Akt signaling pathway in inflammation-induced colorectal carcinogenesis.

Am J Physiol Gastrointest Liver Physiol 306: G229–G243, 2014. First published December 12, 2013; doi:10.1152/ajpgi.00484.2012.

Identification of a microRNA landscape targeting the PI3K/Akt signaling pathway in inflammation-induced colorectal carcinogenesis Claire Josse,1* Nassim Bouznad,1* Pierre Geurts,2 Alexandre Irrthum,2 Vân Anh Huynh-Thu,2 Laurence Servais,1 Alexandre Hego,1 Philippe Delvenne,3 Vincent Bours,1 and Cécile Oury1 1

GIGA-Research, Human Genetics Unit, University of Liège, Liège, Belgium; 2GIGA-Research, Systems and Modeling, University of Liège, Liège, Belgium; and 3GIGA-Research, Laboratory of Experimental Pathology, University of Liège, Liège, Belgium Submitted 19 December 2012; accepted in final form 4 December 2013

Josse C, Bouznad N, Geurts P, Irrthum A, Huynh-Thu VA, Servais L, Hego A, Delvenne P, Bours V, Oury C. Identification of a microRNA landscape targeting the PI3K/Akt signaling pathway in inflammation-induced colorectal carcinogenesis. Am J Physiol Gastrointest Liver Physiol 306: G229 –G243, 2014. First published December 12, 2013; doi:10.1152/ajpgi.00484.2012.—Inflammation can contribute to tumor formation; however, markers that predict progression are still lacking. In the present study, the well-established azoxymethane (AOM)/dextran sulfate sodium (DSS)-induced mouse model of colitis-associated cancer was used to analyze microRNA (miRNA) modulation accompanying inflammation-induced tumor development and to determine whether inflammation-triggered miRNA alterations affect the expression of genes or pathways involved in cancer. A miRNA microarray experiment was performed to establish miRNA expression profiles in mouse colon at early and late time points during inflammation and/or tumor growth. Chronic inflammation and carcinogenesis were associated with distinct changes in miRNA expression. Nevertheless, prediction algorithms of miRNAmRNA interactions and computational analyses based on ranked miRNA lists consistently identified putative target genes that play essential roles in tumor growth or that belong to key carcinogenesisrelated signaling pathways. We identified PI3K/Akt and the insulin growth factor-1 (IGF-1) as major pathways being affected in the AOM/DSS model. DSS-induced chronic inflammation downregulates miR-133a and miR-143/145, which is reportedly associated with human colorectal cancer and PI3K/Akt activation. Accordingly, conditioned medium from inflammatory cells decreases the expression of these miRNA in colorectal adenocarcinoma Caco-2 cells. Overexpression of miR-223, one of the main miRNA showing strong upregulation during AOM/DSS tumor growth, inhibited Akt phosphorylation and IGF-1R expression in these cells. Cell sorting from mouse colons delineated distinct miRNA expression patterns in epithelial and myeloid cells during the periods preceding and spanning tumor growth. Hence, cell-type-specific miRNA dysregulation and subsequent PI3K/ Akt activation may be involved in the transition from intestinal inflammation to cancer. microRNA; mouse model; inflammation; colorectal cancer; computational analyses MICRORNAS (miRNAs) are short (⬃22 nucleotides) noncoding single-stranded RNAs (7), which are highly conserved among mammals. miRNAs modulate gene expression by destabilizing mRNA and/or inhibiting translation. They predominantly bind to the 3= untranslated region (UTR) of their target mRNA, with an imperfect base complementarity. For this reason, each

* C. Josse and N. Bouznad contributed equally to this work. Address for reprint requests and other correspondence: C. Oury, Univ. of Liège, GIGA-Research, Ave. de l’Hopital 1, B-4000 Liège, Belgium (e-mail: [email protected]). http://www.ajpgi.org

miRNA can bind to many mRNA targets, thereby regulating around one-third of the genome (40, 66). miRNAs are transcribed and processed in response to extracellular stimuli, or during developmental stages, in a tightly regulated manner. They are implicated in many cellular processes, including inflammation and cancer. Several studies show variations in miRNA expression in human colorectal cancer (CRC), which largely depend on the origin and characteristics of the tumor (2, 59). Inflammatory conditions are increasingly being acknowledged to contribute to tumor formation. Inflammation is even included as the seventh hallmark of cancer development (13, 27). Patients diagnosed with inflammatory bowel disease (IBD) have an 18% increased risk of developing a CRC during the following 30 years (71). Chronic inflammation is associated with an increased risk of progression to epithelial dysplasia and CRC in patients with IBD, but there is a limited understanding of the mechanisms that are involved in the transition from intestinal inflammation to cancer (4). The azoxymethane (AOM)/dextran sulfate sodium (DSS) mouse model is a widely used and well-characterized model of human colitis-associated cancer (CAC) (15, 23). AOM/DSSinduced tumors display very similar histological and molecular features to human CRC (15). Depending on the genetic background of the mice, the tumor-induction method comprises a single intraperitoneal injection of a carcinogen, AOM, followed by one or three cycles of DSS administration via the drinking water. DSS treatment leads to intestinal epithelial barrier disruption and the establishment of chronic inflammation. After 10 –20 wk, and only when AOM is administered, colorectal tumors grow in a manner that recapitulates the aberrant crypt foci-adenoma-carcinoma sequence seen in human CRC (15). Using this mouse model, Greten et al. (23) and Grivennikov et al. (24) showed that interactions among tumor cells, the inflammatory microenvironment, and immune cells appear to be critical for tumor progression (52, 58). In the present study, we used this mouse model to establish miRNA expression profiles in mouse whole colon at early and late time points during inflammation and/or tumor growth. We identified 77 differentially expressed miRNAs, defined as miRNA landscape. Prediction algorithms and computational analyses based on ranked miRNA lists revealed potential regulation of genes that play essential roles in tumor growth or that are involved in key carcinogenesis networks or signaling pathways. We found that inflammation-triggered dysregulation of key miRNAs can affect the cancer-related PI3K/Akt and IGF-1 signaling pathways. In addition, we showed variations

0193-1857/14 Copyright © 2014 the American Physiological Society

G229

G230

MICRORNA IN COLITIS-ASSOCIATED CANCER

of miRNA expression patterns between epithelial and myeloid cell types during the periods preceding and spanning tumor growth. MATERIALS AND METHODS

Animal procedures. Male C57Bl/6J mice (5 wk old) were obtained from Charles River Laboratories. Carcinogen (AOM; Sigma-Aldrich) was diluted in PBS and injected intraperitoneally at 12 mg/kg on the first day of the procedure. The animals were maintained on a standard rodent diet, and chronic inflammation was induced by three cycles of DSS (2%) administration via the drinking water for 5 days, followed by 16 days of regular water. On day 65 or day 100, mice were euthanized and the colons were excised, flushed with PBS, and 1) fixed as “Swiss rolls” (rectum part inside and cecal part outside) in 10% formalin at 4°C overnight and paraffin embedded (FFPE); 2) conserved in RNALater (Ambion-Applied Biosystems) until total RNA extraction; or 3) embedded as Swiss rolls in Tissue-Tek (Gentaur) and conserved at ⫺80°C. For some animals, macroscopic tumors were dissected at day 100 and compared with adjacent tissue taken from an ⬃1-cm2 surface area around the tumor. FFPE tissue sections (4 ␮m) were stained with hematoxylin/eosin and observed by a pathologist. Ethical statements. This study was carried out in strict accordance with the guidelines for the care and use animals set out by the European Union. The protocol was approved by the Committee on the Ethics of Animal Experiments of the University of Liège (permit no. 1220). RNA extraction. Colons were removed from the RNALater solution, wiped, and cut into small pieces. Homogenization/lysis was performed by using Tripur (Roche, Vilvoorde, Belgium) in a Qiagen Tissue Lyser with a stainless steel bead (Qiagen, Leiden, The Netherlands). Total RNA was extracted according to the manufacturer’s instructions. Total RNAs were extracted from Caco-2 cells or sorted colon cell populations by using the miRNeasy mini Kit (Qiagen) according to the manufacturer’s instruction. miRNA microarray. miRNA microarray analysis was performed by Miltenyi Biotech using 1 ␮g of total RNA per sample and MiRXplore (miRBase 13.0). Each RNA sample was hybridized in dual color to enable comparison with a commercial internal reference standard (miRXplore Universal Reference). The use of an internal reference enables the comparison of each sample or group with all others. The signal ratio of sample to internal reference was calculated. After examination of the raw data, miRNAs that did not show a ⬎1.3-fold increase in expression above background levels in all four technical replicates of the 50 RNA samples were discarded from the analyses. In total, 211 miRNAs remained after this filtering process. The miRXplore Universal Reference (Miltenyi Biotech) did not contain every miRNA present in the miRBase 13.0 database. Because the distribution of the ratio of miRNAs present in (172 over 211), or absent from (39 over 211), the Universal Reference was different, normalization of the ratios was performed separately on these two groups as follows: the ratios were log-transformed and then standardized (i.e., mean-subtracted and divided by the standard deviation, with the means and standard deviation calculated separately for the miRNA on each array). Significant differences in miRNA expression under the two different conditions were detected via the Mann-Whitney U-test. Raw P values were then false discovery rate (FDR) adjusted for multiple testing by Benjamini and Hochberg’s method. miRNAs with P ⬍0.05 and at least 1.6-fold expression variation were considered to be significantly affected by the treatment. All analyses were performed with use of MATLAB (The MathWorks, Natick, MA). The data presented in this publication have been deposited in NCBI’s Gene Expression Omnibus (18) and are accessible through

