Inhibition of adhesion of Clostridium difficile to Caco-2 cells

Abstract For many microorganisms, including Clostridium difficile , mucosal association is an important factor influencing intestinal colonisation and subsequent infection. Inhibition of adhesion of C. difficile to intestinal mucosa could be a new promising strategy for prevention and treatment of antibiotic-associated diarrhoea. We investigated the possibilities of influencing the adhesion of C. difficile by xylitol and bovine colostrum.whey. Caco-2 cells and C. difficile cells were incubated with 1%, 5% and 10% solutions of xylitol and colostrum. Our study revealed that both xylitol and colostrum inhibited the adhesion of C. difficile to Caco-2 cells. Inhibition by xylitol was dose-dependent. When compared to the control, the count of adherent C. difficile decreased 3.4 times when treated with 1% xylitol, 12 times when 5% xylitol was applied, and 18.7 times when treated with 10% xylitol. The inhibition of adherence by colostrum was partially dose-dependent: 3.1 times in the case of 1%, and 5.5 times in the cases of 5% and 10% colostrum. Further experimental and clinical studies are needed for the application of xylitol and colostrum in the treatment and prophylaxis of pseudomembraneous colitis.     

Comparison of the Caco-2, HT-29 and the mucus-secreting HT29-MTX intestinal cell models to investigate Salmonella adhesion and invasion

Human intestinal cell models are widely used to study host-enteric pathogen interactions, with different cell lines exhibiting specific characteristics and functions in the gut epithelium. In particular, the presence of mucus may play an important role in adhesion and invasion of pathogens. The aim of this study was to evaluate the suitability of the mucus-secreting HT29-MTX intestinal epithelial cell model to test adhesion and invasion of Salmonella strains and compare with data obtained with the more commonly used Caco-2 and HT-29 models. Adhesion of Salmonella to HT29-MTX cell model was significantly higher, likely due to high adhesiveness to mucins present in the native human mucus layer covering the whole cell surface, compared to the non- and low-mucus producing Caco-2 and HT-29 cell models, respectively. In addition, invasion percentages of some clinical Salmonella strains to HT29-MTX cultures were remarkably higher than to Caco-2 and HT-29 cells suggesting that these Salmonellae have subverted the mucus to enhance pathogenicity. The transepithelial electrical resistances of the infected HT29-MTX cell model decreased broadly and were highly correlated with invasion ability of the strain. Staining of S. Typhimurium-infected cell epithelium confirmed the higher invasion by Salmonella and subsequent disruption of tight junctions of HT29-MTX cell model compared with the Caco-2 and HT-29 cell models. Data from this study suggest that the HT29-MTX cell model, with more physiologically relevant characteristics with the mucus layer formation, could be better suited for studying cells–pathogens interactions.     

Lactococci as probiotic strains: adhesion to human enterocyte-like Caco-2 cells and tolerance to low pH and bile

There have been few studies on the probiotic activity of Lactococcus strains although they are commonly used as starter bacteria in manufacturing many kinds of fermented dairy products. Nine strains of the genus Lactococcus were examined for their probiotic properties, such as adherence to human enterocyte-like Caco-2 cells and tolerance to acid and bile. Six strains were adhesive and the highest adhesion was observed with Lactcoccus lactis ssp. lactis NIAI527. This strain adhered to the microvilli of cells as observed by scanning electron microscopy and also tolerated low pH and bile. These properties should make strain 527 a potential new probiotic strain.     

Adhesion of different bifidobacteria strains to human enterocyte-like Caco-2 cells and comparison with in vivo study

The validity of the in vitro adhesion tests performed with cultured cell lines, was determined in this study by comparison with results obtained in vivo , in a previous study. To make this experiment the in vitro adhesion tests were performed during a long period by utilization of an appropriate medium, to determine the capacity of the adhered strain to colonize the intestinal tract. It was demonstrated that the ability of the strain to adhere and colonize the intestinal cell in vivo or the cultured intestinal cells in intro was similar.     

