The RNA-binding protein Hfq of Listeria monocytogenes: role in stress tolerance and virulence

Abstract

In gram-negative bacteria, the RNA-binding protein Hfq has emerged as an important regulatory factor in a variety of physiological processes, including stress resistance and virulence. In Escherichia coli, Hfq modulates the stability or the translation of mRNAs and interacts with numerous small regulatory RNAs. Here, we studied the role of Hfq in the stress tolerance and virulence of the gram-positive food-borne human pathogen Listeria monocytogenes. We present evidence that Hfq is involved in the ability of L. monocytogenes to tolerate osmotic and ethanol stress and contributes to long-term survival under amino acid-limiting conditions. However, Hfq is not required for resistance to acid and oxidative stress. Transcription of hfq is induced under various stress conditions, including osmotic and ethanol stress and at the entry into the stationary growth phase, thus supporting the view that Hfq is important for the growth and survival of L. monocytogenes in harsh environments. The stress-inducible transcription of hfq depends on the alternative sigma factor sigmaB, which controls the expression of numerous stress- and virulence-associated genes in L. monocytogenes. Infection studies showed that Hfq contributes to pathogenesis in mice, yet plays no role in the infection of cultured cell lines. This study provides, for the first time, information on the role of Hfq in the stress tolerance and virulence of a gram-positive pathogen.

FIG. 1.

Genetic organization of the hfq locus in L. monocytogenes EGD-e. See the text for a detailed description of the gene products. Putative transcriptional terminators are indicated by lollipop structures. The locations and designations of the primer pairs used for RT-PCR analyses are indicated. Regions that were successfully amplified by RT-PCR are indicated by solid lines, whereas the dashed lines indicate the specific regions that could not be amplified by RT-PCR.

FIG. 2.

Ethanol and osmotic stress tolerance of wild-type and Δhfq strains. (A) Cells grown in BHI medium to the early exponential phase were diluted in BHI medium containing 4.5% ethanol. (B) Cells grown in BHI medium to the stationary phase were diluted in BHI medium containing 4.5% ethanol. (C) Cells grown in BHI medium to the early exponential phase were diluted in BHI medium containing 7% NaCl. (D) Cells grown in BHI medium to the stationary phase were diluted in BHI medium containing 7% NaCl. The arrows indicate the time of dilution. Symbols: •, wild-type strain; ○, Δhfq mutant. The error bars indicate standard deviations based on duplicate experiments.

FIG. 3.

Viability during long-term amino acid starvation of wild-type and Δhfq strains. Cells were grown in IMM containing amino acids at a concentration of 0.002%. After the cells entered the stationary phase, they were incubated for 20 days. Samples were taken at the times indicated to determine viability. The arrow indicates the sampling time for the Δhfq mutant when no survivors were detected. The limit of detection was 25 CFU ml⁻¹. Symbols: •, wild-type strain; ○, Δhfq mutant. The error bars indicate standard deviations based on duplicate experiments.

FIG. 4.

Transcription of hfq is induced by various stress conditions in a σ^B-dependent manner. (A) Sequence of the hfq promoter region. The translation start and stop codons are indicated by boldface type. The transcription start site is indicated by boldface type. Putative −35 and −10 sequences for σ^B are underlined. (B) Primer extension analysis of transcription originating from the hfq promoter under various stress conditions. The analysis was performed by using RNA purified from the wild-type strain or the ΔsigB strain. Cells were grown in BHI medium to an OD₆₀₀ of 0.3. The cell cultures were split and stressed as indicated at the top. Controls without stress treatment were included. Δ, EGDΔsigB; wt, wild type; EtOH, ethanol. (C) Expression of hfq-lacZ transcriptional fusions in response to various stress conditions. The wild-type and ΔsigB strains containing phfq(−109)-lacZ or phfq(−25)-lacZ were grown in BHI medium until the OD₆₀₀ was 0.3. The cell cultures were split and subjected to stresses as indicated at the bottom for 1 h. For stationary-phase cells, cultures were grown in BHI medium until the OD₆₀₀ was 2.7. Cell pellets were harvested and subjected to β-galactosidase assays. Controls without stress treatment were included. The data are the means for three experiments in which the observed variation did not exceed 10%. Stat. phase, stationary phase.

FIG. 5.

Infection studies of wild-type and Δhfq strains: effect of hfq on the intracellular replication of L. monocytogenes in the murine macrophage-like cell line J774A.1 (A) or the intestinal epithelial cell line INT-407 (B). Cell monolayers were infected with approximately 1 bacterium per cell (A) or 20 bacteria per cell (B). After 1 h of incubation, cells were incubated for 1 h in the presence of gentamicin (time zero). The data are the means for two independent experiments, each performed in triplicate. The error bars indicate standard deviations. Symbols: •, wild-type strain; ○, Δhfq strain. (C) Intraperitoneal infection of mice with L. monocytogenes: growth and survival of the wild-type strain (black bars) and the Δhfq strain (grey bars) in the spleens and livers of infected mice on day 3 after injection. The log₁₀ CFU in the organs are averages for five mice. The experiments were repeated twice with similar results.