Interferon-γ Inhibits Intestinal Restitution by Preventing Gap Junction Communication Between Enterocytes

Background & Aims: Necrotizing enterocolitis (NEC) is characterized by interferon-gamma (IFN-γ) release and inadequate intestinal restitution. Because enterocytes migrate together, mucosal healing may require interenterocyte communication via connexin 43-mediated gap junctions. We hypothesize that enterocyte migration requires interenterocyte communication, that IFN impairs migration by impairing connexin 43, and that impaired healing during NEC is associated with reduced gap junctions. Methods: NEC was induced in Swiss-Webster or IFN−/− mice, and restitution was determined in the presence of the gap junction inhibitor oleamide, or via time-lapse microscopy of IEC-6 cells. Connexin 43 expression, trafficking, and localization were detected in cultured or primary enterocytes or mouse or human intestine by confocal microscopy and 35S-labeling, and gap junction communication was assessed using live microscopy with oleamide or connexin 43 siRNA. Results: Enterocytes expressed connexin 43 in vitro and in vivo, and exchanged fluorescent dye via gap junctions. Gap junction inhibition significantly reduced enterocyte migration in vitro and in vivo. NEC was associated with IFN release and loss of enterocyte connexin 43 expression. IFN inhibited enterocyte migration by reducing gap junction communication through the dephosphorylation and internalization of connexin 43. Gap junction inhibition significantly increased NEC severity, whereas reversal of the inhibitory effects of IFN on gap junction communication restored enterocyte migration after IFN exposure. Strikingly, IFN−/− mice were protected from the development of NEC, and showed restored connexin 43 expression and intestinal restitution. Conclusions: IFN inhibits enterocyte migration by preventing interenterocyte gap junction communication. Connexin 43 loss may provide insights into the development of NEC, in which restitution is impaired. Background & Aims: Necrotizing enterocolitis (NEC) is characterized by interferon-gamma (IFN-γ) release and inadequate intestinal restitution. Because enterocytes migrate together, mucosal healing may require interenterocyte communication via connexin 43-mediated gap junctions. We hypothesize that enterocyte migration requires interenterocyte communication, that IFN impairs migration by impairing connexin 43, and that impaired healing during NEC is associated with reduced gap junctions. Methods: NEC was induced in Swiss-Webster or IFN−/− mice, and restitution was determined in the presence of the gap junction inhibitor oleamide, or via time-lapse microscopy of IEC-6 cells. Connexin 43 expression, trafficking, and localization were detected in cultured or primary enterocytes or mouse or human intestine by confocal microscopy and 35S-labeling, and gap junction communication was assessed using live microscopy with oleamide or connexin 43 siRNA. Results: Enterocytes expressed connexin 43 in vitro and in vivo, and exchanged fluorescent dye via gap junctions. Gap junction inhibition significantly reduced enterocyte migration in vitro and in vivo. NEC was associated with IFN release and loss of enterocyte connexin 43 expression. IFN inhibited enterocyte migration by reducing gap junction communication through the dephosphorylation and internalization of connexin 43. Gap junction inhibition significantly increased NEC severity, whereas reversal of the inhibitory effects of IFN on gap junction communication restored enterocyte migration after IFN exposure. Strikingly, IFN−/− mice were protected from the development of NEC, and showed restored connexin 43 expression and intestinal restitution. Conclusions: IFN inhibits enterocyte migration by preventing interenterocyte gap junction communication. Connexin 43 loss may provide insights into the development of NEC, in which restitution is impaired. Necrotizing enterocolitis (NEC) is the leading cause of death from gastrointestinal disease in premature infants.1Blakely M.L. Lally K.P. McDonald S. Brown R.L. Barnhart D.C. Ricketts R.R. Thompson W.R. Scherer L.R. Klein M.D. Letton R.W. Chwals W.J. Touloukian R.J. Kurkchubasche A.G. Skinner M.A. Moss R.L. Hilfiker M.L. Postoperative outcomes of extremely low birth-weight infants with necrotizing enterocolitis or isolated intestinal perforation: a prospective cohort study by the NICHD Neonatal Research Network.Ann Surg. 2005; 241 (discussion 989–994): 984-989Crossref PubMed Scopus (236) Google Scholar, 2Henry M.C. Moss R.L. Current issues in the management of necrotizing enterocolitis.Semin Perinatol. 2004; 28: 221-233Abstract Full Text Full Text PDF PubMed Scopus (69) Google Scholar, 3Moss R.L. Dimmitt R.A. Barnhart D.C. Sylvester K.G. Brown R.L. Powell D.M. Islam S. Langer J.C. Sato T.T. Brandt M.L. Lee H. Blakely M.L. Lazar E.L. Hirschl R.B. Kenney B.D. Hackam D.J. Zelterman D. Silverman B.L. Laparotomy versus peritoneal drainage for necrotizing enterocolitis and perforation.N Engl J Med. 2006; 354: 2225-2234Crossref PubMed Scopus (321) Google Scholar NEC is characterized by a disruption in the intestinal barrier, which leads to the interaction with and subsequent activation of immune cells by pathogenic bacteria found in the intestinal lumen.4Hackam D.J. Upperman J.S. Grishin A. Ford H.R. Disordered enterocyte signaling and intestinal barrier dysfunction in the pathogenesis of necrotizing enterocolitis.Semin Pediatr Surg. 2005; 14: 49-57Abstract Full Text Full Text PDF PubMed Scopus (118) Google Scholar, 5Caplan M.S. Simon D. Jilling T. The role of PAF, TLR, and the inflammatory response in neonatal necrotizing enterocolitis.Semin Pediatr Surg. 2005; 14: 145-151Abstract Full Text Full Text PDF PubMed Scopus (114) Google Scholar The subsequent release of proinflammatory cytokines—including interferon-gamma (IFN)—leads to further intestinal injury and ultimately to the intestinal necrosis that characterizes NEC.6Ford H. Mechanism of nitric oxide-mediated intestinal barrier failure: insight into the pathogenesis of necrotizing enterocolitis.J Pediatr Surg. 2006; 41: 294-299Abstract Full Text Full Text PDF PubMed Scopus (45) Google Scholar, 7Anand R. Leaphart C.L. Mollen K. Hackam D. The role of the intestinal barrier in the pathogenesis of necrotizing enterocolitis.Shock. 2007; 27: 124-133Crossref PubMed Scopus (182) Google Scholar, 8Markel T. Crisostomo P. Wairiuko G. Pitcher J. Tsai B. Meldrum D. Cytokines in necrotizing enterocolitis.Shock. 2006; 25: 329-337Crossref PubMed Scopus (124) Google Scholar In response to epithelial injury, the process of healing is initiated when enterocytes from healthy areas migrate to sites of mucosal disruption to bridge the defects in the intestinal barrier. Studies from our laboratory have demonstrated that the process of intestinal healing—termed intestinal restitution—is significantly impaired in experimental NEC, which leads to persistent mucosal defects and progression of the disease.9Cetin S. Ford H.R. Sysko L.R. Agarwal C. Wang J. Neal M.D. Baty C. Apodaca G. Hackam D.J. Endotoxin inhibits intestinal epithelial restitution through activation of Rho-GTPase and increased focal adhesions.J Biol Chem. 2004; 279: 24592-24600Crossref PubMed Scopus (121) Google Scholar, 10Qureshi F.G. Leaphart C. Cetin S. Li J. Grishin A. Watkins S. Ford H.R. Hackam D.J. Increased expression and function of integrins in enterocytes by endotoxin impairs epithelial restitution.Gastroenterology. 