High Mobility Group Box 1 Promotes Small Intestinal Damage Induced by Nonsteroidal Anti-Inflammatory Drugs through Toll-Like Receptor 4

Release of high mobility group box 1 (HMGB1) from damaged cells, which is involved in many types of tissue injuries, activates inflammatory pathways by stimulating multiple receptors, including Toll-like receptor 2 (TLR2), TLR4, and receptor for advanced glycation end-products (RAGE). Our objective was to determine the role of HMGB1 in nonsteroidal anti-inflammatory drug (NSAID)-induced damage of the small intestine. Oral indomethacin (10 mg/kg) induced damage to the small intestine and was associated with increases in intestinal HMGB1 expression and serum HMGB1 levels. In wild-type mice, recombinant human HMGB1 aggravated indomethacin-induced small intestinal damage; enhanced the mRNA expression levels of tumor necrosis factor α (TNF-α), monocyte chemotactic protein 1, and KC; activated nuclear factor kappa B; and stimulated phosphorylation of the mitogen-activated protein kinases p38, extracellular signal-regulated kinase (ERK), and c-Jun N-terminal kinase (JNK). In contrast, blocking HMGB1 action with neutralizing antibodies prevented damage and inhibited both inflammatory cytokine overexpression and activation of these intracellular signaling pathways. TLR2-knockout (KO) and RAGE-KO mice exhibited high sensitivities to indomethacin-induced damage, similar to wild-type mice, whereas TLR4-KO mice exhibited less severe intestinal damage and lower levels of TNF-α mRNA expression. Exogenous HMGB1 aggravated the damage in TLR2- and RAGE-KO mice but did not affect the damage in TLR4-KO mice. Thus, our results suggest that HMGB1 promotes NSAID-induced small intestinal damage through TLR4-dependent signaling pathways. Release of high mobility group box 1 (HMGB1) from damaged cells, which is involved in many types of tissue injuries, activates inflammatory pathways by stimulating multiple receptors, including Toll-like receptor 2 (TLR2), TLR4, and receptor for advanced glycation end-products (RAGE). Our objective was to determine the role of HMGB1 in nonsteroidal anti-inflammatory drug (NSAID)-induced damage of the small intestine. Oral indomethacin (10 mg/kg) induced damage to the small intestine and was associated with increases in intestinal HMGB1 expression and serum HMGB1 levels. In wild-type mice, recombinant human HMGB1 aggravated indomethacin-induced small intestinal damage; enhanced the mRNA expression levels of tumor necrosis factor α (TNF-α), monocyte chemotactic protein 1, and KC; activated nuclear factor kappa B; and stimulated phosphorylation of the mitogen-activated protein kinases p38, extracellular signal-regulated kinase (ERK), and c-Jun N-terminal kinase (JNK). In contrast, blocking HMGB1 action with neutralizing antibodies prevented damage and inhibited both inflammatory cytokine overexpression and activation of these intracellular signaling pathways. TLR2-knockout (KO) and RAGE-KO mice exhibited high sensitivities to indomethacin-induced damage, similar to wild-type mice, whereas TLR4-KO mice exhibited less severe intestinal damage and lower levels of TNF-α mRNA expression. Exogenous HMGB1 aggravated the damage in TLR2- and RAGE-KO mice but did not affect the damage in TLR4-KO mice. Thus, our results suggest that HMGB1 promotes NSAID-induced small intestinal damage through TLR4-dependent signaling pathways. Patients with chronic arthritic conditions, such as rheumatoid arthritis and osteoarthritis, take nonsteroidal anti-inflammatory drugs (NSAIDs) for long periods of time. Although these drugs are effective in reducing joint pain, stiffness, and swelling, their use is associated with a broad spectrum of adverse reactions in the liver, kidney, cardiovascular system, and gastrointestinal (GI) tract. Although it has been believed that the gastroduodenal mucosa is most susceptible to NSAID toxicology in the GI tract, recent advances in small intestinal endoscopy, such as video capsule endoscopy and double-balloon endoscopy, have revealed that NSAIDs have strong ulcerogenic effects in the small intestine. Graham et al1Graham D.Y. Opekun A.R. Willingham F.F. Qureshi W.A. Visible small-intestinal mucosal injury in chronic NSAID users.Clin Gastroenterol Hepatol. 