A bacterial virulence protein promotes pathogenicity by inhibiting the bacterium's own F1Fo ATP synthase

Abstract

Several intracellular pathogens, including Salmonella enterica and Mycobacterium tuberculosis, require the virulence protein MgtC to survive within macrophages and to cause a lethal infection in mice. We now report that, unlike secreted virulence factors that target the host vacuolar ATPase to withstand phagosomal acidity, the MgtC protein acts on Salmonella's own F1Fo ATP synthase. This complex couples proton translocation to ATP synthesis/hydrolysis and is required for virulence. We establish that MgtC interacts with the a subunit of the F1Fo ATP synthase, hindering ATP-driven proton translocation and NADH-driven ATP synthesis in inverted vesicles. An mgtC null mutant displays heightened ATP levels and an acidic cytoplasm, whereas mgtC overexpression decreases ATP levels. A single amino acid substitution in MgtC that prevents binding to the F1Fo ATP synthase abolishes control of ATP levels and attenuates pathogenicity. MgtC provides a singular example of a virulence protein that promotes pathogenicity by interfering with another virulence protein.

Figure 1. In Vitro-Synthesized MgtC Interacts with ATP Synthase F_o a Subunit in Proteoliposomes

Western blot analysis of proteoliposomes reconstituted from in vitro synthesized F₁F_o ATP synthase containing F_o a-HA and in vitro synthesized MgtC-FLAG, YhiD-FLAG or MgtC N92T-FLAG proteins. At the end of the reconstitution reaction, an aliquot (input) and fractions immunoprecipitated with either anti-HA or anti-FLAG antibodies were analyzed using anti-HA and anti-FLAG antibodies. Proteoliposomes were prepared as described in Experimental Procedures. The data are representative of two independent experiments, which gave similar results. See also Figures S1, S2 and S3.

Figure 2. MgtC Inhibits ATP-Coupled Proton Translocation and ATP Hydrolysis in an atpB-Dependent Manner

(A) Schematic cartoon of the ATP-driven proton translocation assay (see text for details). (B–E) Fluorescence quenching of the pH-dependent dye ACMA driven by ATP (B, C) or NADH (D, E) in inverted vesicles prepared from wild -type (14028s), mgtC (EL4), atpB (MP23), mgtC atpB (MP25) Salmonella or wild -type Salmonella harboring a plasmid expressing either the wild -type mgtC gene or the mgtC_N92T variant from a heterologous promoter (pmgtC or pmgtC_N92T) or the plasmid vector (pUHE21-2lacI^q). Proton-translocation was initiated by adding ATP and terminated by adding NH₄Cl as indicated by the arrow. % ACMA fluorescence corresponds to the relative change in fluorescence intensity before adding ATP or NADH when fluorescence was set to 100%. (F) Schematic cartoon of the ATP hydrolysis assay (see text for details). (G–I) ATP hydrolysis measured by phosphate release in inverted vesicles prepared from wild-type (14028s), mgtC (EL4) (G), atpB (MP23), and mgtC atpB (MP25) Salmonella (H), or wild-type Salmonella harboring a plasmid expressing the mgtC gene or its derivative from a heterologous promoter (pmgtC or pmgtC_N92T) or the plasmid vector (I). The reaction was initiated by adding ATP and monitored for 5 min as described in Experimental Procedures. For Figures B, D, G and H, vesicles were prepared from cells grown in 10 μM Mg²⁺ to induce mgtC expression from the normal chromosomal location (Soncini et al., 1996) and for Figures C, E and I, vesicles were prepared from cells grown in 50 μM Mg²⁺ in the presence of 0.25 mM IPTG to induce mgtC expression from the heterologous promoter in the plasmid -borne mgtC. Data are represented as mean ± SEM. See also Figure S4 and S5.

Figure 3. MgtC Inhibits NADH-Driven ATP Synthesis in an atpB-Dependent Manner

(A) Schematic cartoon of the NADH-driven ATP synthesis assay (see text for details). (B–E) NADH-driven ATP synthesis assay in inverted membrane vesicles prepared from wild -type (14028s), mgtC (EL4), atpB (MP23), mgtC atpB (MP25) Salmonella (B, D) or wild -type Salmonella harboring a plasmid expressing either the wild -type mgtC gene or the mgtC_N92T variant from a heterologous promoter (pmgtC or pmgtC_N92T) or the plasmid vector (pUHE21-2lacI^q) (C, E) in the presence (+Pi) or absence (−Pi) of phosphate. The ATP synthesis reaction was initiated by adding NADH and monitored by the luciferase reaction for 210 sec as described in Experimental Procedures. Light (100% or 50%) corresponds to the initial luminescence intensity before adding NADH. Vesicles were prepared from cells grown as described in Figure 2. See also Figure S4 and S5.

