Advertisement

Cecal Tumorigenesis in Aryl Hydrocarbon Receptor–Deficient Mice Depends on Cecum-Specific Mitogen-Activated Protein Kinase Pathway Activation and Inflammation

Open ArchivePublished:November 14, 2019DOI:https://doi.org/10.1016/j.ajpath.2019.10.005
      The aryl hydrocarbon receptor (AhR) is a transcription factor known as a dioxin receptor. Recently, Ahr−/− mice were revealed to develop cecal tumors with inflammation and Wnt/β-catenin pathway activation. However, whether β-catenin degradation is AhR dependent remains unclear. To determine whether other signaling pathways function in Ahr−/− cecal tumorigenesis, we investigated histologic characteristics of the tumors and cytokine/chemokine production in tumors and Ahr−/− peritoneal macrophages. AhR expression was also assessed in human colorectal carcinomas. Of the 28 Ahr−/− mice, 10 developed cecal lesions by 50 weeks of age, an incidence significantly lower than previously reported. Cecal lesions of Ahr−/− mice developed from serrated hyperplasia to adenoma/dysplasia-like neoplasia with enhanced proliferation. Macrophage and neutrophil infiltration into the lesions was also observed early in serrated hyperplasia, although adjacent mucosa was devoid of inflammation. Il1b, Il6, Ccl2, and Cxcl5 were up-regulated at lesion sites, whereas only IL-6 production increased in Ahr−/− peritoneal macrophages after lipopolysaccharide + ATP stimulation. Neither Myc (alias c-myc) up-regulation nor β-catenin nuclear translocation was observed, unlike previously reported. Interestingly, enhanced phosphorylation of extracellular signal-regulated kinase, Src, and epidermal growth factor receptor and Amphiregulin up-regulation at Ahr−/− lesion sites were detected. In human serrated lesions, however, AhR expression in epithelial cells was up-regulated despite morphologic similarity to Ahr−/− cecal lesions. Our results suggest novel mechanisms underlying Ahr−/− cecal tumorigenesis, depending primarily on cecum-specific mitogen-activated protein kinase pathway activation and inflammation.
      The aryl hydrocarbon receptor (AhR) is a ligand-activated transcription factor known as a dioxin receptor and is a member of the basic helix-loop-helix/Per-AhR nuclear translocator-Sim homology superfamily. On ligand binding, AhR translocates to the nucleus and induces transcription of target genes, like CYP1A1, to detoxify low molecular organic compounds.
      • Nguyen L.P.
      • Bradfield C.A.
      The search for endogenous activators of the aryl hydrocarbon receptor.
      However, AhR cannot detoxify dioxins, such as 2,3,7,8-tetrachlorobenzo-p-dioxin, and mediate their harmful effects oppositely through downstream targets.
      • Nakatsuru Y.
      • Wakabayashi K.
      • Fujii-Kuriyama Y.
      • Ishikawa T.
      • Kusama K.
      • Ide F.
      Dibenzo[A,L]pyrene-induced genotoxic and carcinogenic responses are dramatically suppressed in aryl hydrocarbon receptor-deficient mice.
      • Shimizu Y.
      • Nakatsuru Y.
      • Ichinose M.
      • Takahashi Y.
      • Kume H.
      • Mimura J.
      • Fujii-Kuriyama Y.
      • Ishikawa T.
      Benzo[a]pyrene carcinogenicity is lost in mice lacking the aryl hydrocarbon receptor.
      • Mimura J.
      • Yamashita K.
      • Nakamura K.
      • Morita M.
      • Takagi T.N.
      • Nakao K.
      • Ema M.
      • Sogawa K.
      • Yasuda M.
      • Katsuki M.
      • Fujii-Kuriyama Y.
      Loss of teratogenic response to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in mice lacking the Ah (dioxin) receptor.
      AhR reportedly suppresses the immune response in the context of innate and adaptive immunity.
      • Monteleone I.
      • Pallone F.
      • Monteleone G.
      Aryl hydrocarbon receptor and colitis.
      In particular, AhR suppresses caspase-1 activation through plasminogen activator inhibitor-2 and consequently antagonizes IL-1β production.
      • Sekine H.
      • Mimura J.
      • Oshima M.
      • Okawa H.
      • Kanno J.
      • Igarashi K.
      • Gonzalez F.J.
      • Ikuta T.
      • Kawajiri K.
      • Fujii-Kuriyama Y.
      Hypersensitivity of aryl hydrocarbon receptor-deficient mice to lipopolysaccharide-induced septic shock.
      Moreover, AhR was recently reported to regulate intracellular β-catenin protein levels by serving as a ligand-dependent E3 ubiquitin ligase independently of adenomatous polyposis coli (APC).
      • Ohtake F.
      • Baba A.
      • Takada I.
      • Okada M.
      • Iwasaki K.
      • Miki H.
      • Takahashi S.
      • Kouzmenko A.
      • Nohara K.
      • Chiba T.
      • Fujii-Kuriyama Y.
      • Kato S.
      Dioxin receptor is a ligand-dependent E3 ubiquitin ligase.
      ,
      • Kawajiri K.
      • Kobayashi Y.
      • Ohtake F.
      • Ikuta T.
      • Matsushima Y.
      • Mimura J.
      • Pettersson S.
      • Pollenz R.S.
      • Sakaki T.
      • Hirokawa T.
      • Akiyama T.
      • Kurosumi M.
      • Poellinger L.
      • Kato S.
      • Fujii-Kuriyama Y.
      Aryl hydrocarbon receptor suppresses intestinal carcinogenesis in ApcMin/+ mice with natural ligands.
      Previous reports showed that most AhR-deficient mice developed cecal tumors by 10 weeks of age.
      • Kawajiri K.
      • Kobayashi Y.
      • Ohtake F.
      • Ikuta T.
      • Matsushima Y.
      • Mimura J.
      • Pettersson S.
      • Pollenz R.S.
      • Sakaki T.
      • Hirokawa T.
      • Akiyama T.
      • Kurosumi M.
      • Poellinger L.
      • Kato S.
      • Fujii-Kuriyama Y.
      Aryl hydrocarbon receptor suppresses intestinal carcinogenesis in ApcMin/+ mice with natural ligands.
      ,
      • Ikuta T.
      • Kobayashi Y.
      • Kitazawa M.
      • Shiizaki K.
      • Itano N.
      • Noda T.
      • Pettersson S.
      • Poellinger L.
      • Fujii-Kuriyama Y.
      • Taniguchi S.
      • Kawajiri K.
      ASC-associated inflammation promotes cecal tumorigenesis in aryl hydrocarbon receptor-deficient mice.
      Two mechanisms were suggested as underlying tumorigenesis in these mice: Wnt/β-catenin pathway activation and severe inflammation accompanied by IL-1β and IL-6 up-regulation and subsequent Stat3 phosphorylation. Mice deficient in both AhR and apoptosis-associated speck-like protein containing a caspase recruitment domain showed considerably reduced tumor incidence relative to AhR single knockouts due to suppressed inflammasome activation.
      • Ikuta T.
      • Kobayashi Y.
      • Kitazawa M.
      • Shiizaki K.
      • Itano N.
      • Noda T.
      • Pettersson S.
      • Poellinger L.
      • Fujii-Kuriyama Y.
      • Taniguchi S.
      • Kawajiri K.
      ASC-associated inflammation promotes cecal tumorigenesis in aryl hydrocarbon receptor-deficient mice.
      Also, germ-free AhR-deficient mice do not show tumor development, suggesting that inflammation caused by gut microbiota is important for tumorigenesis.
      • Ikuta T.
      • Kobayashi Y.
      • Kitazawa M.
      • Shiizaki K.
      • Itano N.
      • Noda T.
      • Pettersson S.
      • Poellinger L.
      • Fujii-Kuriyama Y.
      • Taniguchi S.
      • Kawajiri K.
      ASC-associated inflammation promotes cecal tumorigenesis in aryl hydrocarbon receptor-deficient mice.
      Initially, AhR activity was reported to promote β-catenin degradation,
      • Ohtake F.
      • Baba A.
      • Takada I.
      • Okada M.
      • Iwasaki K.
      • Miki H.
      • Takahashi S.
      • Kouzmenko A.
      • Nohara K.
      • Chiba T.
      • Fujii-Kuriyama Y.
      • Kato S.
      Dioxin receptor is a ligand-dependent E3 ubiquitin ligase.
      ,
      • Kawajiri K.
      • Kobayashi Y.
      • Ohtake F.
      • Ikuta T.
      • Matsushima Y.
      • Mimura J.
      • Pettersson S.
      • Pollenz R.S.
      • Sakaki T.
      • Hirokawa T.
      • Akiyama T.
      • Kurosumi M.
      • Poellinger L.
      • Kato S.
      • Fujii-Kuriyama Y.
      Aryl hydrocarbon receptor suppresses intestinal carcinogenesis in ApcMin/+ mice with natural ligands.
      but later studies reported discrepant results.
      • Braeuning A.
      • Köhle C.
      • Buchmann A.
      • Schwarz M.
      Coordinate regulation of cytochrome P450 1A1 expression in mouse liver by the aryl hydrocarbon receptor and the β-catenin pathway.
      • Jin U.-H.
      • Lee S.-O.
      • Sridharan G.
      • Lee K.
      • Davidson L.A.
      • Jayaraman A.
      • Chapkin R.S.
      • Alaniz R.
      • Safe S.
      Microbiome-derived tryptophan metabolites and their aryl hydrocarbon receptor-dependent agonist and antagonist activities.
      • Kasai S.
      • Ishigaki T.
      • Takumi R.
      • Kamimura T.
      • Kikuchi H.
      β-Catenin signaling induces CYP1A1 expression by disrupting adherens junctions in Caco-2 human colon carcinoma cells.
      For example, Jin et al
      • Jin U.-H.
      • Lee S.-O.
      • Sridharan G.
      • Lee K.
      • Davidson L.A.
      • Jayaraman A.
      • Chapkin R.S.
      • Alaniz R.
      • Safe S.
      Microbiome-derived tryptophan metabolites and their aryl hydrocarbon receptor-dependent agonist and antagonist activities.
      reported that 2,3,7,8-tetrachlorobenzo-p-dioxin and tryptophan metabolites did not alter β-catenin levels in CaCo-2 cells through AhR, suggesting that other signaling pathways underlie cecal tumorigenesis in Ahr−/− mice.
      To investigate these discrepancies, we undertook detailed analysis of tumorigenesis in AhR-deficient mice. The data revealed that mitogen-activated protein kinase (MAPK) signaling was activated primarily at lesion sites of AhR−/− mice on the basis of the morphologic similarity between AhR−/− cecal lesions and human colorectal serrated lesions.

      Materials and Methods

      Animal Experiments

      Wild-type and AhR-deficient mice
      • Mimura J.
      • Yamashita K.
      • Nakamura K.
      • Morita M.
      • Takagi T.N.
      • Nakao K.
      • Ema M.
      • Sogawa K.
      • Yasuda M.
      • Katsuki M.
      • Fujii-Kuriyama Y.
      Loss of teratogenic response to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in mice lacking the Ah (dioxin) receptor.
      on a C57BL/6 background were bred and maintained under conventional conditions in the animal house of Shinshu University (Matsumoto, Japan). AhR-deficient mice were mated with apoptosis-associated speck-like protein containing a caspase recruitment domain–deficient mice
      • Yamamoto M.
      • Yaginuma K.
      • Tsutsui H.
      • Sagara J.
      • Guan X.
      • Seki E.
      • Yasuda K.
      • Yamamoto M.
      • Akira S.
      • Nakanishi K.
      • Noda T.
      • Taniguchi S.
      ASC is essential for LPS-induced activation of procaspase-1 independently of TLR-associated signal adaptor molecules.
      to generate Ahr−/− Asc−/− [double knockout (DKO)] mice. Because Ahr−/− mice were maintained by breeding DKO males with Ahr+/− Asc+/− females, experiments were performed using Ahr−/− Asc+/− (Ahr−/−), Ahr+/− Asc+/− [wild-type (WT)], and Ahr+/− Asc−/− (Asc−/−) mice. Sequences of PCR primers used for genotyping are listed in Table 1. In a time course experiment, 28 Ahr−/−, 23 WT, 12 DKO, and 16 Asc−/− mice were sacrificed at indicated times up to 50 weeks of age. Ileocecal regions of mice were opened to confirm the presence of cecal tumors, and intestinal tissues were used for histologic and quantitative RT-PCR (RT-qPCR) analysis. All animal experiments were approved by the Shinshu University Animal Care and Use Committee.
      Table 1Oligonucleotides Used in Genotyping
      NameSequence
      AhR-com-5sF: 5′-GGCGCGGGCACCATGAGCAG-3′
      AhR-wt3-3asR: 5′-GTAGGTCAGAGCTGTCAACGAACC-3′
      AhR-LacZ-3asR: 5′-GCGGATTGACCGTAATGGGATAGG-3′
      ASC-F25SLF: 5′-GCCATATGTGGCCCAGTGGTAG-3′
      ASC-R40R: 5′-TGGCTTTGGTTGGCATTGCATG-3′
      ASC-LacZF1F: 5′-GTAGGGTTTTTCACAGACCGCT-3’
      F, Forward; R, Reverse.

