The Hydroxylase Inhibitor Dimethyloxalylglycine Is Protective in a Murine Model of Colitis

Background & Aims: Prolyl and asparaginyl hydroxylases are key oxygen-sensing enzymes that confer hypoxic sensitivity to transcriptional regulatory pathways including the hypoxia inducible factor 1 (HIF-1) and nuclear factor-κB (NF-κB). Knockout of either HIF-1 or (IKKβ-dependent) NF-κB pathways in intestinal epithelial cells promotes inflammatory disease in murine models of colitis. Both HIF-1 and NF-κB pathways are repressed by the action of hydroxylases through the hydroxylation of key regulatory molecules. Methods: In this study we have investigated the effects of the hydroxylase inhibitor dimethyloxalylglycine (DMOG) on Caco-2 intestinal epithelial cells in vitro and in a dextran sodium sulfate–induced model of murine colitis. Results: DMOG induces both HIF-1 and NF-κB activity in cultured intestinal epithelial cells, and is profoundly protective in dextran-sodium sulfate colitis in a manner that is at least in part reflected by the development of an anti-apoptotic phenotype in intestinal epithelial cells, which we propose reduces epithelial barrier dysfunction. Conclusions: These data show that hydroxylase inhibitors such as DMOG represent a new strategy for the treatment of inflammatory bowel disease. Background & Aims: Prolyl and asparaginyl hydroxylases are key oxygen-sensing enzymes that confer hypoxic sensitivity to transcriptional regulatory pathways including the hypoxia inducible factor 1 (HIF-1) and nuclear factor-κB (NF-κB). Knockout of either HIF-1 or (IKKβ-dependent) NF-κB pathways in intestinal epithelial cells promotes inflammatory disease in murine models of colitis. Both HIF-1 and NF-κB pathways are repressed by the action of hydroxylases through the hydroxylation of key regulatory molecules. Methods: In this study we have investigated the effects of the hydroxylase inhibitor dimethyloxalylglycine (DMOG) on Caco-2 intestinal epithelial cells in vitro and in a dextran sodium sulfate–induced model of murine colitis. Results: DMOG induces both HIF-1 and NF-κB activity in cultured intestinal epithelial cells, and is profoundly protective in dextran-sodium sulfate colitis in a manner that is at least in part reflected by the development of an anti-apoptotic phenotype in intestinal epithelial cells, which we propose reduces epithelial barrier dysfunction. Conclusions: These data show that hydroxylase inhibitors such as DMOG represent a new strategy for the treatment of inflammatory bowel disease. See editorial on page 346. See editorial on page 346. In a range of inflammatory diseases, chronically inflamed sites are hypoxic relative to normal healthy tissue.1Saadi S. Wrenshall L.E. Platt J.L. Regional manifestations and control of the immune system.FASEB J. 2002; 16: 849-856Crossref PubMed Scopus (100) Google Scholar In the gastrointestinal tract, the impact of hypoxia on transepithelial permeability has been implicated in inflammatory bowel disease (IBD), which comprises Crohn’s disease (CD) and ulcerative colitis (UC).2Taylor C.T. Dzus A.L. Colgan S.P. Autocrine regulation of epithelial permeability by hypoxia: role for polarized release of tumor necrosis factor alpha.Gastroenterology. 1998; 114: 657-668Abstract Full Text Full Text PDF PubMed Scopus (175) Google Scholar The development of new therapeutic strategies for IBD are focused on addressing the compromised epithelial barrier and resultant inflammation.3Sartor R.B. Mechanisms of disease: pathogenesis of Crohn’s disease and ulcerative colitis.Nat Clin Pract Gastroenterol Hepatol. 2006; 3: 390-407Crossref PubMed Scopus (1349) Google Scholar, 4Strober W. Fuss I. Mannon P. The fundamental basis of inflammatory bowel disease.J Clin Invest. 