In Silico Prediction of O⁶-Methylguanine-DNA Methyltransferase Inhibitory Potency of Base Analogs with QSAR and Machine Learning Methods.

Tengjiao Fan,Yuxing Hao,Xin Cui,Rugang Zhong,Lijiao Zhao,Yue Zhou,Yongzhen Peng,Guohui Sun,Xiaodong Sun,Ting Ren

doi:10.3390/molecules23112892

Abstract

O6-methylguanine-DNA methyltransferase (MGMT), a unique DNA repair enzyme, can confer resistance to DNA anticancer alkylating agents that modify the O6-position of guanine. Thus, inhibition of MGMT activity in tumors has a great interest for cancer researchers because it can significantly improve the anticancer efficacy of such alkylating agents. In this study, we performed a quantitative structure activity relationship (QSAR) and classification study based on a total of 134 base analogs related to their ED50 values (50% inhibitory concentration) against MGMT. Molecular information of all compounds were described by quantum chemical descriptors and Dragon descriptors. Genetic algorithm (GA) and multiple linear regression (MLR) analysis were combined to develop QSAR models. Classification models were generated by seven machine-learning methods based on six types of molecular fingerprints. Performances of all developed models were assessed by internal and external validation techniques. The best QSAR model was obtained with Q2Loo = 0.83, R2 = 0.87, Q2ext = 0.67, and R2ext = 0.69 based on 84 compounds. The results from QSAR studies indicated topological charge indices, polarizability, ionization potential (IP), and number of primary aromatic amines are main contributors for MGMT inhibition of base analogs. For classification studies, the accuracies of 10-fold cross-validation ranged from 0.750 to 0.885 for top ten models. The range of accuracy for the external test set ranged from 0.800 to 0.880 except for PubChem-Tree model, suggesting a satisfactory predictive ability. Three models (Ext-SVM, Ext-Tree and Graph-RF) showed high and reliable predictive accuracy for both training and external test sets. In addition, several representative substructures for characterizing MGMT inhibitors were identified by information gain and substructure frequency analysis method. Our studies might be useful for further study to design and rapidly identify potential MGMT inhibitors.

Highlights

DNA alkylating agents, such as temozolomide (TMZ) and carmustine (BCNU), have been widely used for treating various malignant tumors [1,2]
MGMTinhibitory inhibitorypotency potencyofofbase baseanalogs, analogs,respectively; respectively;(2)(2)gain gainsome someimportant important predictions descriptors or substructure information that can be used for discovering novel compounds with descriptors or substructure information that can be used for discovering novel compounds with desirable activities
A total of 42 predictive models were generated based on the training set. 10-Fold cross-validation was performed to evaluate the performance of based on the training set. 10-Fold cross-validation was performed to evaluate the performance of all all models, and the best models were chosen according to the values of classification accuracy (CA)

Summary

Introduction

DNA alkylating agents, such as temozolomide (TMZ) and carmustine (BCNU), have been widely used for treating various malignant tumors [1,2]. These agents can undergo enzymatic hydrolysis or spontaneous decomposition to generate reactive intermediates, which act as electrophilic reagents to alkylate DNA, RNA or proteins, resulting in the loss of normal physiological function of these biomacromolecules [3,4,5,6] They exert their anticancer activity through producing lesions at O6 -position of DNA guanine. QSAR and classification models can be used for rapidly screening potent drug candidates from chemical databases before their synthesis, which can reduce unnecessary chemical synthesis, biological activity tests and animal experiments [18] This appears attractive to chemical and drug manufacturers, and government agencies, especially in times of shrinking resources. In the context of design or discovery of novel compounds with desired MGMT inhibitory activity, these models offer a meaningful mechanistic interpretation, and provide some crucial information between trends in structural modifications and respective changes of biological activity

2.2.Results

Outliers Analysis and Applicability Domain of QSAR Models

Data Set Analysis

Performances of 10-Fold Cross-Validation

Performances of External Test Set

Identification and Analysis of Privileged Substructures

Representative

Data Set

Calculation of Molecular Descriptors

Model Development and Evaluation

Applicability Domain

Data Collection and Preparation