Multiple clades of regulators contribute to bacterial phosphate homeostasis and pathogenesis

Abstract

Phosphate is both essential for life and toxic, necessitating the tight regulation of its acquisition. Based on Escherichia coli, most bacteria are thought to use a single accessory protein that monitors import to regulate phosphate homeostasis. This work reveals that most bacteria possess multiple distinct families of accessory regulators with each family regulating homeostasis in conjunction with a unique importer family. The antibiotic-resistant pathogen Staphylococcus aureus can obtain phosphate from divergent environments and possesses accessory-transporter pairs from all three identified groups. Investigations with S. aureus revealed that all three accessory proteins can regulate phosphate homeostasis, but that there is a hierarchy, which is dictated by the environment. Multiple accessory regulators are independently necessary for S. aureus to cause infection. Thus, microbes possess not one, but multiple distinct groups of accessory regulatory proteins and this diversity enables them to control phosphate homeostasis across environments, including those encountered during infection.

Keywords: PhoPR; PhoU; Staphylococcus aureus; phosphate homeostasis; regulation; two-component system; virulence.

Figure 1.. Unrooted phylogram shows the inferred evolutionary relationships between PhoU-and PitR-related proteins.

PhoU- and PitR-related sequences were obtained from the InterPro database (Materials and Methods). After filtering by sequence identity at 70% and taxonomy at the genus level, 200 and 100 protein sequences were randomly chosen among the filtered representative sequences from each genus for both groups. Protein sequences were aligned using MAFFT v7 and a phylogenetic tree was reconstructed using RAxML v8.2.12 with LG+G substitution model. Branch lengths in the unrooted tree are proportional to the sequence divergence. S. aureus PhoU and PitR sequences are labeled.

Figure 2.. The PitR and NptA clades of accessory regulators control activation of phosphate-responsive two-component signal transduction systems.

(A & B) S. aureus wild-type and the indicated strains containing (A) P_pstS-yfp or (B) P_nptA-yfp reporters were grown in PFM9, pH 7.4, supplemented with 5 mM, 500 μM, or 50 μM P_i. The expression of pstSCAB or nptA was assessed by measuring fluorescence at T = 8 h. * = P ≤ 0.05 relative to wild-type at the same P_i concentration via one-way ANOVA with Tukey’s posttest. # = P ≤ 0.05 relative to ΔpitR at the same P_i concentration via one-way ANOVA with Tukey’s posttest. & = P ≤ 0.05 relative to ΔpitRΔphoUnptA_ΔphoU via one-way ANOVA with Tukey’s posttest. n ≥ 3. Error bars = SEM. (C, D, E, F) S. aureus wild-type and the indicated strains containing the P_pstS-yfp reporter and an empty vector (P_empty), (C) P_pitR, (D, E) P_nptA, (F) P_lgtphoU plasmids were grown in PFM9, pH 7.4, supplemented with 5 mM (C, D, F) or 500 μM (E) P_i. The expression of pstSCAB or nptA was assessed by measuring fluorescence at T = 8 h. “+” designates the presence of the indicated plasmid, while “−” designates its absence. * = P ≤ 0.05 relative to wild-type at the same P_i concentration via one-way ANOVA with Tukey’s posttest. # = P ≤ 0.05 relative to the parent strain carrying P_empty at the same P_i concentration via one-way ANOVA with Tukey’s posttest. n ≥ 3. Error bars = SEM.

Figure 3.. The staphylococcal PhoU homologs are not essential for viability.

(A-F) S. aureus wild-type and the indicated strains were grown in PFM9, pH 7.4, supplemented with 5 mM or 500 μM P_i. Growth was assessed by measuring absorbance at OD₆₀₀. n ≥ 3. Error bars = SEM. C-F) Wild-type S. aureus and the indicated mutants containing an empty vector (pKK30), P_pitR (C-D) or P_lgtphoU (E-F) expression constructs were grown in PFM9 supplement with 5 mM or 500 μM P_i as indicated. n ≥ 3. Error bars = SEM. (G) S. aureus wild-type and the indicated strains were grown in PFM9, supplemented with 5 mM or 500 μM P_i, at pH 7.4 and intracellular P_i was assessed. * = P ≤ 0.05 relative to wild-type at the same P_i concentration via two-way ANOVA with Dunnett’s posttest. n ≥ 5. Error bars = SEM.

Figure 4.. Environment dictates the contribution of accessory proteins to controlling P_i homeostasis.

(A, B) S. aureus wild-type and the indicated strains were grown in PFM9, supplemented with 5 mM or 500 μM P_i, at pH 6.4 (A) or 8.4 (B), and intracellular P_i was assessed. * = P ≤ 0.05 relative to wild-type at the same P_i concentration via two-way ANOVA with Dunnett’s posttest. n ≥ 5. Error bars = SEM. ND, not determined due to an inability to generate sufficient biomass. (C, D, E, F) S. aureus wild-type and the indicated strains containing (C & D) P_pstS-yfp or (E & F) P_nptA-yfp reporters were grown in PFM9, at pH 6.4 (C & E) or 8.4 (D & F), supplemented with 5 mM, 500 μM, or 50 μM P_i. The expression of pstSCAB or nptA was assessed by measuring fluorescence at T = 8 h. * = P ≤ 0.05 relative to wild-type at the same P_i concentration via one-way ANOVA with Tukey’s posttest. n ≥ 3. Error bars = SEM. # = P ≤ 0.05 relative to ΔpitR at the same P_i concentration via one-way ANOVA with Tukey’s posttest.

Figure 5.. Multiple accessory proteins contribute to the ability of Staphylococcus aureus to cause infection.

Wild-type C57BL/6J mice were systemically infected with S. aureus wild-type and the indicated strains and weight loss was monitored (A) and bacterial burdens in the liver (B) and heart (C) were enumerated 4 days post-infection by plating for colony forming units. (A) * = P ≤ 0.05 for ΔpitRnptA_ΔphoU mutant compared to the wild-type (Day 3 p.i.) and # = P ≤ 0.05 for ΔpitR, ΔpitRnptA_ΔphoU, ΔphoUnptA_ΔphoU, and ΔpitRΔphoUnptA_ΔphoU mutants compared to the wild-type (Day 4 p.i.) via two-way ANOVA with Dunnett’s posttest. Error bars = SEM. (B & C) * = P < 0.05 relative to wild-type by Mann-Whitney test. # = P < 0.05 relative to ΔphoU by Mann-Whitney test. & = P < 0.05 relative to ΔpitRΔphoUnptA_ΔphoU by Mann-Whitney test. Only significant P values are shown. Lines indicate medians. The data are results from three independent experiments. n ≥ 14 for each group.

Figure 6.. Model of the regulatory mechanisms controlling phosphate homeostasis.

In P_i-deplete environments, PhoR is in kinase conformation and P_i import is accomplished by the PstSCAB, PitA, and NptA transporters. In P_i-replete environments, the accessory regulatory proteins exert repression of PhoR, which is in phosphatase conformation. The transporters PitA and NptA maintain P_i import activities, however PstSCAB expression and its subsequent P_i import is drastically decreased. The classical accessory regulatory protein PhoU associated with the PstSCAB P_i-transporter represses PhoR in vivo during infection, along with the PitR protein associated with the PitA P_i-transporter. In neutral and acidic conditions in vitro, PitR is the primary accessory regulatory protein that represses PhoR. In alkaline conditions in vitro, the PhoU-domain of NptA represses PhoR.