Advertisement

Prostaglandin E2 Produced by Entamoeba histolytica Signals via EP4 Receptor and Alters Claudin-4 to Increase Ion Permeability of Tight Junctions

Open ArchivePublished:May 30, 2011DOI:https://doi.org/10.1016/j.ajpath.2011.05.001
      Entamoeba histolytica is a protozoan parasite that causes amebic dysentery characterized by severe watery diarrhea. Unfortunately, the parasitic factors involved in the pathogenesis of diarrhea are poorly defined. Prostaglandin E2 (PGE2) is a host lipid mediator associated with diarrheal diseases. Intriguingly, E. histolytica produces and secretes this inflammatory molecule. We investigated the mechanism whereby ameba-derived PGE2 induces the onset of diarrhea by altering ion permeability of paracellular tight junctions (TJs) in colonic epithelia. PGE2 decreased barrier integrity of TJs in a dose- and time-dependent manner, as measured by transepithelial resistance. PGE2 signals were selectively transduced via the EP4 receptor. Furthermore, PGE2 signaling decreased TJ integrity, as revealed by EP receptor-specific agonist and antagonist studies. Loss of mucosal barrier integrity corresponded with increased ion permeability across TJs. Subcellular fractionation and confocal microscopy studies highlighted a significant spatial alteration of an important TJ protein, claudin-4, that corresponded with increased sodium ion permeability through TJs toward the lumen. Moreover, PGE2-induced luminal chloride secretion was a prerequisite for alterations at TJs. Thus, the gradient of NaCl created across epithelia could serve as a trigger for osmotic water flow that leads to diarrhea. Our results highlight a pathological role for E. histolytica-derived PGE2 in the onset of diarrhea.
      Entamoeba histolytica is an enteric-dwelling protozoan parasite that infects an estimated 500 million people worldwide.
      • Schuster F.L.
      • Visvesvara G.S.
      Amebae and ciliated protozoa as causal agents of waterborne zoonotic disease.
      Alarmingly, E. histolytica accounts for more than 100,000 deaths annually.
      • Anane S.
      • Khaled S.
      [Entamoeba histolytica and Entamoeba dispar: differentiation methods and implications].
      The disease amebiasis, which manifests as amebic colitis in 10% to 20% of infected subjects, is characterized by acute watery and/or bloody diarrhea.
      • Schuster F.L.
      • Visvesvara G.S.
      Amebae and ciliated protozoa as causal agents of waterborne zoonotic disease.
      • Haque R.
      • Mondal D.
      • Duggal P.
      • Kabir M.
      • Roy S.
      • Farr B.M.
      • Sack R.B.
      • Petri Jr, W.A.
      Entamoeba histolytica infection in children and protection from subsequent amebiasis.
      • Haque R.
      • Mondal D.
      • Kirkpatrick B.D.
      • Akther S.
      • Farr B.M.
      • Sack R.B.
      • Petri Jr, W.A.
      Epidemiologic and clinical characteristics of acute diarrhea with emphasis on Entamoeba histolytica infections in preschool children in an urban slum of Dhaka, Bangladesh.
      • Niyogi S.K.
      • Saha M.R.
      • De S.P.
      Enteropathogens associated with acute diarrhoeal diseases.
      The mechanism by which E. histolytica causes acute diarrhea is not clearly understood. E. histolytica in contact with colonic epithelial cells induces a robust inflammatory response involving proinflammatory cytokines and other inflammatory mediators that causes what is known as invasive or inflammatory diarrhea.
      • Navaneethan U.
      • Giannella R.A.
      Mechanisms of infectious diarrhea.
      However, the acute nature of the disease indicates the possibility of pathogenesis even in the absence of parasite-epithelial contact. We speculate that E. histolytica can cause diarrhea in an epithelial contact-independent manner; however, the parasite factor or factors responsible for this effect is not clearly understood.
      Prostaglandin E2 (PGE2) is an important host inflammatory mediator that is associated with various diarrheal pathologies, including cholera, entero-invasive bacterial diseases, and inflammatory bowel diseases.
      • Beubler E.
      • Schuligoi R.
      Mechanisms of cholera toxin-induced diarrhea.
      • Resta-Lenert S.
      • Barrett K.E.
      Enteroinvasive bacteria alter barrier and transport properties of human intestinal epithelium: role of iNOS and COX-2.
      • Alcantara C.
      • Stenson W.F.
      • Steiner T.S.
      • Guerrant R.L.
      Role of inducible cyclooxygenase and prostaglandins in Clostridium difficile toxin A-induced secretion and inflammation in an animal model.
      • Ahrenstedt O.
      • Hällgren R.
      • Knutson L.
      Jejunal release of prostaglandin E2 in Crohn's disease: relation to disease activity and first-degree relatives.
      PGE2 levels are increased by 10-fold in the colonic mucosa during E. histolytica infection.
      • Stenson W.F.
      • Zhang Z.
      • Riehl T.
      • Stanley Jr, S.L.
      Amebic infection in the human colon induces cyclooxygenase-2.
      Amebic trophozoites induce host cells such as macrophages, polymorphonuclear cells, and colonic epithelial cells to synthesize PGE2.
      • Stenson W.F.
      • Zhang Z.
      • Riehl T.
      • Stanley Jr, S.L.
      Amebic infection in the human colon induces cyclooxygenase-2.
      • Sánchez-Ramírez B.
      • Ramírez-Gil M.
      • Vázquez-Moctezuma I.
      • Ramos-Martínez E.
      • Talamás-Rohana P.
      Entamoeba histolytica: induction of cyclooxygenase-2 expression during amoebic liver abscess formation in hamsters (Mesocricetus auratus).
      Remarkably, E. histolytica trophozoites also constitutively produce and secrete PGE2.
      • Belley A.
      • Chadee K.
      Eicosanoid production by parasites: from pathogenesis to immunomodulation?.
      The identity/similarity of the parasite PGE2 with that of the host has been confirmed by gas chromatography/mass spectrometry analysis.
      • Belley A.
      • Chadee K.
      Production of prostaglandin E(2) by Entamoeba histolytica via a novel cyclooxygenase.
      Nevertheless, PGE2 is the only prostanoid present in the parasite SPs. Parasitic production of PGE2 was significantly increased in the presence of arachidonic acid (AA),
      • Belley A.
      • Chadee K.
      Eicosanoid production by parasites: from pathogenesis to immunomodulation?.
      • Dey I.
      • Keller K.
      • Belley A.
      • Chadee K.
      Identification and characterization of a cyclooxygenase-like enzyme from Entamoeba histolytica.
      the precursor for PGE2, which is usually present in high concentrations in the gut.
      • Meyer B.J.
      • Mann N.J.
      • Lewis J.L.
      • Milligan G.C.
      • Sinclair A.J.
      • Howe P.R.
      Dietary intakes and food sources of omega-6 and omega-3 polyunsaturated fatty acids.
      • Sioen I.A.
      • Pynaert I.
      • Matthys C.
      • De Backer G.
      • Van Camp J.
      • De Henauw S.
      Dietary intakes and food sources of fatty acids for Belgian women, focused on n-6 and n-3 polyunsaturated fatty acids.
      More importantly, E. histolytica has a COX-like gene that encodes a functional cyclooxygenase enzyme for the biosynthesis of PGE2; however, E. histolytica COX is primitive, and is unique in its pharmacological sensitivity to nonsteroidal anti-inflammatory drugs in being sensitive only to aspirin (ASA).
      • Dey I.
      • Keller K.
      • Belley A.
      • Chadee K.
      Identification and characterization of a cyclooxygenase-like enzyme from Entamoeba histolytica.
      Ameba COX has little homology to COX-1/2 enzymes from different species, in that the AA-binding domain and heme-coordinating and catalytic sites are absent.
      • Dey I.
      • Keller K.
      • Belley A.
      • Chadee K.
      Identification and characterization of a cyclooxygenase-like enzyme from Entamoeba histolytica.
      At present, we do not know the exact functional/biological significance of ameba-derived PGE2 in the pathogenesis of amebiasis. We have recently shown that E. histolytica PGE2 signals via the host epithelial EP4 receptors to induce robust IL-8 secretion that could augment inflammation.
      • Dey I.
      • Chadee K.
      Prostaglandin E2 produced by Entamoeba histolytica binds to EP4 receptors and stimulates interleukin-8 production in human colonic cells.
      Because PGE2 is strongly associated with various diarrheal diseases, however, we reasoned that ameba-derived PGE2 might play a major role in the onset of diarrhea during acute amebic colitis.
      In general, secretory diarrhea is characterized by an alteration of the ion barrier of intestinal epithelium
      • Berkes J.
      • Viswanathan V.K.
      • Savkovic S.D.
      • Hecht G.
      Intestinal epithelial responses to enteric pathogens: effects on the tight junction barrier, ion transport, and inflammation.
      through increased luminal secretion of chloride ions (anions), a phenomenon that is observed with cholera toxin in the ileal mucosa.
      • Field M.
      • Fromm D.
      • al-Awqati Q.
      • Greenough 3rd, W.B.
      Effect of cholera enterotoxin on ion transport across isolated ileal mucosa.
      To maintain charge balance in the lumen, sodium ions (cations) are pulled across the epithelium, also drawing water molecules and thus causing diarrhea.
      • Berkes J.
      • Viswanathan V.K.
      • Savkovic S.D.
      • Hecht G.
      Intestinal epithelial responses to enteric pathogens: effects on the tight junction barrier, ion transport, and inflammation.
      PGE2 is a potent colonic epithelial chloride ion secretagogue
      • Keeler R.
      • Wong N.L.
      Evidence that prostaglandin E2 stimulates chloride secretion in cultured A6 renal epithelial cells.
      • Bunce K.T.
      • Spraggs C.F.
      Stimulation of electrogenic chloride secretion by prostaglandin E2 in guinea-pig isolated gastric mucosa.
      • Deachapunya C.
      • O'Grady S.M.
      Regulation of chloride secretion across porcine endometrial epithelial cells by prostaglandin E2.
      ; however, it is not known whether it induces sodium ion flux across the epithelium. Colonic epithelial tight junctions (TJs) act as an important barrier for paracellular movement of macromolecules and ions across the epithelium.
      • Schneeberger E.E.
      • Lynch R.D.
      The tight junction: a multifunctional complex.
      • Harhaj N.S.
      • Antonetti D.A.
      Regulation of tight junctions and loss of barrier function in pathophysiology.
      • González-Mariscal L.
      • Betanzos A.
      • Nava P.
      • Jaramillo B.E.
      Tight junction proteins.
      Among the 40 different proteins that constitute the TJ complex, the claudin family is responsible for ion barrier maintenance of the paracellular space.
      • Turksen K.
      • Troy T.C.
      Barriers built on claudins [Erratum appeared in J Cell Sci 2004, 117:4341].
      • Balkovetz D.F.
      Claudins at the gate: determinants of renal epithelial tight junction paracellular permeability.
      We speculate that PGE2 secreted by E. histolytica breaks the TJ ion barrier by selectively altering claudin family members, allowing free paracellular permeability to sodium ions. Any alteration in the TJ ion barrier will markedly affect epithelial transport mechanisms associated with electrolyte and water imbalance. Here, we present a novel mechanism by which E. histolytica disrupts colonic TJ barrier function and offer important insight into the cellular and molecular basis of epithelial barrier/ion transport function in intestinal amebiasis. To date, there has been no report suggesting a contact-independent disruption of TJs by E. histolytica. The focus of our study was therefore to identify a role for E. histolytica-derived PGE2 in altering the colonic epithelial TJ ion barrier and to define the mechanism by which PGE2 causes diarrhea.