2007; 117: 514-521Crossref PubMed Scopus (1046) Google Scholar Thus, manipulation of the regulatory mechanisms that determine the hypoxic response in the gut has therapeutic potential in the treatment of IBD. The best understood of the transcriptional pathways underlying the adaptive response to hypoxia is that of the hypoxia inducible factor (HIF). The oxygen sensitivity of the HIF pathway is conferred by a family of hydroxylases including at least 3 prolyl hydroxylases (PHDs) and 1 asparagine hydroxylase (FIH) (factor inhibiting HIF).5Schofield C.J. Ratcliffe P.J. Signalling hypoxia by HIF hydroxylases.Biochem Biophys Res Commun. 2005; 338: 617-626Crossref PubMed Scopus (284) Google Scholar Under steady-state conditions in which cellular adenosine triphosphate demand is matched by mitochondrial generation and sufficient surplus oxygen is available, hydroxylases suppress the activity of HIF-1 through hydroxylation of Pro402 and Pro564 of HIF-1α, leading to von Hipple Lindau protein-dependent ubiquitination and subsequent proteasomal degradation by the 26S proteasome.6Kaelin W.G. Proline hydroxylation and gene expression.Annu Rev Biochem. 2005; 74: 115-128Crossref PubMed Scopus (370) Google Scholar A further hydroxylation of asparagine 803 by factor inhibiting HIF prevents HIF transcriptional activity through inhibition of interaction with the transcriptional co-activator cyclic AMP response element binding protein-binding protein.7Lando D. Peet D.J. Whelan D.A. et al.Asparagine hydroxylation of the HIF transactivation domain: a hypoxic switch.Science. 2002; 295: 858-861Crossref PubMed Scopus (1280) Google Scholar Thus, hydroxylases repress both the expression and transcriptional activity of HIF-1. In hypoxia hydroxylation is reversed, leading to HIF stabilization and transactivation with the subsequent induction of an adaptive response. We and others recently have shown that hydroxylase activity is not confined to the regulation of HIF pathways.8Cummins E.P. Berra E. Comerford K.M. et al.Prolyl hydroxylase-1 negatively regulates IkappaB kinase-beta, giving insight into hypoxia-induced NFkappaB activity.Proc Natl Acad Sci U S A. 2006; 103: 18154-18159Crossref PubMed Scopus (604) Google Scholar, 9Cockman M.E. Lancaster D.E. Stolze I.P. et al.Posttranslational hydroxylation of ankyrin repeats in IkappaB proteins by the hypoxia-inducible factor (HIF) asparaginyl hydroxylase, factor inhibiting HIF (FIH).Proc Natl Acad Sci U S A. 2006; 103: 14767-14772Crossref PubMed Scopus (228) Google Scholar Hydroxylases also modulate nuclear factor-κB (NF-κB) activity, likely through hydroxylation of IKKβ and ankyrin-repeat–containing proteins such as p105 and IKKβ by prolyl and asparaginyl hydroxylases, respectively. Although it is not currently known what the functional consequences of asparagine hydroxylation of ankyrin-repeat domains of proteins such as p105 and IκBα are,9Cockman M.E. Lancaster D.E. Stolze I.P. et al.Posttranslational hydroxylation of ankyrin repeats in IkappaB proteins by the hypoxia-inducible factor (HIF) asparaginyl hydroxylase, factor inhibiting HIF (FIH).Proc Natl Acad Sci U S A. 2006; 103: 14767-14772Crossref PubMed Scopus (228) Google Scholar it appears that proline hydroxylation of IKKβ leads to repression of NF-κB signaling.8Cummins E.P. Berra E. Comerford K.M. et al.Prolyl hydroxylase-1 negatively regulates IkappaB kinase-beta, giving insight into hypoxia-induced NFkappaB activity.Proc Natl Acad Sci U S A. 2006; 103: 18154-18159Crossref PubMed Scopus (604) Google Scholar In response to hypoxia, the hydroxylase-dependent repression of NF-κB activity is reversed, leading to a mild induction of basal NF-κB activity as well as increased sensitivity of this pathway to activation by inflammatory mediators such as tumor necrosis factor α (TNFα). Protective roles for both the HIF-1 and NF-κB pathways in intestinal epithelial cells have been shown recently using conditional intestinal epithelial cell knockout mice for HIF-1α and IKKβ, respectively.10Karhausen J. Furuta G.T. Tomaszewski J.E. et al.Epithelial hypoxia-inducible factor-1 is protective in murine experimental colitis.J Clin Invest. 