The disruption of ung, the unique uracil-DNA-glycosylase-encoding gene in Bacillus subtilis, slightly increased the spontaneous mutation frequency to rifampin resistance (Rif r ), suggesting that additional repair pathways counteract the mutagenic effects of uracil in this microorganism. An alternative excision repair pathway is involved in this process, as the loss of YwqL, a putative endonuclease V homolog, significantly increased the mutation frequency of the ung null mutant, suggesting that Ung and YwqL both reduce the mutagenic effects of base deamination. Consistent with this notion, sodium bisulfite (SB) increased the Rif r mutation frequency of the single ung and double ung ywqL strains, and the absence of Ung and/or YwqL decreased the ability of B. subtilis to eliminate uracil from DNA. Interestingly, the Rif r mutation frequency of single ung and mutSL (mismatch repair [MMR] system) mutants was dramatically increased in a ung knockout strain that was also deficient in MutSL, suggesting that the MMR pathway also counteracts the mutagenic effects of uracil. Since the mutation frequency of the ung mutSL strain was significantly increased by SB, in addition to Ung, the mutagenic effects promoted by base deamination in growing B. subtilis cells are prevented not only by YwqL but also by MMR. Importantly, in nondividing cells of B. subtilis, the accumulations of mutations in three chromosomal alleles were significantly diminished following the disruption of ung and ywqL. Thus, under conditions of nutritional stress, the processing of deaminated bases in B. subtilis may normally occur in an error-prone manner to promote adaptive mutagenesis.

GSTD1 is an insect glutathione S-transferase that has received considerable attention because of its role in detoxification of xenobiotic compounds, specifically pesticides and plant allelochemicals involved in detoxification, or in the use of some substrates as a nutritional source. GSTD1 has been implicated in the adaptation to a new cactus host in Drosophila mojavensis and thus constitutes an interesting candidate to study ecological genetics of adaptation in Drosophila. We conducted population genetic and molecular evolution analyses of the GstD1 gene in the context of association with different cactus hosts (Opuntia sp. vs. Columnar) in nine Drosophila species from the repleta group. We observed strong evidence of selection in GstD1 from D. hexastigma. This species is associated with a diverse set of columnar cacti with very complex chemistries. GstD1 sequences from D. hexastigma show evidence of a recent selective sweep, and positive selection at one residue just outside of the active site of the enzyme. The substitution (Q116T) at the site under selection leads to a conformational change in the enzyme that could have important consequences for substrate binding efficiency. Our results suggest that GSTD1 from D. hexastigma may have evolved improved substrate binding in order to adapt to the diverse chemical environments that this species encounters in the wild.