      Histologic Analysis and Immunohistochemistry

      Histologic and immunohistochemical studies were performed using sections (3 μm thick) from formalin-fixed, paraffin-embedded tissue blocks. Sections were stained with hematoxylin and eosin. Antibodies used for immunohistochemistry and procedures related to antibodies used in immunohistochemistry are shown in Table 2.
      Table 2Antibodies Used in Immunohistochemistry and Procedures of Respective Antibodies
      AntibodiesClone name or catalog no.ManufacturerAntigen retrievalSecondary antibodyNote
      Anti–Ki-67B56BD Pharmingen (San Diego, CA)MicrowavingHistofine mouse stain kitMonM
      Anti–β-catenin9562Cell Signaling Technology (Danvers, MA)MicrowavingEnvision (anti-rabbit)
      Anti-p53FL-393Santa Cruz Biotechnology Inc. (Dallas, TX)MicrowavingEnvision (anti-rabbit)
      Anti-F4/80CI:A3-1Novus Biologicals (Centennial, CO)TrypsinizationAnti-rat Ig with HRP
      Anti-MPOab9535Abcam (Cambridge, UK)MicrowavingEnvision (anti-rabbit)
      Anti-B220PA3-6B2BD PharmingenTrypsinizationAnti-rat Ig with HRP
      Anti-CD3LN10Leica Biosystems (Wetzlar, Germany)MicrowavingHistofine mouse stain kitMonM
      Anti–p-Erk1/2D13.14.4ECell Signaling TechnologyMicrowavingEnvision (anti-rabbit)
      Anti–p-Src family2101Cell Signaling TechnologyMicrowavingEnvision (anti-rabbit)
      Anti-AhRA-3Santa Cruz Biotechnology Inc.MicrowavingEnvision (anti-mouse)
      Anti-MLH1ES05Dako (Santa Clara, CA)MicrowavingEnvision (anti-mouse)
      Anti–p-EGFREP774YBiocare Medical (Pacheco, CA)MicrowavingEnvision (anti-rabbit)
      Anti–p-Stat3D3A7Cell Signaling TechnologyMicrowavingEnvision (anti-rabbit)
      AhR, aryl hydrocarbon receptor; HRP, horseradish peroxidase; MLH1, mutL homolog 1; MonM, mouse-on-mouse immunostaining; MPO, myeloperoxidase; p-EGFR, phosphorylated epidermal growth factor receptor; p-Erk1/2, phosphorylated extracellular signal-regulated kinase 1/2; p-Src, phosphorylated Src; p-Stat3, phosphorylated Stat3.
      Tissue sections were deparaffinized and rehydrated through a series of xylene and ethanol. Endogenous peroxidase activity was blocked using 3% H2O2 in methanol for 10 minutes. Antigen was retrieved by microwaving [in 10 mmol/L Tris-HCl buffer (pH 8.0) containing 1 mmol/L EDTA for 30 minutes] or trypsinization [digested with 0.25% trypsin 250 (Difco, Franklin Lakes, NJ) at 37°C for 30 minutes], according to primary antibodies listed in Table 2. Slides were incubated with primary antibodies for 1 hour. Envision+ System (anti-rabbit) (Dako, Santa Clara, CA), Envision+ System (anti-mouse) (Dako), or anti-rat Ig antibody conjugated with horseradish peroxidase (Dako) was used as secondary antibody, according to primary antibodies. Slides were incubated with secondary antibodies for 30 minutes. For anti–Ki-67 and anti-CD3 antibodies, Histofine mouse stain kit (Nichirei Biosciences, Tokyo, Japan) was used for mouse-on-mouse immunostaining, following manufacturer's instructions. Peroxidase activity was visualized with diaminobenzidine–hydrogen peroxide solution.

      RT-qPCR

      Total RNA was extracted from mouse intestinal tissues using an RNeasy Mini Kit (Qiagen, Hilden, Germany), and 250 ng RNA was reverse transcribed using SuperScript III (Invitrogen, Waltham, MA), random primers (Promega, Fitchburg, WI), and oligo-dT primers (Promega), following the manufacturers' instructions. RT-qPCR analysis was performed using a 7300 Real-Time PCR System (Applied Biosystems, Waltham, MA). Premixed reagents containing primers and TaqMan probes (Applied Biosystems) were used for genes listed in Table 3. SYBR Premix Ex Taq (Takara, Kusatsu, Japan) and primers were used for genes shown in Table 4. mRNA expression was normalized to that of glyceraldehyde-3-phosphate dehydrogenase. ΔΔCT values and fold expression of target genes were determined by defining mRNA expression in a WT 45-week–old female mouse as 1.0. Data are represented as means ± SD (n = 6 for Ahr−/− mice; n = 7 for WT mice).
      Table 3Target Genes and Assay IDs Used in Quantitative RT-PCR Analysis with Premixed Reagents Containing Primers and TaqMan Probes
      GenesAssay IDs
      Il1bMm01336189_m1
      Il6Mm00446190_m1
      Ccl2Mm00441242_m1
      Cxcl5Mm00436451_g1
      IfngMm01168134_m1
      Il10Mm00439614_m1
      GapdhMm99999915_g1
      Assay IDs available at https://www.thermofisher.com/taqman-gene-expression (last accessed January 5, 2020).
      Ccl2, chemokine (C-C motif) ligand 2; Gapdh, glyceraldehyde-3-phosphate dehydrogenase; ID, identification; Ifng, interferon-γ; RT-qPCR, quantitative RT-PCR.
      Table 4Target Genes and Sequences of the Primers Used in Quantitative RT-PCR Analysis with SYBR Premix Ex Taq
      GenesSenseAntisense
      Ccl35′-CCATGACACTCTGCAACCAAG-3′5′-AATCTTCCGGCTGTAGGAGAA-3′
      Ccl45′-TTTCTCTTACACCTCCCGGC-3′5′-GTCTGCCTCTTTTGGTCAGGA-3′
      Ccl55′-GTGCTCCAATCTTGCAGTCG-3′5′-CAGGGAAGCTATACAGGGTCAG-3′
      Ccl65′-TTATCCTTGTGGCTGTCCTTGG-3′5′-GGCATAAGAGAAGCAGCAGTC-3′
      Ccl85′-CTACGCAGTGCTTCTTTGCC-3′5′-CGTAGCTTTTCAGCACCCGA-3′
      Il45′-TCTCGAATGTACCAGGAGCCATATC-3′5′-AGCACCTTGGAAGCCCTACAGA-3′
      Il175′-TCAACCGTTCCACGTCACCCT-3′5′-AGCTTTCCCTCCGCATTGACAC-3′
      Il12p405′-CCACTCATGGCCATGTGGG-3′5′-GCGTGTCACAGGTGAGGTTC-3′
      Myc (alias c-myc)5′-GTGCTGCATGAGGAGACACC-3′5′-GACCTCTTGGCAGGGGTTTG-3′
      Gapdh5′-GATGGGTGTGAACCACGAGA-3′5′-GCCCTTCCACAATGCCAAAG-3′
      Areg5′-GGGGACTACGACTACTCAGAG-3′5′-TCTTGGGCTTAATCACCTGTTC-3′

      PCR Amplification and Direct Sequencing of Mouse Braf

      Genomic DNA was extracted from mouse intestinal tissues and tails using a High Pure PCR Template Preparation Kit (Roche, Basel, Switzerland), followed by phenol chloroform extraction and ethanol precipitation. Exon 18 of mouse Braf was amplified by PCR using the primers 5′-GACAGCTTAAAAGTAGGTCGT-3′ (forward) and 5′-AGCCCTTCAGTGTATTTCTCG-3′ (reverse). The amplification protocol was as follows: initial denaturation at 98°C for 10 seconds; 30 cycles of denaturation at 98°C for 10 seconds, annealing at 55°C for 10 seconds, and extension at 72°C for 40 seconds; followed by a final extension at 72°C for 4 minutes using PrimeSTAR HS DNA Polymerase (Takara). PCR products were electrophoresed on 2.0% (w/v) agarose gels and purified using a QIAquick Gel Extraction Kit (Qiagen). Subsequently, DNA sequencing reaction was performed using the above forward or reverse primer with a BigDye Terminator version 3.1 Cycle Sequencing Kit (Thermo Fisher Scientific, Waltham, MA). Sequences were determined using a capillary automatic sequencer ABI 3130XL PRISM Genetic Analyzer (Thermo Fisher Scientific).

      Macrophage Preparation and Stimulation

      Ahr−/− and WT peritoneal macrophages were prepared, as described.
      • Fujii C.
      • Shiratsuchi A.
      • Manaka J.
      • Yonehara S.
      • Nakanishi Y.
      Difference in the way of macrophage recognition of target cells depending on their apoptotic states.
      ,
      • Shiratsuchi A.
      • Osada S.
      • Kanazawa S.
      • Nakanishi Y.
      Essential role of phosphatidylserine externalization in apoptosing cell phagocytosis by macrophages.
      More than 90% of cells obtained by this procedure were confirmed to be macrophages on the basis of F4/80 staining. Macrophages were cultivated in RPMI 1640 medium with 10% fetal calf serum, 100 μg/mL streptomycin, and 100 U/mL penicillin G with 500 ng/mL lipopolysaccharide (LPS; Sigma-Aldrich, St. Louis, MO) for 6 or 24 hours and then incubated with 2.5 mmol/L ATP (Sigma-Aldrich) for 30 minutes.

      Bacterial Infection

      Fusobacterium nucleatum (ATCC, Manassas, VA; ATCC 25586) was cultured on Anaero Columbia Blood Agar plates (BD Biosciences, Franklin Lakes, NJ) under anaerobic conditions, and Escherichia coli (ATCC 25922) was cultured on blood agar plates (BD Biosciences). Fusobacterium nucleatum and E. coli were transferred to new plates 3 and 1 day before infection, respectively. Fusobacterium nucleatum and E. coli were individually suspended in physiological saline, the number of bacteria in suspension was determined on the basis of McFarland turbidity, and bacterial suspensions were diluted with antibiotic-free medium to the desired multiplicity of infection. Ahr−/− and WT peritoneal macrophages were prepared in antibiotic-free medium, and infections were performed by adding diluted F. nucleatum or E. coli to macrophages for 6 or 24 hours. Gentamycin (10 μg/mL) was added to cell cultures 1 hour after infection.

      Cytokine Assays

      Culture supernatants of peritoneal macrophages were harvested at indicated time points. IL-1β, tumor necrosis factor (TNF), IL-6, and chemokine (C-C motif) ligand 2 (CCL2) concentrations were measured using a cytometric beads array flex set (BD Biosciences), following the manufacturer's instructions. Data are represented as means ± SD of four (LPS + ATP stimulation) and three (bacterial infection) independent experiments.

      Patients and Tissue Samples

      Specimens were obtained from 15 patients with colon tumors who underwent surgical or endoscopic resection at Ina Central Hospital (Ina, Japan) between 2015 and 2018. Lesions included one tubular adenoma, two sessile serrated adenoma/polyps (SSA/Ps), six adenocarcinomas with adenoma components, three adenocarcinomas with SSA/P components, and five adenocarcinomas without adenoma or SSA/P components. In the last five cases without benign components, MLH1 immunostaining was performed to determine microsatellite instability status; and three negatively stained cases were considered to be microsatellite instability high and to have developed through the sessile serrated pathway. Use of retrospective tissue samples was approved by the Ethics Committee of Ina Central Hospital.

      Immunohistochemistry Evaluation

      For immunohistochemistry evaluation of mice samples, two high-power fields (×400; 0.4 mm2) of the Ahr−/− lesion site or the WT ileocecal region per mouse were examined for each staining, except for phosphorylated extracellular signal-regulated kinase (p-Erk). For Ki-67, phosphorylated Src (p-Src), and phosphorylated epidermal growth factor receptor (p-EGFR) immunostaining, positive ratios were calculated as the proportion of positively stained epithelial cells relative to total number of epithelial cells. For F4/80, myeloperoxidase, and phosphorylated Stat3 immunostaining, the total number of positively stained cells per 1 mm2 was calculated. For β-catenin immunostaining, the nuclear translocation ratio was determined as the proportion of epithelial cells showing β-catenin nuclear localization relative to total epithelial cells. For p-Erk immunostaining, h-scores of the entire Ahr−/− lesion site or the corresponding WT ileocecal region were assigned. To calculate the h-score, the staining intensity in epithelial cells [none (0), weak (1), moderate (2), or strong (3)] was multiplied by the percentage of positively stained cells (0% to 100%) of each intensity, resulting in a score ranging from 0 to 300. Data are represented as means ± SD.
      For AhR immunostaining of human colorectal specimens, h-scores of normal colorectal mucosa, benign component (tubular adenoma and SSA/Ps), and the adenocarcinoma component were assigned, as described above. Data are represented as means ± SD.