GEO Series accession number GSE38443 (http://www.ncbi.nlm.nih. gov/geo/query/acc.cgi?acc⫽GSE38443). Identification of genes regulated by pools of miRNA. Given a ranked list of N miRNAs, a P value is associated to each gene that is a potential target of the miRNA in the list as follows: 1) the miRNAs targeting a particular gene are retrieved from TargetScan Version 5.0 (conserved species, http://www.targetscan.org) and miRanda software packages (43), only genes predicted by the two algorithms are retained; (2) the sum of inverted ranks of the miRNAs targeting the gene is computed, where the inverted rank of a miRNA is defined as N⫺i⫹1, with i being the rank of the miRNA in the ranked list of N miRNAs (i is taken as N⫹1 for miRNAs not present in the ranked list); 3) this sum is then compared with equivalent sums obtained for 100,000 random sets of miRNAs (drawn from miRNAs present on the array) of the same size as the set of miRNAs targeting the gene, and the P value is calculated as the number of these 100,000 sums that are greater than the original sum; 4) the resulting P values for all the genes are FDR adjusted for multiple testing by the method of Benjamini and Hochberg. These analyses were performed with in-house software developed in Python. Reverse-transcription and real-time PCR. The miRCURY LNA Universal RT microRNA PCR system (Exiqon) was used to perform reverse transcription and quantification of miRNA. Primer sets specific for mmu-miR-223, mmu-miR-29c, mmu-miR-148a, mmu-miR-34a, mumiR-142–5p, mmu-miR-146a, mmu-miR-193, mmu-miR-216a, mmumiR-133a, mmu-miR-133b, mmu-miR-143, mmu-miR-145, hsa-miR223, hsa-miR-133a, hsa-miR-133b, hsa-miR-143, hsa-miR-145, and endogenous control primers 5S rRNA, RNU1A1, and U6 snRNA or SNORD65 and SNORD110, were purchased from Exiqon. Reactions were performed according to the manufacturer instructions. Normalization was performed by using the mean of three endogenous reference genes. The RevertAid H Minus First Strand cDNA Synthesis Kit (Fermentas) was used to perform reverse transcription. Quantification of murine IGF1R mRNA was done by real-time PCR using the SYBR Premix exTaq II, Perfect Real Time (Takara, Lonza, Verviers, Belgium) according to the manufacturer instructions. Normalization was performed using the endogenous reference Gapdh gene expression level. Primers sequences were for Igf1r forward 5=-AGCTGATGTGTACGTGCCTGATGA-3= and reverse 5=-TGATGGCCACTCTGGTTTCGGGT-3=, and for Gapdh forward 5=-TGTGTCCGTCGTGGATCTGA-3= and reverse 5=-CCTGCTTCACCACCTTCTTGA-3=. Real-time PCR reactions were carried out on a LightCycler480 (Roche). Human cell lines. The human monocytic leukemia cell line THP-1 was grown in RPMI 1640 culture medium (Lonza) supplemented with 10% fetal bovine serum (FBS, Lonza) and 1% penicillin and streptomycin (Lonza) at 37°C in 5% CO2 in a humidified incubator. For conditioned medium preparation, cells were stimulated with 1 ␮g/ml of Ultra Pure Escherichia coli LPS (InvivoGen, Toulouse, France) for 24 h. Cells were harvested and supernatant was collected. Human colorectal adenocarcinoma cells (Caco-2) were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% FBS and 1% penicillin and streptomycin. Cells were cultured for 48 h in the presence of the conditioned medium of the LPS-stimulated THP-1 cells. Western blotting. Proteins were extracted from 60 ␮M tissue slices. In this case, colons were cryopreserved and embedded in Tissue-Tek (Gentaur) without fixation. Proteins were extracted in RIPA buffer [150 mM NaCl; 1% Nonidet P-40 (NP-40); 0.5% sodium deoxycholate, 2% SDS, 50 mM Tris pH 8, 1 mM EDTA pH 8] containing an anti-protease cocktail (Complete Protease Inhibitor Cocktail Tablets, Roche). Extracts were heated for 20 min at 98°C followed by 2 h at 60°C. Caco-2 cell proteins were extracted on ice by using the following lysis buffer: 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 50 mM Tris, 0.1% SDS, containing Halt Protease Inhibitor Cocktail and Halt Phosphatase Inhibitor Cocktail (Fisher

AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00484.2012 • www.ajpgi.org


Scientific, Erembodegem, Belgium). Protein concentrations were measured using the MicroBCA kit (Pierce, Thermo Scientific). Protein extracts (20 ␮g) were electrophoresed on 8% SDS-PAGE gels and subjected to classical Western blotting with an anti-MEKK5/ ASK1 (NH2-terminal) rabbit monoclonal antibody (Epitomics, BioConnect, The Netherlands), anti-phospho-Akt (Ser473) antibody (Cell Signaling Technology, Danvers, MA), or anti-IGF1R rabbit polyclonal antibody (Sigma). Immunohistochemistry. Cryopreserved Tissue-Tek (Gentaur)-embedded or FFPE 4-␮m colon sections mounted on SuperFrost Plus slides were used along with the DAKO EnVision⫹ Detection SystemHRP (DAB) (Dako) according to the manufacturer’s instructions. The primary antibodies were the anti-NUFIP2 polyclonal rabbit antibody (Novus Biological, Bio-Connect, The Netherlands), antiIGF1R (Ab-1161) rabbit polyclonal antibody (Sigma-Aldrich), and anti-phospho-Akt (Ser473) antibody (Cell Signaling Technology). Sections were counterstained with hematoxylin. Mouse colon cell sorting. Cell sorting from mouse colon of mice was realized mainly as described (64). In brief, the colon segment of the gastrointestinal tract was removed and flushed with Ca2⫹- and Mg2⫹-free HBSS. The colon was cut longitudinally, placed in Ca2⫹and Mg2⫹-free HBSS containing 10% FBS-5 mM EDTA-25 mM HEPES, and shaken vigorously at 37°C for 30 min. The tissue was cut into 1-cm segments and incubated in digestion buffer containing 2.4 mg/ml collagenase A (Roche Diagnostics, Indianapolis, IN) and 0.2 mg/ml DNase I (Roche Diagnostics) in RPMI 1640 for 45 min on a shaker at 37°C. Following incubation, the cell aggregates were dissociated by filtering thorough a 19-gauge needle and 70-mm filter and centrifuged at 1,200 rpm for 20 min at 4°C. The supernatant was decanted and the cell pellet resuspended in PBS. Cells were labeled with the following antibodies from eBioscience company (Vienna, Austria): anti-CD326 (EpCAM) eFluor 450, anti-CD11b PerCP-cyanine 5.5, anti-Gr-1-PE (clone RB6 – 8C5), anti-mouse CD45, and LY-6G (clone 1A8) FITC (STEMCELL Technologies, Grenoble, France). The FACSAria IIIu 4L SORP sorter was used (BD Biosciences, Erembodegem, Belgium). Lentivirus and lentiviral transduction. A genomic sequence of 342 bp encompassing the hsa-pre-miR-223 and a shuffled sequence of the hsa-pre-miR-223 were cloned with use of the In-Fusion Advantage PCR Cloning Kit (Clontech) into a lentiviral plasmid derived from pLentiLox 3.7, under the control of the EF-1␣ promoter. The genomic sequence of hsa-miR-223 was amplified from Caco-2 cell gDNA with the following primers: forward, 5=-CCAGAACACAGGCCAACCTGGCCTGCTGCCCAGTGGAGGTTCC-3= containing a SfiI site; reverse, 5=-GACTGCATAAGCTAGCATGCACCCCAGAGCTGCATCGTTCCT-3= containing a SphI site. Two complementary DNA oligomers containing the shuffled sequence of hsa-pre-miR-223 were synthesized (Integrated DNA Technologies) and annealed. The sequence was scramble hsa-pre-mir-223/5Phos/TGGCCTAAGATATGGTTTCCCACTAACCGAGGAGTCAGCCGGGTAATTATGCCATCCGAGTCTTTCCCTCGGAACCGTGATAACCGGTGCGCGCTGAACTTTTGGCTGCCCGTACGCATG (120 bp) and contained a SfiI site at the 5= end and a XhoI site at the 3= end. VSV-G pseudotyped second generation lentivirus was produced by cotransfection of either the hsa-pre-miR-223 plasmid or the scr-pre-miR-223 plasmid described above, the VSV-G-containing plasmid psPAX2, and the packaging plasmid R8.91. Viruses were concentrated by high-speed centrifugation, and titers were determined by quantitative PCR (qPCR) using the qPCR Lentiviral titer kit from ABGood (Gentaur). Caco-2 cells (ATCC, LGC Standards) and LentiX cells (Fisher Scientific) were grown in DMEM (Lonza) supplemented with 10% FBS (Lonza). Caco-2 cells transductions were performed at MOI ⫽ 20 on 7.5 ⫻ 105 target cells with 8 ␮g/ml polybrene (Sigma) at 37°C for 16 h. Target Protector. miScript Target Protectors and negative control Target Protector were purchased from Qiagen (Leiden). According to predicted microRNA target and target downregulation scores (www.

G231

microrna.org, miRanda and miRSVR), two Target Protectors were designed: the first miScript Target protects the binding site for miR-223 in the IGF-1R mRNA 3=-UTR (GeneID: 3480 chr15: 99500881–99500902). The second miScript Target protects a common binding site for the miR-133a, miR133b, and miR-145 in the IGF-1R mRNA 3=-UTR (Chr15: 99504664 –99504685). Transfections of Caco-2 cells were performed with the HiPerfect Transfection Reagent according to the manufacturer’s instructions. RESULTS

