Mechanistic divergence between SOS response activation and antibiotic-induced plasmid conjugation in Escherichia coli

Abstract

The SOS response is a critical DNA damage repair mechanism in bacteria, designed to counteract genotoxic stress and ensure survival. This system can be activated by different classes of antimicrobial agents, each inducing the SOS response through different mechanisms. Moreover, it has been observed that certain antibiotics can enhance conjugative plasmid transfer frequencies. However, while previous studies have suggested that the SOS response contributes to horizontal transfer of certain genes, its role in plasmid conjugation remains unclear. In this study, we investigated the relationship between the SOS response and conjugation of IncI1 and IncFII plasmids harboring various bla_CTX-M resistance genes. Results showed that cefotaxime and mitomycin C induced both the SOS response and conjugation, while ciprofloxacin induced the SOS response without affecting conjugation frequencies. Further analysis of SOS mutants, ranging from constitutively inactive to hyper-induced states, revealed no correlation between SOS levels and conjugation frequencies, despite upregulation of tra gene expression in a SOS hyper-induced strain. Proteomic analysis revealed that cefotaxime-induced conjugation was associated with increased transfer and pilus protein expression. In contrast, the SOS hyper-induced strain displayed limited upregulation of plasmid-encoded proteins, suggesting post-transcriptional regulation. Additionally, putative LexA binding sites on the IncI1 plasmid revealed potential SOS-mediated regulation of plasmid genes, highlighting the interaction between the SOS response and plasmid, although it did not significantly affect conjugation.IMPORTANCEPlasmids play a critical role in the dissemination of antibiotic resistance through conjugation. Recent research suggests that the use of antibiotics not only selects for already resistant variants but further increases the rate of plasmid-encoded conjugative transmission by increasing expression of the conjugative system. At the same time, these antibiotics are known to induce the stress-related SOS response in bacteria. To be able to counteract an antibiotic-induced increase in conjugative transfer of resistance plasmid, there is a need for a fundamental understanding of the regulation of transmission, including whether this happens through activation of the SOS response. In this research, we show that antibiotic-induced conjugation and induction of the SOS response happen through different mechanisms, and thus that future strategies to control the spread of antibiotics cannot interfere with the SOS response as its target.

Keywords: Escherichia coli; IncI1; SOS response; antibiotics; conjugation.

Fig 1

Investigation of SOS response, tra-gene expression, and conjugation when WT is exposed to sub-inhibitory concentrations of antimicrobial agents (½-MIC of CTX [128 µg/mL], ½-MIC of MMC [2 µg/mL], combination of 1/8-MIC of CTX [32 µg/mL] and ¼-MIC of MMC [1 µg/mL], and ½-MIC of CIP [0.004 µg/mL]). (A) Antimicrobials induce bacterial SOS response to varying degrees. (B) CTX and MMC, as well as their combined treatment, lead to the upregulation of tra-genes expression, whereas CIP treatment has no effect on tra expression. (C) CTX and MMC up-regulated the conjugation frequency of pTF2, whereas CTX+MMC and CIP treatment did not affect conjugation. Data are presented as fold change relative to WT without antibiotic treatment (dotted line), with the values presented as average plus standard deviation. Each dot in the visual representation corresponds to an individual biological replicate. The fold changes were compared to WT without antibiotic treatment to analyze the differences: *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, and ****P ≤ 0.0001.

Fig 2

(A) SOS response gene expression levels and (B) tra-gene expression levels in SOS variants relative to WT. Data are presented as fold change relative to WT (dotted line), and the values are presented as average plus standard deviation. (C) Conjugation frequency of SOS mutants. Each dot in the visual representation corresponds to individual biological replicates. The stars indicate statistical significance between the mutants and the WT at different levels: *P ≤ 0.05, ***P ≤ 0.001, and ****P ≤ 0.0001.

Fig 3

Proteomic analysis of SOS* and the WT ± ½-MIC of CTX or CIP. The data are presented as fold change relative to WT without antibiotic treatment. (A) The violin plot demonstrates a relative abundance of identified proteins encoded by plasmid pTF2. Each dot represents the abundance ratio of a specific identified protein. The relative protein expression levels were compared to WT without antibiotic treatment to analyze the differences: **P ≤ 0.01 and ***P ≤ 0.001. (B) The heatmap of the relative abundance of conjugation-related proteins. (C) Relative protein ratio of the four pTF2-encoded proteins up-regulated in SOS*.

Fig 4

(A) Deletion of pTF2-encoded upregulated genes in SOS* did not affect SOS-response levels (sulA and recN) when compared to SOS*. (B) tra gene expression (traF and traM) was reduced in SOS*-derived mutants. (C) Conjugation frequency of SOS* and SOS*-derived mutants revealed no significant changes. The values are presented as average plus standard deviation, and the fold changes were compared to that of SOS* to analyze the differences: *P ≤ 0.05, **P ≤ 0.01.