The first two authors contributed equally to this work.
DMBT1 is a novel gene induced by IL-22 in ulcerative colitis
Article first published online: 7 SEP 2010
Copyright © 2010 Crohn's & Colitis Foundation of America, Inc.
Inflammatory Bowel Diseases
Volume 17, Issue 5, pages 1177–1188, May 2011
How to Cite
Fukui, H., Sekikawa, A., Tanaka, H., Fujimori, Y., Katake, Y., Fujii, S., Ichikawa, K., Tomita, S., Imura, J., Chiba, T. and Fujimori, T. (2011), DMBT1 is a novel gene induced by IL-22 in ulcerative colitis. Inflamm Bowel Dis, 17: 1177–1188. doi: 10.1002/ibd.21473
- Issue published online: 11 APR 2011
- Article first published online: 7 SEP 2010
- Manuscript Accepted: 30 JUL 2010
- Manuscript Received: 19 JUN 2010
- Ministry of Education, Culture, Sports, Science and Technology, Japan. Grant Numbers: 20590747, 21790684
- IL-22 receptor;
- ulcerative colitis
Interleukin (IL)-22 is a recently identified cytokine that is suggested to play pivotal roles in various inflammatory diseases. Although the IL-22 receptor 1 (IL-22R1) is restrictively expressed in epithelial cells in the colon, the role of IL-22 in colonic diseases still remains unclear. In this study microarray analyses revealed that deleted in malignant brain tumors 1 (DMBT1) is a novel upregulated gene in IL-22-stimulated colon cancer cells. Therefore, we investigated the involvement of DMBT1 and IL-22 in ulcerative colitis (UC) tissues and examined the mechanism regulating the expression of DMBT1 in response to IL-22 stimulation.
Changes of gene expression in IL-22-stimulated SW403 cells were investigated by microarray analyses. The effects of IL-22 on DMBT1 expression were examined in SW403 cells using a small interfering RNA (si)RNA for STAT3 or inhibitors for MEK, PI3K, and nuclear factor kappa B (NF-κB). The element responsible for IL-22-induced DMBT1 promoter activation was determined by a promoter deletion and electrophoretic mobility shift assay (EMSA). Expression of IL-22, IL-22R1, and DMBT1 in UC tissues was analyzed by real-time reverse-transcription polymerase chain reaction (RT-PCR) and immunohistochemistry.
IL-22 treatment enhanced the expression of DMBT1 through STAT3 tyrosine phosphorylation and NF-κB activation in colon cancer cells. The IL-22-responsive element was located between −187 and −179 in the DMBT1 promoter region. In the UC mucosa the levels of DMBT1 and IL-22 mRNA expression were significantly enhanced and positively correlated, the numbers of IL-22-positive lymphocytes were increased, and the expression of IL-22R1 and DMBT1 was enhanced in the inflamed epithelium.
The IL-22/DMBT1 axis may play a pivotal role in the pathophysiology of UC. (Inflamm Bowel Dis 2010;)
Interleukin-22 (IL-22) is a recently identified cytokine produced mainly by activated lymphocytes in various inflammatory and autoimmune diseases.1–4 Although the biological roles of IL-22 still remain to be elucidated, IL-22 has been reported to induce the production of proinflammatory mediators such as IL-8 and IL-6 in skin and gastrointestinal mucosa,5–7 suggesting that IL-22 may play roles in the pathophysiology of tissue inflammation. The receptor for IL-22 consists of two chains: IL-22 receptor 1 (IL-22R1) and IL-10R2.8, 9 At present, it is clear that IL-10R2 is expressed ubiquitously in various organs, whereas expression of IL-22R1 is restricted to epithelial cells, and not immune cells, in the skin, pancreas, kidney, liver, and colorectum.10, 11 Thus, IL-22 may play pivotal roles as a biological mediator in signaling from the immune system to epithelial cells in such organs. Interestingly, two groups have recently reported that IL-22 ameliorates intestinal inflammation using a mouse colitis model.12, 13 Therefore, in order to screen the target molecules for IL-22 signaling, we preliminarily stimulated colon cancer cells with IL-22 and investigated the changes of gene expression in the cells by microarray analyses. Subsequently, we found that deleted in malignant brain tumors 1 (DMBT1) is a novel candidate gene whose expression is induced by IL-22 stimulation.
