GASTROENTEROLOGY 2009;137:1736 –1745 Th1/Th17 Immune Response Is Induced by Mesenteric Lymph Node Dendritic Cells in Crohn’s Disease ATSUSHI SAKURABA,* TOSHIRO SATO,* NOBUHIKO KAMADA,* MINA KITAZUME,* AKIRA SUGITA,‡ and TOSHIFUMI HIBI* *Division of Gastroenterology, Department of Internal Medicine, Keio University School of Medicine, Tokyo; and ‡Department of Surgery, Yokohama Municipal Hospital, Yokohama, Japan See editorial on page 1566. BASIC– ALIMENTARY TRACT BACKGROUND & AIMS: Dendritic cells (DCs) possess the most potent ability to induce acquired immunity. However, their involvement in the pathogenesis of Crohn’s disease (CD) has not yet been determined. We aimed to establish the immune status of mesenteric lymph nodes, the major gut-associated lymphoid tissue, and isolated DCs and determine their involvement in the pathogenesis of CD. METHODS: CD4⫹ T cells and DCs were isolated from mesenteric lymph nodes of CD, ulcerative colitis, and normal control. The immune status of CD4⫹ T cells was analyzed by cytokine production and transcriptional profile. Surface phenotype of DCs was analyzed by flow cytometry. Cytokine production by myeloid DCs was analyzed by realtime polymerase chain reaction and exogenous bacterial stimulation. Immune stimulating activity of DCs was determined by mixed lymphocyte reaction. RESULTS: In CD, mesenteric lymph node CD4⫹ T cells produced higher amounts of interferon-␥ and interleukin (IL)-17 compared with ulcerative colitis and normal control, and this was dictated by increased T-bet and retinoic acid-related orphan receptor-␥ expression. Three subtypes of DCs, myeloid DC, plasmacytoid DC, and mature DC, were identified in all groups. When stimulated with exogenous bacterial derivative, myeloid DCs from CD produced a higher amount of IL-23 and a lower amount of IL-10. Myeloid DCs from CD induced stronger T helper cell (Th)1 immune response in mixed lymphocyte reaction compared with those from ulcerative colitis and normal control. CONCLUSIONS: Our findings revealed that mesenteric lymph node is the key pathogenic location of CD elicited by the unique cytokine milieu produced by DCs leading to a dysregulated Th1/Th17 immune response. I ntensive research in recent years has revealed that the pathogenesis and etiology of Crohn’s disease (CD) involve complex interactions among genetic predisposition, environmental factors, and immune abnormalities.1 For example, we and others have shown that interleukin (IL)-12/IL-18 secretion from intestinal macrophages aug- ments T helper cell (Th) 1 commitment of intestinal lamina propria T cells in CD.2– 6 Although intestinal macrophages induce proliferation of lamina propria T cells, which mainly consist of effector memory T cells, they do not efficiently prime naïve T cells. It thus remains unknown where and how naïve T cells are primed to form disease-inducible memory T cells. Recent studies have proposed that dendritic cells (DCs) are the most potent type of antigen-presenting cells.7,8 Furthermore, only DCs can prime naïve T cells in lymphoid tissues, and, in the gut, T cells are thought to be primed by DCs in gut-associated lymphoid tissues. Before antigen presentation, DCs reside in the intestinal lamina propria, Peyer’s patches, or isolated lymph follicles and sample commensal as well as pathogenic microorganisms.9,10 After antigen uptake, intestinal DCs mature and migrate to mesenteric lymph nodes (MLNs),11 where they encounter and prime naïve T cells. In this engagement, DCs provide 3 signals. The first is delivered through the T-cell receptor on engagement of an appropriate peptide-major histocompatibility complex (MHC).7,8 The second is referred to as costimulation and involves expression of several costimulatory molecules: CD80, CD86, and CD40, representatively.12–15 Costimulation is essential for priming of naïve T cells, preventing anergy. The third signal was proposed recently and is thought to be delivered from DCs to naïve T cells, determining their differentiation into specific effector cells through production of Th1, Th2, or the recently identified Th17 polarization factors.16 In humans, at least 2 distinct DC precursor subsets have been identified based on their origin, phenotype, and function.17 Myeloid DCs (mDCs), Abbreviations used in this paper: CD, Crohn’s disease; cDNA, complementary DNA; DC, dendritic cells; EDTA, ethylenediaminetetraacetic acid; ELISA, enzyme-linked immunosorbent assay; FITC, fluorescein isothiocyanate; GATA-3, GATA binding protein 3; IFN, interferon; IL, interleukin; mDCs, myeloid DCs; MHC, major histocompatibility complex; MLN, mesenteric lymph nodes; MLR, mixed lymphocyte reaction; NC, normal controls; PCR, polymerase chain reaction; pDCs, plasmacytoid DCs; PE, phycoerythrin; RORc, retinoic acid-related orphan receptor-␥; Th, T helper cell; UC, ulcerative colitis. © 2009 by the AGA Institute 0016-5085/09/$36.00 doi:10.1053/j.gastro.2009.07.049 which have a CD11c⫹CD123⫺ (IL-3R␣) phenotype, distribute