Figure 4. MgtC Controls Intracellular ATP Levels and the Proton Gradient (ΔpH) in an atpB-Dependent Manner

(A) Intracellular ATP levels (determined at an OD₆₀₀: 0.1) of wild -type (14028s), mgtC (EL4), mgtB (EL5), mgtC_N92T (EL551) Salmonella or wild -type (14028s) Salmonella harboring the plasmid vector (pUHE21-2lacI^q) or derivatives with the mgtC (pmgtC), mgtB (pmgtB), or mgtC_N92T (pmgtC_N92T) genes, or an atpB mutant (EL515) and mgtC atpB (EL516) mutant Salmonella or atpB mutant (EL515) harboring either the plasmid vector or the mgtC gene. Bacteria were grown for 4 h in N-minimal media pH 7.7 containing low Mg²⁺ (10 μM), and for plasmid -harboring strains IPTG (0.2 mM). Nucleic acids were extracted as described in Experimental Procedures. Intracellular ATP levels correspond to picomoles of ATP per ml of cells at given OD₆₀₀. (B) In tracellular ATP levels of mgtC (EL4) Salmonella harboring the plasmid vector (pUHE21-2lacI^q) or a derivative with the mgtC coding region (pmgtC) determined at an OD₆₀₀: 0.177. Bacteria were grown as described in (A). (C) Change in intracellular pH of wild -type (14028s), mgtC (EL4), mgtB (EL5), atpB (EL515), and mgtC atpB (EL516) Salmonella that had been grown at pH 7.7 and switched to pH 5.1 for 1 h. Intracellular pH values in brackets are representative of two measurements, which gave similar results. Values for wild -type and atpB mutant are similar to those previously reported using a different method (Foster and Hall, 1991). (D) Intracellular ATP levels of wild -type (14028s) and mgtC (EL4) Salmonella grown as described in (C). Control experiment was carried out at pH 5.1 in the presence of the uncoupler carbonyl cyanide m-chlorophenyl hydrazone (CCCP). Data in (A–D) are represented as mean ± SEM. (E) Intracellular pH of wild -type (14028s) and mgtC (EL4) Salmonella inside the macrophage-like cell line J774A.1. Number represents the average pH of five independent replicates for wild -type (14028s) and six replicates for mgtC (EL4) Salmonella, which gave similar bacterial colony counts when plated on LB agar plates. See also Figure S5.

Figure 5. MgtC Affects Membrane Potential

(A) Membrane potential of Salmonella strains listed and grown as described in Figure 4A. Red/ green fluorescence ratio was determined after incubation with the membrane potential-dependent dye DiOC₂ for 30 min in either the presence or absence of the uncoupler carbonyl cyanide m-chlorophenyl hydrazone (CCCP). (B) Same as in (A) but bacteria were grown in N -minimal medium with 10 mM Mg²⁺. Data are represented as mean ± SEM. See also Figure S5.

Figure 6. The Defect in Macrophage Survival and Virulence in Mice of the mgtC Mutant Salmonella is atpB-Dependent

(A) Survival of wild-type (14028s), mgtC (EL4), phoP (MS7953s), atpB (EL515), mgtC atpB (EL516), and phoP atpB (EL535) Salmonella inside J774A.1 macrophages at 18 h after infection. The inset shows survival values below 20%. Data are represented as mean ± SEM. (B–C) Survival of C3H/ HeN mice inoculated intraperitoneally with ~ 10⁴ (B) or ~10⁵ (C) colony forming units of wild -type (14028s), mgtC (EL4), atpB (EL515), and mgtC atpB (EL516) Salmonella. The data are representative of two independent experiments, which gave similar results. See also Figure S5.

Figure 7. Intracellular Pathogens Utilize Different Strategies to Cope with Phagosome Acidification

(A) S. enterica harbors the inner membrane protein MgtC, which targets the bacterium's own F₁F_o ATPase to decrease intracellular ATP levels heightened by phagosomal acidic pH. (B) Legionella pneumophila secretes the effector protein SidK to the host phagosomal membrane, where it inhibits phagosome acidification driven by the host vacuolar ATPase (Xu et al., 2010). (C) M. tuberculosis secretes the PtpA protein, which affects the host vacuolar ATPase to hinder phagosome acidification (Wong et al., 2011), and harbors a functional mgtC gene (Buchmeier et al., 2000), which is expected to reduce intracellular ATP levels.