miRNA expression profiles in the AOM/DSS mouse model of CAC. Three cycles of DSS subsequent to a single AOM pretreatment resulted in the development of colon tumors in nearly 100% of treated mice (Fig. 1, A and B). From experimental days 65 to 100, tumors (mainly tubular adenomas with low- or high-grade dysplasia, average size 3– 4 mm) were observed in the median and rectal part of the colon. Few tumors exhibited features of adenocarcinomas or infiltrated the submucosa. In agreement with previous studies, colonic tumors were never found in animals treated with three DSS cycles without AOM pretreatment (51). These animals developed chronic colitis characterized by a thickened mucosa and muscularis, crypt loss, submucosal edema, immune cell infiltration, and disruption of the intestinal barrier (Fig. 1B). Mice treated with AOM alone displayed normal colon histology, identical to that of PBS-treated mice. To study the evolution of miRNA expression during the periods preceding and spanning tumor development, AOM/ DSS-treated mice were euthanized on experimental days 15, 29, 50, and 65 (Fig. 1C). Whole colons were excised and processed for RNA/protein extraction and histological analyses. Animals treated with DSS, AOM alone, or vehicle (PBS) were euthanized on days 15 and 65. MiRXplore microRNA microarray (miRBase 13.0) was used to establish miRNA expression profiles of whole colons from each group of mice. Twenty-one pairs of conditions were compared by the Mann-Whitney U-test (Supplemental Table S1; supplemental material for this article is available online at the Journal website). We found that the expression of 71 miRNAs was affected by AOM/DSS treatment, most in a time-dependent manner (Table 1). The greatest number of changes (53 miRNAs) occurred between days 15 and 65, when tumors developed. Thirty-five miRNAs were differentially regulated between days 50 and 65, whereas 25 miRNAs expression changes were identified when the effect of 65-day treatment with AOM/DSS was compared with PBS. The expression of 14 miRNAs was significantly different between mice treated with DSS or PBS for 65 days (Table 2). DSS treatment for 15 days did not cause significant modulation of miRNA expression. miRNA expression profiles under AOM/DSS conditions clearly differed from those obtained under DSS conditions (Tables 1 and 2). Of the 71 carcinogenesis-associated miRNAs identified, only eight were modified by DSS treatment, one of them showing opposite expression changes (mmu-miR-451). Five of the six chronic colitis-associated miRNAs (mmu-miR133A, mmu-miR-133B, mmu-miR-143, mmu-miR-145, and mmu-miR-345–5p) were all negatively regulated, apart from the mmu-miR-21, which was upregulated. These results suggest that specific miRNA pools may regulate tumor growth.


G232


Fig. 1. Mouse model of colitis-associated cancer. A: azoxymethane (AOM)/dextran sulfate sodium (DSS) treatment protocol. B: hematoxylin/eosin-stained colon sections from mice treated with vehicle (PBS), AOM (AOM), DSS (3 cycles, DSS), or with a combination of AOM and DSS. PBS and AOM treatment did not alter colon histology. Chronic DSS treatment led to epithelial disruption, inflammatory infiltration, submucosal edema, and disruption of the intestinal barrier. AOM/DSS led to the development of tubular adenocarcinoma. Scale bars: 140 ␮m. C: experimental design of the microRNA (miRNA) microarray experiment. D: Venn diagram representing the distribution of modulated miRNAs among the indicated experimental conditions.

To evaluate the relevance of our data to human pathology, we compared miRNAs differentially expressed in whole colon of AOM/DSS-treated mice with those associated with human CRC in the literature (Table 3) (3, 5, 6, 10, 16, 17, 20, 26, 39, 41, 45, 46, 48, 56, 57, 63). As many as 24 miRNAs identified in the present AOM/DSS mouse model are reportedly associated with human CRC, with only a few differences in expression variation (up- or downregulation). Ten AOM/DSS-regulated miRNAs are altered in human IBDs (Table 4) (8, 9, 19, 33, 53, 60, 65). Among these miRNAs, miR-146a, miR-146b, miR-150, and miR-223 are also differentially expressed upon mouse treatment with three DSS cycles, indicating that the inflammatory insult is sufficient to alter these miRNAs. Inversely, DSS treatment diminishes the expression of miR-133a, miR-143, and miR-145, which are associated with human CRC (30, 36). We next used miRCURY LNA Universal RT microRNA PCR to validate our microarray results. We chose eight miRNAs: miR-223, miR-142–5p, miR-148a, mir-34a, miR-29c, miR-146a, miR-193, and miR-216a (Fig. 2A), five of which

were associated with human CRC pathology (Table 3) and three with human IBD (Table 4). For most of these miRNAs, the PCR analyses confirmed the microarray results. The expression levels of mir-34a, miR-142–5p, miR-146a, miR-148a, and miR-223 were significantly increased in whole colon at day 65 of AOM/DSS protocol compared with day 15, whereas a downregulation of miR-216a was observed under the same conditions (Fig. 2A). miR-142–5p, miR-146a, miR-148a, and miR-223 were significantly upregulated at day 65 compared with age-matched PBS-treated control mice. These PCR experiments also confirmed that three DSS cycles significantly upregulated miR-223 and miR-142–5p and downregulated miR-133a, mir-133b, miR-143, and miR-145 compared with control mice (Fig. 2B). Genes potentially regulated by pools of miRNA. miRNAs can be functionally included in networks. We investigated whether we could identify proteins involved in CAC on the basis of networking miRNA properties. To achieve this, we considered the modulated miRNAs as a whole, later referred to as the “miRNA climate.” Using predicted miRNA-mRNA


G233


Table 1. Changes in mouse colon microRNA expression induced by AOM/DSS treatment AOM/DSS d65 vs. AOM/DSS d15

AOM/DSS d65 vs. PBS d65

AOM/DSS d65 vs. AOM/DSS d50

miRNA

Ratio

P value

Ratio

P value

Ratio

P value

MMU-LET-7E MMU-LET-7I* MMU-MIR-15A MMU-MIR-17 MMU-MIR-18A MMU-MIR-19A MMU-MIR-24-2* MMU-MIR-29C MMU-MIR-34A MMU-MIR-96 MMU-MIR-99A MMU-MIR-99B MMU-MIR-101A MMU-MIR-106A MMU-MIR-124 MMU-MIR-125A-3P MMU-MIR-139-5P MMU-MIR-140* MMU-MIR-142-5P MMU-MIR-146A MMU-MIR-146B MMU-MIR-148A MMU-MIR-150 MMU-MIR-151-5P MMU-MIR-152 MMU-MIR-182 MMU-MIR-183 MMU-MIR-193 MMU-MIR-193B MMU-MIR-196B MMU-MIR-199A-5P MMU-MIR-203 MMU-MIR-205 MMU-MIR-212 MMU-MIR-214 MMU-MIR-216A MMU-MIR-223 MMU-MIR-292-3P MMU-MIR-302B* MMU-MIR-322/424 MMU-MIR-328-5P MMU-MIR-337-3P MMU-MIR-342-3P MMU-MIR-361 MMU-MIR-375 MMU-MIR-376C MMU-MIR-384-5P MMU-MIR-421-3P MMU-MIR-425 MMU-MIR-451 MMU-MIR-494 MMU-MIR-497 MMU-MIR-503 MMU-MIR-574-3P MMU-MIR-654-3P MMU-MIR-675-3P MMU-MIR-679 MMU-MIR-685 MMU-MIR-686 MMU-MIR-705 MMU-MIR-714 MMU-MIR-762 MMU-MIR-871 MMU-MIR-1194 MMU-MIR-1224 MMU-MIR-1895 MMU-MIR-1931 MMU-MIR-1937A MMU-MIR-1940 MMU-MIR-1959 MMU-MIR-1983

0.88 1 1.38 1.31 1.84 1.8 0.6 0.76 1.04 1.22 1.1 0.94 0.66 1.47 0.88 2.94 1.05 1.87 5.14 2.56 4.38 0.98 3.98 1.08 1.02 1.13 1.78 0.62 0.75 1.2 1.2 1.96 1.28 2.1 2.49 0.83 3.97 0.71 0.58 0.61 3.36 0.68 2.64 1.38 0.98 0.56 0.52 0.86 1.32 1.43 2.44 1.27 0.85 0.63 0.65 0.73 0.72 2.07 0.72 3.34 0.76 3.13 0.7 0.94 5.48 2.1 6 0.99 0.93 4.96 2.1

0.054 0.0335 0.054 0.0335 0.054 0.0335 0.0335 0.2392 0.6179 0.5014 1 0.5014 0.319 0.319 0.1196 0.0335 1 0.0335 0.0335 0.0335 0.0335 0.3982 0.0335 1 0.8701 0.6179 0.0335 0.0335 0.319 0.7628 0.0827 0.0827 0.1196 0.0335 0.0335 0.5014 0.0335 0.0335 0.1196 0.2392 0.1196 0.319 0.0335 0.054 0.319 0.5014 0.1196 0.2392 0.1689 0.1689 0.0335 0.1689 0.1689 0.054 0.1196 0.0335 0.1196 0.1689 0.0827 0.0335 0.0335 0.0827 0.2392 0.2392 0.0335 0.319 0.0335 0.1196 0.0335 0.0335 0.2392

1.77 0.72 2.41 1.52 1.25 1.82 1.86 2.57 3.8 4.26 2.85 2.27 0.4 2.49 0.54 1.86 2.42 2.28 3.88 4.51 2.05 4.1 2.69 2.6 2.66 3.44 1.56 1.7 2.72 2.61 1.8 2.67 4.26 1.61 1.8 0.4 1.65 0.58 0.55 0.55 1.65 0.48 2.44 2.05 2.38 0.42 0.53 0.51 2.45 3.69 1.13 2.18 0.44 0.52 0.44 0.43 0.55 2.15 0.46 2.53 0.57 2.46 0.52 0.51 2.38 2 3.61 0.9 0.56 4.36 2.58

0.0465 0.0342 0.0465 0.751 0.8701 0.0465 0.0985 0.0465 0.0342 0.0342 0.0342 0.0342 0.0342 0.0342 0.0465 0.2053 0.0342 0.0465 0.0342 0.0342 0.0342 0.0342 0.0698 0.0342 0.0342 0.0342 0.2912 0.2053 0.0342 0.0465 0.0465 0.0698 0.0342 0.2912 0.1467 0.0342 0.2912 0.0465 0.0342 0.0342 0.3797 0.0342 0.0465 0.0342 0.0465 0.0342 0.0342 0.0342 0.0985 0.0342 0.751 0.0342 0.0465 0.0465 0.0342 0.0342 0.0465 0.0342 0.0465 0.1467 0.0342 0.0698 0.0465 0.0698 0.0465 0.0465 0.0342 0.0342 0.0465 0.0342 0.1467

1.12 0.59 1.45 1.78 2.33 2.36 1.17 0.89 1.57 2.11 1.62 1.35 0.28 1.78 0.48 1.71 1.34 1.87 2.64 2.19 2.65 1.43 2.83 1.72 1.62 1.9 1.59 0.91 0.96 1.63 1.15 2.58 1.74 1.88 1.76 0.51 2.53 0.6 0.62 0.49 2.17 0.66 2.09 1.81 1.26 0.62 0.64 0.61 2.08 1.41 2.45 1.28 0.53 0.42 0.51 0.44 0.57 2.46 0.5 2.14 0.63 5.08 0.57 0.4 1.89 2.2 3.24 0.58 0.52 6.96 3.63