throughout the peripheral tissue and migrate via lymphatic vessels into lymph nodes.18 In contrast, plasmacytoid DCs (pDCs) are CD11c⫺CD123⫹ and enter directly from the systemic circulation into lymph nodes.19 mDCs possess a significant Th1-inducing ability through abundant production of IL-12 in response to pathogenic stimuli.20 –22 In contrast, pDCs release lower amounts of IL-12 than mDCs and provide a milieu favoring a Th2biased immune reaction.17,23,24 The factors inducing Th17 cells remain elusive, but IL-23 facilitates the expansion and stabilization of Th17-cell responses.25,26 Despite the importance of DCs in innate immunity, only a few studies have investigated their function in CD. Here, we first sorted and characterized DCs from CD MLNs and examined their contribution to the pathogenesis of CD. Materials and Methods Patients and Samples MLNs were obtained from 16 patients with CD and 15 patients with ulcerative colitis (UC) who underwent surgery because of refractoriness to medical treatment. The ileum was the primary site of CD involvement in 10 patients, the ileocolonic in 4, and the colon in 2. All of the patients had clinically severely active disease as defined by Crohn’s disease activity index of ⬎250. The resected intestine showed macroscopically severe inflammation in all patients. All patients were taking 5-aminosalicylates at the time of surgery, and 4 were taking corticosteroids. Eight of the 15 UC patients had total colitis, and the remaining 7 had left-sided colitis. Fourteen patients were taking 5-aminosalicylates, and 4 were on corticosteroids at the time of surgery. All UC patients were refractory to medical therapy and had active disease in the resected colon. None of the patients were receiving anti-tumor necrosis factor agents. MLNs were obtained from the mesentery closest to the inflamed mucosa in all CD and UC patients. As a normal control, MLNs devoid of metastasis or inflammation were obtained from 4 colonic cancer patients undergoing surgery. The study was approved by the institute’s Ethics Committee, and written informed consent was obtained from all patients before surgery. Cytokines and Antibodies The following were purchased from BD Pharmingen (San Diego, CA): purified phycoerythrin (PE) and fluorescein isothiocyanate (FITC)-conjugated anti-human lineage 1 (CD3, CD16, CD19, CD20) monoclonal antibodies (mAb) (SK7, 3G8, SJ25C1, L27, mouse IgG1), PE-conjugated anti-human CD80 mAb (2D3, mouse IgG2b), control mouse IgG1 mAb (MOPC-21), control mouse IgG2b (G155–178), allophycocyanin (APC)-conjugated anti-human CD3 mAb (UCHT1, mouse IgG1), PE-conjugated Th1/Th17 IMMUNE RESPONSE IN CROHN’S DISEASE 1737 anti-human CD86 mAb (CD28.2, mouse IgG1), purified and FITC-conjugated anti-human CD3 (HIB19, mouse IgG1), purified anti-human CD28 (HIB28, mouse IgG1), PE-conjugated CD45 (HI30, mouse IgG1), PE-conjugated anti-human CD123 mAb (BNI3, mouse IgG2a), PE-conjugated anti-human CD45RA mAb (HI100, mouse IgG2b), PE-conjugated anti-human CD45RO mAb (UCHL1, mouse IgG2a), FITC, PE and peridinin chlorophyll proteinconjugated anti-human CD4 mAb (RPA-T4, mouse IgG1), and FITC and peridinin chlorophyll protein-conjugated anti-human HLA-DR (HIT8a, mouse IgG1). FITC-conjugated anti-human interferon (IFN)-␥, PE-conjugated antihuman IL-4, and FITC/PE-conjugated anti-mouse IgG2a⫹b (X57) were purchased from BD Bioscience (San Jose, CA). PE-IL-17A (64DEC17, Mouse IgG1) was purchased from eBioscience (San Diego, CA). Recombinant human IL-10 and IL-23p19 were purchased from eBioscience and used at a concentration of 20 ng/mL. Phorbol 12-myristate 13-acetate and ionomycin were purchased from Sigma– Aldrich Japan (Tokyo, Japan). Isolation of CD4ⴙ T Cells and DCs From MLNs MLNs closest to the inflamed intestine were immediately dissected from surgical specimens, followed by blunt dissociation with slide glasses in calcium- and magnesium-free Hank’s balanced salt solution (CMF-HBSS) containing 2.5% fetal bovine serum, 1 mmol/L dithiothreitol, and ethylenediaminetetraacetic acid (EDTA) (Sigma–Aldrich, St. Louis, MO) to obtain a single cell suspension at 4°C. For T cells, the fraction was pelleted twice and resuspended in 5 mL of 40% Percoll (Pharmacia Biotech, Piscataway, NJ), which was then layered over 60% Percoll before centrifugation at 1500 rpm for 30 minutes at 18°C. Cells in the top 40%/60% layer interface contained ⬎95% pure viable mononuclear cells. CD4⫹ T cells were obtained using a magnetic cell sorting system (MACS, CD4 multi-sort kit; Miltenyi Biotec, Bergisch Gladbach, Germany) according to the manufacturer’s instructions. The purification of CD4⫹ T cells was ⬎95% (data not shown). To obtain DCs, the fraction was pelleted twice and resuspended in 5 mL of 13% Optiprep (Pharmacia Biotech, Piscataway, NJ). Three milliliters of CMF-HBSS containing 2.5% fetal bovine serum, 1 mmol/L dithiothreitol, and EDTA was then layered over the top and centrifuged at 1500 rpm for 30 minutes at 18°C. Cells