2004; 114: 1098-1106Crossref PubMed Scopus (502) Google Scholar, 11Greten F.R. Eckmann L. Greten T.F. et al.IKKbeta links inflammation and tumorigenesis in a mouse model of colitis-associated cancer.Cell. 2004; 118: 285-296Abstract Full Text Full Text PDF PubMed Scopus (2122) Google Scholar, 12Chen L.W. Egan L. Li Z.W. et al.The two faces of IKK and NF-kappaB inhibition: prevention of systemic inflammation but increased local injury following intestinal ischemia-reperfusion.Nat Med. 2003; 9: 575-581Crossref PubMed Scopus (437) Google Scholar Conditional intestinal epithelial HIF-1α knockout mice have increased susceptibility in a hapten-induced murine model of colitis. This was proposed to be owing to the loss of expression of HIF-1–dependent genes that are essential for the protective barrier function.10Karhausen J. Furuta G.T. Tomaszewski J.E. et al.Epithelial hypoxia-inducible factor-1 is protective in murine experimental colitis.J Clin Invest. 2004; 114: 1098-1106Crossref PubMed Scopus (502) Google Scholar Furthermore, IKKβ conditional intestinal epithelial cell knockout mice that lack a functional NF-κB pathway show increased susceptibility to a number of inflammatory stimuli. The mechanism(s) underlying the protective effects of intestinal epithelial IKKβ is possibly through the inhibition of epithelial apoptosis through enhanced NF-κB–dependent expression of anti-apoptotic genes that confer enhanced intestinal epithelial barrier function.11Greten F.R. Eckmann L. Greten T.F. et al.IKKbeta links inflammation and tumorigenesis in a mouse model of colitis-associated cancer.Cell. 2004; 118: 285-296Abstract Full Text Full Text PDF PubMed Scopus (2122) Google Scholar, 13Egan L.J. Eckmann L. Greten F.R. et al.IkappaB-kinase beta-dependent NF-kappaB activation provides radioprotection to the intestinal epithelium.Proc Natl Acad Sci U S A. 2004; 101: 2452-2457Crossref PubMed Scopus (176) Google Scholar Alternatively, a recent study has proposed that the protective effects of epithelial IKKβ in the context of Trichuris muris infection is through tissue-specific conditioning of dendritic cells to limit proinflammatory cytokines in the intestinal mucosa.14Zaph C. Troy A.E. Taylor B.C. et al.Epithelial-cell-intrinsic IKK-beta expression regulates intestinal immune homeostasis.Nature. 2007; 446: 552-556Crossref PubMed Scopus (436) Google Scholar Furthermore, mice with NF-κB essential modulator protein (IKKγ) ablated in intestinal epithelial cells have compromised intestinal immune homeostasis and develop an IBD-like phenotype.15Nenci A. Becker C. Wullaert A. et al.Epithelial NEMO links innate immunity to chronic intestinal inflammation.Nature. 2007; 446: 557-561Crossref PubMed Scopus (856) Google Scholar Intestinal epithelial cells are among the most short-lived cells of the body. After their generation from crypt stem cells, epithelial cells migrate up the mucosal crypt and villus and are programmed from an early stage for cell death through tightly regulated apoptotic pathways.16Edelblum K.L. Yan F. Yamaoka T. et al.Regulation of apoptosis during homeostasis and disease in the intestinal epithelium.Inflamm Bowel Dis. 2006; 12: 413-424Crossref PubMed Scopus (139) Google Scholar The process of intestinal epithelial apoptosis is tightly regulated. Excessive epithelial apoptosis leads