- Email: [email protected]
Caspases-3, -8, and -9 are required for induction of epithelial cell apoptosis by enteropathogenic E. coli but are dispensable for increased paracellular permeability
ARTICLE IN PRESS
MICROBIAL PATHOGENESIS Microbial Pathogenesis 44 (2008) 311–319 www.elsevier.com/locate/micpath
Caspases-3, -8, and -9 are required for induction of epithelial cell apoptosis by enteropathogenic E. coli but are dispensable for increased paracellular permeability Andrew N. Flynn, Andre G. Buret Inflammation Research Network, University of Calgary, 2500 University Drive NW, Calgary, Alta., Canada T2N 1N4 Received 25 August 2007; received in revised form 10 October 2007; accepted 11 October 2007 Available online 24 October 2007
Abstract Enteropathogenic Escherichia coli (EPEC) is an important cause of diarrhea, particularly among infants in developing countries. An increase in intestinal permeability due to EPEC infection has been suggested as a factor in the development of diarrhea. Abnormally high levels of programmed cell death (apoptosis) of intestinal epithelial cells can lead to increased intestinal permeability. The effects of EPEC on cell apoptosis remain incompletely understood. This study characterized the mechanisms of EPEC-induced epithelial apoptosis and examined whether this effect contributes to heightened permeability in an in vitro model of infection. We report that EPEC-induced apoptosis in T84 intestinal epithelial cells via a mechanism involving caspases-3, -6, -8, and -9, the cleavage of PARP, and oligonucleosome formation. In addition, EPEC time-dependently increased paracellular permeability as assessed by transepithelial resistance and the apical-to-basolateral movement of 3000 MW dextran. Furthermore, EPEC infection led to the cleavage and mislocalization of tight junctional ZO-1 and occludin. However, pharmacological inhibition of caspases did not prevent the EPECinduced disruptions in epithelial barrier structure and function. Taken together, these results suggest that a caspase-dependent upregulation in epithelial cell apoptosis during EPEC infection occurs independent of impaired intestinal barrier function. Crown Copyright r 2007 Published by Elsevier Ltd. All rights reserved. Keywords: Tight junction; Intestinal permeability; Programmed cell death
1. Introduction Enteropathogenic Escherichia coli (EPEC) is an important cause of infant diarrhea in developing countries, and is associated with significant levels of mortality [1]. Infection with EPEC is characterized by intimate attachment of the bacteria to intestinal epithelial cells (IECs). This leads to the formation of an actin-rich host cell pedestal upon which the bacterium rests, and the effacement of the surrounding brush border. The results of these effects are termed attaching and effacing (A/E) lesions. A/E lesions are the direct result of the secretion of EPEC effector proteins into the host cell cytosol via a type three secretion system (TTSS). The injection of these proteins by EPEC leads to diverse phenotypic changes in cultured Corresponding author. Tel.: +1 403 220 2817; fax: +1 403 289 9311.
E-mail address: [email protected] (A.G. Buret).
IECs, including loss of polarity [2], altered ion transport [3], activation of apoptosis [4,5], and loss of barrier function [3,6]. The epithelial layer of the gastrointestinal tract represents an integral barrier between the internal environment of the body and potentially antigenic constituents of the lumen. Paracellular migration of these constituents is prevented by the apical junctional complex (AJC), which consists of transmembrane proteins, including occludin and the claudin family, and of cytoplasmic plaque proteins such as zonula occludens (ZO)-1 [7]. In concert, these proteins provide a physical barrier linking cells together. This barrier is tightly regulated to allow subtle alterations in permeability due to physiological stimuli while preventing inappropriate activation of the mucosal immune system [8]. Dramatic increases in intestinal permeability have been associated with a number of pathological conditions, including Crohn’s disease [9] and celiac disease [10].
0882-4010/$ - see front matter Crown Copyright r 2007 Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.micpath.2007.10.007
ARTICLE IN PRESS 312
A.N. Flynn, A.G. Buret / Microbial Pathogenesis 44 (2008) 311–319
Previous studies have demonstrated that in vitro infections with EPEC lead to increases in the permeability of IEC monolayers, but the mechanisms involved remain obscure [3]. This effect may contribute to the watery diarrhea that is a result of infection in humans. Although the bacterial effector proteins EspF, Map, and intimin and the host protein myosin light chain kinase (MLCK) appear to be important for increases in permeability [6,11,12], other factors await investigation. Upregulated IEC apoptosis has been found to lead to increases in monolayer permeability in various systems [13,14]. We have previously demonstrated that the protozoan parasite Giardia lamblia increases the permeability of IEC monolayers in a caspase-3-dependent manner [15]. More recently, we have shown the same effect using high doses of E. coli lipopolysaccharide [16]. Surprisingly, although EPEC may induce apoptosis in IECs [4,17], the contribution of this to increases in cell monolayer permeability has not been investigated. In addition, the pathways by which IECs undergo apoptosis when challenged with EPEC remain largely unknown. Therefore, the purposes of this study were to investigate intracellular pathways involved in EPEC-mediated IEC apoptosis, and to determine whether increases in IEC apoptosis contribute to increases in cell monolayer permeability during EPEC infection. 2. Materials and methods 2.1. Antibodies Antibodies directed against the following proteins, at the concentrations specified, were used for immunoblotting or immunocytochemistry: ZO-1, occludin (1:250; Zymed Laboratories, San Francisco, CA); PARP (1:1000; Cell Signaling Technology, Beverly, MA). Secondary antibodies conjugated to Alexa fluorophores or horseradish peroxidase were obtained from Invitrogen (1:2000; Carlsbad, CA) and Santa Cruz Biotechnology (1:10 000; Santa Cruz, CA), respectively. 2.2. Cell culture Human intestinal T84 cells (ATCC, Manassas, VA) were grown in a 1:1 mixture of DMEM and Ham’s F-12 supplemented with 5% fetal bovine serum, 100 mg/ml streptomycin, 100 U/ml penicillin, and 2 mM L-glutamine (all from Sigma, St. Louis, MO). Cells were incubated at 37 1C and 5% CO2 in 96% humidity. For all experiments, T84 cells were grown for 7 days on Lab-Tek chamber slides (Nalge Nunc International, Naperville, IL), culture-treated plates and dishes, or Transwell filter units (Costar, Corning, NY). Three hours prior to infection, monolayers were washed with HBSS followed by addition of antibioticfree medium containing 0.5% fetal bovine serum (hereafter referred to as infection medium).
2.3. Infection of monolayers Overnight cultures of EPEC strain E2348/69 (kindly provided by R. DeVinney, University of Calgary) were diluted 1:50 in LB broth and grown to early log growth phase (3 h) at 37 1C. Bacteria were then centrifuged at 1000 g for 10 min and resuspended in the appropriate volume of infection medium to give a multiplicity of infection of 100. Bacteria were added to the apical side of cell monolayers and were co-incubated for indicated lengths of time. In some experiments, T84 cells were pretreated for 1 h with 50 mM caspase inhibitors (Calbiochem, La Jolla, CA), which remained in the culture medium for the duration of the experiment. 2.4. MTT cell viability assay Viability of T84 cells was assessed using the thiazolyl blue tetrazolium bromide (MTT) method as previously described [16]. The effect of EPEC on epithelial cell viability was expressed as a percent of surviving cells compared to vehicle-treated controls. 2.5. Apoptotic nucleosome quantification T84 cell apoptosis was quantified using a Cell Death Detection ELISA kit (Roche Molecular Biochemicals, Laval, PQ) according to the manufacturer’s instructions as previously described [16]. Apoptosis was expressed as absorbance fold increase of the experimental cell lysates over controls, arbitrarily set at 1.0. 2.6. Caspase activity assays Caspase activity of T84 cells was assessed using the ApoTarget Caspase Assay (BioSource, Camarillo, CA). Briefly, cell monolayers grown in 12-well plates were washed with HBSS before being lysed with the supplied buffer. After brief sonication and centrifugation, lysates were stored at 70 1C until used. The protein content of thawed samples was normalized using the DC protein assay (Bio-Rad, Hercules, CA) before incubation with synthetic caspase substrates conjugated to p-nitroanilide. Chromophore liberation was detected at 405 nm using a SpectraMax M2e microplate reader (Molecular Devices, Sunnyvale, CA). 2.7. Lactate dehydrogenase release assay T84 cells grown in 48-well plates were infected with EPEC for 6 h. Fifty microliter samples of the culture medium were centrifuged (5 min, 10,000g) and assayed for lactate dehydrogenase (LDH) activity using the CytoTox 96 Non-Radioactive Cytotoxicity Assay according to the manufacturer’s instructions (Promega Corporation, Madison, WI). Pure cultures of EPEC did not demonstrate any LDH activity.
