Impact of subinhibitory concentrations of metronidazole on proteome of Clostridioides difficile strains with different levels of susceptibility

Abstract

Clostridioides difficile is responsible for various intestinal symptoms from mild diarrhea to severe pseudomembranous colitis and is the primary cause of antibiotic-associated diarrhea in adults. Metronidazole was the first-line treatment for mild to moderate C. difficile infections for 30 years. However, clinical failure and recurrence rates of metronidazole is superior to oral vancomycin and metronidazole is now recommended only as an alternative to vancomycin or fidaxomicin, for an initial non-severe infection. The mechanisms of treatment failure and infection recurrence remain unclear. Given the poor fecal concentrations of metronidazole, the bacteria may be exposed to subinhibitory concentrations of metronidazole and develop adaptation strategy, which is likely to be the origin of an increase in treatment failures. In this study, a proteomic approach was used to analyze changes in the proteome of two strains with different levels of susceptibility to metronidazole in the presence of subinhibitory concentrations of this antibiotic. The two strains were grown to stationary phase: CD17-146, a clinical C. difficile isolate with reduced susceptibility to metronidazole, and VPI 10463, a metronidazole susceptible strain. Our study revealed that, whatever the strain, subinhibitory concentrations of metronidazole modified the amount of proteins involved in protein biosynthesis, glycolysis, and protection against stress induced by metronidazole, as well as in DNA repair. Several proteins involved in stress response are known to be synthesized under the control of Sigma factor B, which suggests a close link between Sigma factor B and metronidazole. Interestingly, impact of metronidazole on protein production for VPI 10463 strain differed from CD17-146 strain, for which the amount of two proteins involved in biofilm formation of CD17-146 were modified by metronidazole.

Fig 1

Effect of different concentrations of MTZ on growth of C. difficile VPI 10463 (A, B) and CD17-146 (C, D) strains. The bacteria were grown without MTZ (blue) or exposed to MIC/4 (orange) or MIC/2 (grey) of MTZ in BHISG medium. MIC for VPI 10463 and CD17-146 were 0.5 and 1 mg.mL^-1, respectively. The growth rates were determined by measuring OD_600nm (A, C) and numeration of bacterial cells on BHI agar supplemented with 3% defibrinated horse blood (B, D). We observed that subinhibitory concentrations of MTZ had a direct inhibitory effect on the bacterial growth for both strains and CD17-146 grew more slowly than VPI 10463 in the presence of MTZ.

Fig 2. Hierarchical clustering analysis based on protein expressions of VPI 10463 and CD17-146 strains after culture in BHISG without MTZ, at MIC/4 and at MIC/2.

Each column corresponds to a single sample: VPI-0 (VPI 10463, without MTZ), VPI-MIC/4 (VPI 10463, with MTZ at MIC/4), VPI-MIC/2 (VPI 10463, with MTZ at MIC/2), 146–0 (CD17-146, without MTZ), 146-MIC/4 (CD17-146, with MTZ at MIC/4), 146-MIC/2 (CD17-146, with MTZ at MIC/2). Each row corresponds to a single protein, blue indicating a reduced expression and red an increased expression. The proteome profiles of the strains CD17-146 and VPI 10463 were close without MTZ but reached opposite directions with MTZ at MIC/2, suggesting two types of MTZ stress response as stress gets more important.

Fig 3. Differences in expressions of proteins involved in biofilm formation between CD17-146 and VPI 10463 strains after culture in BHISG under exposure to MTZ.

The protein abundance displayed by the number of spectra that has been identified to peptides belonging to a defined protein. VPI-0 (VPI 10463, without MTZ), VPI-MIC/4 (VPI 10463, with MTZ at MIC/4), VPI-MIC/2 (VPI 10463, with MTZ at MIC/2), 146–0 (CD17-146, without MTZ), 146-MIC/4 (CD17-146, with MTZ at MIC/4), 146-MIC/2 (CD17-146, with MTZ at MIC/2). In the presence of MTZ at MIC/2, CD17-146 decreased Cwp84 production and increased an aminotransferase belonged to MocR family 2. These modifications may implicate in enhanced biofilm production.

Fig 4. Differential production in electron transport and redox-active proteins of CD17-146 and VPI 10463 after culture in BHISG without MTZ and with MTZ.

The protein abundance displayed by the number of spectra that has been identified to peptides belonging to a defined protein.VPI-0 (VPI 10463, without MTZ), VPI-MIC/4 (VPI 10463, with MTZ at MIC/4), VPI-MIC/2 (VPI 10463, with MTZ at MIC/2), 146–0 (CD17-146, without MTZ), 146-MIC/4 (CD17-146, with MTZ at MIC/4), 146-MIC/2 (CD17-146, with MTZ at MIC/2). In the presence of MTZ at MIC/2, the amount of Pfor, which activate MTZ in bacteria, decreased in both strains. To establish alternative route for pyruvate metabolism, the production of two pyruvate-formate-lyase increased.

Fig 5. Modification in the amount of proteins involved in glycolysis for CD17-146 and VPI 10463 after culture in BHISG with subinhibitory concentrations of MTZ.

A. The protein abundance displayed by the number of spectra that has been identified to peptides belonging to a defined protein.VPI-0 (VPI 10463, without MTZ), VPI-MIC/4 (VPI 10463, with MTZ at MIC/4), VPI-MIC/2 (VPI 10463, with MTZ at MIC/2), 146–0 (CD17-146, without MTZ), 146-MIC/4 (CD17-146, with MTZ at MIC/4), 146-MIC/2 (CD17-146, with MTZ at MIC/2). The amount of several enzymes involved in glycolysis decreased in the presence of MTZ, except for some enzymes involved in substrate phosphorylation which were overproduced. B. Modification in the amount of proteins involved in fermentation for CD17-146 and VPI 10463 after culture in BHISG with subinhibitory concentrations of MTZ. The protein abundance displayed by the number of spectra that has been identified to peptides belonging to a defined protein.VPI-0 (VPI 10463, without MTZ), VPI-MIC/4 (VPI 10463, with MTZ at MIC/4), VPI-MIC/2 (VPI 10463, with MTZ at MIC/2), 146–0 (CD17-146, without MTZ), 146-MIC/4 (CD17-146, with MTZ at MIC/4), 146-MIC/2 (CD17-146, with MTZ at MIC/2). Differences in the production of proteins involved in fermentation pathways were also observed in the presence of MTZ: increase of AdhE and diminution of Bcd and Hbd.

Fig 6. Schematic overview of the glycolysis showing the changes on glycolysis of CD17-146 and VPI 10463 after culture in BHISG with subinhibitory concentrations of MTZ.

The increase or decrease in production of proteins were marked by dot green or red, respectively.

Fig 7. Differential expressions of SigB target genes involved in oxidative/nitrosative stress and DNA repair in CD17-146 and VPI 10463 strains after culture in BHI-SG with subinhibitory concentrations of MTZ.

Fold change in expression of genes of CD17-146 (A) and VPI 10463 (B) in culture with MTZ at MIC/4 (orange) and MIC/2 (grey) compared to culture without MTZ (blue). Ruberythrine in this chart represents two copies of revRbr because the primers for q-RT-PCR cannot distinguish them. Error bars represent standard deviation. Significantly different (p<0.05) ratios are indicated by asterisks (Man-Whitney test). Data are representative of three independent experiments, each performed in duplicate. We found that norV, pflB, trxB1, gluD and uvrAB genes were up-regulated around 2-3-fold for both strains in the presence of MTZ.