to a loss of the protective intestinal epithelial barrier, with consequential exposure of the intestinal immune system to nonspecific luminal antigens.16Edelblum K.L. Yan F. Yamaoka T. et al.Regulation of apoptosis during homeostasis and disease in the intestinal epithelium.Inflamm Bowel Dis. 2006; 12: 413-424Crossref PubMed Scopus (139) Google Scholar A breakdown in the integrity of the intestinal barrier is a critical event in the initiation and development of IBD.4Strober W. Fuss I. Mannon P. The fundamental basis of inflammatory bowel disease.J Clin Invest. 2007; 117: 514-521Crossref PubMed Scopus (1046) Google Scholar In this study we postulated that as the expression of both HIF-1 and NF-κB are implicated in preventing intestinal mucosal insult, the use of hydroxylase inhibition to evoke their transcriptional activity may have protective effects in colitis. We used the dextran-sodium sulfate (DSS)-induced model of colitis in mice, which has many of the characteristics of human UC.17Okayasu I. Hatakeyama S. Yamada M. et al.A novel method in the induction of reliable experimental acute and chronic ulcerative colitis in mice.Gastroenterology. 1990; 98: 694-702Abstract PubMed Google Scholar In this model, DSS damages colon epithelial cells, thereby compromising the intestinal barrier and allowing intestinal bacteria to penetrate the injured mucosa and perpetuate colonic inflammation. Dimethyloxalylglycine (DMOG) was used as a cell-permeable inhibitor of both proline and asparaginyl hydroxylases.18Elvidge G.P. Glenny L. Appelhoff R.J. et al.Concordant regulation of gene expression by hypoxia and 2-oxoglutarate-dependent dioxygenase inhibition: the role of HIF-1alpha, HIF-2alpha, and other pathways.J Biol Chem. 2006; 281: 15215-15226Crossref PubMed Scopus (367) Google Scholar DMOG has been shown previously to activate both HIF-1 and NF-κB–dependent gene expression in vitro, likely through the reversal of hydroxylation-dependent repression.8Cummins E.P. Berra E. Comerford K.M. et al.Prolyl hydroxylase-1 negatively regulates IkappaB kinase-beta, giving insight into hypoxia-induced NFkappaB activity.Proc Natl Acad Sci U S A. 2006; 103: 18154-18159Crossref PubMed Scopus (604) Google Scholar When used in vivo, DMOG has been shown to induce expression of HIF-1α protein in ischemic skeletal muscle.19Milkiewicz M. Pugh C.W. Egginton S. Inhibition of endogenous HIF inactivation induces angiogenesis in ischaemic skeletal muscles of mice.J Physiol. 2004; 560: 21-26Crossref PubMed Scopus (122) Google Scholar Here, we report the protective effects of hydroxylase inhibition by DMOG in a murine model of colitis. Female C57BL/6 strain mice were bred under specific pathogen-free conditions at the BioResources Unit (Trinity College, Dublin, Ireland). Mice were fed an irradiated diet and housed on irradiated bedding and kept in individually ventilated and filtered cages (Tecniplast, Northants, UK). All animal experiments were performed in compliance with Irish Department of Health and Children regulations and approved by the Trinity College Bioresources ethical review board. DMOG (Caymen Chemicals, Ann Arbor, MI) was prepared in endotoxin-free phosphate-buffered saline (PBS) (Sigma, St. Louis, MO). Mice were treated with 8 mg of DMOG injected intraperitoneally every second day, a regimen previously shown to cause HIF-1 activation in vivo.19Milkiewicz M. Pugh C.W. Egginton S. Inhibition of endogenous HIF inactivation induces angiogenesis in ischaemic skeletal muscles of mice.J Physiol. 