Groups of coevolving positions provide drug resistance to Mycobacterium tuberculosis: A study using targets of first-line antituberculosis drugs

Drug resistance has been associated with point mutations in coding regions leading to an altered protein sequence and structure. Such changes have been seen as isolated events occurring at various positions in a sequence. However, we hypothesise that it is not a single mutation at a specific position but a group of positions that coevolve in a correlated fashion to increase the fitness of a target protein against a drug. To prove the hypothesis, selected protein sequences of Mycobacterium tuberculosis drug resistance genes were successfully screened using a bioinformatics approach to detect groups of coevolving amino acids at important structural and functional positions in the targets of first-line antituberculosis drugs (isoniazid, rifampicin, ethambutol and pyrazinamide). The algorithmically characterised genetic mutations and the lineage-specific single nucleotide polymorphisms (SNPs) detected previously in drug resistance genes of M. tuberculosis complex genomes were also found in the identified coevolving groups. Mapping of coevolving positions to the secondary structure of proteins clearly indicates the preference of amino acid residues in the helix to coevolve. Moreover, active-site residues of some candidate proteins were also found in coevolving groups. The coevolving groups detected in this study will be useful to gain new insights into the molecular and evolutionary basis of drug resistance. This work provides an important first step towards finding solutions to the multidrug resistance problem through coevolution analysis of proteins, in turn helping to develop new drug regimens against pathogens, including M. tuberculosis.