in the top Optiprep/CMF-HBSS layer interface contained ⬎95% pure viable mononuclear cells. CD4⫹ cells were obtained using MACS and then sorted using a FACS vantage (Beckton Dickson, Franklin Lakes, NJ) to obtain mature DCs (Lin⫺/MHC IIhi/CD11c⫹/CD123⫹), mDCs (Lin⫺/ MHC II⫹/CD11c⫹/CD123⫺), and pDCs (Lin⫺/MHC II⫹/ CD11c⫺/CD123⫹). The purification of each DC subset was ⬎95% confirmed by flow cytometry (data not shown). In some experiments, to obtain mature DCs and mDCs as a combined fraction, BDCA-1⫹ cells were mag- BASIC– ALIMENTARY TRACT November 2009 1738 SAKURABA ET AL netically sorted (MACS; BDCA-1 isolation kit; Miltenyi Biotec) according to the manufacturer’s instructions. Flow Cytometry Flow cytometric analysis was performed as previously described.27 Viable lymphocyte and DC populations were gated using forward scatter/side scatter and negative staining with propidium iodide. For staining, 1 ⫻ 106 freshly isolated or cultured cells were incubated with 20 ␮L of mentioned antibodies or isotype-matched mouse immunoglobulin (Ig)G for 20 minutes on ice. After washing, the fluorescence intensity on the cell surfaces was analyzed using a FACScan (Becton Dickinson). Mixed Lymphocyte Reaction and Intracellular Cytokine Staining BASIC– ALIMENTARY TRACT Peripheral blood was obtained from a single healthy volunteer, and CD4⫹CD45RA⫹ naïve T cells were isolated according to the manufacturer’s protocol (MACS, CD4⫹CD45RA⫹ multi-sort kit; Miltenyi Biotec). Freshly isolated mature MLN DCs were incubated with allogeneic CD4⫹CD45RA⫹ T cells for 5 days in 96-well plates. To assay proliferation, the DC/T-cell ratio was varied from 1:5 to 1:1, 3H-thymidine at a dose of 5.0 ␮Ci/mL (Amersham Life Science, Buckinghamshire, United Kingdom) was added for the last 24 hours, and its uptake was examined using a liquid scintillation counter (Pharmacia, Peapack, NJ). For intracellular cytokine staining, T-cell cultures were collected on day 5, and 1 ⫻ 106 cells were restimulated for 4 hours with 50 ng/mL phorbol 12myristate 13-acetate plus 2 ␮g/mL ionomycin and then incubated with 10 ␮g/mL brefeldin A (Sigma Aldrich Japan) for an additional 3 hours. Cells were collected, fixed, and permeabilized with Cyto Fix/Perm (Becton Dickson). Intracellular staining was performed with FITC-conjugated anti-human IFN-␥ and PE-conjugated anti-human IL-4 or IL-17. Stained cells were analyzed using a FACScalibur. For intracellular cytokine staining of CD4⫹ T cells from MLNs, 1 ⫻ 106 cells were stimulated with plate bound CD3/CD28 in 96-well plates for 72 hours, followed by restimulation for 4 hours with 50 ng/mL phorbol 12-myristate 13-acetate plus 2 ␮g/mL ionomycin. Cells were collected and processed for the intracellular staining as described above. Reverse-Transcription Polymerase Chain Reaction Analysis Total RNA was isolated from 1 ⫻ 107 freshly purified MLN CD4⫹ T cells using RNeasy columns (Qiagen, Valencia, CA) according to the manufacturer’s instructions. First-strand complementary DNA (cDNA) was synthesized from 2 ␮g of total RNA with an oligo (dT) primer using an Omniscript RT kit (Qiagen). Real-time polymerase chain reaction (PCR) was conducted with the GASTROENTEROLOGY Vol. 137, No. 5 ABI Prism 7700 sequence detection system (Applied Biosystems, Foster City, CA). Taqman Probes and the primer for T-bet, GATA binding protein 3 (GATA-3), retinoic acidrelated orphan receptor-␥ (RORc), and ␤-actin were purchased from Applied Biosystems. Values were calculated on the basis of standard curves generated for each gene. Samples were normalized by dividing the number of copies of T-bet, GATA-3, or RORc cDNA by the number of copies of ␤-actin cDNA. Cytokine Analysis To detect cytokines produced by MLN CD4⫹ T cells, supernatants were collected after 72-hour stimulation with plate bound CD3/CD28 (eBioscience) in 96well plates. The concentrations of IL-2, IL-4, IL-5, IFN-␥, and IL-10 were then determined by Cytokine Beads Array (BD Pharmingen, San Diego, CA) according to the manufacturer’s instructions. The IL-17 contents in culture supernatants were measured using a specific sandwich enzyme-linked immunosorbent assay (ELISA) (eBioscience). To detect cytokines produced in mixed lymphocyte reactions (MLR), supernatants were collected after 60hour culture of DC and CD4⫹CD45RA⫹ T cells. The concentrations of IL-2, IL-4, IL-5, IFN-␥, and IL-10 were measured as described above. For DCs, isolated DCs were stimulated for 36 hours with or without lipopolysaccharides (Sigma Aldrich) and Enterococcus faecalis extract. E faecalis extract was prepared as described previously.28 Supernatants were collected, and the concentrations of IL-12p40/p70 (R&D Systems, Minneapolis, MN), IL-23 (BenderMed Systems, Burlingame, CA), and IL-10 (R&D Systems) were determined by ELISA according to the manufacturer’s instructions. Statistical Analysis Results are expressed as the mean ⫾ SEM. Groups of data were compared using the nonparametric Mann– Whitney U test. Statistical