2004; 560: 21-26Crossref PubMed Scopus (122) Google Scholar Control mice received PBS. Mice were treated with vehicle (PBS) or DMOG for 2 days. Colons were excised and epithelial cells were scraped and snap-frozen in liquid nitrogen. To prepare lysates the colonic scrapings were thawed on ice and were immersed in 200 μL of modified RIPA buffer supplemented with Protease inhibitor cocktail (Sigma) + β-glycerophosphate (0.5 mmol/L) + NaF (1 mmol/L) + sodium orthovanadate (1 mmol/L). Samples were disrupted on ice with pipette tips before a series of 3 sonication pulses of 10 seconds at increasing intensity on ice. Samples were triterated again as before. Tissue debris was removed by centrifugation at 12,000 rpm at 4°C for 6 minutes × 2. The supernatant was retained each time and protein determination was performed (BioRad, Hercules, CA). Samples were diluted in reducing sample buffer and examined by Western blot. Nuclear lysates, whole-cell lysates, whole-colon homogenates, and lysates from colonic scrapings were generated and separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis, transferred to nitrocellulose membranes, and immunoblotted as described previously8Cummins E.P. Berra E. Comerford K.M. et al.Prolyl hydroxylase-1 negatively regulates IkappaB kinase-beta, giving insight into hypoxia-induced NFkappaB activity.Proc Natl Acad Sci U S A. 2006; 103: 18154-18159Crossref PubMed Scopus (604) Google Scholar using the following specific antibodies: p65 NF-κB (Santa Cruz, Santa Cruz, CA), HIF-1α (BD Transduction Labs, Franklin Lakes, NJ; in vitro experiments), HIF-1α (Calbiochem, Nottingham, UK; H1α67, mouse tissue extracts), P-IKK α/β (Cell Signaling, Danvers, MA), TATA box binding protein (Abcam, Cambridge, UK), cyclooxygenase-2 (Santa Cruz), Bcl-XL (BD Pharmingen, Franklin Lakes, NJ), cellular inhibitor of apoptosis protein 2 (cIAP-2; Santa Cruz), and β-actin (Sigma). Densitometric analysis was performed using the Scion Image (Scion Image, Frederick, MD) for Windows Software. Arbitrary values for experimental samples were expressed relative to values obtained for loading controls (β-actin or TATA box binding protein). These data are expressed as raw arbitrary values or as a percentage of the control (normoxia or PBS vehicle). Figures represent the mean of several experiments ± SD (in vitro experiments) or mean of several samples ± SEM (in vivo experiments). Colons were homogenized using methods adapted from processing lung tissue as described above. Protein normalization was performed (BioRad). Whole-colon homogenate was incubated with an immobilized oligonucleotide on a 96-well plate (containing a specific NF-κB binding site) and detected by a specific p65 NF-κB subunit enzyme-linked immunosorbent assay. Colitis was induced by DSS treatment of mice.17Okayasu I. Hatakeyama S. Yamada M. et al.A novel method in the induction of reliable experimental acute and chronic ulcerative colitis in mice.Gastroenterology. 1990; 98: 694-702Abstract PubMed Google Scholar, 33Smith P. Mangan N.E. Walsh C.M. et al.Infection with a Helminth parasite prevents experimental colitis via a macrophage-mediated mechanism.J Immunol. 2007; 178: 4557-4566Crossref PubMed Scopus (237) Google Scholar Acute disease was induced by treatment of mice with 2.5% DSS (MP Biomedicals, Solon, OH) dissolved in the drinking water for 5 days. For studies on recovery from colitis mice were treated with 3% DSS for 5 days followed by 3 days on normal drinking water.34Melgar S. Karlsson A. Michaelsson E. Acute colitis induced by dextran sulfate sodium progresses to chronicity in C57BL/6 but not in BALB/c mice: correlation between symptoms and inflammation.Am J Physiol. 2005; 288: G1328-G1338Google Scholar Body weight, occult blood in feces, and stool consistency/diarrhea were recorded daily for each mouse to determine the disease activity index (DAI).33Smith P. Mangan N.E. Walsh C.M. et al.Infection with a Helminth parasite prevents experimental colitis via a macrophage-mediated mechanism.J Immunol. 