If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
* This study was supported by National Institutes of Health Grants R01-DK55532 and R01-AA12307 (to R. K. R.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
A recent study (Nusrat, A., Chen, J. A., Foley, C. S., Liang, T. W., Tom, J., Cromwell, M., Quan, C., and Mrsny, R. J. (2000) J. Biol. Chem. 275, 29816–29822) suggested that phosphatidylinositol 3-kinase (PI 3-kinase) may interact with occludin; however, there exists no evidence of direct interaction of PI 3-kinase with the tight junctions. Activation of PI 3-kinase by oxidative stress and its role in disruption of tight junctions was examined in Caco-2 cell monolayer. The oxidative stress-induced decrease in electrical resistance, increase in inulin permeability, and redistribution of occludin and ZO-1 were reduced by a PI 3-kinase inhibitor, LY294002. Oxidative stress-induced tyrosine phosphorylation and dissociation from the actin cytoskeleton of occludin and ZO-1 were reduced by LY294002. The regulatory subunit of PI 3-kinase, p85, and the PI 3-kinase activity were co-immunoprecipitated with occludin, which were rapidly increased by oxidative stress. Oxidative stress resulted in increased translocation of p85 from the intracellular compartment into the intercellular junctions. Pair-wise glutathione S-transferase pull-down assay showed that glutathione S-transferase-occludin (C-terminal tail) binds to recombinant p85. This study shows that oxidative stress increases the association of PI 3-kinase with the occludin, and that PI 3-kinase activity is involved in oxidative stress-induced disruption of tight junction.
Epithelial tight junctions form a barrier to the movement of pathogens, toxins, and allergens from the intestinal lumen into the tissue. The disruption of tight junctions plays an important role in the pathogenesis of a number of gastrointestinal diseases (
). Growing evidence indicates that the activities of intracellular signaling molecules regulate tight junctions. Studies indicate that signaling pathways involving protein kinases, G-proteins, and Rho/Rac GTPases regulate the tight junction permeability (
), raising the possibility of the role of PI 3-kinase-dependent signaling pathway in the regulation of tight junctions. A recent study showed that PI 3-kinase inhibitors prevented vascular endothelial growth factor-induced Ser/Thr-phosphorylation and redistribution of occludin and ZO-1 in bovine aortic endothelial cells (
). Oxidative stress-induced paracellular permeability was inhibited by tyrosine kinase inhibitors and was associated with tyrosine phosphorylation of a wide spectrum of proteins, including occludin, ZO-1, E-cadherin, and β-catenin (
) it was demonstrated that the expression of kinase-inactive c-Src mutant delays the oxidative stress-induced disruption of tight junctions in Caco-2 cell monolayers, indicating the important role of c-Src in regulation of tight junction.
In the present study we examined the role of PI 3-kinase in the oxidative stress-induced disruption of tight junctions in Caco-2 cell monolayers. Results show that: 1) inhibitor of PI 3-kinase activity reduces the oxidative stress-induced disruption of tight junctions; 2) oxidative stress increases the association of PI 3-kinase with occludin, and induces translocation of p85 to the intercellular junction; and 3) occludin interacts directly with p85. This study for the first time shows that PI 3-kinase is associated with occludin in Caco-2 cell monolayers, and plays an important role in the regulation of epithelial tight junctions by oxidative stress.
MATERIALS AND METHODS
Chemicals—Cell culture reagents and supplies were purchased from Invitrogen. FITC-inulin, vanadate, SDS, xanthine oxidase, xanthine, genistein, protease inhibitors, streptavidin agarose, protein A-Sepharose, and protein G-Sepharose were purchased from Sigma. LY294002 was from Calbiochem (San Diego, CA). Phosphatidylinositol was purchased from Avanti Polar Lipids (Alabaster, AL), and [γ-32P]ATP was from ICN Radiochemicals (Irvine, CA). All other chemicals were of analytical grade purchased either from Sigma or Fisher.