Cell surface-associated lipoteichoic acid acts as an adhesion factor for attachment of Lactobacillus johnsonii La1 to human enterocyte-like Caco-2 cells

The influence of pH on the adhesion of two Lactobacillus strains to Caco-2 human intestinal cells was investigated. One strain, Lactobacillus johnsonii La1, was adherent at any pH between 4 and 7. The other one, L. acidophilus La10, did not attach to this cell line under the same experimental conditions. On the basis of these results, we used the monoclonal antibody technique as a tool to determine differences on the surface of these bacteria and to identify a factor for adhesion. Mice were immunized with live La1, and the hybridomas produced by fusion of spleen cells with ONS1 cells were screened for the production of antibodies specific for L. johnsonii La1. A set of these monoclonal antibodies was directed against a nonproteinaceous component of the L. johnsonii La1 surface. It was identified as lipoteichoic acid (LTA). This molecule was isolated, chemically characterized, and tested in adhesion experiments in the same system. The adhesion of L. johnsonii La1 to Caco-2 cells was inhibited in a concentration-dependent way by purified LTA as well as by L. johnsonii La1 culture supernatant that contained LTA. These results showed that the mechanism of adhesion of L. johnsonii La1 to human Caco-2 cells involves LTA.     

Vasoactive intestinal peptide receptor activity and specificity during enterocyte-like differentiation and retrodifferentiation of the human colonic cancerous subclone HT29-18

Commitment of HT29-18 cells to enterocyte-like differentiation by glucose removal is related to a decreased capacity to generate cAMP after treatment with vasoactive intestinal peptide (VIP), forskolin or sodium fluoride. In contrast, the potency of VIP (EC50 = 1.1 - 1.3 X 10(-10) M) and the pharmacological specificity of the VIP receptor (VIP greater than rh GRF 1-43 greater than PHI greater than secretin) are unchanged during differentiation and retrodifferentiation. These results indicate that disturbances in VIP receptor-post-receptor activity, involving cell surface VIP receptors, membrane and intracellular transducers of hormonal information, occur during enterocyte-like differentiation of the HT29-18 subclone.     

Factors involved in adherence of lactobacilli to human Caco-2 cells.

A quantitative assay performed with bacterial cells labelled with [3H]thymidine was used to investigate factors involved in the adherence of human isolates Lactobacillus acidophilus BG2FO4 and NCFM/N2 and Lactobacillus gasseri ADH to human Caco-2 intestinal cells. For all three strains, adherence was concentration dependent, greater at acidic pH values, and significantly greater than adherence of a control dairy isolate, Lactobacillus delbrueckii subsp. bulgaricus 1489. Adherence of L. acidophilus BG2FO4 and NCFM/N2 was decreased by protease treatment of the bacterial cells, whereas adherence of L. gasseri ADH either was not affected or was enhanced by protease treatment. Putative surface layer proteins were identified on L. acidophilus BG2FO4 and NCFM/N2 cells but were not involved in adherence. Periodate oxidation of bacterial cell surface carbohydrates significantly reduced adherence of L. gasseri ADH, moderately reduced adherence of L. acidophilus BG2FO4, and had no effect on adherence of L. acidophilus NCFM/N2. These results indicate that Lactobacillus species adhere to human intestinal cells via mechanisms which involve different combinations of carbohydrate and protein factors on the bacterial cell surface. The involvement of a secreted bridging protein, which has been proposed as the primary mediator of adherence of L. acidophilus BG2FO4 in spent culture supernatant (M.-H. Coconnier, T. R. Klaenhammer, S. Kernéis, M.-F. Bernet, and A. L. Servin, Appl. Environ. Microbiol. 58:2034-2039, 1992), was not confirmed in this study. Rather, a pH effect on Caco-2 cells contributed significantly to the adherence of this strain in spent culture supernatant.(ABSTRACT TRUNCATED AT 250 WORDS)     

Interactions of clinical and environmental Aeromonas isolates with Caco-2 and HT29 intestinal epithelial cells