2007; 178: 4557-4566Crossref PubMed Scopus (237) Google Scholar, 35Cooper H.S. Murthy S.N. Shah R.S. et al.Clinicopathologic study of dextran sulfate sodium experimental murine colitis.Lab Invest. 1993; 69: 238-249PubMed Google Scholar The maximum DAI score was 12 based on assigning a scoring system of 1–4 for each parameter: score of 0, no weight loss, normal stool, no blood; score of 1, 1%–3% weight loss; score of 2, 3%–6% weight loss, loose stool (a loose stool was defined as the formation of a stool that readily becomes paste on handling), blood visible in stool; score of 3, 6%–9% weight loss; and score of 4, less than 9% weight loss, diarrhea, and gross bleeding. Blood in the feces of individual mice was detected using Hemdetect kits, as described.33Smith P. Mangan N.E. Walsh C.M. et al.Infection with a Helminth parasite prevents experimental colitis via a macrophage-mediated mechanism.J Immunol. 2007; 178: 4557-4566Crossref PubMed Scopus (237) Google Scholar When mice were sacrificed the lengths of the colon were measured. When mice were sacrificed an approximately 1-cm length of the distal colon was removed and fixed in 10% formaldehyde-saline. All histology sections were examined in a blinded fashion independently by 2 observers. Arbitrary histologic scoring was used to quantify colon damage.35Cooper H.S. Murthy S.N. Shah R.S. et al.Clinicopathologic study of dextran sulfate sodium experimental murine colitis.Lab Invest. 1993; 69: 238-249PubMed Google Scholar A maximum combined score of 10 was determined from the severity of inflammatory cell infiltration (score of 0, none; score of 1, slightly dispersed cell infiltrate; score of 2, moderately increased cell infiltrates forming occasional cell foci; and score of 3, severely large areas of cell infiltrates causing loss of tissue architecture), extent of injury (score of 0, none; score of 1, mucosal; score of 2, mucosal and submucosal; and score of 3, transmural), and crypt damage (score of 0, none; score of 1, basal one third damaged; score of 2, basal two thirds damaged; score of 3, only surface epithelium intact; and score of 4, loss of entire crypt and epithelium). Colons were homogenized using methods adapted from processing lung tissue.36Mangan N.E. van Rooijen N. McKenzie A.N. et al.Helminth-modified pulmonary immune response protects mice from allergen-induced airway hyperresponsiveness.J Immunol. 2006; 176: 138-147PubMed Google Scholar Levels of interleukin (IL)-1β, IL-6, TNF-α, and IL-12(p70) were detected in colon extracts using an enzyme-linked immunosorbent assay.37Mangan N.E. Fallon R.E. Smith P. et al.Helminth infection protects mice from anaphylaxis via IL-10-producing B cells.J Immunol. 2004; 173: 6346-6356PubMed Google Scholar Myeloperoxidase (MPO) activity was detected using o-phenylediamine dihydrochloride as substrate and data interpolated from an MPO standard curve (Sigma). Levels of cytokines and MPO were expressed as pg/mg or U/mg, respectively, relative to colon protein. Apoptosis of colon epithelial cells was measured by terminal deoxynucleotidyl transferase–mediated deoxyuridine triphosphate nick-end labeling assay on histology sections using the In situ cell death detection kit (Roche Diagnostic, Basal, Switzerland), according to the manufacturer’s instructions. Colon epithelial cell apoptosis was quantified as described,38Fukata M. Chen A. Klepper A. et al.Cox-2 is regulated by Toll-like receptor-4 (TLR4) signaling: role in proliferation and apoptosis in the intestine.Gastroenterology. 