      Statistical Analysis

      Cumulative incidence of cecal lesions in mice was estimated by Kaplan-Meier analysis. Significance was evaluated by an unpaired two-tailed t-test for RT-qPCR analysis and cytokine assays. The immunohistochemical positive ratio (for Ki-67, p-Src, and p-EGFR), the number of positively stained cells per 1 mm2 (for F4/80, myeloperoxidase, and phosphorylated Stat3), the nuclear translocation ratio (for β-catenin), and the h-score (for p-Erk) of Ahr−/− lesion sites and WT ileocecal regions were compared using an unpaired two-tailed t-test. Immunohistochemical h-scores of human colorectal specimens were compared using one-way analysis of variance with a Tukey-Kramer post-hoc test. All statistical data are presented as means ± SD, and P < 0.05 was considered significant. All statistical analyses were performed using EZR 1.37
      • Kanda Y.
      Investigation of the freely available easy-to-use software “EZR” for medical statistics.
      (Saitama Medical Center, Jichi Medical University, Saitama, Japan), which is a graphical user interface for R version 3.4.1 (The R Foundation for Statistical Computing, Vienna, Austria).

      Results

      Cecal Lesions of Ahr−/− Mice Develop from Serrated Hyperplasia to Neoplasia with Macrophage and Neutrophil Infiltration

      Macroscopic and microscopic cecal tumor incidence were studied in Ahr−/− (Ahr−/− Asc+/−), DKO (Ahr−/− Asc−/−), WT (Ahr+/− Asc+/−), and Asc−/− (Ahr+/− Asc−/−) mice to compare their phenotypes with those reported previously.
      • Ikuta T.
      • Kobayashi Y.
      • Kitazawa M.
      • Shiizaki K.
      • Itano N.
      • Noda T.
      • Pettersson S.
      • Poellinger L.
      • Fujii-Kuriyama Y.
      • Taniguchi S.
      • Kawajiri K.
      ASC-associated inflammation promotes cecal tumorigenesis in aryl hydrocarbon receptor-deficient mice.
      Of 28 Ahr−/− mice, 10 developed cecal lesions by 50 weeks of age (Figure 1A). All were protuberant masses located in the cecum near the ileocecal junction (Figure 1B). None of the 12 DKO, 23 WT, or 16 Asc−/− mice developed cecal lesions by 50 weeks of age (Figure 1A). Overall, tumor incidence of DKO mice was reduced compared with that of Ahr−/− mice, as previously reported.
      • Ikuta T.
      • Kobayashi Y.
      • Kitazawa M.
      • Shiizaki K.
      • Itano N.
      • Noda T.
      • Pettersson S.
      • Poellinger L.
      • Fujii-Kuriyama Y.
      • Taniguchi S.
      • Kawajiri K.
      ASC-associated inflammation promotes cecal tumorigenesis in aryl hydrocarbon receptor-deficient mice.
      Furthermore, tumor incidence of Ahr−/− mice was greatly reduced herein relative to previous reports,
      • Kawajiri K.
      • Kobayashi Y.
      • Ohtake F.
      • Ikuta T.
      • Matsushima Y.
      • Mimura J.
      • Pettersson S.
      • Pollenz R.S.
      • Sakaki T.
      • Hirokawa T.
      • Akiyama T.
      • Kurosumi M.
      • Poellinger L.
      • Kato S.
      • Fujii-Kuriyama Y.
      Aryl hydrocarbon receptor suppresses intestinal carcinogenesis in ApcMin/+ mice with natural ligands.
      ,
      • Ikuta T.
      • Kobayashi Y.
      • Kitazawa M.
      • Shiizaki K.
      • Itano N.
      • Noda T.
      • Pettersson S.
      • Poellinger L.
      • Fujii-Kuriyama Y.
      • Taniguchi S.
      • Kawajiri K.
      ASC-associated inflammation promotes cecal tumorigenesis in aryl hydrocarbon receptor-deficient mice.
      even after taking heterozygosity of Asc allele of this experiment into consideration.
      Figure thumbnail gr1
      Figure 1Histologic and immunohistochemical analysis of cecal tumorigenesis in Ahr−/− mice. A: Cecal tumor incidence in Ahr−/− mice. B: Macroscopic appearance of a lesion site from Ahr−/− mice, which is composed of a tumor component and a hyperplastic component. CH: Microscopic analysis of the lesion in B. Panels show immunostaining for Ki-67 (F), F4/80 (H), myeloperoxidase (MPO) (H), and β-catenin (D, G, inset). Black arrowheads indicate protuberant lesion site; orange arrowheads, adjacent mucosa; blue arrowheads, the tumor component; and yellow arrowheads, the hyperplastic component. Scale bars: 200 μm (C and D); 100 μm (EG, main image); 50 μm (G, inset, and H). DKO, double knockout; WT, wild type.
      Of the 10 Ahr−/− mice, microscopically, cecal lesions from nine were composed of a tumor component and a hyperplastic component (Figure 1, C and E). In tumor components, glands were branched and gland density was high (Figure 1G), and this component resembled human colorectal high-grade adenoma/dysplasia. One of the nine cases contained a focal adenocarcinoma component invading submucosal tissue (Supplemental Figure S1A). The hyperplastic component of the nine lesions was located mainly at the periphery of the lesion site, and glands exhibited serrated lumens (Figure 1E). The morphology of this component was similar to that of human SSA/P, a benign precursor of the serrated pathway in colorectal tumorigenesis. Therefore, this histology was referred to as serrated hyperplasia. Severe infiltration of inflammatory cells was observed, consisting mainly of macrophages and neutrophils in and under the lesion site, but there was almost no inflammation apparent in the adjacent mucosa (Figure 1, C, E, and H).
      Immunohistochemical analysis of lesions from Ahr−/− mice showed expansion of Ki-67–positive cells to the full thickness of the mucosa at the tumor component, but at the hyperplastic component, Ki-67–positive cells had also expanded mainly on the basal side of the gland (Figure 1F). Ki-67–positive cells were seen only at the bottom of the crypts in colon mucosa adjacent to the lesions or in normal colon mucosa of WT mice (data not shown). On the basis of these findings, it can be concluded that the two components can be distinguished by the proliferative capacity. In Ahr−/− samples, neither intense nor nuclear immunostaining was observed for β-catenin at the lesion site (Figure 1, D and G), even at the invasive front of the focal adenocarcinoma component (Supplemental Figure S1B). Moreover, β-catenin staining was comparable in Ahr−/− and WT mice (data not shown), unlike previously reported.
      • Kawajiri K.
      • Kobayashi Y.
      • Ohtake F.
      • Ikuta T.
      • Matsushima Y.
      • Mimura J.
      • Pettersson S.
      • Pollenz R.S.
      • Sakaki T.
      • Hirokawa T.
      • Akiyama T.
      • Kurosumi M.
      • Poellinger L.
      • Kato S.
      • Fujii-Kuriyama Y.
      Aryl hydrocarbon receptor suppresses intestinal carcinogenesis in ApcMin/+ mice with natural ligands.
      ,
      • Ikuta T.
      • Kobayashi Y.
      • Kitazawa M.
      • Shiizaki K.
      • Itano N.
      • Noda T.
      • Pettersson S.
      • Poellinger L.
      • Fujii-Kuriyama Y.
      • Taniguchi S.
      • Kawajiri K.
      ASC-associated inflammation promotes cecal tumorigenesis in aryl hydrocarbon receptor-deficient mice.
      All Ahr−/− samples were p53 negative (data not shown). F4/80-positive macrophages and myeloperoxidase-positive neutrophils infiltrated in and under the lesion site in Ahr−/− samples (Figure 1H), but few or no B220- or CD3-positive lymphocytes were observed at this site (data not shown).
      In the other one case of 10 Ahr−/− mice with cecal lesions, the lesion site was composed of only a hyperplastic component (Figure 2, A and B). Even so, Ki-67–positive cells had expanded mainly on the basal side of the gland (Figure 2E), and cecum-specific inflammatory cell infiltration was observed (Figure 2, C and F). β-Catenin nuclear translocation was also not observed (Figure 2D).
      Figure thumbnail gr2
      Figure 2A lesion site of Ahr−/− mice composed only of the hyperplastic component. A: Macroscopic appearance of the lesion site. B and C: Microscopic analysis of the lesion in A. DF: Panels show immunostaining for β-catenin (D), Ki-67 (E), and F4/80 (F). Yellow arrowheads indicate hyperplasia; orange arrowheads, adjacent mucosa in B. Scale bars: 200 μm (B); 100 μm (CE); 50 μm (F).
      Statistical analysis of immunostaining scores revealed that the Ki-67–positive ratio and the number of F4/80- and myeloperoxidase-positive cells in a given area at an Ahr−/− lesion site significantly increased relative to corresponding values in the WT ileocecal region (Figure 3, A–C ). However, the β-catenin nuclear translocation ratio was comparable at Ahr−/− lesion sites and WT ileocecal regions (Figure 3D).
      Figure thumbnail gr3
      Figure 3Statistical analyses of β-catenin, Ki-67, F4/80, and myeloperoxidase (MPO) immunostaining at Ahr−/− lesion sites and wild-type (WT) ileocecal regions. The index used to compare each antibody was calculated as described in . Data are expressed as means ± SD. n = 5 Ahr−/− mice (A); n = 6 WT mice (A) and Ahr−/− mice and WT mice (BD). *P < 0.05, **P < 0.01.
      The distal colons of 10 Ahr−/− mice with cecal lesions showed various degrees of inflammation. Of 10 cases, 2 (20%) showed severe colitis with regenerative atypia covering the distal colon (Supplemental Figure S1C). Four others (40%) showed colitis with regenerative atypia in part of the distal colon (Supplemental Figure S1D). The other four cases (40%) showed only mild colitis (Supplemental Figure S1E). Interestingly, the mucosa adjacent to the lesion site was devoid of inflammation, even in cases of severe colitis. The distal colons of WT mice and Ahr−/− mice without cecal lesions showed almost no inflammation, except for one Ahr−/− mouse with severe inflammation.
      These results indicate overall that cecal lesions of Ahr−/− mice develop from serrated hyperplasia to neoplasia with macrophage and neutrophil infiltration.

      Pro-Inflammatory Cytokines and Chemokines Are Up-Regulated at Lesion Sites in Ahr−/− Mice

      On the basis of the histologic analysis, there was much evidence that cytokines and chemokines that attract inflammatory cells and increase cell proliferation were produced at lesion sites (Figure 4A). Therefore, RT-qPCR analysis of transcripts encoding cytokines or chemokines was performed; and up-regulated Il1b and Il6 expression was observed at lesion sites of Ahr−/− mice compared with the ileocecal region of WT mice (Figure 4B). Ccl2, Cxcl5, Ccl6, and Ccl8 were also up-regulated, consistent with macrophage and neutrophil infiltration at lesion sites. Ifng, Il12p40, Il4, and Il17 cytokines, which primarily regulate lymphocyte function, were not up-regulated at lesion sites, consistent with the scarcity of lymphocytes (Figure 4B). Decreased expression of Il17 in Ahr−/− mice was consistent with previous reports.
      • Veldhoen M.
      • Hirota K.
      • Westendorf A.M.
      • Buer J.
      • Dumoutier L.
      • Renauld J.C.
      • Stockinger B.
      The aryl hydrocarbon receptor links TH17-cell-mediated autoimmunity to environmental toxins.
      ,
      • Kimura A.
      • Naka T.
      • Nohara K.
      • Fujii-Kuriyama Y.
      • Kishimoto T.
      Aryl hydrocarbon receptor regulates Stat1 activation and participates in the development of Th17 cells.
      Myc (alias c-myc) mRNA expression was then quantified to evaluate Wnt/β-catenin pathway activation, and no significant differences were observed between Ahr−/− and WT mice (Figure 4B). These results indicate overall that pro-inflammatory cytokines and chemokines are produced at lesion sites but provide no evidence of Wnt/β-catenin pathway activation.
      Figure thumbnail gr4
      Figure 4Expression of pro-inflammatory cytokines and chemokines at the lesion site of Ahr−/− mice. A: A schematic model showing development of cecal lesions in Ahr−/− mice. Cytokines and chemokines that attract macrophages and neutrophils are continuously produced because of interactions between epithelial and inflammatory cells. Proliferative activity is increased by cytokines, such as IL-1β and IL-6, and serrated hyperplasia progresses to adenoma and adenocarcinoma. B: Quantitative RT-PCR analysis of indicated transcripts at the lesion site of Ahr−/− mice and the ileocecal region of wild-type (WT) mice. ΔΔCT values and fold expression of target genes were determined by defining mRNA expression in a WT 45-week–old female mouse as 1.0. Data are expressed as means ± SD (B). n = 6 Ahr−/− mice (B); n = 7 WT mice (B). *P < 0.05, **P < 0.01. CCL2, chemokine (C-C motif) ligand 2.

