IBD (inflammatory bowel disease)-related tissue damage occurs in areas which are massively infiltrated with monocytes/macrophages. These cells respond to inflammatory stimuli with enhanced production of cytokines/chemokines. In the present study, we analysed the expression and role of IL (interleukin)-34, a regulator of monocyte/macrophage differentiation, survival and function, in IBD. A significant increase in IL-34 mRNA and protein expression was seen in inflamed mucosa of patients with CD (Crohn's disease) and patients with UC (ulcerative colitis) compared with the uninvolved areas of the same patients and normal controls. IL-34 was up-regulated in LPMCs (lamina propria mononuclear cells) isolated from normal colon by TNF-α (tumour necrosis factor α) and TLR (Toll-like receptor) ligands and was down-regulated in intestinal biopsies and LPMCs of IBD patients upon treatment with infliximab. Treatment of normal LPMCs with IL-34 increased TNF-α expression in an ERK1/2 (extracellular-signal-regulated kinase 1/2)-dependent fashion and neutralization of IL-34 in IBD mucosal explants reduced TNF-α and IL-6 synthesis. In conclusion, our results indicate that IL-34 is up-regulated in IBD and suggest a role for this cytokine in sustaining the inflammatory responses in this disease.
Cytokines drive the initiation and progression of the tissue-damaging immune response in IBDs (inflammatory bowel diseases), and are the prime therapeutic targets for these debilitating diseases.
IL (interleukin)-34, a cytokine produced by many cell types, is highly synthesized in inflamed gut of patients with Crohn's disease and patients with ulcerative colitis, the two major IBD in human beings.
In IBD, neutralization of IL-34 function attenuates mucosal production of inflammatory cytokines, and therefore IL-34 could be a promising target for therapeutic intervention in patients with these disorders.
IBDs (inflammatory bowel diseases), which include CD (Crohn's disease) and UC (ulcerative colitis), represent a group of disorders characterized by chronic inflammation and epithelial injury of the gastrointestinal tract [1,2]. The aetiologies of both CD and UC remain unknown, but experimental evidence suggests that IBD-associated tissue damage is driven by an exaggerated immune response against components of the normal flora . In particular, the inflamed gut of IBD patients is massively infiltrated with various cell subsets, such as effector CD4+ T-cells, CD8+ T-cells, B-cells, natural killer (T-) cells and CD14+ monocytes [4–7]. These cell types produce elevated levels of inflammatory cytokines that target both immune and non-immune cells, thus contributing to exacerbate the mucosal inflammation . This hypothesis is supported by the demonstration that blockade of effector cytokines, such as TNF-α (tumour necrosis factor α), attenuates the detrimental response in some subsets of IBD patients [9,10].
IL (interleukin)-34 is a cytokine produced by a wide range of cells, including macrophages, endothelial cells, fibroblasts, neurons, hepatocytes and epithelial cells, and is constitutively expressed in adult human tissues, such as heart, brain, liver, spleen, thymus, testis, ovary, prostate, small intestine and colon [11–15]. Although IL-34 shares no apparent sequence homology with M-CSF (macrophage colony-stimulating factor; also known as CSF1), its biological activity is mediated by interaction with the homodimeric M-CSFR (M-CSF receptor; also known as CFS-1R or Fms) that is mainly expressed on the cell surface of macrophages [11,16,17]. Several observations suggest that IL-34 and M-CSF have non-redundant and complementary roles [11,14,18,19]. IL-34 seems to induce more activation of the M-CSFR that, after ligation, forms a dimer and autophosphorylates its own tyrosine residues, thus recruiting effector proteins and modulating cell activation . In particular, IL-34 regulates myeloid cell differentiation, proliferation and survival, and enhances the secretion of pro-inflammatory cytokines and chemokines [11,18–20]. Consistently, IL-34 was found to be overexpressed in joint fluids, synovial membrane and serum of rheumatoid arthritis patients compared with controls, in serum samples of patients with coronary artery disease and in the inflamed salivary glands of patients with Sjögren's syndrome [21–24].
M-CSF serum levels are increased in IBD patients and M-CSF protein is abundantly produced in inflamed mucosa of IBD patients and non-IBD colitis patients . Moreover, the administration of a neutralizing anti-MCSF-1 antibody in mice is partially protective in dextran sulfate sodium-induced colitis, thus highlighting the ability of M-CSFR to deliver inflammatory signals . However, more recent in vitro studies have demonstrated that IL-34 induces the differentiation of human monocytes into immunosuppressive type II macrophages, through a mechanism that is independent of M-CSF1 . Altogether, these data indicate that IL-34 can be either pro-inflammatory or anti-inflammatory depending on the cell context analysed. The present study aimed to examine the expression and role of IL-34 in IBD.
