Diminishing of Helicobacter pylori adhesion to Cavia porcellus gastric epithelial cells by BCG vaccine mycobacteria

Abstract

Mycobacterium bovis onco-BCG bacilli used in immunotherapy of bladder cancer are candidates for training of immune cells towards microbial pathogens. Increasing antibiotic resistance of gastric pathogen Helicobacter pylori (Hp) prompts the search for new anti-Hp and immunomodulatory formulations. Colonization of gastric mucosa by Hp through mucin 5 AC (MUC5AC) ligands could potentially be a therapeutic target. The aim of this study was to examine the ability of onco-BCG mycobacteria to reduce Hp adhesion to gastric epithelial cells using Cavia porcellus model. Animals were inoculated per os with 0.85% NaCl, Hp alone, onco-BCG alone or with onco-BCG and Hp. After 7/28 days Mucin5AC and Hp binding to gastric epithelium were assessed in gastric tissue specimens by staining with anti-Mucin5AC and anti-Hp antibodies, respectively, both fluorescently labeled. Primary gastric epithelial cells were treated ex vivo with live Hp or Hp surface antigens (glycine extract or lipopolysaccharide) alone or with onco-BCG. In such cells MUC5AC and Hp binding were determined as above. Mycobacteria reduced the amount of MUC5AC animals infected with Hp and in gastric epithelial cells pulsed in vitro with Hp components. Decrease of MUC5AC driven in cell cultures in vitro and in gastric tissue exposed ex vivo to mycobacteria was related to diminished adhesion of H. pylori bacilli. Vaccine mycobacteria by diminishing the amount of MUC5AC in gastric epithelial cells may reduce Hp adhesion.

Figure 1

Biocompatibility of M. bovis onco-BCG. The viability of mouse fibroblasts L929 (A) or guinea pig primary gastric epithelial cells (B) treated with onco-BCG mycobacteria (onco-BCG). The percent of cells, which were able to reduce tetrazolium salt (3-(4,5-Dimethylthiazol-2-yl)-2,5-Diphenyltetrazolium Bromide) (MTT) was determined. NC-negative control (cells treated with 0.03% H₂O₂-non viable cells), PC-positive control (cells in medium alone, 100% cell viability). Results are showed as mean ± standard error of mean (SEM). Three experiments were performed in triplicates for each experimental variant. The blue line indicates the minimal percentage of viable cells (70%) required to confirm the onco-BCG as non-cytotoxic in vitro. Statistical analysis was performed using the nonparametric U Mann–Whitney test with significance, p < 0.05 (*unstimulated cells vs. cells exposed to onco-BCG).

Figure 2

Detection of Mucin 5AC (Mucin 5AC) in vitro (A) in cell cultures of guinea pig primary gastric epithelial cells or in vivo (B) in guinea pig gastric tissue. Evaluation of Mucin 5AC in vitro (A) in cells exposed to onco-BCG and/or H. pylori (i), to H. pylori components and/or onco-BCG mycobacteria (BCG) (ii) or in vivo (B) in gastric tissue of control animals or animals inoculated with H. pylori or onco-BCG alone or first with onco-BCG and then with H. pylori. Mucin 5AC was assessed by measurement of the fluorescence intensity following staining of cells with mouse anti-MUC5AC antibodies and with secondary anti-murine immunoglobulin antibodies labeled with fluorescein isothiocyanate (FITC). Cell nuclei were stained with 4′,6-diamidino-2-fenilindol-DAPI. (A) The fluorescence intensity was measured in multifunctional reader SpectraMaxi3 at the appropriate wavelength: 495 nm (excitation), 519 nm (emission). Mean ± standard error of mean (SEM) are shown. Statistical significance at p < 0.05 in the non-parametric U-Mann–Whitney or Kruskal–Wallis test. *Cells treated with H. pylori or components of these bacteria vs untreated cells; ● cells treated with H. pylori components or H. pylori and onco-BCG as compared to cells treated with H. pylori components alone or with these bacteria. Hp-H. pylori; Ec-Escherichia coli, RFU-relative fluorescence unit, GE-glycine extract, LPS-lipopolysaccharide. (B) (i) The number of luminous cells was obtained with ImageJ 1.48v software at 495 nm (excitation), 519 nm (emission). Mean ± standard error of mean (SEM) are shown. (ii) Representative images of gastric tissue of guinea pigs not receiving onco-BCG bacilli and not infected with H. pylori, receiving onco-BCG bacilli, receiving onco-BCG bacilli and then H. pylori, 7 or 28 days after inoculation with H. pylori (n = 5). Sections were immunohistochemically stained for Mucin 5AC with specific fluorescently labeled antibodies and imaged under a confocal microscope (Leica TCS SPE) at wavelengths for FITC: 495 nm (excitation), 519 nm (emission) or for DAPI: 345 nm (excitation), 455 nm (emission) (10× and 40× magnification). Statistical significance for p < 0.05 in the non-parametric Mann–Whitney or Kruskal–Wallis U test. *Animals infected with H. pylori vs uninfected animals, # animals receiving onco-BCG and infected with H. pylori vs animals receiving onco-BCG or infected with H. pylori.

Figure 3

Modulation of H. pylori adhesion to primary gastric epithelial cells of guinea pig (A) or to gastric tissue specimens of guinea pig (B) by onco-BCG mycobacteria. Binding of H. pylori to guinea pig primary gastric epithelial cells (A) was assessed by staining with anti-H. pylori antibodies labeled with fluorescein isothiocyanate (FITC) while onco-BCG mycobacteria (BCG) using BactLight staining procedure. Cell nuclei were stained with 4′,6-diamidino-2-fenilindol-DAPI while the cytoskeleton with Texas-Red-phalloidin. (i) H. pylori adhesion was expressed as fluorescence intensity measured in a multifunctional reader SpectraMax i3 at wavelengths for FITC 495 (excitation) and 519 (emission). The results are presented as mean ± standard error of mean (SEM). Statistical significance for p < 0.05 in the non-parametric Mann–Whitney or Kruskal–Wallis U test. *Cells treated with onco-BCG or H. pylori vs control cells in medium only, ● cells treated with onco-BCG and H. pylori vs cells treated with H. pylori only. (ii) representative images showing the adherence of H. pylori rods to primary gastric epithelial cells of the guinea pig under a confocal microscope at appropriate wavelengths: for FITC 495 nm (excitation) and 519 (emission), for DAPI 345 (excitation) and 455 (emission), and for phalloidin 591 nm (excitation) and 608 nm (emission), magnification 64×. (B) Binding of H. pylori to guinea pig gastric tissue was assessed by staining of specimens with anti-H. pylori antibodies labeled with fluorescein isothiocyanate (FITC). Onco-BCG mycobacteria were stained with BactLight. Cell nuclei were stained with DAPI. Cells were imaged under the confocal microscope (Leica TCS SPE). (i) the number of glowing cells was calculated using ImageJ version 1.48 software at 495nm (excitation) and 519nm (emission) FITC wavelengths. The results are presented as mean ± standard error of mean (SEM). (ii) representative images of gastric tissue with adjacent H. pylori bacilli from a fluorescence microscope, at 495 nm (excitation) and 519 nm (emission) for FITC, 345 nm (excitation) and 455 nm (emission), (10, 100 × magnification) for DAPI Statistical significance for p  < 0.05 in the non-parametric U Mann–Whitney or Kruskal–Wallis test. Tissue treated: *H. pylori and onco-BCG-versus H. pylori, ^#H. pylori or onco-BCG for 2 h and 4 h.