mir-153 supports colorectal cancer progression via ... · in this study, we hypothesized that...

Molecular and Cellular Pathobiology

miR-153 Supports Colorectal Cancer Progression viaPleiotropic Effects That Enhance Invasion andChemotherapeutic Resistance

Lei Zhang1, Karen Pickard1, Veronika Jenei1, Marc D. Bullock1, Amanda Bruce1, Richard Mitter3,Gavin Kelly3, Christos Paraskeva4, John Strefford1, John Primrose1, Gareth J. Thomas1,Graham Packham1, and Alex H. Mirnezami1,2

AbstractAlthough microRNAs (miRNA) have been broadly studied in cancer, comparatively less is understood about

their role in progression. Here we report that miR-153 has a dual role during progression of colorectal cancer byenhancing cellular invasiveness and platinum-based chemotherapy resistance. miRNA profiling revealed thatmiR-153 was highly expressed in a cellular model of advanced stage colorectal cancer. Its upregulation was alsonoted in primary human colorectal cancer compared with normal colonic epithelium and in more advancedcolorectal cancer stages compared with early stage disease. In colorectal cancer patients followed for 50 months,21 of 30 patients with high levels of miR-153 had disease progression compared with others in this group with lowlevels ofmiR-153. Functional studies revealed thatmiR-153 upregulation increased colorectal cancer invasivenessand resistance to oxaliplatin and cisplatin both in vitro and in vivo. Mechanistic investigations indicated thatmiR-153 promoted invasiveness indirectly by inducing matrix metalloprotease enzyme 9 production, whereas drugresistance was mediated directly by inhibiting the Forkhead transcription factor Forkhead box O3a (FOXO3a). Insupport of the latter finding, we found that levels of miR-153 and FOXO3a were inversely correlated in matchedhuman colorectal cancer specimens. Our findings establish key roles for miR-153 overexpression in colorectalcancer progression, rationalizing therapeutic strategies to target expression of this miRNA for colorectal cancertreatment. Cancer Res; 73(21); 6435–47. �2013 AACR.

IntroductionColorectal cancer is the second commonest cause of cancer-

related death in western societies (1). Disease progression andmetastases are the principal causes of death, and advanceddisease occurs in up to 30% at presentation (2). In addition, inpatients with localized disease who proceed to apparentlycurative surgery, 50% subsequently develop recurrence (2,3). Despite increasingly sophisticated techniques for therapy

to recurrent colorectal cancer, the majority of patients remainincurable, highlighting the need for a continued effort to betterunderstand the complexities of disease progression, and iden-tify new directions for treatment strategies.

MicroRNAs (miRNA) are short noncoding RNA moleculesand play a critical role in malignant transformation, under-scored by the observation that over 50% of miRNA genes arelocated within or close to cancer-associated genomic regions(4, 5). miRNAs regulate gene expression by binding to the 30-untranslated region (30-UTR) of protein-coding mRNAs thr-ough sequences in the 50-end of the miRNA that are onlypartially complementary (6). One consequence of this imper-fect complementarity is that eachmiRNA can regulatemultipletarget mRNAs and thereby impact functionally diverse pro-grams of gene expression. miRNA expression is deregulated incolorectal cancer with a growing number of oncogenes andtumor suppressor genes undermiRNA regulation (7, 8). Impor-tantly, miRNA expression patterns correlate with distinctcolorectal cancer subtypes and clinical phenotype (8–11).Emerging data have also implicated miRNAs in disease pro-gression and acquisition of metastatic capabilities in nonco-lorectal malignancies, through promotion of epithelial–mes-enchymal transition and regulation of metastasis-associatedgenes (12, 13). To date however, little is understood about therole of miRNAs in disease progression in colorectal cancer andfew candidate miRNAs and target genes have been implicated.

Authors' Affiliations: 1University of Southampton Cancer Sciences Divi-sion, Somers Cancer Research Building; 2Department of Colorectal Sur-gery, Southampton University Hospital NHS Trust, Southampton; 3Bioin-formatics Unit, London Research Institute, Cancer Research UK, London;and 4School of Cellular and Molecular Medicine, University of Bristol,Medical Sciences Building, Bristol, United Kingdom

Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).

L. Zhang, K. Pickard, V. Jenei, G.J. Thomas, and G. Packham contributedequally to this work.

Corresponding Author: Alex H. Mirnezami, Cancer Research UK andRoyal College of Surgeons of England Clinician Scientist, Somers CancerResearch Building, University of Southampton Cancer Sciences Division,Southampton University Hospital NHS Trust, Tremona Road, Southamp-ton, Hampshire SO166YD, UK. Fax: 44-23-8079-4020; E-mail:[email protected]

doi: 10.1158/0008-5472.CAN-12-3308

�2013 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org 6435

on October 25, 2020. © 2013 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst August 15, 2013; DOI: 10.1158/0008-5472.CAN-12-3308

http://cancerres.aacrjournals.org/

In this study, we hypothesized that deregulated miRNAscontribute to colorectal cancer progression. miR-153 wasidentified as overexpressed in more advanced colorectal can-cer using cell models and human tumor samples. Functionaland mechanistic studies show that overexpression of miR-153can increase colorectal cancer invasiveness through upregula-tion of matrix metalloprotease enzyme 9 (MMP-9), and addi-tionally lead to enhanced platinum-based chemotherapy resis-tance. This latter effect may be through an inhibitory effect onthe Forkhead box O3 (FOXO3) transcription factor, which weshow is targeted bymiR-153, and inversely correlateswithmiR-153 in human colorectal cancer samples. Together, theseresults point to a significant and novel contribution of upre-gulated miR-153 in colorectal cancer, and suggest that mod-ulation of miR-153 may be a useful strategy in limiting colo-rectal cancer progression.

Materials and MethodsCell andorganotypic cultures, transfections, and cloning

Details of the cell lines, cloning, and transfections areprovided in supplementary methods (14–16, 44). Organotypiccultures were prepared as previously described (17). Gelscomprised a 50:50 mixture of Matrigel and type I collagenwith 5 � 105 human fetal fibroblast cells, to which were added5 � 105 SW480 cells stably transfected with 2 mg plasmidexpression constructs for miR-153 or control scrambledmiRNA (CmiR00001-MR04; GeneCopoeia). Media were chan-ged every 2 days, and gels harvested and formalin fixed after14 days.