2006; 131: 862-877Abstract Full Text Full Text PDF PubMed Scopus (361) Google Scholar with terminal deoxynucleotidyl transferase–mediated deoxyuridine triphosphate nick-end labeling–positive cells checked for morphologic changes consistent with apoptosis. For each mouse the number of apoptotic colon epithelial cells was counted in 200 epithelial cells on 3 fields per section (600 cells checked per mouse). The percentage of terminal deoxynucleotidyl transferase–mediated deoxyuridine triphosphate nick-end labeling–positive epithelial cells was determined. All slides were coded and scored in a blinded fashion. Quantification of apoptosis in T84 cells was performed by fluorescence-activated cell sorter analysis of nuclear propidium iodide fluorescence. Cells (1 × 105) were seeded on 12-well plates and grown to 70% confluence and pretreated with DMOG (1 mmol/L) or dimethyl sulfoxide for 24 hours. Cycloheximide (10 μmol/L) and TNF-related apoptosis-inducing ligand (0–100 ng/mL) then were added for an additional 24 hours. Cells were trypsinized, centrifuged (1100 rpm, 5 min), resuspended in 400 μL hypotonic fluorochrome solution (200 mL PBS, 10 mg propidium iodide, 3.4 mmol/L sodium citrate, 1 mmol/L Tris, 0.1 mmol/L ethylenediaminetetraacetic acid, and 0.1% Triton X-100), and placed on ice for 10 minutes before analysis using a fluorocytometer (Epics XL-mcl; Beckman-Coulter, Fullerton, CA). Apoptosis was quantified as the percentage of cells in the sub-G0 phase. At least 10,000 events were collected per sample and analyzed. All data were tested for normality using GraphPad Instat software (GraphPad, San Diego, CA). The Student t test or the Welch t test was used throughout for statistical comparisons using Graphpad Instat (GraphPad). P values of less than .05 were considered significant. Initially, we investigated the presence of a functionally intact, oxygen-sensing system in cultured enterocytes. Consistent with reports in other cell types, exposure of Caco-2 (colon-derived epithelial) cells to hypoxia resulted in activation of the NF-κB pathway as reflected by the nuclear accumulation of p65 (Figure 1A) and phosphorylation of IKKα/β (Figure 1B). These cells also showed robust activation of the HIF-1 pathway in response to hypoxia (Figure 1C, left). Treatment of the Caco-2 cells with DMOG also resulted in stabilization of HIF-1a (Figure 1C, right) and activation of IKKa/b (Figure 1D). Polymerase chain reaction analysis of mouse colonic tissues showed the presence of all 3 prolyl hydroxylase isoforms (data not shown). DMOG treatment of mice for 48 hours resulted in detectable stabilized HIF-1α in epithelial-enriched colonic mucosal scrapings (Figure 1E), indicating the successful inhibition of hydroxylase activity in those tissues. Furthermore, whole-tissue lysates from DMOG-treated animals (5 days) also showed increased NF-κB activity as determined by TransAm assay for active p65 (Figure 1F). Thus, DMOG activates both NF-κB and HIF-1 signaling in cultured intestinal epithelial cells and in mouse colon. We next investigated the impact of DMOG on intestinal pathophysiology in vivo. Mice were treated with either DMOG or PBS (vehicle control) and exposed to 2.5% DSS in drinking water over a period of 5 days to induce colitis. Within 3 days, control mice developed progressive weight loss and increased composite DAI, accounted for by the development of diarrhea and bloody feces (Figure 2A and B). In contrast, DMOG treatment rendered mice refractory to both DSS-induced weight loss and DSS-induced increases in DAI. Both weight loss and DAI were reduced significantly in DSS-DMOG–treated mice relative to DSS-PBS–treated mice (P < .01–.001; Figure 2A and B). Characteristic colon shortening in DSS-treated animals was attenuated significantly in DMOG-treated mice (P < .0005; Figure 2C and D). Furthermore, the gross appearance of colons from PBS-treated mice exposed to DSS shows