ARTICLE IN PRESS A.N. Flynn, A.G. Buret / Microbial Pathogenesis 44 (2008) 311–319
2.8. Whole cell lysis and subcellular fractionation For whole cell lysis, T84 cells grown on 60 mm dishes were washed with HBSS, and then lysed for 30 min in 4 1C RIPA buffer (1 PBS, 1% Igepal CA-630, 0.5% sodium deoxycholate, and 0.1% SDS) containing a proteinase inhibitor tablet (Complete Mini, Roche Diagnostics, Mannheim, Germany). Following brief sonication, lysates were cleared by centrifugation (10 min, 10,000g). For membrane extractions, cells were washed with HBSS, and then extracted with lysis buffer A (50 mM Tris–HCl, 150 mM NaCl, 0.5% Trition X-100, proteinase inhibitor tablet, pH 7.5) for 30 min at 4 1C. Lysates were then centrifuged (10 min, 10,000g), and supernatants were removed. Pellets were extracted with lysis buffer B (lysis buffer A+0.02% SDS) for 20 min at 4 1C before brief sonication and further centrifugation (10 min, 10,000g). Supernatants were collected as membranous fractions. In all cases, protein levels were normalized using DC protein assay. 2.9. Immunoblotting Protein samples added 1:1 to SDS sample buffer and boiled for 5 min were separated by SDS–PAGE and transferred to polyvinylidene difluoride membranes (Hybond-P, Amersham Biosciences, Buckinghamshire, UK). Membranes, blocked with 5% non-fat skim milk or 5% BSA, were probed overnight at 4 1C with primary antibodies followed by 45 min incubation at room temperature with secondary antibody. Bands were visualized using a chemiluminescence detection system (ECL-Plus, Amersham). The results shown are representative immunoblots of three independent experiments. 2.10. Immunocytochemistry Confluent T84 cell monolayers grown on chamber slides or Transwell filters were washed with HBSS and fixed with 20 1C methanol for 10 min. Cells were blocked with 10% normal goat serum followed by overnight incubation at 4 1C with primary antibodies diluted in Ultra Clean Diluent (Lab Vision). Secondary antibody labeling was carried out with a 45 min incubation at room temperature, after which cells were mounted with Aqua Poly/Mount (Polysciences, Warrington, PA). Cells were visualized using a Leica DMR fluorescence microscope and photomicrographs were taken using a Photometrics CoolSNAP camera (Roper Scientific, Tucson, AZ). 2.2. In vitro permeability Transepithelial resistance (TER) of T84 monolayers grown on Transwell filters was measured using an electrovoltohmeter (World Precision Instruments, Sarasota, FL). Paracellular permeability to a non-absorbable fluorescein isothiocyanate (FITC)-conjugated 3000-mole-
313
cular-weight (MW) dextran probe was assessed as previously described [18]. Values were expressed as fold increase of monolayer permeability over uninfected monolayers, arbitrarily set at 1.0. 2.3. Statistical analysis Numerical values from experiments were expressed as mean7S.E.M. throughout. Analyses were made using a one-way ANOVA with a Student–Neuman–Keuls multiple comparison test. Statistical significance was established at po0.05. 3. Results 3.1. EPEC reduces T84 cell viability and induces cell apoptosis T84 cells infected with EPEC demonstrated timedependent decreases in viability as determined by the MTT method (Fig. 1A). As the MTT method directly measures metabolically active cells, the reduction observed by 4 h represents an early loss of cell viability. In agreement with previous studies using different methods [4,5], EPECinduced apoptosis in T84 cells after an infection time of 6 h. To assess levels of apoptosis, we used both a Cell Death ELISA-based detection of oligonucleosome formation (Fig. 1B) as well as immunoblotting for cleaved PARP (Fig. 1C). Both of these techniques have been extensively validated as methods for detecting cellular apoptosis [15,16]. In further experiments regarding EPEC-induced cell apoptosis, we focused on the Cell Death ELISA due to its rapid and quantitative data generation. 3.2. EPEC activates the extrinsic and intrinsic apoptotic pathways Previous work has determined that EPEC activates the central apoptotic executioner caspase-3 in T84 cells [19]. However, investigations into the activation of upstream caspases have focused solely on caspase-9 using non-IECs [20,21]. Furthermore, the contribution of various caspases to EPEC-mediated IEC apoptosis has not been determined. We used a chromogenic-based assay with various caspase substrates to evaluate caspase activity in T84 cells. After 6 h of infection, T84 cells displayed increased activity of the executioner caspases-3 and -6 (Fig. 2(A) and (B)). Interestingly, at the same time point, both the extrinsic initiator caspase-8 and the intrinsic initiator caspase-9 were activated (Fig. 2(C) and (D)). To determine which initiator caspase was required for the activation of the executioner caspase-3, caspase-3 activity was monitored in infected cells pretreated with various caspase inhibitors. Interestingly, the inhibition of either caspase-8 or -9 completely prevented caspase-3 activation in EPEC-infected T84 cells (Fig. 3A). As expected, the pan-caspase inhibitor Z-VAD ablated caspase-3 activity. We then used the Cell Death
ARTICLE IN PRESS 314
A.N. Flynn, A.G. Buret / Microbial Pathogenesis 44 (2008) 311–319
contribute to EPEC-induced IEC apoptosis. The specificity of caspase inhibitors was verified using staurosporine as an apoptotic stimulus in conjunction with caspase activity assays (data not shown). To ensure that caspase inhibition did not lead to alternate forms of cell death (e.g., necrosis or autophagy), EPEC-infected T84 cells were assessed for LDH release. In agreement with previous studies [6,22], EPEC did not induce the release of LDH from T84 cells after 6 h infection (data not shown). Furthermore, caspase inhibition of T84 cells via Z-VAD did not lead to non-apoptotic death in EPEC-infected cells (data not shown). 3.3. EPEC-mediated T84 cell apoptosis is not required for barrier disruption during infection Further studies aimed to determine whether an upregulation of apoptosis contributes to EPEC-induced disruptions in epithelial barrier function. Confluent monolayers of T84 cells grown on Transwell filters were exposed to EPEC in the presence or absence of Z-VAD. Infected monolayers experienced a drop in TER, which achieved significance at 4 h (Fig. 4A), and continued at least to 8 h. When used at concentrations sufficient to prevent apoptosis (Fig. 3), the pan-caspase inhibitor proved unable to prevent this drop (Fig. 4A), and had no effect on the TER of uninfected cells (data not shown). Similarly, increases in the paracellular flux of a non-absorbable probe, FITClabeled 3000 MW dextran, during EPEC infection were not prevented by the inclusion of Z-VAD (Fig. 4B). 3.4. Cell apoptosis is not required for the cleavage of tight junctional proteins during EPEC infection
Fig. 1. Reduced viability and increased apoptosis in T84 cells infected with EPEC. (A) Cells exposed to EPEC displayed decreased viability by 4 h postinfection as measured by the MTT assay. (B) Increased levels of apoptosis were observed by 6 h postinfection as measured by oligonucleosome formation. (C) Cell apoptosis after 6 h exposure to EPEC was confirmed by immunoblotting for PARP and actin (loading control). Graphed data are representative means7S.E. from three independent experiments, n ¼ 4–8. *po0.0001 compared to time-matched uninfected controls.
ELISA in conjunction with caspase inhibitors to evaluate whether caspase activity was required for EPEC-induced T84 cell apoptosis. Pretreatment of T84 cells with inhibitors to all caspases, caspase-3, -8, or -9, completely blocked apoptosis (Fig. 3B). This indicates that both the extrinsic and the intrinsic pathways of caspase activation
To analyze whether EPEC-induced epithelial apoptosis led to alterations in the expression of tight junctional proteins, T84 cell lysates were collected after infection for 6 h, a time after which the cells exhibit increased apoptosis (Fig. 1). Immunoblotting revealed that membrane-associated ZO-1 from uninfected T84 cells migrated as the expected high molecular weight doublet representing the presence or absence of motif-a [23]. Interestingly, infection with EPEC led to the complete loss of the full-length protein, and the appearance of a 70 kDa cleavage fragment (Fig. 5). Membrane-associated occludin in uninfected cells migrated predominately as the hyperphosphorylated protein, but a 60 kDa band was also present (Fig. 5). After infection, levels of both hyperphosphorylated occludin and the 60 kDa band were dramatically reduced. In addition, occludin cleavage fragments of molecular weight 41 and 37 kDa appeared in infected cells. In both cases (ZO-1 and occludin), treatment of the cells with Z-VAD was unable to prevent the cleavage and loss of tight junctional proteins. Indeed, caspase inhibition appeared to increase the appearance of occludin fragments in EPEC-infected cells (Fig. 5).
ARTICLE IN PRESS A.N. Flynn, A.G. Buret / Microbial Pathogenesis 44 (2008) 311–319
315
Fig. 2. EPEC activates multiple T84 cell caspases. Cells exposed to EPEC for 6 h displayed activated caspases-3 (A), caspases-6 (B), caspases-8 (C), and caspases-9 (D) as determined by chromogenic activity kits. Data are representative means7S.E. from three independent experiments, n ¼ 4. *po0.001 compared to time-matched uninfected controls.