Patients and samplesSamples from patients with biopsy-proven colorectal cancer

were obtained fresh at the time of surgery and snap frozenbefore being deposited in a UK Human Tissue Act approvedtumor bank. In each case, matched tumor and uninvolved pro-ximal mucosa were obtained. All patients provided informedconsent, and the study was approved by the regional researchethics committee. Pathologic verification of diagnosis andstaging was in accordance with the Association of Coloproc-tology of Great Britain and Ireland guidelines (18). All speci-mens were reviewed by clinicians with a specialist interest ingastrointestinal pathology using paraffin-embedded tissuethat was adjacent to or in close proximity to tumor banktissue. Tumors used in this study were adenocarcinomas only,and selected at random from the tumor bank before lasercapture microdissection (LCM) and RNA extraction. Exclusioncriteria included evidence of a hereditary tumor, presence ofmultiple tumors, and tumors with histologically identifiedextensive necrosis. A total of 100 human tumors were exam-ined. RNA extraction from tumors is described in supplemen-tary methods.

miRNA expression profiling, in silico analyses, andbioinformatics

miRNA expression profiling and in silico analyses are detailedin supplementary methods. Array data are MIAME compliantand available in the EBI database (http://www.ebi.ac.uk/arrayexpress/experiments/; accession number E-MEXP-3270).

Array-based comparative genomic hybridizationArray-based comparative genomic hybridization was con-

ducted as previously described and is detailed in supplemen-tary methods (19).

Cell lysis, real-time PCR, Western blotting, andfunctional assays

Cell lysis, downstream analyses, and functional assays todetermine biological effects of overexpression of miR-153 aredescribed in supplementary methods.

Tissue microarray development andimmunohistochemistry

Tissue microarray development and immunostaining isdetailed in supplementary methods.

In vivo xenograft invasion assayDevelopment of a severe combined immunodeficient (SCID)

mouse miR-153 knockdown xenograft model is detailed insupplementary methods.

ResultsMicroRNA expression profiling in SW480 and SW620cells

Expression of 328 human miRNAs was determined in thepaired colorectal cancer lines SW480 and SW620. Figure 1Aand B show that the top 20 miRNAs up- or downregulated>4-fold in the metastatic SW620 cells compared with SW480.Some candidates identified have been previously found to bederegulated in colorectal cancer by other groups (9–11). Tofurther verify microarray data, 9 miRNAs of variable foldchange were selected for real-time PCR validation. Consistentwith the microarray data, similar fold changes in the selectedmiRNAs were found with high correlation between microarrayand PCR results (Supplementary Fig. S1A and S1B).

Expression of miR-153 is increased in colorectal cancerand advanced disease

To prioritize differentially expressed miRNA candidates forfurther study, putative mRNA targets of the top 15 miRNAswere predicted computationally using TargetScan (www.tar-getscan.org). This gene list was annotated and then subjectedto gene-set enrichment analysis using the DAVID bioinformat-ics resource (see supplementary methods) to identify overrep-resented biological processes for each candidate miRNA. Thisdata were collated and compared for all 15 overexpressedmiRNA candidates, to inspect for enrichment of biologicalprocesses associated with disease progression and cancerspread (summarized in Supplementary Fig. S1C). One miRNAidentified in our cell line profiling as upregulated over 12-foldin higher stage colorectal cancer was miR-153. Clustered andnonclustered analyses for miR-153 show that based on a Pvalue of 0.001, 55% of the 36 biological processes identifiedwere cancer associated. In addition, the highest enrichmentscore obtained for all candidate miRNAs was associated withmiR-153 and the biological processes of motility and adhesionwith an enrichment score of 12.5, suggesting that deregulation

Zhang et al.

Cancer Res; 73(21) November 1, 2013 Cancer Research6436




Figure 1. miR-153 is upregulated in advanced stage colorectal cancer cell lines and human tumors. A and B, miRNA expression profiles were determined inSW480andSW620 cells. Upregulated (A) and downregulated (B)miRNAs in themetastatic SW620 cell line comparedwith SW480.miRNAspreviously linked tocolorectal cancer are indicatedwith an asterisk. C, endogenousmiR-153expressionbyquantitative RT-PCR in a panel of colorectal cancer cell lines (�,P < 0.05;��, P < 0.01). D and E, expression levels of miR-153 were examined by qPCR in 83 human samples comprising 23 normal mucosa; 20 stage 1 (T1-2/N0/M0tumors); 20 stage 3 or 4 (any T/N1-2 or M1); and 20metastases (liver and lung). Significantly increased miR-153 expression was noted in tumor compared withnormal tissue andwith increasing disease stage (P < 0.005). F, disease-free survival for patients according to low or high expression ofmiR-153 (Kaplan–Meier).

miR-153 Supports Colorectal Cancer Progression

www.aacrjournals.org Cancer Res; 73(21) November 1, 2013 6437




of miR-153 may have important consequences in carcinogen-esis and disease progression.

Further assessment with Panther, BioCarta, and KEGGannotation categories for miR-153 in DAVID showed enrich-ment of genes associated specifically with colorectal cancer(P < 0.005), and the Wnt, Cadherin, and TGF-b (P < 0.001)signaling pathways, lending additional support for a linkbetween miR-153 deregulation and colorectal tumorigenesisin particular (Supplementary File S1).

Prompted by this, we investigated expression of miR-153 ina panel of human colorectal cancer cell lines. Five of 10 celllines examined showed significant upregulation of miR-153levels compared with SW480 cells (Fig. 1C). The pattern ofmiR-153 expression closely correlated with the expression ofmatching cell lines in the NCI-60 panel as determined throughthe web application CellMiner (43). To determine miR-153expression in vivo, we conducted a small scale pilot to examine10 random tumors of varying stage by LCM and miRNAexpression analysis. Significantly increased expression ofmiR-153 was noted in tumor compared with normal tissue(Supplementary Fig. S2A; P < 0.0001). As transition betweenSW480 and SW620 cells principally reflects a change toward ametastatic phenotype, we examined expression of miR-153 inmetastatic compared with nonmetastatic tumors. Expressionof miR-153 in 5 nonmetastatic stage 1 tumors was comparedwith 5 stage 3 and 5 stage 4 tumors with lymphatic or distantorgan metastases (Supplementary Fig. S2B). Higher expres-sion ofmiR-153was noted inmore advanced colorectal cancer(P < 0.005).

