Threonine Phosphorylation Fine-Tunes the Regulatory Activity of Histone-Like Nucleoid Structuring Protein in Salmonella Transcription

Abstract

Histone-like nucleoid structuring protein (H-NS) in enterobacteria plays an important role in facilitating chromosome organization and functions as a crucial transcriptional regulator for global gene regulation. Here, we presented an observation that H-NS of Salmonella enterica serovar Typhimurium could undergo protein phosphorylation at threonine 13 residue (T13). Analysis of the H-NS wild-type protein and its T13E phosphomimetic substitute suggested that T13 phosphorylation lead to alterations of H-NS structure, thus reducing its dimerization to weaken its DNA binding affinity. Proteomic analysis revealed that H-NS phosphorylation exerts regulatory effects on a wide range of genetic loci including the PhoP/PhoQ-regulated genes. In this study, we investigated an effect of T13 phosphorylation of H-NS that rendered transcription upregulation of the PhoP/PhoQ-activated genes. A lower promoter binding of the T13 phosphorylated H-NS protein was correlated with a stronger interaction of the PhoP protein, i.e., a transcription activator and also a competitor of H-NS, to the PhoP/PhoQ-dependent promoters. Unlike depletion of H-NS which dramatically activated the PhoP/PhoQ-dependent transcription even in a PhoP/PhoQ-repressing condition, mimicking of H-NS phosphorylation caused a moderate upregulation. Wild-type H-NS protein produced heterogeneously could rescue the phenotype of T13E mutant and fully restored the PhoP/PhoQ-dependent transcription enhanced by T13 phosphorylation of H-NS to wild-type levels. Therefore, our findings uncover a strategy in S. typhimurium to fine-tune the regulatory activity of H-NS through specific protein phosphorylation and highlight a regulatory mechanism for the PhoP/PhoQ-dependent transcription via this post-translational modification.

Keywords: bacterial signal transduction; histone-like nucleoid structuring protein (H-NS); post-translational modification; protein threonine phosphorylation; transcriptional regulation.

FIGURE 1

Characterization of T13 phosphorylation of H-NS in Salmonella enterica serovar Typhimurium. (A) The MS/MS spectrum of ¹³(pT)LRAQARECTLETLEEMLEK³². The phosphopeptide from S. Typhimurium wild-type strain (ATCC 14028s) was shown as an ion m/z 849.40485 with phosphorylated 13th threonine (pT). The underlined residues were modified from sample preparation: deamidated Q17, carbamidomethyl C21, oxidated M29, and dimethylated K32. Phosphorylation specific to T13, but not to T22 or T25, was specified by y₁₁⁺ and y₁₂⁺. (B) Immunoprecipitation assay of T13 phosphorylation of the H-NS_HA-WT and H-NS_HA-T13E proteins from the hns-HA wild-type strain (YS17483) and the isogenic hns-HA T13E mutant (YS17485), respectively. Bound represents the H-NS_HA protein that bound to the phosphothreonine-specific antibody. Total lysate represents the H-NS_HA protein present in the lysates before immunoprecipitation. The result was from one of the two independent experiments. Protein amount was quantified by ImageJ. (C) Growth of wild-type strain (YS17483) and T13E mutant (YS17485) in LB medium. Results were from two independent experiments. The OD_600nm value was measured after the original cultures were diluted with the medium used.

FIGURE 2

T13 phosphorylation partially releases the inhibitory effect of H-NS to the PhoP/PhoQ-activated genes. (A) β-Galactosidase activity from a pcgL-lacZ transcriptional fusion was determined in S. Typhimurium hns-HA wild-type strain (WT, YS17529), T13E mutant (YS17531), and T13V mutant (YS17530) grown in N medium with low (0.01 mM) and high (10 mM) Mg²⁺ for 4 h, respectively. ^∗∗∗P < 0.001, vs. WT, t-test. (B) β-Galactosidase activity from a pagC-lacZ transcriptional fusion was determined in S. Typhimurium hns-HA wild-type strain (WT, YS17519), T13E mutant (YS17521), and T13V mutant (YS17520) grown in N medium with low and high Mg²⁺ for 4 h, respectively. ^∗∗P < 0.01, vs. WT, t-test. (C) β-Galactosidase activity from an STM3595-lacZ transcriptional fusion was determined in S. Typhimurium hns-HA wild-type strain (WT, YS17677), T13E mutant (YS17679), and T13V mutant (YS17678) grown in N medium with low and high Mg²⁺ for 4 h, respectively. ^∗∗P < 0.01, vs. WT, t-test. (D) Growth of wild-type strain (YS17483) and T13E mutant (YS17485) in N medium with low and high Mg²⁺, respectively. Results were from one of two independent experiments. The OD_600nm value was measured after the original cultures were diluted with the medium used. (E) Survival rates of S. Typhimurium wild-type strain (WT, YS17483), T13E mutant (YS17485), and T13V mutant (YS17484) after challenged by antimicrobial peptide magainin 2 (left panel) and human defensin HNP-1 (right panel) with indicated concentrations. Data correspond to three independent assays conducted in duplicate, and all values were mean ± standard deviation. ^∗∗P < 0.01, ^∗P < 0.05, vs. the strain untreated with tested peptide, t-test.