The DMBT1 gene encodes a secreted glycoprotein of the scavenger receptor cysteine-rich superfamily and its gene product is expressed in normal brain, alveolar immune cells, and the epithelial cells of salivary glands and the gastrointestinal tract.14–16 Although DMBT1 was originally identified by representational difference analysis as a candidate tumor suppressor gene in medulloblastoma,17 its role in carcinogenesis is still unclear. On the other hand, accumulating evidence suggests that DMBT1 plays a role in innate immunity, epithelial cell differentiation, and binding to bacterial or viral pathogens.18–22 Moreover, DMBT1 protein has been recently reported to inhibit the cytoinvasion of bacteria in colon epithelial cells.23 Together, in the present study we investigated the linkage between DMBT1 and IL-22 in the pathophysiology of ulcerative colitis (UC), and clarified the mechanism regulating the expression of DMBT1 in response to IL-22 stimulation in UC mucosa.
MATERIALS AND METHODS
Cell Culture and Reagents
Human colon cancer cell line SW403 cells were cultured in RPMI1640 medium (Invitrogen, Grand Island, NY) with 10% fetal bovine serum (Sigma, St. Louis, MO) in a humidified incubator at 37°C with an atmosphere of 5% CO2. Human IL-22 was purchased from R&D Systems (Minneapolis, MN). MEK inhibitor PD98059, PI3K inhibitor wortmannin, nuclear factor kappa B (NF-κB) inhibitor N-tosyl-L-phenylalanine chloromethyl ketone (TPCK), and anti-β-actin antibody were purchased from Sigma. Anti-STAT3, -phospho-specific STAT3 (p-STAT3 [Tyr705]), -Akt, -phospho-specific Akt (p-Akt [Ser473]), -ERK, and -phospho-specific ERK (p-ERK) antibodies were from Cell Signaling Technology (Beverly, MA). Anti-DMBT1 antibody was purchased from Santa Cruz Biotechnology (Santa Cruz, CA).
SW403 cells (1 × 106/60 mm-dish) were incubated for 24 hours with or without IL-22 treatment (10 ng/mL) in RPMI1640 medium. Using Trizol reagent (Gibco, Rockville, MD), total RNA was extracted from each dishes separately. cDNA labeling, hybridizations, scanning, and data analysis were performed by Hokkaido System Science (Sapporo, Japan). Briefly, the integrity of total RNA was assessed with an Agilent Bioanalyzer 2100 (Agilent Technologies, Santa Clara, CA). Total RNA (500 ng) was reverse-transcribed into cDNA using a Low RNA Fluorescence Linear Amplification Kit (Agilent Technologies). Synthesized cRNA was labeled with cyanine 5-dCTP (IL-22 stimulated cells) or cyanine 3-dCTP (nonstimulated cells). The labeled cRNA was purified with RNeasy mini spin columns (Qiagen, Hilden, Germany) and then hybridized to a Whole Human Genome Microarray Chip (including 30,484 human genes, Agilent Technologies). The microarray chip was incubated at 65°C for 17 hours in Agilent's microarray hybridization chambers and washed according to the manufacturer's protocol. The microarray chip was scanned at 5-μm resolution using Agilent Technologies Microarray Scanner. The spot intensities were determined with Agilent's Feature Extraction Software (v. 220.127.116.11). Processed signals from the Feature Extraction Software (v. 18.104.22.168) were used for the analysis. This software automatically normalized the signal intensity of each expressed gene using the Linear and Lowess method, subtracted the background intensities, and then flagged any outlier spots. Log ratio was expressed as Log10 (Processed Cy5 signal/Processed Cy3 signal) and P value log ratio was calculated with Rosetta Universal Error Model (Rosetta Inpharmatics).