Antibodies—Mouse monoclonal anti-p85, recombinant HRP-conjugated anti-Tyr(p), biotin-conjugated anti-Tyr(p), anti-mouse IgG, and HRP-conjugated anti-rabbit IgG antibodies were purchased from Transduction Laboratories (Lexington, KY). Rat monoclonal anti-ZO-1 antibody was purchased from Chemicon International Inc. (Temecula, CA). Mouse monoclonal anti-occludin, rabbit polyclonal anti-ZO-1, HRP-conjugated anti-occludin, Cy3-conjugated anti-rabbit IgG, AlexaFluor 488-conjugated anti-mouse IgG, and Oregon Green-conjugated anti-rat IgG antibodies were from Molecular Probes (Eugene, OR).
Cell Culture—Caco-2 and MDCK cells purchased from American Type Cell Collection (Rockville, MD) were grown under standard cell culture conditions as described previously (
). Cells were grown on polycarbonate membranes in transwells (6.5 mm, 12 mm, or 24 mm; Costar, Cambridge, MA), and experiments conducted on 11–13 days (6.5 or 12 mm transwells) or 17–19 days (24 mm transwell) after seeding.
Treatment with Oxidative Stress—Oxidative stress was induced as previously described (
). Briefly, cell monolayers were incubated in phosphate-buffered saline (Dulbecco's saline containing 1.2 mm CaCl2, 1 mm MgCl2, and 0.6% bovine serum albumin) in the absence or presence of a mixture of xanthine oxidase (20 milliunits/ml) and xanthine (0.25 mm) (XO+X) with or without LY294002 (25 μm). Control cell monolayers were incubated in phosphate-buffered saline without XO+X and inhibitors.
Measurement of Transepithelial Electrical Resistance (TER)—TER was measured as described previously (
) using a Millicell-ERS electrical resistance system (Millipore, Bedford, MA). TER calculated as ohms/cm2 by multiplying it with the surface area of the monolayer. The resistance of the polycarbonate membrane in transwells (∼30 ohms/cm2) was subtracted from all readings.
Unidirectional Flux of Inulin—Transwells with the cell monolayers were incubated under different experimental conditions in the presence of FITC-inulin (0.5 mg/ml) in the basal well. At different times after XO+X treatment, 100 μl each of apical and basal media were withdrawn, and fluorescence measured using a fluorescence plate reader (BioTEK Instruments, Winooski, VT). The flux into the apical well was calculated as the percent of total fluorescence administered into the basal well per hour per cm2 surface area.
Immunofluorescence Microscopy—After treatment with XO+X in the absence or presence of LY294002 for varying times Caco-2 cell monolayers (12 mm) were washed in phosphate-buffered saline and fixed in acetone/methanol (1:1) at 0 °C for 5 min. Cell monolayers were blocked in 3% nonfat milk in TBST (20 mm Tris, pH 7.2, and 150 mm NaCl) and incubated for one hour with primary antibodies, rabbit polyclonal anti-occludin, mouse monoclonal anti-p85, and rat monoclonal anti-ZO-1 antibodies, followed by incubation for one hour with secondary antibodies, Oregon Green-conjugated anti-rat IgG, AlexaFluor 488-conjugated anti-mouse IgG, and Cy3-conjugated anti-rabbit IgG antibodies. The fluorescence was examined by a confocal laser scanning microscopy (Bio-Rad MRC1024), and images from Z-series sections (1 μm) were collected by using comos (confocal microscope operating system). Images were stacked using the software, Confocal Assistant 4.02, and processed by Adobe Photoshop (Adobe Systems Inc., San Jose, CA).
Preparation of Cytoskeletal Fractions—Cell monolayers in transwell (24 mm) were washed twice with ice-cold phosphate-buffered saline, and incubated for 5 min with lysis buffer-CS (Tris buffer containing 1.0% Triton X-100, 2 μg/ml leupeptin, 10 μg/ml aprotinin, 10 μg/ml bestatin, 10 μg/ml pepstatin-A, 1 mm vanadate, and 1 mm phenylmethylsulfonyl fluoride). Cell lysates were centrifuged at low speed 15,600 × g for 4 min at 4 °C to sediment the high-density actin cytoskeleton. The pellet was suspended in 200 μl of lysis buffer-CS. Protein contents in different fractions were measured by BCA method (Pierce Biotechnology Inc. (Rockford, IL). Cytoskeletal and Triton-soluble fractions were mixed with equal volumes of Laemmli's sample buffer (2× concentrated) and heated at 100 °C for 5 min.