3.5. Cell apoptosis is not required for the mislocalization of tight junctional proteins during EPEC infection Since we found membrane-associated ZO-1 and occludin to be reduced and cleaved, we next used immunofluorescence microscopy to determine their cellular localization. Control T84 cells displayed the expected ZO-1 and occludin localization to the tight junctions in a chicken wire pattern (Fig. 6(a) and (d)). In contrast, cells incubated with EPEC for 6 h demonstrated a substantial redistribution of ZO-1 and occludin away from the tight junctions (Fig. 6(b) and (e)). In agreement with previous observations [6], the appearance of strand breaks in the staining pattern of ZO-1 and occludin in infected cells was accompanied by the apparent shift of these tight junctional proteins into the cytoplasm. The inclusion of the pancaspase inhibitor was again unable to prevent the redistribution of tight junctional proteins in EPEC-infected T84 cells (Fig. 6(c) and (f)). 4. Discussion In addition to its roles in nutrient and water absorption, the intestinal epithelium acts as a physical barrier between
the external and internal environments. While subtle alterations in barrier function have been noted during physiological stimulation [8], large increases in epithelial permeability often accompany intestinal diseases [9,10]. IEC apoptosis is a physiological mechanism required for normal cell turnover. Indeed, in the human intestine, epithelial cells appear to undergo apoptosis during shedding from the monolayer [24]. Interestingly, the loss of cells from the monolayer during this process does not compromise intestinal barrier function [25]. However, in model epithlia as well as in vivo, increases in cell apoptosis due to immunological or microbial factors may lead to disruptions in barrier function [13,15,16]. Conversely, increases in epithelial permeability observed during exposure to the pro-inflammatory cytokines TNF-a and IFN-g do not require an upregulation in cell apoptosis [26]. EPEC induces an upregulation of apoptosis [4,17] and paracellular permeability [3] in cultured IECs. Indeed, the secreted EPEC effector protein EspF is known to be necessary and/or sufficient to cause both of these effects [5,6]. This may seem to suggest a cause-to-effect relationship between increases in apoptosis and permeability in EPEC-infected cells. However, our results demonstrate that epithelial apoptosis is not required for the increases
ARTICLE IN PRESS 316
A.N. Flynn, A.G. Buret / Microbial Pathogenesis 44 (2008) 311–319
Fig. 3. Caspase activity is required for EPEC-induced T84 cell apoptosis. (A) Caspase-3 activity was determined by a chromogenic activity kit. Cells were left uninfected (control), or infected for 6 h with EPEC in the presence of no inhibitor (–) or inhibitors to all caspases (Z-VAD), caspase8 (Z-IETD), or caspase-9 (Z-LEHD). (B) Cells pretreated without (–), or with inhibitors to caspase-3 (Z-DEVD), all caspases (Z-VAD), caspase-8 (Z-IETD), or caspase-9 (Z-LEHD) were infected with EPEC for 6 h. Apoptosis was monitored via oligonucleosome formation, and compared to uninfected controls (arbitrarily set at 1). Data are representative means7S.E. from three independent experiments, n ¼ 4. *po0.001 compared to time-matched uninfected controls.
in cell monolayer permeability observed during EPEC infection. Crane and colleagues were the first to suggest that EPEC induces host cell apoptosis following non-specific losses in cell viability [4]. In agreement with these findings, we have demonstrated that EPEC causes early disruptions in host cell mitochondrial reductases, as measured by the loss of MTT conversion to formazan. It is likely that mitochondrial disruptions, possibly via EspF or Map activity, contribute to downstream pathways responsible for cellular apoptosis. However, surprisingly few reports have examined the apoptotic pathways involved in EPEC-induced epithelial cell death. Although EPEC has been shown to activate caspase-9 in non-intestinal cells [20], the contributions of intrinsic versus extrinsic pathways to apoptosis had not been investigated. The present results demonstrate for the first time that EPEC activates the intrinsic (i.e., caspase-9) and extrinsic (i.e., caspase-8) apoptotic pathways, and furthermore that both contribute to apoptosis in IECs. In addition, the inclusion of a pan-caspase inhibitor entirely prevented increases in IEC apoptosis due to EPEC infection. The findings indicate that EPEC-induced IEC
Fig. 4. Caspase activity is not required for EPEC-mediated increases in T84 cell paracellular permeability. Cells pretreated with or without a pancaspase inhibitor (Z-VAD) were infected with EPEC and assessed for paracellular permeability via TER (A) or the apical-to-basolateral migration of a 3000 MW FITC-Dextran probe (B). Data are representative means7S.E. from three independent experiments, n ¼ 4. *po0.005 compared to time-matched uninfected controls.
apoptosis is caspase-dependent. Interestingly, a recent paper identified a similar phenomenon in uropathogenic E. coli (UPEC)-mediated apoptosis in urothelial cells [27]. This occurred in association with Bid translocation to the mitochondria, suggesting that apoptosis in this model arises via direct activation of the extrinsic pathway followed by caspase-8 and Bid-dependent mitochondrial release of Cytochrome c [27]. Further research is required to determine whether similar mechanisms are at play with EPEC-mediated IEC apoptosis. Furthermore, and in contrast to several studies using other model systems [28,29], inhibition of caspases in EPEC-infected T84 cells did not initiate non-apoptotic forms of cell death (e.g., necrosis or autophagy). The inability of a pan-caspase inhibitor to prevent increased permeability in EPEC-infected IEC monolayers is in contrast to several other models. For example, we have previously shown that both G. lamblia as well as high levels of lipopolysaccharide increased IEC paracellular permeability in a caspase-3-dependent fashion [15,16]. However, as noted above, and in agreement with our results, increases in IEC permeability caused by TNF-a and IFN-g were not prevented by caspase inhibition [26]. In these systems, apoptosis and paracellular permeability seem to be regulated in parallel, rather than sequentially.
ARTICLE IN PRESS A.N. Flynn, A.G. Buret / Microbial Pathogenesis 44 (2008) 311–319
Interestingly, co-treatment of EPEC-infected T84 cells with the non-pathogenic yeast Saccharomyces boulardii prevented increases in permeability, but only delayed the induction of host cell apoptosis [19]. In conjunction with our current data, this suggests that not only are increases in the permeability of EPEC-infected IECs independent of
Fig. 5. EPEC infection leads to the cleavage and down-regulation of membrane-associated ZO-1 and occludin in a caspase-independent fashion. Cells pretreated with or without a pan-caspase inhibitor (ZVAD) were exposed to EPEC for 6 h and then processed for immunoblotting. Representative blots for ZO-1 and occludin from three independent experiments are shown. Actin serves as a loading control. Molecular weights (kDa) are displayed to the right of the blots.
317
apoptosis, but that increases in apoptosis can occur without increased permeability. Furthermore, recent research indicates that mitochondrial targeting of EspF is required for host cell apoptosis, but not for barrier disruption [30,31]. Our results support and extend these findings by demonstrating that whole EPEC disrupts intestinal barrier function independent of increases in cellular apoptosis. It should be noted that the temporal dissociation we observed between TER and the paracellular movement of the dextran probe is in line with a previous model using T84 cells [32]. Early decreases in T84 monolayer TER may represent chloride secretion rather than barrier disruption. This underlies the importance of performing studies on the apical-to-basolateral movement of molecular probes when examining intestinal barrier function. Therefore, although we found that TER dropped after 4 h of infection, we believe that the observed increase in paracellular flux at 6 h after infection is more reliable as an indicator of increased paracellular permeability. This validates the use of caspase inhibitors to evaluate the contribution of apoptosis, which is also seen after 6 h of infection, to increased monolayer permeability. To correlate functional changes in the IEC barrier with structural changes, we performed immunoblotting for tight junctional proteins during EPEC infection. A 6 h infection with EPEC resulted in the dramatic down-regulation of membrane-associated hyperphosphorylated occludin. This is consistent with a previous study that reported occludin dephosporylation in whole cell lysates from T84 cells infected with EPEC [33]. We also observed, for the first time, cleavage of occludin into 41 and 37 kDa fragments upon EPEC infection. Interestingly, cleavage of occludin has been observed during epithelial cell apoptosis induced
Fig. 6. EPEC infection leads to the mislocalization of tight juntional ZO-1 and occludin in a caspase-independent fashion. Cells pretreated with or without a pan-caspase inhibitor (Z-VAD) were exposed to EPEC for 6 h and then processed for immunocytochemistry. Representative photomicrographs for ZO-1 (a–c) and occludin (d–f) from three independent experiments are shown. Note the dramatic loss of ZO-1 and occludin at the tight junctions in EPECinfected monolayers (arrows), regardless of caspase inhibition. Scale bar, 5 mm.