      Tissues at Lesion Sites of Ahr−/− Mice Exhibit Enhanced Erk Phosphorylation

      Because significant Wnt/β-catenin pathway activation was not observed in lesions from Ahr−/− mice, other signaling pathways that might be activated in Ahr−/− tumorigenesis were studied. Cecal lesions of Ahr−/− mice developed from serrated hyperplasia to neoplasia, and proliferating cells expanded from the bottom of crypts to full thickness in the mucosa (Figures 1 and 2). This morphology was called bottom-up morphogenesis, and it resembles that seen in human colorectal serrated lesions initiated by MAPK pathway activation.
      • Kang M.
      • Mitomi H.
      • Sada M.
      • Tokumitsu Y.
      • Takahashi Y.
      • Igarashi M.
      • Katsumata T.
      • Okayasu I.
      Ki-67, p53, and Bcl-2 expression of serrated adenomas of the colon.
      This pattern differs from so-called top-down morphogenesis of human colorectal adenoma-carcinoma sequence initiated by Wnt/β-catenin pathway activation
      • Shih I.-M.
      • Wang T.-L.
      • Traverso G.
      • Romans K.
      • Hamilton S.R.
      • Ben-Sasson S.
      • Kinzler K.W.
      • Vogelstein B.
      Top-down morphogenesis of colorectal tumors.
      (Figure 5, A and B ). Therefore, MAPK pathway activation was evaluated in Ahr−/− mice on the basis of Erk phosphorylation. p-Erk was strongly positive, mainly in epithelial cells of most lesion sites in Ahr−/− mice (Figure 5, C–E), and the p-Erk h-score was significantly higher at Ahr−/− lesions relative to WT ileocecal regions (Figure 5G). p-Erk was weakly positive in a few epithelial cells in the terminal ileum and in colon mucosa adjacent to lesions (Figure 5F), but there was no significant difference between Ahr−/− and WT mice in normal intestinal mucosa (data not shown). These results suggest that MAPK pathway activation occurs primarily at lesion sites in Ahr−/− mice.
      Figure thumbnail gr5
      Figure 5Extracellular signal-regulated kinase (Erk) phosphorylation at the lesion site of Ahr−/− mice. A and B: Morphogenesis of human colorectal tumors. A: Hematoxylin and eosin staining (left panel) and immunostaining for Ki-67 (right panel) in a human tubular adenoma, representing top-down morphogenesis of the adenoma-carcinoma sequence initiated by Wnt/β-catenin pathway activation. B: Hematoxylin and eosin staining (left panel) and immunostaining for Ki-67 (right panel) in a human sessile serrated adenoma/polyp (SSA/P), representing bottom-up morphogenesis of the human serrated pathway initiated by mitogen-activated protein kinase pathway activation. CF: Immunostaining for phosphorylated Erk (p-Erk). Purple arrowheads indicate lesion sites; orange arrowheads, adjacent mucosa. G: Statistical analyses of h-score of p-Erk staining between Ahr−/− lesion sites and wild-type (WT) ileocecal regions. Data are expressed as means ± SD (G). n = 5 Ahr−/− mice (G); n = 6 WT mice (G). **P < 0.01. Scale bars: 200 μm (C and D); 100 μm (E and F). Original magnification ×100 (A and B).

      Tissues at Ahr−/− Lesion Sites Exhibit Increased Phosphorylation of Src and EGFR as well as Amphiregulin Up-Regulation

      Multiple studies suggest that MAPK activation at lesion sites in Ahr−/− mice depends on various pathways
      • Haarmann-Stemmann T.
      • Bothe H.
      • Abel J.
      Growth factors, cytokines and their receptors as downstream targets of arylhydrocarbon receptor (AhR) signaling pathways.
      • Ikuta T.
      • Kurosumi M.
      • Yatsuoka T.
      • Nishimura Y.
      Tissue distribution of aryl hydrocarbon receptor in the intestine: implication of putative roles in tumor suppression.
      • Takahashi-Tezuka M.
      • Yoshida Y.
      • Fukada T.
      • Ohtani T.
      • Yamanaka Y.
      • Nishida K.
      • Nakajima K.
      • Hibi M.
      • Hirano T.
      Gab1 acts as an adapter molecule linking the cytokine receptor gp130 to ERK mitogen-activated protein kinase.
      (Figure 6, A and B ). To investigate these mechanisms more closely, c-Src and EGFR phosphorylation was evaluated at Ahr−/− lesion sites using immunohistochemistry and Amphiregulin expression was examined by RT-qPCR analysis.
      Figure thumbnail gr6
      Figure 6Src phosphorylation, epidermal growth factor receptor (EGFR) phosphorylation, and amphiregulin expression at lesion sites of Ahr−/− mice. A: Schematic model showing cross talk between AhR and EGFR signaling. A cytosolic multiprotein complex including c-Src prevents AhR nuclear translocation. Ligand binding to AhR leads to the dissociation of a cytosolic multiprotein complex and subsequent release of c-Src, which moves to the cell membrane. Then, c-Src is phosphorylated by the interaction with EGFR, and activated c-Src phosphorylates EGFR conversely. B: Schematic model showing mitogen-activated protein kinase pathway activation after AhR knockdown/knockout. AhR loss frees c-Src from the cytosolic multiprotein complex, allowing c-Src to move to the membrane and phosphorylate EGFR, similar to AhR ligand binding shown in A. C and D: Immunostaining for p-Src. E: Statistical analyses of p-Src positive ratio at Ahr−/− lesion sites and wild-type (WT) ileocecal regions. F and G: Immunostaining for phosphorylated EGFR (p-EGFR). H: Statistical analyses of p-EGFR–positive ratio at Ahr−/− lesion sites and WT ileocecal regions. I: Quantitative RT-PCR analysis of Amphiregulin expression at lesion sites of Ahr−/− mice and ileocecal regions of WT mice. ΔΔCT values and fold expression were determined by defining mRNA expression in a WT 45-week–old female mice as 1.0. J: Immunostaining for phosphorylated Stat3 (p-Stat3). K: Statistical analyses of the number of p-Stat3–positive cells at Ahr−/− lesion sites and WT ileocecal regions. Data are expressed as means ± SD (E, H, I, and K). n = 5 Ahr−/− mice (E and H); n = 6 WT mice (E and H), Ahr−/− mice (I), and Ahr−/− mice and WT mice (K); n = 7 WT mice (I). *P < 0.05, **P < 0.01. Scale bars = 50 μm (C, D, F, G, and J). ERK1/2, extracellular signal-regulated kinase 1/2; FICZ, 5,11-Dihydroindolo[3,2-b]carbazole-6-carboxaldehyde; MEK, MAPK/ERK kinase; TCDD, 2,3,7,8-tetrachlorobenzo-p-dioxin.
      Phosphorylated Src protein was observed in some epithelial cells at lesion sites of Ahr−/− mice at levels higher than that seen in adjacent mucosa (Figure 6C). Inflammatory cells present in that tissue also showed phosphorylated Src protein, although this finding could be explained by cross-reactivity of the antibody to phosphorylated Hck, Lyn, and Fyn (Figure 6C). A small number of epithelial cells in the terminal ileum and adjacent colon mucosa to lesions was also positive for phosphorylated Src (Figure 6D). The p-Src–positive ratio at Ahr−/− lesion sites was significantly increased compared with that of WT ileocecal regions (Figure 6E), although that ratio was comparable in adjacent intestinal mucosa from both Ahr−/− and WT mice (data not shown). Some epithelial and inflammatory cells were positive for p-EGFR at Ahr−/− lesion sites, whereas few epithelial cells in the terminal ileum and adjacent colon mucosa were positive for p-EGFR. p-EGFR distribution was similar to that of p-Src (Figure 6, F and G), and the p-EGFR–positive ratio at Ahr−/− lesion sites was significantly higher than that in the WT ileocecal regions (Figure 6H). Amphiregulin expression at Ahr−/− lesion sites was also up-regulated relative to the WT ileocecal regions (Figure 6I).
      Because Il6 up-regulation at Ahr−/− lesion sites (Figure 4B) likely underlies MAPK activation, phosphorylation of Stat3, a downstream IL-6 target that promotes cell proliferation and invasion,
      • Bollrath J.
      • Phesse T.J.
      • von Burstin V.A.
      • Putoczki T.
      • Bennecke M.
      • Bateman T.
      • Nebelsiek T.
      • Lundgren-May T.
      • Canli Ö.
      • Schwitalla S.
      • Matthews V.
      • Schmid R.M.
      • Kirchner T.
      • Arkan M.C.
      • Ernst M.
      • Greten F.R.
      gp130-Mediated Stat3 activation in enterocytes regulates cell survival and cell-cycle progression during colitis-associated tumorigenesis.
      ,
      • Grivennikov S.
      • Karin E.
      • Terzic J.
      • Mucida D.
      • Yu G.Y.
      • Vallabhapurapu S.
      • Scheller J.
      • Rose-John S.
      • Cheroutre H.
      • Eckmann L.
      • Karin M.
      IL-6 and Stat3 are required for survival of intestinal epithelial cells and development of colitis-associated cancer.
      was investigated. In some Ahr−/− mice, phosphorylated Stat3 was weakly positive, mainly in inflammatory cells but in a few epithelial cells at lesion sites (Figure 6J), as previously reported
      • Ikuta T.
      • Kobayashi Y.
      • Kitazawa M.
      • Shiizaki K.
      • Itano N.
      • Noda T.
      • Pettersson S.
      • Poellinger L.
      • Fujii-Kuriyama Y.
      • Taniguchi S.
      • Kawajiri K.
      ASC-associated inflammation promotes cecal tumorigenesis in aryl hydrocarbon receptor-deficient mice.
      ; and in some Ahr−/− mice, there were few or almost no phosphorylated Stat3–positive inflammatory and epithelial cells at lesion sites. The number of cells positive for phosphorylated Stat3 at Ahr−/− lesion sites was somewhat higher than at WT ileocecal regions, but the difference was not statistically significant (Figure 6K).
      The mouse BRAFV637E mutation in exon 18, which occurs at an orthologous position of the human BRAFV600E mutation in exon 15, reportedly induces intestinal hyperplasia and subsequent dysplasia and adenocarcinoma in mice, phenotypes that resemble the human serrated pathway.
      • Rad R.
      • Cadiñanos J.
      • Rad L.
      • Varela I.
      • Strong A.
      • Kriegl L.
      • Constantino-casas F.
      • Eser S.
      • Hieber M.
      • Seidler B.
      • Price S.
      • Fraga M.F.
      • Calvanese V.
      • Hoffman G.
      • Ponstingl H.
      • Schneider G.
      • Yusa K.
      • Grove C.
      • Schmid R.M.
      • Wang W.
      • Vassiliou G.
      • Kirchner T.
      • Mcdermott U.
      • Liu P.
      • Saur D.
      • Bradley A.
      A genetic progression model of Braf(V600E)-induced intestinal tumorigenesis reveals targets for therapeutic intervention.
      To investigate whether Braf mutation induces MAPK activation and cecal tumorigenesis in Ahr−/− mice, direct sequencing was used to search for BRAFV637E mutations at lesion sites in four Ahr−/− mice but did not detect any (Supplemental Figure S2).
      These results overall suggest that MAPK activation at Ahr−/− lesion sites is likely enhanced by synergistic EGFR phosphorylation via Src phosphorylation, Amphiregulin up-regulation, and IL-6 production.