Real-time Reverse-transcription Polymerase Chain Reaction (RT-PCR)
Total RNA was isolated from colonic biopsy samples and a cell line with Trizol reagent. Total RNA (4 μg) was reverse-transcribed using oligo-dT primer (Applied Biosystems, Branchburg, NJ), and quantitative real-time RT-PCR was performed with the ABI PRISM 7000 Sequence Detection System (Applied Biosystems) as previously reported.24, 25 The following sets of primers for human DMBT, IL-22, IL-22R1, and glyceraldehydes-3-phosphate dehydrogenase (GAPDH) were prepared: human DMBT1 5′-ATTGTGCTGCACCTGGTCAT-3′ (sense), 5′-AGCGGG AAGAGGGGTCATA-3′ (antisense), human IL-22 5′-GCA GGCTTGACAAGTCCAACT-3′ (sense), 5′-GCAGGCTTG ACAAGTCCAACT-3′ (antisense), human IL-22R1 5′-CTA CATGTGCCGAGTGAAGA-3′ (sense), 5′-ACATATCTG TAGCTCAGGTA-3′ (antisense), human GAPDH 5′-GAGT CAACGGATTTGGTCGT-3′ (sense), 5′-TTGATTTTGG AGGGATCTCG-3′ (antisense). Each amplification consisted of a 50-μL reaction mixture with 50 ng of cDNA, 500 nM of gene-specific primers, and 1 × SYBR Green Master Mix (Applied Biosystems). The PCR cycling conditions were 50°C for 2 minutes, 95°C for 10 minutes, followed by 40 cycles at 95°C for 15 seconds and 60°C for 60 seconds. The intensity of the fluorescent dye was determined and the expression levels of DMBT1 mRNA were normalized to those of GAPDH mRNA.
Western Blot Analysis
Western blot analyses were performed using each primary antibody as previously described.26 In brief, after treatment with or without reagents, cells were lysed in protein extraction buffer. Protein extract (25 μg) was fractionated by sodium dodecyl sulfate polyacrylamide gel electrophoresis, transferred to a polyvinylidene difluoride membrane, and detected using an enhanced chemiluminescence system (Amersham Biosciences, Buckinghamshire, UK).
Small interfering RNA for human STAT3 (STAT3-siRNA) was obtained from Qiagen. Cells were seeded in a 6-cm dish (Iwaki, Funabashi, Japan) and maintained for 24 hours. Then the cells were transfected with STAT3-siRNA or nonsilencing siRNA (as a control) using the Oligofectamine reagent (Qiagen) according to the manufacturer's recommendation. After incubation for 48 hours, cells were washed with phosphate-buffered saline (PBS) and treated with IL-22 in serum-free medium. The protein and total RNA were extracted from the cells and subjected to Western blot analysis and real-time RT-PCR, respectively.
The fragment of the human DMBT1 promoter from −872 to +18 (−872/+18) was cloned and inserted in the pGL3-Basic vector (Promega, Madison, WI) as reported previously.27 To obtain the deletion constructs, DMBT1 promoter fragments of −352/+18, −198/+18, −115/+18 were also cloned and inserted in the pGL3-Basic vector. Mutant plasmids (−198M1 and −198M2) were constructed to determine the responsible element for IL-22 in the DMBT1 promoter. The following oligonucleotides were used: for −198M1, 5′-CTTACGCGTTCAAGCCTGTATTC AATCGCC-3′; for −198M2, 5′-CTTACGCGTTCAAG CCTGTACATTAGAAAC-3′, the underlined sequences corresponding to modified regions. The oligonucleotide used for the opposite orientation was 5′-GCAGAT CTTTCTGCTGCTGCTATAAATC-3′ in all cases. After cloning and confirmation of the nucleotides by sequencing, the construct was named DMBT1-Luc. SW403 cells were cotransfected with DMBT1-Luc and Renilla luciferase plasmid pRL-TK and luciferase assays were performed as described previously.27
Electrophoretic Mobility Shift Assay
DNA probes for electrophoretic mobility shift assay (EMSA) were synthesized as oligonucleotides. The sequences of the individual oligonucleotides in the sense orientation were as follows: probe WT, 5′-CCTGTATTCAGG AAACACTTG-3′ corresponding to nucleotides −193 to −173 of the DMBT1 promoter gene; probe M1, 5′-CCT GTATTCAATCGC CACTTG-3′; probe M2, 5′-CCTGTAC ATTAGAAA CACTTG-3′. The probe for NF-κB was from Promega. SW403 cells were stimulated with IL-22 (10 ng/mL) for 30 minutes and nuclear protein was extracted as described previously.26 EMSA was carried out with a Gel Shift Assay System (Promega) according to the manufacturer's recommendation.
