The stringent response of Staphylococcus aureus and its impact on survival after phagocytosis through the induction of intracellular PSMs expression

Abstract

The stringent response is initiated by rapid (p)ppGpp synthesis, which leads to a profound reprogramming of gene expression in most bacteria. The stringent phenotype seems to be species specific and may be mediated by fundamentally different molecular mechanisms. In Staphylococcus aureus, (p)ppGpp synthesis upon amino acid deprivation is achieved through the synthase domain of the bifunctional enzyme RSH (RelA/SpoT homolog). In several firmicutes, a direct link between stringent response and the CodY regulon was proposed. Wild-type strain HG001, rsh(Syn), codY and rsh(Syn), codY double mutants were analyzed by transcriptome analysis to delineate different consequences of RSH-dependent (p)ppGpp synthesis after induction of the stringent response by amino-acid deprivation. Under these conditions genes coding for major components of the protein synthesis machinery and nucleotide metabolism were down-regulated only in rsh positive strains. Genes which became activated upon (p)ppGpp induction are mostly regulated indirectly via de-repression of the GTP-responsive repressor CodY. Only seven genes, including those coding for the cytotoxic phenol-soluble modulins (PSMs), were found to be up-regulated via RSH independently of CodY. qtRT-PCR analyses of hallmark genes of the stringent response indicate that an RSH activating stringent condition is induced after uptake of S. aureus in human polymorphonuclear neutrophils (PMNs). The RSH activity in turn is crucial for intracellular expression of psms. Accordingly, rsh(Syn) and rsh(Syn), codY mutants were less able to survive after phagocytosis similar to psm mutants. Intraphagosomal induction of psmα1-4 and/or psmβ1,2 could complement the survival of the rsh(Syn) mutant. Thus, an active RSH synthase is required for intracellular psm expression which contributes to survival after phagocytosis.

Figure 1. The experimental setup and the interaction of the stringent response with CodY.

(A) Growth inhibition in medium lacking leucine and valine (−leu/val). S. aureus wild-type strain HG001 (square), rsh_Syn mutant (triangle), codY mutant (circle) and rsh_Syn, codY double mutant (diamond) were grown in complete CDM to an optical density of OD₆₀₀ = 0.5 (open arrow), filtered, washed twice and re-suspended in an equal volume of CDM medium without leucine/valine (open symbols) or CDM with leucine (wild-type only, filled square). Aliquots for microarray analysis were harvested after 30 minutes (filled arrow). (B) The interaction of the two regulons CodY and stringent response under leucine/valine starvation (−leu/val) and after serine starvation induced by serine hydroxamate (+SHX) and strain HG001 (WT)), the corresponding rsh_Syn mutant, codY mutant, and the rsh_{Sy n}, codY double mutant were grown in CDM to the exponential growth phase (OD₆₀₀ = 0.5) followed by further incubation in medium containing 1.5 mg/ml SHX respectively in medium without leucine/valine for 30 minutes. RNA was hybridized with digoxigenin-labelled PCR fragments. The 16S rRNA detected in the ethidium bromide-stained gels is indicated as loading control in the lower lane.

Figure 2. Transcriptome analysis of (p)ppGpp dependent genes during stringent response.

Transcriptional changes of S. aureus in response to leucine/valine starvation (−leu/val) and the overlap of the two regulons CodY and stringent response. (A) Venn diagrams showing genes which are down- (blue) and up-regulated (red) during RSH mediated stringent response. A contribution of CodY to the regulation was determined by analyzing the transcriptional changes in a codY-positive background (WT strains vs. rsh_Syn mutant) indicated as codY+ in the figure and in a codY-negative background (codY mutant vs. rsh_Syn, codY double mutant) indicated as codY−. (B) Functional classes of genes which are down- (blue) and up-regulated (red) during RSH mediated stringent response. (C) Heat maps of gene expression ratios for indicated functional classes prominently altered in their expression. The ratios were analyzed in a codY-positive (codY+) and a codY- negative (codY−) background as described for the Venn diagrams. ** saturated signal *** genes with annotated CodY binding motifs .