1 0.0399 0.0399 0.0399 0.0399 0.0399 0.8493 0.0985 0.0985 0.0399 0.2014 0.2014 0.0568 0.0568 0.1446 0.0797 0.4877 0.0399 0.0399 0.0399 0.0399 0.0985 0.0568 0.1446 0.2014 0.0399 0.0399 0.4877 0.605 0.2014 0.2014 0.0399 0.0399 0.0399 0.0399 0.0399 0.0399 0.0797 0.0985 0.0568 0.0399 0.0399 0.0568 0.0568 0.1446 0.0985 0.0985 0.0399 0.0399 0.605 0.0399 0.2774 0.0985 0.0399 0.0568 0.0399 0.0985 0.0399 0.0797 0.0399 0.1446 0.0399 0.0568 0.0399 0.0797 0.0399 0.0399 0.0399 0.0399 0.0399 0.0399

P values were obtained by a Mann-Whitney test and corrected for multitesting by the Benjamini-Hochberg method. MicroRNA (miRNA) with P ⬍ 0.05 in at least 1 comparison, azoxymethane (AOM)/dextran sulfate sodium (DSS) day 65 (d65) vs. PBS d65, AOM/DSS d65 vs. AOM/DSS day 15 (d15), or AOM/DSS d65 vs. AOM/DSS day 50 (d50), and with at least 1.6-fold of expression variation are listed. *miRNA star.


G234


Table 2. Changes in mouse colon microRNA expression induced by DSS treatment DSS treatment for 65 days as compared to control mice (PBS)

Table 4. List of miRNAs differentially expressed in whole colons of AOM/DSS- or DSS-treated mice and in human IBD

DSS d65 vs. PBS d65 miRNA miRNA

Ratio

MMU-MIR-21 MMU-MIR-133A MMU-MIR-133B MMU-MIR-142-5P MMU-MIR-143 MMU-MIR-145 MMU-MIR-146A MMU-MIR-146B MMU-MIR-150 MMU-MIR-223 MMU-MIR-342-3P MMU-MIR-345-5P MMU-MIR-451 MMU-MIR-705

1.61 0.42 0.42 3.82 0.58 0.52 2.07 2.32 4.10 3.69 2.01 0.57 0.40 2.82

0.0465 0.0465 0.0465 0.0465 0.0465 0.0465 0.0465 0.0465 0.0465 0.0465 0.0465 0.0465 0.0465 0.0465

P values were obtained by a Mann-Whitney test and corrected for multitesting by the Benjamini-Hochberg method. miRNA with P ⬍ 0.05 and with at least 1.6 fold of expression variation are listed.

interactions (identified by combining results from TargetScan 5.0 and miRanda algorithms), we searched for genes that would be regulated by the “microRNA climate” of CAC more significantly than would occur by chance. The miRNAs were ranked such that more statistical importance was given to those Table 3. List of miRNAs differentially expressed in whole colons of AOM/DSS-treated mice and in human colorectal cancer miRNA

AOM/DSS Mouse Model

let-7e miR-17 miR-18a miR-19a miR-29c miR-34a

up up up up up up

miR-96 miR-99b miR-101a miR-106a

up up down up

miR-139-5p miR-146a miR-146b miR-148a

up up up up

miR-150

up

miR-182 miR-183 miR-203 miR-205 miR-212

up up up up up

miR-214 miR-223 miR-328-5P miR-375

up up up up

AOM/DSS Mouse Model

Human IBD

Reference

P value

Human CRC

Reference

down up up up down up down up up down up down down up down up down up down up up up up up down down up down down

10 20, 46 20, 5, 48, 46, 10 5, 48, 6, 10 20 5, 6, 45 3 5, 6, 45, 26 63 20, 56 63, 57, 5, 48, 6, 10, 45 16 5, 6, 45, 10, 26 6, 45 56 6, 45 20 45, 63 10 5, 6, 45 46 63, 57, 5, 10, 45 45, 10 57, 56 10 6, 10 20, 57, 48, 45, 10, 17, 63, 39 6 20

Only mouse colon miRNAs significantly modified by AOM/DSS treatment (P ⬍ 0.05, see Table 1) are considered for comparison.

miR-29c

up

miR-106a

up

miR-140* miR-146a miR-146b miR-150

up up up

miR-196b

up

miR-223

up

miR-328-5p miR-375

up up

miRNA

19

up

65 19

up up

33 19

up

19

down down up up

8 19 19 33 9 65 19 19 33 19 65

active UC active UC active CD active CD inactive CD active CD inactive CD active UC

up

down

65 60 65 19 19 19 19 53

up

active CD inactive CD

up

19

up

active CD inactive CD active UC inactive UC active CD active UC inactive CD inactive UC

up

19

down up

8 19 65 19

down

19

DSS Mouse Model

up

miR-133b

down

miR-143 miR-145 miR-146a

down

miR-223

up

down down up down

miR-21

miR-146b miR-150

active CD inactive CD active CD active CD inactive CD active CD active CD inactive CD active CD inactive CD active UC inactive UC inactive UC active UC active CD active CD active UC inactive CD inactive UC active CD active UC active CD

up

Human IBD

up

IBD, inflammatory bowel disease; UC, ulcerative colitis; CD, Crohn’s disease. Only mouse colon miRNA significantly modified by the AOM/DSS or DSS treatment (P ⬍ 0.05, see Table 1 and 2, respectively) are considered for comparison. *miRNA star.

presenting the highest number of significant modulations in our 21 comparisons (Supplemental Table S1). The miRNAs expressed but not modulated, and those that were not expressed, received an equal nearly null rank. Computational analyses (see MATERIALS AND METHODS for details) then identified 299 genes that were potentially regulated (P ⬍0.05) by the miRNA climate expressed under our experimental conditions (Supplemental Table S2). In an attempt to identify genes specifically regulated by carcinogenesis-associated miRNAs, miRNAs were ranked according to five AOM/DSS-based comparisons: AOM/DSS d65 vs. AOM/DSS d15, AOM/DSS d65 vs. PBS d65, AOM/DSS d15 vs. PBS d15, AOM/DSS d15 vs. AOM d15, and AOM/


G235


Fig. 2. RT-quantitative PCR (qPCR) analysis of miRNA expression in mouse whole colons. A: miRNA expression levels in the colon of AOM/DSS- and PBS-treated mice at the indicated times (d65, day 65; d15, day 15). B: expression levels of miRNAs in whole colons of mice that received 3 DSS cycles or not (day 65). n ⫽ 5, *P ⬍0.05; **P ⬍0.01 (Mann-Whitney test). Median and interquartile intervals are represented for each condition.

DSS d65 vs. AOM d65, where d represents day. The same ranking was performed for chronic colitis: miRNAs were ranked on the basis of the following comparisons: DSS d65 vs. DSS d15, DSS d65 vs. PBS d65, DSS d15 vs. PBS d15, DSS d15 vs. AOM d15, and DSS d65 vs. AOM d65. In all, 412 genes were identified as potentially regulated by DSS treatment (Supplemental Table S3) compared with 170 genes regulated by AOM/DSS (Supplemental Table S4). Interestingly, although the pools of miRNAs clearly differed between AOM/ DSS and DSS conditions, 139/170 AOM/DSS genes were found in the list of genes potentially regulated by DSS-related miRNAs. Next, the Ingenuity Pathway software was used to identify the major pathways, biological functions, and protein networks related to the three lists of genes (Fig. 3). The murine database was used to generate the analyses. Interestingly, cancer-associated biological functions were shared by the three lists of genes: abnormal morphology, cell cycle, growth and proliferation, and cell death. Some of the predicted targets were part of common networks involved in tissue development, cell death and survival, cell-to-cell signaling and interaction and cell morphology/movement. The main pathways included IGF-1, insulin receptor, PTEN, and PI3K/Akt signaling, as well as

Myc and MAPK signaling. In addition, the genes potentially regulated by the DSS miRNA pool were involved in Wnt/␤catenin signaling. These results suggest that both inflammation and carcinogenesis-associated miRNA climates regulate the expression of genes involved in cancer development. Relationship between miRNA dysregulation and PI3K/Akt pathway in colorectal adenocarcinoma cells. Previous studies indicated that downregulation of miR-143/145 and miR-133a results in the activation of PI3K/Akt signaling and subsequent cancer cell proliferation (14, 49). Since DSS decreases the expression levels of these miRNAs (Fig. 2B and Table 2), we wanted to assess whether an inflammatory environment alters their expression in cancer colon cells. We found that incubation of Caco-2 cells with conditioned medium of activated monocytic THP-1 cells downregulated miR-143/145 and miR133a in these cells (Fig. 4A) and increased Akt phosphorylation (Fig. 4B). Thus inflammation-triggered miRNA dysregulation could promote cancer cell proliferation through PI3K/Akt. miR-223 is upregulated in human CRC, IBD, and the IL-10 knockout mouse model of IBD (28). In the AOM/DSS mouse model, we found that miR-223 was one of the main miRNAs showing significant upregulation during tumor growth (Table 1). Overexpression of miR-223 in Caco-2 cells was conducted


G236


Fig. 3. Ingenuity Pathway Analysis of genes identified as more likely to be regulated by the pool of miRNAs showing expression changes in any conditions tested (colitis and associated cancer) (Supplemental Table S2) (A), by the pool of miRNAs modulated by AOM/DSS treatment (Supplemental Table S4) (B), or by the miRNA pool modulated by 3 successive DSS treatments (Supplemental Table S3) (C). Only top pathways are shown, which correspond to the first 10 best P values.