2005; 128: 1012-1022Abstract Full Text Full Text PDF PubMed Scopus (80) Google Scholar An understanding of the mechanisms that govern intestinal restitution, both under basal conditions and after stimulation with cytokines such as IFN, remains of paramount importance in understanding the processes leading to the development of diseases like NEC. However, studies designed to investigate the regulation of enterocyte migration have traditionally been modeled after the migration properties of other motile cells, such as fibroblasts and leukocytes.11Dignass A.U. Podolsky D.K. Cytokine modulation of intestinal epithelial cell restitution: central role of transforming growth factor beta.Gastroenterology. 1993; 105: 1323-1332Abstract Full Text PDF PubMed Scopus (414) Google Scholar, 12Goke M. Zuk A. Podolsky D.K. Regulation and function of extracellular matrix intestinal epithelial restitution in vitro.Am J Physiol. 1996; 271: G729-G740PubMed Google Scholar This approach may be somewhat limited, because enterocytes do not exist as single cells, but rather exist as a monolayer in which cells are interconnected to form a functional syncytium. The factors that regulate the connectivity of enterocytes within this syncytium, the ability of cells to move as a unit during the process of intestinal restitution, and the migratory responsiveness of this syncytium to proinflammatory cytokines such as IFN remain largely unexplored. Functional communication between enterocytes could occur via gap junctions—intercellular channels that exist between adjacent cells where they allow the transfer of small molecules (under 1000 daltons) between adjoining cells. Each gap junction channel is comprised of a pair of hexameric arrays of individual subunits called connexins, of which the most widely expressed isoform is connexin-43 (Cx43).13Goodenough D.A. Bulk isolation of mouse hepatocyte gap junctions Characterization of the principal protein, connexin.J Cell Biol. 1974; 61: 557-563Crossref PubMed Scopus (143) Google Scholar, 14Goodenough D.A. The structure of cell membranes involved in intercellular communication.Am J Clin Pathol. 1975; 63: 636-645Crossref PubMed Scopus (18) Google Scholar The function of gap junctions is regulated in part through phosphorylation of the individual connexin molecules, which serves to regulate the localization of the channels at the plasma membrane as well as to regulate the channel through gating.15Laird D.W. Connexin phosphorylation as a regulatory event linked to gap junction internalization and degradation.Biochim Biophys Acta. 2005; 1711: 172-182Crossref PubMed Scopus (245) Google Scholar, 16Lampe P.D. TenBroek E.M. Burt J.M. Kurata W.E. Johnson R.G. Lau A.F. Phosphorylation of connexin43 on serine368 by protein kinase C regulates gap junctional communication.J Cell Biol. 2000; 149: 1503-1512Crossref PubMed Scopus (481) Google Scholar Recent evidence suggests a possible role for gap junctions in cell migration—in particular, in the migration of neural crest populations17Xu X. Francis R. Wei C. Linask K. Lo C. Connexin 43-mediated modulation of polarized cell movement and the directional migration of cardiac neural crest cells.Development. 