MATERIALS AND METHODS
Patients and samples
Biopsies were taken from inflamed mucosa of six patients with active colonic CD, five patients with active ileocolonic CD and 19 patients with active UC undergoing colonoscopy for a clinically active disease at the Gastrointestinal Unit of Tor Vergata University (Rome, Italy) or Fondazione IRCCS Cà Granda, Ospedale Maggiore Policlinico (Milan, Italy). Paired biopsies were also taken from the inflamed and uninflamed mucosa of two patients with colonic CD and six patients with UC. Nine patients (one colonic CD, two ileocolonic CD and six UC) were taking no drug, and biopsies were collected at the time of initial diagnosis. Moreover, biopsies were taken from 21 patients (six colonic CD, two ileocolonic CD and 13 UC) receiving therapy (mesalamine, steroids, immunosuppressive) drugs. In all of these patients, colonoscopy was performed for a clinical relapse of the disease.
Seven steroid-dependent/resistant patients (five UC, two ileocolonic CD) were treated with IFX (infliximab). Each patient received a total of three IFX infusions (5 mg/kg) at weeks 0, 2 and 6. Six colonic biopsies were collected during a flexible sigmoidoscopy from the sigmoid colon before the first and after the third IFX infusion. Two UC patients were receiving concomitant treatment with mesalamine and azathioprine, three patients (one UC, two CD) were treated with mesalamine, and two UC patients were receiving mesalamine, azathioprine and steroids. All of the concomitant therapies were maintained during IFX treatment.
Additionally, surgical specimens were taken from 14 patients with colonic CD and 15 patients with UC undergoing surgery for a chronic active disease poorly responsive to medical treatment and from nine patients with ileocolonic CD undergoing surgery due to stricturing disease. Surgical specimens were also taken from colons of three patients undergoing surgery for diverticulitis.
Controls included biopsies taken from unaffected colonic mucosa of 11 subjects. Additional controls were mucosal specimens taken from macroscopically and microscopically unaffected colonic areas of 48 patients undergoing surgery for colon cancer. Autologous peripheral blood samples were obtained from two healthy donors.
Each patient who took part in the study gave written informed consent and the study protocol was approved by the local Ethics Committees (Tor Vergata University Hospital, Rome; protocol number: 154/12).
Cell isolation and culture
All reagents were from Sigma–Aldrich unless specified. LPMCs (lamina propria mononuclear cells) were isolated as described previously  with the only exception that tissue digestion was performed with Liberase TM (200 μg/ml) and DNase I (200 μg/ml) (both Roche Diagnostics) instead of collagenase. Human PBMCs (peripheral blood mononuclear cells) were isolated from EDTA-stabilized blood samples using Ficoll gradients.
To determine the factors that modulate IL-34 expression, normal colonic LPMCs were either unstimulated or stimulated with the recombinant cytokines TNF-α (20 ng/ml; R&D Systems), IL-6 (50 ng/ml; R&D Systems), IFN-γ (interferon γ) (100 ng/ml; Peprotech), LPS (lipopolysaccharide) (100 ng/ml), PGN (peptidoglycan) (10 μg/ml), unmethylated CpG dinucleotides (1 μg/ml), or poly(I:C) (polyinosinic:polycytidylic acid) (5 μg/ml) for 6–48 h. At the end, cells were used to extract RNA while cell-free supernatants were analysed by ELISA. In order to evaluate whether IFX treatment affects IL-34 gene expression, IL-34 transcripts were evaluated in PBMCs and LPMCs stimulated with TNF-α (20 ng/ml; R&D Systems) in the presence or absence of IFX at 10, 25 or 50 μg/ml or control IgG (50 μg/ml; R&D Systems) for 18 h.