ARTICLE IN PRESS 318
A.N. Flynn, A.G. Buret / Microbial Pathogenesis 44 (2008) 311–319
by a variety of stimuli [34]. However, the cleavage fragments observed in that study were 31 and 55 kDa in mass, and were formed, at least in part, by the proteolytic activity of caspases [34]. This indicates that cleavage of tight junctional occludin in IECs during EPEC infection occurs via mechanisms distinct from classical apoptotic pathways. The same is true of ZO-1, which we found to be cleaved into a distinct 70 kDa fragment during EPEC infection. This, again, is in contrast to a previous report that observed the fragmentation of ZO-1 into at least five separate bands during apoptosis [34]. However, this experiment was performed in breast epithelial cells that expressed a full-length ZO-1 of only 185 kDa, which may explain our differing results. Finally, we have also demonstrated that EPEC-induced disruptions in the localization of tight junctional occludin and ZO-1 are not dependent on apoptosis. Further studies are warranted to determine the bacterial and/or cellular factors responsible for the cleavage and disruption of host cell tight junctional proteins. In conclusion, the current study conclusively demonstrates that EPEC-induced IEC apoptosis requires both intrinsic and extrinsic caspase-dependent pathways. However, epithelial barrier malfunction and tight junction disruption due to EPEC infection occurs independent of host cell apoptosis. Taken together, these findings shed new light on the pathogenesis of EPEC during its interaction with host IECs. Acknowledgments This study was supported by grants from the Natural Sciences and Engineering Research Council of Canada (NSERC) and the Canadian Institutes of Health Research (CIHR). A.N.F. is the recipient of an NSERC postgraduate scholarship. References [1] Chen HD, Frankel G. Enteropathogenic Escherichia coli: unravelling pathogenesis. FEMS Microbiol Rev 2005;29:83–98. [2] Muza-Moons MM, Koutsouris A, Hecht G. Disruption of cell polarity by enteropathogenic Escherichia coli enables basolateral membrane proteins to migrate apically and to potentiate physiological consequences. Infect Immun 2003;71:7069–78. [3] Philpott DJ, McKay DM, Sherman PM, Perdue MH. Infection of T84 cells with enteropathogenic Escherichia coli alters barrier and transport functions. Am J Physiol 1996;270:G634–45. [4] Crane JK, Majumdar S, Pickhardt 3rd. DF. Host cell death due to enteropathogenic Escherichia coli has features of apoptosis. Infect Immun 1999;67:2575–84. [5] Crane JK, McNamara BP, Donnenberg MS. Role of EspF in host cell death induced by enteropathogenic Escherichia coli. Cell Microbiol 2001;3:197–211. [6] McNamara BP, Koutsouris A, O’Connell CB, Nougayrede JP, Donnenberg MS, Hecht G. Translocated EspF protein from enteropathogenic Escherichia coli disrupts host intestinal barrier function. J Clin Invest 2001;107:621–9. [7] Clayburgh DR, Shen L, Turner JR. A porous defense: the leaky epithelial barrier in intestinal disease. Lab Invest 2004;84:282–91.
[8] Turner JR, Rill BK, Carlson SL, Carnes D, Kerner R, Mrsny RJ, et al. Physiological regulation of epithelial tight junctions is associated with myosin light-chain phosphorylation. Am J Physiol 1997;273:C1378–85. [9] Katz KD, Hollander D, Vadheim CM, McElree C, Delahunty T, Dadufalza VD, et al. Intestinal permeability in patients with Crohn’s disease and their healthy relatives. Gastroenterology 1989;97: 927–31. [10] van Elburg RM, Uil JJ, Mulder CJ, Heymans HS. Intestinal permeability in patients with coeliac disease and relatives of patients with coeliac disease. Gut 1993;34:354–7. [11] Dean P, Kenny B. Intestinal barrier dysfunction by enteropathogenic Escherichia coli is mediated by two effector molecules and a bacterial surface protein. Mol Microbiol 2004;54:665–75. [12] Yuhan R, Koutsouris A, Savkovic SD, Hecht G. Enteropathogenic Escherichia coli-induced myosin light chain phosphorylation alters intestinal epithelial permeability. Gastroenterology 1997;113: 1873–82. [13] Abreu MT, Palladino AA, Arnold ET, Kwon RS, McRoberts JA. Modulation of barrier function during Fas-mediated apoptosis in human intestinal epithelial cells. Gastroenterology 2000;119:1524–36. [14] Bojarski C, Gitter AH, Bendfeldt K, Mankertz J, Schmitz H, Wagner S, et al. Permeability of human HT-29/B6 colonic epithelium as a function of apoptosis. J Physiol 2001;535:541–52. [15] Chin AC, Teoh DA, Scott KG, Meddings JB, Macnaughton WK, Buret AG. Strain-dependent induction of enterocyte apoptosis by Giardia lamblia disrupts epithelial barrier function in a caspase3-dependent manner. Infect Immun 2002;70:3673–80. [16] Chin AC, Flynn AN, Fedwick JP, Buret AG. The role of caspase-3 in lipopolysaccharide-mediated disruption of intestinal epithelial tight junctions. Can J Physiol Pharmacol 2006;84:1043–50. [17] Abul-Milh M, Wu Y, Lau B, Lingwood CA, Barnett Foster D. Induction of epithelial cell death including apoptosis by enteropathogenic Escherichia coli expressing bundle-forming pili. Infect Immun 2001;69:7356–64. [18] Fedwick JP, Lapointe TK, Meddings JB, Sherman PM, Buret AG. Helicobacter pylori activates myosin light-chain kinase to disrupt claudin-4 and claudin-5 and increase epithelial permeability. Infect Immun 2005;73:7844–52. [19] Czerucka D, Dahan S, Mograbi B, Rossi B, Rampal P. Saccharomyces boulardii preserves the barrier function and modulates the signal transduction pathway induced in enteropathogenic Escherichia coli-infected T84 cells. Infect Immun 2000;68:5998–6004. [20] Nougayrede JP, Donnenberg MS. Enteropathogenic Escherichia coli EspF is targeted to mitochondria and is required to initiate the mitochondrial death pathway. Cell Microbiol 2004;6:1097–111. [21] Nougayrede JP, Foster GH, Donnenberg MS. Enteropathogenic Escherichia coli effector EspF interacts with host protein Abcf2. Cell Microbiol 2007;9:680–93. [22] Spitz J, Yuhan R, Koutsouris A, Blatt C, Alverdy J, Hecht G. Enteropathogenic Escherichia coli adherence to intestinal epithelial monolayers diminishes barrier function. Am J Physiol 1995;268: G374–9. [23] Willott E, Balda MS, Heintzelman M, Jameson B, Anderson JM. Localization and differential expression of two isoforms of the tight junction protein ZO-1. Am J Physiol 1992;262:C1119–24. [24] Bullen TF, Forrest S, Campbell F, Dodson AR, Hershman MJ, Pritchard DM, et al. Characterization of epithelial cell shedding from human small intestine. Lab Invest 2006;86:1052–63. [25] Watson AJ, Chu S, Sieck L, Gerasimenko O, Bullen T, Campbell F, et al. Epithelial barrier function in vivo is sustained despite gaps in epithelial layers. Gastroenterology 2005;129:902–12. [26] Bruewer M, Luegering A, Kucharzik T, Parkos CA, Madara JL, Hopkins AM, et al. Proinflammatory cytokines disrupt epithelial barrier function by apoptosis-independent mechanisms. J Immunol 2003;171:6164–72. [27] Klumpp DJ, Rycyk MT, Chen MC, Thumbikat P, Sengupta S, Schaeffer AJ. Uropathogenic Escherichia coli induces extrinsic and
ARTICLE IN PRESS A.N. Flynn, A.G. Buret / Microbial Pathogenesis 44 (2008) 311–319 intrinsic cascades to initiate urothelial apoptosis. Infect Immun 2006;74:5106–13. [28] Cao C, Subhawong T, Albert JM, Kim KW, Geng L, Sekhar KR, et al. Inhibition of mammalian target of rapamycin or apoptotic pathway induces autophagy and radiosensitizes PTEN null prostate cancer cells. Cancer Res 2006;66:10040–7. [29] Scheller C, Knoferle J, Ullrich A, Prottengeier J, Racek T, Sopper S, et al. Caspase inhibition in apoptotic T cells triggers necrotic cell death depending on the cell type and the proapoptotic stimulus. J Cell Biochem 2006;97:1350–61. [30] Nagai T, Abe A, Sasakawa C. Targeting of enteropathogenic Escherichia coli EspF to host mitochondria is essential for bacterial pathogenesis: critical role of the 16th leucine residue in EspF. J Biol Chem 2005;280:2998–3011.
319
[31] Quitard S, Dean P, Maresca M, Kenny B. The enteropathogenic Escherichia coli EspF effector molecule inhibits PI-3 kinase-mediated uptake independently of mitochondrial targeting. Cell Microbiol 2006;8:972–81. [32] Parkos CA, Colgan SP, Delp C, Arnaout MA, Madara JL. Neutrophil migration across a cultured epithelial monolayer elicits a biphasic resistance response representing sequential effects on transcellular and paracellular pathways. J Cell Biol 1992;117:757–64. [33] Simonovic I, Rosenberg J, Koutsouris A, Hecht G. Enteropathogenic Escherichia coli dephosphorylates and dissociates occludin from intestinal epithelial tight junctions. Cell Microbiol 2000;2:305–15. [34] Bojarski C, Weiske J, Schoneberg T, Schroder W, Mankertz J, Schulzke JD, et al. The specific fates of tight junction proteins in apoptotic epithelial cells. J Cell Sci 2004;117:2097–107.
