Cold Shock Proteins Promote Nisin Tolerance in Listeria monocytogenes Through Modulation of Cell Envelope Modification Responses

Abstract

Listeria monocytogenes continues to be a food safety challenge owing to its stress tolerance and virulence traits. Several listeriosis outbreaks have been linked to the consumption of contaminated ready-to-eat food products. Numerous interventions, including nisin application, are presently employed to mitigate against L. monocytogenes risk in food products. In response, L. monocytogenes deploys several defense mechanisms, reducing nisin efficacy, that are not yet fully understood. Cold shock proteins (Csps) are small, highly conserved nucleic acid-binding proteins involved in several gene regulatory processes to mediate various stress responses in bacteria. L. monocytogenes possesses three csp gene paralogs; cspA, cspB, and cspD. Using a panel of single, double, and triple csp gene deletion mutants, the role of Csps in L. monocytogenes nisin tolerance was examined, demonstrating their importance in nisin stress responses of this bacterium. Without csp genes, a L. monocytogenes ΔcspABD mutant displayed severely compromised growth under nisin stress. Characterizing single (ΔcspA, ΔcspB, and ΔcspD) and double (ΔcspBD, ΔcspAD, and ΔcspAB) csp gene deletion mutants revealed a hierarchy (cspD > cspB > cspA) of importance in csp gene contributions toward the L. monocytogenes nisin tolerance phenotype. Individual eliminations of either cspA or cspB improved the nisin stress tolerance phenotype, suggesting that their expression has a curbing effect on the expression of nisin resistance functions through CspD. Gene expression analysis revealed that Csp deficiency altered the expression of DltA, MprF, and penicillin-binding protein-encoding genes. Furthermore, the ΔcspABD mutation induced an overall more electronegative cell surface, enhancing sensitivity to nisin and other cationic antimicrobials as well as the quaternary ammonium compound disinfectant benzalkonium chloride. These observations demonstrate that the molecular functions of Csps regulate systems important for enabling the constitution and maintenance of an optimal composed cell envelope that protects against cell-envelope-targeting stressors, including nisin. Overall, our data show an important contribution of Csps for L. monocytogenes stress protection in food environments where antimicrobial peptides are used. Such knowledge can be harnessed in the development of better L. monocytogenes control strategies. Furthermore, the potential that Csps have in inducing cross-protection must be considered when combining hurdle techniques or using them in a series.

Keywords: Listeria monocytogenes; cell envelope; cold shock protein; nisin; tolerance.

FIGURE 1

Csp deficiency severely impairs nisin stress growth efficiency in L. monocytogenes. (A,B) Optical density measurement-based growth curves of L. monocytogenes EGDe wild-type (WT) and ΔcspABD strains cultivated (37°C and 150 rpm) in (A) normal and (B) nisin (5 ppm)-supplemented BHI. (C–F) Box plots summarizing the relative nisin stress growth parameters (area under the curve, lag phase duration, growth rate, and maximum cell density) of EGDe WT and ΔcspABD strains normalized for each strain to the growth parameters of controls grown in BHI without stress. ^∗P < 0.05, significant differences between the WT and ΔcspABD strains identified using the t-test for comparison of independent samples.

FIGURE 2

Nisin stress survival comparison between L. monocytogenes EGDe wild-type (WT) and ΔcspABD strains. Stationary-phase cultures were subjected to 7.5 ppm nisin stress in BHI at 37°C for 60 min. The survival for each strain expressed as a percentage was measured as the number of colony-forming units determined after nisin stress exposure normalized to the number of unstressed cells present at the beginning of stress exposure. Results showing the mean and SEM of six replicates representing three independent biological experiments are presented. ^∗ P < 0.05, significant differences between the WT and ΔcspABD strains identified using the t-test for comparison of independent samples.