      Materials and Methods

      SPs of E. histolytica

      Secreted products (SP) of E. histolytica were prepared and collected as described previously.
      • Yu Y.
      • Chadee K.
      Entamoeba histolytica stimulates interleukin 8 from human colonic epithelial cells without parasite-enterocyte contact.
      Briefly, trophozoites of HM1-IMSS, a highly virulent strain of E. histolytica, were cultured until log phase in TYI-S-33 medium containing 20% serum. Trophozoites were washed three times in ice-cold Hanks' balanced salt solution (HBSS), adjusted to a final concentration of 1 × 107/mL HBSS, and incubated for 2 hours at 37°C under periodic swirling. The resulting SP were collected by centrifugation (1000 × g for 5 minutes) and assayed for protein amount using Bradford's reagent, normalized (0.1 μg/μL of SP), aliquoted, and stored at −80°C until further use. Trophozoite viability after incubation was >95% as determined by trypan blue exclusion assay. In some experiments, E. histolytica trophozoites were incubated with a nonspecific COX-1/2 inhibitor (1 mmol/L ASA) or with a specific COX-1/2 inhibitor (50 μmol/L indomethacin) for 16 hours. COX inhibitors were removed by centrifugation, followed by 2 hours of incubation in fresh HBSS for collecting SP as described previously.
      • Dey I.
      • Chadee K.
      Prostaglandin E2 produced by Entamoeba histolytica binds to EP4 receptors and stimulates interleukin-8 production in human colonic cells.
      In another experiment, 100 μmol/L AA was added to trophozoites in HBSS at the start of the 2-hour incubation and SP was collected.

      Quantification of PGE2 in SPs

      PGE2 in SP was assayed using an Assay Designs Correlate-CLIA kit (no. 910–001; Enzo Life Sciences, Plymouth Meeting, PA). This highly sensitive PGE2 chemiluminescence enzyme immunoassay is designed for quantitative determination of human PGE2 in biological samples with <0.1% cross-reactivity with AA. The assay was performed according to the manufacturer's protocol.

      Colonic Cells and Reagents

      For all experiments, T84 human colonic cells (ATCC, Manassas, VA) were used, maintained in Dulbecco's modified Eagle's medium with Ham's F-12 supplemented with 10% fetal bovine serum, 100 units/mL penicillin, 100 μg/mL streptomycin sulfate and 20 mmol/L HEPES (Sigma-Aldrich, St. Louis, MO). In one experiment, Caco-2 cell line cells were used, maintained in modified Eagle's medium with 10% fetal bovine serum, 100 units/mL penicillin, 100 μg/mL streptomycin sulfate, and 20 mmol/L HEPES (Sigma-Aldrich). PGE2, EP receptor-specific agonists, and antagonists were obtained from Cayman Chemical (Ann Arbor, MI), unless otherwise indicated. Antibodies for TJ proteins, F-actin, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were purchased from Invitrogen (Carlsbad, CA), Abcam (Cambridge, MA), and Calbiochem (San Diego, CA), respectively; all other chemicals were procured from Sigma-Aldrich.

      Transwell Preparation

      T84 cells were incubated at 37°C in 5% CO2 and passaged when the monolayer reached 90% confluency. Cells between passage numbers 64 and 68 were used for plating on culture plates for Transwell studies. To passage cells, the monolayer in a T75 flask was rinsed with 5 mL of sterile Dulbecco's PBS and incubated with 1 mL of trypsin-EDTA for 8 to 10 minutes. Detached cells were resuspended in medium to a final concentration of 1 × 106 cells/mL and the medium was mixed vigorously to avoid clumps. To prepare a 12-well Transwell plate for electrophysiological studies, 5 × 104 cells were seeded on the apical side of the membrane containing 500 μL medium; the basolateral side was bathed with 1.5 mL of medium alone. Transwell plates were fed the day after seeding and every alternate day thereafter and were used for experiments in approximately 8 to 10 days. For TJ protein analysis and immunofluorescence studies, T84 cells plated on 6-well plates/coverslips were used.

      Measurement of TER

      T84 cells grown to 100% confluence on polyethylene membrane inserts (12-mm diameter, 0.4-μm pore size) (catalog no. 3460, Corning Costar Transwell; Corning Life Sciences, Corning, NY) were used. Transepithelial resistance (TER) was measured using a Millicell-ERS apparatus (Millipore, Bedford, MA). When a stable resistance of >2000 Ω·cm2 was reached, the medium from either the apical or basal compartment was replaced with medium containing the desired stimulant and TER was measured. Changes in TER in response to stimuli were normalized to that of the baseline resistance (resistance at the starting time point) and represented as percentage of the control over a desired time course. For experiments involving EP receptor agonists/antagonists, the concentration and the conditions for treatment/pretreatment were determined based on previous reports.
      • Tang C.H.
      • Yang R.S.
      • Fu W.M.
      Prostaglandin E2 stimulates fibronectin expression through EP1 receptor, phospholipase C, protein kinase Calpha, and c-Src pathway in primary cultured rat osteoblasts.
      • Chen B.C.
      • Liao C.C.
      • Hsu M.J.
      • Liao Y.T.
      • Lin C.C.
      • Sheu J.R.
      • Lin C.H.
      Peptidoglycan-induced IL-6 production in RAW 264.7 macrophages is mediated by cyclooxygenase-2, PGE2/PGE4 receptors, protein kinase A, I kappa B kinase, and NF-kappa B.
      • Shao J.
      • Sheng H.
      Prostaglandin E2 induces the expression of IL-1alpha in colon cancer cells.
      • Dey I.
      • Giembycz M.A.
      • Chadee K.
      Prostaglandin E(2) couples through EP(4) prostanoid receptors to induce IL-8 production in human colonic epithelial cell lines.
      • Cherukuri D.P.
      • Chen X.B.
      • Goulet A.C.
      • Young R.N.
      • Han Y.
      • Heimark R.L.
      • Regan J.W.
      • Meuillet E.
      • Nelson M.A.
      The EP4 receptor antagonist, L-161,982, blocks prostaglandin E2-induced signal transduction and cell proliferation in HCA-7 colon cancer cells.

      Paracellular Permeability Study

      The permeability of TJs to macromolecules/solute in response to PGE2 was measured by studying the apical to basal translocation of 14C-labeled ethanolamine (a small paracellular tracer, ∼4.9 Å in diameter), as described previously.
      • Tang V.W.
      • Goodenough D.A.
      Paracellular ion channel at the tight junction.
      The tracer was used at a concentration of 0.33 μCi/well. After incubation with tracer for the desired time intervals, 40-μL samples from both chambers were collected in separate scintillation vials containing 10 mL of scintillation fluid. Radioactivity was measured using a Beckman liquid scintillation counter (Beckman Coulter, Brea, CA). Permeability to [14C]ethanolamine was measured as the apparent permeability coefficient Papp and was expressed as the transport enhancement ratio R, as described previously.
      • Grobovac V.
      • Bernkop-Schnürch A.
      Improvement of the intestinal membrane permeability of low molecular weight heparin by complexation with stem bromelain.

      NaCl Dilution Potential Assay and Measurement of Sodium Potential

      A 20% NaCl dilution potential assay was performed as described previously.
      • Rao R.K.
      • Li L.
      • Baker R.D.
      • Baker S.S.
      • Gupta A.
      Glutathione oxidation and PTPase inhibition by hydrogen peroxide in Caco-2 cell monolayer.
      Briefly, medium was discarded from a Transwell plate with a confluent T84 monolayer and cells were bathed in PBS + bovine serum albumin (BSA) (140 mmol/L NaCl, 2.7 mmol/L KCl, 10 mmol/L Na2HPO4, 2 mmol/L KH2PO4, and 0.6% BSA), with 0.5 and 1.5 mL in the apical and basal chambers, respectively. After apical stimulation with 1 μmol/L PGE2 in PBS + BSA, 20% dilution in either the apical or the basal chamber was developed by replacing 100 or 300 μL of apical or basal solution with a solution of equally osmolar mannitol + BSA (140 mmol/L mannitol, 2.7 mmol/L KCl, 10 mmol/L Na2HPO4, 2 mmol/L KH2PO4 and 0.6% BSA), respectively. The electrical potential was recorded before and after 20% dilution and at 5-minute intervals from the start until 30 minutes. The 20% dilution potential difference was calculated as the difference between the initial potential and the potential after dilution. For measuring sodium potential, T84 monolayer on a Transwell plate was washed with PBS + BSA and stimulated apically with 1 μmol/L PGE2 in mannitol + BSA. At various time points, 25-μL samples were collected from the apical chamber and sodium potential was measured using a highly sensitive sodium electrode (Thermo Orion, Beverly, MA), as described previously.
      • Tang V.W.
      • Goodenough D.A.
      Paracellular ion channel at the tight junction.
      For experiments involving pharmacological inhibition of membrane sodium channel/transporters or cystic fibrosis transmembrane regulator (CFTR), inhibitor concentration and conditions for pretreatment were determined based on previous reports.
      • de Jong J.C.
      • Willems P.H.
      • Mooren F.J.
      • van den Heuvel L.P.
      • Knoers N.V.
      • Bindels R.J.
      The structural unit of the thiazide-sensitive NaCl cotransporter is a homodimer.
      • D'Andrea L.
      • Lytle C.
      • Matthews J.B.
      • Hofman P.
      • Forbush 3rd, B.
      • Madara J.L.
      Na: K:2Cl cotransporter (NKCC) of intestinal epithelial cells Surface expression in response to cAMP.
      • Kim J.A.
      • Kang Y.S.
      • Lee S.H.
      • Lee E.H.
      • Yoo B.H.
      • Lee Y.S.
      Glibenclamide induces apoptosis through inhibition of cystic fibrosis transmembrane conductance regulator (CFTR) Cl(−) channels and intracellular Ca(2+) release in HepG2 human hepatoblastoma cells.

