HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 6, 4247-4254, February 8, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/6/4247    most recent M109254200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Chen, M. L. Articles by LaMont, J. T. Search for Related Content PubMed PubMed Citation Articles by Chen, M. L. Articles by LaMont, J. T. Social Bookmarking                      What's this?

Protein Kinase C Signaling Regulates ZO-1 Translocation and Increased Paracellular Flux of T84 Colonocytes Exposed to Clostridium difficile Toxin A^*

Ming L. Chen, Charalabos Pothoulakis, and J. Thomas LaMont

From the Division of Gastroenterology, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts 02215

Received for publication, September 25, 2001, and in revised form, November 21, 2001

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Clostridium difficile toxin A increases paracellular permeability in colonic epithelial T84 cells by mechanisms involving RhoA glucosylation and actin depolymerization. However, we previously observed that toxin A-mediated decline in transepithelial electrical resistance preceded changes in cell morphology and tight junction ultrastructure (Hecht, G., Pothoulakis, C., LaMont, J. T., and Madara, J. L. (1988) J. Clin. Invest. 82, 1516-1524). Recent studies also showed that C. difficile toxins induce early cellular responses, including activation of mitogen-activated protein kinases, generation of reactive oxygen metabolites, and calcium influx. The aim of this study was to investigate whether toxin A-induced early cellular responses contribute to the permeability changes. We found that toxin A stimulated the activities of membrane and cytosolic protein kinase C (PKC) and cytosolic PKC. A specific PKC/ antagonist (myristoylated PKC/ peptide) blocked toxin A-mediated RhoA glucosylation. Furthermore, decreased transepithelial electrical resistance and increased translocation of ZO-1 from tight junction occurred within 2-3 h of toxin A exposure and were also inhibited by PKC/ antagonist. During this time period, toxin exposure did not induce translocation of ZO-2, dephosphorylation or translocation of occludin, or cell rounding. Our data indicate that PKC signaling regulates toxin A-mediated paracellular permeability changes and ZO-1 translocation.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Clostridium difficile is an anaerobic pathogen that causes antibiotic-associated diarrhea and colitis in humans (1). Two large molecular weight proteins, toxins A and B, secreted by this microbe are glucosyltransferases with substrate specificity for Rho family GTPases (Rho, Rac, and Cdc42) (2, 3). Rho proteins belong to the Ras superfamily of GTPases that regulate cell morphology, gene transcription, and cell proliferation (4). Rho glucosylation by toxins reduces the binding affinity of Rho to its downstream effectors, resulting in inhibition of Rho signaling, disassembly of actin filaments, and cell rounding (5). However, recent studies indicate that C. difficile toxins induce early cellular responses including activation of mitogen-activated protein kinases (6), generation of reactive oxygen metabolites (7), and induction of calcium influx (8) that occur prior to measurable Rho glucosylation, indicating that these and perhaps other Rho-independent signaling events are also involved in toxicity.

Dysfunction of paracellular permeability is a major contributor to the increased fluid secretion and diarrhea in C. difficile infection and other inflammatory bowel diseases (9-11). Tight junctions (TJs),¹ localized in the uppermost basolateral surface of polarized enterocytes, serve as a barrier regulating selective passage of fluid, ions, and lipids through the paracellular pathway (12, 13). Tight junction fibrils are composed of the transmembrane proteins occludin, claudins, and the junctional adhesion molecule whose cytosolic domains interact with the peripheral junctional proteins of the zonula occludens family (ZO-1, -2, and -3) (13). The C-terminal domains of ZOs, in turn, bind to perijunctional actomyosin filaments that support the TJ complex (14). TJ assembly and permeability are regulated by a network of signaling pathways including protein kinase C (PKC), heterotrimeric G proteins, Ras and Rho GTP-binding proteins, protein kinase A, tyrosine kinase, and calcium (15). For example, decreased transepithelial electrical resistance (TER) and increased paracellular flux are observed in cells exposed to the PKC agonist (12-O-tetradecanoylphorbol-13-acetate) (16). Both PKC and - are implicated in the PKC-dependent regulation of TJ assembly and barrier function following activation of a G-protein-coupled receptor (17-20). Freeze fracture electron micrographs reveal abnormal TJ morphology in transfectants overexpressing mutant forms of the Rho family of GTPases, suggesting involvement of Rho proteins in TJ assembly (21). In polarized intestinal epithelial cells (T84 cells) infected with a chimeric protein containing Clostridium botulinum C3 exoenzyme that specifically inactivates RhoA, perijunctional actin rings are disrupted, leading to a decrease in TER and redistribution of ZO-1 (22).

The increase in TJ permeability and redistribution of TJ proteins in T84 cells exposed to C. difficile toxins has been correlated with depolymerization of the actin cytoskeleton (23, 24). However, we previously observed a rapid drop in TER in the absence of any changes in cell morphology or TJ ultrastructure in T84 cells exposed to C. difficile toxin A (24). Furthermore, TJ permeability changes in response to toxin A preceded the disassembly of perijunctional actin filaments, suggesting that the early increase in TJ permeability may be mediated by factor(s) other than actin disassembly (24). In this study, we focused on the early effects of toxin A on TJ permeability and ZO-1 redistribution. Toxin A elevated the activities of membrane and cytosolic PKC and cytosolic PKC. A specific PKC/ antagonist (myristoylated PKC/ peptide (MPKCP), residues 19-27) blocked toxin A-mediated RhoA glucosylation. Furthermore, a significant decrease in TER and increased translocation of ZO-1 from TJ occurred within 3 h of toxin A exposure and were inhibited by MPKCP. These data suggest that PKC signaling regulates the effects of toxin A on TJ function in target cells.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Toxin A Purification and Cell Culture-- Toxin A was purified from C. difficile growth medium by ion exchange chromatography as described (25). Purified toxin A migrated as a single band in SDS-polyacrylamide gel. Toxin A cytotoxicity was monitored by incubating serial dilutions of toxin A with Chinese hamster ovary K1 cells for 24 h. We routinely observed that approximately 0.1 nM toxin A was required to induce rounding of Chinese hamster ovary K1 cells. T84 cells (American Type Culture Collection, Manassas, VA) were propagated in a 1:1 mixture of Dulbecco's modified Eagle's medium and Ham's F-12 medium containing heat-inactivated 10% fetal calf serum and antibiotics (penicillin and streptomycin). Confluent T84 cultures were exposed to toxin A in time- and dose-dependent studies as indicated in the figure legends. To analyze the involvement of PKC signaling in toxin-mediated TJ permeability changes, cells were pretreated with either calphostin C or MPKCP (Calbiochem-Novabiochem) for 30 min prior to toxin exposure.