      LPS-Induced IL-6 Production Increases in Peritoneal Macrophages of Ahr−/− Mice

      Because adjacent mucosal tissues did not show significant Erk, Src, and EGFR phosphorylation in Ahr−/− mice, it was hypothesized that MAPK signaling seen in lesion tissues of Ahr−/− mice might depend on inflammation at the lesion site. To investigate potential functional differences between Ahr−/− and WT macrophages, cytokine and chemokine production was examined by peritoneal macrophages after LPS + ATP stimulation. IL-6 production increased in Ahr−/− macrophages relative to WT macrophages (Figure 7A). Although it fluctuated among samples, no significant difference was observed in IL-1β production between Ahr−/− and WT mice (Figure 7A). CCL2 production was slightly elevated in WT macrophages 6 hours after LPS + ATP stimulation, but no significant difference was observed in TNF production between Ahr−/− and WT mice (Figure 7A).
      Figure thumbnail gr7
      Figure 7Cytokine and chemokine production by mouse peritoneal macrophages. Production of indicated cytokines/chemokines by peritoneal macrophages of wild-type (WT) and Ahr−/− mice after lipopolysaccharide (LPS) + ATP stimulation (A) or Fusobacterium nucleatum infection (B). Fold production is calculated by setting the mean value after 6 hours of LPS stimulation at 500 ng/mL (A) or 6 hours of infection with F. nucleatum at a multiplicity of infection (MOI) of 5 (B) of WT mice in each experiment to 1.0. Data are expressed as means ± SD (A and B). n = 4 independent experiments (A); n = 3 independent experiments (B). *P < 0.05, **P < 0.01, and ***P < 0.001. CCL2, chemokine (C-C motif) ligand 2; TNF, tumor necrosis factor.
      A potential explanation for our observation of decreased tumor incidence could be differences in gut microbiota between different animal facilities. To investigate a potential relationship between gut microbiota and intestinal inflammation, F. nucleatum, a bacteria associated with colorectal carcinogenesis,
      • Dejea C.M.
      • Wick E.C.
      • Hechenbleikner E.M.
      • White J.R.
      • Mark Welch J.L.
      • Rossetti B.J.
      • Peterson S.N.
      • Snesrud E.C.
      • Borisy G.G.
      • Lazarev M.
      • Stein E.
      • Vadivelu J.
      • Roslani A.C.
      • Malik A.A.
      • Wanyiri J.W.
      • Goh K.L.
      • Thevambiga I.
      • Fu K.
      • Wan F.
      • Llosa N.
      • Housseau F.
      • Romans K.
      • Wu X.
      • McAllister F.M.
      • Wu S.
      • Vogelstein B.
      • Kinzler K.W.
      • Pardoll D.M.
      • Sears C.L.
      Microbiota organization is a distinct feature of proximal colorectal cancers.
      • Kostic A.D.
      • Gevers D.
      • Pedamallu C.S.
      • Michaud M.
      • Duke F.
      • Earl A.M.
      • Ojesina A.I.
      • Jung J.
      • Bass A.J.
      • Tabernero J.
      • Baselga J.
      • Liu C.
      • Shivdasani R.A.
      • Ogino S.
      • Birren B.W.
      • Huttenhower C.
      • Garrett W.S.
      • Meyerson M.
      Genomic analysis identifies association of Fusobacterium with colorectal carcinoma.
      • Castellarin M.
      • Warren R.L.
      • Freeman J.D.
      • Dreolini L.
      • Krzywinski M.
      • Strauss J.
      • Barnes R.
      • Watson P.
      • Allen-Vercoe E.
      • Moore R.A.
      • Holt R.A.
      Fusobacterium nucleatum infection is prevalent in human colorectal carcinoma.
      • Yu J.
      • Chen Y.
      • Fu X.
      • Zhou X.
      • Peng Y.
      • Shi L.
      • Chen T.
      • Wu Y.
      Invasive Fusobacterium nucleatum may play a role in the carcinogenesis of proximal colon cancer through the serrated neoplasia pathway.
      was used to stimulate peritoneal macrophages from Ahr−/− and WT mice. IL-6 production increased in Ahr−/− macrophages by F. nucleatum infection similarly to LPS + ATP stimulation, but changes were not statistically significant, possibly due to large fluctuations in values (Figure 7B). IL-1β, CCL2, and TNF production was comparable in Ahr−/− and WT mice (Figure 7B). IL-6 production also increased in Ahr−/− macrophages after E. coli infection, but the difference was not statistically significant (data not shown). These results suggest that the mechanisms that promote IL-6 production in Ahr−/− macrophages explain functional differences between Ahr−/− and WT macrophages.

      AhR Is Up-Regulated in Human Colorectal Carcinomas as Lesions Progress from Benign Precursors in Both Adenoma-Carcinoma Sequence and Sessile Serrated Pathway

      AhR expression in epithelial cells is reportedly elevated in human colorectal carcinoma relative to normal epithelium.
      • Ikuta T.
      • Kurosumi M.
      • Yatsuoka T.
      • Nishimura Y.
      Tissue distribution of aryl hydrocarbon receptor in the intestine: implication of putative roles in tumor suppression.
      However, given that morphology and activated signaling pathways were similar in Ahr−/− cecal lesions and human colorectal serrated lesions, it was studied whether AhR down-regulation occurred differently in tumorigenesis in the serrated pathway versus the adenoma-carcinoma sequence. Immunohistochemistry was used to assess AhR expression in human colorectal carcinomas and benign precursors of both pathways. In normal colorectal mucosa, primarily interstitial cells were AhR positive, whereas epithelial cells were negative (Figure 8A), as previously described.
      • Ikuta T.
      • Kurosumi M.
      • Yatsuoka T.
      • Nishimura Y.
      Tissue distribution of aryl hydrocarbon receptor in the intestine: implication of putative roles in tumor suppression.
      In the adenoma-carcinoma sequence, AhR was weakly positive in epithelial cells of tubular adenoma and adenocarcinoma, and AhR was up-regulated gradually as lesions progressed from tubular adenoma to adenocarcinoma (Figure 8, B and D). In the sessile serrated pathway, AhR was weakly positive in epithelial cells of SSA/P and adenocarcinoma and up-regulated gradually as lesions progressed from SSA/P to adenocarcinoma (Figure 8, C and D). These findings indicate that AhR expression in epithelial cells is elevated in human colorectal carcinomas as lesions progress in both the adenoma-carcinoma sequence and the serrated pathway.
      Figure thumbnail gr8
      Figure 8AhR immunostaining of human colorectal carcinomas and benign precursors in both the adenoma-carcinoma sequence and the sessile serrated pathway. AC: AhR immunostaining in human normal colon mucosa (A), the adenoma-carcinoma sequence (B), and the sessile serrated pathway (C). D: Statistical analyses of the average h-score of AhR staining between normal colon mucosa, the benign precursor component, and the adenocarcinoma component in the two pathways. **P < 0.01. Scale bars = 100 μm (AC). SSA/P, sessile serrated adenoma/polyp.