      Cell Fractionation and Western Blotting for TJ Proteins

      T84 cells grown to confluence on regular 6-well plates were fractionated after appropriate treatment, based on a previously described protocol
      • Sheth P.
      • Basuroy S.
      • Li C.
      • Naren A.P.
      • Rao R.K.
      Role of phosphatidylinositol 3-kinase in oxidative stress-induced disruption of tight junctions.
      with slight modifications. The cell monolayer was washed twice with ice-cold PBS, scraped using a rubber policeman, and collected in an Eppendorf tube. Cells were centrifuged at 1250 × g for 5 minutes at 4°C. The pellet was resuspended in 400 μL of ice cold cell lysis buffer (50 mmol/L Tris-HCl, 140 mmol/L EDTA, 30 mmol/L sodium pyrophosphate, 50 mmol/L sodium fluoride and protease inhibitor cocktail tablet) containing 1% Triton X-100, incubated for 30 minutes on ice and centrifuged at 20,000 × g for 30 minutes at 4°C. The supernatant was labeled as Triton X-100 soluble fraction (SF) or noncytoskeletal fraction. The pellet was again resuspended in 200 μL of cell lysis buffer containing 1% SDS, and then was sonicated and centrifuged at 20,000 × g for 30 minutes at 4°C. The resulting supernatant was labeled as Triton X-100 insoluble fraction (IF) or cytoskeletal fraction. Protein content of subcellular fractions was estimated using the Bradford method and was adjusted for a final concentration of 0.5 μg/μL. Equal volume of 1× sample buffer was added to the samples, boiled at 100°C for 10 minutes, and used for Western blotting. Approximately 2.5 to 5.0 μg of total protein was loaded per well of 7.5% to 15% SDS-PAGE (depending on the molecular weight of the TJ protein being probed) and electrophoresis was performed. Proteins were transferred onto nitrocellulose membranes (Bio-Rad, Hercules, CA) followed by blocking in 5% skim milk powder in TBS-T (20 mmol/L Tris-HCl, pH 7.5, 500 mmol/L NaCl, 0.1% Tween 20) for 1 hour at room temperature. Membranes were incubated with appropriate primary antibodies in 1% skim milk-TBS-T at 4°C, overnight. Blots were washed three times with TBS-T and then incubated in horseradish peroxidase-conjugated secondary antibodies in 1% skim milk-TBS-T for 2 hours at room temperature. The blots were washed with TBS-T and were developed using Immobilon Western chemiluminescent HRP substrate (Millipore, Billerica, MA) according to the manufacturer's instructions.

      Confocal Immunofluorescence Microscopy

      T84 monolayers grown on glass coverslips in regular six-well culture plates were used. After appropriate treatment, monolayer was washed with ice cold PBS and fixed with 3.75% paraformaldehyde at 4°C for 30 minutes. Fixed cells were permeabilized using 0.5% Triton X-100 in PBS for 15 minutes and blocked with 5% BSA in PBS-0.5% Tween 20 (PBS-T) for 1 hour at room temperature. Monolayers were incubated overnight in 1:100 dilution of appropriate primary antibodies at 4°C. After two washes in PBS-T, monolayers were coincubated with fluorescein isothiocyanate or phycoerythrin-conjugated secondary antibodies (1:200 dilution) for 1 hour at room temperature. Coverslip was mounted on clean glass slide using Vectashield (Vector Laboratories, Burlingame, CA). Slides were examined using a FluoView FV1000 confocal immunofluorescence microscope (Olympus). Immunofluorescence signals from the tags for respective proteins were determined in individual en face (xy axes) planes throughout the cellular z axis at 0.35-μm intervals. Z-planes obtained were stacked and analyzed using Volocity software version 5.4 (PerkinElmer, Waltham, MA) to obtain PXZ and three-dimensional (3D) reconstructions of the Z-stacks.

      Statistical Analysis

      Data were analyzed by two-way analysis of variance followed by a Bonferroni post hoc test for comparison between groups using GraphPad Prism version 4.0 (GraphPad Software, San Diego, CA). In the figures, data are reported as means ± SD of three independent experiments.

      Results

      Secreted Components of E. histolytica Alter TER of Colonic Monolayer

      To test the hypothesis that secreted components derived from highly virulent wild-type (Wt) E. histolytica alter the integrity of the colonic epithelial monolayer during pathogenesis of intestinal amebiasis, the effect of SP on TER of T84 colonic epithelial monolayer was analyzed. When the monolayer was exposed on the apical/luminal side to SP, there was a decrease in the integrity of TER that occurred in a dose- and time-dependent manner, compared with the nonstimulated controls (Figure 1). With high doses of SP (48 μg), there was a 55% decrease in TER as early as 5 minutes, followed by a gradual recovery but with a 28% decrease (P < 0.001) remaining even at 60 minutes. This is in contrast to the response observed with low doses of SP (8 μg), which caused a modest 19% decrease in TER at 5 minutes (P < 0.05), followed by recovery to control values as early as 30 minutes. An intermediate dose of SP (24 μg) caused 34% (P < 0.001) and 19% (P < 0.05) decreases at 5 and 30 minutes, respectively, with recovery to control values at 60 minutes. Thus, the effects of E. histolytica SP on TER follow a dose-response relationship. More importantly, these results show that alteration in colonic epithelial barrier function occurred in the absence of parasite-epithelial cell contact.
      Figure thumbnail gr1
      Figure 1Effect of secreted components of E. histolytica on TER of T84 monolayers. T84 monolayers on Transwell plates that reached a stable TER were exposed on the apical side to various amounts (8, 24, 48 μg) of Wt ameba SP (Wt-SP). SP was collected from 1.5 × 107 trophozoites in 1.5 mL of HBSS and protein concentration normalized to 0.1 μg/μL. TER was measured over 5, 30, and 60 minutes and the mean value was expressed as percentage of the control and compared with that of the control. The mean baseline TER of untreated monolayers at the start of experiments was 2744 ± 516 Ω·cm2. *P < 0.05; ***P < 0.001.

      PGE2 Secreted by E. histolytica Decreases Colonic TER

      We have previously shown that trophozoites of E. histolytica constitutively produce and secrete PGE2 and have confirmed its identity with mammalian PGE2 by gas chromatography/mass spectrometry analysis.
      • Belley A.
      • Chadee K.
      Eicosanoid production by parasites: from pathogenesis to immunomodulation?.
      • Belley A.
      • Chadee K.
      Production of prostaglandin E(2) by Entamoeba histolytica via a novel cyclooxygenase.
      • Dey I.
      • Keller K.
      • Belley A.
      • Chadee K.
      Identification and characterization of a cyclooxygenase-like enzyme from Entamoeba histolytica.
      For the present study, we quantified PGE2 release by live trophozoites under the same conditions as used to isolate SP in the previous experiments. PGE2 was constitutively released and present in SP from Wt amebas (635.1 ± 15.67 pg/1.5 × 107 trophozoites) (Figure 2). More importantly, when live trophozoites were incubated in the presence of AA substrate (AA-SP), there was a 754% increase in PGE2 production. In previous studies,
      • Dey I.
      • Keller K.
      • Belley A.
      • Chadee K.
      Identification and characterization of a cyclooxygenase-like enzyme from Entamoeba histolytica.
      • Dey I.
      • Chadee K.
      Prostaglandin E2 produced by Entamoeba histolytica binds to EP4 receptors and stimulates interleukin-8 production in human colonic cells.
      we have shown that among the known COX inhibitors studied only ASA (1 mmol/L) significantly inhibited the cell-free as well as the live E. histolytica COX-like enzyme. Accordingly, in the present study, 16-hour pretreatment of trophozoites with 1 mmol/L ASA inhibited PGE2 production by 32% (P < 0.001), whereas indomethacin had no inhibitory effect. These data clearly show that live E. histolytica not only constitutively synthesize PGE2 but that this synthesis is significantly inhibited by ASA and is markedly increased in the presence of AA substrate.
      Figure thumbnail gr2
      Figure 2Quantification of PGE2 produced by E. histolytica. SP collected from 1.5 × 107 trophozoites in 1.5 mL of HBSS were quantified for PGE2 using a Correlate CLIA kit. AA-SP, SP collected from amebas incubated with 100 μmol/L AA for 2 hours; ASA 16 hours-SP and INDO 16 hours-SP, SPs collected from amebas pretreated for 16 hours with 1 mmol/L ASA or 50 μmol/L indomethacin, respectively; Wt-SP, SP from Wt amebas. ***P < 0.001 versus Wt-SP.
      To determine whether PGE2 produced by E. histolytica was the putative agonist present in SP that altered TER, the effect of SP (48 μg) derived from E. histolytica pretreated with ASA for 16 hours (ASA 16 hours-SP) was compared with that of untreated Wt control (Wt-SP). ASA 16 hours-SP significantly prevented the decrease in TER at all time points measured (Figure 3). These data confirm a critical role for PGE2 produced by E. histolytica in altering TER. Because inhibition of COX by ASA was not 100%, we determined the contribution of other major putative virulence components secreted by amebas in altering TER. Experiments were done with a commercially available analog of serotonin (5-hydroxytryptophan), which is a known SP component, and with SP collected from amebas lacking both amebapore and cysteine proteinase 5 (CP5). Wt-SP treated with the CP5 inhibitor E-64 was also used. None of the virulence factors that can be found in SP significantly modulated TER (see Supplemental Figure S1 at http://ajp.amjpathol.org). Moreover, the inactive PGE2 analog 15-keto PGE2 did not decrease TER, confirming specificity for PGE2.
      Figure thumbnail gr3
      Figure 3Effect of PGE2 secreted by E. histolytica on TER of T84 monolayer. T84 monolayers that reached a stable TER of >2000 Ω·cm2 were exposed on the apical side to 48 μg SP collected from either E. histolytica pretreated with ASA for 16 hours (ASA 16 hours-SP) or from nontreated Wt controls (Wt-SP). TER was measured over 5, 30, and 60 minutes and the mean values were compared with Wt-SP. *P < 0.05; **P < 0.01.

