1 Institute of Comparative Medicine, University of Glasgow Faculty of Veterinary Medicine, University of Glasgow, Bearsden Road, Glasgow G61 1QH, UK
2 Department of Pharmacy and Biomolecular Sciences, University of Brighton, Lewes Road, Brighton BN2 4GJ, UK
Correspondence
Paul H. Everest
phe3d{at}udcf.gla.ac.uk
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
In previous studies, Caco-2 cells have been used as a model of intestinal epithelium for the interaction of C. jejuni with host cells (Everest et al., 1992, Konkel et al., 1992
, Harvey et al., 1999
). The cells have similar properties to colonic enterocytes, in that they form brush borders with microvilli and maintain tight junctions, which dominate the transepithelial resistance values, thus exhibiting the properties of a polarized cell line (Delie & Rubas, 1997
). In addition, they are used as models for absorptive epithelium as they exhibit the property of transporting fluid from their apical surface to their basolateral surface (Delie & Rubas, 1997
). This process is demonstrable in the formation of domes' across the monolayer when cultured on solid supports. The cells also transport fluid when grown on permeable supports but dome formation cannot be assessed when using these filter systems.
We hypothesized that C. jejuni may inhibit absorption of fluid across epithelia as a mechanism of diarrhoeal disease. This hypothesis was examined by studying the effect of infection on dome formation in Caco-2 cell monolayers. In the absence of any increase in detectable epithelial conductance (representing active secretory events), we reasoned that if C. jejuni could disrupt or inhibit dome formation, or collapse domes once formed, then this would inhibit fluid absorption, a potential mechanism of diarrhoeal disease. In this study, we examined the ability of C. jejuni to collapse fluid-transporting domes within Caco-2 cell monolayers and how the collapse of domes relates to electrical resistance of the monolayer and cellular rearrangement of the tight junction protein occludin.
![]() |
METHODS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Bacteria were grown in MuellerHinton broth and agar (Oxoid) and incubated at 37 °C in a variable-atmosphere incubator (VAIN; Don Whitley Scientific) in an atmosphere of 6 % hydrogen, 5 % carbon dioxide, 5 % oxygen and 84 % nitrogen.
Dome formation in Caco-2 cells.
Caco-2 cells were seeded at 2x104 cells cm2 in 12-well plates and incubated until confluent. Cells were maintained in Dulbecco's minimal essential medium (DMEM) with 10 or 20 % fetal calf serum without antibiotics. The confluent monolayers were washed and then inoculated with 10 µl bacterial suspension, containing varying numbers of bacteria for different experiments (between 103 and 109 c.f.u.). Infected monolayers were incubated for up to 7 days at 37 °C in a 6 % carbon dioxide humidified atmosphere. Monolayers for the investigation of bacterial invasion were also treated with 200 µg gentamicin ml1 for 4 h, for experiments involving bacteria invading cells. The bacteria were left on the monolayers for 24 h prior to gentamicin treatment for low (1000 c.f.u.) inoculum experiments. After 4 h, the cells were washed and replaced with DMEM without gentamicin for the remainder of the experiment. Gentamicin kills extracellular bacteria, allowing the effects of intracellular bacteria alone to be investigated. To check that bacteria were indeed intracellular, representative monolayers were lysed using 1 % Triton-X and viability was counted on MuellerHinton agar. Domes were counted daily by microscopy for infected and uninfected monolayers. Results were expressed as domes per low-power field (LPF) (x4 objective). At least 10 microscopic fields were counted per monolayer for each time point. Three replicates of infected and uninfected cells per experiment were performed and each experiment was repeated three times on different days.
Measurement of transepithelial electrical resistance (TEER).
Caco-2 cells were grown on 24 mm diameter semi-permeable filters of 0·4 µm pore size in Transwell units (Costar). Cells were used 1012 days post-confluence when they were fully differentiated. Bacteria were added at 103105 c.f.u. for different experiments. Infected monolayers were incubated for up to 7 days at 37 °C in a 6 % carbon dioxide humidified atmosphere. TEER was measured daily using the Millicell electrical resistance meter (Millicell ERS; Millipore) and monolayer resistance was determined by the calculation monolayer resistance minus blank resistance (Transwell without Caco-2 cells)xarea of the Transwell (4·7 cm2 for 24 mm filters)=monolayer resistance ( cm2). Blank electrical resistance values were usually 30
cm2. Cell monolayers were considered fully differentiated when showing electrical resistances of >200
cm2.
Ussing chambers.
