The ycjU mutation caused cells to be only slightly more susceptible to nalidixic acid than the wild-type strain in our bacteriostatic measurement (Table 1, MIC99 4.1 μg/ml vs. 4.5 μg/ml). Thus, ycjU may not belong in the set previously identified as having a low MIC KU-60019 solubility dmso [5]. The two-fold drop in LD90, from 59 μg/ml to 31 μg/ml (Fig. 1), qualified it as a gene with a hyperlethal phenotype. Mutant susceptibility to other antimicrobial agents and environmental stressors To determine whether the hyperlethal phenotype was restricted to quinolones, we examined the response of the

mutants to a variety of other stresses. When tetracycline was tested, we found that, except for two strains TL24 (ykfM::Tn5) and TL26 (ybcM::Tn5), the mutants were more readily killed (LD90 was about half the wild-type value, Fig. 1). Increased lethality was not observed for the β-lactam ampicillin (Fig. 1). Thus, increased killing of the mutants by antimicrobial agents was not restricted to quinolones, but it was also not universal.

When the DNA damaging agent mitomycin C was tested, all of the mutants were more readily killed SCH 900776 than wild-type cells (for some genes LD90 was 10% of wild-type values, many were in the 20 to 30% range, and two were close to 50%, Fig. 1). More than half of the mutants were more readily killed by UV irradiation than the wild-type strain (Fig. 2). Genes not involved in protecting cells from the effects of UV irradiation were rfbX, ybdA, yfbQ, ykfM, and ycjW. Nine others were involved in protecting cells from the effects Fossariinae of nalidixic acid, mitomycin C, and UV. Thus, many of the genes are involved in facilitating survival of E. coli cells exposed to DNA-damaging agents. Figure 2 Susceptibilities of insertion mutants to physical and chemical stresses. E. coli cultures grown to mid-log phase were treated with 2000 μJ/cm2 of UV; 2 mM H2O2, 10% SDS, or heat shock at 52°C for 15 min. Samples were diluted, applied to agar lacking stressor, and incubated to

determine the fraction of colonies surviving. This fraction was expressed as a percent of an untreated control culture sampled at the time stress was applied. In the case of SDS, some mutants grew during treatment, which caused those samples to have values higher than the control. Values reported are the means of 3 independent experiments. Error bars indicate standard deviations of means. The effect of hydrogen peroxide was also examined, since it has recently been implicated in the lethal action of multiple antibiotics [6, 7]. After treatment with 2 mM H2O2 for 15 min, all mutants showed decreased survival compared to wild-type strain DM4100 (for many mutants survival was 20 to 30% that of the wild-type strain, Fig. 2). We also examined the effects of two other environmental stresses, exposure to high temperature and to the ionic detergent sodium dodecyl sulfate (SDS).