VopX, a novel Vibrio cholerae T3SS effector, modulates host actin dynamics

Abstract

Pathogenic Vibrio cholerae strains cause cholera using different mechanisms. O1 and O139 serogroup strains use the toxin-co-regulated pilus (TCP) and cholera toxin (CT) for intestinal colonization and to promote secretory diarrhea, while non-O1/non-O139 serogroup strains are typically non-toxigenic and use alternate virulence factors to cause a clinically similar disease. An O39 serogroup, TCP/CT-negative V. cholerae strain, named AM-19226, uses a type III secretion system (T3SS) to translocate more than 10 effector proteins into the host cell cytosol. Effectors VopF and VopM directly interact with the host actin and contribute to colonization. Our previous studies using the Saccharomyces cerevisiae model system identified VopX as a third effector that alters cytoskeletal dynamics. Herein, we used complementary approaches to translate yeast findings to a mammalian system and determined the target and mechanism of VopX activity. VopX overexpression in HeLa cells caused dramatic cell rounding. Co-culture of strain AM-19226 with polarized Caco-2/BBE monolayers increased formation of stress fibers and focal adhesions, as well as Caco-2/BBE adherence to extracellular matrix in a VopX-dependent manner. Finally, we demonstrate in vitro that VopX can act as a guanine nucleotide exchange factor for RhoA, which functions upstream of a mitogen-activated protein kinase (MAPK) signaling pathway regulating cytoskeletal dynamics. Our results suggest that VopX activity initiates a signaling cascade resulting in enhanced cell-extracellular matrix adhesion, potentially preventing detachment of host cells, and facilitating sustained bacterial colonization during infection. VopX function is therefore part of a unique pathogenic strategy employed by T3SS-positive V. cholerae, which involves multiple cytoskeletal remodeling mechanisms to support a productive infection.

Importance: Despite different infection strategies, enteric pathogens commonly employ a T3SS to colonize the human host and cause disease. Effector proteins are unique to each T3SS-encoding bacterial species and generally lack conserved amino acid sequences. However, T3SS effectors from diverse pathogens target and manipulate common host cell structures and signaling proteins, such as the actin cytoskeleton and MAPK pathway components. T3SS-encoding Vibrio cholerae strains and effectors have been relatively recently identified, and the mechanisms used to mediate colonization and secretory diarrhea are poorly understood. Two V. cholerae effectors that modify the host actin cytoskeleton were shown to be important for colonization. We therefore sought to determine the target(s) and mechanism of a third actin-reorganizing effector, VopX, based on results obtained from a yeast model system. We recapitulated actin-based phenotypes in multiple mammalian model systems, leading us to identify the molecular function of the V. cholerae VopX effector protein.

Keywords: T3SS; V. cholerae; cytoskeleton; effector; pathogenesis.

Fig 1

VopX overexpression in HeLa cells results in cell shape changes. (A) Immunofluorescence imaging of HeLa cells infected with Vaccinia virus vTF7-3/RFP-A4at a multiplicity of infection of 0.5 and transiently transfected with a vector expressing VopX-HA. Cells were fluorescently labeled for F-actin with Alexa Fluor 647 phalloidin (red), nuclei with 4′,6-diamidino-2-phenylindole (blue), and HA detected using an anti-HA antibody (green). White arrows indicate a transfected cell and yellow brackets indicate an untransfected cell. (B) Parallel experiment of HeLa cells infected with vTF7-3/RFP-A4 and transiently transfected with pGFP. Scale bars represent 20 μm. (C) Quantification of cell perimeter from transfected and untransfected HeLa cells. Measurements are pooled from three images taken from a single biological replicate. Statistics were generated using two-way analysis of variance with Šídák’s multiple comparison test. *P ≤ 0.05. VopX-HA, transfected n = 3; VopX-HA, transfected n = 19; GFP, transfected n = 6; GFP, untransfected n = 14. The study was conducted three times and produced similar results.

Fig 2

VopX activity reduced Caco-2/BBE cell height during AM-19226 co-culture. (A) Confocal Z stacks viewed from the XZ plane showing differentiated Caco-2/BBE co-cultured with the indicated AM-19226 strain at a multiplicity of infection of ~10 or phosphate-buffered saline (PBS; mock) for 3 h. Cells were fluorescently labeled for F-actin with Alexa Fluor 647 phalloidin (red), nuclei (4′,6-diamidino-2-phenylindole, blue), and V. cholerae detected using a polyclonal rabbit antibody developed against whole AM-19226 cells (green). Yellow arrows indicate apical and lateral F-actin signal. White arrows indicate actin-rich signal below bacteria. Scale bars represent 5 μm. (B) Quantification of cell height in microns. Strains are noted on the X axis. Cell height was determined by measuring the length of the actin signal in the Z plane. Data represent the average ± standard deviation of three biological replicates. Statistics were generated by ordinary one-way analysis of variance with Dunnett’s multiple comparison test. *P ≤ 0.05, **P ≤ 0.01. The study was conducted three times and produced similar results.