Confluent monolayers of cells were examined by using Ussing-type chambers as described previously (Hardy et al., 1999). Short-circuit current (ISC) was monitored whilst voltage-clamping (VCC600 amplifier; Physiologic Instruments) the Caco-2 epithelia at 0 mV with the mucosal bath as ground. Transepithelial resistance (RT) was calculated using Ohm's law from voltage pulses of 1 mV for 0·35 ms. All readings were automatically corrected for electrode offsets and solution resistance and recorded online using a Powerlab/8SP (AD Instruments). The bathing solutions contained 113 mM NaCl, 4·5 mM KCl, 25 mM NaHCO3, 1·2 mM Na2HPO4, 1·1 mM CaCl2, 1·2 mM MgCl2, 10 mM glucose, pH 7·4 when gassed at 37 °C. Reduced-chloride Ringer's solution (18 mM chloride) was the same as the standard Ringer's except the NaCl was replaced with equimolar sodium gluconate and the CaCl2 was increased to 5·7 mM (to compensate for the chelating effect of gluconate on calcium). Chloride conductance of the apical membranes was measured by using a chloride gradient with reduced chloride in the mucosal bath. Permeabilization of the basolateral membrane was achieved using nystatin at 0·36 mg ml1 as described by Sheppard et al. (1993)
.
Occludin staining.
Tight junction occludin distribution was investigated by fluorescent antibody staining in infected and uninfected cells. Occludin is a tight junction protein whose functions include maintaining tight junction integrity. When occludin is disrupted, tight junction integrity is lost, electrical resistance is decreased and water may be lost from the paracellular pathway (Simonovic et al., 2000). Immunofluorescent staining was performed on both uninfected and infected monolayers. Infected monolayers were stained when TEER of the monolayer had fallen below 200
cm2, indicating that tight junction integrity was lost. Cells were grown on glass coverslips to confluence, and when differentiated they were infected for the appropriate time, then fixed with paraformaldehyde (3 %), pH 7·4, in PBS for 15 min. Cells were then rinsed and permeabilized with 0·2 % Triton-X 100 in PBS for 15 min and blocked in 1 % BSA in PBS. Monolayers were incubated with anti-occludin antibody (Chemicon; 1 : 1000 dilution) for 1 h, followed by fluorescein-conjugated anti-rabbit IgG (1 : 1000) antibody for 1 h. Monolayers were washed and mounted with Antifade reagent (Molecular Probes). Monolayers were photographed by using a Nikon digital camera on a Leica fluorescent microscope.
Host cell viability.
Host cell viability was determined using live/dead staining (Molecular Probes) and apoptosis by fluorescent terminal deoxynucleotidyl transferase-mediated dUTP nick end-labelling (TUNEL) reaction. DMEM pH was measured for both uninfected and infected monolayers for each time point.
![]() |
RESULTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
|
|
To confirm the results obtained by using the Millicell electrical resistance meter, TEER measurements were also calculated from Caco-2 monolayers mounted in Ussing chambers under short-circuit conditions (Fig. 5). The TEER of the Caco-2 monolayers infected initially with lower numbers of C. jejuni (1000 c.f.u.) was again effectively abolished in monolayers infected with C. jejuni 11168 for 5 days [C. jejuni 11168, (mean) 52
cm2; C. jejuni L115, (mean) 90
cm2; uninfected controls, (mean) 182
cm2, n=6]. Again, the fall in TEER coincided with loss of domes from the infected monolayers. Mutations in cytolethal distending toxin (cdtA) and the proposed haemolysin/regulator/methylase (tlyA) had little effect on TEER in infected cells, indicating no role for these particular mutants in TEER collapse.
|
TEER of infected Caco-2 cells is affected by both extracellular and intracellular bacteria but not intracellular bacteria alone
The fall in TEER of C. jejuni-infected Caco-2 monolayers after 5 days was prevented by exposure of the monolayers to gentamicin for 4 h, applied 24 h after infection of the monolayers (data not shown). This was true for strains 11168, G1, L115 and 81-176. These findings show that intracellular organisms alone were unable to alter TEER. Representative monolayers were lysed after gentamicin treatment to determine if bacteria were intracellular in these assays. Experiments yielded intracellular C. jejuni in numbers similar to those described previously for these strains (Everest et al., 1992).
The tight junction protein occludin is rearranged in infected cells
Fig. 6 demonstrates changes in the cellular distribution of occludin. These changes were observed in Caco-2 cell monolayers infected with C. jejuni at the time points when TEER was reduced but not before.
|
Role of putative bacterial virulence factors in dome collapse and tight junction integrity
All mutants tested, with one exception, behaved like wild-type isolates in our assays. Thus, we have been unable to find a role for a defined bacterial molecule causing the effects we describe. The exception was C. jejuni strain NCTC 12189, a genetically undefined mutant, where infection of doming monolayers showed no collapse of domes, loss of TEER or occludin changes (data not shown).