significance was established at P ⬍ .05. Results CD4ⴙ T Lymphocytes of CD MLNs Overexpress IFN-␥ and IL-17 We analyzed MLNs from CD and frequently observed granuloma formation, the hallmark of the Th1 immune response, as compared with UC or normal controls (NC) (Figure 1A). To analyze the phenotype of effector T cells in CD MLN, we isolated CD4⫹ T lymphocytes from the MLNs. The ratio of CD4⫹/CD8⫹, CD45RA⫹/CD4⫹, and CD25⫹/CD4⫹ in the MLN T cells was constant among the 3 groups: CD, UC, and NC (data not shown). To determine Th1/Th2/Th17 polarization, we isolated MLN CD4⫹ T cells and stimulated them with November 2009 Th1/Th17 IMMUNE RESPONSE IN CROHN’S DISEASE 1739 anti-CD3/CD28 mAbs, followed by phorbol 12-myristate 13-acetate/ionomycin. Intracellular IFN-␥ production by CD4⫹ T cells was significantly increased in CD compared with UC and NC (17.70% ⫾ 5.46% vs 6.07% ⫾ 2.18% and 1.44% ⫾ 0.69%, P ⫽ .012 and P ⫽ .004, respectively), whereas IL-4 production did not differ among the 3 groups (Figure 1B and C). IL-17 producing cells were relatively low in each group but were significantly increased in CD compared with UC (2.87 ⫾ 0.41 vs 0.88 ⫾ 0.27, P ⫽ .032). The proportion of cells double positive for IFN-␥ and IL-17 was small in all 3 groups not exceeding 0.50%. Consistent with the results of intracellular cytokine production, higher production of IFN-␥ and IL-17 was confirmed in CD MLN T cells by cytokine bead array or ELISA (Figure 1D). To further determine Th1/ Th2/Th17 balance, we analyzed the expression of the master regulators of Th1/Th2/Th17 commitment, T-bet (Th1), GATA-3 (Th2), and RORc (Th17).29,30 Transcription of T- BASIC– ALIMENTARY TRACT Figure 1. The immune status of CD4⫹ T cells in MLNs of CD. (A) An MLN from the inflamed intestine of a CD patient is shown here, with abundant non-caseating granulomas (left, original magnification, ⫻10; right, original magnification, ⫻40 of the solid square in the left). (B and C) Intracellular production of IFN-␥, IL-4, and IL17A was assessed in isolated CD4⫹ T cells from MLNs of NC, UC, and CD. Numbers indicate the percentages of cells in each quadrant. (D) Production of IFN-␥, IL-4, and IL-17A was assessed in isolated CD4⫹ T cells from MLNs of NC, UC, and CD by cytokine bead array or ELISA. ND, not detected. (E) Expression of the transcription factors T-bet, GATA-3, and RORc in CD4⫹ T lymphocytes from MLNs was assessed by real-time RT-PCR. A total of 4, 10, and 16 independent experiments for NC, UC, and CD, respectively, were performed for C–E. According to the available cell numbers in each experiment, only certain analyses were performed in some experiments. Horizontal bars show the mean value. bet and RORc was significantly increased in CD4⫹ T lymphocytes from MLN of CD compared with those of UC, whereas transcription of GATA-3 did not differ among the groups (Figure 1E). As is the case with lamina propria CD4⫹ T cells, these results suggested that CD4⫹ T cells show a proinflammatory phenotype in CD MLNs and further led us to seek the mechanisms behind it. Isolation of DCs From MLNs We isolated DCs from MLNs using several different reported methods31–33 and found that blunt dissociation of MLNs followed by density gradient centrifugation of the single cell suspension was the best method in terms of enrichment and yield. After pre-enrichment of DCs using mAb to CD4, a surface molecule expressed on all DCs, we performed flow cytometry. We observed distinct cell population in lineage (B cells [CD19, CD20]; T cells [CD3]; and natural killer (NK) cells [CD56]) nega- 1740 SAKURABA ET AL GASTROENTEROLOGY Vol. 137, No. 5 Next, we assessed the surface phenotype of MLN DCs using flow cytometry. As shown in Figure 3A, mDCs and mature DCs expressed all of the costimulatory molecules, CD40, CD80, and CD86, whereas pDCs only expressed CD86. Among the 3 DC subpopulations, CD40 and CD86 expression was highest in mature DCs, whereas CD80 expression was highest in mDCs. We also investigated the activation status of the DC subpopulation using CD83, the DC activation marker. The results revealed that CD83 was only expressed in mature DCs, not mDCs or pDCs. CD40 and CD80 expression in mDC and mature DCs did not differ among NC, UC, and CD (data not shown). Interestingly, the ratio of mDC to pDC was significantly increased in CD compared with UC and NC (Figure 3B); however, there were no significant differences in the mature DC population or CD83 expression in MLN DCs from the 3 groups (data not shown). BASIC– ALIMENTARY TRACT Figure 2. Isolation of MLN DCs. (A) Mononuclear cells were isolated from MLNs and analyzed by flow cytometry. Dead cells were excluded by observing the fluorescence intensity of propidium iodide (left; gray square). Lineage negative, MHC II high cells were identified as DCs (middle; gray square). These cells were divided into 3 populations using anti-CD11c, CD123 mAb, and fluorescence intensity of MHC II (right). These were myeloid DC (mDC; CD11c⫹/CD123⫺/MHC