Outer Membrane Vesicle Production by Helicobacter pylori Represents an Approach for the Delivery of Virulence Factors CagA, VacA and UreA into Human Gastric Adenocarcinoma (AGS) Cells

Abstract

Helicobacter pylori infection is the etiology of several gastric-related diseases including gastric cancer. Cytotoxin associated gene A (CagA), vacuolating cytotoxin A (VacA) and α-subunit of urease (UreA) are three major virulence factors of H. pylori, and each of them has a distinct entry pathway and pathogenic mechanism during bacterial infection. H. pylori can shed outer membrane vesicles (OMVs). Therefore, it would be interesting to explore the production kinetics of H. pylori OMVs and its connection with the entry of key virulence factors into host cells. Here, we isolated OMVs from H. pylori 26,695 strain and characterized their properties and interaction kinetics with human gastric adenocarcinoma (AGS) cells. We found that the generation of OMVs and the presence of CagA, VacA and UreA in OMVs were a lasting event throughout different phases of bacterial growth. H. pylori OMVs entered AGS cells mainly through macropinocytosis/phagocytosis. Furthermore, CagA, VacA and UreA could enter AGS cells via OMVs and the treatment with H. pylori OMVs would cause cell death. Comparison of H. pylori 26,695 and clinical strains suggested that the production and characteristics of OMVs are not only limited to laboratory strains commonly in use, but a general phenomenon to most H. pylori strains.

Keywords: CagA; Helicobacter pylori; UreA; VacA; gastrointestinal disorders; outer membrane vesicles.

Figure 1

Identification and characterization of outer membrane vesicles (OMVs) isolated from H. pylori 26,695 strain. (a) The scanning election micrograph of H. pylori 26,695 strain taken with 40,000 × magnification. The white arrow indicates that an OMV is being produced from the surface of a bacterium. (b) The growth curve of H. pylori 26,695 strain and the corresponding protein concentration of OMVs isolated during the course of bacterial growth. Data are shown as the mean ± SD (n = 3). (c) Protein profiles of OMVs isolated from H. pylori 26,695 strain at different time points of culture. M: prestained protein markers. (d) The presence of key virulence factors in OMVs isolated from H. pylori 26,695 strain at different time points.

Figure 2

The association of key virulence factors with OMVs isolated from H. pylori 26,695 strain. OMVs isolated from H. pylori 26,695 strain were treated with (a) PBS, 0.8 M, 8 M urea or 1% SDS, (b,c) sonication or 1% SDS, followed by proteinase K or trypsin treatment. The obtained protein samples were then analyzed by 10% SDS-PAGE, followed with Commasie blue staining (b) or immunoblotting detection (a,c) with anti-CagA, anti-VacA and anti-UreA antibodies. M: prestained protein markers.

Figure 3

OMVs isolated from H. pylori 26,695 strain were associated and internalized into human gastric adenocarcinoma (AGS) cells. (a) The association and internalization of OMVs with AGS cells after 1 h of co-incubation were analyzed using LSM510 confocal microscopy. Green: AGS cells labeled with fluorescein isothiocyanate (FITC)-wheat germ agglutinin (WGA); red: BR-18-labeled OMVs isolated from H. pylori 26,695 strain; blue: 4′,6-Diamidino-2-Phenylindole (DAPI) stained nuclei. The (b) time- and (c) dose-dependent association and internalization of BR-18-labeled OMVs isolated from H. pylori 26,695 strain with AGS cells. The addition of trypan blue (final concentration of 0.025%) was used to distinguish the internalized OMVs from the cell-bound but non-internalized OMVs. Data are shown as the mean ± SD (n = 3).

Figure 4

OMVs isolated from H. pylori 26,695 strain entered AGS cells mainly by macropinocytosis/phagocytosis. (a) The association of OMVs with early endosomes after co-incubation of AGS cells with 1,1′-diocatadecyl-3,3,3′,3′-tetramethylindocar-bocyanine perchlorate (DiI)-labeled OMVs for up to 12 h weas analyzed using LSM800 confocal microscopy. Green: early endosomes-GFP; red: DiI-labeled OMVs isolated from H. pylori 26,695 strain. White arrows indicate OMVs that were in close contact with early endosomes. The effect of endocytic inhibitors on the (b) association and (c) internalization of BR-18-labeled OMVs isolated from H. pylori 26,695 strain with AGS cells. Data are shown as the mean ± SD (n = 3, * p < 0.05; ** p < 0.01).

Figure 5

H. pylori OMVs caused cell death after co-incubation with gastric AGS cells. (a,b) OMVs-associated CagA, VacA and UreA were accumulated in the whole cell extracts of AGS cells after the co-incubation study. (a) Whole cell lysates from AGS cells co-incubated with OMVs isolated from H. pylori 26,695 strain for 6 h or 24 h. Mock, a control group without OMVs co-incubation. (b) Whole cell lysates from AGS cells treated with different concentrations of OMVs isolated from H. pylori 26,695 strain for 24 h or with bacteria of H. pylori 26,695 strain (multiplicity of infection (MOI) = 100) for 6 h. (c) The dose-dependent effect of OMVs isolated from H. pylori 26,695 strain on cell viability. Data are shown as the mean ± SD (n = 3, * p < 0.05; ** p < 0.01; *** p < 0.001).

Figure 6

H. pylori clinical strains produced OMVs with characteristics similar but not identical to those of H. pylori 26,695 strain. (a) The production level of OMVs isolated from H. pylori 26,695 and various clinical strains. Data are shown as the mean ± SD (n = 3). (b) Particle size analysis of OMVs isolated from H. pylori 26,695 and various clinical strains. The presence of key virulence factors CagA, VacA and UreA in (c) whole bacterial lysates and (d) OMVs isolated from H. pylori 26,695 and various clinical strains. (e) The effect of OMVs isolated from H. pylori 26,695 and various clinical strains on the viability of AGS cells. Data are shown as the mean ± SD (n = 3, * p < 0.05; ** p < 0.01; *** p < 0.001).