![]() |
DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
One previous study (Bras & Ketley, 1999) has shown no disruption of Caco-2 cell tight junction integrity by C. jejuni, as measured by electrical resistance at early time points (2, 4, 6, 8, 10 h post-infection compared to Salmonella sp., which disrupt tight junction integrity by a fall in electrical resistance as early as 2 h post-infection). However, in the same publication, Bras & Ketley (1999)
showed that C. jejuni can decrease the electrical resistance of Caco-2 cells 24 h post-infection using an inoculum of 12x109 ml1. Our data confirm these findings in that for an initial large inoculum, dome formation was lost 24 h post-infection, which was found to correlate with loss of TEER for infected but not uninfected monolayers. For the results presented in both this paper and that of Bras & Ketley (1999)
, bacteria are present extracellularly (interacting with the apical cell surface) inside the cells (after bacterial invasion) and have passed through or between cells (transcytosis), indicating the importance of extracellular bacteria for disruption of TEER. The time taken for C. jejuni to disrupt cellular tight junctions in vitro is longer than that observed for Salmonella typhimurium (Jepson et al., 1995
) and enteropathogenic Escherichia coli (Canil et al., 1993
; Berkes et al., 2003
). The extended time span may reflect the conditions of a tissue culture system (cultured in 6 % carbon dioxide rather than microaerophilic conditions), which will favour eukaryotic cell survival and growth and may be suboptimal for the organism being studied. C. jejuni takes longer to replicate under the assay conditions we employed and therefore host cellbacteria interaction is delayed, accounting for the longer time points at which the organism exerts a biological effect on epithelial cell function. The small numbers of bacteria used initially (1000 organisms), which then replicate over subsequent days, extend the time for the biological effects of the infection to occur; hence experimental conditions are monitored over 57 days for each experiment. However, using a larger initial inoculum reduces the time for dome collapse and fall in TEER, indicating that a threshold number of organisms must be reached for these phenomena to occur (Bras & Ketley, 1999
). The finding that C. jejuni takes longer to increase the permeability of the paracellular pathway compared with other enteropathogens should not detract from the fact that C. jejuni can alter this important aspect of cell physiology, with all the resulting implications that these events may play an important role in the pathogenesis of diarrhoea caused by these organisms.
C. jejuni strain NCTC 12189, an aflagellate, non-motile spontaneous laboratory mutant (Dolby & Newell, 1985), was the only strain found that did not collapse domes or decrease TEER over the experimental time course (data not shown). Thus, we reasoned that flagella may play a role in the process of dome collapse, perhaps via the secreted Cia effector proteins (Konkel et al., 1999a
, b
, 2001
) previously shown to be actively inserted inside infected cells via flagella. However, the defined flagella mutants flhF and fliD were able to collapse domes in these assays as effectively as the wild-type parent (data not shown). NCTC 12189 is also exquisitely sensitive to antibiotics, suggesting additional physiological defects, so we suspect this strain may also harbour other mutations, which may be responsible for its performance in these experiments. We examined a number of other mutants defective in various bacterial structural and secreted molecules in our assay system but these acted in a manner indistinguishable from the wild-type in our experiments. It has been postulated that CdtA and TlyA are possibly involved in the pathogenesis of diarrhoeal disease due to C. jejuni, but in our assays these mutants behaved like the wild-type clinical isolates. The kpsM mutant has previously been shown to be less virulent in a ferret diarrhoeal model (Bacon et al., 2001
) but behaves like the wild-type strain in our studies. The kpsM mutant is killed by serum and as it lacks a capsule it may be more efficiently killed by phagocytes. So the kpsM mutant may not be so readily able to cause diarrhoea in the ferret model because it is killed by the environment of the host intestinal tract and hence less able to colonize the epithelium. Our epithelial cell models have no such host-selective pressures and hence we would never see such an effect in our system.
The observation that extracellular bacteria seem to be important for the collapse of domes, loss of TEER and rearrangement of occludin suggests an important role for bacteria outside cells, perhaps subverting host cell signalling by an as yet unknown mechanism. It is possible that bacterial secreted proteins from organisms outside cells influence the absorptive transport processes of enterocyte-like cells. Thus, the Cia or FlaC effectors look attractive candidates for such a role in host cell subversion.