To examine whether IL-34 induces the production of inflammatory cytokines, normal colonic LPMCs were treated with increasing doses of recombinant human IL-34 (25–100 ng/ml; R&D Systems) for 6–48 h. At the end, cells were used to extract RNA, and cell-free supernatants were analysed by ELISA and protein array. In parallel, normal LPMCs were stimulated with IL-34 (50 ng/ml) for 15–30 min, and then total proteins were analysed for the content of both phosphorylated and total MAPKs (mitogen-activated protein kinases) by Western blotting. Moreover, normal LPMCs were pre-incubated with PD98059 (20 μM; EMD Millipore Corporation), a selective inhibitor of ERK1/2 (extracellular-signal-regulated kinase 1/2) , or DMSO (vehicle) for 1 h and then stimulated or not with IL-34 (50 ng/ml) for 30 min or 48 h. At the end, ERK, p38 and JNK (c-Jun N-terminal kinase) phosphorylation was evaluated by Western blotting. Additionally, cell-free supernatants were used to assess TNF-α protein content by ELISA.
Ex vivo organ cultures
Freshly obtained IBD mucosal samples were cultured in RPMI 1640 complete medium (Lonza) as described in  in the presence of a neutralizing anti-human IL-34 antibody or control mouse IgG (both used at 10 μg/ml; R&D Systems). After 24 h, biopsies were homogenized with Tissue Lyser II (Qiagen), and total RNA was extracted using PureLink mRNA Mini Kit (Life Technologies), according to the manufacturer's instructions. Culture supernatants were also collected and stored at −80°C until being assessed for cytokine production by ELISA.
A constant amount of RNA (0.5 μg/sample) was retro-transcribed into cDNA and then 1 μl of cDNA/sample was amplified using the following conditions: denaturation for 1 min at 95°C, annealing for 30 s at 62°C for TNF-α, 61°C for IL-6, 60°C for IL-34 and β-actin and 58°C for MCSF-R, followed by 30 s of extension at 72°C. Primer sequences were TNF-α, 5′-AGGCGGTGCTTGTTCCTCAG-3′ (forward) and 5′- GGCTACAGGCTTGTCACTCG-3′ (reverse); IL-6, 5′-CCACTCACCTCTTCAGAACG-3′ (forward) and 5′-GCCTCTTTGCTGCTTTCACAC-3′ (reverse); IL-34, 5′-ACAGGAGCCGACTTCAGTAC-3′ (forward) and 5′-ACCAAGACCCACAGATACCG-3′ (reverse); β-actin, 5′-AAGATGACCCAGATCATGTTTGAGACC-3′ (forward) and 5′-AGCCAGTCCAGACGCAGGAT-3′ (reverse). Human CD206 (Hs 00267207_m1) and arginase 1 (Hs 00968979_m1) gene expression was evaluated using commercial TaqMan probes (Applied Biosystems). mRNA expression was calculated relative to the housekeeping β-actin gene on the base of the ∆∆CT algorithm.
Total protein extraction and Western blotting
LPMCs were lysed on ice in buffer containing 10 mM Hepes (pH 7.9), 10 mM KCl, 0.1 mM EDTA, 0.2 mM EGTA and 0.5% Nonidet P40 supplemented with 1 mM DTT, 10 mg/ml aprotinin, 10 mg/ml leupeptin, 1 mM PMSF, 1 mM Na3VO4 and 1 mM NaF. Lysates were clarified by centrifugation at 12000 g for 30 min at 4°C and separated by SDS/PAGE (10% gel). p-ERK1/2, p-p38 and p-JNK were detected using a mouse anti-human p-ERK1/2 antibody (1:500 dilution; Santa Cruz Biotechnology), rabbit anti-mouse p-p38 antibody (1:1000 dilution; EMD Millipore Corporation) and mouse anti-human p-JNK antibody (1:500 dilution; Santa Cruz Biotechnology) respectively followed by HRP (horseradish peroxidase)-conjugated secondary IgG monoclonal antibodies (all used at 1:20000 final dilution; Dako). The reaction was detected with a sensitive ECL kit (Pierce). After analysis, blots were stripped and incubated with the following internal loading controls: rabbit anti-human total ERK1/2 antibody (1:500 dilution; Santa Cruz Biotechnology), mouse anti-human total p38 antibody and mouse anti-human total JNK antibody (both at 1:500 dilution; Santa Cruz Biotechnology) respectively, followed by HRP-conjugated secondary antibodies (1:20000 dilution; Dako).