via lentivirus infection. Transduced cells expressed lower levels of phosphorylated Akt than cells transduced with scramble miRNA (Fig. 4C), depicting a link between alteration of miR-223 and PI3K/Akt signaling. Activation of the PI3K/Akt signaling pathway during AOM/ DSS tumor growth. Inhibition of PI3K suppresses tumor growth in CRC (44). To experimentally verify that the PI3K/ Akt signaling pathway is activated in our mouse model, we performed immunohistochemical analyses on mouse colon sections using an antibody directed against active Ser473-

phosphorylated Akt. We found that AOM/DSS treatment induced Akt phosphorylation both in tumor and stromal cells (Fig. 5). DSS treatment alone also increased phosphorylated Akt levels in aberrant crypt foci (Fig. 5). Expression changes of three selected candidate target genes in the CAC mouse model. From the lists of target genes potentially regulated by miRNA climates, we selected IGF-1R and MAP3K5 on the basis of the Ingenuity Pathway Analysis. In addition, NUFIP2 was chosen as a new gene candidate of unknown function in cancer development. Immunostaining


G237


Fig. 4. Impact of miRNA dysregulation on PI3K/Akt signaling in Caco-2 cells. A: RT-qPCR analysis of miR-133a and miR-145 expression in Caco-2 cells cultured for 48 h in the presence or absence of conditioned medium (CM) from LPS-activated THP-1 cells. Data are expressed as means ⫾ SD. Results are from 3 independent experiments (n ⫽ 6). ***P ⬍ 0.0001 vs. control. C: Western blotting detection of phospho(Ser473)-Akt performed on protein extracts from these cells. B: Western blotting experiment performed on protein extracts of Caco-2 cells transduced with a miR-223- or scramble miRNA(Scr-miR-223) expressing lentivirus. Gapdh protein expression is used as the loading control. Data are representative of 3 independent experiments.

was performed on colon sections from PBS- and AOM/DSStreated mice (Fig. 6A). Under control conditions, IGF1R expression was detected in the cryptic epithelial cells, which did not change significantly upon AOM/DSS treatment at early

stage of tissue transformation. In tumors, IGF1R expression was significantly decreased, whereas staining of stromal and inflammatory cells was observed between the crypts. The expression of Nufip2 protein in the colon layers was globally increased by AOM/DSS treatment and, to a larger extent, in tumors. AOM/DSS treatment caused a marked decrease of Map3k5 expression, as shown by Western blotting on whole colon extracts (Fig. 6B) and immunohistochemistry (data not shown). These data support tight regulation of the three selected genes during tumor development, which is likely to depend on CAC-related changes in the miRNA climate. To assess whether IGF1R is indeed regulated by miRNAs in colon cells, IGF1R expression was analyzed in Caco-2 cells overexpressing miR-223. Transduced cells expressed lower levels of IGF1R than cells transduced with scramble miRNA (Fig. 6C). The use of a miScript Target Protector interfering with the miR-223-IGF1R mRNA 3=-UTR interaction could rescue IGF1R protein expression in Caco-2 cells, therefore demonstrating that miR-223 directly targets IGF1R (Fig. 6D). On the contrary, a miScript Target that protects a common binding site for miR-133a, miR133b, and miR-145 in the IGF1R mRNA 3=-UTR hardly impacted on IGF1R protein expression. In mice, we observed a striking inverse relationship between changes of IGF1R mRNA and miR-223 expression during AOM/DSS-induced carcinogenesis and in tumors compared with adjacent tissues (Fig. 6E). Cell-type-specific expression of miRNAs in AOM/DSS mouse colons. miRNA function can vary in a cell-type-specific manner, especially considering colon leukocytes and transformed epithelial cells. To determine whether these cells differentially expressed selected miRNAs, we sorted myeloid cells (Gr1⫹ cells), and epithelial cells (CD326⫹) from mouse colons at experimental days 15, 29, 50, and 65 (Fig. 7A). In

Fig. 5. Activation of PI3K/Akt signaling in the colon of DSS- and AOM/DSS-treated mice. Immunohistochemical analyses of phospho(Ser473)-Akt in mouse colon sections after 65 days of treatment with AOM/ DSS or vehicle (PBS). Scale bars: 140 ␮m.


G238


Fig. 6. Expression of candidate target genes in colons of AOM/DSS-treated mice. A: immunohistochemical analyses of protein expression in mouse colon sections after 65 days of treatment with AOM/DSS or vehicle (PBS). Panels on the right depict tumors. Scale bars ⫽ 200 ␮m. B: representative Map3k5 Western blot performed with proteins extracted from colons of one control (PBS-treated) and 2 AOM/DSS-treated mice (65 days). Gapdh protein expression is used as the loading control. C and D: Western blot detection of IGF1R was realized with protein extracts of Caco-2 cells transduced with miR-223 or scramble miRNA (Scr-miR-223) (C) or transfected with 2 different miScript Target Protectors interfering with miR-223-IGF1R mRNA 3=-untranslated region (UTR) (Target Protector 2) or with miR-133a/miR-143/miR-145 binding to IGF1R 3=-UTR (Target Protector 1) (D). E: qPCR analyses of IGF1R mRNA expression and miR-223 expression in whole colons of AOM/DSS-treated mice for 65 days and in tumors resected at day 100. Expression levels measured in healthy tissues (Adj. Tis.) were arbitrarily set to 100 for each set of experiments (n ⫽ 5).

agreement with previous studies (68), early infiltration of myeloid cells of monocytic (Ly6C⫹CD11b⫹) and granulocytic (Ly6G⫹CD11b⫹) origin was observed (Fig. 7B). We then assessed the expression of miR-143/145, miR-133a, and miR-223 in sorted Gr1⫹ myeloid and CD326⫹ epithelial cells (Fig. 7C). In epithelial cells, miR-143/145 expression quickly

decreased at day 15, fluctuated between days 15 and 50, and was downregulated at day 65 when tumors are formed. Epithelial miR-133a was rapidly downregulated at day 15 and remained unchanged until day 65. These three miRNAs were also expressed in myeloid cells. miR-133a expression profile was mainly unchanged during tumor development, whereas a


G239


Fig. 7. miRNA expression in epithelial and myeloid cell types sorted from mouse colons. A: flow cytometry dot plots depicting gating strategy to isolate Gr-1⫹CD11b⫹ myeloid cells and CD326⫹ epithelial cells. B: estimation of percentages of leukocytes (CD45⫹), monocytic (Ly6G-CD11b⫹), and granulocytic (Ly6G⫹CD11b⫹) cells isolated from DSS- or AOM/DSS-treated mice at the indicated experimental time points. C, vehicle-treated mice. Data are expressed as means ⫾ SD (n ⫽ 6). C: RT-qPCR analysis of miR-133a, miR-143, miR-145, and miR223 expression in sorted cell populations at the indicated time points. Data are expressed as means ⫾ SD (n ⫽ 3). Expression levels measured at day 15 were arbitrarily set to 100.

downregulation of miR-145 and miR-143 was observed. miR223 expression was much higher in myeloid cells. In these cells, a time-dependent miR-223 expression profile was observed that followed the same pattern as in whole colon (Fig. 6E). Epithelial cells displayed a distinct expression pattern, characterized by an early upregulation occurring at day 15, followed by a downregulation at day 65. DISCUSSION

Using a well-established mouse model of CAC, we analyzed the miRNA expression changes accompanying chronic inflammation-dependent tumor development. We identified important inflammation- and carcinogenesis-related miRNAs potentially affecting target genes involved in the PI3K/Akt signaling

pathway, which may, therefore, represent molecular links between inflammation and cancer. In vitro experiments performed in human colorectal adenocarcinoma cells confirmed a relationship between miR-133a, miR-143, miR-145, and miR223 dysregulation and alteration of PI3K/Akt signaling. In addition, we found that miR-223 could directly target IGF1R to regulate this pathway. We chose to perform a miRNA microarray analysis on RNA samples extracted from mouse whole colon because 1) both immune and intestinal cells are expected to be involved in carcinogenesis, 2) we wanted to assess miRNA variations at time points preceding tumor development, and 3) secreted miRNA are believed to act on different cell types in a paracrine manner (11).


G240


The majority of miRNA variations mainly occurred between experimental days 15 and 65, and between days 50 and 65, emphasizing the critical importance of events preceding tumor formation. In a similar study, Necela et al. (47) investigated miRNA expression in colorectal tumors from two mouse models: one of genetic origin (APC⫹/min mice), and the second of inflammation-related origin (12 cycles of DSS treatment). They found that miR-215, miR-137, miR-708, miR-31, and miR135b were differentially expressed in APC tumors, whereas the expression of miR-215, miR-133a, miR-467d, miR-218, miR708, miR-31, and miR-135b varied in colitis-associated tumors. Among these miRNAs, miR-133a was the only one identified in the present study. In a more recent study, Gao et al. (21) used a similar AOM/DSS mouse model to ours, in which they analyzed mRNA and miRNA expression profiles. Their study revealed alterations of miRNAs involved in the inflammatory response and in metabolism; in addition, most differentially expressed genes were enriched in metabolic pathways/processes, pointing to a close relationship between metabolic and inflammatory genes in CAC. We identified only a few miRNA reported in this study (miR-21, miR-34a, miR99a, miR-193), probably owing to differences of protocol and/or mouse genetic background (BalB/c vs. C57BL/6). In contrast to our study, the authors of the two studies used scraped colonic epithelial cells at only one final time point when the tumors were already formed. The miRNA identified by Necela et al. and Gao et al. were most likely specific to the epithelial compartment; however, the studies of Greten et al. (23) and Grivennikov et al. (24) showed that immune cells play a role that is as important as that of epithelial cells in the development of CAC. Moreover, our study suggests that celltype-specific early variations of miRNA expression, occurring before tumor formation, may contribute to carcinogenesis. Many groups highlight the cooperative effects of several miRNAs, showing that integrating predicted targets and functional information could identify miRNA clusters and concurrently reveal their underlying functions (67). Yuan et al. (69) demonstrated that clustered miRNAs jointly regulate proteins that are in close proximity within a protein interaction network, and that the amount of coregulation between proteins is negatively correlated with their distance from each other within the network. On the basis of these studies, we considered miRNA modulation in our system as a whole and exploited their networking properties to identify the targets that they most likely share. We postulated that these genes could be of a particular interest in CAC pathology. We identified 170 genes, which are potentially regulated by the “CAC climate” of miRNAs. Using the pool of DSSinduced miRNAs, a list of 421 putative target genes was identified. Interestingly, 139 of the 170 CAC-related genes were on the list of DSS-related genes. Only 31 genes were putative targets of a “carcinogenesis-specific miRNA climate.” Moreover, when the 421 DSS-specific genes were analyzed in terms of pathway involvement and function by use of Ingenuity Pathway software, they appeared to be involved in cancer, apoptosis, and proliferation in the same manner as the carcinogenesis subset of target genes. miRNA-induced inflammation may thus affect many more genes than expected, thereby contributing to carcinogenesis. For instance, genes potentially regulated by the DSS miRNA pool were involved in Wnt/␤-