Immunoprecipitation—After XO+X treatment for varying times Caco-2 cell monolayers (24 mm) were washed with ice-cold 20 mm Tris (pH 7.4) and actin cytoskeleton and Triton-soluble fraction were prepared. Actin suspension was sonicated for 10 s in lysis buffer-CS. Actin lysate and Triton-soluble fraction (1.0 mg protein/ml) were incubated with 2 μg of anti-occludin antibodies at 4 °C for 16 h. Immune complexes were isolated by precipitation using protein A-Sepharose (for 1 h at 4 °C). Washed beads were suspended in 20 μl of assay buffer to measure PI 3-kinase activity, or heated in Laemmli's sample buffer for immunoblot analysis.
For tyrosine phosphorylation studies, cytoskeletal fractions were extracted in lysis buffer D (0.3% SDS in 10 mm Tris buffer, pH 7.4, containing 1 mm vanadate and 0.33 mm phenylmethylsulfonyl fluoride) by heating at 100 °C for 5 min. For co-immunoprecipitation of PI 3-kinase with occludin, cytoskeletal fractions were extracted in lysis buffer N (20 mm Tris, pH 7.4, containing 0.2% NP40, 0.1% sodium deoxycholate and cocktail of protease inhibitors as described above for lysis buffers-CS) at 4 °C for 30 min. Cytoskeletal extracts were incubated overnight at 4 °C with 2 μg of biotin-conjugated anti-Tyr(p) or 2 μg of rabbit polyclonal anti-occludin antibodies. Immunoprecipitation was carried out overnight as described above. Immune complexes were precipitated by incubation for one hour with streptavidin-agarose or protein A-Sepharose at 4 °C. Anti-Tyr(p) immune complexes were immunoblotted for occludin and ZO-1. Anti-occludin immunoprecipitates were immunoblotted for occludin and p85 or used for PI 3-kinase assay.
Immunoblot analysis—Proteins were separated by SDS-polyacrylamide gel (4–12% gradient) electrophoresis and transferred to polyvinylidene difluoride membranes. Membranes blotted for occludin, ZO-1, and p85 by using specific antibodies in combination with HRP-conjugated anti-mouse IgG or HRP-conjugated anti-rabbit IgG antibodies. The blot was developed using ECL chemiluminescence method (Amersham Biosciences).
PI 3-Kinase Assay—PI 3-kinase assay was carried out as described by Avanti Polar Lipids (Alabaster, AL). The occludin immune complexes were incubated in a 50 μl assay system consisting of 25 mm MOPS buffer, pH 7.0, 5 mm MgCl2, 1 mm EGTA, 1 mm sodium orthovanadate, 40 μg of phosphatidylinositol substrate (presonicated), and 150 μm ATP containing 10 μCi [γ-32]PATP. Reaction mixture was incubated at 37 °C for 30 min. Reaction was stopped by the addition of two volumes of 6 m HCl in methanol. Lipids were extracted with chloroform and then separated by thin layer chromatography on calcium depleted, activated silica gel 60 (Whatman, Maidstone, England) using water/n-propanol/acetic acid (34:65:1, v/v/v) solvent system. TLC plates were exposed to x-ray films to determine the level of incorporation of 32P to substrate.
Preparation of GST-C-Occludin and GST-p85—C-terminal tail of chicken occludin as a GST fusion protein, GST-C-occludin, and GST-p85 were prepared in Escherichia coli DH5α cells and purified using glutathione (GSH)-agarose as described previously (
). cDNA for C-terminal tail of occludin (amino acids 354–503) in pGEX vector was a gift from Dr. J. M. Anderson and A. Fanning, University of North Carolina (Chapel Hill, NC), and the cDNA for p85 in pGEX vector was kindly provided by Dr. Marcello Arsura (Department of Pharmacology, University of Tennessee, Memphis, TN).
Pair-wise Binding Assay—This particular assay detects binding between two individually purified proteins. For this purpose, we generated occludin and p85 as GST fusion proteins (e.g. GST-occludin C-tail, amino acids 354–503; and GST-p85, amino acids 1–724). The GST portion of GST-p85 was clipped off with thrombin. Binding assays were performed using GST-C-occludin and thrombin-cleaved p85, and excess glutathione-Sepharose beads was used to “pull down” the bound complex. Complexes were then immunoblotted for p85. This assay determines direct interaction between two proteins.