Immunohistochemistry was performed on archival frozen sections of mucosal samples taken from control and IBD patients. Tissue sections were prepared as follows: samples were embedded in a cryostat mounting medium (Neg-50 Frozen Section Medium, Thermo Scientific), snap-frozen and stored at −80°C. Sections 6-μm-thick were mounted on to Superfrost Plus glass slides (Thermo Scientific) and fixed in 4% neutral buffered formalin for 10 min at room temperature and then in increasing ethanol solutions. After washing in TBS, endogenous peroxidase activity was quenched with 3% H2O2 diluted in methanol for 10 min at room temperature. The slides were incubated with a mouse monoclonal antibody directed against human IL-34 (final dilution 1:50000; Abcam) at room temperature for 30 min, followed by a biotin-free HRP–polymer detection technology (Ultravision Detection System, Thermo Scientific) with 3,3′diaminobenzidine (Dako) as a chromogen. The sections were counterstained with haematoxylin, dehydrated and mounted. Isotype control IgG-stained sections were prepared under identical immunohistochemical conditions as described above, replacing the primary antibody with a purified mouse normal IgG control antibody (both R&D Systems). The IL-34-positive cells were counted in at least six fields per section using IAS 2000 System (Delta Sistemi) and expressed as number of cells in the hpf (high-power field).
Enzyme-linked immunosorbent assay
Human IL-34, TNF-α and IL-6 were measured using sensitive commercial ELISA kits (R&D Systems) according to the manufacturer's instructions.
Human cytokine expression array assay
Human Inflammation Array 3 was purchased from Ray Biotech and used following the manufacturer's instructions. Briefly, supernatants of control LPMCs either left untreated or treated with 50 ng/ml IL-34 were added to membranes, each immobilized with 40 different capture antibodies, and incubated at 4°C overnight. The membranes then were incubated with biotin-conjugated antibodies for 2 h, and then with HRP-conjugated streptavidin for 30 min. Unbound reagents were removed by washing, and the bound antibodies on the membranes were visualized using an ECL system. Finally, quantitative analysis was performed by densitometry scanning of array blots.
Assessment of LPMC proliferation
LPMCs were cultured with or without 50 ng/ml IL-34 for 48 h, and then proliferation was assessed by using a commercially available BrdU (5-bromodeoxyuridine) assay kit (Roche) according to the manufacturer's instructions.
Differences between groups were compared using Student's t test. All analyses were performed using GraphPad Prism 5 software.
IL-34 expression is up-regulated in IBD
To examine whether IL-34 expression is differently regulated during IBD, total RNA was extracted from colonic biopsies and LPMCs of control and IBD patients and examined by real-time PCR. IL-34 RNA transcripts were increased in inflamed colon of both CD and UC patients compared with controls (Figure 1A). Analysis of IL-34 in paired biopsies taken from patients with IBD showed that IL-34 RNA expression was more pronounced in areas with active inflammation (Figure 1B). In IBD, IL-34 expression did not differ between patients receiving no therapy and those on drugs (Figure 1C). Consistent with the above data, IL-34 expression was increased in LPMCs extracted from both CD and UC colonic mucosal specimens compared with controls (Supplementary Figure S1).
IL-34 transcripts are increased in inflamed IBD mucosa
Using immunohistochemistry, we next showed that IL-34-producing cells were more frequent in the lamina propria compartment of both CD and UC patients compared with controls with no apparent difference between CD and UC (Figure 2A). Analysis of total proteins extracted from whole mucosal biopsy samples by ELISA confirmed that IL-34 was overproduced in IBD, whereas there was no difference between samples taken from inflamed mucosa areas of patients with active diverticulitis and controls (Figure 2B).
IL-34 protein expression is increased in inflamed IBD mucosa
IL-34 expression is positively regulated by TNF-α, PGN, poly(I:C) and CpG
Next we examined whether IL-34 gene expression is regulated by inflammatory cytokines produced in excess in IBD [2,31]. Initially, normal colonic LPMCs were stimulated with TNF-α, IL-6 or IFN-γ, and IL-34 RNA transcripts were analysed by real-time PCR. IL-34 mRNA expression and protein production were significantly increased by TNF-α, but not by either IL-6 or IFN-γ stimulation (Figure 3). Consistently, blockade of TNF-α with 50 μg/ml IFX significantly reduced IL-34 mRNA transcripts in IBD LPMC cultures (1.03 ± 0.1 compared with 2.7 ± 0.9 in cells treated with IgG; p=0.02). This dose of IFX was selected from studies performed with normal PBMCs showing that induction of IL-34 by TNF-α was significantly reduced by 25 and 50 μg/ml IFX, with the latter determining a total suppression of IL-34 induction (Supplementary Figure S2). A significant down-regulation of IL-34 RNA expression was also observed in colonic biopsies of IBD patients treated with IFX (5.7 ± 1.9 compared with 15.3 ± 5.5 in biopsies before IFX treatment; p=0.02).