2005; 3: 55-59Abstract Full Text Full Text PDF PubMed Scopus (489) Google Scholar reported that small bowel injury was observed in 15 of 21 chronic NSAID users (71%) by video capsule endoscopy. In addition, Sugimori et al2Sugimori S. Watanabe T. Tabuchi M. Kameda N. Machida H. Okazaki H. Tanigawa T. Yamagami H. Shiba M. Watanabe K. Tominaga K. Fujiwara Y. Oshitani N. Koike T. Higuchi K. Arakawa T. Evaluation of small bowel injury in patients with rheumatoid arthritis by capsule endoscopy: effects of anti-rheumatoid arthritis drugs.Digestion. 2008; 78: 208-213Crossref PubMed Scopus (49) Google Scholar detected mucosal breaks, such as ulcers and erosions, in the small intestine in 13 of 16 (81.3%) rheumatoid arthritis patients who took a NSAID for more than 1 year. Because acid suppressants, such as proton pump inhibitors and H2 receptor antagonists, are effective for treating NSAID-induced injuries in the upper GI tract but do not prevent those injuries in the small intestine,3Watanabe T. Sugimori S. Kameda N. Machida H. Okazaki H. Tanigawa T. Watanabe K. Tominaga K. Fujiwara Y. Oshitani N. Higuchi K. Arakawa T. Small bowel injury by low-dose enteric-coated aspirin and treatment with misoprostol: a pilot study.Clin Gastroenterol Hepatol. 2008; 6: 1279-1282Abstract Full Text Full Text PDF PubMed Scopus (160) Google Scholar the precise mechanisms by which NSAIDs injure the small intestine must be clarified in order to develop new therapies for the prevention and treatment of NSAID-induced enteropathy. We have previously demonstrated that NSAIDs injure the small intestine through activation of the Toll-like receptor 4 (TLR4)-dependent signaling pathway.4Watanabe T. Higuchi K. Kobata A. Nishio H. Tanigawa T. Shiba M. Tominaga K. Fujiwara Y. Oshitani N. Asahara T. Nomoto K. Takeuchi K. Arakawa T. Non-steroidal anti-inflammatory drug-induced small intestinal damage is Toll-like receptor 4 dependent.Gut. 2008; 57: 181-187Crossref PubMed Scopus (206) Google Scholar TLR4 is a member of the TLR family, which plays a crucial role in innate immune responses to microbial pathogens.5Hoshino K. Takeuchi O. Kawai T. Sanjo H. Ogawa T. Takeda Y. Takeda K. Akira S. Cutting edge: Toll-like receptor 4 (TLR4)-deficient mice are hyporesponsive to lipopolysaccharide: evidence for TLR4 as the Lps gene product.J Immunol. 1999; 162: 3749-3752Crossref PubMed Google Scholar TLR4 recognizes primarily lipopolysaccharide (LPS), which is the major cell-wall component of Gram-negative bacteria. The triggering of the TLR4 signaling pathway by LPS leads to activation of nuclear factor κB (NF-κB) through the accessory protein MyD88 and subsequently to regulation of immune and inflammatory genes, including inflammatory cytokines such as tumor necrosis factor α (TNF-α). Several molecules, including high mobility group box 1 (HMGB1), act as endogenous TLR4 ligands.6Dobrovolskaia M.A. Medvedev A.E. Thomas K.E. Cuesta N. Toshchakov V. Ren T. Cody M.J. Michalek S.M. Rice N.R. Vogel S.N. Induction of in vitro reprogramming by Toll-like receptor (TLR)2 and TLR4 agonists in murine macrophages: effects of TLR "homotolerance" versus "heterotolerance" on NF-kappa B signaling pathway components.J Immunol. 