Figure 3. (p)ppGpp dependent psm expression during stringent response.

(A) (p)ppGpp dependent activation of target genes in two different S. aureus strains. Strain HG001 (WT) and strain Newman (WT), the corresponding rsh_Syn mutants, codY mutants, rsh_{sy n}, codY double mutants, the complemented rsh_Syn mutants (compl) and the agr mutants were grown in CDM to the exponential growth phase (OD₆₀₀ = 0.5) followed by further incubation in medium without (−) leu/val for 30 minutes. (B) Wild-type strain HG001 and its isogenic rsh_Syn mutant were incubated with (+) or without (−) leu/val for 30 minutes. RNA was hybridized with digoxigenin-labelled PCR fragments. The 16S rRNA detected in the ethidium bromide-stained gels is indicated as loading control in the lower lane.

Figure 4. Quantitative analysis of the intracellular ATP concentrations.

The intracellular ATP pools of strain HG001, the corresponding rsh_Syn mutant, the rsh_Syn, codY double mutant and the complemented rsh_Syn mutant (compl) were determined by luciferase activity. The bacteria were grown in CDM to exponential growth phase (OD₆₀₀ = 0.5) followed by further incubation in medium with (control) or without (−) leu/val for 30 minutes. The levels of significance to the corresponding control (with leu/val) were determined by the two-tailed Student t test (p<0.05).

Figure 5. Phagocytosis killing assays and a PMN lysis assay with different S. aureus strains.

(A) and (B) Survival of wild-type S. aureus strain HG001 and strain Newman (WT), and their isogenic rsh_Syn mutants, complementation strains (compl), psmα, psmβ double mutants (HG001), rsh_Syn, psmα, psmβ triple mutants (HG001), agr mutant (HG001) and rsh_Syn, agr double mutant (HG001) were determined after phagosomal uptake by counting colony-forming units after 60 min incubation. Bacterial cells used for the experiment were harvested at similar points of growth at an OD₆₀₀ of 1.0. The levels of significance were determined by the two-tailed Student t test (p<0.05). (C) Complementation of survival of the rsh_Syn mutant with tetracycline inducible psmα and psmβ. Therefore membrane permeable anhydrotetracycline (ATc) 0.1 µg/ml was used to induce psm expression after phagosomal uptake of the bacteria. The levels of significance were determined by the two-tailed Student t test (p<0.05). (D) Lysis of PMNs, as measured by release of lactate dehydrogenase (LDH) activity. PMN lysis was determined after a phagocytotic interaction of 60 min with S. aureus strain HG001 wild-type, corresponding rsh_Syn mutant, agr mutant and rsh_Syn, agr double mutant. As a positive control 2% TritonX were used. The levels of significance were determined by the two-tailed Student t test (p<0.05).

Figure 6. Quantitative RT-PCRs of phagocytosed bacteria.

Transcript analyses of rpsB, infB (category I), ilvC (category II), psmα1-4, psmβ1,2 (category III) and RNAIII genes after phagocytosis of the S. aureus strain HG001 and the corresponding rsh_Syn mutant, rsh_Syn, codY double mutant and the complemented rsh_Syn mutant (compl). The transcripts were relatively quantified in reference of gyrB and are shown in a log₁₀ scale relative to the wild-type strain (HG001). RNA was isolated after a 60 and 90-minute interaction of S. aureus with PMNs. Values from four separate experiments were used to calculate the mean expression. The levels of significance compared to the wild-type strain were determined by the two-tailed Student t test (p<0.05).

Figure 7. Overview of the presumed impact of the stringent response on survival after phagocytosis.

The induction of the stringent response (pppGpp) in phagocytosed bacteria leads to increased psms expression. The PSMs in turn mediate a pronounced survival after phagocytosis, most likely through phagosomal escape or intracellular lysis of the phagocytes.