Transepithelial Electrical Resistance and Paracellular Flux-- T84 cells were seeded onto Transwell filters (6.5-mm diameter, polycarbonate membrane, 0.4-µm pore size, Costar, Corning, NY) at a density of 5 × 10⁵ cells/insert as described (23). TER was measured in confluent T84 monolayer before and after apical toxin A exposure using Millicell-ERS (Millipore, Bedford, MA). Data were calculated by subtracting the value of a blank insert and normalized for growth area (ohm·cm²). Paracellular flux was determined by placing 200 µl of F-12/Dulbecco's modified Eagle's medium containing 0.2 mg/ml FITC-conjugated dextran (M_r 3000) (Molecular Probes, Inc., Eugene, OR) in the apical compartment with or without toxin A. The basal chamber was filled with 600 µl of F-12/Dulbecco's modified Eagle's medium containing 10% fetal calf serum. Transepithelial flux of dextran 3K from apical to basal compartments was determined by measuring fluorescence in the basal compartment with a Wallac Victor² 1420 multilabel counter (PerkinElmer Life Sciences). The concentration of FITC-dextran in the basal compartment was calculated from a standard curve, and apical to basal flux was expressed as pmol/h·cm². Statistical significance was evaluated by paired Student's t test.

PKC Activity and Phosphorylation of PLC-- Confluent T84 cells on 100-mm dishes were incubated in serum-free medium overnight. Cells were exposed to 1 nM toxin A for 1 h and collected into ice-cold buffer (10 mM Tris-Cl, pH 7.5, 5 mM MgCl₂, 250 mM sucrose plus protease inhibitors) followed by homogenization using Dounce homogenizer (pestle B) 50 times. The concentrations of protease inhibitors were as follows: 1 µg/ml aprotinin, 10 µg/ml leupeptin, 1 µg/ml pepstatin, and 10 µM phenylmethylsulfonyl fluoride. Cellular debris were removed by centrifugation in a microcentrifuge for 2 min followed by centrifugation at 100,000 × g for 45 min to separate cytosol and membrane fractions. Aliquots with equal amount of protein were incubated with either anti-PKC or anti-PKC (BD Transduction Laboratories, Lexington, KY) overnight at 4 °C followed by incubation with protein G-agarose (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) for 2 h. Enzymatic activities of immunoprecipitates containing different PKC isoforms were determined using the protein kinase C assay system from Invitrogen according to the manufacturer's instructions. This assay utilizes a peptide derived from myelin basic protein (amino acid residues 4-14) as a phosphorylation substrate for PKC. The specificity of this PKC assay relies on inclusion of a pseudosubstrate inhibitor peptide (amino acid residues 19-36 of PKC) to distinguish PKC-dependent and nonspecific substrate phosphorylation. To determine whether PKC phosphorylation was altered in cells exposed to toxin A, aliquots of PKC or - immunoprecipitates were analyzed by Western blot probed with anti-phospho-PKC/_II (threonine 638) (Cell Signaling, Beverly, MA).

To determine PLC activation, T84 cells grown on 35-mm wells were incubated in serum-free medium overnight. Cells were exposed to toxin A and then collected in lysis buffer (150 mM NaCl, 50 mM Tris-Cl, pH 7.5, 1% Nonidet P-40, 0.5% sodium deoxycholate, 30 mM sodium pyrophosphate, 50 mM NaF, 1 mM Na₃VO₄) plus protease inhibitors. Lysate was cleared by centrifugation, and PLC was immunoprecipitated with anti-PLC (BD Transduction Laboratories). Samples were separated in a 7.5% SDS-polyacrylamide gel and transferred to a polyvinylidene difluoride membrane (Polyscreen; PerkinElmer Life Sciences). Blots were probed with antibodies to phosphotyrosine (PY20; Upstate Biotechnology, Inc., Lake Placid, NY) and PLC, successively. Blots were then incubated with species-specific secondary antibodies coupled with horseradish peroxidase (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) and developed with Renaissance enhanced luminol reagent (PerkinElmer Life Sciences).

Determination of Toxin A-induced RhoA Glucosylation-- Cells were collected in lysis buffer (20 mM Tris-Cl, pH 7.5, 3 mM MgCl₂, 1 mM EGTA, and 1% Triton X-100) containing protease inhibitors. Cell lysates were cleared by centrifugation at 20,000 × g for 15 min. The level of glucosylated RhoA in samples was analyzed indirectly using toxin A-associated inhibition of C3-mediated ADP-ribosylation (2). C. difficile toxins specifically transfer a glucosyl residue to threonine 37 of RhoA, which blocks further modification of RhoA by C. botulinum C3 exoenzyme that catalyzes ADP-ribosylation of RhoA on asparagine 41 (2). Highly purified C3 exoenzyme was from Biomol (Plymouth Meeting, PA). Following incubation, samples were separated in 12.5% SDS-PAGE and dried prior to film exposure to identify [³²P]ADP-ribosyl RhoA. The corresponding gel bands were excised, and radioactivity was measured by Cerenkov counting.

