Clinically relevant mutant DNA gyrase alters supercoiling, changes the transcriptome, and confers multidrug resistance

Abstract

Bacterial DNA is maintained in a supercoiled state controlled by the action of topoisomerases. Alterations in supercoiling affect fundamental cellular processes, including transcription. Here, we show that substitution at position 87 of GyrA of Salmonella influences sensitivity to antibiotics, including nonquinolone drugs, alters global supercoiling, and results in an altered transcriptome with increased expression of stress response pathways. Decreased susceptibility to multiple antibiotics seen with a GyrA Asp87Gly mutant was not a result of increased efflux activity or reduced reactive-oxygen production. These data show that a frequently observed and clinically relevant substitution within GyrA results in altered expression of numerous genes, including those important in bacterial survival of stress, suggesting that GyrA mutants may have a selective advantage under specific conditions. Our findings help contextualize the high rate of quinolone resistance in pathogenic strains of bacteria and may partly explain why such mutant strains are evolutionarily successful.

Importance: Fluoroquinolones are a powerful group of antibiotics that target bacterial enzymes involved in helping bacteria maintain the conformation of their chromosome. Mutations in the target enzymes allow bacteria to become resistant to these antibiotics, and fluoroquinolone resistance is common. We show here that these mutations also provide protection against a broad range of other antimicrobials by triggering a defensive stress response in the cell. This work suggests that fluoroquinolone resistance mutations may be beneficial under a range of conditions.

FIG 1

Differences in growth between strains and the impact of the addition of half the MIC for SL1344 of ampicillin, gentamicin, and triclosan on growth. The graph shows average culture OD values at 4 and 8 h postinoculation; averages that are statistically significantly different from those of the parent strain, SL1344, are marked by asterisks. Open bars indicate SL1344, blue bars indicate L821 (GyrA Ser83Phe), and red bars indicate L825 (GyrA Asp87Gly).

FIG 2

Respiratory activity of SL1344 (black lines), L821 (blue lines), and L825 (red lines) in the presence of eight separate antimicrobials in the Biolog Phenomicroarray. Circles and triangles represent separate biological replicates in each experiment.

FIG 3

Supercoiling activity and expression of gyrA and topA by SL1344, L821, and L825. (A) Separation of pBR322 isolated from each strain on a representative 0.9% agarose gel containing 25 mg/liter chloroquine. Red arrows indicate altered supercoiling; black arrows indicate direction of electrophoresis. (B) Densitometry plots of the lanes in the gel from panel A. (C) Expression of gyrA and topA, measured by comparative RT-PCR. Error bars indicate standard deviations, and asterisks indicate values that are significantly different from the value for SL1344.

FIG 4

Expression of genes significantly changed with chromosomal order, with genes organized from start to finish of the SL1344 chromosome, left to right. Expression of “1” indicates no significant change; higher values show induced genes, and lower values show repressed genes. Clusters of at least ten consecutive genes with expression altered in a similar manner belonging to at least 2 separate operons are indicated by red circles. Pop-out boxes show the chromosomal context and GC content of two clusters of genes responsive to gyrA mutation, the extent of the responsive gene cluster is marked by the boundaries of the red boxes, genes with significantly altered expression are red (repressed) or yellow (derepressed).