MICROBIAL PATHOGENESIS Microbial Pathogenesis 44 (2008) 311–319 www.elsevier.com/locate/micpath
Caspases-3, -8, and -9 are required for induction of epithelial cell apoptosis by enteropathogenic E. coli but are dispensable for increased paracellular permeability Andrew N. Flynn, Andre G. Buret Inflammation Research Network, University of Calgary, 2500 University Drive NW, Calgary, Alta., Canada T2N 1N4 Received 25 August 2007; received in revised form 10 October 2007; accepted 11 October 2007 Available online 24 October 2007
Abstract Enteropathogenic Escherichia coli (EPEC) is an important cause of diarrhea, particularly among infants in developing countries. An increase in intestinal permeability due to EPEC infection has been suggested as a factor in the development of diarrhea. Abnormally high levels of programmed cell death (apoptosis) of intestinal epithelial cells can lead to increased intestinal permeability. The effects of EPEC on cell apoptosis remain incompletely understood. This study characterized the mechanisms of EPEC-induced epithelial apoptosis and examined whether this effect contributes to heightened permeability in an in vitro model of infection. We report that EPEC-induced apoptosis in T84 intestinal epithelial cells via a mechanism involving caspases-3, -6, -8, and -9, the cleavage of PARP, and oligonucleosome formation. In addition, EPEC time-dependently increased paracellular permeability as assessed by transepithelial resistance and the apical-to-basolateral movement of 3000 MW dextran. Furthermore, EPEC infection led to the cleavage and mislocalization of tight junctional ZO-1 and occludin. However, pharmacological inhibition of caspases did not prevent the EPECinduced disruptions in epithelial barrier structure and function. Taken together, these results suggest that a caspase-dependent upregulation in epithelial cell apoptosis during EPEC infection occurs independent of impaired intestinal barrier function. Crown Copyright r 2007 Published by Elsevier Ltd. All rights reserved. Keywords: Tight junction; Intestinal permeability; Programmed cell death
1. Introduction Enteropathogenic Escherichia coli (EPEC) is an important cause of infant diarrhea in developing countries, and is associated with significant levels of mortality [1]. Infection with EPEC is characterized by intimate attachment of the bacteria to intestinal epithelial cells (IECs). This leads to the formation of an actin-rich host cell pedestal upon which the bacterium rests, and the effacement of the surrounding brush border. The results of these effects are termed attaching and effacing (A/E) lesions. A/E lesions are the direct result of the secretion of EPEC effector proteins into the host cell cytosol via a type three secretion system (TTSS). The injection of these proteins by EPEC leads to diverse phenotypic changes in cultured Corresponding author. Tel.: +1 403 220 2817; fax: +1 403 289 9311.
E-mail address: [email protected] (A.G. Buret).
IECs, including loss of polarity [2], altered ion transport [3], activation of apoptosis [4,5], and loss of barrier function [3,6]. The epithelial layer of the gastrointestinal tract represents an integral barrier between the internal environment of the body and potentially antigenic constituents of the lumen. Paracellular migration of these constituents is prevented by the apical junctional complex (AJC), which consists of transmembrane proteins, including occludin and the claudin family, and of cytoplasmic plaque proteins such as zonula occludens (ZO)-1 [7]. In concert, these proteins provide a physical barrier linking cells together. This barrier is tightly regulated to allow subtle alterations in permeability due to physiological stimuli while preventing inappropriate activation of the mucosal immune system [8]. Dramatic increases in intestinal permeability have been associated with a number of pathological conditions, including Crohn’s disease [9] and celiac disease [10].
0882-4010/$ - see front matter Crown Copyright r 2007 Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.micpath.2007.10.007
ARTICLE IN PRESS 312
A.N. Flynn, A.G. Buret / Microbial Pathogenesis 44 (2008) 311–319
Previous studies have demonstrated that in vitro infections with EPEC lead to increases in the permeability of IEC monolayers, but the mechanisms involved remain obscure [3]. This effect may contribute to the watery diarrhea that is a result of infection in humans. Although the bacterial effector proteins EspF, Map, and intimin and the host protein myosin light chain kinase (MLCK) appear to be important for increases in permeability [6,11,12], other factors await investigation. Upregulated IEC apoptosis has been found to lead to increases in monolayer permeability in various systems [13,14]. We have previously demonstrated that the protozoan parasite Giardia lamblia increases the permeability of IEC monolayers in a caspase-3-dependent manner [15]. More recently, we have shown the same effect using high doses of E. coli lipopolysaccharide [16]. Surprisingly, although EPEC may induce apoptosis in IECs [4,17], the contribution of this to increases in cell monolayer permeability has not been investigated. In addition, the pathways by which IECs undergo apoptosis when challenged with EPEC remain largely unknown. Therefore, the purposes of this study were to investigate intracellular pathways involved in EPEC-mediated IEC apoptosis, and to determine whether increases in IEC apoptosis contribute to increases in cell monolayer permeability during EPEC infection. 2. Materials and methods 2.1. Antibodies Antibodies directed against the following proteins, at the concentrations specified, were used for immunoblotting or immunocytochemistry: ZO-1, occludin (1:250; Zymed Laboratories, San Francisco, CA); PARP (1:1000; Cell Signaling Technology, Beverly, MA). Secondary antibodies conjugated to Alexa fluorophores or horseradish peroxidase were obtained from Invitrogen (1:2000; Carlsbad, CA) and Santa Cruz Biotechnology (1:10 000; Santa Cruz, CA), respectively. 2.2. Cell culture Human intestinal T84 cells (ATCC, Manassas, VA) were grown in a 1:1 mixture of DMEM and Ham’s F-12 supplemented with 5% fetal bovine serum, 100 mg/ml streptomycin, 100 U/ml penicillin, and 2 mM L-glutamine (all from Sigma, St. Louis, MO). Cells were incubated at 37 1C and 5% CO2 in 96% humidity. For all experiments, T84 cells were grown for 7 days on Lab-Tek chamber slides (Nalge Nunc International, Naperville, IL), culture-treated plates and dishes, or Transwell filter units (Costar, Corning, NY). Three hours prior to infection, monolayers were washed with HBSS followed by addition of antibioticfree medium containing 0.5% fetal bovine serum (hereafter referred to as infection medium).
2.3. Infection of monolayers Overnight cultures of EPEC strain E2348/69 (kindly provided by R. DeVinney, University of Calgary) were diluted 1:50 in LB broth and grown to early log growth phase (3 h) at 37 1C. Bacteria were then centrifuged at 1000 g for 10 min and resuspended in the appropriate volume of infection medium to give a multiplicity of infection of 100. Bacteria were added to the apical side of cell monolayers and were co-incubated for indicated lengths of time. In some experiments, T84 cells were pretreated for 1 h with 50 mM caspase inhibitors (Calbiochem, La Jolla, CA), which remained in the culture medium for the duration of the experiment. 2.4. MTT cell viability assay Viability of T84 cells was assessed using the thiazolyl blue tetrazolium bromide (MTT) method as previously described [16]. The effect of EPEC on epithelial cell viability was expressed as a percent of surviving cells compared to vehicle-treated controls. 2.5. Apoptotic nucleosome quantification T84 cell apoptosis was quantified using a Cell Death Detection ELISA kit (Roche Molecular Biochemicals, Laval, PQ) according to the manufacturer’s instructions as previously described [16]. Apoptosis was expressed as absorbance fold increase of the experimental cell lysates over controls, arbitrarily set at 1.0. 2.6. Caspase activity assays Caspase activity of T84 cells was assessed using the ApoTarget Caspase Assay (BioSource, Camarillo, CA). Briefly, cell monolayers grown in 12-well plates were washed with HBSS before being lysed with the supplied buffer. After brief sonication and centrifugation, lysates were stored at 70 1C until used. The protein content of thawed samples was normalized using the DC protein assay (Bio-Rad, Hercules, CA) before incubation with synthetic caspase substrates conjugated to p-nitroanilide. Chromophore liberation was detected at 405 nm using a SpectraMax M2e microplate reader (Molecular Devices, Sunnyvale, CA). 2.7. Lactate dehydrogenase release assay T84 cells grown in 48-well plates were infected with EPEC for 6 h. Fifty microliter samples of the culture medium were centrifuged (5 min, 10,000g) and assayed for lactate dehydrogenase (LDH) activity using the CytoTox 96 Non-Radioactive Cytotoxicity Assay according to the manufacturer’s instructions (Promega Corporation, Madison, WI). Pure cultures of EPEC did not demonstrate any LDH activity.