We extended our analysis to an additional unselected groupof 83 human samples with known clinical outcome (clinico-pathologic data presented in Supplementary Table S1) com-prising 23 normal mucosa; 20 stage 1 (T1-2/N0/M0 tumors); 20stage 3 or 4 (any T/N1-2 or M1); and 20 metastases (liver andlung). Significantly increased miR-153 expression was noted intumor compared with normal tissue (Fig. 1D; P < 0.005) andwith increasing disease stage. Specifically, higher levels werenoted in tumorswith nodal or distant organ spread (Fig. 1E; P<0.005). When median expression of miR-153 was used tostratify patients, after a follow up time of 50 months, 21 of30 patients in the high miR-153 group developed diseaseprogression compared with 9 of 30 in the low miR-153 group(Fig. 1F). Factors associated with improved disease-free sur-vival by univariate analysis included the absence of extramuralvascular invasion (EMVI) and low miR-153 expression (logrank P ¼ 0.007; x2 ¼ 7.3; Supplementary Table S2). Demo-graphic characteristics, histological grading, and the type ofsurgery did not have a significant impact on survival. Multi-variate analysis indicated that both EMVI and miR-153 statuswere independent prognostic factors [for EMVI, P < 0.001;HR¼ 9.6; 95% confidence interval (CI), 3.63–49.12; for miR-153status, P ¼ 0.024; HR ¼ 4.7; 95% CI, 1.17–8.74].

miR-153 overexpression is not due to genetic copynumber change

Deregulation of miRNAs has been attributed to genomiccopy number changes (5) and miRNAs are overrepresented inregions of genomic gain in colorectal cancer (20). We therefore

sought to determine if miR-153 upregulation in colorectalcancer was due to copy number change. SW620 cells wereanalyzed using a Genome-Wide Human SNP Array 6.0 with thehapmap270.422 dataset as reference. miR-153 is located onchromosome 2q35, however no copy number changes at thislocus were identified (Supplementary Fig. S3). To clarify if themiR-153 locus is subject to copy number change in primarycolorectal cancer, we examined the Mitelman Database ofChromosome Aberrations (21). A total of 346 cases of colo-rectal cancer were identified and examined for recurrent andnonrecurrent cytogenetic band 2q35 abnormalities. No abnor-malities affecting this locus were identified.

Functional consequences of miR-153 overexpressionTo determine any potential functional roles of overex-

pressed miR-153 in promoting colorectal cancer progression,we investigated the effects of transiently overexpressed miR-153 on a series of cancer-relevant in vitro cell-based assaystesting growth and proliferation, invasion and motility, apo-ptosis, and chemosensitivity in SW480 cells. miR-153 over-expression (Supplementary Fig. S4A) had no effect on pro-liferation (Fig. 2A), anchorage-independent growth (Fig. 2B),or cell migration using scratch and transwell invasion assayswithout matrigel (data not shown). Overexpression of miR-153 promoted a more invasive phenotype in SW480 cellshowever, when matrigel-coated transwell invasion assayswere examined (Fig. 2C and D; P < 0.05). Apoptosis wasassessed in the presence or absence of cisplatin. Overexpres-sion of miR-153 had little effect on viability in the absence ofcisplatin, however protection against cell death was observedin cells overexpressing miR-153 when treated with cisplatin(Fig. 2E and F). These findings suggested that miR-153 mayhave a dual role in colorectal cancer progression, promotingenhanced invasiveness and chemosensitivity. Consequentlyboth these processes were assessed further and are describedlater.

Overexpression of miR-153 leads to increasedinvasiveness in colorectal cancer via MMP-9

The effect of miR-153 on colorectal cancer invasion wastested in additional cell lines with low baseline expression ofmiR-153. Transient overexpression of miR-153 in DLD-1, Ht29,and AAC1/82 cells (Supplementary Fig. S4E) led to enhancedinvasiveness in Matrigel-coated transwell assays (Fig. 3A).Using antimiR-153, inhibition of miR-153 in colorectal cancercell lines with high baseline levels of miR-153 (SupplementaryFig. S4F) significantly reduced invasion compared with control(Fig. 3B) without any alteration to proliferation (Supplemen-tary Fig. S4B and S4C). To more closely recreate and model invivo circumstances, we examined the effect of stably trans-fected GFP-tagged miR-153 in SW480 cells using a 3-dimen-sional organotypic coculture of SW480 cells and human fetalfibroblast cells (Fig. 3C). Ectopic expression of miR-153 but notcontrol resulted in increased invasion of SW480 cells into theunderlying stroma of organotypic cultures (Fig. 3C, i and ii).Although stable overexpression was achieved in only 50% ofcells, immunostaining against GFP illustrated enrichment ofmiR-153–expressing cells at the invasive front (Fig. 3C, iii to vi).

Zhang et al.





To determine if miR-153 would stimulate invasion in vivo, weconducted subcutaneous implantation of SW620 cells in SCIDmice. Cells were transfected with antimiR-153 or control, andinjected into opposite flanks of each animal. Inhibition of miR-153 resulted in more spheroid tumors with clean edges com-pared with control tumors with more locally invasive pheno-types (Fig. 3D). No difference in size of tumors was notedbetween control and miR-153 inhibited xenografts (Supple-mentary Fig. S4D). Analysis of depth of invasion and number ofinvasive tumor spikes into surrounding stroma and adjacenttissue showed a significant reduction in invasive ability ofxenografts transfected with antimiR-153 (Fig. 3E).Our initial findings showed that miR-153 overexpression

promoted a more invasive phenotype only in Matrigel-coated

transwell invasion assays but not in non-Matrigel-coatedassays. A principal constituent of Matrigel is type IV collagen,which is also one of themain substrates forMMP-9 (24). MMP-9 is a key effector in extracellular matrix degradation andstrongly implicated in colorectal cancer associated invasion,with levels progressively increasing from early node-negativeto metastatic colorectal cancer, making it an optimal candi-date to examine (22, 23). To test if MMP-9 may play a role inmiR-153–induced invasion, SW480 cells were transfected withmiR-153 or controlmiRNA andmedium supernatants sampledfor MMP-9 activity. Figure 4A and B show that raised levels ofMMP-9 activity were detected after transfection with miR-153,but not control miRNA. We also assessed levels of the othermember of the Gelatinase subfamily, MMP2, however no effect

Figure 2. Transient overexpressionof miR-153 in SW480 cells had noeffect on proliferation inMTS assay(A) or soft agar colony-formingability (B).C andD,Matrigel-coatedtranswell invasion assays showincreased invasion of SW480 cellstransiently transfected with pre-miR-153 (relative levels of miR-153shown in E) compared with controlmiRNA. D, representative phasecontrast micrograph of invadingSW480 cells. F, representative flowcytometry data of SW480 cellstransfected with control or miR-153 miRNAs. Relative expressionlevel of miR-153 is provided inadjacent bar chart. G, analysis ofapoptosis in SW480 cells in thepresence or absence of cisplatin at50 mg/mL. Data presentedrepresents results from at least 3independent experimentsexpressed as means � SD.�, P < 0.05.






was identified (data not shown).We next tested invasiveness ofSW480 cells transduced with miR-153 in the presence of aselectiveMMP-9 inhibitor. Inhibition ofMMP-9using anMMP-

