Predictive Analysis and Validation of Critical Missense SNPs of the ABH2 Gene Using Structural Bioinformatics

Human DNA dioxygenase ABH2 is a key enzyme of the AlkB family of Fe(II)/α-ketoglutarate-dependent oxygenases, which is specialized in removing alkyl groups from damaged DNA bases in the cell nucleus. At the same time, the occurrence of single-nucleotide polymorphisms (SNPs) in the human ABH2 gene can lead to amino acid substitutions that, in turn, may disrupt the normal functioning of the ABH2 enzyme. Currently, databases contain information about more than 2500 nucleotide substitutions in the ABH2 gene. Using a comprehensive bioinformatics approach, in this review, we analyzed over 200 non-synonymous ABH2 SNPs with eleven prediction programs to identify variants capable of negatively affecting its enzymatic activity. The combination of various programs with different evaluation algorithms and scoring approaches allows us to more reliably identify potentially deleterious amino acid substitutions. Moreover, the differences between the programs used allowed for comparison of their tendency to predict amino acid substitutions as deleterious. Structural analysis of the ABH2-substrate complex showed that selected functionally significant SNPs often affect the organization of the active site, reduce the efficiency of substrate binding, and/or disrupt the coordination of Fe²⁺ and α-ketoglutarate cofactors, leading to changes in catalytic efficiency. The data obtained from the conducted analysis suggest that naturally occurring polymorphisms in the ABH2 gene found in the human population may reduce the repair efficiency of DNA dioxygenase ABH2 and, consequently, modulate susceptibility to oncogenesis and influence the effectiveness of antitumor therapy for carriers of these SNPs.