ARTICLE IN PRESS A.N. Flynn, A.G. Buret / Microbial Pathogenesis 44 (2008) 311–319
2.8. Whole cell lysis and subcellular fractionation For whole cell lysis, T84 cells grown on 60 mm dishes were washed with HBSS, and then lysed for 30 min in 4 1C RIPA buffer (1 PBS, 1% Igepal CA-630, 0.5% sodium deoxycholate, and 0.1% SDS) containing a proteinase inhibitor tablet (Complete Mini, Roche Diagnostics, Mannheim, Germany). Following brief sonication, lysates were cleared by centrifugation (10 min, 10,000g). For membrane extractions, cells were washed with HBSS, and then extracted with lysis buffer A (50 mM Tris–HCl, 150 mM NaCl, 0.5% Trition X-100, proteinase inhibitor tablet, pH 7.5) for 30 min at 4 1C. Lysates were then centrifuged (10 min, 10,000g), and supernatants were removed. Pellets were extracted with lysis buffer B (lysis buffer A+0.02% SDS) for 20 min at 4 1C before brief sonication and further centrifugation (10 min, 10,000g). Supernatants were collected as membranous fractions. In all cases, protein levels were normalized using DC protein assay. 2.9. Immunoblotting Protein samples added 1:1 to SDS sample buffer and boiled for 5 min were separated by SDS–PAGE and transferred to polyvinylidene difluoride membranes (Hybond-P, Amersham Biosciences, Buckinghamshire, UK). Membranes, blocked with 5% non-fat skim milk or 5% BSA, were probed overnight at 4 1C with primary antibodies followed by 45 min incubation at room temperature with secondary antibody. Bands were visualized using a chemiluminescence detection system (ECL-Plus, Amersham). The results shown are representative immunoblots of three independent experiments. 2.10. Immunocytochemistry Confluent T84 cell monolayers grown on chamber slides or Transwell filters were washed with HBSS and fixed with 20 1C methanol for 10 min. Cells were blocked with 10% normal goat serum followed by overnight incubation at 4 1C with primary antibodies diluted in Ultra Clean Diluent (Lab Vision). Secondary antibody labeling was carried out with a 45 min incubation at room temperature, after which cells were mounted with Aqua Poly/Mount (Polysciences, Warrington, PA). Cells were visualized using a Leica DMR fluorescence microscope and photomicrographs were taken using a Photometrics CoolSNAP camera (Roper Scientific, Tucson, AZ). 2.2. In vitro permeability Transepithelial resistance (TER) of T84 monolayers grown on Transwell filters was measured using an electrovoltohmeter (World Precision Instruments, Sarasota, FL). Paracellular permeability to a non-absorbable fluorescein isothiocyanate (FITC)-conjugated 3000-mole-
313
cular-weight (MW) dextran probe was assessed as previously described [18]. Values were expressed as fold increase of monolayer permeability over uninfected monolayers, arbitrarily set at 1.0. 2.3. Statistical analysis Numerical values from experiments were expressed as mean7S.E.M. throughout. Analyses were made using a one-way ANOVA with a Student–Neuman–Keuls multiple comparison test. Statistical significance was established at po0.05. 3. Results 3.1. EPEC reduces T84 cell viability and induces cell apoptosis T84 cells infected with EPEC demonstrated timedependent decreases in viability as determined by the MTT method (Fig. 1A). As the MTT method directly measures metabolically active cells, the reduction observed by 4 h represents an early loss of cell viability. In agreement with previous studies using different methods [4,5], EPECinduced apoptosis in T84 cells after an infection time of 6 h. To assess levels of apoptosis, we used both a Cell Death ELISA-based detection of oligonucleosome formation (Fig. 1B) as well as immunoblotting for cleaved PARP (Fig. 1C). Both of these techniques have been extensively validated as methods for detecting cellular apoptosis [15,16]. In further experiments regarding EPEC-induced cell apoptosis, we focused on the Cell Death ELISA due to its rapid and quantitative data generation. 3.2. EPEC activates the extrinsic and intrinsic apoptotic pathways Previous work has determined that EPEC activates the central apoptotic executioner caspase-3 in T84 cells [19]. However, investigations into the activation of upstream caspases have focused solely on caspase-9 using non-IECs [20,21]. Furthermore, the contribution of various caspases to EPEC-mediated IEC apoptosis has not been determined. We used a chromogenic-based assay with various caspase substrates to evaluate caspase activity in T84 cells. After 6 h of infection, T84 cells displayed increased activity of the executioner caspases-3 and -6 (Fig. 2(A) and (B)). Interestingly, at the same time point, both the extrinsic initiator caspase-8 and the intrinsic initiator caspase-9 were activated (Fig. 2(C) and (D)). To determine which initiator caspase was required for the activation of the executioner caspase-3, caspase-3 activity was monitored in infected cells pretreated with various caspase inhibitors. Interestingly, the inhibition of either caspase-8 or -9 completely prevented caspase-3 activation in EPEC-infected T84 cells (Fig. 3A). As expected, the pan-caspase inhibitor Z-VAD ablated caspase-3 activity. We then used the Cell Death
ARTICLE IN PRESS 314
A.N. Flynn, A.G. Buret / Microbial Pathogenesis 44 (2008) 311–319
contribute to EPEC-induced IEC apoptosis. The specificity of caspase inhibitors was verified using staurosporine as an apoptotic stimulus in conjunction with caspase activity assays (data not shown). To ensure that caspase inhibition did not lead to alternate forms of cell death (e.g., necrosis or autophagy), EPEC-infected T84 cells were assessed for LDH release. In agreement with previous studies [6,22], EPEC did not induce the release of LDH from T84 cells after 6 h infection (data not shown). Furthermore, caspase inhibition of T84 cells via Z-VAD did not lead to non-apoptotic death in EPEC-infected cells (data not shown). 3.3. EPEC-mediated T84 cell apoptosis is not required for barrier disruption during infection Further studies aimed to determine whether an upregulation of apoptosis contributes to EPEC-induced disruptions in epithelial barrier function. Confluent monolayers of T84 cells grown on Transwell filters were exposed to EPEC in the presence or absence of Z-VAD. Infected monolayers experienced a drop in TER, which achieved significance at 4 h (Fig. 4A), and continued at least to 8 h. When used at concentrations sufficient to prevent apoptosis (Fig. 3), the pan-caspase inhibitor proved unable to prevent this drop (Fig. 4A), and had no effect on the TER of uninfected cells (data not shown). Similarly, increases in the paracellular flux of a non-absorbable probe, FITClabeled 3000 MW dextran, during EPEC infection were not prevented by the inclusion of Z-VAD (Fig. 4B). 3.4. Cell apoptosis is not required for the cleavage of tight junctional proteins during EPEC infection
Fig. 1. Reduced viability and increased apoptosis in T84 cells infected with EPEC. (A) Cells exposed to EPEC displayed decreased viability by 4 h postinfection as measured by the MTT assay. (B) Increased levels of apoptosis were observed by 6 h postinfection as measured by oligonucleosome formation. (C) Cell apoptosis after 6 h exposure to EPEC was confirmed by immunoblotting for PARP and actin (loading control). Graphed data are representative means7S.E. from three independent experiments, n ¼ 4–8. *po0.0001 compared to time-matched uninfected controls.
ELISA in conjunction with caspase inhibitors to evaluate whether caspase activity was required for EPEC-induced T84 cell apoptosis. Pretreatment of T84 cells with inhibitors to all caspases, caspase-3, -8, or -9, completely blocked apoptosis (Fig. 3B). This indicates that both the extrinsic and the intrinsic pathways of caspase activation
To analyze whether EPEC-induced epithelial apoptosis led to alterations in the expression of tight junctional proteins, T84 cell lysates were collected after infection for 6 h, a time after which the cells exhibit increased apoptosis (Fig. 1). Immunoblotting revealed that membrane-associated ZO-1 from uninfected T84 cells migrated as the expected high molecular weight doublet representing the presence or absence of motif-a [23]. Interestingly, infection with EPEC led to the complete loss of the full-length protein, and the appearance of a 70 kDa cleavage fragment (Fig. 5). Membrane-associated occludin in uninfected cells migrated predominately as the hyperphosphorylated protein, but a 60 kDa band was also present (Fig. 5). After infection, levels of both hyperphosphorylated occludin and the 60 kDa band were dramatically reduced. In addition, occludin cleavage fragments of molecular weight 41 and 37 kDa appeared in infected cells. In both cases (ZO-1 and occludin), treatment of the cells with Z-VAD was unable to prevent the cleavage and loss of tight junctional proteins. Indeed, caspase inhibition appeared to increase the appearance of occludin fragments in EPEC-infected cells (Fig. 5).
ARTICLE IN PRESS A.N. Flynn, A.G. Buret / Microbial Pathogenesis 44 (2008) 311–319
315
Fig. 2. EPEC activates multiple T84 cell caspases. Cells exposed to EPEC for 6 h displayed activated caspases-3 (A), caspases-6 (B), caspases-8 (C), and caspases-9 (D) as determined by chromogenic activity kits. Data are representative means7S.E. from three independent experiments, n ¼ 4. *po0.001 compared to time-matched uninfected controls.