Tissue Specimens and Histological Examination
Colon biopsy samples were obtained by endoscopy from 40 patients with UC (20 men and 20 women; mean age 50.8 years, range 20–77), five patients with Crohn's disease (CD; three men and two women; mean age 37.0 years, range 26–50), eight patients with proctitis (six men and two women; mean age 52.4, range 40–79), and 10 normal controls (six men and four women; mean age 45.6 years, range 33–55) at Kyoto University Graduate School of Medicine and Tokyo Women's Medical University. All the UC patients (n = 40) examined had mesalazine treatment and 19 UC patients additionally received steroid therapy for more than 6 months at the time of tissue sampling. No other medicines including immune suppressive ones were given to any UC patients. Tissue specimens were used for real-time RT-PCR and histological analyses. This work was done with the approval of the Review Board of Kyoto University Hospital and Tokyo Women's Medical University Hospital and the Dokkyo University Surgical Pathology Committee, and informed consent was obtained from all patients. The diagnosis of UC was based on established endoscopic and histological criteria28 and the degree of inflammation was evaluated according to Matt's grade28 throughout the experiments.
Immunohistochemical stainings for IL-22, IL-22R1, and DMBT1 protein were performed as described previously,29, 30 using anti-IL-22 antibody (1:200; Santa Cruz Biotechnology), anti-IL-22R1 antibody (1:25; Abcam, Cambridge, UK), and antihuman DMBT1 antibody (1:20; Lifespan Biosciences, WA). Finally, the sections were incubated in 3,3′-diaminobenzide tetrahydrochloride with 0.05% H2O2 for 5 minutes and then counterstained with Mayer's hematoxylin.
All values are expressed as the mean ± SEM. Statistical differences between the two groups were assessed by the unpaired two-tailed t-test or by the Mann–Whitney U-test when data were not parametric. The relationship between IL-22 and DMBT1 mRNA levels was assessed by linear regression analysis. A P-value of less than 0.05 was considered statistically significant.
Microarray Analyses of Colon Cancer Cells Stimulated with IL-22
To clarify the role of IL-22 in colon epithelial cells, we examined genes whose expression was modulated by IL-22 in human colon cancer cells by microarray analyses. Twenty representative genes that were upregulated or downregulated in IL-22-treated SW403 cells are listed in Table 1. IL-22 markedly increased the gene expression of suppressor of cytokine signaling 3 (SOCS-3), DMBT1, regenerating gene Iα (REG Iα), and lipocalin 2 (LCN2), whereas it significantly decreased the expression of genes encoding a transporter protein (SLC3A1), and a transcription factor (PBX1) (Table 1).