2006; 133: 3629-3633Crossref PubMed Scopus (141) Google Scholar and endodermal cells.18Kwak B.R. Pepper M.S. Gros D.B. Meda P. Inhibition of endothelial wound repair by dominant negative connexin inhibitors.Mol Biol Cell. 2001; 12: 831-845Crossref PubMed Scopus (92) Google Scholar Although the localization and function of gap junctions—if any—in enterocytes remain largely unexplored, we now hypothesize that gap junctions could participate in the regulation of enterocyte migration and affect healing within the intestinal mucosa. We now provide evidence that gap junctions are expressed in enterocytes, that they mediate communication between adjacent enterocytes, and that they regulate the migration of enterocyte monolayers in vitro and in vivo. We further demonstrate that IFN-gamma significantly inhibits enterocyte migration through inhibitory effects on gap junction communication in the pathogenesis of NEC. Cultured intestinal cell lines (small intestine: IEC-6 and large intestine: HT-29, T84, and CaCO-2) were obtained from the American Type Culture Collection (Manassas, VA) and maintained as described.19Cohen M.B. Jensen N.J. Hawkins J.A. Mann E.A. Thompson M.R. Lentze M.J. Giannella R.A. Receptors for Escherichia coli heat stable enterotoxin in human intestine and in a human intestinal cell line (Caco-2).J Cell Physiol. 1993; 156: 138-144Crossref PubMed Scopus (51) Google Scholar, 20Hanski C. Stolze B. Riecken E.O. Tumorigenicity, mucin production and AM-3 epitope expression in clones selected from the HT-29 colon carcinoma cell line.Int J Cancer. 1992; 50: 924-929Crossref PubMed Scopus (17) Google Scholar, 21Quaroni A. Wands J. Trelstad R. Isselbacher K. Epithelioid cell cultures from rat small intestine Characterization by morphologic and immunologic criteria.J Cell Biol. 1979; 80: 248-265Crossref PubMed Scopus (699) Google Scholar Where indicated, cells were treated with IFN-gamma (IFN, Sigma-Aldrich Corp., St. Louis, MO, 1000 IU/mL for 1–14 hours, 37°C), or the following phosphatase inhibitors 20 minutes to 1 hour prior to IFN treatment: calyculin A (2 μmol/L), okadaic acid (100 nmol/L, Calbiochem, San Diego, CA), or fostriecin (1 μmol/L, Calbiochem). Enterocyte proliferation was measured in IEC-6 cells that had been plated to 60% confluence on glass coverslips using a colorimetric XTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] proliferation assay (Sigma-Aldrich, St. Louis, MO) with 5 mg/mL XTT, and absorbance measured at 450 nm as described22Mosmann T. Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays.J Immunol Methods. 1983; 65: 55-63Crossref PubMed Scopus (48661) Google Scholar according to the manufacturer's instructions. IEC-6 cells were transfected with 5 nmol/L Cx43 siRNA or control siRNA (Ambion, Austin, TX) with no known sequence homology, using Lipofectamine 2000 (Invitrogen, Carlsbad, CA) as a carrier, after first establishing inhibition of the housekeeper gene cyclophilin in IEC-6 cells following the manufacturer's instructions. To measure the kinetics of Cx43 transcription and translation, IEC-6 cells—some of which had been treated with IFN (14 hours, 1000 IU/mL)—were treated with cycloheximide (10 μmol/L, Sigma) or actinomycin (5 μmol/L, Sigma) for 1 to 6 hours. In parallel, cells were starved in cysteine- and methionine-free medium, radiolabeled using Tran35Steinhoff I. Leykauf K. Bleyl U. Durst M. Alonso A. Phosphorylation of the gap junction protein Connexin43 in CIN III lesions and cervical carcinomas.Cancer Lett. 2006; 235: 291-297Abstract Full Text Full Text PDF PubMed Scopus (14) Google ScholarS-label (50 UCi/mL; MP Biomedicals, Solon, OH) for 1 hour and chased in serum-free IEC-6 medium. At the indicated chase time, cells were rinsed with PBS and lysed, then after a brief centrifugation to remove nuclei, lysates were immunoprecipitated with Cx43 antibodies and antibody–antigen complexes were collected.23Mohan S. Bruns J.R. Weixel K.M. Edinger R.S. Bruns J.B. Kleyman T.R. Johnson J.P. Weisz O.A. Differential current decay profiles of epithelial sodium channel subunit combinations in polarized renal epithelial cells.J Biol Chem. 2004; 279: 32071-32078Crossref PubMed Scopus (26) Google Scholar After washing with radioimmune precipitation buffer (10 mmol/L Tris-HCl, pH 