      Discussion

      This study revealed that cecal lesions of Ahr−/− mice developed from serrated hyperplasia to neoplasia with macrophage and neutrophil infiltration. The morphology of these lesions was similar to that of human colorectal serrated lesions. Tumorigenesis depended on persistent chemokine and cytokine production, resulting from interactions between epithelial and inflammatory cells, activation of MAPK pathway after AhR knockout, and production of cytokines, such as IL-6. These results suggest novel mechanisms of Ahr−/− cecal tumorigenesis. By contrast, AhR expression did not decrease during tumorigenesis in both the human adenoma-carcinoma sequence and serrated pathway.
      Previous reports showed that most Ahr−/− mice developed cecal lesions by 10 weeks of age, and approximately half of DKO mice developed cecal lesions by 50 weeks of age.
      • Kawajiri K.
      • Kobayashi Y.
      • Ohtake F.
      • Ikuta T.
      • Matsushima Y.
      • Mimura J.
      • Pettersson S.
      • Pollenz R.S.
      • Sakaki T.
      • Hirokawa T.
      • Akiyama T.
      • Kurosumi M.
      • Poellinger L.
      • Kato S.
      • Fujii-Kuriyama Y.
      Aryl hydrocarbon receptor suppresses intestinal carcinogenesis in ApcMin/+ mice with natural ligands.
      ,
      • Ikuta T.
      • Kobayashi Y.
      • Kitazawa M.
      • Shiizaki K.
      • Itano N.
      • Noda T.
      • Pettersson S.
      • Poellinger L.
      • Fujii-Kuriyama Y.
      • Taniguchi S.
      • Kawajiri K.
      ASC-associated inflammation promotes cecal tumorigenesis in aryl hydrocarbon receptor-deficient mice.
      Relative to that, tumor incidence of Ahr−/− mice in our study was significantly reduced (Figure 1A). This reduction may be due to differences in gut microbiota and resultant different susceptibility to intestinal inflammation among animal facilities. Further investigations, such as bacterial transplantation experiments, will be necessary to confirm this possibility. In addition, Wnt/β-catenin pathway activation, which may also account for reduced tumor incidence, was not observed. Several analyses report differing results regarding whether β-catenin degradation occurs through AhR,
      • Ohtake F.
      • Baba A.
      • Takada I.
      • Okada M.
      • Iwasaki K.
      • Miki H.
      • Takahashi S.
      • Kouzmenko A.
      • Nohara K.
      • Chiba T.
      • Fujii-Kuriyama Y.
      • Kato S.
      Dioxin receptor is a ligand-dependent E3 ubiquitin ligase.
      ,
      • Kawajiri K.
      • Kobayashi Y.
      • Ohtake F.
      • Ikuta T.
      • Matsushima Y.
      • Mimura J.
      • Pettersson S.
      • Pollenz R.S.
      • Sakaki T.
      • Hirokawa T.
      • Akiyama T.
      • Kurosumi M.
      • Poellinger L.
      • Kato S.
      • Fujii-Kuriyama Y.
      Aryl hydrocarbon receptor suppresses intestinal carcinogenesis in ApcMin/+ mice with natural ligands.
      ,
      • Braeuning A.
      • Köhle C.
      • Buchmann A.
      • Schwarz M.
      Coordinate regulation of cytochrome P450 1A1 expression in mouse liver by the aryl hydrocarbon receptor and the β-catenin pathway.
      • Jin U.-H.
      • Lee S.-O.
      • Sridharan G.
      • Lee K.
      • Davidson L.A.
      • Jayaraman A.
      • Chapkin R.S.
      • Alaniz R.
      • Safe S.
      Microbiome-derived tryptophan metabolites and their aryl hydrocarbon receptor-dependent agonist and antagonist activities.
      • Kasai S.
      • Ishigaki T.
      • Takumi R.
      • Kamimura T.
      • Kikuchi H.
      β-Catenin signaling induces CYP1A1 expression by disrupting adherens junctions in Caco-2 human colon carcinoma cells.
      and further analysis is needed to determine under what conditions Wnt/β-catenin pathway functions in Ahr−/− cecal tumorigenesis.
      Cecal lesions of most Ahr−/− mice were composed of tumor and hyperplastic components, although one lesion was composed of the hyperplastic component only. Proliferative capacity was enhanced in the tumor component relative to the hyperplastic component (Figure 1, C–H, and Figure 2). These histologic features indicate that serrated hyperplasia occurs initially and progresses to neoplasia by additional mutations or chromosomal aberrations. Macrophage and neutrophil infiltration was observed early in serrated hyperplasia (Figure 1, C–H, and Figure 2), and Il1b, Il6, Ccl2, and Cxcl5 were up-regulated, consistent with persistent inflammation seen at the lesion site (Figure 4B). These results strongly suggest that progression from serrated hyperplasia to neoplasia occurs because of persistent chemokine/cytokine production brought on by interactions between epithelial and inflammatory cells (Figure 4A). The distal colons of most Ahr−/− mice with cecal lesions showed various degrees of inflammation (Supplemental Figure S1, C–E). Therefore, 10 Ahr−/− mice that are especially susceptible to inflammation may simultaneously develop cecal lesions and inflammation over the entire colon. However, some Ahr−/− mice with cecal lesions showed only mild colitis, and the mucosa adjacent to the lesion site was devoid of inflammation, even in the case with severe colitis (Figure 1, C–H, and Supplemental Figure S1, C–E). These histologic features indicate that cecal lesions of Ahr−/− mice develop through cecum-specific recruitment of inflammatory cells rather than from randomly generated dysplasia in the inflamed colon, and that inflammation of the distal colon could be a secondary or concomitant change rather than a direct cause of cecal lesions.
      Correspondence between genetic alterations and morphology seen in human colorectal carcinoma has been reported in mouse models of colorectal cancer as well. Apc-mutant or deficient mice, such as Apcmin/+ mice,
      • Bilger A.
      • Shoemaker A.R.
      • Gould K.A.
      • Dove W.F.
      Manipulation of the mouse germline in the study of Min-induced neoplasia.
      develop intestinal adenomas morphologically similar to so-called top-down morphogenesis of the human adenoma-carcinoma sequence
      • Shih I.-M.
      • Wang T.-L.
      • Traverso G.
      • Romans K.
      • Hamilton S.R.
      • Ben-Sasson S.
      • Kinzler K.W.
      • Vogelstein B.
      Top-down morphogenesis of colorectal tumors.
      ,
      • Oshima H.
      • Oshima M.
      • Kobayashi M.
      • Tsutsumi M.
      • Taketo M.M.
      Morphological and molecular processes of polyp formation in ApcΔ716 knockout mice.
      ,
      • Nalbantoglu I.
      • Blanc V.
      • Davidson N.O.
      Characterization of colorectal cancer development in Apcmin/+ mice.
      (Figure 5A). Enterocyte-specific knock in of oncogenic KRASG12D or BRAFV637E in mice leads to colonic serrated lesions with bottom-up morphogenesis, similar to human serrated lesions
      • Kang M.
      • Mitomi H.
      • Sada M.
      • Tokumitsu Y.
      • Takahashi Y.
      • Igarashi M.
      • Katsumata T.
      • Okayasu I.
      Ki-67, p53, and Bcl-2 expression of serrated adenomas of the colon.
      ,
      • Rad R.
      • Cadiñanos J.
      • Rad L.
      • Varela I.
      • Strong A.
      • Kriegl L.
      • Constantino-casas F.
      • Eser S.
      • Hieber M.
      • Seidler B.
      • Price S.
      • Fraga M.F.
      • Calvanese V.
      • Hoffman G.
      • Ponstingl H.
      • Schneider G.
      • Yusa K.
      • Grove C.
      • Schmid R.M.
      • Wang W.
      • Vassiliou G.
      • Kirchner T.
      • Mcdermott U.
      • Liu P.
      • Saur D.
      • Bradley A.
      A genetic progression model of Braf(V600E)-induced intestinal tumorigenesis reveals targets for therapeutic intervention.
      ,
      • Bennecke M.
      • Kriegl L.
      • Bajbouj M.
      • Retzlaff K.
      • Robine S.
      • Jung A.
      • Arkan M.C.
      • Kirchner T.
      • Greten F.R.
      Ink4a/Arf and oncogene-induced senescence prevent tumor progression during alternative colorectal tumorigenesis.
      (Figure 5B). These reports show that molecular and morphologic changes seen in mouse colorectal tumorigenesis resemble those occurring in humans via these two pathways. We found enhanced Erk phosphorylation mainly at lesion sites of Ahr−/− mice on the basis of the morphologic similarity between Ahr−/− cecal lesions and human colorectal serrated lesions (Figure 5, C–G). These results suggest that MAPK pathway activation is associated with bottom-up morphology in mice colitis-associated carcinogenesis as it is in carcinogenesis of KRASG12D or BRAFV637E knock-in mouse models. On the basis of the morphologic similarity between mouse and human colorectal tumors, MAPK pathway activation may also be associated with bottom-up morphology in human colitis-associated carcinogenesis as it is in human serrated pathway.
      Numerous mechanisms likely underlie MAPK pathway activation after AhR loss. A cytosolic multiprotein complex including c-Src reportedly prevents AhR nuclear translocation.
      • Haarmann-Stemmann T.
      • Bothe H.
      • Abel J.
      Growth factors, cytokines and their receptors as downstream targets of arylhydrocarbon receptor (AhR) signaling pathways.
      Ligand binding to AhR triggers dissociation of a cytoplasmic multiprotein complex and subsequent release of c-Src, which moves to the membrane. Then, c-Src is phosphorylated,
      • Xie G.
      • Peng Z.
      • Raufman J.-P.
      Src-mediated aryl hydrocarbon and epidermal growth factor receptor cross talk stimulates colon cancer cell proliferation.
      presumably by the interaction with EGFR
      • Belsches A.P.
      • Haskell M.D.
      • Parsons S.J.
      Role of c-Src tyrosine kinase in EGF-induced mitogenesis.
      ; and activated c-Src phosphorylates EGFR conversely
      • Haarmann-Stemmann T.
      • Bothe H.
      • Abel J.
      Growth factors, cytokines and their receptors as downstream targets of arylhydrocarbon receptor (AhR) signaling pathways.
      (Figure 6A). AhR loss seems to free c-Src and to enhance EGFR phosphorylation, as is seen after AhR ligand binding (Figure 6B). On the basis of this notion, enhanced Src phosphorylation was found primarily at lesion sites of Ahr−/− mice (Figure 6, C and D). Others have reported that AhR knockdown in colon cancer cells up-regulates genes, including amphiregulin, an EGF family member
      • Ikuta T.
      • Kurosumi M.
      • Yatsuoka T.
      • Nishimura Y.
      Tissue distribution of aryl hydrocarbon receptor in the intestine: implication of putative roles in tumor suppression.
      ; and it is well known that IL-6 induces MAPK pathway activation through gp130 and SHP2.
      • Takahashi-Tezuka M.
      • Yoshida Y.
      • Fukada T.
      • Ohtani T.
      • Yamanaka Y.
      • Nishida K.
      • Nakajima K.
      • Hibi M.
      • Hirano T.
      Gab1 acts as an adapter molecule linking the cytokine receptor gp130 to ERK mitogen-activated protein kinase.
