Adherent-invasive Escherichia coli, strain LF82 disrupts apical junctional complexes in polarized epithelia
© Wine et al. 2009
Received: 14 April 2009
Accepted: 26 August 2009
Published: 26 August 2009
Skip to main content
© Wine et al. 2009
Received: 14 April 2009
Accepted: 26 August 2009
Published: 26 August 2009
Although bacteria are implicated in the pathogenesis of chronic inflammatory bowel diseases (IBD), mechanisms of intestinal injury and immune activation remain unclear. Identification of adherent-invasive Escherichia coli (AIEC) strains in IBD patients offers an opportunity to characterize the pathogenesis of microbial-induced intestinal inflammation in IBD. Previous studies have focused on the invasive phenotype of AIEC and the ability to replicate and survive in phagocytes. However, the precise mechanisms by which these newly identified microbes penetrate the epithelial lining remain to be clarified. Therefore, the aim of this study was to delineate the effects of AIEC, strain LF82 (serotype O83:H1) on model polarized epithelial monolayers as a contributor to intestinal injury in IBD.
Infection of T84 and Madin-Darby Canine Kidney-I polarized epithelial cell monolayers with AIEC, strain LF82 led to a reduction in transepithelial electrical resistance and increased macromolecular (10 kilodalton dextran) flux. Basolateral AIEC infection resulted in more severe disruption of the epithelial barrier. Increased permeability was accompanied by a redistribution of the tight junction adaptor protein, zonula occludens-1, demonstrated by confocal microscopy and formation of gaps between cells, as shown by transmission electron microscopy. After 4 h of infection of intestine 407 cells, bacteria replicated in the cell cytoplasm and were enclosed in membrane-bound vesicles positive for the late endosomal marker, LAMP1.
These findings indicate that AIEC, strain LF82 disrupts the integrity of the polarized epithelial cell barrier. This disruption enables bacteria to penetrate into the epithelium and replicate in the host cell cytoplasm. These findings provide important links between microbes related to IBD, the intestinal epithelial cell barrier and disease pathogenesis.
The inflammatory bowel diseases (IBD), Crohn disease and ulcerative colitis, are relatively common chronic disorders considered to develop due to an aberrant immune response to intestinal microbes in a genetically susceptible host . Human data and murine models both implicate the involvement of luminal bacteria in IBD pathogenesis. For example, inflammation is induced by direct delivery of fecal material into non-inflamed bowel loops in susceptible individuals  and diversion of feces results in distal improvement in mucosal inflammation . In addition, most of the genes associated with susceptibility to IBD, including NOD2/CARD15, Atg16L1 and IRGM encode proteins involved in host-microbial interactions . Further support for the involvement of microbes in the pathogenesis of IBD is based on the observation that colitis does not occur in most gene knock-out models of IBD when animals are reared in germ-free conditions [5, 6].
Recent advances in molecular techniques have identified a reduction in the phyla Firmicutes and Bacteroidetes in IBD patients . Although several organisms have been proposed as a cause of IBD, there is still no compelling evidence that any one specific microbe is the etiologic agent. Mycobacterium avium subspecies paratuberculosis has been suggested by some investigators , although this remains an area of ongoing controversy . Most recently, absence of Faecalibacterium prausnitzii from the ileum of patients with Crohn disease undergoing surgical resection was associated with recurrence of disease, suggesting a protective role for this commensal organism .
Observations linking IBD to an increase in adherent Escherichia coli strains have also been recognized over the past decade . Invasive properties of some of these isolates, including E. coli strain LF82 (serotype O83:H1), led to the proposition that adherent-invasive E. coli strains (also termed AIEC) are involved in disease pathogenesis . Such an association is supported by the isolation of AIEC from 36% of ileal lesions in post-surgical resection Crohn disease patients, compared to just 6% of healthy controls , accompanied by increased prevalence and diversity of AIEC strains in patients with Crohn disease . Although some of the mechanisms by which these bacteria lead to colonization and intestinal injury, such as induction of carcinoembryonic antigen-related cell-adhesion molecule (CEACAM)-6 receptor expression by TNF-α , have been well characterized, other virulence traits remain to be determined.
