BACE1 were both found to be suitable for virtual screening. Assigning calculated protonation states to catalytic Asp32 and Asp228 residues resulted in significant improvement of enrichment factors as calculated at 1% of the ranked database. Using 1FKN we obtained no enrichment by FlexX/D-Score that was improved to 36 when considering calculated protonation states. We also show that combining calculated protonation states with pharmacophore constraints using FlexX-Pharm/D-Score improved enrichment further to 41. Enrichments reported in this study suggest our screening protocol will be effective in the virtual screening of large compound libraries for BACE1 inhibitors.