II⫹), plasmacytoid DC (pDC; CD11c⫺/CD123⫹/MHC II⫹), and mature DC (CD11c⫹/ CD123⫹/MHC II⫹). Representative data of CD are shown. UC and NC showed similar results. (B) MLR between mDCs and allogeneic naïve T cells were performed. The proliferation of T cells was assessed by incorporation of 3H-thymidine. Shown are representative data from 4 individual experiments. tive, MHC II-positive fraction, and the surface molecular phenotypes were identical with DCs in other lymphoid tissue34 (Figure 2A). The capacity to induce allogenenic T-cell proliferation is a functional in vitro hallmark of DCs. Thus, we sorted this population and performed a MLR with allo-donor derived naïve T cells. As compared with peripheral blood monocytes, the sorted population showed higher stimulatory capacity of allo-naïve T cells (Figure 2B), further indicating a DC purification. The MLNs were significantly larger in size in CD and UC, and, thus, the total number of MLN DCs was higher than in NC. However, because of aggressive lymphoid cell accumulation, the ratio of DCs within the MLN mononuclear cells did not significantly differ among the 3 groups (data not shown). We further divided the DCs into 3 populations using anti-CD11c, CD123 mAb, and the fluorescence intensity of MHC II (Figure 2A). The 3 subpopulations showed identical surface phenotypes with previously reported subpopulations in the spleen and thymus31,35; mDC: CD11c⫹/CD123⫺/MHC II⫹, pDC: CD11c⫺/CD123⫹/MHC II⫹, and mature DC: CD11c⫹/ CD123⫹/MHC II⫹⫹. Similar to the peripheral blood DC subpopulation and monocyte-derived DCs, the mDCs showed a spindle-shaped morphology, pDCs a lymphoidshaped morphology, and mature DCs a round-shaped morphology (data not shown). Figure 3. Surface phenotype of MLN DCs. (A) The surface phenotype of MLN DCs was assessed by flow cytometry. The black area shows a histogram plot of each marker and the gray area a histogram plot of respective isotype controls. Representative data from a CD patient are shown. Similar results were obtained from UC and NC. (B) The number of each DC subset was counted and the proportion of each calculated. A total of 3 independent experiments for NC, UC, and CD, respectively, was performed. Data are shown as mean ⫾ SEM. *Indicates P ⬍ .05. Figure 4. mDCs from CD MLNs induce a Th1 response in allogeneic MLR. (A) MLR was performed between the sorted subgroups of DCs and allogeneic CD4⫹ CD45RA⫹ naïve T lymphocytes. Intracellular staining was performed for IFN-␥/IL-4 on the stimulated T cells. Representative data from CD are shown. UC and NC showed similar results. (B) Supernatants of MLR cultures were collected, and the content of IFN-␥, IL-4, and IL-10 were analyzed by CBA. Representative data from CD are shown. UC and NC showed similar results. (C) mDCs and CD4⫹CD45RA⫹ naïve T cells were cocultured, and cytokine production of IFN-␥, IL-4, and IL-17 by T cells was analyzed by intracellular cytokine staining as in A. (D) Cytokine concentrations in culture supernatants of MLR were analyzed by CBA. Representative or pooled data from 3 independent experiments for NC, UC, and CD, respectively, are shown in A–D. Data are shown as mean ⫾ SEM in B and D. Th1/Th17 IMMUNE RESPONSE IN CROHN’S DISEASE 1741 BASIC– ALIMENTARY TRACT November 2009 1742 SAKURABA ET AL DCs From MLNs of CD Preferentially Induce a Th1 Immune Response To seek which DC subset is responsible to induce Th1 immune response, we sorted these populations and performed MLR. Both intracellular cytokine assay and cytokine beads array data showed higher Th1 inducibility in mature DC and mDC than pDC regardless of the origin of patients (Figure 4A and B). This result, together with Figure 3B, suggested that, quantitatively, Th1 favored DCs were more accumulated in MLNs of CD compared with NC or UC. To see whether there is a qualitative difference in mature and mDCs, we sorted these populations from CD, NC, or UC and performed MLR. Interestingly, we detected significantly higher Th1-inducing ability from mature/mDC of CD MLN, compared GASTROENTEROLOGY Vol. 137, No. 5 with those from NC or UC (Figure 4C and D). The level of induction of IL-17 was small in each group, suggesting that there are other factors necessary to induce an effective Th17 response. Taken together, DCs from CD MLNs possess quantitative and qualitative higher capability to induce Th1 immune response. Myeloid DCs From CD MLNs Produce High Amounts of IL-23 and Low Amounts of IL-10 To determine the functional difference of DCs in CD MLNs, we analyzed their cytokine profiles. Because of the paucity of DCs from MLNs and to focus solely on the Th1-inducing ability, we isolated mDCs and mature DCs using the BDCA-1 antibody, a marker for myeloid lineage DCs. We confirmed that isolated BASIC– ALIMENTARY TRACT Figure 5. Isolated DCs from CD MLNs produce high amounts of IL-23 and low amounts of IL-10 upon stimulation. (A) mDCs and mature DCs were isolated from MLNs of UC and CD, and the production of