|Accession No.||Fold Change||P-value||Symbol||Gene Name|
|Upregulated genes in the cells treated with IL-22|
|NM_003955||94.8||4.79 E−23||SOCS3||Suppressor of cytokine signaling 3|
|NM_007329||45.2||1.42 E−22||DMBT1||Deleted in malignant brain tumors 1|
|NM_002909||37.1||2.38 E−22||REG 1α||Regenerating islet-derived 1 alpha|
|NM_024795||35.8||4.10 E−22||TM4SF20||Transmembrane 4 L six family member 20|
|NM_014331||34.5||3.16 E−22||SLC7A11||Solute carrier family 7, member 11|
|NM_002165||30.7||4.17 E−22||ID1||Inhibitor of DNA binding 1|
|NM_005564||25.5||8.02 E−22||LCN2||Lipocalin 2|
|NM_033292||24.2||1.06 E−21||CASP1||Caspase 1|
|NM_006418||21.4||2.53 E−21||OLFM4||Olfactomedin 4|
|NM_001085||20.8||6.30 E−21||SERPINA3||Serine peptidase inhibitor, clade A, member3|
|NM_002089||20.6||2.05 E−21||CXCL2||Chemokine (C-X-C motif) ligand 2|
|NM_018664||18.7||4.07 E−21||SNFT||Jun dimerization protein p21SNFT|
|NM_015892||18.6||6.46 E−21||GALNAC4S-6ST||B cell RAG associated protein|
|NM_001511||18.3||3.79 E−21||CXCL1||Chemokine (C-X-C motif) ligand 1|
|NM_017933||17.4||6.38 E−21||FLJ20701||Hypothetical protein FLJ20701|
|NM_002965||16.6||2.15 E−20||S100A9||S100 calcium binding protein A9|
|THC2652075||16.1||5.36 E−20||THC2652075||Probable oligopeptide binding protein AppA precursor|
|NM_000300||13.4||3.98 E−20||PLA2G2A||Phospholipase A2, group IIA|
|BI710972||13.1||1.11 E−19||BI710972||Insulinoma Homo sapiens cDNA clone|
|NM_002090||13.0||4.37 E−20||CXCL3||Chemokine (C-X-C motif) ligand 3|
|Downregulated genes in the cells treated with IL-22|
|NM_000341||0.105||8.17 E−19||SLC3A1||Solute carrier family 3|
|NM_004923||0.113||1.25 E−17||MTL5||Metallothionein-like 5|
|ENST0000032868||0.115||1.49 E−18||PBX1||Pre-B-cell leukemia transcription factor 1|
|NM_014033||0.143||2.92 E−16||METTL7A||Methyltransferase like 7A|
|NM_001012761||0.149||3.74 E−17||RGMB||RGB domain family, member B|
|NM_144991||0.152||5.61 E−16||C21orf29||Chromosome 21 open reading frame 29|
|NM_007289||0.152||6.00 E−17||MME||Memberane metallo-endopeptidase|
|NM_000142||0.166||2.00 E−16||FGFR3||Fibroblast growth factor receptor 3|
|NM_022138||0.170||2.71 E−16||SMOC2||SPARC related modular calcium binding 2|
|NM_025149||0.173||9.51 E−16||FLJ20920||Hypothetical protein FLJ20920|
|NM_030667||0.177||5.38 E−16||PTPRO||Tyrosine phosphatase, receptor type, O|
|AI359640||0.181||1.53 E−14||AI359640||Qy33b12.x1 NCI_CGAP_Brn23|
|NM_004694||0.181||3.19 E−14||SLC16A6||Solute carrier family 16, member 6|
|NM_017577||0.182||1.05 E−15||GRAMD1C||GRAM domain containing 1C|
|NM_002614||0.182||1.44 E−15||PDZK1||PDZ domain containing 1|
|NM_015515||0.188||1.41 E−15||KRT23||Keratin 23 (histone deacetylase inducible)|
|NM_001010971||0.188||2.32 E−14||SAMD13||Sterile alpha motif domain containing 13|
|AB011115||0.189||6.92 E−15||LOC643641||mRNA for KIAA0543 protein|
|NM_012309||0.189||1.05 E−13||SHANK2||SH3 and multiple ankyrin repeat domains 2|
IL-22 Enhances DMBT1 Expression in Colon Cancer Cells by a STAT3-dependent Mechanism
The effect of IL-22 stimulation on DMBT1 mRNA expression was examined by real-time RT-PCR in colon cancer cells. The expression level of DMBT1 mRNA started to increase significantly from 12 hours after IL-22 stimulation in SW403 cells (Fig. 1A). Western blot analyses also showed that the expression of DMBT1 protein was markedly enhanced by IL-22 stimulation (Fig. 1B). We examined three candidate pathways for IL-22 signaling in colon cancer cells. IL-22 moderately activated phosphorylation of ERK and Akt and strongly induced phosphorylation of STAT3 in SW403 cells (Fig. 2).