7.4, 0.15 mol/L NaCl, 1% Triton X-100, 1% deoxycholate, 1% Nonidet P-40, 0.1% sodium dodecyl sulfate [SDS]), samples were electrophoresed on 10% SDS-PAGE gels and analyzed using a Personal FX phosphorimager (Bio-Rad, Richmond, CA). Primary intestinal epithelial cells were isolated from the terminal ileum of freshly weaned C57BL/6 black mice immediately after sacrifice, using an adaptation of previous methods.24Liu S. Stolz D.B. Sappington P.L. Macias C.A. Killeen M.E. Tenhunen J.J. Delude R. Fink M.P. HMGB1 is secreted by immunostimulated enterocytes and contributes to cytomix-induced hyperpermeability of Caco-2 monolayers.Am J Physiol Cell Physiol. 2006; 290: C990-C999Crossref PubMed Scopus (133) Google Scholar, 25Whitehead R. Demmler K. Rockman S. Watson N. Clonogenic growth of epithelial cells from normal colonic mucosa from both mice and humans.Gastroenterology. 1999; 117: 858-865Abstract Full Text Full Text PDF PubMed Scopus (90) Google Scholar Briefly, after flushing the intestine with ice-cold Ca2+- and Mg2+-free Hanks' balanced salt solution (Invitrogen), tissue was cut into ∼1 mm × 1 mm pieces, decontaminated in 0.04% sodium hypochlorite (20 minutes on ice), soaked in 20 mL of Ca2+- and Mg2+-free Hanks' balanced salt solution containing 3 mmol/L EDTA and 1 mmol/L DTT (90 minutes on ice), then shaken vigorously to detach crypts, which were pelleted by centrifugation at 150 × g for 3 minutes at 4°C. The pellet was washed twice with ice-cold Ca2+- and Mg2+-free Hanks' balanced salt solution and resuspended in DMEM/F12 culture medium (BioWhittaker, Walkersville, MD), which was supplemented with 5% FBS, 100 U/mL penicillin G, 100 μg/mL streptomycin, 1× insulin-transferrin-selenium (Invitrogen), 20 mmol/L HEPES, pH 7.4, 0.25 μg/mL amphotericin B, 1 μg/mL fibronectin, and 1 μg/mL hydrocortisone. For confocal analysis, 25 mm-diameter glass cover slips (Fisherbrand, cat#12-545-102 25 CIR.-1) were coated with collagen I (Purecol, Inamed Biomaterials, Fremont, CA) and poly-L-lysine (Sigma) 1 hour before the cells were seeded in 6-well plates. The plates were seeded with 500 crypts/well and incubated at 37°C under a 5% CO2 atmosphere. In all cases, cellular viability was >95% by Trypan blue exclusion assay prior to use. After 4 to 5 days, cells were fixed with 4% PFA, immunostained with rabbit polyclonal anti-Cx43 (Zymed, South San Francisco, CA), and mouse monoclonal anti-E-Cadherin (Zymed) (see below). For SDS-PAGE, lysates were purified from cultured cells or mucosal scrapings obtained from fresh samples of terminal ileum that were obtained immediately after mice were killed (see animal model). Where indicated, specimens of distal small intestine obtained during the surgical management of patients with inflammatory bowel disease were obtained after deidentification. Specimens of 1-cm diameter from involved and uninvolved regions of the specimen were evaluated. After irrigation of the bowel to remove luminal contents, the mucosa was microdissected from the underlying connective tissue on the stage of an Olympus SZX7 microscope and placed in cold lysis buffer containing the protease inhibitors as previously described.26Cetin S. Dunkelberger J. Li J. Boyle P. Qureshi F. Ford H.R. Upperman J.S. Watkins S. Hackam D.J. Endotoxin differentially modulates the basolateral and apical sodium/proton exchangers (NHE) in enterocytes.Surgery. 