PI 3-Kinase Activity Is Required for the Oxidative Stress-induced Disruption of Tight Junctions—To determine the role of PI 3-kinase in the regulation of tight junctions, the effect of the LY294002, a PI 3-kinase inhibitor, on oxidative stress-induced disruption of tight junctions was evaluated. Treatment of cell monolayers with XO+X resulted in a time-dependent decrease in TER (Fig. 1A) and increase in inulin permeability (Fig. 1B). Pretreatment of cell monolayers with LY294002 significantly reduced the XO+X-induced decrease in TER and increase in inulin permeability. The effect of LY294002 on XO+X-induced changes in TER and inulin permeability was concentration-dependent (Fig. 1, C and D). Immunofluorescence confocal microscopy showed that XO+X induced a redistribution of occludin and ZO-1 from the intercellular junctions (Fig. 2), which was prevented by pretreatment of cell monolayers with LY294002.
PI 3-Kinase Inhibitor Prevents Oxidative Stress-induced Tyrosine Phosphorylation of Occludin and ZO-1 and Their Dissociation from the Actin Cytoskeleton—Previous studies (
) showed that oxidative stress-induced disruption of tight junctions in Caco-2 cell monolayers was associated with tyrosine phosphorylation of occludin and ZO-1 and their release from the actin cytoskeleton. Therefore, the present study evaluated the effect of LY294002 on XO+X-induced changes in tyrosine phosphorylation and cytoskeletal association of tight junction proteins. XO+X treatment resulted in an increase in tyrosine phosphorylation of occludin and ZO-1 in the actin cytoskeleton, the membrane cytoskeleton, and the Triton-soluble fractions. LY294002 reduced XO+X-induced tyrosine phosphorylation of occludin and ZO-1 in all fractions (Fig. 3A). XO+X also induced a time-dependent reduction in the amounts of occludin and ZO-1 associated with the actin cytoskeleton, which was accompanied by an increase in occludin and ZO-1 in the Triton-soluble fraction (Fig. 3B). LY294002 reduced the XO+X-induced decrease in occludin and ZO-1 in the actin cytoskeleton and their increase in Triton-soluble fractions. LY294002 by itself produced no significant effect on cytoskeletal association or tyrosine phosphorylation of occludin and ZO-1 (data not shown).
Oxidative Stress Induces a Rapid Activation of PI 3-Kinase— The above studies indicate that PI 3-kinase activity is required for the oxidative stress-induced disruption of tight junction. Therefore, we evaluated the effect of XO+X on the localization of the regulatory subunit of PI 3-kinase and PI 3-kinase activity in the cytoskeletal fractions. XO+X induced a rapid increase in the amount of PI 3-kinase regulatory subunit associated with the actin cytoskeleton and the membrane cytoskeleton, whereas it was reduced in Triton-soluble fractions (Fig. 4, A and B). XO+X also increased PI 3-kinase activity in the actin cytoskeleton with the maximal increase achieved at 15 min (Fig. 4, C and D). Activity in Triton-soluble fraction was slightly reduced at later time points.
Oxidative Stress Increases Association of PI 3-Kinase with the Occludin—A previous study (
) raised the possibility of association of PI 3-kinase with the tight junction. To determine the interaction of PI 3-kinase with the tight junction complex in Caco-2 cell monolayers we evaluated co-immunoprecipitation of PI 3-kinase with occludin. A considerable amount of p85, the regulatory subunit of PI 3-kinase, was co-immunoprecipitated with occludin precipitated from the actin cytoskeleton (Fig. 5, A and B). The level of p85 co-immunoprecipitated with occludin was rapidly increased by XO+X treatment. Low levels of PI 3-kinase activity were detected in the immune complexes of occludin prepared from the resting epithelial actin cytoskeleton and Triton-soluble fraction (Fig. 5, C and D). Treatment with XO+X rapidly increased PI 3-kinase activity associated with the immune complexes of occludin prepared from both the actin cytoskeleton and the Triton-soluble fraction. Only trace amounts of PI 3-kinase activity were detected in the immune complexes of occludin prepared from the membrane cytoskeleton (data not shown). PI 3-kinase activity was dramatically low in immune complexes of occludin prepared from the cells treated with XO+X after pretreatment with LY294002 (Fig. 5, E and F).