FIGURE 3

Nisin stress growth efficiency comparison between the wild type (WT) and the single (ΔcspA, ΔcspB, and ΔcspD), double (ΔcspBD, ΔcspAD, and ΔcspAB), and triple (ΔcspABD) csp deletion mutants of L. monocytogenes EGDe. The growth parameters area under the curve (A) and maximum growth rate (B) under nisin stress normalized to growth in nisin-free BHI were determined for the csp mutants and expressed relative to the WT strain levels (WT strain levels equivalent to 1.0 are denoted by a broken line on the graph). The results presented are based on three independent experiments that were performed in duplicates. ^∗P < 0.05, significant differences compared to the WT detected using Tukey’s post-hoc test pairwise comparison following one-way ANOVA.

FIGURE 4

Growth under nisin stress variably induces csp mRNA levels. (A–C) Relative csp mRNA amounts were determined for exponentially growing L. monocytogenes EGDe cells cultivated in normal and nisin (5 ppm)-supplemented BHI. The data are presented as scatter plots showing the mean and SD (number of independent biological replicates = 3). 16S rRNA was used for normalization. (D) Bar charts showing the fold induction of different csp mRNAs relative to the abundance under nisin stress (BHI-nisin) with respect to control levels observed without stress (BHI). ^∗P < 0.05, significant differences in mRNA levels and fold induction identified using the (A–C) t-test for independent samples and (D) Tukey’s post-hoc test pairwise comparison following one-way ANOVA.

FIGURE 5

Impact of Csp deficiency on BC and antibiotic susceptibility of L. monocytogenes. Comparison of vancomycin, daptomycin, polymyxin B, ampicillin, and BC stress sensitivities between EGDe wild type (WT) and triple (ΔcspABD) and double (ΔcspBD, ΔcspAD, and ΔcspAB) csp deletion mutants. The minimum inhibitory concentrations (MICs) for vancomycin (A), daptomycin (B,G), polymyxin B (C), and ampicillin (D) were determined using the E-test. (E) WT and ΔcspABD strains were exposed BC (10 ppm) at 25°C for 15 min, and survival expressed as a percentage was measured as the number of colony-forming units determined after BC exposure normalized to the number of unstressed cells present at the beginning of stress exposure. (F) Growth of WT and ΔcspABD strains in BC (1.2 ppm)-supplemented BHI. Results are based on three independent experiments performed in duplicates. ^∗P < 0.05, statistically significant differences between WT and ΔcspABD mutant identified using the t-test for independent samples. Different letters indicate significant differences between WT and the csp mutants daptomycin MICs that were identified using one-way ANOVA and Tukey post-hoc test pairwise comparison of all the strains (p < 0.05).

FIGURE 6

A ΔcspABD mutation reduces the mRNA abundance of nisin response genes and cell surface anionic charge. (A) virR, dltA, and mprF mRNA levels in L. monocytogenes EGDe_ΔcspABD are expressed relative to the wild type (WT) strain (WT level is 1.0 and is denoted by a dotted line). RT-qPCR was performed on mRNA isolated from exponentially growing cells cultivated in nisin (1.5 ppm)-supplemented BHI broth. 16S rRNA was used for mRNA normalization. (B) Bar graph showing the cytochrome C binding level of exponentially growing EGDe_ΔcspABD and WT cells cultivated until OD₆₀₀ 1.0 in BHI at 37°C and 150 rpm. Cytochrome C binding expressed as a percentage was measured as the mean OD₅₃₀ values from replicate samples containing bacteria relative to the mean value of the sample supernatant lacking bacteria. ^∗P < 0.05, statistically significant differences between WT and ΔcspABD mutant identified using the t-test for independent samples.

FIGURE 7

Impact of the ΔcspABD mutation on mRNA abundance of selected penicillin-binding protein (PBP) genes. Bar graphs showing the relative quantities of PBP gene mRNAs in EGDe_ΔcspABD relative to the WT strain whose level is denoted by a dotted line in stationary-phase cells cultivated in BHI for 16 h at 37°C and 150 rpm. RT-qPCR was performed on mRNA samples isolated from cells cultivated for 16 h at 37°C and 150 rpm in BHI. 16S rRNA was used for mRNA normalization. ^∗P < 0.05, significant differences in mRNA levels between the WT and ΔcspABD strains identified using the t-test for comparison of independent samples.