9 inhibitor or siRNA to MMP-9 abrogated the enhancedinvasiveness mediated by miR-153 (Fig. 4C and D). Finally, todetermine if increased activity ofMMP-9 was because of raised

Figure3. A, increased invasionafterupregulationofmiR-153wasnoted incolorectal cancercell lineswith lowbaselineexpressionofmiR-153 (relativeexpressionof miR-153 is shown in Supplementary Fig. S4E). B, miR-153 inhibition using anti-miR-153 in colorectal cancer cell lines with high baseline levels of miR-153significantly reduced invasion comparedwith control (relative expression ofmiR-153 shown in Supplementary Fig. S4E andS4F). C, 3-dimensional organotypiccocultures of SW480 cells and human fetal fibroblast cells show increased invasion of SW480 into stroma after upregulation of miR-153. Adjacent bar chartillustrates comparative expression of miR-153. C, i and ii, representative sections of organotypic gels immunostained with an anti-cytokeratin antibody showenhanced invasiveness of SW480 cells in which stable, ectopic expression of GFP-tagged miR-153 has been induced. C, iii and vi, consecutive sections wereimmunostained with an anti-GFP antibody and show enrichment of miR-153 expressing SW480 cells at the invasive front compared with control miRNA.(transfected cells contain plasmid constructs that coexpress GFP). D, representative images of mouse tumor xenografts with inhibition of miR-153 show aclean edge of tumor spheroid and fewer invasive fronts into the surrounding stroma (D, ii and iii) in contrast to controls with a more locally invasive tumorphenotype (D, i and iv). Arrows inD, i and iii, show invasive tumor fronts extending and involving surroundingmuscle.Arrowheads inD, ii and iv, indicate the sharpand clean edges of the expanding SW620-anti-miR-153 tumor spheroid. Adjacent bar chart illustrates comparative expression of miR-153. E, quantitationof the vertical depth of tumor invasion and the number of invasive tumor spikes. Twenty random high power fields of the tumor stroma interface were analyzedper slide and mean values obtained. Results are expressed relative to control anti-miR findings. Vertical depth of invasion was measured between thedeepest point of invasion and base of the main tumor mass.

Zhang et al.





levels of MMP-9 transcript, we examined for MMP-9 mRNAafter forced expression of miR-153 (Fig. 4E). No change inMMP-9 mRNA was identified, indicating that increased gela-tonylitic activity is not consequent to a change in MMP-9transcript production or stability. Collectively, these studiessuggest that miR-153 may enhance invasiveness through anincrease in MMP-9 activity.

miR-153 overexpression enhances platinum-basedchemoresistance in colorectal cancer via a direct effecton FOXO3aOur initial functional screen identified a possible role formiR-

153 in mediating chemoresistance to cisplatin (Fig. 2E and F).miRNAs are also known to have powerful regulatory effects onviability and target pathways associated with chemosensitivityand chemoresistance (25, 26). Although cisplatin is an effectiveplatinum-based antineoplastic agent with activity against avariety of gastrointestinal and other solid tumors (27, 28), it is

infrequently used in the current management of colorectalcancer. Conversely, oxaliplatin, a newer generation platinum-based compound, now represents a standard of care for patientswith stages III and IV colorectal cancer when combined withfluoropyrimidine-based agents (18, 28). We therefore evaluatedeffect of miR-153 overexpression on both cisplatin- and oxal-platin-mediated apoptosis. Figure 5A shows that overexpressedmiR-153 had a similar effect in protecting against oxalplatin-mediated apoptosis. We also examined sensitivity of SW480 andSW620cells tooxaliplatin. Asmight bepredicted from thehigherlevels ofmiR-153 inSW620cells, SW620cellsweremore resistantto oxaliplatin-mediated apoptosis, even with high doses ofoxaliplatin (Fig. 5B; ref. 16). Upregulation of miR-153 but notscrambled miRNA reduced abundance of activated caspase 3when SW480 cells were treated with cisplatin, indicating thatmiR-153 is functioning as an antiapoptotic factor in this context(Fig. 5C and D). To explore gene targets of miR-153 that maymediate this effect, we examined for putative targets using 4

Figure 4. Upregulation of miR-153enhances invasiveness throughincreased MMP-9 activity. A,representative experimentshowing zymographic activity inmedium supernatants of SW480cells after transfection with pre-miR-153 or control. Adjacent barchart illustrates comparativeexpression of miR-153. B,assessment of MMP-9 activity bydetermination of band intensity.Results are expressed as relativechanges in MMP activity againstcontrol miR. C, transwell invasionassays in the presence of MMP-9inhibitor (100 nmol/L; C) withrepresentative expression of miR-153 shown in D. E, transwellinvasion assays in the presenceof siRNA to MMP-9, withrepresentative expression of miR-153 shown in F, and levels ofMMP-9 inhibition depicted inSupplementary Fig. S5A. G, effectof miR-153 or control pre-miR onMMP-9 transcript levels. Resultsshow data from at least 3independent experiments.�, P < 0.05.






establishedmiRNA target prediction programs,miRanda (www.microrna.org), PicTar (www.pictar.org), TargetScan (www.tar-getscan.org), and Diana-microTv3.0 (http://diana.cslab.ece.ntua.gr/microT/). To reduce false positives, candidates wereonly considered if theywere predictedby at least 3methods.Onecandidate identified by this approachwas the Forkhead/wingedhelix box class O3a (FOXO3a) tumor suppressor protein. FOXOproteins are a conserved subfamily of transcription factors andorchestrate programs of gene expression involved in differen-tiation, apoptosis, and DNA damage responses. Deletion ofFOXO alleles confers a tumorigenic phenotype in mice models(29), whereas inactivation or silencing leads to a well-describedchemoresistance to cisplatin in colon, ovarian, bladder, and oralsquamous cell carcinomas, making it an optimal target toinvestigate (30–36). FOXO3a protein levels were examined inthe SW480/620 cell model, and found to inversely correlatewith miR-153 expression levels (Fig. 6A). To experimentallyverify FOXO3a as a target of miR-153, we examined the effectof miR-153 overexpression on endogenous FOXO3a proteinlevels. Figure 6B and C show a significant reduction in FOXO3aprotein levels in different cell lines aftermiR-153 overexpression.We also evaluated if miR-153 could impact the transcript levelsof FOXO3a, and noted a significant reduction in FOXO3amRNAin the presence of miR-153 (Fig. 6D).