To determine the translocation of PI 3-kinase to the vicinity of tight junction we analyzed the effect of oxidative stress on localization of p85 by immunofluorescence confocal microscopy. P85 was predominantly localized in the intracellular compartments at the apical part of the cell (Fig. 6). Treatment with XO+X induced a minor change in the localization of p85 at 5 min, but at 30 and 60 min after treatment p85 was predominantly localized at the intercellular junctions and co-localized with the occludin (Fig. 6).
Direct Binding of p85 with Occludin—To determine the direct interaction of C-terminal tail of occludin with p85 we generated recombinant GST-fused C-occludin (C-terminal 150 amino acids) and GST-p85. GST-p85 was cleaved with thrombin, and p85 was incubated with varying concentrations of GST-C-occludin. GST pull-down assay showed that recombinant p85 binds C-occludin in a concentration-related manner (Fig. 7). GST alone showed only a minimal binding to p85.
Oxidative Stress Increases Paracellular Permeability in MDCK Cell Monolayer by a PI 3-Kinase-dependent Mechanism—To demonstrate the PI 3-kinase-mediated regulation of tight junction in another epithelial model we evaluated the effect of LY294002 on oxidative stress-induced increase in permeability in MDCK cell monolayers. Incubation of MDCK cell monolayers with XO+X decreased TER (Fig. 8A) and increased inulin flux (Fig. 8B) in a time-dependent manner. This increase in paracellular permeability was associated with reorganization of occludin and ZO-1 from the intercellular junctions (Fig. 8C). Pretreatment of cell monolayers with LY294002 significantly reduced XO+X-induced decrease in TER, increase in inulin permeability and reorganization of occludin and ZO-1 (Fig. 8).
Oxidative stress disrupts tight junction- and adherens junction-based cell-cell adhesion in Caco-2 cell monolayer by a tyrosine kinase-dependent mechanism (
). The oxidative stress-induced disruption of the tight junction and the adherens junction is associated with the tyrosine phosphorylation of occludin, ZO-1, E-cadherin, and β-catenin. Recent studies (
) demonstrated that oxidative stress induces a rapid activation of c-Src, and that c-Src activity is required for the oxidative stress-induced disruption of the tight junction. The present study shows that oxidative stress increases the level of PI 3-kinase associated with the occludin and that PI 3-kinase activity mediates the oxidative stress-induced disruption of tight junctions in Caco-2 cell monolayers. This is the first evidence for the association of PI 3-kinase with a tight junction protein, and for the role of PI 3-kinase activity in the disruption of epithelial tight junctions. On the contrary a previous study (
) showed that PI 3-kinase activity is required for dexamethasone-induced increase in transepithelial resistance.
A significant reduction of oxidative stress-induced decrease in TER, increase in inulin permeability, and redistribution of occludin and ZO-1 by LY294002 indicate that PI 3-kinase activity plays an important role in the oxidative stress-induced disruption of tight junction in Caco-2 cell monolayer. Previous studies demonstrated that occludin, ZO-1, E-cadherin, and β-catenin undergo rapid phosphorylation on tyrosine residues during the oxidative stress-induced disruption of tight junctions (
) also demonstrated that occludin and ZO-1 bound to actin cytoskeleton correlates well with the barrier function of the epithelium. Disruption of tight junctions by oxidative stress reduces the amounts of occludin and ZO-1 bound to the actin cytoskeleton, which was prevented by genistein, a tyrosine kinase inhibitor. The present study shows that LY294002 also reduces the oxidative stress-induced dissociation of occludin and ZO-1 from the actin cytoskeleton.
Therefore, PI 3-kinase activity is required for the oxidative stress-induced tyrosine phosphorylation and dissociation from the actin cytoskeleton of tight junction proteins, and therefore for the disruption of tight junctions. However, the temporal relationship between oxidative stress effect on tyrosine phosphorylation, release from the actin cytoskeleton, and disruption of tight junction is not clear. Significant increase in tyrosine phosphorylation and release of occludin from the actin cytoskeleton occurred by 60 min, whereas less than 20% of TER was reduced. It is likely that mechanisms downstream to activation of PI 3-kinase and c-Src and other lateral mechanisms are involved in oxidative stress-induced disruption of tight junction.