Cell Fractionation and ZO-1 Immunoprecipitation-- T84 cells (35-mm well) were washed three times with ice-cold phosphate-buffered saline and immediately incubated in 500 µl of Nonidet P-40 extraction buffer (25 mM HEPES, pH 7.4, 150 mM NaCl, 4 mM EDTA, 1% Nonidet P-40, 25 mM NaF, 1 mM Na₃VO₄, 10 mM sodium pyrophosphate, and protease inhibitors) at 4 °C for 5 min with gentle shaking. The detergent-soluble fraction was transferred to a microcentrifuge tube. The insoluble fraction was collected in 100 µl of SDS extraction buffer (25 mM HEPES, pH 7.4, 4 mM EDTA, 1% SDS, 25 mM NaF, 1 mM Na₃VO₄, and 10 mM sodium pyrophosphate) and boiled for 5 min. Supernatants (cytoskeleton-associated fractions) were then obtained after centrifugation (20,000 × g for 30 min) and diluted with equal volume of Nonidet P-40 extraction buffer to reduce SDS concentration in the samples. ZO-1 was immunoprecipitated from detergent-soluble and insoluble fractions with ZO-1 antibody (BD Transduction Laboratories), and the immunoprecipitates were separated in a 7.5% SDS-polyacrylamide gel and then transferred to a polyvinylidene difluoride membrane. Levels of ZO-2, occludin, -catenin, and E-cadherin in the Nonidet P-40-soluble and -insoluble fractions were determined using Western blot. Antibodies against ZO-2 and occludin were from Zymed Laboratories Inc. (South San Francisco, CA). Antibodies for -catenin and E-cadherin were from BD Transduction Laboratories. Band intensity was quantified using ImageQuant from Molecular Dynamics, Inc. (Sunnyvale, CA).

Immunofluorescence Detection of TJ Components-- Confluent T84 cultures on glass coverslips were washed three times in phosphate-buffered saline and then fixed with 4% paraformaldehyde in phosphate-buffered saline for 10 min at room temperature. This was followed by a 10-min incubation with sodium borohydride (0.5 mg/ml) in phosphate-buffered saline to remove free aldehyde. Cells were permeabilized by incubation in CSK buffer (10 mM PIPES, pH 6.8, 50 mM NaCl, 300 mM sucrose, 3 mM MgCl₂, 0.5% Triton X-100) for 10 min. Samples were incubated with primary antibodies (ZO-1, occludin, and claudin-1) at 4 °C overnight and then probed with species-specific secondary antibodies coupled with either FITC or rhodamine (Santa Cruz Biotechnology) for 1 h at room temperature. Antibody to claudin-1 was obtained from Zymed Laboratories Inc.. In some experiments, cells were incubated with rhodamine-phalloidin (Molecular Probes) for 1 h to detect actin filaments. Coverslips were mounted with Vectashield (Vector Laboratories, Inc., Burlingame, CA). The distribution of TJ proteins was examined with a confocal microscope (model 1024; Bio-Rad) using a ×100 oil immersion objective.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Protein Kinase C Antagonists Block Toxin A-mediated Changes in Paracellular Permeability and TER-- A number of intracellular signaling pathways have been implicated in regulating paracellular diffusion across epithelium (15). We tested whether any of these signaling pathways were involved in toxin-mediated changes in paracellular permeability. T84 cells were pretreated for 30 min with either calphostin C (1 µM), wortmannin (100 nM), PD98059 (100 µM), or ML-7 (20 µM), inhibitors of PKC, phosphatidylinositol 3-kinase, mitogen-activated protein kinase/extracellular signal-regulated kinase kinase, and myosin light chain kinase, respectively. Paracellular diffusion of FITC-dextran (M_r 3000) was then measured in the presence of 10 nM toxin A for 5 h (Fig. 1A). Toxin A induced a greater than 3-fold increase in paracellular diffusion compared with control (73.1 ± 3.6 versus 19.9 ± 2.1 pmol dextran/h·cm², n = 18, p < 0.05). This toxin-associated increase in permeability was significantly blocked by pretreating cells with calphostin C, a PKC inhibitor with broad specificity, but not with the other kinase inhibitors tested. We next examined the ability of a cell-permeable MPKCP (peptide sequence derived from amino acid residues 19-27 of PKC and -) to block toxin-induced permeability changes. The toxin A-associated increase in dextran 3K diffusion across TJ barrier was completely blocked by MPKCP (10 µM) (Fig. 1B).

View larger version (21K): [in this window] [in a new window]   Fig. 1.   Effects protein kinase C antagonists on toxin A-induced tight junction permeability changes. A, T84 cells were pretreated with either 1 µM calphostin C, 100 nM wortmannin, 100 µM PD98059, or 20 µM ML-7 for 30 min prior to toxin A (10 nM) exposure for 5 h. Paracellular diffusion of FITC-coupled dextran (Mr 3000) from the apical to basal compartments was determined. A significant increase in paracellular diffusion was observed in cells exposed to toxin A, and this increase was blocked by calphostin C, but not other inhibitors tested. Data are mean ± S.E., and n = 18 in each determination. An asterisk denotes a significant difference between cells treated with toxin only and toxin plus inhibitor (p < 0.05). B, the ability of MPKCP, a specific PKC antagonist, to block toxin A-mediated increase in paracellular diffusion was examined. Confluent T84 monolayer was treated with 10 µM MPKCP for 30 min prior to toxin A (10 nM) exposure for time indicated in the figure. Paracellular permeability of dextran 3K was significantly reduced in the presence of MPKCP compared with toxin A alone. Data are mean ± S.E., and n = 12 in each group. An asterisk denotes a significant difference between control and cells treated with toxin A only (p < 0.05). C, TER was measured in control and in T84 monolayer treated with toxin A (1 nM), MPKCP (10 µM), or both. Confluent T84 monolayer registered TER of 1200-1400 ohms·cm2 were used in these experiments. MPKCP significantly inhibited toxin A-mediated decline in TER. Data are mean ± S.E. of 18 culture inserts for each group. An asterisk denotes significant difference between control and cells treated with toxin A only (p < 0.05).

TER, a marker of ionic flow across the TJ, was more sensitive to toxin A then dextran permeability, as shown by a significant reduction in TER in monolayers exposed to only 1 nM toxin A. A significant drop in TER was first observed at 90 min (Fig. 1C, 1216 ± 72 versus 1012 ± 60 ohm·cm², n = 18, p < 0.05) and continued to decline over the subsequent 2.5 h (445 ± 33 ohm·cm²). As expected from the results shown in Fig. 1B, 10 µM MPKCP completely blocked the decline of TER in cells exposed to toxin A (Fig. 1C).

