Fast evolution of antibiotic resistance in E. coli recA mutant strain.

(A) Experimental flow for the single exposures to antibiotics in E. coli strains. Step i: an overnight culture (1 × 109 CFU/mL cells) was diluted 1:50 into 30 mL LB medium supplemented with 50 μg/mL ampicillin and incubated at 37°C with shaking at 250 rpm for 0, 2, 4, 6 and 8 hours; Step ii: after each treatment, the ampicillin containing medium was removed by washing twice in a fresh LB medium; Step iii: the surviving isolates were resuspended in 30 mL fresh LB medium and regrown overnight at 37°C with shaking at 250 rpm; Step iv: cell cultures were plated onto LB agar and incubated for 16 hours at 37°C; Step v: single colonies were inoculated in 30 mL fresh LB medium and cultured at 37°C with shaking at 250 rpm for 4 to 6 hours. (B) MICs of ampicillin were measured against the wild type E coli strain after single exposures to ampicillin. (C) MICs of ampicillin were measured against the ΔrecA strain after single exposures to ampicillin. (D) After the treatment of Step v, cells were continuously cultured in an antibiotic-free medium for seven days. MICs of ampicillin were measured each day. (E) MICs of ampicillin were measured against the ΔrecA strain treated with ampicillin, where the expression of RecA was restored using plasmid-based constitutive expression of recA before the treatment of Step i. (F) MICs of rifampicin were measured after an 8-hour treatment of ampicillin for the wild type and ΔrecA strain. (G) Mutation frequencies using rifampicin as the selector were measured and calculated in the wild type and ΔrecA strain after the single exposure to ampicillin. Each experiment was independently repeated at least six times using parallel replicates, and the data are shown as mean ± SEM. Significant differences among different treatment groups are analyzed by independent t-test, *P < 0.05, **P < 0.01, ***P < 0.001.

Rapid induction of drug resistance associated DNA mutations in the recA mutant strain.

(A) Detection of drug resistance associated DNA mutations in the wild type and ΔrecA strain after the single exposures to ampicillin at 50 µg/mL for 8 hours. (B) MICs of chloramphenicol against the wild type and ΔrecA strain after the single treatment with ampicillin were tested. (C) MICs of kanamycin against the wild type and ΔrecA strain after the single treatment with ampicillin were tested. (D) The wild type, ΔrecA, and ΔrecAacrB mutant resistant isolates were incubated with NMP at different concentrations for 12 hours. Subsequently, MICs were tested in these strains for resistance to ampicillin. Each experiment was independently repeated at least six times, and the data is shown as mean ± SEM. Significant differences among different treatment groups are analyzed by independent t-test, *P < 0.05, **P < 0.01, ***P < 0.001, ns, no significance.

SOS-independent impairment of DNA repair in ΔrecA resistant isolates.

(A) MICs of ampicillin were tested against the wild type strain, ΔrecA strain, and mutants lacking specific genes from the SOS response after single exposures to ampicillin at 50 µg/mL for 8 hours. (B) Filament cell lengths in the wild type (n=253) and the ΔrecA strain (n=216) after single treatments with ampicillin at 50 μg/mL. (C) Multinucleated filaments were observed in the wild type (i) and the ΔrecA (ii) strain after single exposures to ampicillin at 50 μg/mL. Purple: E. coli chromosome; green: DNA Pol I. (D) Relative fold changes of DNA Pol I in the wild type and ΔrecA strain after single treatments with ampicillin at 50 μg/mL. (E) Co-localization between the E. coli chromosome and DNA Pol I in the wild type and ΔrecA strain after the single exposures to ampicillin at 50 μg/mL. Data is shown as mean ± SEM. Significant differences among different treatment groups are analyzed by independent t-test, *P < 0.05, **P < 0.01, ***P < 0.001, ns, no significance.

Overaccumulation of ROS drives the fast evolution of multi-drug resistance in the ΔrecA strain.

(A) Clustered heatmap of relative expression of coding sequences in the wild type and ΔrecA strain with significant fold changes (log2FC > 2 and P value < 0.05). (B) Principal-component analysis (PCA) of normalised read counts for all strains. (C) Venn diagram of differentially expressed genes (log2FC > 2) after treatment with ampicillin at 50 μg/ml for 8 hours in the wild type and ΔrecA strain. (D) The top 10 most differentially expressed genes in the ΔrecA strain after the single treatment with ampicillin are labelled in each plot. Blue dots indicate genes with a significant downregulation compared to the untreated control (log2FC > 2 and P value < 0.05), and yellow dots indicate genes with a significant upregulation compared to the untreated control (log2FC > 2 and P value < 0.05). (E) Levels of transcription of SOS response system-associated genes and gene polA in the wild type and ΔrecA strain after single exposures to ampicillin for 8 hours. (F) Levels of transcription of different antioxidative associated genes in the wild type and ΔrecA strain after single exposures to ampicillin for 8 hours. (G) The addition of 50 mM antioxidative compound GSH prevented the evolution of antibiotic resistance to ampicillin in the ΔrecA strain treated with ampicillin at 50 μg/ml for 8 hours. (H) Survival fraction after a single exposure to ampicillin at 50 μg/ml for 8 hours in the wild type and the ΔrecA strain with or without the addition of GSH at 50 mM. (I) Levels of transcription of proteins involved in the BER DNA repair system in the wild type and ΔrecA strain after single exposures to ampicillin for 8 hours. (J) Whole genome sequencing confirms undetectable DNA mutations in the wild type and ΔrecA strain treated with single exposures to ampicillin with the addition of GSH at 50 mM for 8 hours. (K) Transcription levels of all transcriptional repressors in the wild type and ΔrecA strain after single treatments with ampicillin for 8 hours. Total RNA-seq was performed with three repeats in each group. Each experiment was independently repeated at least six times, and the data are shown as mean ± SEM. Significant differences among different treatment groups are analyzed by independent t-test, *P < 0.05, **P < 0.01, ***P < 0.001, #P < 0.05.