2004; 136: 373-383Abstract Full Text Full Text PDF Scopus (35) Google Scholar Mucosal scrapings were subjected to SDS-PAGE using phospho-Cx43 specific and total connexin antibodies (Chemicon, Temecula, CA), and band density from radiographic film images were quantified using a GS700 Biorad densitometer and QuantityOne analysis software (Hercules, CA). For immunohistochemistry, cells or tissue were processed as described9Cetin S. Ford H.R. Sysko L.R. Agarwal C. Wang J. Neal M.D. Baty C. Apodaca G. Hackam D.J. Endotoxin inhibits intestinal epithelial restitution through activation of Rho-GTPase and increased focal adhesions.J Biol Chem. 2004; 279: 24592-24600Crossref PubMed Scopus (121) Google Scholar and fluorescent images were captured using an Olympus Fluoview1000 confocal microscope under a 60× oil immersion objective using standard filter sets. The nuclear stain DRAQ5 was purchased from AXXORA Platform (San Diego, CA). For immunoelectron microscopy, IEC-6 cells were fixed in cryofix (2% paraformaldehyde, 0.01% glutaraldehyde in 0.1 mol/L PBS) for 1 hour then cryoprotected in polyvinylpyrrolidone cryoprotectant (25% polyvinylpyrrolidone, 2.3 mol/L sucrose, 0.055 mol/L Na2CO3, pH 7.4) overnight at 4°C, as previously described.24Liu S. Stolz D.B. Sappington P.L. Macias C.A. Killeen M.E. Tenhunen J.J. Delude R. Fink M.P. HMGB1 is secreted by immunostimulated enterocytes and contributes to cytomix-induced hyperpermeability of Caco-2 monolayers.Am J Physiol Cell Physiol. 2006; 290: C990-C999Crossref PubMed Scopus (133) Google Scholar Blocks were situated on ultracryotome stubs that were cross-sectioned, frozen with liquid nitrogen, and stored in liquid nitrogen until use. Ultrathin sections (70 nm) were cut with the use of a Reichert Ultracut U ultramicrotome with an FC4S cryo-attachment, lifted on a small drop of 2.3 mol/L sucrose in PBS, and mounted on Formvar-coated copper grids. Sections were washed 3 times with PBS, then 3 times with PBG followed by a 30-minute blocking incubation with 5% normal goat serum in PBG. Sections were labeled with mouse anti-Cx43 for 1 hour. Sections were washed 4 times in PBG and labeled with goat antimouse antibody conjugated with 5 nm of colloidal gold (Amersham, Piscataway, NJ). Sections were washed 3 times in PBG, 3 times in PBS, and then fixed in 2.5% glutaraldehyde in PBS for 5 minutes. The sections were washed 2 times in PBS, and then washed 6 times in double-distilled H2O. Sections were poststained in 2% neutral uranyl acetate for 7 minutes, washed 3 times in double distilled H2O, stained for 2 minutes in 4% uranyl acetate, and then embedded in 1.25% methyl cellulose. Labeling was observed with a TEM (model JEM 1210, Jeol, Peabody, MA) at 80 kV. Gap junction communication was assessed using 2 parallel techniques: fluorescence recovery after photobleaching (FRAP), and single cell microinjection. This technique measures the movement of a fluorescent tracer through gap junctions into an area that has been previously photobleached using a laser, such that the rate and extent to which the photobleached cells fill with the fluorescent dye (termed the "fluorescence recovery") provide a direct measure of gap junction channel activity. Specifically, after exposure to IFN (1000 IU/mL, 14 hours), the gap junction inhibitors oleamide (10 μmol/L, 15 minutes) or 18α-glyceric acid (18αGA, 10 μmol/L), or after transfection with Cx43 siRNA or control siRNA, IEC-6 were loaded with the gap junction tracer calcein-AM (Molecular Probes, Carlsbad, CA, 20 μmol/L, 20 minutes). Upon entry into the cells, calcein-AM becomes lipid impermeable, leaving it free to diffuse between cells via gap junctions. Cells were thus placed upon the stage of an Olympus Fluoview F1000V inverted confocal microscope with a 40× oil objective (Center Valley, PA) and a small area was photobleached with a 20-second pulse from an argon laser. The recovery of fluorescent dye from adjacent cells to the photobleached area was then monitored using live cell confocal imaging over the next 5 minutes. Images were analyzed by measuring mean pixel density of