2003; 170: 508-519Crossref PubMed Scopus (276) Google Scholar HMGB1, which is a member of the high mobility group protein superfamily, exists in all eukaryotic cells. It is a highly conserved protein, with >95% amino acid identity between rodents and humans. It was initially discovered as a nuclear protein that interacts with DNA as a chromatin-associated nonhistone protein7Bustin M. Regulation of DNA-dependent activities by the functional motifs of the high-mobility-group chromosomal proteins.Mol Cell Biol. 1999; 19: 5237-5246Crossref PubMed Scopus (757) Google Scholar and that stabilizes nucleosomes and regulates transcription of many genes in the nucleus.8Park J.S. Arcaroli J. Yum H.K. Yang H. Wang H. Yang K.Y. Choe K.H. Strassheim D. Pitts T.M. Tracey K.J. Abraham E. Activation of gene expression in human neutrophils by high mobility group box 1 protein.Am J Physiol Cell Physiol. 2003; 284: C870-C879Crossref PubMed Scopus (386) Google Scholar Once HMGB1 is released into the extracellular environment, it acts as an inflammatory cytokine (an alarmin).9Wang H. Bloom O. Zhang M. Vishnubhakat J.M. Ombrellino M. Che J. Frazier A. Yang H. Ivanova S. Borovikova L. Manogue K.R. Faist E. Abraham E. Andersson J. Andersson U. Molina P.E. Abumrad N.N. Sama A. Tracey K.J. HMG-1 as a late mediator of endotoxin lethality in mice.Science. 1999; 285: 248-251Crossref PubMed Scopus (2957) Google Scholar HMGB1 leaks out of cells during necrotic cell death,10Scaffidi P. Misteli T. Bianchi M.E. Release of chromatin protein HMGB1 by necrotic cells triggers inflammation [Erratum appeared in Nature 2010, 467:622].Nature. 2002; 418: 191-195Crossref PubMed Scopus (3358) Google Scholar and it is actively secreted by monocytes and macrophages.10Scaffidi P. Misteli T. Bianchi M.E. Release of chromatin protein HMGB1 by necrotic cells triggers inflammation [Erratum appeared in Nature 2010, 467:622].Nature. 2002; 418: 191-195Crossref PubMed Scopus (3358) Google Scholar, 11Andersson U. Wang H. Palmblad K. Aveberger A.C. Bloom O. Erlandsson-Harris H. Janson A. Kokkola R. Zhang M. Yang H. Tracey K.J. High mobility group 1 protein (HMG-1) stimulates proinflammatory cytokine synthesis in human monocytes.J Exp Med. 2000; 192: 565-570Crossref PubMed Scopus (1229) Google Scholar HMGB1 exerts these proinflammatory effects through TLR2, which is a receptor for components of the Gram-positive bacterial cell wall, and through the receptor for advanced glycation end-products (RAGE), which is a multiligand receptor belonging to the immunoglobulin superfamily, as well as through TLR4.12Yu M. Wang H. Ding A. Golenbock D.T. Latz E. Czura C.J. Fenton M.J. Tracey K.J. Yang H. HMGB1 signals through Toll-like receptor (TLR) 4 and TLR2.Shock. 2006; 26: 174-179Crossref PubMed Scopus (668) Google Scholar Similarly to TLR4, triggering of the TLR2 signaling pathway results in NF-κB activation. Binding of RAGE to its ligands activates mitogen-activated protein kinases (MAPKs), including extracellular signal-regulated kinase (ERK) and p38, which results in NF-κB activation.13Taguchi A. Blood D.C. del Toro G. Canet A. Lee D.C. Qu W. Tanji N. Lu Y. Lalla E. Fu C. Hofmann M.A. Kislinger T. Ingram M. Lu A. Tanaka H. Hori O. Ogawa S. Stern D.M. Schmidt A.M. Blockade of RAGE-amphoterin signalling suppresses tumour growth and metastases.Nature. 2000; 405: 354-360Crossref PubMed Scopus (1106) Google Scholar HMGB1 causes inflammatory responses and tissue injuries in various organ systems, including colitis,14Maeda S. Hikiba Y. Shibata W. Ohmae T. Yanai A. Ogura K. Yamada S. Omata M. Essential roles of high-mobility group box 1 in the development of murine colitis and colitis-associated cancer.Biochem Biophys Res Commun. 