Defects in the structure and function of apical junctional complexes (AJCs) are implicated in both patients with IBD and in animal models of IBD [16, 17]. In this context, the adverse effects of microbes on intercellular junctions offer potential bridges connecting bacteria to the pathogenesis of IBD. Barrier dysfunction precedes the relapse of Crohn disease in asymptomatic patients  and is also seen in unaffected first-degree relatives, who are at increased risk of subsequently developing the illness . Recent studies demonstrate specific distribution patterns of the tight junction proteins claudin 2, 3, 4, 5, & 8 in IBD patients, which correlate with increased gut permeability [20, 21]. For these reasons, the aim of this study was to define the ability of AIEC strain LF82 to disrupt model epithelial cell polarized monolayers. We describe herein increased permeability of polarized epithelia infected with AIEC as well as morphologic disruption of apical junction complexes.
T84 and Madin-Darby Canine Kidney (MDCK)-I cells are polarized epithelial cells that form AJCs, resulting in high electrical resistance, and are widely used for studying the effects of bacteria on permeability [22, 23]. T84 human colon cancer epithelial cells were cultured in Dulbecco's minimal essential medium (DMEM)/F-12, 10% heat-inactivated fetal bovine serum (FBS), 2% penicillin-streptomycin, 2% sodium bicarbonate and 0.6% L-glutamine. MDCK-I cells were grown in DMEM, 10% FBS and 2% penicillin-streptomycin (all from Gibco, Grand Island, NY). Cells were maintained in 25 cm2 flasks (Corning Glass Works, Corning, NY) and then grown on 12-well Transwells (6.5 mm diameter; 3 μm pore size; 37°C; 5% CO2; Corning) or 24-well plates (Corning). Non-polarized intestine 407 (human fetal intestine) cells were cultivated in Minimal Essential Medium (MEM), 10% FBS and 2% penicillin-streptomycin (Gibco).
Enterohemorrhagic E. coli (EHEC), strain CL56 serotype O157:H7 , non-pathogenic E. coli, laboratory strain HB101, used as a negative control, and adherent-invasive E. coli (AIEC), strain LF82 serotype O83:H1, a generous gift from Dr. Darfeuille-Michaud (Université d'Auvergne, Clermont-Ferrand, France)  were stored at -80°C and re-grown on 5% sheep blood agar plates at 37°C. Colonies were transferred from plates into Penassay broth and incubated at 37°C for 18 h, and re-grown in 10:1 fresh Penassay broth (3 h; 37°C). MultipliCity of infection (MOI) used for all experiments was 100:1. To determine whether live bacteria were required for the observed effects, bacterial suspensions were either boiled at 100°C for 30 min or fixed with formaldehyde for 6 h prior to infection of cell monolayers.
MDCK-I and T84 cells were plated onto Transwells (5 × 104 or 2 × 105 cells/well, respectively; 6.5 mm diameter; 0.4 μm-pore size; Corning) and grown until AJCs developed (as indicated by a TER > 1,000 Ω·cm2). Twenty four hours prior to infection the tissue culture medium was removed and fresh medium without antibiotics, but with FBS, was added. FBS was maintained throughout the infection period. Transwells were then infected with either EHEC O157:H7, E. coli HB101 or AIEC (MOI: 100:1; 37°C; 5% CO2) introduced either to the apical or basolateral aspect of the Transwell. Sham control monolayers were treated in an identical fashion, excluding the addition of bacteria. TER was measured prior to and 16 h after infection, using a Millicell-ERS Voltmeter and chopstick electrodes (Millpore, Bedford, MA). TER of Transwells without cells was 32 Ω·cm2. Results are expressed as a percentage, relative to sham control wells.