We next determined if the effect of miR-153 in promotingcolorectal cancer chemoresistance tooxaliplatin could be alteredby introductionof exogenousFOXO3a.As shownbyothergroups,FOXO3a was found to mediate chemosensitivity to oxaliplatin(Supplementary Fig. S5B; refs. 30–36). In addition, Fig. 6E showsthat overexpression of FOXO3a can reverse the observed effect ofmiR-153 on oxaliplatin chemosensitivity. To determine if themiR-153–FOXO3a interaction may also be contributing to theenhanced invasiveness observed with miR-153 overexpression,we first examined the effect of FOXO3a overexpression inmediating colorectal cancer invasiveness. Figure 6F shows thatalthough miR-153 can successfully mediate increased invasive-ness in transwell invasion assays, overexpression or siRNA-mediated inhibition of FOXO3a is unable to replicate this effect.In addition, when FOXO3a was overexpressed in SW480 cells,mediumsupernatants sampled forMMP-9 showedno increase inMMP-9 activity compared with controls, unlike miR-153 over-expression (Fig. 6G). Similarly, other groups have noted nochanges in invasiveness or MMP-9 activity from overexpressedwild-type FOXO3a, further supporting the hypothesis that theobserved effects of miR-153 on invasion and MMP-9 are unlikelyto be mediated through FOXO3a (30, 37).

The 30-UTR of FOXO3a contains one phylogeneticallyconserved miR-153 binding site (Fig. 6H) with a 7mer-A1

Figure 5. A, overexpression of miR-153 in SW480 cells providesresistance to apoptosis from bothcisplatin (50 mg/mL) and oxaliplatin(30 mg/mL), compared with nochemotherapy. Data presentedrepresents results from at least3 independent experimentsexpressed as means � SD.�, P < 0.05. Relative expressionof miR-153 is shown inSupplementary Fig. S4G. B, effectof oxaliplatin treatment at variousdosesoncell viability inSW620andSW480 cells. SW620 cells aremoreresistant to oxaliplatin-mediatedcell death when compared withSW480 cells. C, Western blotanalysis for levels of pro- andcleaved caspase-3 after SW480cells were transfected with miR-153 or control for 48 hours aftertreatment with cisplatin. D, relativesignal intensities for pro- andcleaved caspase 3 weredetermined by densitometry andnormalized to actin. Results arepresented relative to control-miRexpression, whichwas set at 1, andrelative expression levels of miR-153 are provided in SupplementaryFig. S4H.

Zhang et al.





Figure 6. FOXO3a levels are reduced in SW620 cells and FOXO3a is a target of miR-153. A, Western blotting for FOXO3a in SW480 and SW620 cells withcorresponding analysis of signal intensities using polyclonal anti-FOXO3a. Results are presented relative to SW480 cells, which were set at 1.0. B, Westernblot analysis for endogenous FOXO3a in SW480, DLD1, and Ht29 cells after transfection with control or miR-153 precursor miRNA. Representative figure formiR-153 relative levels is presented in Supplementary Fig. S4I. C, relative signal intensities for FOXO3a in Western blots (n ¼ 3) were determined bydensitometry. D, effect of miR-153 or control miRNAs on FOXO3a transcript levels. Total RNA was extracted 48 hours after transfection of SW480 cells andqRT-PCRconducted. Results illustrate data from3 independent experiments and are expressed asmeans�SD. E, overexpressionofmiR-153 inSW480cellsprovides resistance to apoptosis from oxaliplatin (30 mg/mL), which can be reversed by exogenous FOXO3a. Relative expression of miR-153 is shown inSupplementary Fig. S4J. F, knock-in or knockdown of FOXO3a does not influence invasion compared with miR-153. G, zymographic activity in mediumsupernatants of SW480 cells after transfection with pre-miR-153, FOXO3a expression construct, or control. Comparative expression of miR-153 is shown inSupplementary Fig. S4K. H, the 30-UTR of mammalian FOXO3a mRNA contains a phylogenetically preserved miR-153 binding site. F, dual luciferasereporter assayswere conductedwith FOXO3A-30-UTRwt (containingwild-type30-UTRof FOXO3a) andFOXO3A-30-UTRmut (containing amutation of 4 of theamino acids in the predicted miR-153 binding site) vectors. �, P < 0.05.






seed-matched site in the 50 region of miR-153, ranking as oneof the top scoring probabilities of conserved targeting (PCT)as described by Friedman and colleagues (38). To determineif the effect on FOXO3a is mediated via this predicted miR-153 binding site, we cloned a 542bp region of the 30-UTR ofFOXO3a containing the single putative miR-153 binding sitedownstream of the Renilla luciferase open reading frame.Wild-type 30-UTR and a mutant form in which the putativeseed-binding site was mutated were evaluated. As shownin Fig. 6I, when miR-153 precursor was cotransfected withwild-type FOXO3a 30-UTR reporter construct, significantrepression in activity was noted. Mutation of the miR-153binding site eliminated the observed repression, supportinga direct association between miR-153 and FOXO3a. Similarresults were obtained in HCT116 colorectal cancer cells,indicating this is not unique to SW480 cells.

We next sought to determine if expression of FOXO3a wasaltered in human colorectal cancer. Using a custom-designedtissue microarray, we examined and scored the intensity andproportion of cells staining for FOXO3a in 10 normal colorectalmucosa samples, 10 adenomatous polyps, 20 stage 1 colorectalcancer, and 20 stage 3/4 tumors. Representative sections arepresented inFig. 7Aandanalysis of the scoringpresented inFig.7B. FOXO3a expression was significantly lower in advancedcancer stages (P < 0.0005). Analysis of matched tumors forexpression of miR-153 and FOXO3a showed a significant and

inverse correlation betweenmiR-153 and FOXO3a (Fig. 7C; R¼�0.75; P < 0.0001; 95% CI ¼ �0.86 to �0.52). Taken together,these results suggest that miR-153–mediated resistance toplatinum-based chemotherapy could be mediated throughreduced levels of FOXO3a.

DiscussionDisease progression in colorectal cancer is a complex mul-

tistep process. During progression, cancer cells acquire theability to invade beyond normal cellular boundaries, intrava-sate into blood and lymphatics, journey to distant organs,extravasate and proliferate in a different microenvironment,concomitantly eluding antitumor host immunity and defyingchemotherapeutic agents. Although to datemiRNAs have beenprincipally identified as mediators of tumorigenesis, emergingdata is also uncovering their role as important modulators ofdifferent steps in metastasis and disease progression, servingas promoters (7, 12), or inhibitors (13), of these processes.

Here we report on the pleiotropic actions of a poorlyunderstood miRNA, miR-153 in promoting colorectal cancerprogression. We used the SW480/SW620 cell model for colo-rectal cancer progression and conducted miRNA profiling toidentify differentially expressed miRNAs between the 2 celllines. One candidate identified by our screen was miR-153,which was upregulated by over 12-fold in the more advancedSW620 cells. miR-153 is a conserved miRNA first detected at

Figure 7. FOXO3a expression incolorectal cancer. A,representative images fromimmunostaining on a tissuemicroarray, showing FOXO3aexpression in normal, stage 1, andstage 3/4 colorectal cancer. B,FOXO3a expression is lower inmore advanced disease stages(P < 0.0005). C, correlation ofFOXO3a expression and miR-153levels in human tumor samples.

