Trans-translation is essential in the human pathogen Legionella pneumophila

Abstract

Trans-translation is a ubiquitous bacterial mechanism for ribosome rescue in the event of translation stalling. Although trans-translation is not essential in several bacterial species, it has been found essential for viability or virulence in a wide range of pathogens. We describe here that trans-translation is essential in the human pathogen Legionella pneumophila, the etiologic agent of Legionnaire's disease (LD), a severe form of nosocomial and community-acquired pneumonia. The ssrA gene coding for tmRNA, the key component of trans-translation, could not be deleted in L. pneumophila. To circumvent this and analyse the consequences of impaired trans-translation, we placed ssrA under the control of a chemical inducer. Phenotypes associated with the inhibition of ssrA expression include growth arrest in rich medium, hampered cell division, and hindered ability to infect eukaryotic cells (amoebae and human macrophages). LD is often associated with failure of antibiotic treatment and death (>10% of clinical cases). Decreasing tmRNA levels led to significantly higher sensitivity to ribosome-targeting antibiotics, including to erythromycin. We also detected a higher sensitivity to the transcription inhibitor rifampicin. Both antibiotics are recommended treatments for LD. Thus, interfering with trans-translation may not only halt the infection, but could also potentiate the recommended therapeutic treatments of LD.

Figure 1. Contruction of ssrA^ind, an L. pneumophila mutant in which ssrA is under repression by the LacI^q repressor.

(A) Schematic diagram of the IPTG-inducible allele ssrA^ind. A gentamicin resistance gene aacC1, the lacIq gene and a modified ssrA promoter with two LacO sites (red boxes) were inserted upstream of the predicted ssrA transcription start site. (B) PCR verification of the construction using primers ssrA-P1 and ssrA-P4. (C) Northern-blot analysis of tmRNA expression in the wild-type strain and the ssrA^ind mutant. Bacteria were cultivated with IPTG in solid medium then cultivated for the indicated duration in IPTG-free liquid medium or with 500 μM IPTG. Ethidium-bromide stained ribosomal RNA were used as loading controls.

Figure 2. tmRNA is essential for growth of L. pneumophila in rich medium.

(A) Serial, ten-fold dilutions of suspensions of the wild-type and the ssrA^ind mutant grown with IPTG were spotted on CYE solid medium containing various concentrations of IPTG. (B) Growth curve of the wild-type and the ssrA^ind mutant in AYE liquid medium containing various concentrations of IPTG. (C) Northern-blot and RT-qPCR analysis of the relative abundance of tmRNA after growth on solid medium. Ethidium-bromide stained ribosomal RNA were used as loading controls. RT-qPCR expression data are average (+/− standard deviation) from three quantifications.

Figure 3. Ribosome release systems can compensate for the loss of tmRNA.

Growth of the ssrA^ind mutant expressing the E. coli arfA gene in liquid AYE medium (A) and on solid CYE medium (B). (C) Northern-blot and RT-qPCR analysis of the relative abundance of tmRNA in the ssrA^ind mutant expressing the E. coli arfA. (D) Growth of the ssrA^ind mutant in the presence of subinhibitory concentrations of puromycin and in the absence of IPTG.

Figure 4. tmRNA-depleted L. pneumophila exhibit filamentation and impaired membrane integrity.

(A) Fluorescence microscopy of the wild-type strain, ssrA^ind mutant and ssrA^ind/pArfA grown for 24 hours (h) in AYE medium and stained with Hoescht 33258 and Propidium Iodide. (B) Electron micrographs of the wild-type strain and ssrA^ind mutant grown for 24 hours in AYE in the absence of IPTG.

Figure 5. Growth of tmRNA-depleted Legionella pneumophila is impaired in eukaryotic hosts.

Intracellular multiplication in A. castellanii (A) and U937 monocyte-derived human macrophages (B) was followed by plating samples on CYE plates containing IPTG and counting colony-forming units. Both charts represent the average of technical triplicates and are representative of three independent experiments. Error bars indicate standard deviation between the triplicates.

Figure 6. Antibiotic sensitivity of L. pneumophila with limited ssrA expression.

The ssrA^ind strain was grown in AYE with either 500 or 4 μM IPTG and antibiotic sensitivity was assayed as described in Materials and Methods. The chart represents the average IC50 obtained in six independent experiments, and error bars indicate standard deviation between experiments. (**) = p < 0.005, (*) = p < 0.01, (n.s.) = non significative, Mann-Wilcoxon U test, one-tailed.