Tight junctional integrity per se has never been determined in natural human disease due to C. jejuni. However, the histopathology of intestinal biopsies of acute disease shows intense neutrophil infiltration in the infected mucosa and neutrophils in diarrhoeal faeces (Skirrow & Blaser, 2000). For these neutrophils to be present in mucosa and faeces they presumably have had to traverse intestinal epithelial tight junctions in response to bacterial invasion, which in turn must lose their tight junctional integrity in order to allow neutrophils to migrate through the epithelial monolayer. Using a macaque model of C. jejuni diarrhoeal disease, Russell et al. (1993)
demonstrated colonic damage with intercellular junctions widened by electron microscopy. As previously mentioned, this may be a consequence of infiltrating neutrophils or the direct effects of the organism on the mucosa.
Our data suggest that C. jejuni inhibits absorption in infected Caco-2 cells, as we could detect no secretory activity in infected cell monolayers. If the tight junctional integrity of the intestinal epithelium is lost, electrolyte and fluid absorption are likely to be compromised. If these observations are manifested in vivo it is likely that they contribute to the clinical manifestations of diarrhoea. The in vitro experimental system employed here could provide a useful model for investigation of how C. jejuni effector molecules mediate this process.
![]() |
ACKNOWLEDGEMENTS |
---|
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Berkes, J., Viswanathan, V. K., Savkovic, S. D. & Hecht, G. (2003). Intestinal epithelial responses to enteric pathogens: effects on the tight junction barrier, ion transport, and inflammation. Gut 52, 43951.
Black, R. E., Levine, M. M., Clements, M. L., Hughes, T. P. & Blaser, M. J. (1988). Experimental Campylobacter jejuni infection in humans. J Infect Dis 157, 472479.[Medline]
Bras, A. M. & Ketley, J. M. (1999). Transcellular translocation of Campylobacter jejuni across human polarised epithelial cell monolayers. FEMS Microbiol Lett 179, 209215.[CrossRef][Medline]
Canil, C., Rosenshine, I., Ruschkowski, S., Donnenberg, M. S., Kaper, J. B. & Finlay, B. B. (1993). Enteropathogenic Escherichia coli decreases the transepithelial electrical resistance of polarized epithelial monolayers. Infect Immun 61, 27552762.[Abstract]
Coker, A. O., Isokpehi, R. D., Thomas, B. N., Amisu, K. O. & Obi, C. L. (2002). Human Campylobacteriosis in developing countries. Emerg Infect Dis 8, 237243.[Medline]
Delie, F. & Rubas, W. (1997). A human colonic cell line sharing similarities with enterocytes as a model to examine oral absorption: advantages and limitations of the Caco-2 model. Crit Rev Ther Drug Carrier Syst 14, 221286.[Medline]
Dolby, J. M. & Newell, D. G. (1986). The protection of infant mice from colonization with Campylobacter jejuni by vaccination of the dams. J Hyg 96, 143151.
Everest, P. H., Goossens, H., Butzler, J. P., Lloyd, D., Knutton, S., Ketley, J. M. & Williams, P. H. (1992). Differentiated Caco-2 cells as a model for enteric invasion by Campylobacter jejuni and Campylobacter coli. J Med Microbiol 37, 319325.[Abstract]
Everest, P. H., Goossens, H., Sibbons, P., Lloyd, D. R., Knutton, S., Leece, R., Ketley, J. M. & Williams, P. H. (1993). Pathological changes in the rabbit ileal loop model caused by Campylobacter jejuni from human colitis. J Med Microbiol 38, 316321.[Abstract]
Hardy, S. P., Denmead, M., Parekh, N. & Granum, P. E. (1999). Cationic currents induced by Clostridium perfringens type A enterotoxin in human intestinal Caco-2 cells. J Med Microbiol 48, 235243.[Abstract]
Harvey, P., Battle, T. & Leach, S. (1999). Different invasion phenotypes of Campylobacter isolates in Caco-2 cell monolayers. J Med Microbiol 48, 461469.[Abstract]
Jepson, M. A., Collares-Buzato, C. B., Clark, M. A., Hirst, B. H. & Simmons, N. L. (1995). Rapid disruption of epithelial barrier function by Salmonella typhimurium is associated with structural modification of intercellular junctions. Infect Immun 63, 356359.[Abstract]
