Zięba Agata, Laitinen Tuomo, Patel Jayendra Z, Poso Antti, Kaczor Agnieszka A
Department of Synthesis and Chemical Technology of Pharmaceutical Substances with Computer Modeling Laboratory, Faculty of Pharmacy, Medical University of Lublin, PL-20059 Lublin, Poland.
School of Pharmacy, University of Eastern Finland, FI-70211 Kuopio, Finland.
Int J Mol Sci. 2021 Jun 6;22(11):6108. doi: 10.3390/ijms22116108.
This work aimed to construct 3D-QSAR CoMFA and CoMSIA models for a series of 31 FAAH inhibitors, containing the 1,3,4-oxadiazol-2-one moiety. The obtained models were characterized by good statistical parameters: CoMFA Q = 0.61, R = 0.98; CoMSIA Q = 0.64, R = 0.93. The CoMFA model field contributions were 54.1% and 45.9% for steric and electrostatic fields, respectively. In the CoMSIA model, electrostatic, steric, hydrogen bond donor, and hydrogen acceptor properties were equal to 34.6%, 23.9%, 23.4%, and 18.0%, respectively. These models were validated by applying the leave-one-out technique, the seven-element test set (CoMFA r = 0.91; CoMSIA r = 0.91), a progressive scrambling test, and external validation criteria developed by Golbraikh and Tropsha (CoMFA r = 0.98, k = 0.95; CoMSIA r = 0.98, k = 0.89). As the statistical significance of the obtained model was confirmed, the results of the CoMFA and CoMSIA field calculation were mapped onto the enzyme binding site. It gave us the opportunity to discuss the structure-activity relationship based on the ligand-enzyme interactions. In particular, examination of the electrostatic properties of the established CoMFA model revealed fields that correspond to the regions where electropositive substituents are not desired, e.g., in the neighborhood of the 1,3,4-oxadiazol-2-one moiety. This highlights the importance of heterocycle, a highly electronegative moiety in this area of each ligand. Examination of hydrogen bond donor and acceptor properties contour maps revealed several spots where the implementation of another hydrogen-bond-donating moiety will positively impact molecules' binding affinity, e.g., in the neighborhood of the 1,3,4-oxadiazol-2-one ring. On the other hand, there is a large isopleth that refers to the favorable H-bond properties close to the terminal phenoxy group of a ligand, which means that, generally speaking, H-bond acceptors are desired in this area.
这项工作旨在为一系列含有1,3,4-恶二唑-2-酮部分的31种脂肪酸酰胺水解酶(FAAH)抑制剂构建3D-QSAR CoMFA和CoMSIA模型。所获得的模型具有良好的统计参数:CoMFA的Q = 0.61,R = 0.98;CoMSIA的Q = 0.64,R = 0.93。在CoMFA模型中,空间场和静电场的贡献分别为54.1%和45.9%。在CoMSIA模型中,静电、空间、氢键供体和氢键受体性质分别为34.6%、23.9%、23.4%和18.0%。这些模型通过留一法、七元测试集(CoMFA的r = 0.91;CoMSIA的r = 0.91)、逐步重排测试以及由戈尔布赖赫和特罗普沙开发的外部验证标准(CoMFA的r = 0.98,k = 0.95;CoMSIA的r = 0.98,k = 0.89)进行了验证。由于所获得模型的统计学意义得到确认,CoMFA和CoMSIA场计算结果被映射到酶结合位点上。这使我们有机会基于配体 - 酶相互作用来讨论构效关系。特别是,对所建立的CoMFA模型的静电性质进行检查时发现了一些区域,这些区域对应于不希望有正电取代基的地方,例如在1,3,4-恶二唑-2-酮部分附近。这突出了杂环的重要性,杂环是每个配体这一区域中高度电负性的部分。对氢键供体和受体性质等高线图的检查揭示了几个位置,在这些位置引入另一个氢键供体部分将对分子的结合亲和力产生积极影响,例如在1,3,4-恶二唑-2-酮环附近。另一方面,在配体末端苯氧基附近有一个较大的等值线,表示有利的氢键性质,这意味着一般来说,该区域需要氢键受体。