catenin signaling, a pathway dysregulated in AOM/DSS-induced tumors (42, 61) and in human CRC. In addition, the target genes of the Wnt/␤-catenin signaling pathway include the cellular myelocytomatosis oncogene (c-Myc), similarly found to be dysregulated in mouse and human tumors. c-Myc functions as a global mediator of the oncogenic process, linking the seemingly heterogeneous pool of molecular mechanisms that underlies cancer development (37). Overall, the predicted targets of DSS and AOM/DSS miRNA climates were part of common cancer-related networks, and the main pathways included IGF-1, insulin receptor, PTEN, and PI3K/Akt signaling, as well as Myc and MAPK signaling. These findings are in agreement with the previously reported function of PI3K in colitis and CAC (22, 32, 70). Immunohistochemical analyses indeed confirmed that this pathway is activated during AOM/DSS tumor growth. Interestingly, we found that chronic DSS treatment downregulates miR-133a, miR-143, and miR145. These miRNA were also downregulated in macrodissected tumors compared with adjacent tissues (data not shown). Downregulation of miR-133a and miR-143/145 is reported in human CRC (38, 50, 54) and results in PI3K/Akt signaling activation in breast (14) and bladder (49) cancer cells, respectively. Accordingly, we found that conditioned medium from inflammatory cells decreases the expression of these miRNA and increased Akt phosphorylation in colorectal adenocarcinoma Caco-2 cells. Validated targets of miR-143 and miR-145 that may affect the PI3K/Akt pathway include KRAS and RREB1 (36), miR-143 could directly target AKT, and miR-145 targets ILK (49). miR-133a possesses many targets that have been involved in cancer (50). Particularly, this miRNA can directly target IGF1R, as shown in myoblasts (31). IGF1R mediates PI3K/Akt activation. Deregulation of IGF1R activity is implicated in tumor cell proliferation in many human malignant diseases, including CRC (1). Currently, no data are available regarding the role of IGF1R in the AOM/DSS mouse model. The study by Knowlton et al. (38) indicates that miRNA alterations modify kinase activation in the IGF-1 pathway and correlate with CRC stage and progression in patients. Qian et al. (54) have shown that miR-143 directly targets IGF1R, and IGF1R expression inversely correlated with miR-143 levels in human CRC specimens. In our study, interfering with miR-143-IGF1R mRNA 3=-UTR interaction hardly

Fig. 8. Model depicting a proposed mechanism linking inflammation and cancer. Carcinogenesis may be triggered by inflammation-induced miRNA dysregulation in colon cells and leukocytes that would impact on proteins involved in the PI3K/Akt signaling pathway, thereby contributing to cancer cell proliferation and tumor growth.



impacted on IGF1R expression, suggesting that miR-143 can affect the PI3K/Akt pathway independently of IGF-1R signaling (see above). In the present mouse model, we found that miR-223 is one of the main miRNA showing strong upregulation when AOM/ DSS-induced tumors are formed. Strikingly, a clear downregulation is observed during the periods preceding tumor formation (days 15 to 50). miR-223 is upregulated in human CRC and IBD (Tables 3 and 4) and in mouse models of IBD (55). miR-223 is fairly specific for the hematopoietic lineage where it acts to limit inflammation and prevent oncogenic myeloid transformation (28). In our study, colon cell sorting enabled us to confirm high expression of miR-223 in myeloid cells compared with epithelial cells. Differential expression patterns were observed in the two cell types during tumor development. miR-223 has been involved in solid cancers in which opposing roles in regulating migration and invasion of cancer cells have been described (28). Injection of HeLa cells overexpressing miR-223 into nude mice suppressed proliferation and tumor growth (35). Interestingly, this suppression was mainly caused by reducing expression of the miR-223 target IGF1R and downstream Akt/mTOR/p70S6K signaling, which was also seen in leukemia and hepatoma cell lines (25). In our study, we observed a striking inverse relationship between changes of Igf1r mRNA and miR-223 expression in whole colon of AOM/DSS-treated mice and in tumors compared with adjacent tissues. The expression of miR-223 is indeed dramatically increased at stages of tumor growth when Igf1r expression decreases. Immunohistochemistry revealed that treatment with AOM/DSS decreased IGF1R expression in tumor cells whereas expression was detected in stromal and infiltrating immune cells. Expression of the Igf1r gene thus appeared to be tightly controlled during carcinogenesis in a cell-type-specific manner. We wondered whether IGF1R could be a direct target of miR-223 in colon cancer cells. Using two methods, lentivirus-based overexpression and transfection of Target Protectors, we found that IGF1R expression was directly regulated by miR-223 in colorectal adenocarcinoma cells. Colon cancer cells overexpressing miR-223 also displayed increased levels of phosphorylated Akt, indicating that targeting IGF1R with miR-223 can inhibit PI3K/Akt activation. Evaluating the importance of cell-type-specific miR-223 regulation in the development of colitis-associated cancer requires further in vivo investigation. To validate our computational method, two other genes were selected from our list of 170 genes potentially regulated by carcinogenesis-related miRNAs. These genes code for the Map3k5 (Ask1) MAP3 kinase and for the Nufip2 RNA binding protein. Map3k5 is a ubiquitously expressed MAP3 kinase activated by various stress stimuli, including reactive oxygen species. Map3k5 activates the JNK and p38 signaling pathways and induces oxidative stress and cytokine-induced apoptosis. Map3k5 is also implicated in a variety of cellular functions, including immune responses (29, 34). Using a very similar CAC mouse model, Hayakawa et al. (29) identified a protective role for Map3k5 against DSS-induced inflammation and AOM/ DSS-induced tumors, and demonstrated that Map3k5⫺/⫺ mice develop more numerous and larger tumors than wild-type mice during increased colonic inflammation. In accordance with this study, we showed reduced Map3k5 expression in whole colon of AOM/DSS-treated mice, which confirms a

G241

tumor suppressor role of this protein. Interestingly, Map3k5 is a target of miR-17 (12) that was upregulated when tumors developed in our mouse model (Table 1). Nufip2 is a RNA binding protein involved in fragile X mental retardation. Little is known about its function. The 3=-UTR of Nufip2 mRNA is unusually long, encompassing 8,231 bp (NCBI Gene ID: 68564). In humans, the 3=-UTR of Nufip2 can bind up to 19 miRNAs (TarBase, DIANA Lab) (62). The expression of Nufip2 protein was markedly increased in the colons of mice treated with AOM/DSS and remained elevated in tumors. Further investigations are required to elucidate the role of this protein in CRC and to validate regulation of its expression by miRNAs. In conclusion, on the basis of our microarray data, networking, and target prediction analyses, as well as in vitro and in vivo validation experiments, we propose a model in which chronic inflammation could induce miRNA dysregulation in colon cells and/or leukocytes that would affect proteins involved in the PI3K/Akt signaling pathway, thereby contributing to cancer cell proliferation (Fig. 8). Thus variations in miRNA expression within a biological system can be exploited to generate lists of potential target genes by a ranking and permutation method. In our study, we demonstrate that this method is a valuable tool for identifying novel epigenetic mechanisms that regulate the genes or signaling pathways involved in cancer pathobiology. ACKNOWLEDGMENTS We thank the GIGA-Genotranscriptomics, GIGA-Imaging, GIGA-Lentiviral Vector, and GIGA-Bioinformatics platforms, and the GIGA-Animal Facility. GRANTS N. Bouznard and L. Servais are Televie PhD fellows. C. Oury is a Research Associate at the Belgian Funds for Scientific Research (F.R.S.-FNRS). This work was supported by French Community of Belgium (ARC07/12-02); F.R.S.-FNRS (FRSM 3.4600.09); and F.R.S.-FNRS-Televie (7.4625.10). DISCLOSURES No conflicts of interest, financial or otherwise, are declared by the author(s). AUTHOR CONTRIBUTIONS C.J., N.B., L.S., and A.H. performed experiments; C.J., N.B., P.G., A.I., V.A.H.-T., P.D., and C.O. analyzed data; C.J., N.B., and C.O. interpreted results of experiments; C.J., N.B., and C.O. prepared figures; C.J., N.B., P.G., A.I., V.A.H.-T., P.D., V.B., and C.O. approved final version of manuscript; V.B. and C.O. edited and revised manuscript; C.O. conception and design of research; C.O. drafted manuscript. REFERENCES 1. Adachi Y, Yamamoto H, Ohashi H, Endo T, Carbone DP, Imai K, Shinomura Y. A candidate targeting molecule of insulin-like growth factor-I receptor for gastrointestinal cancers. World J Gastroenterol 16: 5779 –5789, 2010. 2. Agostini M, Pucciarelli S, Calore F, Bedin C, Enzo M, Nitti D. miRNAs in colon and rectal cancer: a consensus for their true clinical value. Clin Chim Acta 411: 1181–1186, 2010. 3. Akao Y, Nakagawa Y, Hirata I, Iio A, Itoh T, Kojima K, Nakashima R, Kitade Y, Naoe T. Role of anti-oncomirs miR-143 and -145 in human colorectal tumors. Cancer Gene Ther 17: 398 –408, 2010. 4. Anderson EC, Wong MH. Caught in the Akt: regulation of Wnt signaling in the intestine. Gastroenterology 139: 718 –722, 2010. 5. Arndt GM, Dossey L, Cullen LM, Lai A, Druker R, Eisbacher M, Zhang C, Tran N, Fan H, Retzlaff K, Bittner A, Raponi M. Characterization of global microRNA expression reveals oncogenic potential of miR-145 in metastatic colorectal cancer. BMC Cancer 9: 374, 2009.