Zhang et al.





high levels in brain tissue, and implicated in development ofneurodegenerative conditions (39, 40). Intriguingly, miR-153levels are reduced in human Glioblastoma Multiforme (41),and increased in endometrial adenocarcinomas (42). To ourknowledge, miR-153 has not been implicated in colorectalcancer before. To determine the relevance of this miRNA, weexamined miR-153 expression in a panel of colorectal cancercell lines, in human colorectal cancer tumor samples com-pared to normal mucosa, and subsequently across a panel ofpredefined colorectal cancers of different clinical stage. miR-153 expression in the colorectal cancer cell lines examined inour study showed similar expression patterns to matchingcolorectal cancer lines in the NCI-60 panel, further verifyingour findings (43). miR-153 was overexpressed in tumors com-pared to normal tissue, and consistent with our cellular model,expressed at higher levels in metastatic compared with non-metastatic tumors. As themainmechanism for development ofcolorectal cancer is chromosomal instability (44), andmiRNAsare overrepresented in regions of genomic copy numberchange in colorectal cancer, we examined the miR-153 locusfor copy number change. Little is understood about the reg-ulation of miR-153 and systematic evaluation of the literaturedid not identify any studies describing genetic or epigeneticregulation of miR-153 specifically. miR-153 resides in the genelocus for the PTPRN gene, which encodes a member of theprotein tyrosine phosphatase family (Supplementary Fig. S3).Analysis of our cell line model and the Mitelman databasesuggest that copy number change in this region is a rare eventhowever. Intriguingly, recent data from DNA methylationprofiling in ovarian cancer has shown that promoter hyper-methylation of the PTPRN gene is significantly associated withlonger patient survival, supporting a model in which miR-153silencingmay be protective against disease progression (46). Atpresent however, no data showing coregulation of PTPRN andmiR-153 has been described.To better understand how deregulation of miR-153 impacts

colorectal cancer, we conducted functional studies to examineeffects of ectopic miR-153 and inhibition of miR-153. Weobserved 2 broad effects of miR-153 upregulation, leading toincreased cellular invasiveness, and enhanced platinum-basedchemotherapy resistance. To dissect the mechanisms behindmiR-153-mediated increase in invasiveness, we examined pro-duction of MMPs after miR-153 overexpression. The gelatinaseMMP-9 is strongly implicated in colorectal cancer progression,and secretion is higher in metastatic murine colorectal cancercompared with nonmetastatic tumors (45), whereas in theSW480/620 human colorectal cancer model, MMP-9 expres-sion is higher in metastatic SW620 cells (14). SW620 cells aremore invasive than SW480 cells in a variety of in vitro assays,and overexpression of MMP-9 in SW480 cells leads to a moreinvasive phenotype (14, 47). In addition,multiple in vivo studieshave showed increased MMP-9 transcript (23), protein (48),and bioactivity (49) inmore advanced human colorectal cancercompared with earlier stages, and increased MMP-9 is an ind-ependent predictor of overall, cancer specific, and disease-freesurvival (23, 49). In this study, we noted increased MMP-9activity after transfection of cells with miR-153. miR-153 tra-nsfected cells showed enhanced invasion through matrigel, a

basement membrane-like matrix rich in collagen type IV (themain substrate of MMP-9), and inhibition of MMP-9 wassufficient to abrogate this effect. The mechanism by whichmiR-153 upregulates MMP-9 activity remains unclear at pres-ent, however, the results of our gene set enrichment analysis forputative miR-153 targets showed overrepresentation of a num-berof transcription factors,whichmayprovide insights into thisprocess (Supplementary Files S1 and Supplementary Fig. S1C).

To better understand how miR-153 increases resistance toplatinum-based agents, in silico analysis was conducted withmiRNA target predicting algorithms. One candidate identifiedwas the FOXO3a transcription factor. Functionally, FOXO3a hasstriking similarities to p53, and the 2 transcription factors sharemany downstream targets (50). Although inactivatingmutationsin FOXO3a in human cancer have not been described, loss-of-function of FOXO3a frequently occurs by posttranslational mod-ifications. FOXO3a is a critical molecule in initiating apoptoticprograms, and upregulates proapoptotic genes such as Bim andPUMA, while downregulating antiapoptotic genes (50). FOXO3ais awell-described regulatorof the response to cisplatin in severalmalignancies including colorectal cancer, and silencing endog-enous FOXO3a impairs cytotoxicity of this chemotherapeuticagent (31–36). Accordingly, we tested if miR-153 could be medi-ating its effect through FOXO3a. Our results show a consistentand inverse relationship between miR-153 and FOXO3a levels incolorectal cancer cell lines and tumors, and support a model inwhich tumor overexpression of miR-153 reduces FOXO3a tran-script and protein levels, impairing the apoptotic response tocisplatin, and mimicking the role of oncogenic kinases such asAkt in inactivating FOXO3a. Our results verify that the miR-153/FOXO3a interaction is a direct one, asmiR-153 interacts with the30-UTR of FOXO3a through a specific seed-binding site to bringabout repression of a FOXO3a-30-UTR luciferase fusion gene.Expression of FOXO3a is clearly a tightly regulated process, andour findings show that miR-153 represents a further layer ofregulation and providing furtherfine tuning of FOXO3a function.

Collectively, our findings suggest that overexpression ofmiR-153 has an important and dual role in promoting diseaseprogression in colorectal cancer through enhanced cellularinvasion and reduced chemosensitivity. The ability of onemiRNA to have a dual function in disease promotion is notunprecedented (14) and indeed the predicted promiscuity ofmiRNAs by target prediction has long hinted at this. Suchpleiotropy also exposes opportunities for exploiting themiRNAsystem for therapeutic manipulation with potential to correctmore than one corrupted pathway by targeting one moleculeonly, giving added value in the design and development of anynovel-targeted therapy. Our present findings also add to theweight of evidence incriminating miRNAs in the mechanismsbehind cancer progression and metastasis, and identify miR-153 as a furthermolecule in the new class of "metastamirs" (51).