IL-12p40, IL-12p70, IL-23, and IL-10 in the supernatants was assessed by ELISA. Lipopolysaccharides (LPS) (10 ␮g/mL) and E faecalis cell extract were used for stimulation. UC, gray bar; CD, black bar. Three independent experiments for UC and CD, respectively, were performed. ND, not detected. (B) The ratio of IL-23p19 to IL-10 was calculated in UC and CD. (C) Mature/mDCs were isolated from MLNs of CD by magnetic sorting. MLR was performed with allogeneic CD4⫹ CD45RA⫹ naïve T lymphocytes. MLR was performed with or without the addition of 20 ng/mL of IL-23 or IL-10. (⫺): medium only. Representative of 3 independent experiments. Data are shown as mean ⫾ SEM, where applicable. DC population consisted of mDCs and mature DCs, but not pDCs, by flow cytometry (data not shown). The isolated DC population was stimulated with lipopolysaccharide and E faecalis cell extract, both potent stimulators of cytokine production in lamina propria macrophages.28 Interestingly, isolated DCs from CD MLNs produced higher amounts of IL-12p40 as well as IL23p19 compared with those from UC, whereas the production of IL-12p70 was comparable between the 2 (Figure 5A). In addition, IL-10 production was lower in CD MLN DCs compared with those from UC. The ratio of IL-23 to IL-10, which has been reported to have an effect on the outcome of Th response,36, was higher in CD compared with UC. To confirm that the unbalance in IL-23 and IL-10 leads to the Th1 dominant phenotype in T cells, we performed MLR between mature and mDCs from CD MLN and naïve CD4⫹CD45RA⫹ T cells, with or without the addition of IL-23 or IL-10. When IL-23 was added to MLR, the level of IFN-␥-producing cells increased, whereas it decreased when IL-10 was present in the culture. Interestingly, exogenous IL-23 or IL-10 had little effect on the outcome of IL-17-producing cells. This suggests that additional factors, such as IL-6, IL-1, or transforming growth factor-␤, which may be produced by different cell types, are necessary to induce sufficient Th17 response in naïve T cells.25,26,37 Taken together, DCs from CD MLNs preferentially induce a Th1 dominant response through their increased IL-23 and decreased IL-10-producing property. Discussion In the present study, we evaluated the immune status of T cells in MLN, the major gut-associated lymphoid tissue, in human inflammatory bowel disease. We showed that lymphocytes from MLNs show a Th1 and Th17 feature in CD. They produced high amounts of IFN-␥ and IL-17, expressed T-bet and RORc, the master transcription factor of Th1 and Th17 immune response, respectively. Furthermore, we sought the mechanism of Th1/Th17 commitment in MLNs and found that mDCs from CD MLNs induce a proinflammatory response. mDCs from CD MLNs secreted high amounts of IL-23 and low amounts of IL-10, which promoted mainly IFN-␥ production by CD4⫹ T cells. Previous studies have mainly focused on the lamina propria, the effector site of inflammation,5,27,38 – 40 elucidating that Th1 T cells and macrophages massively migrate into the intestinal mucosa, where their engagement causes proinflammatory cytokine production, further recruiting inflammatory cells and inducing mucosal damage. Recently, IL-17-producing helper T cells, designated Th17 cells, have been shown to infiltrate the gut mucosa in inflammatory bowel disease.41,42 We therefore aimed to resolve where and how Th1/Th17 IMMUNE RESPONSE IN CROHN’S DISEASE 1743 the effector T cells are induced in CD. Considering the fact that granuloma, the hallmark of the Th1 immune response, is often observed in CD MLNs,43 MLN is deemed a feasible primary induction site of CD, and, thus, we focused on the involvement of MLN in CD pathophysiology. We found that CD4⫹ T cells in the MLNs of CD show a predominant Th1/Th17 phenotype. This was partially explained by the fact that mDCs and/or mature DCs from CD MLNs possess a strong Th1-inducing ability compared with those from UC and NC. Although it is difficult to show directly the origin of human MLN DCs, circumstantial evidence supports the notion that they are derived from the gut and induce naïve T cells to transform into effector T cells in the MLNs. The T cells then return to the intestinal mucosa where, in the effector phase, they are reactivated by intestinal macrophages, augmenting the Th1 immune response in situ. We found 3 functional subsets of DCs in MLN. mDCs and mature DCs were more frequently seen in CD MLNs than their counterparts from UC and NC, and they preferentially induced Th1 cells. We confirmed that mDCs and mature DCs from MLNs of CD were more potent to induce Th1 response than those from MLNs of UC or NC. Furthermore, we found that mDCs and mature DCs from MLNs of CD produced a higher amount of IL-12p40 and IL-23p19 and a lower amount of IL-10 compared with UC. Although the mechanism leading to this increased production remains unresolved, it is notably a distinct feature of CD. Our previous study showed that the differentiation status of macrophages/DCs