Toxin A Elevates PKC Activity-- The ability of MPKCP to block toxin A effects suggested the possibility that toxin A activated PKC or - signaling. Since a marked decline in TER was first observed 90 min after toxin A exposure (Fig. 1C), we measured the enzymatic activities of PKC and - in control and in T84 cells after a 1-h exposure to toxin A (1 nM). Significant increases in cytosolic PKC activity (7.3 ± 0.7 versus 13.0 ± 0.7 pmol/min/mg, n = 3, p < 0.05) as well as membrane PKC activity (6.0 ± 0.1 versus 8.6 ± 0.2 pmol/min/mg, n = 3, p < 0.05) were found in T84 cells exposed to toxin A compared with control (Fig. 2A). The activity of PKC in the cytosolic fraction was also increased significantly following toxin A exposure (0.12 ± 0.02 versus 0.29 ± 0.01 pmol/min/mg, n = 3, p < 0.05), while membrane PKC activity was unaltered in response to toxin A (Fig. 2A). These data suggested that the effect of toxin A on TJ permeability required PKC signaling related to either increased PKC phosphorylation (26) or elevated levels of intracellular Ca²⁺ and diacylglycerol, known cofactors for conventional PKC isoforms (27). Western blot analysis of PKC immunoprecipitates with anti-PKC revealed a slight but significant increase in membrane PKC level in cells exposed to toxin A (10 ± 2%, n = 3, p < 0.05), while the cytosolic PKC level was unaltered. In addition, toxin A treatment did not alter the ratio of phosphorylated PKC (threonine 638) versus total PKC in either membrane or cytosolic fractions (data not shown). Consistent with a depressed level of PKC activity (Fig. 2A), we were unable to detect antigen in PKC immunoprecipitates by Western blot, suggesting a low expression level of PKC in T84 cells. PLC, a well characterized regulator of intracellular Ca²⁺ and diacylglycerol, is phosphorylated following agonist-stimulated activation (27). Since increased Ca²⁺ influx and PLC activity have been reported in cells exposed to toxin B (8, 28), we examined whether toxin A altered the phosphorylation status of PLC in T84 cells. PLC phosphorylation was increased by 42 ± 6% (n = 3, p < 0.05) over control within 30 min of toxin A exposure (Fig. 2B), suggesting that toxin A induced PKC activity via activation of PLC.

View larger version (34K): [in this window] [in a new window]   Fig. 2.   Effects of toxin A on PKC activity and PLC phosphorylation. A, the enzymatic activities of PKC and - in the cytosolic and membrane fractions were determined. T84 cells were exposed to 1 nM toxin A for 1 h and separated into cytosol and membrane fractions. PKC and - were immunoprecipitated from each fraction and their activities were determined as described under "Materials and Methods." Data are mean ± S.E. of three determinations. Significant difference (p < 0.05) between control and toxin A-treated groups is denoted by an asterisk. B, phosphorylation of PLC were determined in T84 cells exposed to 1 nM toxin A for time indicated. The phosphorylation level of PLC was determined by probing PLC immunoprecipitates with anti-phosphotyrosine antibody (PY). Representative images of three experiments are shown.

PKC Antagonist Blocks Toxin A-mediated RhoA Glucosylation-- We next examined the role of PKC in toxin A-mediated Rho A glucosylation, since inactivation of RhoA is known to increase TJ permeability in T84 cells (22). The ability of toxin A to glucosylate RhoA was measured indirectly using the C3 exoenzyme assay. C3 exoenzyme specifically ADP-ribosylates native RhoA but not glucosylated RhoA (2). Thus, a decrease in the level of C3 ribosylation of RhoA indicates an increase in the level of glucosylated RhoA in cells exposed to toxin A. Inhibition of ADP-ribosylation in T84 cell lysates in cells exposed to toxin A (1 nM) increased progressively from 2 to 4 h (39 ± 4% of base line) (Fig. 3, A and B). In contrast, ADP-ribosyl RhoA level remained unchanged in cells treated with toxin A plus MPCKP (10 µM) or calphostin C (1 µM) (Fig. 3, A and B). MPKCP or calphostin C alone did not affect the level of ADP-ribosyl RhoA (Fig. 3A).

View larger version (22K): [in this window] [in a new window]   Fig. 3.   PKC antagonists block toxin A-mediated RhoA glucosylation. A, Rho A glucosylation in T84 cells exposed to 1 nM toxin A was measured indirectly using C. botulinum C3-mediated ADP-ribosylation of RhoA. A decreased level of ADP-ribosyl RhoA indicated an increased level of glucosylated Rho A, since glucosylated RhoA is not a substrate of C3 enzyme. The decline in ADP-ribosyl RhoA level was initially observed in T84 cells exposed to toxin A for 2 h. Pretreating T84 cells with 10 µM MPKCP or 1 µM calphostin C abolished the decrease in ADP-ribosyl RhoA induced by toxin A. This experiment was repeated three times. B, [32P]ADP-ribosyl RhoA bands were excised and counted in a liquid scintillation counter. The amount of radioactivity in RhoA in cells exposed to toxin A for 4 h was compared with their respective time 0 controls and data (mean ± S.E., n = 3) are expressed as percentage of control. Significant difference (p < 0.05) between toxin A only and toxin A plus inhibitor was denoted by an asterisk.