G242


6. Bandres E, Cubedo E, Agirre X, Malumbres R, Zarate R, Ramirez N, Abajo A, Navarro A, Moreno I, Monzo M, Garcia-Foncillas J. Identification by real-time PCR of 13 mature microRNAs differentially expressed in colorectal cancer and non-tumoral tissues. Mol Cancer 5: 29, 2006. 7. Bartel DP. MicroRNAs: genomics, biogenesis, mechanism, function. Cell 116: 281–297, 2004. 8. Bian Z, Li L, Cui J, Zhang H, Liu Y, Zhang CY, Zen K. Role of miR-150-targeting c-Myb in colonic epithelial disruption during dextran sulphate sodium-induced murine experimental colitis and human ulcerative colitis. J Pathol 225: 544 –553, 2011. 9. Brest P, Lapaquette P, Souidi M, Lebrigand K, Cesaro A, VouretCraviari V, Mari B, Barbry P, Mosnier JF, Hebuterne X, HarelBellan A, Mograbi B, Darfeuille-Michaud A, Hofman P. A synonymous variant in IRGM alters a binding site for miR-196 and causes deregulation of IRGM-dependent xenophagy in Crohn’s disease. Nat Genet 43: 242–245, 2011. 10. Chen X, Guo X, Zhang H, Xiang Y, Chen J, Yin Y, Cai X, Wang K, Wang G, Ba Y, Zhu L, Wang J, Yang R, Zhang Y, Ren Z, Zen K, Zhang J, Zhang CY. Role of miR-143 targeting KRAS in colorectal tumorigenesis. Oncogene 28: 1385–1392, 2009. 11. Chen X, Liang H, Zhang J, Zen K, Zhang CY. Secreted microRNAs: a new form of intercellular communication. Trends Cell Biol 22: 125–132, 2012. 12. Chi SW, Zang JB, Mele A, Darnell RB. Argonaute HITS-CLIP decodes microRNA-mRNA interaction maps. Nature 460: 479 –486, 2009. 13. Colotta F, Allavena P, Sica A, Garlanda C, Mantovani A. Cancerrelated inflammation, the seventh hallmark of cancer: links to genetic instability. Carcinogenesis 30: 1073–1081, 2009. 14. Cui W, Zhang S, Shan C, Zhou L, Zhou Z. microRNA-133a regulates the cell cycle and proliferation of breast cancer cells by targeting epidermal growth factor receptor through the EGFR/Akt signaling pathway. FEBS J 280: 3962–3974, 2013. 15. De Robertis M, Massi E, Poeta ML, Carotti S, Morini S, Cecchetelli L, Signori E, Fazio VM. The AOM/DSS murine model for the study of colon carcinogenesis: from pathways to diagnosis and therapy studies. J Carcinog 10: 9, 2011. 16. Diaz R, Silva J, Garcia JM, Lorenzo Y, Garcia V, Pena C, Rodriguez R, Munoz C, Garcia F, Bonilla F, Dominguez G. Deregulated expression of miR-106a predicts survival in human colon cancer patients. Genes Chromosomes Cancer 47: 794 –802, 2008. 17. Earle JS, Luthra R, Romans A, Abraham R, Ensor J, Yao H, Hamilton SR. Association of microRNA expression with microsatellite instability status in colorectal adenocarcinoma. J Mol Diagn 12: 433–440, 2010. 18. Edgar R, Domrachev M, Lash AE. Gene Expression Omnibus: NCBI gene expression and hybridization array data repository. Nucleic Acids Res 30: 207–210, 2002. 19. Fasseu M, Treton X, Guichard C, Pedruzzi E, Cazals-Hatem D, Richard C, Aparicio T, Daniel F, Soule JC, Moreau R, Bouhnik Y, Laburthe M, Groyer A, Ogier-Denis E. Identification of restricted subsets of mature microRNA abnormally expressed in inactive colonic mucosa of patients with inflammatory bowel disease. PloS One 5:2010. 20. Gaedcke J, Grade M, Camps J, Sokilde R, Kaczkowski B, Schetter AJ, Difilippantonio MJ, Harris CC, Ghadimi BM, Moller S, Beissbarth T, Ried T, Litman T. The rectal cancer microRNAome—microRNA expression in rectal cancer and matched normal mucosa. Clin Cancer Res 18: 4919 –4930, 2012. 21. Gao Y, Li X, Yang M, Zhao Q, Liu X, Wang G, Lu X, Wu Q, Wu J, Yang Y, Zhang Y. Colitis-accelerated colorectal cancer and metabolic dysregulation in a mouse model. Carcinogenesis 34: 1861–1869, 2013. 22. Gonzalez-Garcia A, Sanchez-Ruiz J, Flores JM, Carrera AC. Phosphatidylinositol 3-kinase gamma inhibition ameliorates inflammation and tumor growth in a model of colitis-associated cancer. Gastroenterology 138: 1374 –1383, 2010. 23. Greten FR, Eckmann L, Greten TF, Park JM, Li ZW, Egan LJ, Kagnoff MF, Karin M. IKKbeta links inflammation and tumorigenesis in a mouse model of colitis-associated cancer. Cell 118: 285–296, 2004. 24. Grivennikov S, Karin E, Terzic J, Mucida D, Yu GY, Vallabhapurapu S, Scheller J, Rose-John S, Cheroutre H, Eckmann L, Karin M. IL-6 and Stat3 are required for survival of intestinal epithelial cells and development of colitis-associated cancer. Cancer Cell 15: 103–113, 2009.

25. Gusscott S, Kuchenbauer F, Humphries RK, Weng AP. Notch-mediated repression of miR-223 contributes to IGF1R regulation in T-ALL. Leuk Res 36: 905–911, 2012. 26. Hamfjord J, Stangeland AM, Hughes T, Skrede ML, Tveit KM, Ikdahl T, Kure EH. Differential expression of miRNAs in colorectal cancer: comparison of paired tumor tissue and adjacent normal mucosa using high-throughput sequencing. PloS One 7: e34150, 2012. 27. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell 144: 646 –674, 2011. 28. Haneklaus M, Gerlic M, O’Neill LA, Masters SL. miR-223: infection, inflammation and cancer. J Intern Med 274: 215–226, 2013. 29. Hayakawa Y, Hirata Y, Nakagawa H, Sakamoto K, Hikiba Y, Otsuka M, Ijichi H, Ikenoue T, Tateishi K, Akanuma M, Ogura K, Yoshida H, Ichijo H, Omata M, Maeda S. Apoptosis signal-regulating kinase 1 regulates colitis and colitis-associated tumorigenesis by the innate immune responses. Gastroenterology 138: 1055–1067.e1– e4, 2010. 30. Hu G, Chen D, Li X, Yang K, Wang H, Wu W. miR-133b regulates the MET proto-oncogene and inhibits the growth of colorectal cancer cells in vitro and in vivo. Cancer Biol Ther 10: 190 –197, 2010. 31. Huang MB, Xu H, Xie SJ, Zhou H, Qu LH. Insulin-like growth factor-1 receptor is regulated by microRNA-133 during skeletal myogenesis. PloS One 6: e29173, 2011. 32. Huang XL, Xu J, Zhang XH, Qiu BY, Peng L, Zhang M, Gan HT. PI3K/Akt signaling pathway is involved in the pathogenesis of ulcerative colitis. Inflamm Res 60: 727–734, 2011. 33. Iborra M, Bernuzzi F, Correale C, Vetrano S, Fiorino G, Beltran B, Marabita F, Locati M, Spinelli A, Nos P, Invernizzi P, Danese S. Identification of serum and tissue micro-RNA expression profiles in different stages of inflammatory bowel disease. Clin Exp Immunol 173: 250 –258, 2013. 34. Iriyama T, Takeda K, Nakamura H, Morimoto Y, Kuroiwa T, Mizukami J, Umeda T, Noguchi T, Naguro I, Nishitoh H, Saegusa K, Tobiume K, Homma T, Shimada Y, Tsuda H, Aiko S, Imoto I, Inazawa J, Chida K, Kamei Y, Kozuma S, Taketani Y, Matsuzawa A, Ichijo H. ASK1 and ASK2 differentially regulate the counteracting roles of apoptosis and inflammation in tumorigenesis. EMBO J 28: 843–853, 2009. 35. Jia CY, Li HH, Zhu XC, Dong YW, Fu D, Zhao QL, Wu W, Wu XZ. MiR-223 suppresses cell proliferation by targeting IGF-1R. PloS One 6: e27008, 2011. 36. Kent OA, Fox-Talbot K, Halushka MK. RREB1 repressed miR-143/145 modulates KRAS signaling through downregulation of multiple targets. Oncogene 32: 2576 –2585, 2013. 37. Knoepfler PS. Myc goes global: new tricks for an old oncogene. Cancer Res 67: 5061–5063, 2007. 38. Knowlton DL, Tang K, Henstock PV, Subramanian RR. miRNA alterations modify kinase activation in the IGF-1 pathway and correlate with colorectal cancer stage and progression in patients. J Cancer 2: 490 –502, 2011. 39. Lanza G, Ferracin M, Gafa R, Veronese A, Spizzo R, Pichiorri F, Liu CG, Calin GA, Croce CM, Negrini M. mRNA/microRNA gene expression profile in microsatellite unstable colorectal cancer. Mol Cancer 6: 54, 2007. 40. Lewis BP, Shih IH, Jones-Rhoades MW, Bartel DP, Burge CB. Prediction of mammalian microRNA targets. Cell 115: 787–798, 2003. 41. Ma Y, Zhang P, Yang J, Liu Z, Yang Z, Qin H. Candidate microRNA biomarkers in human colorectal cancer: systematic review profiling studies and experimental validation. Int J Cancer 130: 2077–2087, 2012. 42. Maltzman T, Whittington J, Driggers L, Stephens J, Ahnen D. AOMinduced mouse colon tumors do not express full-length APC protein. Carcinogenesis 18: 2435–2439, 1997. 43. Maziere P, Enright AJ. Prediction of microRNA targets. Drug Discov Today 12: 452–458, 2007. 44. Migliardi G, Sassi F, Torti D, Galimi F, Zanella ER, Buscarino M, Ribero D, Muratore A, Massucco P, Pisacane A, Risio M, Capussotti L, Marsoni S, Di Nicolantonio F, Bardelli A, Comoglio PM, Trusolino L, Bertotti A. Inhibition of MEK and PI3K/mTOR suppresses tumor growth but does not cause tumor regression in patient-derived xenografts of RAS-mutant colorectal carcinomas. Clin Cancer Res 18: 2515–2525, 2012. 45. Monzo M, Navarro A, Bandres E, Artells R, Moreno I, Gel B, Ibeas R, Moreno J, Martinez F, Diaz T, Martinez A, Balague O, GarciaFoncillas J. Overlapping expression of microRNAs in human embryonic colon and colorectal cancer. Cell Res 18: 823–833, 2008.