      Dose-Response Study Highlights a Differential Role for PGE2 in Altering TER

      Given that bioactive E. histolytica-derived PGE2 is difficult to purify, and given that it is identical to mammalian PGE2, all subsequent experiments were done with commercially available purified PGE2. This strategy also eliminated any confounding effects of other ameba molecules that could affect TER function. Having identified a dose-response relationship for SP in altering TER (Figure 1), we determined whether purified PGE2 could also mimic the same effect. More importantly, we analyzed a dose-response of PGE2, considering the fact that PGE2 can exert diverse biological functions based on the amount released around target cells.
      • Dey I.
      • Lejeune M.
      • Chadee K.
      Prostaglandin E2 receptor distribution and function in the gastrointestinal tract.
      Various doses of purified PGE2 had a differential effect on TER (Figure 4). A low dose of 8 nmol/L PGE2 did not significantly alter TER. At higher concentrations, there was a precipitous decrease in TER as early as 5 minutes. At the highest concentration tested (1 μmol/L), there was a 70% decrease in TER as early as 5 minutes (P < 0.001). At an intermediate dose of 100 nmol/L there was a temporal fluctuation, but TER remained well below that of control. For all doses tested, TER tended to be restored to normal levels by 12 hours (720 minutes). These results highlight the importance of a concentration-dependent differential effect of PGE2 in altering TER of the colonic epithelium. Moreover, a PGE2-induced decrease in TER was noted in a different colonic epithelial cell line, Caco-2, demonstrating that the effect is not cell line-specific (see Supplemental Figure S2 at http://ajp.amjpathol.org). Nonetheless, the magnitude of PGE2-induced alterations in TER varied between T84 and Caco-2 cells. This discrepancy could be due to a difference in the baseline TER between these cell lines (∼2700 Ω·cm2 for T84, compared with ∼1100 Ω·cm2 for Caco-2).
      Figure thumbnail gr4
      Figure 4Dose-response of purified PGE2 on TER. T84 monolayers that reached a stable TER of >2000 Ω·cm2 were exposed on the apical side to various concentrations (8 nmol/L to 1 μmol/L) of commercially available purified PGE2. The mean values of each group for 5, 30, 60, and 720 minutes were expressed as percentage of the control and were statistically compared with controls. **P < 0.01; ***P < 0.001.

      High-Dose PGE2 Signals via EP4 Receptors to Decrease TER

      PGE2 binds and signals through four different subtypes of EP receptors: EP1, EP2, EP3, and EP4.
      • Dey I.
      • Lejeune M.
      • Chadee K.
      Prostaglandin E2 receptor distribution and function in the gastrointestinal tract.
      To elucidate the EP receptor subtype specificity of PGE2 for decreasing TER, we used a panel of EP receptor subtype-specific agonists: 17 phenyl trinor PGE2 for EP1/3, butaprost for EP2, sulprostone for EP3, and PGE1OH for EP4/2. Only the EP4 receptor agonist PGE1OH caused a rapid decrease in TER (71% at 5 minutes, P < 0.001) (Figure 5A). The EP2 receptor agonist butaprost exerted a modest decrease in TER (21%) at 5 minutes, but recovered shortly thereafter and was then identical to the control. Notably, the effect of the EP4 receptor agonist was similar to that observed with high-dose PGE2 (Figure 4). Specificity for signaling via the EP4 receptor was confirmed with a highly specific EP4 receptor agonist, ONO-AE1 329, which caused a rapid decrease in TER (60% at 5 minutes, P < 0.001) and also with an EP4 receptor-specific antagonist, L161982, which completely abrogated the effect of 1 μmol/L PGE2 (Figure 5B).
      Figure thumbnail gr5
      Figure 5PGE2 signaling via EP4 receptor decreases TER. T84 monolayers on Transwell plates with stable TER (mean, 2422 ± 262 Ω·cm2) were used. The mean values at 5, 30, and 60 minutes were expressed as percentage of control and compared statistically with that of control. A: The monolayers were treated separately with various EP receptor agonists (3 μmol/L 17 phenyl trinor PGE2, an EP1/3 agonist; 3 μmol/L sulprostone, an EP3 agonist; 10 μmol/L butaprost, an EP2 agonist; and 500 nmol/L PGE1-OH, an EP4/2 agonist) and TER was recorded. B: The monolayers were exposed to 1 μmol/L ONO-AE1-329, a highly specific EP4 receptor agonist. They were also exposed to 1 μmol/L PGE2 after 6 hours pretreatment with 10 μmol/L of the EP4 receptor antagonist L161982. *P < 0.05; ***P < 0.001.

      PGE2 Secreted by E. histolytica Preferentially Signals via EP4 Receptor to Alter TER

      To identify whether ameba PGE2 preferentially signaled via the EP4 receptor to alter TER, T84 monolayers were pretreated for 6 hours with the EP4 receptor-specific antagonist L161982 and then were exposed to high concentrations (48 μg) of Wt ameba SP. As predicted, L161982 significantly attenuated the decrease in TER caused by ameba SP (P < 0.01 at 5 minutes; P < 0.05 at 30 minutes) (Figure 6). In contrast, the EP2 receptor-specific antagonist AH6809 did not inhibit the barrier-compromising effect of SP. These data clearly demonstrate that PGE2 present in SP of E. histolytica was responsible for decreasing TER by selectively activating the EP4 receptor on colonic epithelia. Moreover, these results also confirm the specificity of ameba-derived PGE2 in altering colonic TER. Although L161982 significantly prevented SP-induced decrease in TER, it could not completely abrogate the effect of SP, indicating that other ameba-derived factors can play a role in altering TER.
      Figure thumbnail gr6
      Figure 6PGE2 secreted by E. histolytica preferentially signals via EP4 receptor to alter TER. T84 monolayers with stable TER of >2000 Ω·cm2 were pretreated individually with 10 μmol/L of the EP4 receptor antagonist L161982 for 6 hours and with 10 μmol/L of the EP2 receptor antagonist AH6809 for 45 minutes and then were exposed to 48 μg SP from Wt E. histolytica. The difference in mean TER value at 5, 30, and 60 minutes was compared with that of Wt-SP. *P < 0.05; **P < 0.01.

      High-Dose PGE2 Alters Charge but Not Size Selectivity of TJs

      TER is the gold standard for the measurement of TJ integrity of contiguous epithelial monolayers. In fact, TJs act as dynamic barriers that selectively allow macromolecules or ions to permeate the paracellular space. Thus, to determine whether the decrease in TER induced by high-dose PGE2 corresponded with permeation of macromolecules across TJs, a paracellular tracer study using the lowest molecular weight tracer [14C]ethanolamine was performed. The Papp ratio (PGE2 treated versus control) did not differ significantly, indicating that the TJ barrier was not altered for permeability to macromolecules (Figure 7A); however, a 20% NaCl dilution potential assay, which tracks ion permeation across TJs, indicated that the TJ ion barrier was altered (Figure 7B). In fact, when 20% NaCl dilution was made on the apical side, the difference in dilution potential was significantly altered in the PGE2-treated group, compared with that of controls (P < 0.05 at 15 minutes; P < 0.001 at 30 minutes), indicating basal to apical ion flow across the TJ. Notably, alteration in dilution potential was not symmetrical, as evident from a 20% NaCl dilution made on the basal side. Indeed, the dilution potential difference was not significant at any measured time point, which rules out apical to basal ion flow. These results clearly demonstrate a unidirectional (basal to apical) flow/flux of ions across paracellular TJs in response to high-dose PGE2.
      Figure thumbnail gr7
      Figure 7Characterization of the effect of high-dose PGE2 on paracellular permeability. A: Paracellular tracer study. Confluent T84 monolayers on Transwell plates that reached a TER of >2000 Ω·cm2 were treated apically with or without 1 μmol/L PGE2 in medium containing the paracellular tracer [14C]ethanolamine (0.33 μCi/mL). At 0.5, 3, and 12 hours, radioactivity in both chambers was measured and apical to basal translocation of the tracer, calculated as the apparent permeability coefficient (Papp) and expressed as transport enhancement ratio R (Papp PGE2 treated versus Papp control). B: 20% NaCl dilution potential assay. After exposure to 1 μmol/L PGE2, 20% of apical/basal NaCl was replaced with an equal volume of equimolar mannitol. Potential difference was measured using a Millicell ERS apparatus every 5 minutes from 5 to 30 minutes and the dilution potential difference was compared with that of the control. *P < 0.05; **P < 0.01; ***P < 0.001.