Toxin A Induces ZO-1 Translocation via a PKC-dependent Pathway-- To determine whether translocation of TJ proteins was involved in toxin A-mediated and PKC-dependent increase in paracellular permeability, we analyzed the time-dependent distribution of TJ proteins (ZO-1, ZO-2, and occludin) between Nonidet P-40-soluble and -insoluble fractions (Fig. 4). We observed a marked increase in the level of Nonidet P-40-soluble ZO-1 and a decrease in Nonidet P-40-insoluble ZO-1 that began at 2 h of toxin A exposure (Fig. 4A). Densitometric analysis revealed an 89 ± 6% (n = 5) increase in the Nonidet P-40-soluble ZO-1 level and 38 ± 7% (n = 5) decrease in the Nonidet P-40 insoluble ZO-1 level after a 4-h exposure to toxin A (Fig. 4B). MPKCP blocked the translocation of ZO-1 from the cytoskeletal to the detergent-soluble fraction (Fig. 4, A and B). Exposure of T84 cells to MPKCP alone did not affect ZO-1 level in either Nonidet P-40-soluble or -insoluble fractions (Fig. 4A). In contrast to the effect of toxin A on ZO-1 translocation, the level of ZO-2 in either detergent-soluble or -insoluble fraction was unchanged (Fig. 4C). We next examined phosphorylation and translocation of occludin from cytoskeletal to detergent-soluble fraction. In control T84 cells (time 0), both phosphorylated (high M_r form) and nonphosphorylated (low M_r form) occludin was detected in the Nonidet P-40-insoluble fraction, while only the nonphosphorylated form was found in the detergent-soluble fraction (Fig. 4C). Toxin A exposure (1-4 h) did not induce either translocation or dephosphorylation of occludin (Fig. 4C). Moreover, the levels of E-cadherin and -catenin in either Nonidet P-40-soluble or -insoluble fraction remained unchanged in cells exposed to toxin A (Fig. 4C). Thus, ZO-1 translocation was selectively induced by toxin A and correlated temporally with TJ permeability changes. Both ZO-1 translocation and TJ permeability changes were dependent on PKC signaling.

View larger version (38K): [in this window] [in a new window]   Fig. 4.   Distribution of ZO-1 in Nonidet P-40-soluble and insoluble fractions in response to toxin A. A, T84 cells were pretreated with PKC antagonist (10 µM MPKCP) for 30 min followed by toxin A (1 nM) exposure for the time indicated. Cells were fractionated into Nonidet P-40-soluble and -insoluble fractions. ZO-1 was immunoprecipitated and analyzed by Western blot. The ZO-1 level in the Nonidet P-40 soluble fraction was increased in cells exposed to toxin A in a time-dependent manner with a concomitant decrease in the Nonidet P-40-insoluble fraction. Toxin A-induced ZO-1 translocation was blocked by MPKCP. A representative experiment of six analyses is shown. B, ZO-1 levels (band intensity) in cells exposed to toxin A for 0 and 4 h were quantified with a densitometer. Data (mean ± S.E., n = 6) were expressed as percentage of their respective time 0 controls. An asterisk indicates significant difference (p < 0.05) from toxin A-treated samples. C, aliquots of cell lysate obtained in A were used to determine the levels of TJ proteins (ZO-2 and occludin) and adherence junction (AJ) proteins (E-cadherin and -catenin) using Western blot. Occludin was separated into high molecular weight (HMW) and low molecular weight (LMW) bands representing phosphorylated and nonphosphorylated forms of occludin, respectively. There was no apparent change in the levels of these proteins and occludin phosphorylation following toxin A exposure in either Nonidet P-40-soluble or -insoluble fractions. Data are representative of six determinations.

Actin Filaments and TJ Proteins Are Redistributed in Cells Exposed to Toxin A-- The cytotoxicity of toxin A in fibroblasts is characterized by disassembly of actin filaments, detachment of cells from the underlying matrix, and eventually cell rounding (3). Extensive cell rounding was observed in T84 cells exposed to 1 nM toxin A for 8 h (data not shown). Cell rounding was first observed in scattered clusters of ~100-200 cells (Fig. 5). MPKCP completely blocked this toxin A-mediated cytotoxic effect (Fig. 5). The distribution of TJ proteins in cells surrounding these clusters was examined by immunofluorescence staining, since these neighboring cells might provide an early indication of TJ protein distribution prior to cell rounding. Confocal analyses of actin filaments, claudin-1, ZO-1, and occludin were performed in control T84 cells and in T84 cells treated with toxin A alone, MPKCP alone, or toxin A plus MPKCP (Figs. 6 and 7). T84 cells treated with MPKCP alone were identical to control monolayers (data not shown). Subapical actin filaments formed ringlike structures distributed along the cell periphery in control cells (Fig. 6A) in contrast to scattered aggregates or clumps in cells exposed to 1 nM toxin A for 4 h (Fig. 6B). Actin filaments at the base of control cells appeared as elongated microfilaments (Fig. 6D), which became disaggregated after toxin A exposure (Fig. 6E). Claudin-1 formed organized fibrils at the cell-cell junction in control cells (Fig. 6G) but formed aggregates with fringed extensions in cells exposed to toxin A (Fig. 6H). As shown in the bottom panels of Fig. 6, MPKCP completely blocked the effects of toxin A on disruption of actin filaments and claudin-1 (Fig. 6, C, F, and I).

View larger version (69K): [in this window] [in a new window]   Fig. 5.   Effects of MPKCP on toxin A-induced morphological change. Phase-contrast images of control T84 monolayer and cells exposed to either toxin A (1 nM) alone or toxin A plus MPKCP (10 µM) for 4 h were taken using a ×20 objective. Toxin A-induced cell rounding was observed in local regions (marked with asterisks) ranging from 100 to 200 cells in size, while greater than 95% of the monolayer remained intact. The rounded cells were out of focus compared with the surrounding area. No cell rounding was observed in cells pretreated with MPKCP. Data presented are representative images of six experiments.

View larger version (170K): [in this window] [in a new window]   Fig. 6.   Effects of MPKCP on toxin A-induced reorganization of actin filaments and claudin-1. Distribution of subapical (A-C) and basal (D-F) actin filaments and claudin-1 (G-I) in the area immediately surrounding rounded cells was analyzed using a confocal microscope (×100 objective). Images of control T84 cells (A, D, and G) and cells exposed to 1 nM toxin A alone (B, E, and H) or toxin A plus 10 µM MPKCP (C, F, and I) for 4 h were shown. Data presented are representative images of eight experiments.