the absence of formation of intact fecal pellets, whereas DMOG treatment attenuated this effect with colons forming normal fecal pellets (arrows; Figure 2C). Thus, DMOG treatment profoundly attenuated the severity of disease in DSS-induced colitis. Histologic examination of colons from mice treated with DMOG alone showed no overt alterations in the distal colon (Figure 2E), which was confirmed by blinded scoring of colon sections (Figure 2F). DSS treatment caused extensive colonic damage with epithelial loss and collapse of crypt structures accompanied by prominent edema and inflammatory cell infiltration (Figure 2E). In contrast, DMOG-treated mice had a dramatically reduced severity of DSS-induced colon injury. Crypt architecture again was observed and ulceration was not a feature. Furthermore, lower magnitude edema and inflammatory cell infiltrate also were noted (Figure 2E). Scoring the histopathologic changes in the colon illustrated the significantly reduced damage in the colon of DMOG-treated mice relative to control PBS-treated animals (P < .0001; Figure 2F). We next investigated the impact of DMOG treatment on the expression of markers of colonic inflammation that are increased in mice exposed to DSS. In control mice DSS induced a significant increase in MPO activity, a marker for inflammation and leukocyte infiltration (P < .005; Figure 3A). In contrast, exposure of DMOG-treated mice to DSS did not increase colon MPO levels (Figure 3A). Furthermore, although colonic levels of the proinflammatory cytokines IL-1β, TNF-α, IL-6, and IL-12 were increased significantly in mice exposed to DSS (P < .05–.001), co-administration of DMOG prevented the changes in TNF-α, IL-6, and IL-12, with a significant increase only in IL-1β (P < .05) (Figure 3B). Furthermore, the DSS-induced increases in MPO, IL-1β, TNF-α, IL-6, and IL-12 all were diminished significantly in DMOG-treated animals (P < .05). Treatment of mice with DMOG alone had no effects on basal MPO, IL-1β, TNF-α, IL-6, and IL-12 levels in the colons of control mice (Figure 3). To determine whether DMOG treatment is solely effective in the prevention of DSS-induced colitis or whether it may be useful for the promotion of recovery, we used a higher dose of DSS (3%) to induce more severe disease over 5 days, followed by a 3-day period of recovery returning mice to normal drinking water. Three percent DSS for 5 days induced a more severe colitis in control mice (Figure 4A; maximum DAI score is 12) when compared with 2.5% DSS (Figure 2), DMOG treatment again significantly attenuated disease (Figure 4A and supplementary Figure 1; supplementary material online at www.gastrojournal.org). Furthermore, despite the extent of gross clinical scoring in DMOG-treated mice being nonsignificantly different from control mice at day 6, these mice cotreated with DSS and DMOG showed significant signs of recovery with reduced DAI values within 48 hours of removing DSS (P < .05; Figure 4A). The significant reduction in DAI of DMOG-treated mice during the recovery phase was the result of reduction in the DAI scores for fecal blood, with no blood detected in the feces of DMOG-treated animals by day 8, whereas in the PBS-treated group 5 of 6 mice had blood in the feces (Figure 4B).Supplementary Figure 1Alterations in weight and colon length in DMOG- and DSS-treated mice. Mice were exposed to 3% DMOG for 5 days to induce severe pathology and returned to normal drinking water for 3 more days. (A) Animal weight and (B) colon length were measured. (C) Twenty-four–hour DMOG treatment activates HIF in mouse intestinal tissue as shown by Western blotting of 4 independent animals. Black triangle, PBS; red triangle, DMOG; black circle, DSS-PBS; red circle, DSS-DMOG.View