Fig 3

VopX is required for stress fiber formation in polarized Caco-2/BBE cells. (A) Three planes, taken at 0.30 μm step size, shown as maximum intensity projections of differentiated Caco-2/BBE monolayers co-cultured with the indicated AM-19226 strain for 3 h (multiplicity of infection of ~10 or phosphate-buffered saline). The top row shows the apical cell surface (a); the middle row shows them midpoint of the cells (m); and the bottom row shows the basal cell surface (b). F-actin was labeled with Alexa Fluor 647 phalloidin (red); nuclei were labeled with 4′,6-diamidino-2-phenylindole (DAPI) (blue); and bacteria were detected using a polyclonal rabbit antibody developed against whole cell AM-19226 (green). Scale bars represent 10 μm. (B) Parallel experiment involving co-labeling of F-actin with Alexa Fluor 647 phalloidin (red), nuclei (DAPI, blue), and actin motor, NM IIB (green). Images depict the basal cell surface. Scale bars represent 10 μm. The study was conducted three times and produced similar results.

Fig 4

VopX activity increases Caco-2/BBE adherence to the extracellular matrix. (A) Maximum intensity projections of three planes, taken at 0.30 μm step size, showing differentiated Caco-2/BBE monolayers co-cultured with the indicated AM-19226 strain for 3 h (multiplicity of infection ~10 or PBS). Cells were fluorescently labeled for F-actin with Alexa Fluor 647 phalloidin (red), nuclei (DAPI, blue), and zyxin (green). Images depict the basal cell surface. Scale bars represent 10 μm. (B) Maximum intensity values of zyxin staining from three 0.9 μm confocal projections. Data represent the average of three biological replicates. Statistics were generated by ordinary one-way analysis of variance (ANOVA) with Dunnett’s multiple comparison test. ***P ≤ 0.001. (C) Optical density at 450 nm of Caco-2/BBE cells stained with crystal violet, following an inverted spin after 3 h co-culture with the indicated AM-19226 strain or PBS (mock). Data represent the average ±standard deviation of four technical replicates. Statistics were generated by ordinary one-way ANOVA with Dunnett’s multiple comparison test. ***P ≤ 0.001, ****P ≤ 0.0001. The study was conducted three times with similar results. ns, not significant.

Fig 5

VopX activity does not alter Caco-2/BBE junctional integrity after co-culture at a multiplicity of infection (MOI) of ~10. (A) Relative transepithelial electrical resistance (TEER) assay of Caco-2/BBE cells co-cultured for 3 h at an MOI of 10 with the indicated V. cholerae strain or PBS (mock). Data are presented as the average ± standard deviation of three biological replicates. (B) Relative transepithelial electrical resistance assay of Caco-2/BBE cells co-cultured for 3 h at an MOI of ~1,000 with the indicated V. cholerae strain or PBS (mock). Data are presented as the average ± standard deviation of three biological replicates. Statistics were generated by ordinary one-way ANOVA with Dunnett’s multiple comparison test. ****P ≤ 0.0001. (C) Three planes, taken at 0.30 μm step size, shown as a maximum intensity projection of differentiated Caco-2/BBE monolayers co-cultured with the indicated AM-19226 strain for 3 h (MOI of ~10 or PBS). Cells were fluorescently labeled for F-actin with phalloidin (red), nuclei with DAPI (blue), and occludin (green). Images show Z slices from the basal cell surface. Scale bars represent 10 μm. The study was repeated with similar results.

Fig 6

Purified VopX acts as a GEF for RhoA in vitro. Relative fluorescence of N-MAR-GTP when combined with purified Rac1 (A), Cdc42 (B), or RhoA (C). The indicated GEF or water was added at T = 5:00. Gray lines indicate activity due to diH₂O; black lines indicate hDBS activity; and dotted lines indicate VopX activity. (D) RhoGEF GTPase exchange rates represented as micromole N-MAR-GTP per micromole GTPase per second. Statistics were generated using two-way ANOVA with Dunnett’s multiple comparison test. ****P < 0.0001. The assay was conducted three times and produced similar results.

Fig 7

VopX and RhoA can be modeled to form a heterodimeric complex. ColabFold-AF2 modeling of the heterodimeric interaction between VopX and RhoA. (A) RhoA is shown in pink and VopX in purple. (B) Duplicate image of RhoA and VopX colored by confidence level: dark blue, pLDTT >90; blue, 70 < pLDTT < 90; yellow, 50 < pLDTT < 70; orange, pLDTT < 50. (C) Enlarged image showing the amino acids at the site of the protein-protein interface. Atoms from amino acids ³⁴PTVFEN⁴¹ (RhoA), ¹⁴⁸WLSFD¹⁵² (VopX), and ¹⁶⁶ANKSH¹⁷⁰ (VopX) are shown at the predicted protein-protein interface.