ZO-1 is a scaffolding protein connecting perijunctional actin filaments with TJ fibril proteins, claudins, and occludin (13). Since we observed severe disruption of claudin-1 and actin filaments in cells exposed to toxin A (Fig. 6) as well as dissociation of ZO-1 from the cytoskeletal fraction (Fig. 4A), we expected both ZO-1 and occludin to be displaced from their respective TJ localizations. TJ-associated ZO-1, localized to the cell periphery in control cells (Fig. 7A) in a pattern similar to that observed for actin (Fig. 6A). This pattern was completely lost in cells exposed to toxin A, in which ZO-1 was distributed in irregular scattered clumps (Fig. 7B). Occludin displayed a similar distribution at the cell periphery (Fig. 7D) where it was colocalized with ZO-1 in control T84 cells (Fig. 7G). Toxin A produced disaggregation and clumping of occludin (Fig. 7E) and loss of association of ZO-1 and occludin from the TJ (Fig. 7H). The extensive disorganization and redistribution of ZO-1 and occludin in T84 cells exposed to toxin A was completely blocked in cells pretreated with MPKCP (Fig. 7, C, F, and I).

View larger version (83K): [in this window] [in a new window]   Fig. 7.   Effects of MPKCP on toxin A-induced ZO-1 and occludin distribution. Distribution of ZO-1 and occludin in control T84 cells (A, D, and G) and in cells exposed to either 1 nM toxin A (B, E, and H) alone or toxin A plus 10 µM MPKCP (C, F, and I) for 4 h were analyzed. Cells were double-stained with antibodies against ZO-1 (A, B, and C) and occludin (D, E, and F). This was followed by staining with species-specific secondary antibodies coupled with either FITC (green; ZO-1) or rhodamine (red; occludin). Images were taken and focused at the surrounding area of rounded cells. Data presented are representative images of eight experiments.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The present study provides evidence that the PKC signaling pathway mediates the increase in paracellular permeability of T84 colonocyte monolayers exposed to C. difficile toxin A. We and others have previously reported that this toxin alters TJ permeability in intestinal monolayers (23, 24), as well as in ileal and colonic loops of intact animals exposed to toxin A intraluminally (29, 30). The toxic effect on paracellular permeability has been attributed to the inactivation of RhoA, an essential regulator of filamentous actin (23). Loss of organization in the perijunctional F-actin ring following inactivation of RhoA by toxin A or by C. botulinum C3 transferase exoenzyme has been identified as a critical event in the mechanism of lowered TER and increased paracellular flux of solutes in toxin-exposed intestinal monolayers (22-24). We also observed in the current study that activation of PLC and PKC/ preceded the toxin-associated alterations of paracellular flux and TER. Moreover, TJ dysfunction as well as cytoskeletal changes involving actin, ZO-1, occludin, and claudin-1 were ameliorated or prevented when monolayers were incubated with the PKC inhibitors, MPKCP or calphostin C.

A substantial body of experimental data indicates that PKC regulates paracellular permeability in epithelial (17, 18) and endothelial (31) monolayers. The PKC activators diacylglycerol and phorbol esters increase paracellular permeability and diminish TER (15, 32). PKC regulates the sorting and assembly of the TJ proteins ZO-1 and the development of TER in confluent renal epithelial cells (17, 18). In pulmonary epithelial monolayers, PKC activation modulates TNF-induced increase in paracellular permeability (31). Similarly, it has been reported (17, 18) that PKC and PKC expression were correlated with tight junctional leakiness in renal epithelial cells. Marano et al. (33) reported that phorbol ester increased paracellular permeability in IEC-18 intestinal cells by a mechanism involving translocation of PKC from the cytosolic to membrane compartment and phosphorylation of the tight junctional protein occludin. Caco-2 monolayers express the classic PKC isoforms , I and II, , and  (34). Dodane and Kachar (35) reported that PKC was the only isoform that colocalized with ZO-1 at tight junction of Madin-Darby canine kidney and Caco-2 cells. However, colonocyte crypt cells of rat and mouse express PKC isoforms , II, , and  predominantly in the cytosolic compartment (36). As colonocytes migrate to the surface level and differentiate, PKC isoenzymes expression increases especially in the membrane/cytoskeletal compartment (36). In this study, we found elevated PLC phosphorylation after 30 min of toxin A exposure that might be responsible for the observed increases in PKC activity (Fig. 2). Toxin A caused an increase in ZO-1 translocation from TJ to the detergent soluble fraction (Fig. 4) that correlated temporally with increased paracellular permeability and diminished TER within 2-3 h of toxin exposure (Fig. 1). The distribution and localization of tight junction proteins claudin-1, ZO-1, and occludin underwent profound disorganization in scattered clusters of T84 cells following toxin A exposure for 4 h (Figs. 6 and 7). All these toxin A effects were blocked by the PKC inhibitor MPKCP, a relatively specific PKC/ inhibitor. In control T84 cells, PKC activities were 60- and 10-fold greater than that of PKC in cytosol and membrane fractions, respectively (Fig. 2A). In addition, overexpressing dominant negative PKC in a renal epithelial cell line (LLC-PK₁) blocked the loss of transepithelial electrical resistance induced by PKC agonist (17). These data suggested that the effects of toxin A on TJ functions and organization in T84 cells are most likely mediated by PKC. However, a significant role of PKC cannot be ruled out on the basis of the current data.

We previously suggested that a subtle alteration in TJ composition might be responsible for the toxin A-induced decline in TER that occurred in the absence of any detectable changes in cell morphology and TJ ultrastructure (24). In this study, toxin A-induced ZO-1 translocation correlated temporally with a decline in TER (Fig. 1C), suggesting that the loss of ZO-1 from TJ in T84 cells might mediate the decrease in TJ barrier function. Since ZO-1 binds to the carboxyl terminus of claudins (37), it is possible that toxin A-mediated translocation of ZO-1 could trigger the loosening of claudin contacts between adjacent epithelial cells, resulting in decreased TER. The association of ZO-2 or occludin with actin cytoskeleton might stabilize TJ structure and maintain cell morphology in cells exposed to toxin A for 2-3 h, since toxin A did not induce translocation in any of these TJ proteins (Fig. 4C).