Karlyshev, A. V., Linton, D., Gregson, N. A., Lastovica, A. J. & Wren, B. W. (2000). Genetic and biochemical evidence of a Campylobacter jejuni capsular polysaccharide that accounts for Penner serotype specificity. Mol Microbiol 35, 529541.[CrossRef][Medline]
Karlyshev, A. V., Everest, P., Linton, D., Cawthraw, S., Newell, D. G. & Wren, B. W. (2004). The Campylobacter jejuni general glycosylation system is important for attachment to human epithelial cells and in the colonization of chicks. Microbiology 150, 19571964.[CrossRef][Medline]
Konkel, M. E., Mead, D. J., Hayes, S. F. & Cieplak, W. Jr (1992). Translocation of Campylobacter jejuni across human polarized epithelial cell monolayer cultures. J Infect Dis 166, 308315.[Medline]
Konkel, M. E., Kim, B. J., Rivera-Amill, V. & Garvis, S. G. (1999a). Bacterial secreted proteins are required for the internalization of Campylobacter jejuni into cultured mammalian cells. Mol Microbiol 32, 691701.[CrossRef][Medline]
Konkel, M. E., Kim, B. J., Rivera-Amill, V. & Garvis, S. G. (1999b). Bacterial secreted proteins are required for the internalization of Campylobacter jejuni into cultured mammalian cells. Mol Microbiol 32, 691701.[CrossRef][Medline]
Konkel, M. E., Monteville, M. R., Rivera-Amill, V. & Joens, L. A. (2001). The pathogenesis of Campylobacter jejuni-mediated enteritis. Curr Issues Intest Microbiol 2, 5571.[Medline]
Larsen, J. C., Szymanski, C. & Guerry, P. (2004). N-linked protein glycosylation is required for full competence in Campylobacter jejuni 81-176. J Bacteriol 186, 65086514.
Linton, D., Allan, E., Karlyshev, A. V., Cronshaw, A. D. & Wren, B. W. (2002). Identification of N-acetylgalactosamine-containing glycoproteins PEB3 and CgpA in Campylobacter jejuni. Mol Microbiol 43, 497508.[CrossRef][Medline]
Martino, M. C., Stabler, R. A., Zhang, Z. W., Farthing, M. J., Wren, B. W. & Dorrell, N. (2001). Helicobacter pylori pore-forming cytolysin orthologue TlyA possesses in vitro hemolytic activity and has a role in colonization of the gastric mucosa. Infect Immun 69, 16971703.
Newell, D. G., McBride, H. & Dolby, J. M. (1985). Investigations on the role of flagella in the colonization of infant mice with Campylobacter jejuni and attachment of Campylobacter jejuni to human epithelial cell lines. J Hyg 95, 217227.
Parkhill, J., Wren, B. W., Mungall, K. & 18 other authors (2000). The genome sequence of the food-borne pathogen Campylobacter jejuni reveals hypervariable sequences. Nature 403, 665668.[CrossRef][Medline]
Pei, Z. H., Ellison, R. T. III & Blaser, M. J. (1991). Identification, purification, and characterization of major antigenic proteins of Campylobacter jejuni. J Biol Chem 266, 1636316369.
Russell, R. G., O'Donnoghue, M., Blake, D. C. Jr, Zulty, J. & DeTolla, L. J. (1993). Early colonic damage and invasion of Campylobacter jejuni in experimentally challenged infant Macaca mulatta. J Infect Dis 168, 210215.[Medline]
Sheppard, D. N., Rich, D. P., Ostegaard, L. S., Gregory, R. J., Smith, A. E. & Welsh, M. J. (1993). Mutations in CFTR associated with mild disease form Cl channels with altered pore properties. Nature 362, 160164.[CrossRef][Medline]
Simonovic, I., Rosenberg, J., Koutsouris, A. & Hecht, G. (2000). Enteropathogenic Escherichia coli dephosphorylates and dissociates occludin from intestinal epithelial tight junctions. Cell Microbiol 2, 305315.[CrossRef][Medline]
Skirrow, M. & Blaser, M. (2000). Clinical aspects of Campylobacter infection. In Campylobacter, pp. 6888. Edited by I. Nachamkin & M. J. Blaser. Washington, DC: American Society for Microbiology.
Wassenaar, T. M. & Blaser, M. J. (1999). Pathophysiology of Campylobacter jejuni infections of humans. Microbes Infect 1, 10231033.[CrossRef][Medline]
Zhang, Z. W., Dorrell, N., Wren, B. W. & Farthingt, M. J. (2002). Helicobacter pylori adherence to gastric epithelial cells: a role for non-adhesin virulence genes. J Med Microbiol 51, 495502.
Received 7 February 2005;
revised 1 April 2005;
accepted 20 April 2005.
HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
INT J SYST EVOL MICROBIOL | MICROBIOLOGY | J GEN VIROL |
J MED MICROBIOL | ALL SGM JOURNALS |