Dextran flux was used to measure paracellular macromolecular permeability . After 16 h of infection, monolayers were washed four times with phosphate-buffered saline (PBS) and infrared-labeled dextran (10-kDa; 0.2 ml of 0.1 mg/ml in DMEM; Alexa-Fluor 647, Molecular Probes, Eugene, OR) was then inserted into the apical compartment of Transwells. After 5 h at 37°C, the basal compartment was sampled, diluted 1:20, and loaded into 96-well plates for infrared signal quantification using an imaging system at 700 nm (Odyssey®, Licor, Rockford, IL). Integrated intensities were expressed relative to sham control polarized monolayers.
For ZO-1 staining, MDCK-I cell monolayers were grown to confluence (TER >1,000 Ω·cm2) on 6.5 mm Transwells and then infected with AIEC, strain LF82 at a MOI of 100:1 for 16 h at 37°C. Monolayers were then washed 4 times with PBS and fixed with 100% cold methanol for 10 min at 4°C, blocked with 5% skim milk (Santa Cruz Biotechnology, Santa Cruz, CA) for 1 h at room temperature and then incubated with primary rabbit anti-ZO-1 antibodies (1:50, Zymed, Burlington, Ontario, Canada) for 1 h at room temperature. After rinsing 3 times for 10 min with PBS, cell monolayers were incubated with secondary antibodies, Cy2-goat anti-rabbit (1:200, Zymed), for 1 h at 20°C. After two further washes, 300 nM of 4',6-diamidino-2-phenylindole (DAPI, 1:36,000, Invitrogen, Eugene, ON) was added for 5 min, and rinsed off twice. Membranes supporting the monolayers were then excised and mounted onto glass slides (using DakoCytomation Mounting Medium, Carpentaria, CA).
For LAMP1 staining, intestine 407 cells were grown on glass cover slips in 24-well plates overnight and then either left uninfected or infected with AIEC, strain LF82 for 4 h at 37°C (MOI 100:1). Wells were washed 3 times with PBS (pH 7.0) and fixed with 4% paraformaldehyde in PBS for 20 min at 20°C. Wells were then washed with PBS and permeabilized with Triton-X 100 (0.1% in PBS; 20 min at 20°C) and blocked overnight with 5% skim milk (Santa Cruz) at 4°C. Wells were incubated with mouse monoclonal anti-LAMP1 antibodies (1 in 1,000 dilution; Developmental Studies Hybridoma Bank, Iowa City, IA) for 1 h at 20°C, washed 5 times in PBS and then incubated with secondary antibody, Cy3-goat anti-mouse (1:100, Zymed) for 1 h at 20°C. DAPI staining was performed, as detailed above, and coverslips mounted onto glass slides.
All samples were examined using a Leica DMIRE2 Quorum spinning disk confocal scan head inverted fluorescence microscope (Wetzlar, Germany), equipped with a Hamamatsu Back-Thinned EM-CCD camera (Hamamatsu, Japan), at 63× objective. Images were acquired and analyzed using VeloCity 3.7.0 acquisition software (Improvision, Coventry, England).
Confluent MDCK-I Transwells were left uninfected or infected with AIEC, strain LF82 (MOI: 100:1; 4 h or 48 h; 37°C). Support membranes were washed, excised and cells fixed in formaldehyde (4%) and glutaraldehyde (1%) in phosphate buffer, and then post-fixed in osmium tetroxide (1%; 2 h; 20°C). Specimens were dehydrated in a graded series of acetone, and subsequently infiltrated and embedded in Epon-Araldite epoxy resin. The processing steps from post fixation to polymerization of resin blocks were carried out in a microwave oven (Pelco BioWave 34770, Pelco International, Redding, CA). Ultrathin sections were cut with a diamond knife (Reichert Ultracut E, Leica Inc., Wetzlar, Germany), stained with uranyl acetate and lead citrate and then examined by transmission electron microscopy (JEM-1011, JEOL USA Corp., Peabody, MA) at 75 kV. Digital electron micrographs were acquired directly with a 1024 × 1024 pixels CCD camera system (AMT Corp., Denver, MA).