AIM: Evaluation of adherence and invasion of Aeromonas spp. to human colon carcinoma cell lines Caco-2 and HT29 and assessment of cytotoxic activity. METHODS AND RESULTS: A number of 27 strains of Aeromonas caviae and 23 strains of Aeromonas hydrophila was analysed. All strains were capable to adhere to sub-confluent monolayers of Caco-2 and HT29 cell types, presenting aggregative and diffuse adherence patterns cells, respectively. In the cytotoxic assays all strains showed cytopathic and/or cytotoxic activities to Vero cells. The evaluation of the tetrazolium salt (MTT test) reduction capability was carried out in Vero, Caco-2, and HT29 cells. MTT test showed that Vero cell line was the most sensitive cell type. In the invasion test, 13 strains were analysed on Caco-2 and HT29 monolayers. Only two (15%) of the 13 strains, A. hydrophila and A. caviae species, both isolated from vegetables were invasive to Caco-2 cells. No strains were able to invade the HT29 cells. CONCLUSIONS: A. hydrophila and A. caviae isolated from human diarrhoeic faeces, vegetables, and water, were able to adhere to and produce cytotoxic/cytopathic effects in intestinal epithelial cell lines. SIGNIFICANCE AND IMPACT OF THE STUDY: The presence of Aeromonas spp. in food and water samples expressing virulence factors suggest that these sources may act as dissemination vehicles of human pathogen with implication in the public health.     

Binding of Clostridium difficile to Caco-2 epithelial cell line and to extracellular matrix proteins

Adhesion of Clostridium difficile to Caco-2 was examined as a function of monolayers polarization and differentiation. The number of adherent C. difficile C253 bacteria per cell strongly decreased when postconfluent 15-day-old monolayers were used (1.7 bacteria per cell versus 17.3 with 3-day-old monolayers). Following disruption of intercellular junctions by ethylene glycol-bis(beta-aminoethyl ether)-N,N,N',N',-tetraacetic acid, a significant rise in the level of bacterial adhesion was observed, above all in postconfluent monolayers. Immunofluorescence studies of bacteria and transferrin receptor, a marker of basolateral pole of polarized monolayers, showed that C. difficile C253 adheres mainly to the basolateral surface of differentiated and undifferentiated polarized Caco-2 cells. Furthermore, binding of C. difficile C253 to several extracellular matrix proteins in vitro was demonstrated by an ELISA-based assay.     

Characterization of Na+/K+/Cl- cotransport in cultured HT29 human colonic adenocarcinoma cells

A Na+/K+/Cl- cotransport pathway has been examined in the HT29 human colonic adenocarcinoma cell line using 86Rb as the K congener. Ouabain-resistant bumetanide-sensitive (OR-BS) K+ influx in attached HT29 cells was 17.9 +/- 0.9 nmol/min per mg protein at 25 degrees C. The identity of this pathway as a Na+/K+/Cl- cotransporter has been deduced from the following findings: (a) OR-BS K+ influx ceased if the external Cl- (Cl-o) was replaced by NO3- or the external Na+ (Na+o) by choline; (b) neither OR-BS 24Na+ nor 36Cl- influx was detectable in the absence of external K+ (K+o); and (c) concomitant measurements of 86Rb+, 22Na+, and 36Cl- influx indicated that the stoichiometry of the cotransport system approached a ratio of 1N+:1K+:2Cl-. In addition, OR-BS K+ influx was exquisitely sensitive to cellular ATP levels. Depletion of the normal ATP content of 35-40 nmol/mg protein to 10-15 nmol/mg protein, a concentration at which the ouabain-sensitive K+ influx was unaffected, completely abolished K+ cotransport. OR-BS K+ influx was slightly reduced by the divalent cations Ca2+, Ba2+, Mg2+ and Mn2+. Although changes in cell volume, whether shrinking or swelling, did not influence OR-BS K+ influx, ouabain-sensitive K+ influx was activated by cell swelling. As in T84 cells, we found that the OR-BS K+ influx in HT29 cells was stimulated by exogenous cyclic AMP analogues and by augmented cyclic AMP content in response to vasoactive intestinal peptide, forskolin, norepinephrine and forskolin or prostaglandin E1.     