FIGURE 3

T13 phosphorylation lowers the affinity of H-NS to the PhoP/PhoQ-activated genes. (A) Immunoblot analysis of the total levels of H-NS_HA protein in the hns-HA strain (WT, YS17483) and the isogenic T13E mutant (YS17485) and the total level of PhoP_HA protein in the phoP-HA strain (WT, YS11591) and the isogenic T13E mutant (YS18776). Bacterial cells were grown in low and high Mg²⁺ for 4 h, respectively. Mg²⁺ transporter protein CorA was used as negative control. Protein amount was quantified by ImageJ and used to calculate ratios of H-NS_HA-T13E vs. H-NS_HA-WT with the formula: [H-NS_HA-T13E]/[H-NS_HA-WT] (middle panel); and ratios of the PhoP_HA and CorA_FLAG proteins (Pr) in wild-type strain (WT) and T13E mutant grown in low and high Mg²⁺ for 4 h, respectively, with the formula: [Pr]_highMg2+/[Pr]_lowMg2+ (right panel). (B) Chromatin immunoprecipitation-qRT-PCR assay for comparison of binding between H-NS_HA-WT in the wild-type strain (WT, YS17483) and H-NS_HA-T13E in T13E mutant (YS17485) to the pcgL, pagC and STM3595 promoters in low and high Mg²⁺. The results were from three independent experiments and quantified by qRT-PCR (see section “Materials and Methods”). Input represents the total amount of target promoter DNA for the assay. IP presents the amount of target promoter binding to the H-NS_HA proteins. All WT results were taken to 1.0, respectively. ^∗∗P < 0.01, vs. WT, t-test. (C) Chromatin immunoprecipitation-qRT-PCR assay for comparison of PhoP_HA protein to the pcgL, pagC and STM3595 promoters between wild-type strain (WT, YS11591) and T13E mutant (YS18776) in low and high Mg²⁺. The results were from three independent experiments and quantified by qRT-PCR. Input represents the total amount of target promoter DNA for the assay. IP presents the amount of target promoter binding to the PhoP_HA protein. All Input results from wild-type strain (WT), and also the IP result from wild-type strain (WT) in low Mg²⁺ were taken to 1.0, respectively. ^∗∗∗P < 0.001, ^∗∗P < 0.01, ^∗P < 0.05, vs. WT, t-test. (D) β-Galactosidase activity from pcgL-lacZY, pagC-lacZY, and STM3595-lacZY transcriptional fusions was determined in T13E mutants (YS17531, YS17521, and YS17679) harboring plasmid pUHE21-hns-HA_WT grown in low Mg²⁺ with or without 0.1 mM IPTG. The results were from three independent experiments. ^∗∗∗P < 0.001, ^∗P < 0.05, vs. 0 mM IPTG, t-test. (E) β-Galactosidase activity from pcgL-lacZY, pagC-lacZY, and STM3595-lacZY transcriptional fusions was determined in T13E mutants (YS17531, YS17521, and YS17679) harboring plasmid pUHE21-hns-HA_WT grown in high Mg²⁺ with or without 0.1 mM IPTG. The results were from three independent experiments. ^∗∗∗P < 0.001, ^∗∗P < 0.01, vs. 0 mM IPTG, t-test.

FIGURE 4

T13 phosphorylation reduces H-NS dimerization in vivo by inducing a conformational change. (A) In vivo cross-linking and immunoblot to monitor dimerization of H-NS_HA proteins in the wild-type strain (YS17483) and T13E mutant (YS17485). H-NS bands were quantified with ImageJ, and the percentage of H-NS dimer in each strain was calculated using the formula: Dimer/(Monomer + Dimer). (B) Partial trypsin digestion of H-NS_HA proteins. Crude extracts of the wild-type strain and T13E mutant were digested with trypsin (30 μg/ml) for 0, 10, 20, 30 min, respectively. Full-length H-NS_HA protein from each aliquot was analyzed and quantified by immunoblot. The results were from three independent experiments (^∗∗P < 0.01; ^∗∗∗P < 0.001).