Mechanism of rapid development of multi-drug resistance in the recA mutant E. coli strain.

The deletion of recA results in a decrease in DNA damage repair capability. Additionally, the absence of recA upregulates the transcription levels of the transcriptional repressor H-NS, although the mechanism of this regulation is still unclear. Upregulation of H-NS potentially inhibits the transcription levels of several downstream genes associated with antioxidant functions, leading to an excessive accumulation of ROS. The excessive production of ROS causes an increased occurrence of genetic mutations. Under antibiotic pressure, mutations related to drug resistance are selected, ultimately leading to the emergence of rapid multidrug resistance.

Other mutations detected in the ΔrecA resistant isolates

Strains used in this study

Plasmids used in this study

Primers used in this study

Long-term exposures to ampicillin induced the evolution of resistance in the wild type E. coli strain.

(A) The experimental flow of ALE antibiotic treatment experiment. An overnight culture (0.6 mL; 1 × 109 CFU/mL cells) was diluted 1:50 into 30 mL LB medium supplemented with 50 μg/mL ampicillin and incubated at 37°C with shaking at 250 rpm for 4 or 8 hours. After treatment, the antibiotic-containing medium was removed by washing twice (20 min centrifugation at 1500 g) in fresh LB medium. The surviving isolates were resuspended in 30 mL LB medium and grown overnight at 37°C with shaking at 250 rpm. Ampicillin treatment was applied to the regrown culture and repeated until resistance was established. (B) Changes in the MICs of ampicillin in the wild type strain after 21 days of treatment with ampicillin at 50 μg/ml for 4 hours each day. (C) Changes of MICs of ampicillin in the wild type strain after 21 days of treatment with ampicillin at 50 μg/ml for 8 hours each day. MICs were measured after each daily treatment. Each experiment was independently repeated at least twice, and the data are shown as mean ± SEM.

Single exposures to ampicillin induced the evolution of resistance in the ΔrecACGSCstrain (JW2669-1).

(A) MICs of ampicillin were measured against the wild typeCGSC E coli strain after single exposures to ampicillin. (B) MICs of ampicillin were measured against the ΔrecACGSC strain after single exposures to ampicillin. Each experiment was independently repeated at least six times using parallel replicates, and the data are shown as mean ± SEM. Significant differences among different treatment groups are analyzed by independent t-test, *P < 0.05, **P < 0.01, ***P < 0.001.

Single exposures to other β-lactam antibiotics induced the evolution of resistance in the ΔrecA strain.

(A) MICs of penicillin G in the wild type, ΔrecA, and complemented ΔrecA strain, where the expression of RecA was restored, after single exposures to penicillin G at 1 mg/mL for 8 hours. (B) MICs of carbenicillin in the wild type, ΔrecA, and complemented ΔrecA strain, where the expression of RecA was restored, after single exposures to carbenicillin at 200 µg/mL for 8 hours. Each experiment was independently repeated at least six, and the data are shown as mean ± SEM.

The activity of β-lactamase was increased in the ΔrecA culture supernatants.

The ΔrecA strain was treated with ampicillin at 50 µg/ml for 8 hours, and surviving isolates harbouring the ampC mutations were selected. The level of β-lactamase in cell culture supernatants was determined by the absorbance at OD490. The levels of β-lactamase in the wild type or ΔrecA strain culture supernatants without exposure to ampicillin were tested as a control. Each experiment was independently repeated six times. Each experiment was independently repeated at least six, and the data are shown as mean ± SEM.

Transcriptional responses of the wild type and ΔrecA strain after single treatments with ampicillin for 8 hours.

GO analysis was performed following the GOseq approach, and different genes in the wild type (A, left) and the ΔrecA strain (B, left) were plotted. KEGG pathway enrichment was assigned according to the KEGG database, and different genes in the wild type (C, left) and the ΔrecA strain (D, left) were plotted. The top 20 enrichment pathways are listed in the GO and KEGG enrichment analysis (A-D, right). Total RNA-seq was performed with three repeats in each group.