3.5. Cell apoptosis is not required for the mislocalization of tight junctional proteins during EPEC infection Since we found membrane-associated ZO-1 and occludin to be reduced and cleaved, we next used immunofluorescence microscopy to determine their cellular localization. Control T84 cells displayed the expected ZO-1 and occludin localization to the tight junctions in a chicken wire pattern (Fig. 6(a) and (d)). In contrast, cells incubated with EPEC for 6 h demonstrated a substantial redistribution of ZO-1 and occludin away from the tight junctions (Fig. 6(b) and (e)). In agreement with previous observations [6], the appearance of strand breaks in the staining pattern of ZO-1 and occludin in infected cells was accompanied by the apparent shift of these tight junctional proteins into the cytoplasm. The inclusion of the pancaspase inhibitor was again unable to prevent the redistribution of tight junctional proteins in EPEC-infected T84 cells (Fig. 6(c) and (f)). 4. Discussion In addition to its roles in nutrient and water absorption, the intestinal epithelium acts as a physical barrier between
the external and internal environments. While subtle alterations in barrier function have been noted during physiological stimulation [8], large increases in epithelial permeability often accompany intestinal diseases [9,10]. IEC apoptosis is a physiological mechanism required for normal cell turnover. Indeed, in the human intestine, epithelial cells appear to undergo apoptosis during shedding from the monolayer [24]. Interestingly, the loss of cells from the monolayer during this process does not compromise intestinal barrier function [25]. However, in model epithlia as well as in vivo, increases in cell apoptosis due to immunological or microbial factors may lead to disruptions in barrier function [13,15,16]. Conversely, increases in epithelial permeability observed during exposure to the pro-inflammatory cytokines TNF-a and IFN-g do not require an upregulation in cell apoptosis [26]. EPEC induces an upregulation of apoptosis [4,17] and paracellular permeability [3] in cultured IECs. Indeed, the secreted EPEC effector protein EspF is known to be necessary and/or sufficient to cause both of these effects [5,6]. This may seem to suggest a cause-to-effect relationship between increases in apoptosis and permeability in EPEC-infected cells. However, our results demonstrate that epithelial apoptosis is not required for the increases
ARTICLE IN PRESS 316
A.N. Flynn, A.G. Buret / Microbial Pathogenesis 44 (2008) 311–319
Fig. 3. Caspase activity is required for EPEC-induced T84 cell apoptosis. (A) Caspase-3 activity was determined by a chromogenic activity kit. Cells were left uninfected (control), or infected for 6 h with EPEC in the presence of no inhibitor (–) or inhibitors to all caspases (Z-VAD), caspase8 (Z-IETD), or caspase-9 (Z-LEHD). (B) Cells pretreated without (–), or with inhibitors to caspase-3 (Z-DEVD), all caspases (Z-VAD), caspase-8 (Z-IETD), or caspase-9 (Z-LEHD) were infected with EPEC for 6 h. Apoptosis was monitored via oligonucleosome formation, and compared to uninfected controls (arbitrarily set at 1). Data are representative means7S.E. from three independent experiments, n ¼ 4. *po0.001 compared to time-matched uninfected controls.
in cell monolayer permeability observed during EPEC infection. Crane and colleagues were the first to suggest that EPEC induces host cell apoptosis following non-specific losses in cell viability [4]. In agreement with these findings, we have demonstrated that EPEC causes early disruptions in host cell mitochondrial reductases, as measured by the loss of MTT conversion to formazan. It is likely that mitochondrial disruptions, possibly via EspF or Map activity, contribute to downstream pathways responsible for cellular apoptosis. However, surprisingly few reports have examined the apoptotic pathways involved in EPEC-induced epithelial cell death. Although EPEC has been shown to activate caspase-9 in non-intestinal cells [20], the contributions of intrinsic versus extrinsic pathways to apoptosis had not been investigated. The present results demonstrate for the first time that EPEC activates the intrinsic (i.e., caspase-9) and extrinsic (i.e., caspase-8) apoptotic pathways, and furthermore that both contribute to apoptosis in IECs. In addition, the inclusion of a pan-caspase inhibitor entirely prevented increases in IEC apoptosis due to EPEC infection. The findings indicate that EPEC-induced IEC
Fig. 4. Caspase activity is not required for EPEC-mediated increases in T84 cell paracellular permeability. Cells pretreated with or without a pancaspase inhibitor (Z-VAD) were infected with EPEC and assessed for paracellular permeability via TER (A) or the apical-to-basolateral migration of a 3000 MW FITC-Dextran probe (B). Data are representative means7S.E. from three independent experiments, n ¼ 4. *po0.005 compared to time-matched uninfected controls.
apoptosis is caspase-dependent. Interestingly, a recent paper identified a similar phenomenon in uropathogenic E. coli (UPEC)-mediated apoptosis in urothelial cells [27]. This occurred in association with Bid translocation to the mitochondria, suggesting that apoptosis in this model arises via direct activation of the extrinsic pathway followed by caspase-8 and Bid-dependent mitochondrial release of Cytochrome c [27]. Further research is required to determine whether similar mechanisms are at play with EPEC-mediated IEC apoptosis. Furthermore, and in contrast to several studies using other model systems [28,29], inhibition of caspases in EPEC-infected T84 cells did not initiate non-apoptotic forms of cell death (e.g., necrosis or autophagy). The inability of a pan-caspase inhibitor to prevent increased permeability in EPEC-infected IEC monolayers is in contrast to several other models. For example, we have previously shown that both G. lamblia as well as high levels of lipopolysaccharide increased IEC paracellular permeability in a caspase-3-dependent fashion [15,16]. However, as noted above, and in agreement with our results, increases in IEC permeability caused by TNF-a and IFN-g were not prevented by caspase inhibition [26]. In these systems, apoptosis and paracellular permeability seem to be regulated in parallel, rather than sequentially.
ARTICLE IN PRESS A.N. Flynn, A.G. Buret / Microbial Pathogenesis 44 (2008) 311–319
Interestingly, co-treatment of EPEC-infected T84 cells with the non-pathogenic yeast Saccharomyces boulardii prevented increases in permeability, but only delayed the induction of host cell apoptosis [19]. In conjunction with our current data, this suggests that not only are increases in the permeability of EPEC-infected IECs independent of
Fig. 5. EPEC infection leads to the cleavage and down-regulation of membrane-associated ZO-1 and occludin in a caspase-independent fashion. Cells pretreated with or without a pan-caspase inhibitor (ZVAD) were exposed to EPEC for 6 h and then processed for immunoblotting. Representative blots for ZO-1 and occludin from three independent experiments are shown. Actin serves as a loading control. Molecular weights (kDa) are displayed to the right of the blots.
317
apoptosis, but that increases in apoptosis can occur without increased permeability. Furthermore, recent research indicates that mitochondrial targeting of EspF is required for host cell apoptosis, but not for barrier disruption [30,31]. Our results support and extend these findings by demonstrating that whole EPEC disrupts intestinal barrier function independent of increases in cellular apoptosis. It should be noted that the temporal dissociation we observed between TER and the paracellular movement of the dextran probe is in line with a previous model using T84 cells [32]. Early decreases in T84 monolayer TER may represent chloride secretion rather than barrier disruption. This underlies the importance of performing studies on the apical-to-basolateral movement of molecular probes when examining intestinal barrier function. Therefore, although we found that TER dropped after 4 h of infection, we believe that the observed increase in paracellular flux at 6 h after infection is more reliable as an indicator of increased paracellular permeability. This validates the use of caspase inhibitors to evaluate the contribution of apoptosis, which is also seen after 6 h of infection, to increased monolayer permeability. To correlate functional changes in the IEC barrier with structural changes, we performed immunoblotting for tight junctional proteins during EPEC infection. A 6 h infection with EPEC resulted in the dramatic down-regulation of membrane-associated hyperphosphorylated occludin. This is consistent with a previous study that reported occludin dephosporylation in whole cell lysates from T84 cells infected with EPEC [33]. We also observed, for the first time, cleavage of occludin into 41 and 37 kDa fragments upon EPEC infection. Interestingly, cleavage of occludin has been observed during epithelial cell apoptosis induced
Fig. 6. EPEC infection leads to the mislocalization of tight juntional ZO-1 and occludin in a caspase-independent fashion. Cells pretreated with or without a pan-caspase inhibitor (Z-VAD) were exposed to EPEC for 6 h and then processed for immunocytochemistry. Representative photomicrographs for ZO-1 (a–c) and occludin (d–f) from three independent experiments are shown. Note the dramatic loss of ZO-1 and occludin at the tight junctions in EPECinfected monolayers (arrows), regardless of caspase inhibition. Scale bar, 5 mm.