Altered expression of sialylated carbohydrate antigens in HT29 colonic carcinoma cells

To determine whether cell growth conditions impacted carbohydrate expression, HT29 cells were gradually transferred from a conventional glucose-containing media to a glucose-free galactose containing media. Indirect immunofluorescence on acetone fixed cells showed increased expression of sialyl Lewis A antigen (CA19-9), sialyl Lewis C (DUPAN2) and Tn/sialyl-Tn on the surface of HT29 cells grown in the glucose-free galactose containing media compared to those grown in the glucose containing media. Sialyltransferases responsible for the synthesis for these sialylated epitopes were increased in the galactose-fed HT29 cells. Media overlying the cells was subjected to isopycnic ultracentrifugation in cesium chloride and the fractions derived from both glucose and galactose media with equivalent buoyant densities of 1.56 g/L, which are predicted to contain mucin glycoforms, were further separated by HPLC using a Mono-Q anion exchange column. The chromatograph of eluent from the sample derived from the cells growing in the galactose containing media showed an increased peak that reacted with the anti-sialyl Lewis A antibody, CA19-9. These results show that alteration of in vitro culture conditions may cause HT29 colonic carcinoma cells to alter the expression of sialylated carbohydrates.     

Characterization of Caco-2 and HT29-MTX cocultures in an in vitro digestion/cell culture model used to predict iron bioavailability

Cocultures of two human cell lines, Caco-2 and HT29-MTX mucus-producing cells, have been incorporated into an in vitro digestion/cell culture model used to predict iron bioavailability. A range of different foods were subjected to in vitro digestion, and iron bioavailability from digests was assessed with Caco-2, Caco-2 overlaid with porcine mucin, HT29-MTX or cocultures of Caco-2 and HT29-MTX at varying ratios. It was found that increasing the ratio of HT29-MTX cells decreased the amount of ferritin formed and resulted in an overall decline in the ability of the model to detect differences in iron bioavailability. At the physiologically relevant ratios of 90% Caco-2/10% HT29-MTX and 75% Caco-2/25% HT29-MTX, however, a mucus layer completely covered the cell monolayer and the in vitro digestion model was nearly as responsive to changes in sample iron bioavailability as pure Caco-2 cultures. The in vitro digestion/Caco-2 cell culture model correlates well with human iron bioavailability studies, but, as mucus appears to play a role in iron absorption, the addition of a physiologically realistic mucus layer and goblet-type cells to this model may give more accurate iron bioavailability predictions.     

Expression levels of heat shock proteins in enterocyte-like Caco-2 cells after exposure to Salmonella enteritidis

The enterocytes of the small intestine are occasionally exposed to pathogenic bacteria, such as Salmonella enteritidis 857, an etiologic agent of intestinal infections in humans. The expression of the heat shock response by enterocytes may be part of a protective mechanism developed against pathogenic bacteria in the intestinal lumen. We aimed at investigating whether S. enteritidis 857 is able to induce a heat shock response in crypt- and villus-like Caco-2 cells and at establishing the extent of the induction. To establish whether S. enteritidis 857 interfered with the integrity of the cell monolayer, the transepithelial electrical resistance (TEER) of filter-grown, differentiated (villus-like) Caco-2 cells was measured. We clearly observed damage to the integrity of the cell monolayer by measuring the TEER. The stress response was screened in both crypt- and villus-like Caco-2 cells exposed to heat (40-43 degrees C) or to graded numbers (10(1)-10(8)) of bacteria and in villus-like cells exposed to S. enteritidis 857 endotoxin. Expression of the heat shock proteins Hsp70 and Hsp90 was analyzed by polyacrylamide gel electrophoresis and immunoblotting with monoclonal antibodies. Exposure to heat or Salmonella resulted in increased levels of Hsp70 and Hsp90 in a temperature-effect or Salmonella-dose relationship, respectively. Incubation of Caco-2 cells with S. enteritidis 857 endotoxin did not induce heat shock gene expression. We conclude that S. enteritidis 857 significantly increases the levels of stress proteins in enterocyte-like Caco-2 cells. However, our data on TEER clearly indicate that this increase is insufficient to protect the cells.     