Our results indicate that PKC is an upstream regulator of toxin A-mediated inactivation of RhoA, since toxin A-induced PKC activation (1 h) preceded RhoA glucosylation (2 h) (Figs. 2 and 3). Moreover, PKC inhibitors blocked RhoA glucosylation in cells exposed to toxin A (Fig. 3). It has been shown that guanidine nucleotide dissociation inhibitor (GDI) blocks the translocation of RhoA from cytosol to membrane (GTP-bound, active form) by binding with high affinity to GDP-bound RhoA in the cytosol (38). In addition, RhoA complexed with GDI is not glucosylated by toxin (38). A recent report showed that PKC induces GDI phosphorylation and dissociation of GDI-RhoA complex, leading to RhoA activation (39). Thus, inactivation of PKC signaling by exposing T84 cells to MPKCP might block the dissociation of GDI-RhoA complex and make RhoA unavailable for toxin A-mediated glucosylation. Alternatively, toxin A could be activated by phosphorylation via a PKC-dependent pathway. Several potential PKC phosphorylation sites are present in toxin A by scanning its sequence against PROSITE (Swiss Institute of Bioinformatics; available on the Internet at www.expasy.ch). Phosphorylation of other bacterial proteins or toxins by host cell kinases has also been reported (40, 41).

Several other important enteric pathogens produced alterations in TJ permeability via PKC-dependent pathways. Philpott et al. (10) reported that enterohemorrhagic Escherchia coli increased transepithelial flux and lowered TER of T84 monolayers. These functional changes were preceded by an increase in membrane-associated PKC activity and could be prevented by a PKC inhibitor as well as by inhibitors of calmodulin and myosin light chain kinase (10). As in our study, the effect of enterohemorrhagic E. coli on barrier function was mediated by disruption of tight junction and diminished intensity of staining for ZO-1 but not for E-cadherin. PKC inhibition also completely inhibits the ability of T84 cells exposed to Salmonella typhimurium to direct the transepithelial migration of neutrophils (42). Helicobacter pylori disrupted the barrier function of T84 cells in a time-dependent course (43) similar to the one described here for C. difficile toxin A. However, in this model system, the H. pylori-associated drop in TER was blocked by the PKC activator phorbol myristate acetate, suggesting that H. pylori alters barrier function by interfering with a normal PKC-dependent pathway that maintains epithelial barrier properties (43). The current observations on the effects of toxin A on TJ permeability in T84 cells are consistent with our prior observations using human colonic tissues (9). However, future studies will be necessary to confirm and extend these observations to intact human tissues.

The data presented here on PKC signaling as well as previous studies by us and others allow us to propose a scheme of the early cellular signaling events that follow toxin A binding to its membrane receptor. During the first 30 min, phosphorylation of PLC occurs. In HT29 intestinal cells, we have also observed generation of reactive oxygen species and a drop in intracellular ATP levels between 15 and 30 min following toxin A exposure (7). In the current study, we observed significant increases in membrane and cytosolic PKC as well as cytosolic PKC activities at 1 h, followed by a significant decline in TER at 90 min. Translocation of ZO-1 from the cytoskeletal to the detergent-soluble fraction was observed at 2 h. Between 2 and 3 h, glucosylation of RhoA was observed. Morphological evidence of cell damage was not observed until 4 h, when both ZO-1 and occludin were clumped and disorganized, compared with their typical localization in subapical belts at the periphery of T84 cells. We also observed disorganization of actin filament structure and claudin-1 after 4 h of toxin A exposure. Inhibition of PKC/ activity blocked toxin A-induced RhoA glucosylation, ZO-1 translocation, and subsequently disorganization of TJ structure. Together, these data suggest that PKC signaling plays an important role in toxin A-mediated damage on TJ structure and functions.

	FOOTNOTES

^* This work was supported by National Institute of Health Grants (DK 34583 and DK 47343) and a pilot feasibility study award from the Center for the Study of Inflammatory Bowel Disease at Massachusetts General Hospital (DK 43351) (to M. L. C.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

This work has been presented in a preliminary form at the American Gastroenterological Association Digestive Disease Week in Atlanta, Georgia, May 20, 2001.

To whom correspondence should be addressed: Division of Gastroenterology, DANA 501, Beth Israel Deaconess Medical Center, 330 Brookline Ave., Boston, MA 02215. Tel.: 617-667-8377; Fax: 617-975-5071; E-mail: jlamont@caregroup.harvard.edu.

Published, JBC Papers in Press, November 29, 2001, DOI 10.1074/jbc.M109254200

	ABBREVIATIONS

The abbreviations used are: TJ, tight junction; PKC, protein kinase C; TER, transepithelial electrical resistance; MPKCP, myristoylated PKC/ peptide; FITC, fluorescein isothiocyanate; FCS, fetal calf serum; PLC, phospholipase C; PVDF, polyvinylidene difluoride; PIPES, 1,4-piperazinediethanesulfonic acid; GDI, guanidine nucleotide dissociation inhibitor.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				M. Moreto and A. Perez-Bosque Dietary plasma proteins, the intestinal immune system, and the barrier functions of the intestinal mucosa J Anim Sci, April 1, 2009; 87(14_suppl): E92 - E100. [Abstract] [Full Text] [PDF]

				Y. Grumbach, N. V. T. Quynh, R. Chiron, and V. Urbach LXA4 stimulates ZO-1 expression and transepithelial electrical resistance in human airway epithelial (16HBE14o-) cells Am J Physiol Lung Cell Mol Physiol, January 1, 2009; 296(1): L101 - L108. [Abstract] [Full Text] [PDF]

				X. Na, H. Kim, M. P. Moyer, C. Pothoulakis, and J. T. LaMont gp96 Is a Human Colonocyte Plasma Membrane Binding Protein for Clostridium difficile Toxin A Infect. Immun., July 1, 2008; 76(7): 2862 - 2871. [Abstract] [Full Text] [PDF]