2007; 360: 394-400Crossref PubMed Scopus (95) Google Scholar, 15Davé S.H. Tilstra J.S. Matsuoka K. Li F. DeMarco R.A. Beer-Stolz D. Sepulveda A.R. Fink M.P. Lotze M.T. Plevy S.E. Ethyl pyruvate decreases HMGB1 release and ameliorates murine colitis.J Leukoc Biol. 2009; 86: 633-643Crossref PubMed Scopus (146) Google Scholar pancreatitis,16Sawa H. Ueda T. Takeyama Y. Yasuda T. Shinzeki M. Nakajima T. Kuroda Y. Blockade of high mobility group box-1 protein attenuates experimental severe acute pancreatitis.World J Gastroenterol. 2006; 12: 7666-7670Crossref PubMed Scopus (119) Google Scholar and ischemia-reperfusion (I-R) injuries of liver,17Tsung A. Sahai R. Tanaka H. Nakao A. Fink M.P. Lotze M.T. Yang H. Li J. Tracey K.J. Geller D.A. Billiar T.R. The nuclear factor HMGB1 mediates hepatic injury after murine liver ischemia-reperfusion.J Exp Med. 2005; 201: 1135-1143Crossref PubMed Scopus (960) Google Scholar heart,18Andrassy M. Volz H.C. Igwe J.C. Funke B. Eichberger S.N. Kaya Z. Buss S. Autschbach F. Pleger S.T. Lukic I.K. Bea F. Hardt S.E. Humpert P.M. Bianchi M.E. Mairbaurl H. Nawroth P.P. Remppis A. Katus H.A. Bierhaus A. High-mobility group box-1 in ischemia-reperfusion injury of the heart.Circulation. 2008; 117: 3216-3226Crossref PubMed Scopus (521) Google Scholar and kidney.19Wu H. Chen G. Wyburn K.R. Yin J. Bertolino P. Eris J.M. Alexander S.I. Sharland A.F. Chadban S.J. TLR4 activation mediates kidney ischemia/reperfusion injury.J Clin Invest. 2007; 117: 2847-2859Crossref PubMed Scopus (681) Google Scholar In this study, we investigated the roles of HMGB1 and its receptors TLR2, TLR4, and RAGE in the pathogenesis of NSAID-induced small intestinal inflammation and damage in mice. TLR2- and TLR4-knockout mice (KO), which were originally generated by Dr. Shizuo Akira (Osaka University, Osaka, Japan)5Hoshino K. Takeuchi O. Kawai T. Sanjo H. Ogawa T. Takeda Y. Takeda K. Akira S. Cutting edge: Toll-like receptor 4 (TLR4)-deficient mice are hyporesponsive to lipopolysaccharide: evidence for TLR4 as the Lps gene product.J Immunol. 1999; 162: 3749-3752Crossref PubMed Google Scholar and backcrossed eight times on a C57BL/6 background, were obtained from Oriental Bioservice (Kyoto, Japan). RAGE-KO mice, which had been backcrossed eight times on a C57BL/6 background, were originally generated by and were a gift from Dr. Yasuhiko Yamamoto (Kanazawa Medical University, Kanazawa, Japan).20Myint K.M. Yamamoto Y. Doi T. Kato I. Harashima A. Yonekura H. Watanabe T. Shinohara H. Takeuchi M. Tsuneyama K. Hashimoto N. Asano M. Takasawa S. Okamoto H. Yamamoto H. RAGE control of diabetic nephropathy in a mouse model: effects of RAGE gene disruption and administration of low-molecular weight heparin.Diabetes. 2006; 55: 2510-2522Crossref PubMed Scopus (219) Google Scholar Wild-type C57BL/6 mice were purchased from Charles River Japan (Atsugi, Japan) as the control strain for TLR2-KO, TLR4-KO, and RAGE-KO mice. In the animal experiments, specific pathogen-free 12-week-old male animals were used. All animals were housed in polycarbonate cages with paper-chip bedding. The cages were located in an air-conditioned biohazard room under a 12-hour light-dark cycle. To induce small intestinal injury, we administered 10 mg/kg of indomethacin, 40 mg/kg of naproxen, or 50 mg/kg NS-398 (a selective cyclooxygenase-2 inhibitor) in a 0.5% carboxymethylcellulose solution by gavage to nonfasting animals. Animals were sacrificed 3, 12, or 24 hours later. To evaluate tissue damage, 1% Evans Blue was injected intravenously 30 minutes before sacrifice, and the small intestine was opened along the antimesenteric attachment. The areas (mm2) of the macroscopically visible lesions were measured and summed per small intestine; this sum was used as the lesion index. All animals had free access