Rotation about the C6-O6 bond in O6-methylguanine: the syn and anti conformers can be of similar energies in duplex DNA as estimated by molecular modeling techniques.

O6-Methylguanine (O6MeGua) is generally regarded as the most important premutagenic lesion formed from carcinogenic methylating agents, so its structure and mechanism of action have received considerable attention. Two conformations for the methyl group in O6MeGua are possible: in one the methyl group is syn with respect to the N(1)-position of the purine, points into the helix, and disrupts hydrogen bonding potential; in the second the methyl group is anti with respect to the N1-position of the purine, and points into the major groove. Syn-O6MeGua has been observed when paired with thymine in duplex DNA as determined by NMR, while anti-O6MeGua has been observed when paired with thymine in X-ray diffraction studies. Herein, molecular modeling/computational chemistry is used to evaluate this apparent discrepancy. [N6-Methyladenine (N6MeAde) was also studied, because it is isoelectronic with O6MeGua, and because more information is available about its energetics. Syn-N6MeAde is computed to be favored in small molecules; however, the fraction of the anti-conformer is computed to be approximately 7%, which agrees well with experimentally determined values (4-12%). In contrast, the anti conformation for N6MeAde is calculated to be favored in duplex DNA, which is consistent with what has been observed experimentally using both NMR and X-ray diffraction techniques. The agreement between the calculated and experimentally determined results with N6MeAde suggest that the methods are reasonable.] For O6MeGua, a syn/anti ratio of approximately 10(3) is computed for small molecules. In duplex DNA, syn-O6MeGua is computed to be favored, but the anti-conformer is less than approximately 1 kcal/mol higher in energy. Whether syn- or anti-O6MeGua predominates may depend upon sequence context, as well as environmental factors. The comparison between O6MeGua and N6MeAde suggests a rationale for the puzzling observation that O6MeGua appears to be a cytotoxic lesion in eukaryotic, but not prokaryotic, cells.