DNA interaction of the CcrM DNA methyltransferase: a mutational and modeling study

Abstract

Caulobacter crescentus CcrM is a DNA-(adenine N6)-methyltransferase that methylates adenine in the sequence GANTC with high specificity. To investigate its mechanism of DNA recognition, we used the crystal structure of a related methyltransferase (M1.MboII, which modifies GAAGA) as a starting point, and docked into it a DNA substrate to identify the protein regions that approach the DNA. After alignment of CcrM and M1.MboII, we identified four candidate regions in CcrM to contain residues involved in DNA recognition. We mutated 20 amino acid residues within these regions, purified the CcrM variants, and determined their DNA-binding and catalytic activity on a cognate GANTC substrate and on nine near-cognate substrates, each of which contained a single base-pair substitution in the recognition sequence. Altogether, we identified four residues in two of the regions, mutations of which resulted in a strong (>100-fold) reduction of methylation activity. Our data show that DNA recognition by CcrM is a cooperative process, because disruption of critical contacts led to loss of catalytic activity but not to a relaxation in specificity. In addition, we identified a change in the readout of the fifth base pair in the GANTC sequence with two other CcrM variants that showed smaller reductions in overall activity. Based on this and the sequence alignment of CcrM with other DNA methyltransferases of same or related recognition sequence, we propose roles for these two regions in DNA recognition by CcrM.