to food and water. All experimental procedures were approved by the Animal Care Committee of the Osaka City University Graduate School of Medicine. To clarify the involvement of HMGB1 in indomethacin-induced small intestinal injury, mice received intraperitoneal injections of 100 to 1000 μg/kg human recombinant HMGB1 (rHMGB1; Sigma-Aldrich, St. Louis, MO) or vehicle (PBS) at 0 and 3 hours after indomethacin treatment. (Each mouse was given two injections.) Additionally, mice were intraperitoneally administered neutralizing chicken anti-HMGB1 polyclonal antibody (50 mg/kg; Shino-Test, Tokyo, Japan), normal chicken IgY (50 mg/kg; Sigma-Aldrich), or ethyl pyruvate (40 mg/kg; Sigma-Aldrich), which is an inhibitor of HMGB1 release,21Ulloa L. Ochani M. Yang H. Tanovic M. Halperin D. Yang R. Czura C.J. Fink M.P. Tracey K.J. Ethyl pyruvate prevents lethality in mice with established lethal sepsis and systemic inflammation.Proc Natl Acad Sci USA. 2002; 99: 12351-12356Crossref PubMed Scopus (538) Google Scholar at 0 and 3 hours after indomethacin treatment. To determine the HMGB1 receptors responsible for the damage, TLR2-KO, TLR4-KO, and RAGE-KO mice were administered 10 mg/kg of indomethacin by gavage with or without intraperitoneal injections of 1000 μg/kg of rHMGB1. Experiments were performed using four to eight samples. Total RNA was isolated from small intestinal tissue using an ISOGEN kit (Nippon Gene, Tokyo, Japan) according to the manufacturer's protocol. Complementary DNA was acquired using a high-capacity RNA-to-cDNA kit (Life Technologies, Carlsbad, CA) according to the manufacturer's protocol. Real-time quantitative RT-PCR analyses were performed using an Applied Biosystems ABI 7500 Fast RT-PCR system and software (Life Technologies, Foster City, CA). The reaction mixture was prepared according to the manufacturer's protocol using the ABI TaqMan Fast universal PCR master mixture (Life Technologies). Thermal cycling conditions were 45 cycles of amplification at 95°C for 15 seconds and 60°C for 1 minute. Total RNA was subjected to real-time quantitative RT-PCR for the measurement of target genes using ABI TaqMan glyceraldehyde-3-phosphate dehydrogenase (GADPH) control reagents, which were used as an internal standard. mRNA expression levels for HMGB1, TLR2, TLR4, RAGE, and a number of inflammatory mediators [TNF-α, monocyte chemotactic protein-1 (MCP-1), and the mouse IL-8 homolog (KC)] in injured and normal small intestinal tissue were quantified using real-time RT-PCR and were standardized to GAPDH mRNA levels. mRNA expression levels are reported relative to the mean value in normal intestinal tissue. The primers and probes used for RT-PCR are given in Table 1. RAGE primers and probes were purchased from Applied Biosystems.Table 1Primers and ProbesMurine gene (protein)Primers and probesSequenceTnf (TNF-α)Forward5′-TCATGCACCACCATCAAGGA-3′Reverse5′-GAGGCAACCTGACCACTCTCC-3′Probe5′-FAM-AATGGGCTTTCCGAATTCACTGGAGC-TAMRA-3′Mcp1 (MCP-1)Forward5′-CCACTCACCTGCTGCTACTCAT-3′Reverse5′-GGTGATCCTCTTGTAGCTCTCCA-3′Probe5′-FAM-CACCAGCAAGATGATCCCAATGAGTAGGTAMRA-TAMRA-3′Cxcl1 (KC)Forward5′-TCATCGATTTCTCCCCTGTGA-3′Reverse5′-CTCGCGACCATTCTTGAGTGT-3′Probe5′-FAM-CCCACTGCACCCAAACCGAAGTCATA-TAMRA-3′Tlr2 (TLR2)Forward5′-CTCTGGAGCATCCGAATTGC-3′Reverse5′-GCTGAAGAGGACTGTTATGGC-3′Probe5′-CCTCAGACAAAGCGTCAAATCTCAGAGGA-TAMRA-3′Tlr4 (TLR4)Forward5′-GGCTGGATTTATCCAGGTGTGA-3′Reverse5′-CTGTCAGTATCAAGTTTGAGAGGTG-3′Probe5′-AGCCATGCCATGCCTTGTCTTCAATTGT-TAMRA-3′Hmgb1 (HMBG1)Forward5′-CAGCCATTGCAGTACATTGAGC-3′Reverse5′-TCTCCTTTGCCCATGTTTAGTTG-3′Probe5′-GACAGAGTCGCCCAGTGCCCGTCC-TAMRA-3′ Open table in a new tab Tissue samples were fixed with 0.1 mol/L phosphate buffer (pH 7.4) containing 4% paraformaldehyde and were embedded in