Vibrio cholerae requires the type VI secretion system virulence factor VasX to kill Dictyostelium discoideum

Abstract

The type VI secretion system (T6SS) is recognized as an important virulence mechanism in several Gram-negative pathogens. In Vibrio cholerae, the causative agent of the diarrheal disease cholera, a minimum of three gene clusters--one main cluster and two auxiliary clusters--are required to form a functional T6SS apparatus capable of conferring virulence toward eukaryotic and prokaryotic hosts. Despite an increasing understanding of the components that make up the T6SS apparatus, little is known about the regulation of these genes and the gene products delivered by this nanomachine. VasH is an important regulator of the V. cholerae T6SS. Here, we present evidence that VasH regulates the production of a newly identified protein, VasX, which in turn requires a functional T6SS for secretion. Deletion of vasX does not affect export or enzymatic function of the structural T6SS proteins Hcp and VgrG-1, suggesting that VasX is dispensable for the assembly of the physical translocon complex. VasX localizes to the bacterial membrane and interacts with membrane lipids. We present VasX as a novel virulence factor of the T6SS, as a V. cholerae mutant lacking vasX exhibits a phenotype of attenuated virulence toward Dictyostelium discoideum.

Fig. 1.

Identification of VasX. (A) A V. cholerae vasH mutant carrying plasmid pBAD24-vasH that allows arabinose-inducible expression of vasH was grown in the absence or presence of 0.1% arabinose. Total protein of culture supernatants was resolved by 10% SDS-PAGE and visualized by silver staining. The 100- to 140-kDa regions from both lanes were excised and subjected to LC-MS/MS analysis. (B) Proteins identified by mass spectrometry as present in culture supernatant after vasH expression was induced. Swiss ID, Swiss-Prot identification number. (C) VasX belongs to a T6SS satellite gene cluster on the V. cholerae small chromosome immediately downstream of hcp-2, vgrG-2, and VCA0019.

Fig. 2.

VasX expression depends on the T6SS regulator VasH. (A) Western blot demonstrating that trans expression of vasH restores production of both Hcp and VasX. Bacterial cultures were grown to late logarithmic phase in the presence or absence of 0.1% arabinose. Pellet and supernatant fractions were separated and subjected to 10% SDS-PAGE followed by Western blotting with anti-VasX, -Hcp, and -DnaK antibodies. Results are representative of three independent experiments. (B) VasH regulates the expression of vasX. V52ΔvasH carrying pBAD24-vasH to allow arabinose-controlled complementation was grown to mid-logarithmic phase in the presence or absence of arabinose. VasX transcript levels were determined by RT-PCR using the gene encoding the V. cholerae outer membrane protein OmpW (ompW) as an internal control. Results are representative of three experimental data sets.

Fig. 3.

Secretion of VasX is T6SS dependent. (A) Western blot of VasX and Hcp in culture supernatants and bacterial pellets. Bacterial cultures were grown to late logarithmic phase; supernatant and pellet fractions were isolated and subjected to Western blotting using the indicated antibody types (listed to the left of the blot). Molecular masses are shown to the right of both blots. Results are representative of at least three experimental data sets. (B) VasX is not required for T6SS-mediated cross-linking of murine macrophage actin. RAW 264.7 macrophages were infected at a multiplicity of infection of 10 for 2 h. Cell lysates were subjected to SDS-PAGE followed by Western blotting using an antiactin antibody. Molecular mass markers are shown to the left of the blot. The hcp deletion strain V52Δhcp-1,2 has both chromosomal copies of hcp deleted. Results are representative of at least three experimental data sets.

Fig. 4.

VasX localizes to the bacterial membrane. V52/pBAD24 was grown to mid-logarithmic phase and subjected to subcellular fractionation. Various fractions (whole cell [WC], permeabilized V52 [PERM], membrane [M], periplasm [PP], and cytosol [CY]) were separated by SDS-PAGE followed by Western blotting with anti-VasX, -Hcp, -OmpU (membrane control), -DnaK (cytosol control), and -β-lactamase (Bla; periplasm control) primary antibodies. Results are representative of three independent experiments.

Fig. 5.

VasX is required for virulence toward D. discoideum. (A) Plaque assay comparing strains V52, V52ΔvasK, V52ΔvasX, and V52ΔvgrG-1. Bacteria were mixed with approximately 500 amoebae, spread on nutrient SM/5 agar, and incubated at 22°C for 4 days to allow for plaque formation. Results are representative of three independent experiments. (B) Graphical representation comparing plaques in lawns of V52 and other T6SS mutants, including V52ΔvasX, V52ΔvasK, V52ΔvgrG-1, and V52ΔvgrG-1ΔvasX. Bacteria and amoebae were mixed on SM/5 nutrient agar and incubated at 22°C for 4 days to allow for plaque formation. Results are representative of three experimental data sets. ***, P value < 0.001; **, P value < 0.01. (C) trans expression of vasX restores virulence toward D. discoideum. V52ΔvasX/pBAD24-vasX and the plasmid control (V52ΔvasX/pBAD24) were mixed with approximately 500 amoebae, spread on nutrient SM/5 agar (with or without 0.1% arabinose), and incubated for 4 days at 22°C to allow for plaque formation. All results are representative of at least three experimental data sets.

Fig. 6.

VasX binds phosphoinositides but not LPS. (A) Schematic representation of the various biological membrane lipids present on PIP Strips, including phosphatidic acid (PA), phosphatidyl inositol (PI), phosphatidylinositol phosphate (PIP), phosphatidylserine (PS), phosphatidylethanolamine (PE), phosphatidylcholine (PC), lysophosphatidic acid (LPA), lysophosphocholine (LPC), and sphingosine-1-phosphate (S1P). (B and C) Purified full-length VasX or a truncated version consisting of residues 1 to 200 (containing the PH domain) was used to probe PIP Strips for lipid binding. Bound protein was detected using anti-6×His primary antibody and anti-mouse-HRP secondary antibody. The positive control (+) included purified VasX spotted directly onto the PIP Strip membrane. In panel C, E. coli LPS was spotted in place of the positive control on the PIP Strip. Results are representative of three independent experiments. (D) MLV pulldown of purified recombinant VasX and the N-terminal fragment of VasX(1–200). Full-length VasX, VasX(1–200), and BSA (negative control) were mixed with MLV of total liver extract (TLE), phosphatidic acid (PA), lysophosphatidic acid (LPA), or PBS (as a technical control) and divided into total (T), pellet (P), and supernatant (S) fractions by centrifugation. The partitioning of protein into each fraction was visualized by SDS-PAGE and Coomassie blue staining. Results are representative of three independent experiments.