Guanosine tetraphosphate relieves the negative regulation of Salmonella pathogenicity island-2 gene transcription exerted by the AT-rich ssrA discriminator region

Abstract

The repressive activity of ancestral histone-like proteins helps integrate transcription of foreign genes with discrepant AT content into existing regulatory networks. Our investigations indicate that the AT-rich discriminator region located between the -10 promoter element and the transcription start site of the regulatory gene ssrA plays a distinct role in the balanced expression of the Salmonella pathogenicity island-2 (SPI2) type III secretion system. The RNA polymerase-binding protein DksA activates the ssrAB regulon post-transcriptionally, whereas the alarmone guanosine tetraphosphate (ppGpp) relieves the negative regulation imposed by the AT-rich ssrA discriminator region. An increase in the GC-content of the ssrA discriminator region enhances ssrAB transcription and SsrB translation, thus activating the expression of downstream SPI2 genes. A Salmonella strain expressing a GC-rich ssrA discriminator region is attenuated in mice and grows poorly intracellularly. The combined actions of ppGpp and DksA on SPI2 expression enable Salmonella to grow intracellularly, and cause disease in a murine model of infection. Collectively, these findings indicate that (p)ppGpp relieves the negative regulation associated with the AT-rich discriminator region in the promoter of the horizontally-acquired ssrA gene, whereas DksA activates ssrB gene expression post-transcriptionally. The combined effects of (p)ppGpp and DksA on the ssrAB locus facilitate a balanced SPI2 virulence gene transcription that is essential for Salmonella pathogenesis.

Figure 1

Contributions of DksA and (p)ppGpp to the ability of Salmonella to replicate intracellularly. Intracellular replication of wild-type (wt) and mutant Salmonella was quantified over time (A) or 18 h post-infection (B–D) in J774A.1 macrophage-like cells. Where indicated, J774A.1 cells were stimulated with 200 U/ml IFNγ 24 h prior to infection, or treated since the time of infection with 960 μM of the selective iNOS inhibitor L-NIL. Intracellular expression of sifA::luc 8 h after J774A.1 cells were infected with Salmonella (E). Non-significant (ns), *p < 0.05, **p < 0.01, or ***p < 0.001 compared to wild-type controls. The data represent the mean ± S.D. from 3–19 biological replicates.

Figure 2

Codependence of SsrB, ppGpp, and DksA in Salmonella pathogenesis. Competitive indexes of Salmonella strains recovered from spleens of C57BL/6 mice 3 days after infection. Mice were inoculated intraperitoneally with 10² (A) or 10⁵ (B) CFU of the indicated Salmonella strains. No detectable (nd) CFU were isolated for the ΔrelA ΔspoT strain under the experimental conditions used in panel A. Competitive index was determined according to the equation: (strain 1/strain 2)_output/(strain 1/strain 2)_input. Non-significant (ns), **p < 0.01.

Figure 3

The alarmone ppGpp directly activates ssrA gene transcription. The abundance of ssrA (A) and ssrB (B) mRNA was quantified by qRT-PCR in Salmonella grown for 3 h in 8 μM (low) or 10 mM (high) MgCl₂ N9 media. The data, which are from 4 biological replicates and are plotted as the mean ± S.D., represent transcripts levels normalized to the housekeeping gene rpoD. Western blot of SsrB-FLAG in Salmonella grown for 3 h in high or low MgCl₂ N9 medium (C). The blot is representative of 4 biological replicates. The abundance of the DnaK internal control was measured for comparison. The relative amounts of SsrB protein were compared between Salmonella grown in low and high Mg⁺⁺ (SsrB^low/SsrB^high). The amount of ssrA mRNA induced by serine hydroxamate (SHX) was quantified by qRT-PCR in Salmonella grown to log-phase in M9 minimal medium (D). Data represent mean ± S.D. transcripts levels normalized to the housekeeping gene rpoD. Effects of increasing ppGpp concentrations on PssrA in vitro transcription using the pTIM-ssrA template (E). The data, which are plotted as the mean ± S.D., represent absolute copy number determined by qRT-PCR from 6–8 biological replicates. Non-significant (ns), *p < 0.05, **p < 0.01, or ***p < 0.001 as compared to high Mg²⁺ (A,B) or untreated (D,E) controls. An independent, uncropped, blot of panel C can be seen in Fig. S6A.

Figure 4

The AT-rich discriminator region of ssrA serves as a negative regulator of SPI2 expression. A mutant discriminator region of the ssrA promoter was expressed from the native locus in the Salmonella chromosome (A). The abundance of ssrA (B), ssrB (C,E) and ssaG (F) transcripts was measured by qRT-PCR in Salmonella grown for the indicated times in LB broth. Data, which are depicted as mean ± S.D. from 6–8 biological replicates, were normalized to the mRNA levels of the rpoD housekeeping gene. Western blot analysis the SsrB protein in Salmonella grown in LB broth for 5 h (D). **p < 0.01; ***p < 0.001. The abundance of the DnaK chaperone was measured as an internal control. Data are representative of 3 independent experiments. An independent, uncropped blot, of panel D can be seen in Fig. S6B.

Figure 5

Importance of the AT-rich discriminator region of ssrA in Salmonella virulence. (A) The virulence of wild-type (wt) and mutant Salmonella was recorded in a C3H/HeN murine model of oral infection. (B) SsrB protein in Salmonella expressing the pWSK29 or pWSK29-ssrB plasmids. Virulence of pWSK29-ssrB⁺ Salmonella in a C3H/HeN murine model of acute oral infection. Data in A and C are from 9–10 mice per group. Growth of the indicated Salmonella strains in J774 A.1 cells after 20 h of infection (D). The data are shown as the mean ± S.D. of 6 biological replicates **p < 0.01, or ***p < 0.001 compared to wild-type controls.