TNF-α induces IL-34 in normal colonic LPMCs
Since IL-34 is produced by a variety of cells that can recognize and respond to bacterial components/products , we next evaluated the effect of TLR (Toll-like receptor) ligands on control LPMC-derived IL-34. PGN, poly(I:C) and CpG, but not LPS, significantly increased IL-34 and this was evident at both the RNA and protein levels (Figures 4A and 4B respectively).
PGN, poly(I:C) and CpG enhance IL-34 RNA and protein expression in normal LPMCs
IL-34 enhances synthesis of TNF-α and IL-6 in the human gut
To determine the contribution of IL-34 to the IBD-associated cytokine response, normal LPMCs were stimulated with increasing doses of IL-34, and TNF-α gene and protein expression was evaluated by real-time PCR and ELISA respectively. We initially focused the study on TNF-α given the importance of this cytokine in IBD-related inflammation . IL-34 dose-dependently enhanced TNF-α mRNA expression (Figure 5A). Since the maximum induction of TNF-α was observed stimulating with 50 ng/ml IL-34, this dose was selected for the subsequent experiments. Treatment of control LPMCs with IL-34 significantly increased TNF-α protein secretion (Figure 5B) without affecting cell proliferation (results not shown). Since IL-34 induces ERK1/2 activation in human monocytes/macrophages [11,12], we evaluated whether IL-34-driven TNF-α induction was mediated by ERK1/2 signalling. Control LPMCs stimulated with IL-34 exhibited enhanced phosphorylation of ERK1/2, but no increase in p38 or JNK activation (Figure 5C). Pre-incubation of normal LPMCs with PD98059, an inhibitor of activation of ERK1/2, but not of p38 or JNK (Figures 5D and 5E), drastically abrogated IL-34-induced TNF-α production (Figure 5F).
IL-34 enhances TNF-α expression through an ERK-dependent mechanism in normal LPMCs
To evaluate whether IL-34 regulates the synthesis of other inflammatory molecules, control LPMCs were stimulated with IL-34, and the culture supernatants were screened for the content of 40 different proteins using a human protein antibody array. As shown in Figure 6(A), stimulation of cells with IL-34 enhanced the synthesis of IL-6, whereas the secretion of other chemokines and cytokines remained unchanged. Analysis of mRNA samples extracted from control LPMCs by real-time PCR confirmed the IL-34-inducing effect on IL-6 also at the gene level (Figure 6B). In contrast, IL-34 did not affect RNA expression of CD206 and arginase 1 (Supplementary Figures S3A and S3B respectively), two markers of type II macrophages, thus excluding the possibility that the increased expression of IL-6 and TNF-α following IL-34 stimulation is due to shifts in the differentiation/development of type II macrophages.
IL-34 enhances IL-6 in normal LPMCs
To confirm the positive regulation by IL-34 of TNF-α and IL-6, mucosal explants taken from three IBD patients were cultured with a neutralizing anti-IL-34 or control antibody and TNF-α and IL-6 were analysed by ELISA. In each experiment, treatment with anti-IL-34 reduced TNF-α and IL-6 secretion (Table 1).
|Treatment .||Patient .||TNF-α (ng/ml) .||IL-6 (ng/ml) .|
|Treatment .||Patient .||TNF-α (ng/ml) .||IL-6 (ng/ml) .|
In the present study, we evaluated the expression and function of IL-34 in IBD. IL-34 is a cytokine produced by several cell types which is primarily involved in the regulation of the differentiation, survival and function of monocytes/macrophages . IL-34 shares a common receptor with M-CSF, M-CSFR, even though the two cytokines can exert non-redundant functions . The present study shows that IBD-related inflammation is marked by an increased gene expression and protein production of IL-34. Interestingly, no significant increase in IL-34 was observed in inflamed colonic samples of patients with active diverticular disease, thus suggesting that up-regulation of the cytokine in IBD is not simply an epiphenomenon of the ongoing mucosal inflammation.