      Because activation of MAPK signaling, Src, and EGFR phosphorylation as well as Amphiregulin up-regulation were seen mainly at lesion sites and not to a significant extent in adjacent mucosa, and no Braf mutations were detected (Figures 5 and 6 and Supplemental Figure S2), it was concluded that MAPK activation is enhanced synergistically by these mechanisms, rather than only by AhR loss in epithelial cells (Figure 6B). A similar distribution of p-EGFR– and p-Src–positive cells at Ahr−/− lesion sites (Figure 6, C–H) supports this idea.
      EGFR phosphorylation was also observed in inflammatory cells at Ahr−/− lesion sites, suggesting that similar mechanisms underlie EGFR phosphorylation in both epithelial and inflammatory cells. However, EGFR is reportedly phosphorylated in inflammatory cells during human gastric carcinogenesis and in mice gastric Helicobacter pylori infection model.
      • Hardbower D.M.
      • Singh K.
      • Asim M.
      • Verriere T.G.
      • Olivares-Villagómez D.
      • Barry D.P.
      • Allaman M.M.
      • Washington Jr., M.K.
      • Peek R.M.
      • Piazuelo M.B.
      • Wilson K.T.
      EGFR regulates macrophage activation and function in bacterial infection.
      Moreover, EGFR is reportedly phosphorylated in inflammatory cells in human inflammatory bowel disease (IBD)–associated carcinogenesis and in a mouse model of azoxymethane-dextran sodium sulfate colitis–associated carcinogenesis.
      • Hardbower D.M.
      • Coburn L.A.
      • Asim M.
      • Singh K.
      • Sierra J.C.
      • Barry D.P.
      • Gobert A.P.
      • Piazuelo M.B.
      • Washington M.K.
      • Wilson K.T.
      EGFR-mediated macrophage activation promotes colitis-associated tumorigenesis.
      In these human gastrointestinal carcinogenesis and mice models, AhR does not always seem to be down-regulated. Therefore, EGFR phosphorylation in inflammatory cells seems to be an important process in human and mice gastrointestinal inflammation or inflammation-associated carcinogenesis, which occurs without necessarily AhR down-regulation.
      IL-6 production by Ahr−/− peritoneal macrophages increased after LPS + ATP stimulation compared with WT macrophages, but IL-1β, CCL2, and TNF production did not (Figure 7). Ahr knockout reportedly enhances IL-6 and TNF production by peritoneal macrophages after LPS stimulation.
      • Kimura A.
      • Naka T.
      • Nakahama T.
      • Chinen I.
      • Masuda K.
      • Nohara K.
      • Fujii-Kuriyama Y.
      • Kishimoto T.
      Aryl hydrocarbon receptor in combination with Stat1 regulates LPS-induced inflammatory responses.
      ,
      • Masuda K.
      • Kimura A.
      • Hanieh H.
      • Nguyen N.T.
      • Nakahama T.
      • Chinen I.
      • Otoyo Y.
      • Murotani T.
      • Yamatodani A.
      • Kishimoto T.
      Aryl hydrocarbon receptor negatively regulates LPS-induced IL-6 production through suppression of histamine production in macrophages.
      In those reports, AhR-deficient mice all died within 5 weeks of age under conventional conditions, and all mice were maintained under specific pathogen-free conditions. However, in this study, Ahr−/− mice were maintained under conventional conditions, and only some Ahr−/− mice died within several weeks after birth. Therefore, difference of the environment used to house the mice and resultant macrophage differentiation between the facilities may underlie variations in TNF production. Others showed that AhR-deficient mice were hypersensitive to LPS-induced septic shock, and bone marrow–derived macrophages of AhR-deficient mice produced relatively high levels of IL-1β due to reduced plasminogen activator inhibitor-2 expression.
      • Sekine H.
      • Mimura J.
      • Oshima M.
      • Okawa H.
      • Kanno J.
      • Igarashi K.
      • Gonzalez F.J.
      • Ikuta T.
      • Kawajiri K.
      • Fujii-Kuriyama Y.
      Hypersensitivity of aryl hydrocarbon receptor-deficient mice to lipopolysaccharide-induced septic shock.
      Differences in IL-1β production may also be due to the environment used to house the mice. Moreover, in that report, investigators stimulated macrophages with LPS only to examine IL-1β production. Although monocytes (and presumably some immature macrophages) reportedly produce IL-1β after LPS stimulation as they release ATP,
      • Netea M.G.
      • Nold-Petry C.A.
      • Nold M.F.
      • Joosten L.A.B.
      • Opitz B.
      • van der Meer J.H.M.
      • van de Veerdonk F.L.
      • Ferwerda G.
      • Heinhuis B.
      • Devesa I.
      • Funk C.J.
      • Mason R.J.
      • Kullberg B.J.
      • Rubartelli A.
      • van der Meer J.W.M.
      • Dinarello C.A.
      Differential requirement for the activation of the inflammasome for processing and release of IL-1β in monocytes and macrophages.
      • Piccini A.
      • Carta S.
      • Tassi S.
      • Lasiglie D.
      • Fossati G.
      • Rubartelli A.
      ATP is released by monocytes stimulated with pathogen-sensing receptor ligands and induces IL-1β and IL-18 secretion in an autocrine way.
      • Elliott E.I.
      • Sutterwala F.S.
      Monocytes take their own path to IL-1β.
      IL-1β production via this mechanism may differ somewhat from that seen in mature macrophages with ATP stimulation, which activate inflammasomes overtly. Taken together, these findings suggest that IL-6 production is a key functional difference between Ahr−/− and WT macrophages, and IL-1β, TNF, and CCL2 are up-regulated at lesion sites mainly by persistent inflammation itself.
      It was found that AhR is already up-regulated in SSA/P, precursor lesions of human sessile serrated pathway initiated by activating BRAF mutations (Figure 8). Previous reports showed that AhR was up-regulated in papillary thyroid carcinoma after establishment of activating BRAF mutations.
      • Mian C.
      • Ceccato F.
      • Barollo S.
      • Watutantrige-Fernando S.
      • Albiger N.
      • Regazzo D.
      • De Lazzari P.
      • Pennelli G.
      • Rotondi S.
      • Nacamulli D.
      • Pelizzo M.R.
      • Jaffrain-Rea M.L.
      • Grimaldi F.
      • Occhi G.
      • Scaroni C.
      AHR over-expression in papillary thyroid carcinoma: clinical and molecular assessments in a series of Italian acromegalic patients with a long-term follow-up.
      ,
      • Occhi G.
      • Barollo S.
      • Regazzo D.
      • Bertazza L.
      • Galuppini F.
      • Guzzardo V.
      • Jaffrain-Rea M.L.
      • Vianello F.
      • Ciato D.
      • Ceccato F.
      • Watutantrige-Fernando S.
      • Bisognin A.
      • Bortoluzzi S.
      • Pennelli G.
      • Boscaro M.
      • Scaroni C.
      • Mian C.
      A constitutive active MAPK/ERK pathway due to BRAFV600E positively regulates AHR pathway in PTC.
      However, this study shows that AhR is up-regulated in a human adenoma-carcinoma sequence in the likely absence of BRAF mutations (Figure 8). It is well known that activating KRAS mutation is an essential step for progression after APC mutations in this pathway. In addition, others reported that AhR was up-regulated in tumor tissue of lung
      • Lin P.
      • Chang H.
      • Tsai W.-T.
      • Wu M.-H.
      • Liao Y.-S.
      • Chen J.-T.
      • Su J.-M.
      Overexpression of aryl hydrocarbon receptor in human lung carcinomas.
      and pancreatic cancer,
      • Koliopanos A.
      • Kleeff J.
      • Xiao Y.
      • Safe S.
      • Zimmermann A.
      • Büchler M.W.
      • Friess H.
      Increased arylhydrocarbon receptor expression offers a potential therapeutic target for pancreatic cancer.
      both of which frequently exhibit activating EGFR or KRAS mutations, and in HER2-overexpressing breast cancer cells.
      • Zhao S.
      • Ohara S.
      • Kanno Y.
      • Midorikawa Y.
      • Nakayama M.
      • Makimura M.
      • Park Y.
      • Inouye Y.
      HER2 overexpression-mediated inflammatory signaling enhances mammosphere formation through up-regulation of aryl hydrocarbon receptor transcription.
      Therefore, AhR is likely up-regulated downstream of MAPK signaling and prevents tumorigenesis in these two pathways. In Ahr−/− mice, the MAPK pathway is apparently activated by AhR knockout, whereas in the human serrated pathway and the adenoma-carcinoma sequence, AhR is likely up-regulated by BRAF or KRAS mutation and subsequent MAPK pathway activation. However, MAPK signaling is activated initially in Ahr−/− mice and in the human serrated pathway, whereas in the human adenoma-carcinoma sequence, MAPK pathway is activated in the middle phase of carcinogenesis by KRAS mutation after the initial APC mutation.
      • Morkel M.
      • Riemer P.
      • Bläker H.
      • Sers C.
      Similar but different: distinct roles for KRAS and BRAF oncogenes in colorectal cancer development and therapy resistance.
      The morphology of lesions in these cases is likely determined by the initially activated signaling pathways. Therefore, the lesions in Ahr−/− mice and in the human serrated pathway likely show similar bottom-up serrated morphology that is somewhat different from top-down tubular morphology in the human adenoma-carcinoma sequence.
      In patients with IBD, such as ulcerative colitis or Crohn disease, colorectal carcinomas develop through a dysplasia-carcinoma sequence,
      • Warren S.
      • Sommers S.C.
      Pathogenesis of ulcerative colitis.
      a pathway of colitis-associated carcinogenesis due to chronic inflammation and resultant dysplasia formation. A classification system for subtypes of dysplasia in IBD has been proposed,
      • Riddell R.H.
      • Goldman H.
      • Ransohoff D.F.
      • Appelman H.D.
      • Fenoglio C.M.
      • Haggitt R.C.
      • Ahren C.
      • Correa P.
      • Hamilton S.R.
      • Morson B.C.
      • Sommers S.C.
      • Yardley J.H.
      Dysplasia in inflammatory bowel disease: standardized classification with provisional clinical applications.
      but the genetic basis of variations in dysplasia morphology has not been studied in detail. AhR is reported a newly identified candidate gene associated with IBD pathogenesis.
      • Liu J.Z.
      • Van Sommeren S.
      • Huang H.
      • Ng S.C.
      • Alberts R.
      • Takahashi A.
      • Ripke S.
      • Lee J.C.
      • Jostins L.
      • Shah T.
      • Abedian S.
      • Cheon J.H.
      • Cho J.
      • Daryani N.E.
      • Franke L.
      • Fuyuno Y.
      • Hart A.
      • Juyal R.C.
      • Juyal G.
      • Kim W.H.
      • Morris A.P.
      • Poustchi H.
      • Newman W.G.
      • Midha V.
      • Orchard T.R.
      • Vahedi H.
      • Sood A.
      • Sung J.Y.
      • Malekzadeh R.
      • Westra H.J.
      • Yamazaki K.
      • Yang S.K.
      • Barrett J.C.
      • Franke A.
      • Alizadeh B.Z.
      • Parkes M.
      • Thelma B.K.
      • Daly M.J.
      • Kubo M.
      • Anderson C.A.
      • Weersma R.K.
      International Multiple Sclerosis Genetics ConsortiumInternational IBD Genetics Consortium
      Association analyses identify 38 susceptibility loci for inflammatory bowel disease and highlight shared genetic risk across populations.
      Therefore, further investigation is needed to determine whether AhR down-regulation leads to IBD-associated dysplasia or carcinoma, similar to cecal lesions in AhR−/− mice.

