Bakal Ravindra L, Jawarkar Rahul D, Manwar J V, Jaiswal Minal S, Ghosh Arabinda, Gandhi Ajaykumar, Zaki Magdi E A, Al-Hussain Sami, Samad Abdul, Masand Vijay H, Mukerjee Nobendu, Nasir Abbas Bukhari Syed, Sharma Praveen, Lewaa Israa
Department of Medicinal Chemistry, Dr. Rajendra Gode Institute of Pharmacy, University-Mardi Road, Amravati, Maharashtra, India.
Department of Medicinal Chemistry and Pharmacognosy, Dr. Rajendra Gode College of Pharmacy, University-Mardi Road, Amravati, Maharashtra, India.
Saudi Pharm J. 2022 Jun;30(6):693-710. doi: 10.1016/j.jsps.2022.04.003. Epub 2022 Apr 7.
The aldose reductase (AR) enzyme is an important target enzyme in the development of therapeutics against hyperglycaemia induced health complications such as retinopathy, etc. In the present study, a quantitative structure activity relationship (QSAR) evaluation of a dataset of 226 reported AR inhibitor (ARi) molecules is performed using a genetic algorithm - multi linear regression (GA-MLR) technique. Multi-criteria decision making (MCDM) analysis furnished two five variables based QSAR models with acceptably high performance reflected in various statistical parameters such as, R = 0.79-0.80, Q = 0.78-0.79, Q = 0.78-0.79. The QSAR model analysis revealed some of the molecular features that play crucial role in deciding inhibitory potency of the molecule against AR such as; hydrophobic Nitrogen within 2 Å of the center of mass of the molecule, non-ring Carbon separated by three and four bonds from hydrogen bond donor atoms, number of sp2 hybridized Oxygen separated by four bonds from sp2 hybridized Carbon atoms, etc. 14 in silico generated hits, using a compound 18 (a most potent ARi from present dataset with pIC = 8.04 M) as a template, on QSAR based virtual screening (QSAR-VS) furnished a scaffold 5 with better ARi activity (pIC = 8.05 M) than template compound 18. Furthermore, molecular docking of compound 18 (Docking Score = -7.91 kcal/mol) and scaffold 5 (Docking Score = -8.08 kcal/mol) against AR, divulged that they both occupy the specific pocket(s) in AR receptor binding sites through hydrogen bonding and hydrophobic interactions. Molecular dynamic simulation (MDS) and MMGBSA studies right back the docking results by revealing the fact that binding site residues interact with scaffold 5 and compound 18 to produce a stable complex similar to co-crystallized ligand's conformation. The QSAR analysis, molecular docking, and MDS results are all in agreement and complementary. QSAR-VS successfully identified a more potent novel ARi and can be used in the development of therapeutic agents to treat diabetes.
醛糖还原酶（AR）是开发针对高血糖诱导的健康并发症（如视网膜病变等）治疗药物的重要靶标酶。在本研究中，使用遗传算法 - 多元线性回归（GA-MLR）技术对226个已报道的AR抑制剂（ARi）分子数据集进行了定量构效关系（QSAR）评估。多标准决策（MCDM）分析提供了两个基于五个变量的QSAR模型，其在各种统计参数（如R = 0.79 - 0.80，Q = 0.78 - 0.79，Q = 0.78 - 0.79）中表现出可接受的高性能。QSAR模型分析揭示了一些在决定分子对AR的抑制效力中起关键作用的分子特征，例如：分子质心2Å范围内的疏水氮、与氢键供体原子相隔三个和四个键的非环碳、与sp2杂化碳原子相隔四个键的sp2杂化氧原子数量等。以化合物18（本数据集中最有效的ARi，pIC = 8.04 M）为模板，在基于QSAR的虚拟筛选（QSAR-VS）中通过计算机生成了14个命中化合物，得到了一个支架5，其ARi活性（pIC = 8.05 M）比模板化合物18更好。此外，化合物18（对接分数= -7.91 kcal/mol）和支架5（对接分数= -8.08 kcal/mol）与AR的分子对接表明，它们都通过氢键和疏水相互作用占据AR受体结合位点中的特定口袋。分子动力学模拟（MDS）和MMGBSA研究通过揭示结合位点残基与支架5和化合物18相互作用以产生类似于共结晶配体构象的稳定复合物这一事实，支持了对接结果。QSAR分析、分子对接和MDS结果均一致且相互补充。QSAR-VS成功鉴定出一种更有效的新型ARi，可用于开发治疗糖尿病的治疗药物。