Results are expressed as means ± SEM. N represents the number of individual experiments. Replicates within experiments are expressed as a mean for a single experiment. ANOVA and unpaired Student's t-test were conducted using InStat3 (GraphPad, San Diego, CA). Means were compared using ANOVA and Tukey's post-hoc test.
Apical and basolateral infections of canine kidney-derived MDCK-I polarized monolayers with EHEC and AIEC caused a comparable reduction of 53–73% in TER (Figure 1B; ANOVA: p < 0.01). Live bacteria were required, because there was no drop in TER with either heat-inactivated or formaldehyde-fixed bacteria (Figure 1B). The effects were not due to the metabolic activity of bacteria on epithelial cells, since incubation with tissue culture medium corrected to pH 6 (the pH of medium after 16 h of infection) did not reduce TER (N = 2).
Sham control MDCK-I cells (Figure 2A) demonstrated a normal distribution of ZO-1, delineating intact apical cellular junction complexes . Consistent with effects on permeability, 16 h infection of MDCK-I monolayers with AIEC, strain LF82 (Figure 2B) led to profound disruption of ZO-1 with large gaps between cells with punctate and interrupted distribution of ZO-1, indicating disruption of this integral tight junction protein . Nevertheless, cells in the monolayer remained viable, as demonstrated by the presence of nuclei and maintenance of normal cells shape and morphology.
The ability of invasive microbes to survive in cells is dependent on creating a protective niche for replication . Invasive AIEC were found in membrane-bound compartments 4 h after infection (Figure 3D). Presence of multiple organisms in one compartment suggests that they can effectively replicate within these vacuoles. Since the membrane appeared to be partially missing, it is possible that bacteria were escaping the vacuole.
The intestinal barrier is comprised of a single layer of polarized epithelial cells serving to separate the luminal content, including microbes, from the underlying mucosa. Breaches in the epithelial barrier integrity result in penetration of luminal antigens and microbes, which stimulate pro-inflammatory responses, leading to chronic intestinal and systemic diseases, including IBD . The importance of barrier maintenance in IBD is further highlighted by the development of colitis in mice expressing constitutively active myosin light chain kinase, which is involved in regulating the epithelial barrier . AJCs are common targets of bacterial virulence, as displayed by multiple infection models affecting the integrity of the epithelial barrier . Targeting of AJCs can be mediated by either bacterial toxins or effectors, by direct contact of the pathogen and by indirect effects on signaling pathways involved in host regulation of junction-associated proteins.
Recent research related to microbes isolated from patients with Crohn disease highlight a role for adherent-invasive E. coli strains, including strain LF82, in the pathogenesis of IBD . Studies have focused on bacterial adhesion, invasion and replication in both epithelial cells and macrophages, as well as the accompanying inflammatory response . For example, epithelial surface adhesions, such as CEACAMs, mediate attachment of various bacterial pathogens . The association between LF82 and Crohn disease is linked to up-regulation of CEACAM5 and CEACAM6 by intestinal epithelial cells, which is induced by LF82 through TNF-α secretion . Nevertheless, the ability of these microbes to disrupt the integrity of the epithelial barrier has not been extensively studied. Only a single published study describes AIEC-induced barrier disruption of Caco-2 cells . Therefore, we employed transformed human colonic T84 cells and canine kidney MDCK-I cells as model polarized epithelia, which both express mature apical junctional complex proteins and maintain cell polarity , and are used extensively to study host-microbial interactions [22, 36]. Furthermore, the utility of polarized epithelial monolayers in the study of AIEC infection was recently reported by Eaves-Pyles et al.  that demonstrated chemokine secretion by AIEC-infected Caco-2 and T84 monolayers leading to transmigration of immune cells.
