Studies on Aminoglycoside Susceptibility Identify a Novel Function of KsgA To Secure Translational Fidelity during Antibiotic Stress

Abstract

Antibiotic resistance has become a global crisis. Studies on the mechanism of bacterial tolerance to antibiotics will not only increase our conceptual understanding of bacterial death but also provide potential targets for novel inhibitors. We screened a mutant library containing a full set of in-frame deletion mutants of Escherichia coli K-12 and identified 140 genes that possibly contribute to gentamicin tolerance. The deletion of ksgA increased the inhibition and killing potency against mid-log-phase bacteria by aminoglycosides. Initially identified as a 16S rRNA methyltransferase, KsgA also has additional functions as a ribosomal biogenesis factor and a DNA glycosylase. We found that the methyltransferase activity of KsgA is responsible for the tolerance, as demonstrated by a site-directed mutagenesis analysis. In contrast to the mechanism for cold sensitivity, the decreased tolerance to aminoglycoside is not related to the failure of ribosomal biogenesis. Furthermore, the DNA glycosylase activity of KsgA contributes minimally to kanamycin tolerance. Importantly, we discovered that KsgA secures protein translational fidelity upon kanamycin killing, in contrast to its role during cold stress and kasugamycin treatment. The results suggest that the compromise in protein translational fidelity in the absence of KsgA is the root cause of an increased sensitivity to a bactericidal aminoglycoside. In addition, KsgA in the pathogenic Acinetobacter baumannii contributes not only to the tolerance against aminoglycoside killing but also to virulence in the host, warranting its potential application as a target for inhibitors that potentiate aminoglycoside therapeutic killing as well as disarm bacterial virulence simultaneously.

Keywords: Acinetobacter baumannii; KsgA; aminoglycosides; translational fidelity; virulence.

FIG 1

Deletion of ksgA results in decreased tolerance to aminoglycosides. (A) Results of biological process (BP) category of GO enrichment analysis. (B) Survival of E. coli BW25113 and three mutants (mraW, rrmJ, and ksgA) from the Keio collection treated with 20 μg/ml of gentamicin for 2 h. Survival of E. coli MG1655 and the ΔksgA mutant exposed to 100 μg/ml of kanamycin (C) and 20 μg/ml of gentamicin (D).

FIG 2

Deficiency of methyltransferase activity results in decreased kanamycin tolerance. Survival of E. coli MG1655 and ΔksgA mutants complemented with copies of KsgA with different mutations in the presence of 100 μg/ml kanamycin after 3 h of treatment. Equivalent results were obtained at least in triplicates. ev, empty vector.

FIG 3

Deficiency in KsgA did not affect ribosomal biogenesis. (A) Ratio of nonmature terminus to total 16S rRNA in E. coli MG1655 and ΔksgA cells via qRT-PCR at 0 h and 3 h after 100 μg/ml of kanamycin treatment. Error bars indicate standard deviations. ns, no statistically significant difference. (B) Western blots showing the expression of inducible FLAG-tagged protein upon 100 μg/ml kanamycin treatment. (C) Western blots showing the expression of constitutive eGFP in wild type and the ΔksgA mutant upon 100 μg/ml kanamycin treatment.

FIG 4

DNA glycosylase activity does not contribute to bacterial tolerance against kanamycin. Killing efficiency of E. coli MG1655 and the ΔksgA mutant upon kanamycin in the presence of 1,500 μM thiourea (A) and 50 μM 2,2′-dipyridyl (B). Log₁₀ changes in CFU/ml following exposure are shown.

FIG 5

Dual-luciferase assay of read-through and frameshift upon kanamycin treatment. (A) Reporter constructs that monitor the frequency of stop codon read-through (UGA and UAG) and frameshifting (+1 and −1) are shown. (B) Assessment of stop codon read-through and frameshifts in E. coli MG1655 and the ΔksgA mutant in the presence of 1.5 μg/ml of kanamycin, 100 μg/ml of kasugamycin at 37°C, or incubation at 25°C. The translational efficiency of each construct was measured by the chemiluminescence ratio of Fluc to Rluc (F/R), and the fold change of translational efficiency of the ΔksgA mutant to that of the wild type was measured by dividing the F/R value of the ΔksgA mutant by that of the wild type. The P values for comparisons between the control and different culture conditions were calculated by one-way analysis of variance (ANOVA). ***, P < 0.001; **, P < 0.01; *, P < 0.05.

FIG 6

Membrane disruption measurement by flow cytometry. The fluorescence intensity was profiled for E. coli MG1655 and the ΔksgA mutant at 0 h (A and E), 3 h after 100 μg/ml kanamycin (B and F) or 100 μg/ml kasugamycin (C and G) treatment, or incubation at 25°C (D and H). The horizontal axes indicate the intracellular fluorescence intensity of DiBAC₄ (3), and the vertical axes indicate the numbers of cells.

FIG 7

KsgA contributes to kanamycin tolerance and bacterial virulence in A. baumannii. (A) Survival of A. baumannii wild-type ATCC 17978 and the ΔksgA_AB mutant in the presence of 25 μg/ml of kanamycin. Equivalent results were obtained at least in triplicates, and the representative results are shown. (B) Survival of C. elegans upon infection by A. baumannii wild type and the ΔksgA_AB mutant.