      PGE2 Increases Paracellular TJ Sodium Ion Permeability

      As evident from our studies conducted by creating a NaCl gradient across epithelial monolayer, it is clear that either sodium (cation) or chloride (anion) passes through the TJ. Given the selectivity of TJs for cations,
      • Tang V.W.
      • Goodenough D.A.
      Paracellular ion channel at the tight junction.
      we examined if there was an increase in the basal to apical, paracellular flow of sodium ions in response to high-dose PGE2. The 1 μmol/L PGE2 treatment caused an increase in sodium potential on the apical side, with significance noted at 15 minutes, compared with control (increase of 18.63 mV, P < 0.001) (Figure 8A). To rule out the transcellular release of sodium ions, however, important membrane sodium channels/transporters such as epithelial Na+ channels, the Na+K+2Cl cotransporter, and the NaCl cotransporter were pharmacologically inhibited using amiloride, bumetanide, and hydrochlorothiazide, respectively, and the effect of their inhibition on 1 μmol/L PGE2-induced decrease in TER was analyzed (Figure 8B). Inhibition of these channels/transporters did not prevent PGE2-induced decrease of TER, which indicates a lesser role for membrane sodium channel/transporters in the apical release of sodium ions. These data indicate that the high-dose PGE2-induced apical flux of sodium ions goes essentially through the TJ.
      Figure thumbnail gr8
      Figure 8Role of TJs in sodium ion permeability. A: Assessment of sodium ion flux. After treatment with or without 1 μmol/L PGE2, 20 μL of apical medium was collected and examined for Na+ potential using the Orion Na+-sensitive electrode at 0, 5, 15, and 30 minutes, compared with control. B: The monolayers that reached a mean baseline TER of 3980 ± 458 Ω·cm2 were individually pretreated for 45 minutes with 100 μmol/L each of amiloride (a Na+ channel inhibitor) and hydrochloride thiazide (a NaCl cotransporter inhibitor), apically, and with 10 μmol/L bumetanide (a Na+K+2Cl cotransporter inhibitor), basolaterally. Apical membrane chloride channels (CFTR) were inhibited by 6 minutes pretreatment on the apical side with 1 mmol/L glibenclamide. TER was measured at 5, 30, and 60 minutes after 1 μmol/L PGE2 exposure and was compared with that of the group treated with 1 μmol/L PGE2 alone. ***P < 0.001.

      Activation of CFTR Is Essential for PGE2-Induced Alterations in TER

      PGE2 is well known for activating apical membrane chloride channels that increase luminal secretion of chloride ions. In fact, apical stimulation of T84 monolayers caused a robust increase in potential difference, indicating luminal chloride secretion (see Supplemental Figure S3 at http://ajp.amjpathol.org). To identify whether crosstalk occurs between PGE2-induced secretion of chloride ions (increase in potential difference) and a decrease in TER (alteration at the TJ), the activity of CFTR, which is an abundantly expressed colonic epithelial chloride channel, was inhibited with glibenclamide, and the effect of that inhibition on high-dose PGE2-induced decrease in TER was analyzed (Figure 8B). Of note, inactivation of CFTR significantly prevented the rapid decrease in TER, suggesting crosstalk between apical chloride secretion and alteration at the TJ.

      High-Dose PGE2 Shifts Claudin-4 from Cell Membrane to Cytoskeleton

      TJ dynamics are regulated by a group of proteins that constitute the TJ complex. To correlate rapid changes in TER induced by PGE2 with changes in TER due to structural alterations in TJs, we used a cellular fractionation technique that enables detection of TJ proteins that either dissociate from or remain integrated with the membrane complex. T84 monolayers treated with 1 μmol/L PGE2 for 5 minutes were fractionated into Triton-X 100 SF or IF. PGE2-untreated monolayers were similarly processed in parallel. Validity of the subcellular fractions was confirmed by probing for the localization of subcellular markers (Figure 9). GAPDH was used as a cytoplasmic marker, F-actin as a cytoskeletal marker, and NaKATPase as a membrane marker. GAPDH and NaKATPase were localized in the SF and F-actin was detected in the IF. We assumed that any membrane-bound TJ proteins not linked to the actin cytoskeleton would fractionate in the SF, and that those linked to actin, even if present on the membrane, would segregate to the IF. Based on this assumption, we examined the subcellular status of the major TJ proteins, with special emphasis on claudins, because they play a critical role in regulating paracellular ion transport.
      • Turksen K.
      • Troy T.C.
      Barriers built on claudins [Erratum appeared in J Cell Sci 2004, 117:4341].
      • Balkovetz D.F.
      Claudins at the gate: determinants of renal epithelial tight junction paracellular permeability.
      There was a significant shift in claudin-4 from the noncytoskeletal (SF) to cytoskeletal (IF) fraction in response to 1 μmol/L PGE2, whereas the other TJ proteins tested were unaltered (Figure 9).
      Figure thumbnail gr9
      Figure 9Subcellular fractionation studies highlighting PGE2-induced cytoskeletal shift of claudin-4. A: After 5 minutes treatment with 1 μmol/L PGE2, T84 cells were fractionated as Triton-X 100 soluble/noncytoskeletal fraction (SF) and Triton-X 100 insoluble/cytoskeletal fraction (IF) and were screened by Western blotting to detect localization of subcellular markers and analyzed for shift in six TJ proteins. B: Corresponding densitometry for the Western blots for the TJ proteins.
      To visualize the nature of the spatial alteration in claudin-4, immunofluorescence confocal microscopy was performed (Figure 10A). T84 monolayers treated with 1 μmol/L PGE2 for 5 minutes were probed for claudin-4 and ZO-1 (a marker for TJs), and the Z-stacks obtained were visualized for claudin-4 localization. Untreated T84 monolayers were used as controls. As seen in control monolayer, claudin-4 was predominantly present at TJs and a weak cytoplasmic localization of this protein was also observed. More importantly, claudin-4 was colocalized with ZO-1. However, in response to PGE2, claudin-4 markedly dissociated from the TJ, and the redistribution of claudin-4 followed a mosaic-like pattern. The 3D reconstruction of Z-stacks confirmed that PGE2 induced the dissociation of claudin-4 from the TJ (Figure 10B).
      Figure thumbnail gr10
      Figure 10Confocal microscopic studies indicating high-dose PGE2-induced dissociation of claudin-4 from TJs. T84 monolayers treated with or without 1 μmol/L PGE2 for 5 minutes were probed for cellular localization of claudin-4. Claudin-4 is indicated by green fluorescence tag and ZO-1 (a marker for TJs) by red fluorescence. A: Z-stacks of slices along xy planes throughout the cellular z axis at 0.35-μm intervals, as well as PXZ. In merged images, the blue arrow indicates sites where claudin-4 localizes with TJs and the white arrow points to dissociation from the TJ. Scale bar = 26.0 μm. B: 3D reconstruction of Z-stacks for the same experiments. Scale: 1 grid unit = 5.30951 μm.

      PGE2 Secreted by E. histolytica Dissociates Claudin-4 from TJs

      To determine specificity for PGE2 secreted by E. histolytica in inducing spatial alteration of claudin-4, T84 monolayers were treated for 5 minutes with 48 μg SP derived from either Wt amebas or from amebas pretreated for 16 hours with ASA. Monolayers were then examined for claudin-4 localization as determined by confocal microscopy. SP dissociated claudin-4 from the TJ (with ZO-1 used as a marker for TJs); however, redistribution of claudin-4 followed a punctate-like pattern (Figure 11). As predicted, TJ integrity of claudin-4 was partially altered in T84 monolayers exposed to SP derived from trophozoites treated with ASA to inhibit PGE2 production (Figure 11, A and B). Curiously, however, noticeable punctate-like patterns were still observed. Nonetheless, the chicken-wire-like TJ appearance of claudin-4 can be clearly seen in the ASA-SP treated panel and in the 3D reconstruction of Z-stacks (Figure 11B). These data demonstrate that PGE2 secreted by E. histolytica can selectively dissociate claudin-4 from the TJ complex.
      Figure thumbnail gr11
      Figure 11PGE2 secreted by E. histolytica dissociates claudin-4 from TJs. T84 monolayers treated or not for 5 minutes with 48 μg SP either from Wt amebas (Wt-SP) or from amebas pretreated for 16 hours with aspirin (ASA 16 hours-SP) were checked for localization of claudin-4 (green). ZO-1 was used as a marker for TJs (red). The red and green images were then merged. A: Z-stacks of slices along xy planes throughout the cellular z axis at 0.35-μm intervals. Scale bar = 26.0 μm). Blue arrows indicate colocalization and white arrows point to dissociation of claudin-4 from the TJ. B: 3D reconstruction of Z-stacks for the same experiments. Scale: 1 grid unit = 5.30951 μm.