determines the level of IL-12/IL-23 production.44 Thus, unknown factors such as intestinal luminal antigens may alter the maturation of DCs and increase the production of IL-12/ IL-23. Further study is needed to address this issue. IL-17 has recently been recognized to play an important role in immune pathologic responses. The proportion of T cells producing IL-17 was relatively low in each disease but was significantly up-regulated in CD. However, IFN-␥/IL-17 double-positive cells were hardly detected. Furthermore, DCs from CD MLNs induced only a small number of IL-17⫹ T cells. We presume that IL-17⫹ T cells are mainly induced in the lamina propria, and/or there are other factors that are produced by different cell types involved in the induction of Th17 cells. Further investigation is needed to elucidate the precise mechanism of Th17 induction. Although several lines of evidence have shown that IL-23 is involved in Th1-mediated colitis, IL-23 cannot directly induce Th1 cells. Here, we have shown that IL-23 possesses Th1-inducing effect through its effect on DCs. Further studies are needed to determine how IL-23 auto-stimulated DCs acquire higher Th1 inducibility. BASIC– ALIMENTARY TRACT November 2009 1744 SAKURABA ET AL In conclusion, we propose that MLN may be the new disease induction location in CD. Furthermore, we found the existence of Th1/Th17 unbalance in MLNs of CD and that this was partially due to the cytokine milieu produced by MLN DCs. It is expected that DCs or T cells in MLN are feasible therapeutic targets in CD. References BASIC– ALIMENTARY TRACT 1. Podolsky DK. Inflammatory bowel disease. N Engl J Med 2002; 347:417– 429. 2. Takagi H, Kanai T, Okazawa A, et al. Contrasting action of IL-12 and IL-18 in the development of dextran sodium sulphate colitis in mice. Scand J Gastroenterol 2003;38:837– 844. 3. Okazawa A, Kanai T, Watanabe M, et al. Th1-mediated intestinal inflammation in Crohn’s disease may be induced by activation of lamina propria lymphocytes through synergistic stimulation of interleukin-12 and interleukin-18 without T-cell receptor engagement. Am J Gastroenterol 2002;97:3108 –3117. 4. Kanai T, Watanabe M, Okazawa A, et al. Interleukin 18 is a potent proliferative factor for intestinal mucosal lymphocytes in Crohn’s disease. Gastroenterology 2000;119:1514 –1523. 5. Matsuoka K, Inoue N, Sato T, et al. T-bet up-regulation and subsequent interleukin 12 stimulation are essential for induction of Th1-mediated immunopathology in Crohn’s disease. Gut 2004;53:1303–1308. 6. Monteleone G, Biancone L, Marasco R, et al. Interleukin 12 is expressed and actively released by Crohn’s disease intestinal lamina propria mononuclear cells. Gastroenterology 1997;112: 1169 –1178. 7. Banchereau J, Steinman RM. Dendritic cells and the control of immunity. Nature 1998;392:245–252. 8. Banchereau J, Briere F, Caux C, et al. Immunobiology of dendritic cells. Annu Rev Immunol 2000;18:767– 811. 9. Rescigno M, Urbano M, Valzasina B, et al. Dendritic cells express tight junction proteins and penetrate gut epithelial monolayers to sample bacteria. Nat Immunol 2001;2:361–367. 10. Niess JH, Brand S, Gu X, et al. CX3CR1-mediated dendritic cell access to the intestinal lumen and bacterial clearance. Science 2005;307:254 –258. 11. Kobayashi H, Miura S, Nagata H, et al. In situ demonstration of dendritic cell migration from rat intestine to mesenteric lymph nodes: relationships to maturation and role of chemokines. J Leukoc Biol 2004;75:434 – 442. 12. Inaba K, Witmer-Pack M, Inaba M, et al. The tissue distribution of the B7-2 costimulator in mice: abundant expression on dendritic cells in situ and during maturation in vitro. J Exp Med 1994;180: 1849 –1860. 13. Inaba K, Inaba M, Witmer-Pack M, et al. Expression of B7 costimulator molecules on mouse dendritic cells. Adv Exp Med Biol 1995;378:65–70. 14. Caux C, Vanbervliet B, Massacrier C, et al. B70/B7-2 is identical to CD86 and is the major functional ligand for CD28 expressed on human dendritic cells. J Exp Med 1994;180:1841–1847. 15. Young JW, Koulova L, Soergel SA, et al. The B7/BB1 antigen provides one of several costimulatory signals for the activation of CD4⫹ T lymphocytes by human blood dendritic cells in vitro. J Clin Invest 1992;90:229 –237. 16. Kalinski P, Moser M. Consensual immunity: success-driven development of T-helper 1 and T-helper 2 responses. Nat Rev Immunol 2005;5:251–260. 17. Rissoan MC, Soumelis V, Kadowaki N, et al. Reciprocal control of T helper cell and dendritic cell differentiation. Science 1999;283: 1183–1186. GASTROENTEROLOGY Vol. 137, No. 5 18. Ingulli E, Ulman DR, Lucido MM, et al. In situ analysis reveals physical interactions between CD11b⫹ dendritic cells and antigen-specific CD4 T cells after subcutaneous injection of antigen. J Immunol 2002;169:2247–2252. 19. Cella M, Jarrossay D, Facchetti F, et al. Plasmacytoid monocytes migrate to inflamed lymph nodes and produce large amounts of type I interferon. Nat Med 1999;5:919 –923. 