MICRORNA IN COLITIS-ASSOCIATED CANCER 46. Motoyama K, Inoue H, Takatsuno Y, Tanaka F, Mimori K, Uetake H, Sugihara K, Mori M. Over- and under-expressed microRNAs in human colorectal cancer. Int J Oncol 34: 1069 –1075, 2009. 47. Necela BM, Carr JM, Asmann YW, Thompson EA. Differential expression of microRNAs in tumors from chronically inflamed or genetic (APCMin/⫹) models of colon cancer. PloS One 6: e18501, 2011. 48. Ng EK, Tsang WP, Ng SS, Jin HC, Yu J, Li JJ, Rocken C, Ebert MP, Kwok TT, Sung JJ. MicroRNA-143 targets DNA methyltransferases 3A in colorectal cancer. Br J Cancer 101: 699 –706, 2009. 49. Noguchi S, Yasui Y, Iwasaki J, Kumazaki M, Yamada N, Naito S, Akao Y. Replacement treatment with microRNA-143 and -145 induces synergistic inhibition of the growth of human bladder cancer cells by regulating PI3K/Akt and MAPK signaling pathways. Cancer Lett 328: 353–361, 2013. 50. Nohata N, Hanazawa T, Enokida H, Seki N. microRNA-1/133a and microRNA-206/133b clusters: dysregulation and functional roles in human cancers. Oncotarget 3: 9 –21, 2012. 51. Okayasu I, Yamada M, Mikami T, Yoshida T, Kanno J, Ohkusa T. Dysplasia and carcinoma development in a repeated dextran sulfate sodium-induced colitis model. J Gastroenterol Hepatol 17: 1078 –1083, 2002. 52. Pages F, Galon J, Dieu-Nosjean MC, Tartour E, Sautes-Fridman C, Fridman WH. Immune infiltration in human tumors: a prognostic factor that should not be ignored. Oncogene 29: 1093–1102, 2010. 53. Pekow JR, Hetzel JT, Rothe JA, Hanauer SB, Turner JR, Hart J, Noffsinger A, Huo D, Rubin DT. Outcome after surveillance of lowgrade and indefinite dysplasia in patients with ulcerative colitis. Inflamm Bowel Dis 16: 1352–1356, 2010. 54. Qian X, Yu J, Yin Y, He J, Wang L, Li Q, Zhang LQ, Li CY, Shi ZM, Xu Q, Li W, Lai LH, Liu LZ, Jiang BH. MicroRNA-143 inhibits tumor growth and angiogenesis and sensitizes chemosensitivity to oxaliplatin in colorectal cancers. Cell Cycle 12: 1385–1394, 2013. 55. Schaefer JS, Montufar-Solis D, Vigneswaran N, Klein JR. Selective upregulation of microRNA expression in peripheral blood leukocytes in IL-10⫺/⫺ mice precedes expression in the colon. J Immunol 187: 5834 – 5841, 2011. 56. Schepeler T, Reinert JT, Ostenfeld MS, Christensen LL, Silahtaroglu AN, Dyrskjot L, Wiuf C, Sorensen FJ, Kruhoffer M, Laurberg S, Kauppinen S, Orntoft TF, Andersen CL. Diagnostic and prognostic microRNAs in stage II colon cancer. Cancer Res 68: 6416 –6424, 2008. 57. Schetter AJ, Leung SY, Sohn JJ, Zanetti KA, Bowman ED, Yanaihara N, Yuen ST, Chan TL, Kwong DL, Au GK, Liu CG, Calin GA, Croce CM, Harris CC. MicroRNA expression profiles associated with prognosis and therapeutic outcome in colon adenocarcinoma. JAMA 299: 425– 436, 2008. 58. Sica A, Allavena P, Mantovani A. Cancer related inflammation: the macrophage connection. Cancer Lett 267: 204 –215, 2008.

G243

59. Slattery M, Wolff E, Hoffman M, Pellatt D, Milash B, Wolff R. MicroRNAs and colon and rectal cancer: differential expression by tumor location and subtype. Genes Chromosomes Cancer 50: 196 –206, 2011. 60. Takagi T, Naito Y, Mizushima K, Hirata I, Yagi N, Tomatsuri N, Ando T, Oyamada Y, Isozaki Y, Hongo H, Uchiyama K, Handa O, Kokura S, Ichikawa H, Yoshikawa T. Increased expression of microRNA in the inflamed colonic mucosa of patients with active ulcerative colitis. J Gastroenterol Hepatol 25, Suppl 1: S129 –S133, 2010. 61. Takahashi M, Nakatsugi S, Sugimura T, Wakabayashi K. Frequent mutations of the beta-catenin gene in mouse colon tumors induced by azoxymethane. Carcinogenesis 21: 1117–1120, 2000. 62. Vergoulis T, Vlachos IS, Alexiou P, Georgakilas G, Maragkakis M, Reczko M, Gerangelos S, Koziris N, Dalamagas T, Hatzigeorgiou AG. TarBase 6.0: capturing the exponential growth of miRNA targets with experimental support. Nucleic Acids Res 40: D222–D229, 2012. 63. Volinia S, Calin GA, Liu CG, Ambs S, Cimmino A, Petrocca F, Visone R, Iorio M, Roldo C, Ferracin M, Prueitt RL, Yanaihara N, Lanza G, Scarpa A, Vecchione A, Negrini M, Harris CC, Croce CM. A microRNA expression signature of human solid tumors defines cancer gene targets. Proc Natl Acad Sci USA 103: 2257–2261, 2006. 64. Waddell A, Ahrens R, Steinbrecher K, Donovan B, Rothenberg ME, Munitz A, Hogan SP. Colonic eosinophilic inflammation in experimental colitis is mediated by Ly6C(high) CCR2(⫹) inflammatory monocyte/ macrophage-derived CCL11. J Immunol 186: 5993–6003, 2011. 65. Wu F, Zikusoka M, Trindade A, Dassopoulos T, Harris ML, Bayless TM, Brant SR, Chakravarti S, Kwon JH. MicroRNAs are differentially expressed in ulcerative colitis and alter expression of macrophage inflammatory peptide-2 alpha. Gastroenterology 135: 1624 –1635.e24, 2008. 66. Xie X, Lu J, Kulbokas EJ, Golub TR, Mootha V, Lindblad-Toh K, Lander ES, Kellis M. Systematic discovery of regulatory motifs in human promoters and 3=-UTRs by comparison of several mammals. Nature 434: 338 –345, 2005. 67. Xu J, Li CX, Li YS, Lv JY, Ma Y, Shao TT, Xu LD, Wang YY, Du L, Zhang YP, Jiang W, Li CQ, Xiao Y, Li X. miRNA-miRNA synergistic network: construction via co-regulating functional modules and disease miRNA topological features. Nucleic Acids Res 39: 825–836, 2011. 68. Yang XD, Ai W, Asfaha S, Bhagat G, Friedman RA, Jin G, Park H, Shykind B, Diacovo TG, Falus A, Wang TC. Histamine deficiency promotes inflammation-associated carcinogenesis through reduced myeloid maturation and accumulation of CD11b⫹Ly6G⫹ immature myeloid cells. Nat Med 17: 87–95, 2011. 69. Yuan X, Liu C, Yang P, He S, Liao Q, Kang S, Zhao Y. Clustered microRNAs’ coordination in regulating protein-protein interaction network. BMC Syst Biol 3: 65, 2009. 70. Zhang J, Roberts TM, Shivdasani RA. Targeting PI3K signaling as a therapeutic approach for colorectal cancer. Gastroenterology 141: 50 –61, 2011. 71. Zisman TL, Rubin DT. Colorectal cancer and dysplasia in inflammatory bowel disease. World J Gastroenterol 14: 2662–2669, 2008.


Serrated pathway in colorectal carcinogenesis.

HOTAIR signaling pathway.

MACC1 promotes carcinogenesis of colorectal cancer via β-catenin signaling pathway.

Mismatch repair pathway: molecules, functions, and role in colorectal carcinogenesis.

AKT signaling pathway.

STAT3 Signaling in Colorectal Carcinogenesis.

CD95L signaling pathway: a role in carcinogenesis.

β-Catenin Signaling Pathway in Skin Carcinogenesis and Therapy.

AKT pathway in colorectal cancer.

Colorectal Carcinogenesis, Radiation Quality, and the Ubiquitin-Proteasome Pathway.

CD80-CD28 signaling controls the progression of inflammatory colorectal carcinogenesis.

Signaling specificity in the Akt pathway in biology and disease.

Colorectal carcinogenesis.

Colorectal cancer and NF-κB signaling pathway.

Apoptosis prediction via inhibition of AKT signaling pathway by neogrifolin.

GRP94-mediated signaling in gastric cancer.

Baicalein inhibits the migration and invasion of colorectal cancer cells via suppression of the AKT signaling pathway.

β-catenin signaling pathway.

AKT signaling pathway.

The Epigenomics of Embryonic Pathway Signaling in Colorectal Cancer.

Hedgehog signaling pathway in colorectal cancer: function, mechanism, and therapy.

g silencing activates AKT signaling to promote gastric carcinogenesis.

PI3K-AKT signaling is a downstream effector of retinoid prevention of murine basal cell carcinogenesis.

IL-6 Pathway in the Processing of Colorectal Carcinogenesis in Colitis Mice.