The fast migrating occludin bands that appeared in oxidative stress-treated cells (Fig. 3) are likely a result of proteolytic degradation of occludin. In our previous study (
) we showed that oxidative stress induces proteolytic degradation of occludin, and the low molecular weight degradation products are predominantly present in MCS and TS fractions. The metallo-proteinase inhibitor, 1,10-phenanthroline, significantly reduced the oxidative stress-induced degradation of occludin. However, this inhibitor did not prevent oxidative stress-induced disruption of tight junction, and therefore occludin degradation was considered uninvolved in oxidative stress-induced disruption of tight junction and increase in permeability. Additionally, genistein prevented oxidative stress-induced increase in permeability without an effect on occludin degradation. Similarly, the present study shows that LY294002 reduces oxidative stress-induced tyrosine phosphorylation of occludin (Fig. 3A) with no effect on proteolytic degradation of occludin (Fig. 3B).
The requirement of PI 3-kinase activity in the regulation of tight junction, and the previous in vitro study that raised the possibility of interaction between the regulatory subunit of PI 3-kinase and the C-terminal sequence of occludin (
) suggest that PI 3-kinase may be associated with the tight junction complex in Caco-2 cells, and that oxidative stress may alter this interaction. The present study shows that the regulatory subunit of PI 3-kinase and the PI 3-kinase activity are associated with the immune complexes of occludin prepared from the actin cytoskeleton of the resting epithelium. Oxidative stress rapidly increases the level of regulatory subunit of PI 3-kinase and the PI 3-kinase activity in the immune complexes of occludin. These results demonstrate that PI 3-kinase does interact with tight junction complex, and it can be increased by physiologic or pathophysiologic conditions such as oxidative stress. This observation was further confirmed by confocal immunofluorescence localization of p85 at the junctions. Oxidative stress increased the localization of p85 at the intercellular junctions, with a concomitant decrease in p85 stain at the intracellular compartments. Double staining for p85 and occludin indicates that p85 is co-localized with occludin. At 5 min of XO+X treatment the PI 3-kinase activity associated with occludin was found to be higher than control in both ACS and TS fraction, suggesting an activation of PI 3-kinase at the early time period. Therefore, it is possible that the initial effect of oxidative stress is activation of PI 3-kinase bound to occludin, which was immediately followed by a translocation of PI 3-kinase into occludin.
The binding of PI 3-kinase to occludin does not necessarily mean that it is associated with the tight junction. However, our previous study (
) showed that oxidative stress-induced increase in paracellular permeability and its reduction by genistein correlated well with the changes in the levels of actin-bound occludin. Changes in occludin present in MCS fraction or TS fraction did not correlate with the permeability changes, which indicated that actin-bound occludin is the pool of occludin that is most relevant t the tight junctions. Therefore, it is likely that PI 3-kinase activity present in occludin immunoprecipitates prepared from the ACS fraction is associated with the tight junction.
The rapid increase in the activity of PI 3-kinase in the tight junctions strongly indicates that this pool of PI 3-kinase plays a crucial role in the oxidative stress-induced disruption of tight junctions. However, it is not clear if the co-immunoprecipitation of p85 with occludin is a direct interaction. As immune complexes of occludin prepared under native conditions are expected to co-precipitate many of the tight junction proteins, it is possible that PI 3-kinase interacts indirectly with the occludin through other proteins. Therefore, we investigated the direct interaction of occludin with p85 by GST pull-down assays using GST-C-occludin and thrombin-cleaved GST-p85. Concentration-related binding of recombinant p85 with recombinant GST-C-occludin demonstrates that p85 directly interacts with the C-terminal tail of occludin. These studies suggest a possibility of direct interaction of p85 with occludin within the cell, and that this interaction is enhanced by oxidative stress. Such an enhancement of interaction between p85 and occludin may involve post-translational modifications, such as phosphorylation.
In summary, this study shows that oxidative stress increases the association of PI 3-kinase regulatory subunit and PI 3-kinase activity with the tight junction protein complex, and that PI 3-kinase activity is required for the oxidative stress-induced disruption of tight junctions.
Sleisenger M. Fordstran J. Gastrointestinal Disease. WB Saunders,