cells in and surrounding the photobleached area using Fluoview Systems software. Recovery was calculated based on the fluorescence intensity in the photobleached area relative to the nonphotobleached surrounding area over time, as described.27Cushing P. Bhalla R. Johnson A.M. Rushlow W.J. Meakin S.O. Belliveau D.J. Nerve growth factor increases connexin43 phosphorylation and gap junctional intercellular communication.J Neurosci Res. 2005; 82: 788-801Crossref PubMed Scopus (21) Google Scholar In all cases, oleamide- or 18α-GA-treated cells served as controls. Microinjection allows the detection of the extent to which the 0.4-kilodalton fluorescent gap junction tracer Lucifer yellow passes from an injected cell to adjacent cells through gap junctions. IEC-6 cells were grown to approximately 75% confluence on Labteks cell chambers (Nalge Nunce International, Naperville, IL) in the presence or absence of IFN (1000 IU/mL, 14 hours) or oleamide (25 μmol/L, 15 minutes) then bathed in injection buffer (130 mmol/L NaCl, 4 mmol/L KCl, 1.8 mmol/L CaCl2, 0.56 mmol/L MgCl2, 10 mmol/L glucose, 1.2 mmol/L NaH2PO4, 14.3 mmol/L, HEPES, pH 7.4) and transferred to the stage of an inverted microscope (Nikon Diaphot TMD). Individual enterocytes were then loaded onto the stage of a IX81 Olympus microscope and visualized with a 60× oil objective, then microinjected using an InjectMan NI2 Femtojet injector to which the fluorescent tracers were loaded into individual Femtotips II (Eppendorf, Hamburg, Germany). Each cell was injected for 0.2 seconds and an injection pressure of 30 hpa with both the gap junction tracer 0.4-kilodalton Lucifer yellow (25 mg/mL) and the 70-kilodalton gap junction impermeant fluorescent compound Texas red dextran (Molecular Probes, Eugene, OR, 10 mg/mL), which serves as a control for nonspecific dye passage. Fluorescent emission was quantified using MetaMorph 6.1 (Universal Imaging Corp, Downington, PA) under standard filter sets. Cells were imaged prior to microinjection to assess for the presence of autofluorescence, which was subtracted from further analysis. The extent of dye coupling was determined by examining the level of transfer of Lucifer yellow to increasing tiers of neighboring cells. In all experiments, dye coupling was tested in a minimum of 10 microinjected cells.18Kwak B.R. Pepper M.S. Gros D.B. Meda P. Inhibition of endothelial wound repair by dominant negative connexin inhibitors.Mol Biol Cell. 2001; 12: 831-845Crossref PubMed Scopus (92) Google Scholar To measure enterocyte migration, IEC-6 cells were grown on glass coverslips to 100% confluence, scraped with a cell scraper, then transferred to the stage of an Olympus 1X71 inverted microscope (Melville, NY) and perfused with DMEM + 10 mmol/L HEPES pH 7.4 at 0.5 mL/h. Where indicated, cells were transfected with Cx43 siRNA or control siRNA that has no known targets. Images were taken every 5 minutes for 18 hours and analyzed using Metamorph software (Universal Imaging Corporation, Downingtown, PA). A calibration scale was obtained and the migration rate was determined by measuring the mean distance traveled by 15 cells per field over the course of the experiment. Measurements were obtained from cells that were selected both at the leading edge and several rows back. To induce NEC, the following experimental protocol was approved by the Animal Research and Care Committee of the Children's Hospital of Pittsburgh (protocol #0805). NEC was induced in 10-day-old mice (Swiss-Webster) or freshly weaned IFN−/− or wild-type littermates (Jackson Labs, Bar Harbor, ME) by the administration of enteral formula and the induction of hypoxia (5% oxygen for 1 minute prior to each feeding) twice daily28Nadler E.P. Dickinson E. Knisely