The relevance of the increased IL-34 expression in IBD tissue is highlighted by further observations. The up-regulation of IL-34 in patients with CD and patients with UC could reflect a similar pathophysiological mechanism which accounts for the induction of this cytokine in the two IBDs. Indeed, IL-34 expression was enhanced in normal LPMCs by TNF-α, a cytokine that is overproduced in both CD and UC and that is involved in the amplification of the inflammatory response in these disorders [9,10,34]. Consistently, the neutralization of TNF-α with IFX reduced IL-34 production in IBD LPMC cultures and in vivo in IBD patients. Our results also indicate that TLR ligands, such as PGN, poly(I:C) and CpG, but not LPS, positively regulate IL-34 at the gene and protein levels. These data are consistent with the demonstration that IL-34 expression is induced by TLR ligands in cell lines and primary head-kidney macrophages in rainbow trout . The reason IL-34 production was not increased in control LPMCs by LPS remains to be clarified, even though it is well known that normal intestinal macrophages exhibit a defective response to LPS in terms of inflammatory cytokines . Previous studies indicated that NF-κβ (nuclear factor κB) and MAPK pathways regulate production of IL-34 in macrophages, synovial fibroblasts and osteoblasts [12,20,22]. Since both NF-κβ and MAPK are hyperactivated in IBD LPMCs [36,37], it is likely that induction of IL-34 in the gut of patients with CD and patients with UC is driven by these signalling pathways. Since, in IBD mucosa, IL-34 is produced by many immune and non-immune cells, studies are now ongoing to identify which signals control IL-34 synthesis in the different cell types.
Our functional studies showed that IL-34 positively regulates TNF-α production in control LPMCs through an ERK-dependent mechanism and that the neutralization of endogenous IL-34 with a blocking antibody reduced TNF-α in IBD mucosal explants. Furthermore, IL-34 induced IL-6 production in control LPMCs, and neutralization of IL-34 decreased IL-6 production in IBD mucosal explants. In contrast, treatment of control LPMCs with IL-34 did not change the production of other cytokines, such as IL-8 and GM-CSF (granulocyte/macrophage colony-stimulating factor), perhaps reflecting the inability of IL-34 to activate intracellular pathways that govern the synthesis of such cytokines [38,39].
Our results are in line with previous studies documenting high expression of IL-34 in pathological conditions. For example, IL-34 is overexpressed in the inflamed synovium of patients with rheumatoid arthritis, where it acts as a downstream effector of TNF-α and IL-1β, induces osteoclastogenesis and contributes to tissue inflammation and bone erosion . Moreover, baseline serum IL-34 levels would seem to serve as a prognostic factor for progression in such patients . IL-34 is also overproduced in the inflamed salivary glands of patients with Sjögren's syndrome, a chronic autoimmune disease characterized by keratoconjunctivitis sicca and dryness . More recent studies have shown that IL-34 is increased in serum and visceral adipose tissue of obese patients and circulating levels of IL-34 correlate not only with adiposity parameters, such as abdominal fat areas and body mass index, but also with the signs associated with insulin resistance and chronic inflammation . We feel that the pro-inflammatory role of IL-34 described in the present paper needs to be confirmed in vivo using animal models of colitis. Unfortunately, however, no commercial IL-34-deficient or -transgenic mouse is yet available.
In conclusion, we have shown for the first time that IL-34 is abundantly produced in inflamed gut mucosa of patients with IBD and acts a positive regulator of TNF-α and IL-6 synthesis in the gut. This is intriguing and pinpoints a potential pivotal role for IL-34 in the amplification and perpetuation of the ongoing mucosal inflammation in IBD.
Eleonora Franzè performed experiments, analysed data and wrote the paper. Ivan Monteleone, Maria Laura Cupi, Irene Marafini, Pamela Mancia, Federica Laudisi, Alfredo Colantoni and Angela Ortenzi performed experiments. Flavio Caprioli, Giuseppe Sica and PierPaolo Sileri contributed reagents. Francesco Pallone analysed data. Giovanni Monteleone designed experiments and wrote the paper.
This work was supported by the Fondazione Umberto di Mario, Rome, Italy.
granulocyte/macrophage colony-stimulating factor
inflammatory bowel disease
c-Jun N-terminal kinase
lamina propria mononuclear cell
mitogen-activated protein kinase
macrophage colony-stimulating factor
nuclear factor κB
peripheral blood mononuclear cell
tumour necrosis factor α
G. M. has contributed to the patent application entitled ‘Methods and compositions for diagnosing and treating inflammatory bowel disorders’.