ARTICLE IN PRESS 318
A.N. Flynn, A.G. Buret / Microbial Pathogenesis 44 (2008) 311–319
by a variety of stimuli [34]. However, the cleavage fragments observed in that study were 31 and 55 kDa in mass, and were formed, at least in part, by the proteolytic activity of caspases [34]. This indicates that cleavage of tight junctional occludin in IECs during EPEC infection occurs via mechanisms distinct from classical apoptotic pathways. The same is true of ZO-1, which we found to be cleaved into a distinct 70 kDa fragment during EPEC infection. This, again, is in contrast to a previous report that observed the fragmentation of ZO-1 into at least five separate bands during apoptosis [34]. However, this experiment was performed in breast epithelial cells that expressed a full-length ZO-1 of only 185 kDa, which may explain our differing results. Finally, we have also demonstrated that EPEC-induced disruptions in the localization of tight junctional occludin and ZO-1 are not dependent on apoptosis. Further studies are warranted to determine the bacterial and/or cellular factors responsible for the cleavage and disruption of host cell tight junctional proteins. In conclusion, the current study conclusively demonstrates that EPEC-induced IEC apoptosis requires both intrinsic and extrinsic caspase-dependent pathways. However, epithelial barrier malfunction and tight junction disruption due to EPEC infection occurs independent of host cell apoptosis. Taken together, these findings shed new light on the pathogenesis of EPEC during its interaction with host IECs. Acknowledgments This study was supported by grants from the Natural Sciences and Engineering Research Council of Canada (NSERC) and the Canadian Institutes of Health Research (CIHR). A.N.F. is the recipient of an NSERC postgraduate scholarship. References [1] Chen HD, Frankel G. Enteropathogenic Escherichia coli: unravelling pathogenesis. FEMS Microbiol Rev 2005;29:83–98. [2] Muza-Moons MM, Koutsouris A, Hecht G. Disruption of cell polarity by enteropathogenic Escherichia coli enables basolateral membrane proteins to migrate apically and to potentiate physiological consequences. Infect Immun 2003;71:7069–78. [3] Philpott DJ, McKay DM, Sherman PM, Perdue MH. Infection of T84 cells with enteropathogenic Escherichia coli alters barrier and transport functions. Am J Physiol 1996;270:G634–45. [4] Crane JK, Majumdar S, Pickhardt 3rd. DF. Host cell death due to enteropathogenic Escherichia coli has features of apoptosis. Infect Immun 1999;67:2575–84. [5] Crane JK, McNamara BP, Donnenberg MS. Role of EspF in host cell death induced by enteropathogenic Escherichia coli. Cell Microbiol 2001;3:197–211. [6] McNamara BP, Koutsouris A, O’Connell CB, Nougayrede JP, Donnenberg MS, Hecht G. Translocated EspF protein from enteropathogenic Escherichia coli disrupts host intestinal barrier function. J Clin Invest 2001;107:621–9. [7] Clayburgh DR, Shen L, Turner JR. A porous defense: the leaky epithelial barrier in intestinal disease. Lab Invest 2004;84:282–91.
[8] Turner JR, Rill BK, Carlson SL, Carnes D, Kerner R, Mrsny RJ, et al. Physiological regulation of epithelial tight junctions is associated with myosin light-chain phosphorylation. Am J Physiol 1997;273:C1378–85. [9] Katz KD, Hollander D, Vadheim CM, McElree C, Delahunty T, Dadufalza VD, et al. Intestinal permeability in patients with Crohn’s disease and their healthy relatives. Gastroenterology 1989;97: 927–31. [10] van Elburg RM, Uil JJ, Mulder CJ, Heymans HS. Intestinal permeability in patients with coeliac disease and relatives of patients with coeliac disease. Gut 1993;34:354–7. [11] Dean P, Kenny B. Intestinal barrier dysfunction by enteropathogenic Escherichia coli is mediated by two effector molecules and a bacterial surface protein. Mol Microbiol 2004;54:665–75. [12] Yuhan R, Koutsouris A, Savkovic SD, Hecht G. Enteropathogenic Escherichia coli-induced myosin light chain phosphorylation alters intestinal epithelial permeability. Gastroenterology 1997;113: 1873–82. [13] Abreu MT, Palladino AA, Arnold ET, Kwon RS, McRoberts JA. Modulation of barrier function during Fas-mediated apoptosis in human intestinal epithelial cells. Gastroenterology 2000;119:1524–36. [14] Bojarski C, Gitter AH, Bendfeldt K, Mankertz J, Schmitz H, Wagner S, et al. Permeability of human HT-29/B6 colonic epithelium as a function of apoptosis. J Physiol 2001;535:541–52. [15] Chin AC, Teoh DA, Scott KG, Meddings JB, Macnaughton WK, Buret AG. Strain-dependent induction of enterocyte apoptosis by Giardia lamblia disrupts epithelial barrier function in a caspase3-dependent manner. Infect Immun 2002;70:3673–80. [16] Chin AC, Flynn AN, Fedwick JP, Buret AG. The role of caspase-3 in lipopolysaccharide-mediated disruption of intestinal epithelial tight junctions. Can J Physiol Pharmacol 2006;84:1043–50. [17] Abul-Milh M, Wu Y, Lau B, Lingwood CA, Barnett Foster D. Induction of epithelial cell death including apoptosis by enteropathogenic Escherichia coli expressing bundle-forming pili. Infect Immun 2001;69:7356–64. [18] Fedwick JP, Lapointe TK, Meddings JB, Sherman PM, Buret AG. Helicobacter pylori activates myosin light-chain kinase to disrupt claudin-4 and claudin-5 and increase epithelial permeability. Infect Immun 2005;73:7844–52. [19] Czerucka D, Dahan S, Mograbi B, Rossi B, Rampal P. Saccharomyces boulardii preserves the barrier function and modulates the signal transduction pathway induced in enteropathogenic Escherichia coli-infected T84 cells. Infect Immun 2000;68:5998–6004. [20] Nougayrede JP, Donnenberg MS. Enteropathogenic Escherichia coli EspF is targeted to mitochondria and is required to initiate the mitochondrial death pathway. Cell Microbiol 2004;6:1097–111. [21] Nougayrede JP, Foster GH, Donnenberg MS. Enteropathogenic Escherichia coli effector EspF interacts with host protein Abcf2. Cell Microbiol 2007;9:680–93. [22] Spitz J, Yuhan R, Koutsouris A, Blatt C, Alverdy J, Hecht G. Enteropathogenic Escherichia coli adherence to intestinal epithelial monolayers diminishes barrier function. Am J Physiol 1995;268: G374–9. [23] Willott E, Balda MS, Heintzelman M, Jameson B, Anderson JM. Localization and differential expression of two isoforms of the tight junction protein ZO-1. Am J Physiol 1992;262:C1119–24. [24] Bullen TF, Forrest S, Campbell F, Dodson AR, Hershman MJ, Pritchard DM, et al. Characterization of epithelial cell shedding from human small intestine. Lab Invest 2006;86:1052–63. [25] Watson AJ, Chu S, Sieck L, Gerasimenko O, Bullen T, Campbell F, et al. Epithelial barrier function in vivo is sustained despite gaps in epithelial layers. Gastroenterology 2005;129:902–12. [26] Bruewer M, Luegering A, Kucharzik T, Parkos CA, Madara JL, Hopkins AM, et al. Proinflammatory cytokines disrupt epithelial barrier function by apoptosis-independent mechanisms. J Immunol 2003;171:6164–72. [27] Klumpp DJ, Rycyk MT, Chen MC, Thumbikat P, Sengupta S, Schaeffer AJ. Uropathogenic Escherichia coli induces extrinsic and
ARTICLE IN PRESS A.N. Flynn, A.G. Buret / Microbial Pathogenesis 44 (2008) 311–319 intrinsic cascades to initiate urothelial apoptosis. Infect Immun 2006;74:5106–13. [28] Cao C, Subhawong T, Albert JM, Kim KW, Geng L, Sekhar KR, et al. Inhibition of mammalian target of rapamycin or apoptotic pathway induces autophagy and radiosensitizes PTEN null prostate cancer cells. Cancer Res 2006;66:10040–7. [29] Scheller C, Knoferle J, Ullrich A, Prottengeier J, Racek T, Sopper S, et al. Caspase inhibition in apoptotic T cells triggers necrotic cell death depending on the cell type and the proapoptotic stimulus. J Cell Biochem 2006;97:1350–61. [30] Nagai T, Abe A, Sasakawa C. Targeting of enteropathogenic Escherichia coli EspF to host mitochondria is essential for bacterial pathogenesis: critical role of the 16th leucine residue in EspF. J Biol Chem 2005;280:2998–3011.
319
[31] Quitard S, Dean P, Maresca M, Kenny B. The enteropathogenic Escherichia coli EspF effector molecule inhibits PI-3 kinase-mediated uptake independently of mitochondrial targeting. Cell Microbiol 2006;8:972–81. [32] Parkos CA, Colgan SP, Delp C, Arnaout MA, Madara JL. Neutrophil migration across a cultured epithelial monolayer elicits a biphasic resistance response representing sequential effects on transcellular and paracellular pathways. J Cell Biol 1992;117:757–64. [33] Simonovic I, Rosenberg J, Koutsouris A, Hecht G. Enteropathogenic Escherichia coli dephosphorylates and dissociates occludin from intestinal epithelial tight junctions. Cell Microbiol 2000;2:305–15. [34] Bojarski C, Weiske J, Schoneberg T, Schroder W, Mankertz J, Schulzke JD, et al. The specific fates of tight junction proteins in apoptotic epithelial cells. J Cell Sci 2004;117:2097–107.