A. Zhang X.R. Boyle P. Beer-Stolz D. Watkins S.C. Ford H.R. Expression of inducible nitric oxide synthase and interleukin-12 in experimental necrotizing enterocolitis.J Surg Res. 2000; 92: 71-77Abstract Full Text PDF PubMed Scopus (216) Google Scholar for 4 days. We and others have demonstrated that this experimental protocol induces intestinal inflammation in animals that resembles clinical NEC.28Nadler E.P. Dickinson E. Knisely A. Zhang X.R. Boyle P. Beer-Stolz D. Watkins S.C. Ford H.R. Expression of inducible nitric oxide synthase and interleukin-12 in experimental necrotizing enterocolitis.J Surg Res. 2000; 92: 71-77Abstract Full Text PDF PubMed Scopus (216) Google Scholar, 29Dvorak B. Halpern M.D. Holubec H. Dvorakova K. Dominguez J.A. Williams C.S. Meza Y.G. Rat milk decreases necrotizing enterocolitis in a rat model.Adv Exp Med Biol. 2004; 554: 471-473Crossref PubMed Scopus (5) Google Scholar Control animals remained with their mothers and received breast milk. Where indicated, animals were injected IP with oleamide (5 mg/kg), IFN (1000 IU/g), or saline daily for 4 days. To measure enterocyte migration, animals were injected with Bromodeoxyuridine (50 mg/kg 5′-BrdU; Sigma) intraperitoneally then sacrificed 14 hours later. Samples of terminal ileum were then immunostained using anti-BrDU antibodies as described.9Cetin S. Ford H.R. Sysko L.R. Agarwal C. Wang J. Neal M.D. Baty C. Apodaca G. Hackam D.J. Endotoxin inhibits intestinal epithelial restitution through activation of Rho-GTPase and increased focal adhesions.J Biol Chem. 2004; 279: 24592-24600Crossref PubMed Scopus (121) Google Scholar, 10Qureshi F.G. Leaphart C. Cetin S. Li J. Grishin A. Watkins S. Ford H.R. Hackam D.J. Increased expression and function of integrins in enterocytes by endotoxin impairs epithelial restitution.Gastroenterology. 2005; 128: 1012-1022Abstract Full Text Full Text PDF PubMed Scopus (80) Google Scholar Enterocyte migration was expressed by measuring the distance from the bottom of the crypt to the foremost labeled enterocyte and expressing the distance as a percentage of total villus height. In parallel experiments, mucosal scrapings were harvested from the terminal ileum to assess for protein expression by SDS-PAGE. Data presented are mean ± SEM, and comparisons are by 2-tailed Student's t test or analysis of variance, with statistical significance accepted for P < .05. We first sought to determine whether enterocytes express the gap junction protein Cx43 (Cx43), and whether gap junctions are capable of exchanging molecules between adjacent enterocytes. As shown in Figure 1, Cx43, which is the most widely expressed connexin isoform, was detected in several cultured enterocyte cell lines, including the small intestinal epithelial cell line IEC-6, the colonic epithelial lines CaCO-2, HT-29, and T84, and in mucosal scrapings from the terminal ilea of Swiss-Webster mice (Figure 1A). By immunofluorescence, Cx43 was detected at the interface between adjacent IEC-6 cells (arrows, Figure 1B). Cx43 was also found to be localized between primary enterocytes that were derived from the terminal ilea of mice and maintained in culture, indicating that the expression of Cx43 in enterocytes is not solely a property of cell lines (Figure 1C–E). The appearance of Cx43 on the small intestine of Swiss-Webster mice is shown in Figure 1F, where a higher magnification of the inset area corresponding to regions of crypts and villi is shown in Figure 1G and H, respectively, revealing the presence of Cx43 between enterocytes, as well as in a stromal compartment. Having shown that gap junctions are expressed between enterocytes in vitro and in vivo, we next sought to determine whether they could mediate interenterocyte