      Discussion

      Intestinal amebiasis typically manifests as watery and/or bloody diarrhea. However, the parasite-associated factors and the mechanisms by which E. histolytica causes diarrhea have not been elucidated in detail. In the present study, we identified PGE2 secreted by E. histolytica as one of the major factors that alter the colonic epithelial barrier and transport functions. We also highlight a role for PGE2 in generating a NaCl gradient across epithelia that acts as a potential trigger for osmotic water flow across the paracellular space. Thus, we have unraveled a pathological role for E. histolytica-derived PGE2 in the onset of watery diarrhea, the most common manifestation of the intestinal amebiasis that occurs in 50 million cases annually during pathogenesis of amebic colitis. This is the first report identifying ameba-derived PGE2 as a factor that triggers events potentially leading to diarrhea in diseased subjects.
      Previous studies suggest that a direct contact between E. histolytica and the host epithelium is required for increased permeability of the colonic mucosa
      • Li E.
      • Stenson W.F.
      • Kunz-Jenkins C.
      • Swanson P.E.
      • Duncan R.
      • Stanley Jr, S.L.
      Entamoeba histolytica interactions with polarized human intestinal Caco-2 epithelial cells.
      • Leroy A.
      • Lauwaet T.
      • De Bruyne G.
      • Cornelissen M.
      • Mareel M.
      Entamoeba histolytica disturbs the tight junction complex in human enteric T84 cell layers.
      and hemorrhage, the causal events of inflammatory/bloody diarrhea. The present findings show that E. histolytica is capable of initiating disease pathogenesis even in a contact-independent manner. Recent studies have also highlighted the significance of contact-independent mechanisms in the pathogenesis of amebiasis. For example, CP5 secreted by E. histolytica has been shown to degrade colonic MUC2 mucin
      • Lidell M.E.
      • Moncada D.M.
      • Chadee K.
      • Hansson G.C.
      Entamoeba histolytica cysteine proteases cleave the MUC2 mucin in its C-terminal domain and dissolve the protective colonic mucus gel.
      • Moncada D.
      • Keller K.
      • Chadee K.
      Entamoeba histolytica cysteine proteinases disrupt the polymeric structure of colonic mucin and alter its protective function.
      and to alter innate host defense, thus setting the stage for amebic invasion of the mucosa. Similarly, the roles of other SP components, such as serpins (serine proteinase inhibitors), that manipulate the host pathophysiological function have been identified.
      • Riahi Y.
      • Siman-Tov R.
      • Ankri S.
      Molecular cloning, expression and characterization of a serine proteinase inhibitor gene from Entamoeba histolytica.
      Among the components secreted by E. histolytica, PGE2 assumes greater significance, because it is a well-known lipid mediator produced by the host and is associated with various physiological and pathophysiological functions of the gastrointestinal tract.
      • Dey I.
      • Lejeune M.
      • Chadee K.
      Prostaglandin E2 receptor distribution and function in the gastrointestinal tract.
      In particular, the pathological association of host-derived PGE2 with various diarrheal diseases is well documented.
      • Beubler E.
      • Schuligoi R.
      Mechanisms of cholera toxin-induced diarrhea.
      • Resta-Lenert S.
      • Barrett K.E.
      Enteroinvasive bacteria alter barrier and transport properties of human intestinal epithelium: role of iNOS and COX-2.
      • Alcantara C.
      • Stenson W.F.
      • Steiner T.S.
      • Guerrant R.L.
      Role of inducible cyclooxygenase and prostaglandins in Clostridium difficile toxin A-induced secretion and inflammation in an animal model.
      Notably, E. histolytica is the only known gut pathogen that synthesizes and secretes PGE2,
      • Dey I.
      • Keller K.
      • Belley A.
      • Chadee K.
      Identification and characterization of a cyclooxygenase-like enzyme from Entamoeba histolytica.
      and the identity of the SP with host PGE2 was confirmed by gas chromatography/mass spectrometry analysis.
      • Belley A.
      • Chadee K.
      Production of prostaglandin E(2) by Entamoeba histolytica via a novel cyclooxygenase.
      In the present study, using a highly sensitive CLIA kit designed to assay host PGE2 in biological fluids, we confirmed the presence of PGE2 in SP derived from live amebas. Why E. histolytica should produce a host mediator is intriguing. Given that E. histolytica requires AA for the biosynthesis of PGE2 and that host diet is a rich source of AA, we speculate that the parasite is naturally selected to produce high levels of PGE2 in the gut. In this regard, it was not unexpected that trophozoites cultured in the presence of 100 μmol/L AA for 2 hours in HBSS exhibited a robust increase in PGE2 production. Because increased production of PGE2 in the gut is invariably associated with pathophysiological functions, we hypothesized that high-output PGE2 secreted by E. histolytica can play a significant role in the development of acute diarrhea by altering the host epithelial barrier and transport functions. Our results identify PGE2 as the bioactive factor present in SP responsible for the disruption of the colonic epithelial TJ barrier.
      We used T84 colonic monolayers grown on Transwell plates as an in vitro epithelial model. This cell line forms a polarized monolayer with effective TJs and with an inherent monolayer resistance similar to that observed in vivo.
      • Silvis M.R.
      • Bertrand C.A.
      • Ameen N.
      • Golin-Bisello F.
      • Butterworth M.B.
      • Frizzell R.A.
      • Bradbury N.A.
      Rab11b regulates the apical recycling of the cystic fibrosis transmembrane conductance regulator in polarized intestinal epithelial cells.
      • McKay D.M.
      • Watson J.L.
      • Wang A.
      • Caldwell J.
      • Prescott D.
      • Ceponis P.M.
      • Di Leo V.
      • Lu J.
      Phosphatidylinositol 3′-kinase is a critical mediator of interferon-gamma-induced increases in enteric epithelial permeability.
      • Wills N.K.
      • Lewis S.A.
      • Eaton D.C.
      Active and passive properties of rabbit descending colon: a microelectrode and nystatin study.
      Moreover, T84 is one of the few transformed colonic cell lines that produce very low endogenous PGE2,
      • Resta-Lenert S.
      • Barrett K.E.
      Enteroinvasive bacteria alter barrier and transport properties of human intestinal epithelium: role of iNOS and COX-2.
      making it the cell line of choice for studying the effects of exogenous PGE2. Because excretory and secretory components of E. histolytica freely interact with the colonic epithelium on the luminal side (apical side of epithelia), we investigated the effects of SP on colonic epithelial integrity by exposing the T84 monolayer to SP on its apical side. Contrary to previous reports,
      • Li E.
      • Stenson W.F.
      • Kunz-Jenkins C.
      • Swanson P.E.
      • Duncan R.
      • Stanley Jr, S.L.
      Entamoeba histolytica interactions with polarized human intestinal Caco-2 epithelial cells.
      we found that secreted components consistently decreased TER. We speculate that SP collected from E. histolytica that were serum-starved before the collection of SP, as others have reported,
      • Li E.
      • Stenson W.F.
      • Kunz-Jenkins C.
      • Swanson P.E.
      • Duncan R.
      • Stanley Jr, S.L.
      Entamoeba histolytica interactions with polarized human intestinal Caco-2 epithelial cells.
      could account for the observed differences. An exogenous supply of AA is required for the parasite to produce PGE2,
      • Dey I.
      • Keller K.
      • Belley A.
      • Chadee K.
      Identification and characterization of a cyclooxygenase-like enzyme from Entamoeba histolytica.
      and serum is a rich source of AA. To circumvent this problem, we cultured amebas in TYI-S-33 medium containing 20% serum and the SP was collected without subjecting amebas to serum starvation. SP was collected in HBSS after thorough washing of trophozoites, to avoid serum contamination, even though serum alone had no adverse effects on TER. Moreover, we used 0.4-μm pore-sized polyethylene Transwell inserts as a substratum to grow T84 monolayers. In this system, T84 monolayers consistently reached a TER of >2000 Ω·cm2. However, the same cells when grown on 3-μm pore-sized polycarbonate plates, as described previously,
      • Li E.
      • Stenson W.F.
      • Kunz-Jenkins C.
      • Swanson P.E.
      • Duncan R.
      • Stanley Jr, S.L.
      Entamoeba histolytica interactions with polarized human intestinal Caco-2 epithelial cells.
      did not reach a TER of even 250 Ω·cm2 (data not shown). We propose that consistent maintenance of high TER values is critical for studies on monolayer resistance.
      We confirmed the specificity of ameba-derived PGE2 in altering TER by studying the effect of SP collected from trophozoites that were inhibited (with ASA) in their ability to synthesize PGE2, as described previously.
      • Dey I.
      • Keller K.
      • Belley A.
      • Chadee K.
      Identification and characterization of a cyclooxygenase-like enzyme from Entamoeba histolytica.
      • Dey I.
      • Chadee K.
      Prostaglandin E2 produced by Entamoeba histolytica binds to EP4 receptors and stimulates interleukin-8 production in human colonic cells.
      SP derived from ASA-treated amebas significantly prevented the decrease in TER. To prove specificity for PGE2 in SP in decreasing TER, a neutralizing antibody against PGE2 can be used. Specific neutralizing antibodies for PGE2 are not commercially available, however, and unfortunately a neutralizing antibody procured from Cayman Chemical (catalog no. 10009814) did not yield positive results in the present study (see Supplemental Figure S4 at http://ajp.amjpathol.org). Nevertheless, pretreatment of colonic epithelial monolayer with EP4 receptor-specific antagonist significantly abrogated SP-induced decrease in TER, confirming the specificity for ameba-derived PGE2 in altering TER. However, given that ASA or EP4 antagonist did not completely abrogate the decrease in TER, other ameba-secreted components may also play a role in decreasing epithelial resistance. Such molecules remain to be identified, because the major virulent components of the parasite (including CP5, amebapore, and serotonin) did not alter TER. At present, we cannot rule out a role for other putative secreted components apart from PGE2 in altering epithelial monolayer resistance.
      PGE2 is well known for its dual role in exerting both physiological and pathophysiological functions in the gastrointestinal tract, based on amounts produced in a local milieu.
      • Dey I.
      • Lejeune M.
      • Chadee K.
      Prostaglandin E2 receptor distribution and function in the gastrointestinal tract.
      To determine whether increased parasite production of PGE2 resulted in a corresponding decrease in TER, we tested graded doses of SP on the resistance of colonic epithelial monolayer. SP followed a dose-response relationship, indicating a concentration dependency for its effects on TER. This effect was mirrored by commercially available purified PGE2. Nonetheless, we determined the molar concentration of PGE2 present in the high dose of SP (48 μg) used in our experiments. Surprisingly, the concentration was approximately 4 nmol/L. However, SP followed a dose-response in decreasing TER very similar to that of exogenous PGE2. ASA treatment of amebas significantly decreased the amount of PGE2 in the SP and had a corresponding effect on TER. Pretreatment of colonic epithelial monolayer with an EP4 receptor-specific antagonist significantly prevented the decrease in TER caused by a high dose of SP. Thus, we could establish the significance of ameba-derived PGE2 signaling via the EP4 receptor of epithelial cells in decreasing TER. The potential for other known (see Supplemental Figure S1 at http://ajp.amjpathol.org) and unknown SP factors that can enhance the effects of PGE2 cannot be discounted.
      TER is the gold standard for the measurement of integrity of paracellular TJs. Loss of TER in contiguous monolayers strongly indicates a break in the TJ barrier for macromolecules and/or ions. In the present study, a PGE2-induced decrease in TER corresponded with TJ permeability to ions. This effect of PGE2 is in complete contrast to the observed macromolecule permeability when E. histolytica physically contacts and perturbs the epithelial monolayer.
      • Li E.
      • Stenson W.F.
      • Kunz-Jenkins C.
      • Swanson P.E.
      • Duncan R.
      • Stanley Jr, S.L.
      Entamoeba histolytica interactions with polarized human intestinal Caco-2 epithelial cells.
      • Leroy A.
      • Lauwaet T.
      • De Bruyne G.
      • Cornelissen M.
      • Mareel M.
      Entamoeba histolytica disturbs the tight junction complex in human enteric T84 cell layers.
      Thus, E. histolytica-induced alterations in epithelial monolayer resistance are distinct for contact-dependent versus contact-independent modes of interaction. Given that high-output PGE2 rapidly decreases TER, we speculate that the EP4 receptors activated by PGE2 and through which PGE2 signals must be localized on the apical plasma membrane of colonic epithelial cells. We confirmed apical membrane localization of EP4 receptors in human colonic mucosal biopsies by immunohistochemistry and on T84 monolayers by confocal microscopy.
      • Lejeune M.
      • Leung P.
      • Beck P.L.
      • Chadee K.
      Role of EP4 receptor and prostaglandin transporter in prostaglandin E2-induced alteration in colonic epithelial barrier integrity.
      We also propose that the rapidity of alteration in TER can be attributed to a sudden spatial alteration of TJ proteins, rather than to their comparatively complex transcriptional/translational regulation. Cell fractionation and immunofluorescence confocal microscopic studies confirmed spatial alteration for claudin-4. In fact, fractionation studies indicate an increased shift of claudin-4 from the membrane to the cytoskeletal fraction. Using an inhibitor for actin polymerization, we demonstrate that claudin-4 is associated with the cytoskeleton when it is dissociated from TJs. Surprisingly, claudin-4 is not colocalized with F-actin, indicating its indirect association with the actin cytoskeleton (see Supplemental Figure S5 at http://ajp.amjpathol.org). At present, it is not clear how the actin cytoskeleton pulls claudin-4 away from the TJ.
      It is well established that the claudin family of proteins regulates TJ ion permeability. More importantly, claudin-4 regulates paracellular sodium ion permeability. Its presence at the TJ decreases sodium ion permeability across paracellular space.
      • Van Itallie C.
      • Rahner C.
      • Anderson J.M.
      Regulated expression of claudin-4 decreases paracellular conductance through a selective decrease in sodium permeability.
      The present findings show selective dissociation of claudin-4 from TJ in response to ameba-derived PGE2 or a high dose of purified PGE2. We note that crosstalk occurs between PGE2-induced activation of apical chloride channels (CFTR) and alteration at the TJ. More importantly, PGE2 induced basal to apical (unidirectional) movement/flux of sodium ion across the TJ, as evident from bidirectional assessment of 20% NaCl dilution potential and by quantification of apical sodium potential. It seems that PGE2 induces unidirectional passive flow of sodium across the TJ in response to a robust presence of chloride on the apical side. The rapidity with which PGE2 induces apical chloride secretion, spatial alteration of claudin-4 at TJs (occurs at 5 minutes), and the increase in apical sodium potential (occurs at 15 minutes) in the absence of activation of major membrane sodium channels is highly suggestive of this phenomenon. Thus, PGE2 disrupts the TJ barrier for sodium ions, which, along with increased apical chloride secretion, would draw water toward the luminal side to maintain osmotic balance in vivo.
      Taken together, these results demonstrate that the observed alteration in TJs is an effect of PGE2 secreted by E. histolytica. The present findings further reinforce the concept that this parasite can initiate disease pathogenesis even in the absence of parasite-epithelial contact, and highlight a critical role for E. histolytica-derived PGE2 in the onset of events that lead to diarrhea in intestinal amebiasis. Identification of EP receptor subtype 4 through which parasite-derived PGE2 signals to alter TER widens the scope for pharmacological intervention in amebic diarrhea. This is of paramount significance, in light of the fact that amebic diarrhea often causes life-threatening complications, especially in young children. Finally, the present findings can be extrapolated to various other diarrheal diseases in which increased production of PGE2 is a hallmark.