We next examined the contribution of STAT3 signaling to IL-22-induced DMBT1 expression using an siRNA system for STAT3. As shown in Figure 3A, the increased level of p-STAT3 in IL-22-treated SW403 cells was decreased by the treatment with STAT3-siRNA. Similarly, the increased level of DMBT1 mRNA expression in IL-22-treated colon cancer cells was markedly decreased by treatment with STAT3-siRNA (Fig. 3B). Moreover, inhibition of STAT3 signaling with STAT3-siRNA reduced the increase of DMBT1 protein expression by IL-22 (Fig. 3C). These findings suggest that IL-22 enhances DMBT1 expression via the STAT3 signaling pathway in colon cancer cells.
Determination of the Element Responsive to IL-22-induced DMBT1 Promoter Activation in Colon Cancer Cells
The element responsive to IL-22-induced DMBT1 promoter activation was determined in a series of promoter deletion assays. While deletion of the 5′ DMBT1 promoter gene was extended from position −872 to −198, a significant increase of luciferase activity was sustained. However, the luciferase activity returned to the control level when the promoter region was additionally deleted up to −115 (Fig. 4A). These findings suggest that the promoter region between −198 and −115 is critical for DMBT1 promoter activation by IL-22 stimulation. To identify the precise consensus sequences for IL-22-induced DMBT1 promoter activation, we used the mutant constructs (Fig. 4B). Neither −198M1 nor −198M2 responded to IL-22 stimulation, suggesting that the element located between −187 and −179 (TTCAGGAAA) is responsible for IL-22-induced DMBT1 promoter activation (Fig. 4C). The binding of the nuclear extract to candidate nucleotides for the IL-22-responsive element was then examined by EMSA. IL-22 enhanced the binding of nuclear extracts to the probe; however, disruption of the responsive element resulted in complete loss of the binding activity (Fig. 4D). Moreover, supershift analysis clearly revealed a supershifted band when the responsive element-binding nuclear factors were treated with anti-STAT3 antibody, suggesting that IL-22-induced STAT3 protein binds to the responsive element in the DMBT1 promoter region (Fig. 4E).
NF-κB Signaling Is Partially Responsible for DMBT1 Expression in Colon Cancer Cells
The effect of IL-22 stimulation on NF-κB signaling was then examined by EMSA. As shown in Figure 5A, IL-22 stimulation induced NF-κB activation in colon cancer cells. However, concomitant treatment with the NF-κB inhibitor TPCK abolished the IL-22-induced NF-κB activation (Fig. 5A). We next examined whether inhibition of NF-κB signaling affected the expression of DMBT1 mRNA and protein. Treatment with TPCK partially decreased IL-22-induced expression of DMBT1 mRNA in colon cancer cells (Fig. 5B). Consistent with this, concomitant administration of TPCK partially decreased the IL-22-induced expression of DMBT1 protein in colon cancer cells (Fig. 5C). These findings suggest that NF-κB signaling is partially involved in IL-22-induced expression of DMBT1 in colon cancer cells.
Neither MEK-ERK nor PI3K-Akt Signaling Is Responsible for DMBT1 Expression in Colon Cancer Cells
We then examined whether MAPK or PI3K-Akt signaling is involved in IL-22-induced DMBT1 expression in colon cancer cells. The increase in the level of p-ERK and p-Akt in IL-22-treated cells was inhibited by concomitant administration of the MEK inhibitor PD 98059 and the PI3-K inhibitor wortmannin, respectively (Fig. 6A). We then examined whether inhibition of MEK-ERK or PI3K-Akt signaling affected the expression of DMBT1 mRNA and protein. IL-22 stimulation significantly enhanced the expression of DMBT1 mRNA and protein, but treatment with neither PD 98059 nor wortmannin attenuated this effect of IL-22 in colon cancer cells (Fig. 6B,C). These findings suggest that IL-22 stimulation activates both MEK-ERK and PI3K-Akt signaling, but that neither MEK-ERK nor PI3K-Akt signaling is involved in IL-22-induced DMBT1 expression in colon cancer cells.
Gene Expression of DMBT1 Is Correlated with that of IL-22 in UC Mucosa
The level of IL-22, IL-22R1 and DMBT1 mRNA expression was significantly higher in UC tissues than in normal colonic tissue. Furthermore, levels of those mRNA expressions were significantly increased in CD but not in proctitis compared with healthy control (Fig. 7A).