Tissue-Tek optimal cutting temperature compound (Sakura Finetek Japan, Tokyo, Japan). Serial cryostat sections (5 μm thick) were mounted on silanized slides (Dako, Tokyo, Japan). The specimens were immersed in a solution of 3% H2O2 in absolute methanol to inhibit endogenous peroxidase activity for 5 minutes and then were incubated in 5% nonfat milk for 10 minutes. A rabbit monoclonal anti-HMGB1 antibody (diluted 1:250; Abcam, Cambridge, MA), anti-phosphorylated-p38 (Santa Cruz Biotechnology, Santa Cruz, CA), anti-phosphorylated c-Jun N-terminal kinase (JNK) (Cell Signaling Technology, Danvers, MA), and anti-phosphorylated-ERK (Santa Cruz Biotechnology) were applied as the primary antibodies and were incubated overnight at 4°C with the specimens. A Histofine Simple Stain MAX peroxidase kit (Nichirei Biosciences, Tokyo, Japan) was used as the secondary antibody, incubated with the specimens for 1 hour according to the manufacturer's instructions. Immunoreactivity was visualized by treating the sections with a Histofine Simple Stain diaminobenzidine solution (Nichirei Biosciences). The specimens were then counterstained with hematoxylin. H&E staining was performed for morphological observations. Expression levels of TLR2, TLR4, and RAGE and the colocalization of these receptors with leukocytes or endothelial cells were determined with an immunofluorescence method. The primary antibodies used in the immunofluorescent staining were as follows (all diluted 1:200; Abcam): mouse monoclonal antibodies against TLR2 and TLR4, a rat monoclonal antibody against RAGE, rabbit monoclonal antibodies against CD68 (a monocyte/macrophage marker) and CD3, and a rabbit polyclonal antibody against CD31 (a marker of endothelial cells). The tissue samples, prepared as described above, were incubated overnight at 4°C with the primary antibodies and then reacted with the corresponding secondary fluorescent dye-conjugated antibodies (Alexa Fluor 488 and 594; Invitrogen, Carlsbad, CA) for 2 hours. Samples were examined under a confocal microscope equipped with argon and argon-krypton laser sources. Tissue sections stained with H&E were viewed under a light microscope. For each mouse, at least 10 random villi at injured areas were scored in a masked fashion by two investigators independently (Y.N and T.W). For evaluation, we used a modified histological scoring system.22Kaczmarek A. Brinkman B.M. Heyndrickx L. Vandenabeele P. Krysko D.V. Severity of doxorubicin-induced small intestinal mucositis is regulated by the TLR-2 and TLR-9 pathways.J Pathol. 2012; 226: 598-608Crossref PubMed Scopus (90) Google Scholar, 23de Koning B.A. van Dieren J.M. Lindenbergh-Kortleve D.J. van der Sluis M. Matsumoto T. Yamaguchi K. Einerhand A.W. Samsom J.N. Pieters R. Nieuwenhuis E.E. Contributions of mucosal immune cells to methotrexate-induced mucositis.Int Immunol. 2006; 18: 941-949Crossref PubMed Scopus (65) Google Scholar The histology score ranged from 0 to 13 and was subdivided into the following six categories: epithelium (0 = normal, 1 = flattened, 2 = loss of epithelial continuity, 3 = severe denudation), villus shape (0 = normal, 1 = short and rounded, 2 = extremely short and thick), villus tip (0 = normal, 1 = damaged, 2 = severely damaged), stroma (0 = normal, 1 = slightly retracted, 2 = severely retracted), inflammation (0 = no infiltration, 1 = mild infiltration, 2 = severe infiltration), and crypt status (0 = normal, 1 = mild crypt loss, 2 = severe crypt loss). Blood samples (1000 μL) were obtained in serum separator tubes by cardiac puncture. After centrifugation at 780 × g for 10 minutes, the serum was collected and stored at −80°C. Serum levels of HMGB1 were measured using an HMGB1 sandwich ELISA kit (Shino-Test) according