Our findings indicate that infection of polarized monolayers with AIEC, strain LF82 leads to disruption of epithelial cell monolayers, as demonstrated by both reduced transepithelial electrical resistance and increased macromolecular permeability, as well as morphological defects in the structure of the AJCs of infected monolayers. The ability of invasive bacteria to disrupt monolayer integrity is described for some intestinal pathogens, such as Shigella flexneri, Listeria monocytogenes  and Campylobacter jejuni , while other bacteria, such as Helicobacter pylori, appear to alter AJCs without entering into the cytoplasm . Since AIEC strains are associated with IBD, host cell invasion and barrier disruption, as presented in this study, are mechanisms that could contribute to intestinal injury and immune stimulation in affected patients.
Sasaki et al.  demonstrated the ability of LF82, as well as other AIEC strains, to reduce TER of Caco-2 monolayers and displace both ZO-1 and E-cadherin from AJCs. Our results confirm these findings in additional polarized epithelial cell lines and also reveal an increase in macromolecular permeability of infected monolayers. In addition, we show that introducing bacteria to the basolateral surface of T84 monolayers leads to a more profound reduction in TER. The significance of this finding is highlighted by the suggestion that other enteric pathogens, such as C. jejuni, enter epithelial cells through the basolateral membrane .
Our findings show that AIEC can replicate in membrane-bound vesicles, which positively stain with the late endosomal marker LAMP1. Similar to these findings, previous work suggests that strain LF82 is present in vacuoles in epithelial cells after invasion, but is also seen in the cytoplasm, suggesting that these bacteria can escape from the vacuoles . Nevertheless, the phagocytic pathway involved in AIEC invasion of epithelial cells has not been characterized. Similar to our findings in epithelial cells, LF82 co-localizes with LAMP1 in infected macrophages , suggesting that there are common features in the intracellular fate of these microorganisms in different cell types.
The ability of AIEC to survive and replicate within the cytoplasm of epithelial cells is of relevance in IBD, since defects in the handling of intercellular microbes are considered to contribute to disease pathogenesis . For example, absence of NOD2 in transgenic mice results in increased susceptibility to infection with intracellular pathogens, such as Mycobacterium tuberculosis . Furthermore, the autophagy protein Atg16L1, which is also implicated in the pathogenesis of IBD , is involved in inflammatory responses to invasive microbes. Mice lacking Atg16L1 are more susceptible to chemically-induced colitis than wild-type animals subject of the same stress . Therefore, it is plausible that defective handling of invasive AIEC strains in patients with IBD who have genetic mutations linked to defects in microbial processing contributes to intestinal injury, as suggested by increased response of monocytes from Crohn disease patients with NOD2 mutations to AIEC infection in vitro . The findings of our study support the ability of AIEC to subvert one of the first lines of host innate defence, the epithelial cell barrier.
Taken together, these findings provide an improved understanding of mechanisms leading to intestinal injury and chronic immune stimulation by an AIEC bacterial strain that has been linked to IBD pathogenesis. Further insight into the mechanisms of epithelial barrier disruption and subversion of host defenses by intestinal pathogens is essential for developing novel strategies to interrupt the infectious process and thereby prevent its complications, including IBD.
The invasive E. coli strain LF82, which is linked to IBD, disrupts AJCs of polarized epithelial monolayers and leads to increased macromolecular permeability and morphological interruption of intercellular tight junctions. After invasion into epithelial cells, the bacteria replicate within late endosomes. These findings contribute to current understanding of bacterial-mediated processes related to the pathogenesis of IBD and offer potential targets for intervening early in the course of the disease process.
adherent-invasive Escherichia coli
apical junctional complexes
carcinoembryonic antigen-related cell-adhesion molecules
Dulbecco's minimal essential medium
Enterohemorrhagic Escherichia coli
fetal bovine serum
inflammatory bowel diseases
lysosomal-associated membrane protein
Madin-Darby canine kidney
multipliCity of infection
transepithelial electrical resistance
This work was supported by operating grants from the Crohn's and Colitis Foundation of Canada (Fay Shapiro Cutler Grant in Aid of Research) and the Canadian Institutes of Health Research (CIHR). EW is supported by a fellowship award from the Canadian Association of Gastroenterology/CIHR/Astra Zeneca. PMS is the recipient of a Canada Research Chair in Gastrointestinal Disease.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.