				I. Beau, J. Cotte-Laffitte, R. Amsellem, and A. L. Servin A Protein Kinase A-Dependent Mechanism by Which Rotavirus Affects the Distribution and mRNA Level of the Functional Tight Junction-Associated Protein, Occludin, in Human Differentiated Intestinal Caco-2 Cells J. Virol., August 15, 2007; 81(16): 8579 - 8586. [Abstract] [Full Text] [PDF]

				H. Kohler, T. Sakaguchi, B. P. Hurley, B. J. Kase, H.-C. Reinecker, and B. A. McCormick Salmonella enterica serovar Typhimurium regulates intercellular junction proteins and facilitates transepithelial neutrophil and bacterial passage Am J Physiol Gastrointest Liver Physiol, July 1, 2007; 293(1): G178 - G187. [Abstract] [Full Text] [PDF]

				B. Zheng and L. C. Cantley Regulation of epithelial tight junction assembly and disassembly by AMP-activated protein kinase PNAS, January 16, 2007; 104(3): 819 - 822. [Abstract] [Full Text] [PDF]

				A. Perez-Bosque, C. Amat, J. Polo, J. M. Campbell, J. Crenshaw, L. Russell, and M. Moreto Spray-Dried Animal Plasma Prevents the Effects of Staphylococcus aureus Enterotoxin B on Intestinal Barrier Function in Weaned Rats J. Nutr., November 1, 2006; 136(11): 2838 - 2843. [Abstract] [Full Text] [PDF]

				S. M. Stamatovic, O. B. Dimitrijevic, R. F. Keep, and A. V. Andjelkovic Protein Kinase C{alpha}-RhoA Cross-talk in CCL2-induced Alterations in Brain Endothelial Permeability J. Biol. Chem., March 31, 2006; 281(13): 8379 - 8388. [Abstract] [Full Text] [PDF]

				A. Hashim, G. Mulcahy, B. Bourke, and M. Clyne Interaction of Cryptosporidium hominis and Cryptosporidium parvum with Primary Human and Bovine Intestinal Cells Infect. Immun., January 1, 2006; 74(1): 99 - 107. [Abstract] [Full Text] [PDF]

				C. Jacob, P.-C. Yang, D. Darmoul, S. Amadesi, T. Saito, G. S. Cottrell, A.-M. Coelho, P. Singh, E. F. Grady, M. Perdue, et al. Mast Cell Tryptase Controls Paracellular Permeability of the Intestine: ROLE OF PROTEASE-ACTIVATED RECEPTOR 2 AND {beta}-ARRESTINS J. Biol. Chem., September 9, 2005; 280(36): 31936 - 31948. [Abstract] [Full Text] [PDF]

				The PKC-DRS Study Group The Effect of Ruboxistaurin on Visual Loss in Patients With Moderately Severe to Very Severe Nonproliferative Diabetic Retinopathy: Initial Results of the Protein Kinase C {beta} Inhibitor Diabetic Retinopathy Study (PKC-DRS) Multicenter Randomized Clinical Trial Diabetes, July 1, 2005; 54(7): 2188 - 2197. [Abstract] [Full Text] [PDF]

				D. E. Voth and J. D. Ballard Clostridium difficile Toxins: Mechanism of Action and Role in Disease Clin. Microbiol. Rev., April 1, 2005; 18(2): 247 - 263. [Abstract] [Full Text] [PDF]

				B. Ozdamar, R. Bose, M. Barrios-Rodiles, H.-R. Wang, Y. Zhang, and J. L. Wrana Regulation of the Polarity Protein Par6 by TGF{beta} Receptors Controls Epithelial Cell Plasticity Science, March 11, 2005; 307(5715): 1603 - 1609. [Abstract] [Full Text] [PDF]

				J. M. Smith, P. A. Johanesen, M. K. Wendt, D. G. Binion, and M. B. Dwinell CXCL12 activation of CXCR4 regulates mucosal host defense through stimulation of epithelial cell migration and promotion of intestinal barrier integrity Am J Physiol Gastrointest Liver Physiol, February 1, 2005; 288(2): G316 - G326. [Abstract] [Full Text] [PDF]

				L. S. Bertelsen, G. Paesold, S. L. Marcus, B. B. Finlay, L. Eckmann, and K. E. Barrett Modulation of chloride secretory responses and barrier function of intestinal epithelial cells by the Salmonella effector protein SigD Am J Physiol Cell Physiol, October 1, 2004; 287(4): C939 - C948. [Abstract] [Full Text] [PDF]

				J. Yoo, A. Nichols, J. Mammen, I. Calvo, J. C. Song, R. T. Worrell, K. Matlin, and J. B. Matthews Bryostatin-1 enhances barrier function in T84 epithelia through PKC-dependent regulation of tight junction proteins Am J Physiol Cell Physiol, August 1, 2003; 285(2): C300 - C309. [Abstract] [Full Text] [PDF]

				F. Hollande, D. J. Lee, A. Choquet, S. Roche, and G. S. Baldwin Adherens junctions and tight junctions are regulated via different pathways by progastrin in epithelial cells J. Cell Sci., April 1, 2003; 116(7): 1187 - 1197. [Abstract] [Full Text] [PDF]

				J. Yoo, A. Nichols, J. C. Song, J. Mammen, I. Calvo, R. T. Worrell, J. Cuppoletti, K. Matlin, and J. B. Matthews Bryostatin-1 attenuates TNF-induced epithelial barrier dysfunction: role of novel PKC isozymes Am J Physiol Gastrointest Liver Physiol, April 1, 2003; 284(4): G703 - G712. [Abstract] [Full Text] [PDF]

				J Berkes, V K Viswanathan, S D Savkovic, and G Hecht Intestinal epithelial responses to enteric pathogens: effects on the tight junction barrier, ion transport, and inflammation Gut, March 1, 2003; 52(3): 439 - 451. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 277/6/4247    most recent M109254200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Chen, M. L. Articles by LaMont, J. T. Search for Related Content PubMed PubMed Citation Articles by Chen, M. L. Articles by LaMont, J. T. Social Bookmarking                      What's this?