Fur negatively regulates hns and is required for the expression of HilA and virulence in Salmonella enterica serovar Typhimurium

Abstract

Iron is an essential element for the survival of living cells. However, excess iron is toxic, and its uptake is exquisitely regulated by the ferric uptake regulator, Fur. In Salmonella, the Salmonella pathogenicity island 1 (SPI-1) encodes a type three secretion system, which is required for invasion of host epithelial cells in the small intestine. A major activator of SPI-1 is HilA, which is encoded within SPI-1. One known regulator of hilA is Fur. The mechanism of hilA regulation by Fur is unknown. We report here that Fur is required for virulence in Salmonella enterica serovar Typhimurium and that Fur is required for the activation of hilA, as well as of other HilA-dependent genes, invF and sipC. The Fur-dependent regulation of hilA was independent of PhoP, a known repressor of hilA. Instead, the expression of the gene coding for the histone-like protein, hns, was significantly derepressed in the fur mutant. Indeed, the activation of hilA by Fur was dependent on 28 nucleotides located upstream of hns. Moreover, we used chromatin immunoprecipitation to show that Fur bound, in vivo, to the upstream region of hns in a metal-dependent fashion. Finally, deletion of fur in an hns mutant resulted in Fur-independent activation of hilA. In conclusion, Fur activates hilA by repressing the expression of hns.

FIG. 1.

Fur is required for virulence in mice. (A) C3H/HeN (Nramp^+/+) mice were inoculated i.p. with 3,000 CFU of WT 14028s or fur mutant. Mice were observed for 21 days after infection. (B) C57BL/6 (Nramp^−/−) mice were inoculated i.p. with 250 CFU of WT 14028s or fur mutant. Mice were observed for 30 days after infection.

FIG. 2.

Fur is required for the expression of hilA-lacZ. WT and the fur mutant strains carrying a single-chromosomal copy of hilA-lacZ were used. Expression of hilA-lacZ was determined in cells harboring empty vector (pACYC184) and compared to cells harboring pfur-ha. Cultures were grown anaerobically for ∼14 h before β-galactosidase was determined. A paired Student t test was used to determine significance (*, P ≤ 0.05; **, P ≤ 0.01; n = 5). Mean values ± the standard deviations (SD) are shown.

FIG. 3.

Fur regulates hilA-lacZ independent of the response regulator PhoP. β-Galactosidase activity was determined in the ΔphoP, Δfur, or ΔphoP Δfur mutant strains with single-copy hilA-lacZ. Activity was determined as in Fig. 2. A paired Student t test was used to determine significance (*, P ≤ 0.05; **, P ≤ 0.01; n = 4). Means ± the SD are shown.

FIG. 4.

Fur binds upstream of hns in a metal-dependent manner. (A) Expression of hns was significantly elevated in the fur mutant determined by qRT-PCR. A paired Student t test was used to determine significance (*, P ≤ 0.05; n = 3). (B) Fur enrichment upstream of hns in S. Typhimurium. ChIP was performed on chromatin from the fur mutant that contained the pfur-ha plasmid. Resultant DNAs were analyzed by qPCR for antibody-mediated enrichment of Fur-HA at rpoD, dinP, and hns. Amplicon enrichment with anti-HA pAb was normalized to matched signal in nonspecific IgG control immunoprecipitations and expressed as the fold enrichment over that obtained from cells treated with dip (200 μM). Bars indicate means ± the SD of triplicate qPCRs and are representative of four independent ChIP experiments.

FIG. 5.

Fur regulates hilA-lacZ through nucleotides upstream of hns. (A) The truncated constructs used to determine the role of the upstream region of the hns promoter in the Fur-dependent regulation of hilA-lacZ. A filled triangle depicted a predicted Fur binding site. (B) The β-galactosidase activity was determined in the parent strain and in the fur mutant that contained single-copy hilA-lacZ and harbored the empty vector (pACYC184) or in the different truncated constructs listed in panel A. β-Galactosidase activity was determined as in Fig. 2. A paired Student t test was used to determine significance (*, P ≤ 0.05; **, P ≤ 0.01; n = 6). Means ± the SD are shown.

FIG. 6.

Deletion of hns ablates Fur-dependent regulation of hilA-lacZ. β-Galactosidase activity was determined under anaerobic conditions. The data are presented as slopes from the differential plots (see Fig. S3 in the supplemental material). Means ± the SD are shown (*, P ≤ 0.05; ***, P ≤ 0.001; n = 6). The expression of hilA-lacZ in rpoS^low was biphasic (see the text and Fig. S3 for details).

FIG. 7.

Model depicting the role of Fur in the regulation of hilA. (A) Conditions that activate Fur (i.e., the presence of iron) repress the expression of hns (solid arrow), and allow the expression of hilA (blocked dotted arrows) and increased HilA. (B) Mutation in Fur or lack of the cofactor (i.e., Fe²⁺) results in the expression of hns and the H-NS protein (blocked dotted arrows) and the reduced expression of hilA (solid arrows) and HilA. Implicit in this model is the role of ferrous iron transport and iron homeostasis in the activation of hilA/HilA and virulence.