to the manufacturer's protocol. Mice received intraperitoneal injections of rHMG1 (1000 μg/kg), neutralizing chicken anti-HMGB1 polyclonal antibody (50 mg/kg), or vehicle at 0 and 3 hours after treatment with 10 mg/kg of indomethacin and were sacrificed 3 hours after the indomethacin challenge. The small intestinal tissues of the mice were assayed for NF-κB activation. Nuclear proteins of the small intestine were extracted using an NE-PER nuclear protein extraction kit (Thermo Fisher Scientific, Rockford, IL). Binding activity of the NF-κB p65 subunit to the NF-κB DNA-binding consensus sequence 5′-GGGACTTTCC-3′ was measured using an ELISA-based transcription factor activation assay kit (TransAM; Active Motif, Carlsbad, CA) according to the manufacturer's instructions. The kit uses a 96-well microtiter plate that is coated with an oligonucleotide containing the NF-κB consensus sequence. After incubation with an antibody to p65, a horseradish peroxidase-conjugated secondary antibody was added, followed by addition of a developing solution. Absorbance was read on a spectrophotometer (MTP-500; Corona Electronics, Ibaraki, Japan) at 450 nm with a reference wavelength of 655 nm on a spectrophotometer. Small intestinal tissues were homogenized and lysed on ice in a buffer containing 0.5% NP-40, 40 mmol/L Tris-HCl (pH 8.0), 120 mmol/L NaCl, phosphatase inhibitor cocktail (PhosSTOP; Roche Applied Science, Indianapolis, IN), and a Complete Mini protease cocktail inhibitor (Thermo Fisher Scientific). Protein levels in the lysate were measured with a modified bicinchoninic acid method (Thermo Fisher Scientific). Proteins were denatured with sample buffer at 95°C for 5 minutes, subjected to 10% SDS-PAGE, and transferred to a polyvinylidene difluoride membrane. Membranes were blocked in Tris-buffered saline buffer (10 mmol/L Tris-HCl pH 7.5, 100 mmol/L NaCl, 0.1% Tween-20) containing 5% nonfat milk and then were incubated overnight with one of the following antibodies: anti-phosphorylated-p38, anti-p38, anti-phosphorylated JNK (Cell Signaling Technology), anti-JNK, anti-phosphorylated-ERK, or anti-ERK (Santa Cruz Biotechnology). Bound antigen-antibody complexes were detected with anti-rabbit IgG-HRP with Amersham enhanced chemiluminescence (GE Healthcare Life Sciences, Arlington Heights, IL) according to the manufacturer's instructions. Relevant bands were quantified with laser-scanning densitometry. Data are expressed as means ± SEM. One-way analysis of variance was used to test for significance of differences among treatment group means, and the results were analyzed with Fisher's protected least significant difference test. P values of < 0.05 were considered significant. Macroscopic small intestinal damage, visualized as dark blue staining with 1% Evans Blue (Figure 1, A and B), was observed beginning 3 hours after indomethacin administration. The lesion index increased in a time-dependent manner (Figure 1C). Similarly, naproxen induced small intestinal damage (Figure 1C), whereas administration of NS-398 did not cause small intestinal injury. mRNA expression levels for TNF-α, KC, and MCP-1 increased after indomethacin administration (Figure 1, D–F). mRNA levels for HMGB1 in the small intestine peaked 12 hours after indomethacin administration (Figure 2A), and serum levels of HMGB1 peaked by 3 hours (Figure 2B). Similar dynamics of serum HMGB1 levels were observed in the naproxen-treated mice. The administration of NS-398, however, did not affect serum levels of HMGB1. Immunohistochemically, HMGB1 localization was limited to inside the nuclei of epithelial cells and interstitial cells in normal intestinal mucosa (Figure 2C). In injured areas, however, indomethacin treatment induced prominent cytoplasmic