      Acknowledgments

      We thank Dr. Togo Ikuta for offering aryl hydrocarbon receptor–deficient mice and Dr. Elise Lamar for critical review and editing of the manuscript.

      Supplemental Data

      Figure thumbnail figs1
      Supplemental Figure S1Histologic and immunohistochemical analysis of cecal tumorigenesis in AhR−/− mice. A and B: Focal adenocarcinoma component of an AhR−/− cecal lesion (A) and immunostaining for β-catenin (B). CE: Distal colons of AhR−/− mice with cecal lesions. There were various degrees of inflammation. Scale bars: 100 μm (A); 50 μm (B); 200 μm (CE). Original magnification: ×100 (A); ×200 (B); ×40 (CE).
      Figure thumbnail figs2
      Supplemental Figure S2Representative DNA sequencing of mouse BRAF exon 18. BrafV637E mutations at lesion sites in four AhR−/− mice were screened by direct sequencing, and no mutations were found.

      References

        • Nguyen L.P.
        • Bradfield C.A.
        The search for endogenous activators of the aryl hydrocarbon receptor.
        Chem Res Toxicol. 2008; 21: 102-116
        • Nakatsuru Y.
        • Wakabayashi K.
        • Fujii-Kuriyama Y.
        • Ishikawa T.
        • Kusama K.
        • Ide F.
        Dibenzo[A,L]pyrene-induced genotoxic and carcinogenic responses are dramatically suppressed in aryl hydrocarbon receptor-deficient mice.
        Int J Cancer. 2004; 112: 179-183
        • Shimizu Y.
        • Nakatsuru Y.
        • Ichinose M.
        • Takahashi Y.
        • Kume H.
        • Mimura J.
        • Fujii-Kuriyama Y.
        • Ishikawa T.
        Benzo[a]pyrene carcinogenicity is lost in mice lacking the aryl hydrocarbon receptor.
        Proc Natl Acad Sci U S A. 2000; 97: 779-782
        • Mimura J.
        • Yamashita K.
        • Nakamura K.
        • Morita M.
        • Takagi T.N.
        • Nakao K.
        • Ema M.
        • Sogawa K.
        • Yasuda M.
        • Katsuki M.
        • Fujii-Kuriyama Y.
        Loss of teratogenic response to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in mice lacking the Ah (dioxin) receptor.
        Genes Cells. 1997; 2: 645-654
        • Monteleone I.
        • Pallone F.
        • Monteleone G.
        Aryl hydrocarbon receptor and colitis.
        Semin Immunopathol. 2013; 35: 671-675
        • Sekine H.
        • Mimura J.
        • Oshima M.
        • Okawa H.
        • Kanno J.
        • Igarashi K.
        • Gonzalez F.J.
        • Ikuta T.
        • Kawajiri K.
        • Fujii-Kuriyama Y.
        Hypersensitivity of aryl hydrocarbon receptor-deficient mice to lipopolysaccharide-induced septic shock.
        Mol Cell Biol. 2009; 29: 6391-6400
        • Ohtake F.
        • Baba A.
        • Takada I.
        • Okada M.
        • Iwasaki K.
        • Miki H.
        • Takahashi S.
        • Kouzmenko A.
        • Nohara K.
        • Chiba T.
        • Fujii-Kuriyama Y.
        • Kato S.
        Dioxin receptor is a ligand-dependent E3 ubiquitin ligase.
        Nature. 2007; 446: 562-566
        • Kawajiri K.
        • Kobayashi Y.
        • Ohtake F.
        • Ikuta T.
        • Matsushima Y.
        • Mimura J.
        • Pettersson S.
        • Pollenz R.S.
        • Sakaki T.
        • Hirokawa T.
        • Akiyama T.
        • Kurosumi M.
        • Poellinger L.
        • Kato S.
        • Fujii-Kuriyama Y.
        Aryl hydrocarbon receptor suppresses intestinal carcinogenesis in ApcMin/+ mice with natural ligands.
        Proc Natl Acad Sci U S A. 2009; 106: 13481-13486
        • Ikuta T.
        • Kobayashi Y.
        • Kitazawa M.
        • Shiizaki K.
        • Itano N.
        • Noda T.
        • Pettersson S.
        • Poellinger L.
        • Fujii-Kuriyama Y.
        • Taniguchi S.
        • Kawajiri K.
        ASC-associated inflammation promotes cecal tumorigenesis in aryl hydrocarbon receptor-deficient mice.
        Carcinogenesis. 2013; 34: 1620-1627
        • Braeuning A.
        • Köhle C.
        • Buchmann A.
        • Schwarz M.
        Coordinate regulation of cytochrome P450 1A1 expression in mouse liver by the aryl hydrocarbon receptor and the β-catenin pathway.
        Toxicol Sci. 2011; 122: 16-25
        • Jin U.-H.
        • Lee S.-O.
        • Sridharan G.
        • Lee K.
        • Davidson L.A.
        • Jayaraman A.
        • Chapkin R.S.
        • Alaniz R.
        • Safe S.
        Microbiome-derived tryptophan metabolites and their aryl hydrocarbon receptor-dependent agonist and antagonist activities.
        Mol Pharmacol. 2014; 85: 777-788
        • Kasai S.
        • Ishigaki T.
        • Takumi R.
        • Kamimura T.
        • Kikuchi H.
        β-Catenin signaling induces CYP1A1 expression by disrupting adherens junctions in Caco-2 human colon carcinoma cells.
        Biochim Biophys Acta. 2013; 1830: 2509-2516
        • Yamamoto M.
        • Yaginuma K.
        • Tsutsui H.
        • Sagara J.
        • Guan X.
        • Seki E.
        • Yasuda K.
        • Yamamoto M.
        • Akira S.
        • Nakanishi K.
        • Noda T.
        • Taniguchi S.
        ASC is essential for LPS-induced activation of procaspase-1 independently of TLR-associated signal adaptor molecules.
        Genes Cells. 2004; 9: 1055-1067
        • Fujii C.
        • Shiratsuchi A.
        • Manaka J.
        • Yonehara S.
        • Nakanishi Y.
        Difference in the way of macrophage recognition of target cells depending on their apoptotic states.
        Cell Death Differ. 2001; 8: 1113-1122
        • Shiratsuchi A.
        • Osada S.
        • Kanazawa S.
        • Nakanishi Y.
        Essential role of phosphatidylserine externalization in apoptosing cell phagocytosis by macrophages.
        Biochem Biophys Res Commun. 1998; 246: 549-555
        • Kanda Y.
        Investigation of the freely available easy-to-use software “EZR” for medical statistics.
        Bone Marrow Transplant. 2013; 48: 452-458
        • Veldhoen M.
        • Hirota K.
        • Westendorf A.M.
        • Buer J.
        • Dumoutier L.
        • Renauld J.C.
        • Stockinger B.
        The aryl hydrocarbon receptor links TH17-cell-mediated autoimmunity to environmental toxins.
        Nature. 2008; 453: 106-109
        • Kimura A.
        • Naka T.
        • Nohara K.
        • Fujii-Kuriyama Y.
        • Kishimoto T.
        Aryl hydrocarbon receptor regulates Stat1 activation and participates in the development of Th17 cells.
        Proc Natl Acad Sci U S A. 2008; 105: 9721-9726
        • Kang M.
        • Mitomi H.
        • Sada M.
        • Tokumitsu Y.
        • Takahashi Y.
        • Igarashi M.
        • Katsumata T.
        • Okayasu I.
        Ki-67, p53, and Bcl-2 expression of serrated adenomas of the colon.
        Am J Surg Pathol. 1997; 21: 417-423
        • Shih I.-M.
        • Wang T.-L.
        • Traverso G.
        • Romans K.
        • Hamilton S.R.
        • Ben-Sasson S.
        • Kinzler K.W.
        • Vogelstein B.
        Top-down morphogenesis of colorectal tumors.
        Proc Natl Acad Sci U S A. 2001; 98: 2640-2645
        • Haarmann-Stemmann T.
        • Bothe H.
        • Abel J.
        Growth factors, cytokines and their receptors as downstream targets of arylhydrocarbon receptor (AhR) signaling pathways.
        Biochem Pharmacol. 2009; 77: 508-520
        • Ikuta T.
        • Kurosumi M.
        • Yatsuoka T.
        • Nishimura Y.
        Tissue distribution of aryl hydrocarbon receptor in the intestine: implication of putative roles in tumor suppression.
        Exp Cell Res. 2016; 343: 126-134
        • Takahashi-Tezuka M.
        • Yoshida Y.
        • Fukada T.
        • Ohtani T.
        • Yamanaka Y.
        • Nishida K.
        • Nakajima K.
        • Hibi M.
        • Hirano T.
        Gab1 acts as an adapter molecule linking the cytokine receptor gp130 to ERK mitogen-activated protein kinase.
        Mol Cell Biol. 1998; 18: 4109-4117
        • Bollrath J.
        • Phesse T.J.
        • von Burstin V.A.
        • Putoczki T.
        • Bennecke M.
        • Bateman T.
        • Nebelsiek T.
        • Lundgren-May T.
        • Canli Ö.
        • Schwitalla S.
        • Matthews V.
        • Schmid R.M.
        • Kirchner T.
        • Arkan M.C.
        • Ernst M.
        • Greten F.R.
        gp130-Mediated Stat3 activation in enterocytes regulates cell survival and cell-cycle progression during colitis-associated tumorigenesis.
        Cancer Cell. 2009; 15: 91-102
        • Grivennikov S.
        • Karin E.
        • Terzic J.
        • Mucida D.
        • Yu G.Y.
        • Vallabhapurapu S.
        • Scheller J.
        • Rose-John S.
        • Cheroutre H.
        • Eckmann L.
        • Karin M.
        IL-6 and Stat3 are required for survival of intestinal epithelial cells and development of colitis-associated cancer.
        Cancer Cell. 2009; 15: 103-113
        • Rad R.
        • Cadiñanos J.
        • Rad L.
        • Varela I.
        • Strong A.
        • Kriegl L.
        • Constantino-casas F.
        • Eser S.
        • Hieber M.
        • Seidler B.
        • Price S.
        • Fraga M.F.
        • Calvanese V.
        • Hoffman G.
        • Ponstingl H.
        • Schneider G.
        • Yusa K.
        • Grove C.
        • Schmid R.M.
        • Wang W.
        • Vassiliou G.
        • Kirchner T.
        • Mcdermott U.
        • Liu P.
        • Saur D.
        • Bradley A.
        A genetic progression model of Braf(V600E)-induced intestinal tumorigenesis reveals targets for therapeutic intervention.
        Cancer Cell. 2013; 24: 15-29
        • Dejea C.M.
        • Wick E.C.
        • Hechenbleikner E.M.
        • White J.R.
        • Mark Welch J.L.
        • Rossetti B.J.
        • Peterson S.N.
        • Snesrud E.C.
        • Borisy G.G.
        • Lazarev M.
        • Stein E.
        • Vadivelu J.
        • Roslani A.C.
        • Malik A.A.
        • Wanyiri J.W.
        • Goh K.L.
        • Thevambiga I.
        • Fu K.
        • Wan F.
        • Llosa N.
        • Housseau F.
        • Romans K.
        • Wu X.
        • McAllister F.M.
        • Wu S.
        • Vogelstein B.
        • Kinzler K.W.
        • Pardoll D.M.
        • Sears C.L.
        Microbiota organization is a distinct feature of proximal colorectal cancers.
        Proc Natl Acad Sci U S A. 2014; 111: 18321-18326
        • Kostic A.D.
        • Gevers D.
        • Pedamallu C.S.
        • Michaud M.
        • Duke F.
        • Earl A.M.
        • Ojesina A.I.
        • Jung J.
        • Bass A.J.
        • Tabernero J.
        • Baselga J.
        • Liu C.
        • Shivdasani R.A.
        • Ogino S.
        • Birren B.W.
        • Huttenhower C.
        • Garrett W.S.
        • Meyerson M.
        Genomic analysis identifies association of Fusobacterium with colorectal carcinoma.
        Genome Res. 2012; 22: 292-298
        • Castellarin M.
        • Warren R.L.
        • Freeman J.D.
        • Dreolini L.
        • Krzywinski M.
        • Strauss J.
        • Barnes R.
        • Watson P.
        • Allen-Vercoe E.
        • Moore R.A.
        • Holt R.A.
        Fusobacterium nucleatum infection is prevalent in human colorectal carcinoma.
        Genome Res. 2012; 22: 299-306
        • Yu J.
        • Chen Y.
        • Fu X.
        • Zhou X.
        • Peng Y.
        • Shi L.
        • Chen T.
        • Wu Y.
        Invasive Fusobacterium nucleatum may play a role in the carcinogenesis of proximal colon cancer through the serrated neoplasia pathway.
        Int J Cancer. 2016; 139: 1318-1326
        • Bilger A.
        • Shoemaker A.R.
        • Gould K.A.
        • Dove W.F.
        Manipulation of the mouse germline in the study of Min-induced neoplasia.
        Semin Cancer Biol. 1996; 7: 249-260
        • Oshima H.
        • Oshima M.
        • Kobayashi M.
        • Tsutsumi M.
        • Taketo M.M.
        Morphological and molecular processes of polyp formation in ApcΔ716 knockout mice.
        Cancer Res. 1997; 57: 1644-1649
        • Nalbantoglu I.
        • Blanc V.
        • Davidson N.O.
        Characterization of colorectal cancer development in Apcmin/+ mice.
        Methods Mol Biol. 2016; 1422: 309-327
        • Bennecke M.
        • Kriegl L.
        • Bajbouj M.
        • Retzlaff K.
        • Robine S.
        • Jung A.
        • Arkan M.C.
        • Kirchner T.
        • Greten F.R.
        Ink4a/Arf and oncogene-induced senescence prevent tumor progression during alternative colorectal tumorigenesis.
        Cancer Cell. 2010; 18: 135-146
        • Xie G.
        • Peng Z.
        • Raufman J.-P.
        Src-mediated aryl hydrocarbon and epidermal growth factor receptor cross talk stimulates colon cancer cell proliferation.
        Am J Physiol Gastrointest Liver Physiol. 2012; 302: G1006-G1015
        • Belsches A.P.
        • Haskell M.D.
        • Parsons S.J.
        Role of c-Src tyrosine kinase in EGF-induced mitogenesis.
        Front Biosci. 1997; 2: 501-518
        • Hardbower D.M.
        • Singh K.
        • Asim M.
        • Verriere T.G.
        • Olivares-Villagómez D.
        • Barry D.P.
        • Allaman M.M.
        • Washington Jr., M.K.
        • Peek R.M.
        • Piazuelo M.B.
        • Wilson K.T.
        EGFR regulates macrophage activation and function in bacterial infection.
        J Clin Invest. 2016; 126: 3296-3312
        • Hardbower D.M.
        • Coburn L.A.
        • Asim M.
        • Singh K.
        • Sierra J.C.
        • Barry D.P.
        • Gobert A.P.
        • Piazuelo M.B.
        • Washington M.K.
        • Wilson K.T.
        EGFR-mediated macrophage activation promotes colitis-associated tumorigenesis.
        Oncogene. 2017; 36: 3807-3819
        • Kimura A.
        • Naka T.
        • Nakahama T.
        • Chinen I.
        • Masuda K.
        • Nohara K.
        • Fujii-Kuriyama Y.
        • Kishimoto T.
        Aryl hydrocarbon receptor in combination with Stat1 regulates LPS-induced inflammatory responses.
        J Exp Med. 2009; 206: 2027-2035
        • Masuda K.
        • Kimura A.
        • Hanieh H.
        • Nguyen N.T.
        • Nakahama T.
        • Chinen I.
        • Otoyo Y.
        • Murotani T.
        • Yamatodani A.
        • Kishimoto T.
        Aryl hydrocarbon receptor negatively regulates LPS-induced IL-6 production through suppression of histamine production in macrophages.
        Int Immunol. 2011; 23: 637-645
        • Netea M.G.
        • Nold-Petry C.A.
        • Nold M.F.
        • Joosten L.A.B.
        • Opitz B.
        • van der Meer J.H.M.
        • van de Veerdonk F.L.
        • Ferwerda G.
        • Heinhuis B.
        • Devesa I.
        • Funk C.J.
        • Mason R.J.
        • Kullberg B.J.
        • Rubartelli A.
        • van der Meer J.W.M.
        • Dinarello C.A.
        Differential requirement for the activation of the inflammasome for processing and release of IL-1β in monocytes and macrophages.
        Blood. 2009; 113: 2324-2335
        • Piccini A.
        • Carta S.
        • Tassi S.
        • Lasiglie D.
        • Fossati G.
        • Rubartelli A.
        ATP is released by monocytes stimulated with pathogen-sensing receptor ligands and induces IL-1β and IL-18 secretion in an autocrine way.
        Proc Natl Acad Sci U S A. 2008; 105: 8067-8072
        • Elliott E.I.
        • Sutterwala F.S.
        Monocytes take their own path to IL-1β.
        Immunity. 2016; 44: 713-715
        • Mian C.
        • Ceccato F.
        • Barollo S.
        • Watutantrige-Fernando S.
        • Albiger N.
        • Regazzo D.
        • De Lazzari P.
        • Pennelli G.
        • Rotondi S.
        • Nacamulli D.
        • Pelizzo M.R.
        • Jaffrain-Rea M.L.
        • Grimaldi F.
        • Occhi G.
        • Scaroni C.
        AHR over-expression in papillary thyroid carcinoma: clinical and molecular assessments in a series of Italian acromegalic patients with a long-term follow-up.
        PLoS One. 2014; 9: e101560
        • Occhi G.
        • Barollo S.
        • Regazzo D.
        • Bertazza L.
        • Galuppini F.
        • Guzzardo V.
        • Jaffrain-Rea M.L.
        • Vianello F.
        • Ciato D.
        • Ceccato F.
        • Watutantrige-Fernando S.
        • Bisognin A.
        • Bortoluzzi S.
        • Pennelli G.
        • Boscaro M.
        • Scaroni C.
        • Mian C.
        A constitutive active MAPK/ERK pathway due to BRAFV600E positively regulates AHR pathway in PTC.
        Oncotarget. 2015; 6: 32104-32114
        • Lin P.
        • Chang H.
        • Tsai W.-T.
        • Wu M.-H.
        • Liao Y.-S.
        • Chen J.-T.
        • Su J.-M.
        Overexpression of aryl hydrocarbon receptor in human lung carcinomas.
        Toxicol Pathol. 2003; 31: 22-30
        • Koliopanos A.
        • Kleeff J.
        • Xiao Y.
        • Safe S.
        • Zimmermann A.
        • Büchler M.W.
        • Friess H.
        Increased arylhydrocarbon receptor expression offers a potential therapeutic target for pancreatic cancer.
        Oncogene. 2002; 21: 6059-6070
        • Zhao S.
        • Ohara S.
        • Kanno Y.
        • Midorikawa Y.
        • Nakayama M.
        • Makimura M.
        • Park Y.
        • Inouye Y.
        HER2 overexpression-mediated inflammatory signaling enhances mammosphere formation through up-regulation of aryl hydrocarbon receptor transcription.
        Cancer Lett. 2013; 330: 41-48
        • Morkel M.
        • Riemer P.
        • Bläker H.
        • Sers C.
        Similar but different: distinct roles for KRAS and BRAF oncogenes in colorectal cancer development and therapy resistance.
        Oncotarget. 2015; 6: 20785-20800
        • Warren S.
        • Sommers S.C.
        Pathogenesis of ulcerative colitis.
        Am J Pathol. 1949; 25: 657-679
        • Riddell R.H.
        • Goldman H.
        • Ransohoff D.F.
        • Appelman H.D.
        • Fenoglio C.M.
        • Haggitt R.C.
        • Ahren C.
        • Correa P.
        • Hamilton S.R.
        • Morson B.C.
        • Sommers S.C.
        • Yardley J.H.
        Dysplasia in inflammatory bowel disease: standardized classification with provisional clinical applications.
        Hum Pathol. 1983; 14: 931-968
        • Liu J.Z.
        • Van Sommeren S.
        • Huang H.
        • Ng S.C.
        • Alberts R.
        • Takahashi A.
        • Ripke S.
        • Lee J.C.
        • Jostins L.
        • Shah T.
        • Abedian S.
        • Cheon J.H.
        • Cho J.
        • Daryani N.E.
        • Franke L.
        • Fuyuno Y.
        • Hart A.
        • Juyal R.C.
        • Juyal G.
        • Kim W.H.
        • Morris A.P.
        • Poustchi H.
        • Newman W.G.
        • Midha V.
        • Orchard T.R.
        • Vahedi H.
        • Sood A.
        • Sung J.Y.
        • Malekzadeh R.
        • Westra H.J.
        • Yamazaki K.
        • Yang S.K.
        • Barrett J.C.
        • Franke A.
        • Alizadeh B.Z.
        • Parkes M.
        • Thelma B.K.
        • Daly M.J.
        • Kubo M.
        • Anderson C.A.
        • Weersma R.K.
        • International Multiple Sclerosis Genetics ConsortiumInternational IBD Genetics Consortium
        Association analyses identify 38 susceptibility loci for inflammatory bowel disease and highlight shared genetic risk across populations.
        Nat Genet. 2015; 47: 979-986