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: J. Primrose, G.J. Thomas, G. Packham, A.H. MirnezamiDevelopment of methodology: K. Pickard, J. Strefford, G.J. Thomas, A.H.Mirnezami






Acquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): K. Pickard, V. Jenei, M.D. Bullock, C. Paraskeva,J. Strefford, J. Primrose, A.H. MirnezamiAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): L. Zhang, K. Pickard, V. Jenei, M.D. Bullock, R.Mitter,G. Kelly, J. Strefford, G.J. Thomas, G. Packham, A.H. MirnezamiWriting, review, and/or revision of the manuscript: K. Pickard, V. Jenei,J. Strefford, J. Primrose, G.J. Thomas, G. Packham, A.H. MirnezamiAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): K. Pickard, M.D. Bullock, A. Bruce,C. ParaskevaStudy supervision: A.H. Mirnezami

Grant SupportK. Pickard and A.H. Mirnezami are supported by grant funding from

Wessex Medical Research and Cancer Research UK/RCS (England; C28503/A10013).

The costs of publication of this article were defrayed in part by the payment ofpage charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received August 27, 2012; revised July 8, 2013; accepted August 10, 2013;published OnlineFirst August 15, 2013.

References1. Boyle P, Ferlay J. Cancer incidence andmortality in Europe. AnnOncol

2005;16:481–8.2. Andre N, Schmiegel W. Chemoradiotherapy for colorectal cancer. Gut

2005;54:1194–202.3. Manfredi S, Bouvier AM, Lepage C, Hatem C, Dancourt V, Faivre J.

Incidence and patterns of recurrence after resection for cure ofcolonic cancer in a well defined population. Br J Surg 2006;93:1115–22.

4. Iorio MV, Croce CM. MicroRNAs in cancer: small molecules with ahuge impact. J Clin Oncol 2009;27:5848–56.

5. Calin GA, Sevignani C, Dumitru CD, Hyslop T, Noch E, Yendamuri S,et al. HumanmicroRNAgenes are frequently locatedat fragile sites andgenomic regions involved in cancers. Proc Natl Acad Sci U S A2004;101:2999–004.

6. Bartel DP. MicroRNAs: target recognition and regulatory functions.Cell 2009;136:215–33.

7. Asangani IA, Rasheed SA, Nikolova DA, Leupold JH, Colburn NH,Post S, et al. MicroRNA-21 (miR-21) post-transcriptionally down-regulates tumor suppressor Pdcd4 and stimulates invasion, intra-vasation and metastasis in colorectal cancer. Oncogene 2008;27:2128–36.

8. Valeri N, Gasparini P, Fabbri M, Braconi C, Veronese A, Lovat F, et al.Modulation ofmismatch repair and genomic stability bymiR-155. ProcNatl Acad Sci U S A 2010;107:6982–87.

9. Lanza G, Ferracin M, Gaf�a R, Veronese A, Spizzo R, Pichiorri F, et al.mRNA/microRNA gene expression profile in microsatellite unstablecolorectal cancer. Mol Cancer 2007;6:54.

10. Rossi S, Di Narzo AF,Mestdagh P, Jacobs B, Bosman FT, GustavssonB, et al. MicroRNAs in colon cancer: a roadmap for discovery. FEBSLett 2012;586:3000–07.

11. SchetterAJ, LeungSY,SohnJJ, Zanetti KA,BowmanED,YanaiharaN,et al. MicroRNA expression profiles associated with prognosis andtherapeutic outcome in colon adenocarcinoma. JAMA 2008;299:425–36.

12. Ma L, Teruya-Feldstein J, Weinberg RA. Tumour invasion and metas-tasis initiated by microRNA-10b in breast cancer. Nature 2007;449:682–88.

13. ValastyanS,Reinhardt F,BenaichN,CalogriasD,Sz�aszAM,WangZC,et al. A pleiotropically actingmicroRNA,miR-31, inhibits breast cancermetastasis. Cell 2009;137:1032–46.

14. Bergmann-Leitner ES, Abrams SI. Differential role of Fas/Fas ligandinteractions in cytolysis of primary and metastatic colon carcinomacell lines by human antigen-specific CD8þ CTL. J Immunol 2000;164:4941–54.

15. Hewitt RE, McMarlin A, Kleiner D, Wersto R, Martin P, Tsokos M, et al.Validation of a model of colon cancer progression. J Pathol 2000;192:446–54.

16. Huerta S, Baay-Guzman G, Gonzalez-Bonilla CR, Livingston EH,Huerta-YepezS,BonavidaB. In vitroand in vivo sensitization ofSW620metastatic colon cancer cells to CDDP-induced apoptosis by the nitricoxide donor DETANONOate: involvement of AIF. Nitric Oxide 2009;20:182–94.

17. Nystr€omML, ThomasGJ, StoneM,Mackenzie IC, Hart IR,Marshall JF.Development of a quantitative method to analyse tumour cell invasionin organotypic culture. J Pathol 2005;205:468–75.

18. Association of Coloproctology of Great Britain and Ireland. Guidelinesfor themanagement of colorectal cancer. 3rd ed. London: Associationof Coloproctology of Great Britain and Ireland, 2007.

19. Parker H, Rose-Zerilli MJ, Parker A, Chaplin T, Wade R, Gardiner A,et al. 13q deletion anatomy and disease progression in patients withchronic lymphocytic leukemia. Leukemia 2011;25:489–97.

20. Lamy P, Andersen CL, Dyrskjøt L, Tørring N, �rntoft T, Wiuf C. AremicroRNAs located in genomic regions associated with cancer? Br JCancer 2006;95:1415–18.

21. Mitelman F, Johansson B, Mertens F. Mitelman database of chromo-some aberrations and gene fusions in cancer. Available from: http://cgap.nci.nih.gov/Chromosomes/Mitelman. Access Feb 13, 2012.

22. KessenbrockK,PlaksV,WerbZ.Matrixmetalloproteinases: regulatorsof the tumor microenvironment. Cell 2010;141:52–67.

23. Zeng ZS, Huang Y, Cohen AM, Guillen JG. Prediction of colorectalcancer relapse and survival via tissue RNA levels of matrix metallo-proteinase-9. J Clin Oncol 1996;14:3133–40.

24. Benton G, Kleinman HK, George J, Arnaoutova I. Multiple uses ofbasement membrane-like matrix (BME/Matrigel) in vitro and in vivowith cancer cells. Int J Cancer 2011;128:1751–57.

25. Ma J, Dong C, Ji C. MicroRNA and drug resistance. Cancer Gene Ther2010;17:523–31.

26. Ory B, Ramsey MR, Wilson C, Vadysirisack DD, Forster N, Rocco JW,et al. A microRNA-dependent program controls p53-independentsurvival and chemosensitivity in human and murine squamous cellcarcinoma. J Clin Invest 2011;121:809–20.

27. Valle J, Wasan H, Palmer DH, Cunningham D, Anthoney A, MaraveyasA, et al. Cisplatin plus gemcitabine versus gemcitabine for biliary tractcancer. N Engl J Med 2010;362:1273–81.