staining of HMGB1 in epithelial cells by 3 hours (Figure 2D). The administration of rHMGB1 at a dose of 1000 μg/kg significantly aggravated small intestinal damage and increased mRNA expression levels for TNF-α, KC, and MCP-1 by 24 hours (Figure 3, A–E). In contrast, neutralizing antibodies to HMGB1 prevented the damage by reducing mRNA expression levels of these inflammatory cytokines (Figure 3, F–J). Furthermore, the administration of ethyl pyruvate, which is an inhibitor of HMGB1 release, markedly inhibited the indomethacin-induced small intestinal damage (by 55.0%) and inhibited mRNA expression levels for TNF-α, KC, and MCP-1 by 17.6%, 70.9%, and 45.4%, respectively, 24 hours after the indomethacin challenge (data not shown). Histologically, indomethacin caused intestinal sloughing and the destruction of the upper part of the epithelium and the infiltration of inflammatory cells by 3 hours (Figure 4A). By 12 hours, these mucosal injuries had progressed, and necrosis and destruction of the lower part of the epithelium were observed. By 24 hours, the intestinal ulcers extended into the submucosal layer, with a massive infiltration of inflammatory cells (Figure 4, B and C). In the rHMGB1-treated group, indomethacin caused more severe injuries by 3 hours, with more destructive and necrotic changes of the epithelium and inflammatory cell infiltration (Figure 4D) compared with the vehicle-treated group, whereas it caused less severe injuries in the anti-HMGB1 antibody-treated group (Figure 4E). Similarly, the mean histological score was significantly higher in the rHMGB1-treated group after 24 hours (Figure 4F), compared with the vehicle-treated group, whereas it was lower in the anti-HMGB1 antibody-treated group (Figure 4G). Indomethacin increased the binding activity of NF-κB by 3 hours in the small intestine, an increase that was further enhanced by exogenous HMGB1 (Figure 5A). The neutralizing antibody to HMGB1 prevented the increase in NF-κB binding activity by indomethacin (Figure 5B). The indomethacin challenge induced the phosphorylation of p38, JNK, and ERK by 3 hours (Figure 5, C–E). Exogenous HMGB1 further increased the phosphorylation of p38 and JNK and generally increased the phosphorylation of ERK (P = 0.07) (Figure 5, F–H). In contrast, anti-HMGB1 antibodies inhibited the phosphorylation of the MAPKs that were induced by indomethacin treatment, although the differences in phosphorylation levels between the anti-HMGB1 antibody-treated groups and the control antibody-treated group were not significant (Figure 5, I–K). Immunohistological analysis showed that phospho-p38 and phospho-JNK were expressed mainly in the cytoplasm of inflammatory cells in the injured mucosa. The cytoplasm of some epithelial cells was also weakly stained for phospho-p38 and phospho-JNK. Treatment with rHMGB1 induced stronger nuclear staining in inflammatory cells than cytoplasmic staining in epithelial cells. Conversely, immunoneutralization of HMGB1 reduced expression in both inflammatory cells and epithelial cells (Figure 6, A–H). Expression of phospho-ERK was restricted almost exclusively to inflammatory cells. Treatment with rHMGB1 induced stronger staining in the nucleus and cytoplasm of inflammatory cells. Conversely, immunoneutralization of HMGB1 reduced expression in inflammatory cells (Figure 6, I–L). TLR4 deficiency resulted in the inhibition of small intestinal injury and prevented the increase of TNF-α mRNA expression by 24 hours after indomethacin treatment (Figure 7, A and B), whereas neither TLR2 deficiency nor RAGE deficiency affected injury and expression of TNF-α mRNA (Figure 7, C–F). Exogenous HMGB1 aggravated small intestinal injury in both TLR2-KO and RAGE-KO mice and increased th