Rotavirus impairs the biosynthesis of brush-border-associated dipeptidyl peptidase IV in human enterocyte-like Caco-2/TC7 cells

Rotavirus is the leading cause of severe dehydrating diarrhoea in infants and young children worldwide. This virus infects mature enterocytes in the small intestine, and induces structural and functional damage. In the present study, we have identified a new mechanism by which rotavirus impairs a brush border-associated intestinal protein. We show that infection of enterocyte-like Caco-2/TC7 cells by rhesus monkey rotavirus (RRV) impairs the biosynthesis of dipeptidyl peptidase IV (DPP IV), an important hydrolase in the digestion of dietary proline-rich proteins. We show that the enzyme activity of DPP IV was reduced, and that rearrangements of the protein occurred at the apical domain of the RRV-infected cells. Using pulse-chase experiments and cell surface immunoprecipitation, we have demonstrated that RRV infection did not affect the stability or apical targeting of DPP IV, but did induce a dramatic decrease in its biosynthesis. Using quantitative RT-PCR, we showed that RRV had no effect on the level of expression of DPP IV mRNA, suggesting that the observed decrease in the biosynthesis of the protein is related to an effect of the virus at the translational level.     

Different responses of Fe transporters in Caco-2/HT29-MTX cocultures than in independent Caco-2 cell cultures

The human intestinal epithelium is composed of several cell types, mainly enterocytes and goblet (mucin-secreting) cells. This study compares the cellular response of Fe transporters in Caco-2, HT29-MTX, and Caco-2/HT29-MTX co-culture models for Fe bioavailability. Caco-2 cells in vitro differentiate into enterocyte-like cells and HT29-MTX cell lineage into a mucin-secreting cellular population. Cell cultures were exposed to digests of Fe⁺³, Fe⁺³/ascorbic acid, cooked fish (high-available Fe) or white beans (low-available Fe). Cell responses as shown by mRNA expression of the main Fe transporters, DMT1 and DcytB, and cell ferritin formation were monitored. In Caco-2/HT29-MTX co-cultures, the mucin layer lowered the pool of free Fe to diffuse towards the cell brush border membrane of enterocytes, which was accompanied of an upregulation of DMT1 mRNA expression. In contrast, cultures exposed to digests of fish or white beans showed no significant differences in the regulation of Fe transporters.     

Interaction of Clostridium difficile toxin A with L cells in culture

Toxin A of Clostridium difficile was purified by column chromatography and acetic acid precipitation. Cells exposed to toxin A showed polarization of nuclei towards one pole of the cells. Toxin A was conjugated to ferritin and applied to L cells to localize binding sites of this toxin to the cell surface. It was found that toxin A conjugate attached to the cell membrane in aggregated form. Antibody specific to toxin A was prepared and used for localization of intracellular toxins in intoxicated cells. Toxin A was found inside the cytoplasm 6 h after cell treatment, mainly in the form of aggregates inside the cytoplasmic vacuoles. At 24 h after exposure, toxin A could be detected within the cytoplasm. Tunicamycin treatment of cells reduced the cell-binding efficiency of toxin A to 50%, but neuraminidase did not effect toxin binding significantly.     

Protein-mediated adhesion of Lactobacillus acidophilus BG2FO4 on human enterocyte and mucus-secreting cell lines in culture.

The adhesion of Lactobacillus acidophilus BG2FO4, a human stool isolate, to two human enterocytelike cell lines (Caco-2 and HT-29) and to the mucus secreted by a subpopulation of mucus-secreting HT29-MTX cells was investigated. Scanning electron microscopy revealed that the bacteria interacted with the well-defined apical microvilli of Caco-2 cells without cell damage and with the mucus secreted by the subpopulation of HT29-MTX cells. The adhesion to Caco-2 cells did not require calcium and involved an adhesion-promoting factor that was present in the spent supernatant of L. acidophilus cultures. This factor promoted adhesion of poorly adhering human Lactobacillus casei GG but did not promote adhesion of L. casei CNRZ 387, a strain of dairy origin. The adherence components on the bacterial cells and in the spent supernatant were partially characterized. Carbohydrates on the bacterial cell wall appeared to be partly responsible for the interaction between the bacteria and the extracellular adhesion-promoting factor. The adhesion-promoting factor was proteinaceous, since trypsin treatment dramatically decreased the adhesion of the L. acidophilus strain. The adhesion-promoting factor may be an important component of Lactobacillus species that colonize the gastrointestinal tract.