28. Kelland L. The resurgence of platinum-based cancer chemotherapy.Nat Rev Cancer 2007;7:573–84.

29. Paik JH, Kollipara R, Chu G, Ji H, Xiao Y, Ding Z, et al. FoxOs arelineage-restricted redundant tumor suppressors and regulate endo-thelial cell homeostasis. Cell 2007;128:309–23.

30. Fern�andez de Mattos S, Villalonga P, Clardy J, Lam EW. FOXO3amediates the cytotoxic effects of cisplatin in colon cancer cells. MolCancer Ther 2008;7:3237–46.

31. Arimoto-Ishida E, Ohmichi M, Mabuchi S, Takahashi T, Ohshima C,Hayakawa J, et al. Inhibition of phosphorylation of a forkhead tran-scription factor sensitizes human ovarian cancer cells to cisplatin.Endocrinology 2004;145:2014–22.

32. Shiota M, Yokomizo A, Kashiwagi E, Tada Y, Inokuchi J, Tatsugami K,et al. Foxo3a expression and acetylation regulate cancer cell growthand sensitivity to cisplatin. Cancer Sci 2010;101:1177–85.

33. Chock KL, Allison JM, Shimizu Y, El Shamy WM. BRCA1-IRIS over-expression promotes cisplatin resistance in ovarian cancer cells.Cancer Res 2010;70:8782–91.

34. Fang L, Wang H, Zhou L, Yu D. FOXO3a reactivation mediates thesynergistic cytotoxic effects of rapamycin and cisplatin in oral squa-mous cell carcinoma cells. Toxicol Appl Pharmacol 2011;251:8–15.

35. Gao J, Yang X, Yin P, Hu W, Liao H, Miao Z, et al. The involvement ofFoxO in cell survival and chemosensitivity mediated byMirk/Dyrk1B inovarian cancer. Int J Oncol 2012;40:1203–9.

36. Yuan Z,Wang F, Zhao Z, Zhao X,Qiu J, NieC, et al. BIM-mediated AKTphosphorylation is a key modulator of arsenic trioxide-induced

Zhang et al.





apoptosis in cisplatin-sensitive and -resistant ovarian cancer cells.PLoS One 2011;6:e20586.

37. Motta MC, Divecha N, Lemieux M, Kamel C, Chen D, Gu W, et al.Mammalian SIRT1 represses Forkhead transcription factors. Cell2004;116:551–63.

38. FriedmanRC, FarhKK,BurgeCB,Bartel DP.MostmammalianmRNAsare conserved targets of microRNAs. Genome Res 2009;19:92–105.

39. Sempere LF, Freemantle S, Pitha-Rowe I, Moss E, Dmitrovsky E,Ambros V. Expression profiling of mammalian microRNAs uncoversa subset of brain-expressed microRNAs with possible roles in murineand human neuronal differentiation. Genome Biol 2004;5:R13.

40. Doxakis E. Post-transcriptional regulation of alpha-synuclein expres-sion by mir-7 and mir-153. J Biol Chem 2010;285:12726–34.

41. Silber J, Lim DA, Petritsch C, Persson AI, Maunakea AK, Yu M, et al.miR-124 and miR-137 inhibit proliferation of glioblastoma multiformecells and induce differentiation of brain tumor stem cells. BMC Med2008;6:14.

42. Myatt SS, Wang J, Monteiro LJ, Christian M, Ho KK, Fusi L, et al.Definition of microRNAs that repress expression of the tumor suppres-sor gene FOXO1 in endometrial cancer. Cancer Res 2010;70:367–77.

43. Reinhold WC, Sunshine M, Liu H, Varma S, Kohn KW, Morris J, et al.CellMiner: a web-based suite of genomic and pharmacologic tools toexplore transcript and drug patterns in the NCI-60 cell line set. CancerRes 2012;72:3499–511.

44. Pino MS, Chung DC. The chromosomal instability pathway in coloncancer. Gastroenterology 2010;138:2059–72.

45. Yamagata S, Ito Y, Tanaka R, Shimizu S. Gelatinases of metastatic celllines of murine colonic carcinoma as detected by substrate-gel elec-trophoresis. Biochem Biophys Res Commun 1988;151:158–62.

46. Bauerschlag DO, Ammerpohl O, Br€autigam K, Schem C, Lin Q, WeigelMT, et al. Progression-free survival in ovarian cancer is reflected inepigenetic DNA methylation profiles. Oncology 2011;80:12–20.

47. Kim HR, Wheeler MA, Wilson CM, Iida J, Eng D, Simpson MA, et al.Hyaluronan facilitates invasion of colon carcinoma cells in vitro viainteraction with CD44. Cancer Res 2004;64:4569–76.

48. Bendardaf R, Buhmeida A, Hilska M, Laato M, Syrj€anen S, Syrj€anen K,et al. MMP-9 (gelatinase B) expression is associated with disease-freesurvival and disease-specific survival in colorectal cancer patients.Cancer Invest 2010;28:38–43.

49. Waas ET, Wobbes T, Lomme RM, DeGroot J, Ruers T, Hendriks T.Matrix metalloproteinase 2 and 9 activity in patients with colorectalcancer liver metastasis. Br J Surg 2003;90:1556–64.

50. You H, Mak TW. Crosstalk between p53 and FOXO transcriptionfactors. Cell Cycle 2005;4:37–38.

51. Hurst DR, Edmonds MD, Scott GK, Benz CC, Vaidya KS, WelchDR. Breast cancer metastasis suppressor 1 up-regulates miR-146,which suppresses breast cancer metastasis. Cancer Res 2009;69:1279–83.






2013;73:6435-6447. Published OnlineFirst August 15, 2013.Cancer Res Lei Zhang, Karen Pickard, Veronika Jenei, et al. Effects That Enhance Invasion and Chemotherapeutic ResistancemiR-153 Supports Colorectal Cancer Progression via Pleiotropic

Updated version

10.1158/0008-5472.CAN-12-3308doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2013/08/23/0008-5472.CAN-12-3308.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/73/21/6435.full#ref-list-1

This article cites 49 articles, 13 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/73/21/6435.full#related-urls

This article has been cited by 2 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/73/21/6435To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-12-3308

http://cancerres.aacrjournals.org/content/suppl/2013/08/23/0008-5472.CAN-12-3308.DC1

http://cancerres.aacrjournals.org/content/73/21/6435.full#ref-list-1

http://cancerres.aacrjournals.org/content/73/21/6435.full#related-urls

http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://cancerres.aacrjournals.org/content/73/21/6435


mir-153 supports colorectal cancer progression via ... · in this study, we hypothesized that...

Documents