20. Macatonia SE, Hosken NA, Litton M, et al. Dendritic cells produce IL-12 and direct the development of Th1 cells from naive CD4⫹ T cells. J Immunol 1995;154:5071–5079. 21. Cella M, Scheidegger D, Palmer-Lehmann K, et al. Ligation of CD40 on dendritic cells triggers production of high levels of interleukin-12 and enhances T cell stimulatory capacity: T-T help via APC activation. J Exp Med 1996;184:747–752. 22. Reis e Sousa C, Hieny S, Scharton-Kersten T, et al. In vivo microbial stimulation induces rapid CD40 ligand-independent production of interleukin 12 by dendritic cells and their redistribution to T-cell areas. J Exp Med 1997;186:1819 –1829. 23. Pulendran B, Smith JL, Caspary G, et al. Distinct dendritic cell subsets differentially regulate the class of immune response in vivo. Proc Natl Acad Sci U S A 1999;96:1036 –1041. 24. Ito T, Amakawa R, Inaba M, et al. Differential regulation of human blood dendritic cell subsets by IFNs. J Immunol 2001;166:2961– 2969. 25. Monteleone I, Pallone F, Monteleone G. Interleukin-23 and Th17 cells in the control of gut inflammation. Mediators Inflamm 2009; 297645. 26. Bettelli E, Korn T, Oukka M, et al. Induction and effector functions of T(H)17 cells. Nature 2008;453:1051–1057. 27. Sato T, Kanai T, Watanabe M, et al. Hyperexpression of inducible costimulator and its contribution on lamina propria T cells in inflammatory bowel disease. Gastroenterology 2004;126:829 – 839. 28. Kamada N, Hisamatsu T, Okamoto S, et al. Unique CD14 intestinal macrophages contribute to the pathogenesis of Crohn’s disease via IL-23/IFN-␥ axis. J Clin Invest 2008;118:2269 – 2280. 29. Szabo SJ, Kim ST, Costa GL, et al. A novel transcription factor, T-bet, directs Th1 lineage commitment. Cell 2000;100:655– 669. 30. Chakir H, Wang H, Lefebvre DE, et al. T-bet/GATA-3 ratio as a measure of the Th1/Th2 cytokine profile in mixed cell populations: predominant role of GATA-3. J Immunol Methods 2003; 278:157–169. 31. Stent G, Reece JC, Baylis DC, et al. Heterogeneity of freshly isolated human tonsil dendritic cells demonstrated by intracellular markers, phagocytosis, and membrane dye transfer. Cytometry 2002;48:167–176. 32. Tsunoda R, Nakayama M, Onozaki K, et al. Isolation and longterm cultivation of human tonsil follicular dendritic cells. Virchows Arch 1990;59:95–105. 33. Hart DN, McKenzie JL. Isolation and characterization of human tonsil dendritic cells. J Exp Med 1988;168:157–170. 34. Bell SJ, Rigby R, English N, et al. Migration and maturation of human colonic dendritic cells. J Immunol 2001;166:4958 – 4967. 35. McIlroy D, Troadec C, Grassi F, et al. Investigation of human spleen dendritic cell phenotype and distribution reveals evidence of in vivo activation in a subset of organ donors. Blood 2001;97: 3470 –3477. 36. Depaolo RW, Tang F, Kim I, et al. Toll-like receptor 6 drives differentiation of tolerogenic dendritic cells and contributes to LcrV-mediated plague pathogenesis. Cell Host Microb 2008;4: 350 –361. 37. Harrington LE, Hatton RD, Mangan PR, et al. Interleukin 17producing CD4⫹ effector T cells develop via a lineage distinct from the T helper type 1 and 2 lineages. Nat Immunol 2005;6: 1123–1132. 38. Pizarro TT, Michie MH, Bentz M, et al. IL-18, a novel immunoregulatory cytokine, is up-regulated in Crohn’s disease: expression and localization in intestinal mucosal cells. J Immunol 1999; 162:6829 – 6835. 39. Sato T, Nakai T, Tamura N, et al. Osteopontin/Eta-1 up-regulated in Crohn’s disease regulates the Th1 immune response. Gut 2005;54:1254 –1262. 40. Inoue N, Watanabe M, Sato T, et al. Restricted V(H) gene usage in lamina propria B cells producing anticolon antibody from patients with ulcerative colitis. Gastroenterology 2001;121:15–23. 41. Fujino S, Andoh A, Bamba S, et al. Increased expression of interleukin 17 in inflammatory bowel disease. Gut 2003;52:65– 70. 42. Kleinschek MA, Boniface K, Sadekova S, et al. Circulating and gut-resident human Th17 cells express CD161 and promote intestinal inflammation. J Exp Med 2009;206:525–534. 43. Assarsson N, Raf L. Incidence of granuloma in Crohn’s disease. Acta Chir Scand 1974;140:249 –251. 44. Kamada N, Hisamatsu T, Okamoto S, et al. Abnormally differentiated subsets of intestinal macrophage play a key role in Th1-dominant chronic colitis through excess production of Th1/Th17 IMMUNE RESPONSE IN CROHN’S DISEASE 1745 IL-12 and IL-23 in response to bacteria. J Immunol 2005; 175:6900 – 6908. Received October 22, 2008. Accepted July 16, 2009. Reprint requests Address requests for reprints to: Toshifumi Hibi, MD, PhD, Division of Gastroenterology, Department of Internal Medicine, Keio University School of Medicine, 35 Shinanomachi, Shinjuku-ku, Tokyo 160-8582, Japan. e-mail: thibi@sc.itc.keio.ac.jp; fax: (81) 3-33576156. Acknowledgments A.S. and T.S. contributed equally to this work. Conflicts of interest The authors disclose no conflicts. Funding Supported by the Japanese Ministry of Health, Labor and Welfare. BASIC– ALIMENTARY TRACT November 2009