      Acknowledgments

      We thank Dr. Mark Giembyk (University of Calgary, Canada) for the generous gift of ONO-AE1-329 and L-161982, Dr. David Mirelman (Weizmann Institute of Science, Israel) for providing E. histolytica that lacks amebapore and/or CP5, Tehmeena Malik and Dr. Pina Colarusso (University of Calgary Live Cell Imaging Facility) for technical support, and Dr. Pinaki Bose for critically reviewing the manuscript.

      Supplementary data

      • Supplemental Figure S1

        Putative E. histolytica virulence components or inactive PGE2 analog does not affect TER. T84 monolayers were exposed for 5, 30, and 60 minutes to 48 μg SP from wild-type E. histolytica (Wt-SP) or with SP from amebas genetically deficient in amebapore and cysteine CP5 (RB8-SP, a gift from David Mirelman), as well as Wt-SP treated with 100 μmol/L E-64, a cysteine proteinase inhibitor (Wt-SP+E-64). Monolayers were also individually exposed to 1 μmol/L of a commercially available analog of serotonin (5-HT; a known E. histolytica secreted component), to PGE2 (as a positive control), and to an inactive PGE2 analog, 15-keto PGE2.

      • Supplemental Figure S2

        Effect of high-dose PGE2 on TER of Caco-2 epithelial monolayer. Caco-2 monolayer grown on Transwell plates that reached a stable TER of ∼1100 Ω.cm2 was exposed on the apical side to 1 μmol/L PGE2 for 5, 30, or 60 minutes. The mean values were compared with control and expressed as percentage of control. The effect of 1 μmol/L PGE2 on TER of T84 monolayer is shown for comparison. ***P < 0.001.

      • Supplemental Figure S3

        Effect of high-dose PGE2 on potential difference of T84 epithelial monolayer. T84 monolayers grown on Transwell plates that reached a stable TER of >2000 Ω·cm2 were bathed in PBS + BSA and were stimulated or not on the apical side with 1 μmol/L PGE2. Potential difference in millivolts was measured using Millipore ERS apparatus.

      • Supplemental Figure S4

        Effect of PGE2 neutralizing antibody (Ab) on SP-induced TER. T84 monolayers were exposed for 5 minutes to purified PGE2 or E. histolytica SP incubated for 6 hours or not with various concentrations of PGE2 neutralizing Ab. In one experiment, the PGE2 neutralizing antibody was immunoprecipitated (IP) after incubation with PGE2 and the effect of the supernatant on TER analyzed [(PGE2 + 20 μg Ab) IP]. E64 pretreatment of SP was done to inhibit cysteine proteinases of amebas that could degrade the neutralizing PGE2 antibody.

      • Supplemental Figure S5

        Role of actin cytoskeleton in PGE2-induced dissociation of claudin-4. T84 monolayers treated with or without 20 μmol/L cytochalasin D for 1 hour and then were exposed or not to 1 μmol/L PGE2 for 5 minutes. A: Cellular localization of claudin-4 was probed by green fluorescence tag and ZO-1 (a marker for TJ) by red fluorescence. Z-stacks of slices along xy planes throughout the cellular z axis at 0.35-μm intervals. B: Analysis for colocalization of claudin-4 with F-actin. Claudin-4 was tagged with green fluorescence and F-actin was stained with phalloidin (red). Z-stacks of slices along xy planes throughout the cellular z axis at 0.35-μm intervals. Scale: 1 grid unit = 5.30951 µm.