We next analyzed interrelationships among IL-22, IL-22R1, and DMBT1 expression and clinicopathological factors in patients with UC, and found that use of steroid therapy was associated with reduced expression of IL-22 and DMBT1 in UC mucosa. Therefore, we further subdivided UC patients according to use of steroid therapy and histological findings and analyzed the gene expression of IL-22, IL-22R1, and DMBT1 in each group. In groups with both high and low Matts scores, IL-22 and DMBT1 gene expression levels were significantly lower in UC patients with steroid therapy than in those without (Fig. 7B). In addition, in both of these groups the level of IL-22R1 gene expression tended to be lower in UC patients who had received steroid therapy. In both groups with and without steroid therapy, levels of IL-22, IL-22R1, and DMBT1 mRNA tended to become higher as the severity of inflammation increased, although not to a statistically significant degree. Neither age nor sex was significantly correlated with the expression of IL-22, IL-22R1, and DMBT1 (data not shown). We then analyzed the relationship between the expression of IL-22 mRNA and that of DMBT1 mRNA in UC tissues, and found that the two were significantly correlated (Fig. 7C).
Localization of IL-22, IL-22R1, and DMBT1 Expression in Normal Colonic and UC Mucosa
In normal colonic mucosa, hardly any immunoreactivity for IL-22 and DMBT1 was detected in the lamina propria and in the epithelium, respectively (Fig. 8A,C). The immunoreactivity for IL-22R1 was expressed in a few epithelial cells in the basal portion of crypts (Fig. 8B). In UC mucosa the number of IL-22-positive lymphocytes was increased (Fig. 8D). The numbers of IL-22R1- and DMBT1-positive cells were simultaneously increased in the basal portion of crypts in UC mucosa (Fig. 8E,F).
IL-22, secreted mainly by IL-17-producing helper (Th17) and Th1 lymphocytes,2, 31 is a cytokine unique in that its receptor, IL-22R1, is not expressed in immune cells but in epithelial cells and fibroblasts.6, 11 In the present study, as well as that of Andoh et al,6 the numbers of IL-22-positive cells were found to be increased in UC mucosa. Moreover, we found that the expression of IL-22R1 was enhanced in crypt epithelial cells in UC mucosa. These findings strongly suggest that IL-22 plays a pivotal role in the signal crosstalk between immune and crypt epithelial cells in UC mucosa. However, as the effect of IL-22 on colonic epithelial cells remained unclear, we initially carried out a microarray analysis to screen for changes in the gene expression of colon cancer cells subjected to IL-22 stimulation. Subsequently, in addition to SOCS3 (a negative regulator of STAT3 signaling), we found candidates for antimicrobial molecules (REG Iα and LCN2) as novel genes that were upregulated by IL-22 stimulation in colon cancer cells. Of note, Raffatellu et al32 have also screened IL-22-induced genes in T84 colon cancer cells using gene chip array and subsequently demonstrated that antimicrobial molecules, such as lipocalin-2 and MUC4, were strongly induced in IL-22-stimulated cells. This may suggest that IL-22 signaling plays a central role in induction of antimicrobial molecules in colonic epithelial cells.
Although the precise mechanism of IL-22 signaling has not been fully clarified, it is possible that binding of IL-22 to its receptor complex activates not only NF-κB and ERK but also STAT3 signaling. On the other hand, Rosenstiel et al23 have shown that the putative elements for NF-κB and AP-1 binding are present in the DMBT1 promoter region. Accordingly, it is reasonable to speculate that in the present study IL-22-induced NF-κB activation was partly involved in the expression of DMBT1 in colon cancer cells, although ERK activation was not involved under the experimental conditions we employed. Interestingly, our series of in vitro studies clearly showed that IL-22 stimulation enhanced the expression of DMBT1 via activation of STAT3 tyrosine phosphorylation. Moreover, we determined the element responsible for IL-22-induced DMBT1 promoter activation and confirmed that STAT3 indeed binds to its consensus oligonucleotides, suggesting that IL-22/STAT3 signaling is directly linked to DMBT1 expression. Although the dominant pathway may differ in accordance with experimental conditions, all these findings strongly suggest that DMBT1 is a novel target gene that acts downstream in the IL-22 signaling pathway.
To date, little is known about the role of DMBT1 in human diseases. However, Mollenhauer and coworkers33 have recently raised an original antibody against DMBT1 and demonstrated immunohistochemically that DMBT1 expression is enhanced in the inflamed colonic epithelium in UC mucosa, being compatible with our quantitative data for DMBT1 gene expression. Interestingly, the expression of IL-22, IL-22R1, and DMBT1 mRNA was concurrently enhanced in UC and CD but not in proctitis, and moreover, the gene expression of DMBT1 was strongly correlated with that of IL-22 in UC mucosa. Thus, these clinical data, together with our in vitro studies, suggest that the IL-22/DMBT1 axis may be specifically involved in the pathophysiology of inflammatory bowel disease. Additionally, we observed that steroid treatment was associated with a decrease of IL-22 and DMBT1 expression in UC mucosa, contrary to a previous report that IL-22 production by Th17 cells was not affected by dexamethasone treatment in bronchial tissues.34 Although we have no explanation for this discrepancy, it will be necessary to examine whether steroid directly decreases lymphocytic IL-22 and/or epithelial DMBT1 expression directly, or indirectly by modulation of innate immunity in UC mucosa.
What role does the IL-22/DMBT1 axis play in UC mucosa? Lisle et al35 and Renner et al33 independently generated a DMBT1 knockout mouse and examined the role of DMBT1 in DSS-induced colitis. Lisle et al showed that loss of the DMBT1 gene did not affect the severity of DSS-induced colitis.35 In contrast, Renner et al showed that loss of DMBT1 gene enhanced susceptibility to such colitis.33 In this regard, it still remains conflicting whether DMBT1 plays a role in intestinal mucosal protection and prevention of inflammation. However, accumulating evidence indicates that DMBT1 binds and aggregates a broad spectrum of microorganisms in the gastrointestinal tract and inhibits the cytoinvasion of pathogenic bacteria.23, 36–38 Furthermore, DMBT1 has been reported to suppress the production of inflammation in the colonic mucosa.33 These findings suggest that DMBT1 may play a cytoprotective role in UC mucosa. On the other hand, it is noteworthy that IL-22 has been reported to ameliorate intestinal inflammation in a mouse model of UC.12, 13 Moreover, it is interesting that all of DMBT1, REG Iα, and LCN2, which were isolated as novel genes upregulated by IL-22 in our microarray analysis, are reported to confer an antimicrobial effect on colonic epithelial cells.23, 39, 40 Taken together, the evidence suggests that IL-22 signaling may play a pivotal role in the antimicrobial response of UC mucosa.
In summary, DMBT1 and other antimicrobial genes have been shown to be upregulated by IL-22 stimulation in colon cancer cells. IL-22 stimulation enhanced DMBT1 expression through activation of STAT3 and NF-κB. Moreover, we have shown that the expressions of IL-22 and DMBT1 are concurrently enhanced and positively correlated in the inflamed UC mucosa. These data suggest that the IL-22/DMBT1 axis plays an important role in innate immunity in UC mucosa.
We thank Chiaki Matsuyama, Ayako Shimizu, Takako Ono, Midori Katayama, Atsuko Kikuchi, and Nozomi Nagashima (Department of Surgical and Molecular Pathology, Dokkyo University School of Medicine, Tochigi, Japan) for excellent technical and secretarial assistance.
- 20CRP-ductin, the mouse homologue of gp-340/deleted in malignant brain tumors 1 (DMBT1), binds gram-positive and gram-negative bacteria and interacts with lung surfactant protein D. Eur J Immunol. 2003; 33: 2327–2336., , , et al.
- 28The value of rectal biopsy in the diagnosis of ulcerative colitis. Q J Med. 1961; 120: 393–407..
- 35Effect of Muclin (Dmbt1) deficiency on the gastrointestinal system. Am J Physiol. 2008; 294: G717–G727., , , et al.