      References

        • Schuster F.L.
        • Visvesvara G.S.
        Amebae and ciliated protozoa as causal agents of waterborne zoonotic disease.
        Vet Parasitol. 2004; 126: 91-120
        • Anane S.
        • Khaled S.
        [Entamoeba histolytica and Entamoeba dispar: differentiation methods and implications].
        Ann Biol Clin (Paris). 2005; 63 (French): 7-13
        • Haque R.
        • Mondal D.
        • Duggal P.
        • Kabir M.
        • Roy S.
        • Farr B.M.
        • Sack R.B.
        • Petri Jr, W.A.
        Entamoeba histolytica infection in children and protection from subsequent amebiasis.
        Infect Immun. 2006; 74: 904-909
        • Haque R.
        • Mondal D.
        • Kirkpatrick B.D.
        • Akther S.
        • Farr B.M.
        • Sack R.B.
        • Petri Jr, W.A.
        Epidemiologic and clinical characteristics of acute diarrhea with emphasis on Entamoeba histolytica infections in preschool children in an urban slum of Dhaka, Bangladesh.
        Am J Trop Med Hyg. 2003; 69: 398-405
        • Niyogi S.K.
        • Saha M.R.
        • De S.P.
        Enteropathogens associated with acute diarrhoeal diseases.
        Indian J Public Health. 1994; 38: 29-32
        • Navaneethan U.
        • Giannella R.A.
        Mechanisms of infectious diarrhea.
        Nat Clin Pract Gastroenterol Hepatol. 2008; 5: 637-647
        • Beubler E.
        • Schuligoi R.
        Mechanisms of cholera toxin-induced diarrhea.
        Ann N Y Acad Sci. 2000; 915: 339-346
        • Resta-Lenert S.
        • Barrett K.E.
        Enteroinvasive bacteria alter barrier and transport properties of human intestinal epithelium: role of iNOS and COX-2.
        Gastroenterology. 2002; 122: 1070-1087
        • Alcantara C.
        • Stenson W.F.
        • Steiner T.S.
        • Guerrant R.L.
        Role of inducible cyclooxygenase and prostaglandins in Clostridium difficile toxin A-induced secretion and inflammation in an animal model.
        J Infect Dis. 2001; 184: 648-652
        • Ahrenstedt O.
        • Hällgren R.
        • Knutson L.
        Jejunal release of prostaglandin E2 in Crohn's disease: relation to disease activity and first-degree relatives.
        J Gastroenterol Hepatol. 1994; 9: 539-543
        • Stenson W.F.
        • Zhang Z.
        • Riehl T.
        • Stanley Jr, S.L.
        Amebic infection in the human colon induces cyclooxygenase-2.
        Infect Immun. 2001; 69: 3382-3388
        • Sánchez-Ramírez B.
        • Ramírez-Gil M.
        • Vázquez-Moctezuma I.
        • Ramos-Martínez E.
        • Talamás-Rohana P.
        Entamoeba histolytica: induction of cyclooxygenase-2 expression during amoebic liver abscess formation in hamsters (Mesocricetus auratus).
        Exp Parasitol. 2004; 106: 119-125
        • Belley A.
        • Chadee K.
        Eicosanoid production by parasites: from pathogenesis to immunomodulation?.
        Parasitol Today. 1995; 11: 327-334
        • Belley A.
        • Chadee K.
        Production of prostaglandin E(2) by Entamoeba histolytica via a novel cyclooxygenase.
        Arch Med Res. 2000; 31: S74-S75
        • Dey I.
        • Keller K.
        • Belley A.
        • Chadee K.
        Identification and characterization of a cyclooxygenase-like enzyme from Entamoeba histolytica.
        Proc Natl Acad Sci USA. 2003; 100: 13561-13566
        • Meyer B.J.
        • Mann N.J.
        • Lewis J.L.
        • Milligan G.C.
        • Sinclair A.J.
        • Howe P.R.
        Dietary intakes and food sources of omega-6 and omega-3 polyunsaturated fatty acids.
        Lipids. 2003; 38: 391-398
        • Sioen I.A.
        • Pynaert I.
        • Matthys C.
        • De Backer G.
        • Van Camp J.
        • De Henauw S.
        Dietary intakes and food sources of fatty acids for Belgian women, focused on n-6 and n-3 polyunsaturated fatty acids.
        Lipids. 2006; 41: 415-422
        • Dey I.
        • Chadee K.
        Prostaglandin E2 produced by Entamoeba histolytica binds to EP4 receptors and stimulates interleukin-8 production in human colonic cells.
        Infect Immun. 2008; 76: 5158-5163
        • Berkes J.
        • Viswanathan V.K.
        • Savkovic S.D.
        • Hecht G.
        Intestinal epithelial responses to enteric pathogens: effects on the tight junction barrier, ion transport, and inflammation.
        Gut. 2003; 52: 439-451
        • Field M.
        • Fromm D.
        • al-Awqati Q.
        • Greenough 3rd, W.B.
        Effect of cholera enterotoxin on ion transport across isolated ileal mucosa.
        J Clin Invest. 1972; 51: 796-804
        • Keeler R.
        • Wong N.L.
        Evidence that prostaglandin E2 stimulates chloride secretion in cultured A6 renal epithelial cells.
        Am J Physiol. 1986; 250: F511-F515
        • Bunce K.T.
        • Spraggs C.F.
        Stimulation of electrogenic chloride secretion by prostaglandin E2 in guinea-pig isolated gastric mucosa.
        J Physiol. 1988; 400: 381-394
        • Deachapunya C.
        • O'Grady S.M.
        Regulation of chloride secretion across porcine endometrial epithelial cells by prostaglandin E2.
        J Physiol. 1998; 508: 31-47
        • Schneeberger E.E.
        • Lynch R.D.
        The tight junction: a multifunctional complex.
        Am J Physiol Cell Physiol. 2004; 286: C1213-C1228
        • Harhaj N.S.
        • Antonetti D.A.
        Regulation of tight junctions and loss of barrier function in pathophysiology.
        Int J Biochem Cell Biol. 2004; 36: 1206-1237
        • González-Mariscal L.
        • Betanzos A.
        • Nava P.
        • Jaramillo B.E.
        Tight junction proteins.
        Prog Biophys Mol Biol. 2003; 81: 1-44
        • Turksen K.
        • Troy T.C.
        Barriers built on claudins [Erratum appeared in J Cell Sci 2004, 117:4341].
        J Cell Sci. 2004; 117: 2435-2447
        • Balkovetz D.F.
        Claudins at the gate: determinants of renal epithelial tight junction paracellular permeability.
        Am J Physiol Renal Physiol. 2006; 290: F572-F579
        • Yu Y.
        • Chadee K.
        Entamoeba histolytica stimulates interleukin 8 from human colonic epithelial cells without parasite-enterocyte contact.
        Gastroenterology. 1997; 112: 1536-1547
        • Tang C.H.
        • Yang R.S.
        • Fu W.M.
        Prostaglandin E2 stimulates fibronectin expression through EP1 receptor, phospholipase C, protein kinase Calpha, and c-Src pathway in primary cultured rat osteoblasts.
        J Biol Chem. 2005; 280: 22907-22916
        • Chen B.C.
        • Liao C.C.
        • Hsu M.J.
        • Liao Y.T.
        • Lin C.C.
        • Sheu J.R.
        • Lin C.H.
        Peptidoglycan-induced IL-6 production in RAW 264.7 macrophages is mediated by cyclooxygenase-2, PGE2/PGE4 receptors, protein kinase A, I kappa B kinase, and NF-kappa B.
        J Immunol. 2006; 177: 681-693
        • Shao J.
        • Sheng H.
        Prostaglandin E2 induces the expression of IL-1alpha in colon cancer cells.
        J Immunol. 2007; 178: 4097-4103
        • Dey I.
        • Giembycz M.A.
        • Chadee K.
        Prostaglandin E(2) couples through EP(4) prostanoid receptors to induce IL-8 production in human colonic epithelial cell lines.
        Br J Pharmacol. 2009; 156: 475-485
        • Cherukuri D.P.
        • Chen X.B.
        • Goulet A.C.
        • Young R.N.
        • Han Y.
        • Heimark R.L.
        • Regan J.W.
        • Meuillet E.
        • Nelson M.A.
        The EP4 receptor antagonist, L-161,982, blocks prostaglandin E2-induced signal transduction and cell proliferation in HCA-7 colon cancer cells.
        Exp Cell Res. 2007; 313: 2969-2979
        • Tang V.W.
        • Goodenough D.A.
        Paracellular ion channel at the tight junction.
        Biophys J. 2003; 84: 1660-1673
        • Grobovac V.
        • Bernkop-Schnürch A.
        Improvement of the intestinal membrane permeability of low molecular weight heparin by complexation with stem bromelain.
        Int J Pharm. 2006; 326: 153-159
        • Rao R.K.
        • Li L.
        • Baker R.D.
        • Baker S.S.
        • Gupta A.
        Glutathione oxidation and PTPase inhibition by hydrogen peroxide in Caco-2 cell monolayer.
        Am J Physiol Gastrointest Liver Physiol. 2000; 279: G332-G340
        • de Jong J.C.
        • Willems P.H.
        • Mooren F.J.
        • van den Heuvel L.P.
        • Knoers N.V.
        • Bindels R.J.
        The structural unit of the thiazide-sensitive NaCl cotransporter is a homodimer.
        J Biol Chem. 2003; 278: 24302-24307
        • D'Andrea L.
        • Lytle C.
        • Matthews J.B.
        • Hofman P.
        • Forbush 3rd, B.
        • Madara J.L.
        Na: K:2Cl cotransporter (NKCC) of intestinal epithelial cells.
        J Biol Chem. 1996; 271: 28969-28976
        • Kim J.A.
        • Kang Y.S.
        • Lee S.H.
        • Lee E.H.
        • Yoo B.H.
        • Lee Y.S.
        Glibenclamide induces apoptosis through inhibition of cystic fibrosis transmembrane conductance regulator (CFTR) Cl(−) channels and intracellular Ca(2+) release in HepG2 human hepatoblastoma cells.
        Biochem Biophys Res Commun. 1999; 261: 682-688
        • Sheth P.
        • Basuroy S.
        • Li C.
        • Naren A.P.
        • Rao R.K.
        Role of phosphatidylinositol 3-kinase in oxidative stress-induced disruption of tight junctions.
        J Biol Chem. 2003; 278: 49239-49245
        • Dey I.
        • Lejeune M.
        • Chadee K.
        Prostaglandin E2 receptor distribution and function in the gastrointestinal tract.
        Br J Pharmacol. 2006; 149: 611-623
        • Li E.
        • Stenson W.F.
        • Kunz-Jenkins C.
        • Swanson P.E.
        • Duncan R.
        • Stanley Jr, S.L.
        Entamoeba histolytica interactions with polarized human intestinal Caco-2 epithelial cells.
        Infect Immun. 1994; 62: 5112-5119
        • Leroy A.
        • Lauwaet T.
        • De Bruyne G.
        • Cornelissen M.
        • Mareel M.
        Entamoeba histolytica disturbs the tight junction complex in human enteric T84 cell layers.
        FASEB J. 2000; 14: 1139-1146
        • Lidell M.E.
        • Moncada D.M.
        • Chadee K.
        • Hansson G.C.
        Entamoeba histolytica cysteine proteases cleave the MUC2 mucin in its C-terminal domain and dissolve the protective colonic mucus gel.
        Proc Natl Acad Sci USA. 2006; 103: 9298-9303
        • Moncada D.
        • Keller K.
        • Chadee K.
        Entamoeba histolytica cysteine proteinases disrupt the polymeric structure of colonic mucin and alter its protective function.
        Infect Immun. 2003; 71: 838-844
        • Riahi Y.
        • Siman-Tov R.
        • Ankri S.
        Molecular cloning, expression and characterization of a serine proteinase inhibitor gene from Entamoeba histolytica.
        Mol Biochem Parasitol. 2004; 133: 153-162
        • Silvis M.R.
        • Bertrand C.A.
        • Ameen N.
        • Golin-Bisello F.
        • Butterworth M.B.
        • Frizzell R.A.
        • Bradbury N.A.
        Rab11b regulates the apical recycling of the cystic fibrosis transmembrane conductance regulator in polarized intestinal epithelial cells.
        Mol Biol Cell. 2009; 20: 2337-2350
        • McKay D.M.
        • Watson J.L.
        • Wang A.
        • Caldwell J.
        • Prescott D.
        • Ceponis P.M.
        • Di Leo V.
        • Lu J.
        Phosphatidylinositol 3′-kinase is a critical mediator of interferon-gamma-induced increases in enteric epithelial permeability.
        J Pharmacol Exp Ther. 2007; 320: 1013-1022
        • Wills N.K.
        • Lewis S.A.
        • Eaton D.C.
        Active and passive properties of rabbit descending colon: a microelectrode and nystatin study.
        J Membr Biol. 1979; 45: 81-108
        • Lejeune M.
        • Leung P.
        • Beck P.L.
        • Chadee K.
        Role of EP4 receptor and prostaglandin transporter in prostaglandin E2-induced alteration in colonic epithelial barrier integrity.
        Am J Physiol Gastrointest Liver Physiol. 2010; 299: G1097-G1105
        • Van Itallie C.
        • Rahner C.
        • Anderson J.M.
        Regulated expression of claudin-4 decreases paracellular conductance through a selective decrease in sodium permeability.
        J Clin Invest. 2001; 107: 1319-1327