Overexpression of ZEB2 at the Invasion Front of ColorectalCancer Is an Independent Prognostic Marker and RegulatesTumor Invasion In Vitro
Christoph Kahlert1, Saleh Lahes1, Praveen Radhakrishnan1, Shamik Dutta1, Carolin Mogler2,Esther Herpel2, Karsten Brand2, Gunnar Steinert1, Martin Schneider1, Martin Mollenhauer1,Christoph Reissfelder1, Fee Klupp1, Johannes Fritzmann1, Christina Wunder3, Axel Benner3,Matthias Kloor2, Cathrin Huth2, Pietro Contin1, Alexis Ulrich1, Moritz Koch1, and J€urgen Weitz1
AbstractPurpose: Epithelial-to-mesenchymal transition (EMT) plays a pivotal role in tumor invasion and
dissemination. EMT occurs predominantly at the tumor edge where it is induced by cytokines, the
extracellularmatrix environment, or hypoxia. In the tumor cell, it is furthermediatedby several transcription
factors and microRNAs. The aim of this study was to explore the expression of EMT-associated genes at the
invasive front in colorectal cancer and to evaluate their prognostic significance.
Experimental Design:We evaluated the expression of 13 EMT-associated genes at the invasion front of
30 colorectal liver metastases by quantitative real-time PCR. Immunostaining against zinc finger E-box–
binding homeobox 2 (ZEB2) was carried out on 175 primary colorectal cancer specimens and 30 colorectal
livermetastases and correlated to clinical and histopathologic data.DLD-1 cells were transfectedwith siRNA
and subjected to migration and invasion assays.
Results: Gene expression analysis and immunohistochemistry showed an upregulation of ZEB2 at the
invasion front in primary colorectal cancer and liver metastases. Overexpression of ZEB2 at the invasion
front correlated significantly with tumor stage in primary colorectal cancer. Moreover, univariate and
multivariate analysis revealed overexpression of ZEB2 at the invasion front as an independent prognostic
marker for cancer-specific survival.Downregulation of ZEB2by siRNAdecreased themigration and invasion
capacity of DLD-1 cells in vitro.
Conclusions: Overexpression of ZEB2 at the invasion front correlates with tumor progression and
predicts cancer-specific survival in primary colorectal cancer. Therefore, ZEB2 may be interesting as
biomarker and potential target for treatment of colorectal cancer. Clin Cancer Res; 17(24); 7654–63.
�2011 AACR.
Introduction
The ability of tumor cells to detach from the main tumorbulk and to invade into the surrounding tissue is a crucialstep for tumor dissemination and metastasis. At the tumorinvasion front, these processes are often associatedwith loss
of an epithelial cell phenotype and acquisition of a mes-enchymal cell phenotype (1, 2). By this epithelial-to-mes-enchymal transition (EMT), tumor cells are endowed withan increased migratory capacity and an augmented inva-siveness (3).
Although epithelial cells undergo EMT, loss of E-cadherinand concomitant expression of distinct mesenchymal mar-kers like vimentin (VIM), cadherin-11 (CDH11), or fibro-blast-specific protein (FSP1) play a central role in thisreversible transdifferentiation (4–7). Besides to cytokineslike TGF-b, extracellular proteins of the tumor microenvi-ronment or hypoxia can induce the downregulation ofE-cadherin and the upregulation of mesenchymal markers(8). This process is further mediated intracellularly byseveral transcription factors such as Snail1, Snail2, zincfinger E-box–binding homeobox 1 (ZEB1) and 2 (ZEB2)and the basic helix-loop-helix transcription factor E47.Theses transcription factors can directly bind the E-cadherinpromoter and inhibit E-cadherin transcription (9–13).
Authors' Affiliations: 1Department of General, Visceral and Transplanta-tionSurgery, 2Institute ofPathology,University ofHeidelberg, and 3GermanCancer Research Center, Division of Biostatistics, Heidelberg, Germany
Note: Supplementary data for this article are available at Clinical CancerResearch Online (http://clincancerres.aacrjournals.org/).
C. Kahlert and S. Lahes contributed equally to this work.
Corresponding Author: Christoph Kahlert, Department of General, Vis-ceral and Transplantation Surgery, University of Heidelberg, Im Neuenhei-mer Feld 110, 69120 Heidelberg, Germany. Phone: 49-6221-56-36498;Fax: 49-6221-565506; E-mail: [email protected]
Because the process of EMT most probably occurs at theedge of a tumor, we hypothesized an overexpression of apattern of genes at the invasion front, which are involved inthe control of EMT in colorectal cancer. Therefore, westarted our studywith an expression analysis of EMT-relatedgenes in the tumor invasion front of colorectal liver metas-tases. These genes included the extracellular matrix-associ-ated genes COL1A1, COL3A1, the mesenchymal markersVimentin, FSP1 and OB-Cadherin, the transcription factorsHIF1a, ZEB1, and ZEB2, EMT-repressing microRNAs(miRNA) miR-141, miR-200a, miR-200b, miR-200c, andMMP3. Among other genes, we could show ZEB2 to beoverexpressed in the tumor invasion front on mRNA andprotein expression level. On the basis of these findings, wehave further investigated the impact of ZEB2 on migrationand invasion in vitro in DLD-1 tumor cells. In this study, weshow that repression of ZEB2 by siRNA in DLD-1 cellsresults in a reduced capacity of migration and invasion invitro. Furthermore, we show the prognostic significance ofZEB2 in a large group of patients with primary colorectalcancer.
Materials and Methods
PatientsParaffin-embedded samples of primary colorectal adeno-
carcinomas were included from a consecutive series of 175patients, who underwent tumor resection between 1996and 2004 at the Department of General, Visceral, andTransplantation Surgery, University of Heidelberg. Sixty-seven patients died during follow-up by a cancer-relateddeath. Prior to our immunohistochemical study, we select-ed only patients with sporadic colorectal cancer. Patientswith a positive medical history for hereditary nonpolyposiscolorectal cancer or familial adenomatous polyposis wereexcluded for our analysis. Paraffin-embedded and snap-frozen samples of colorectal liver metastases were obtainedfrom 30 patients who underwent curative liver resectionbetween 2005 and 2009 at the Department of General,Visceral, and Transplantation Surgery, University of Heidel-berg. The median follow-up time for patients with primary
colorectal cancer was 124 months, the median follow-uptime for patients with colorectal liver metastases was 32months. A written informed consent from all patientsregarding tissue sampling had been obtained. Our studyprotocol was approved by the local ethics committee. Table1 A and B displays the clinical and histopathologic char-acteristics of the patients.
Tissue materialParaffin-embedded tumor samples were provided from
the tissue bank of the National Center for Tumor Disease(NCT) Heidelberg. Frozen tissue of colorectal liver metas-tases was taken immediately after resection. Normal livertissue without evidence of liver metastases was acquired atleast 10 cm away from the tumor. One piece of pure livertissue and1piece consisting ofmetastases and adjacent livertissue were fixed in Tissue-Tek (Satura Finektek), frozenin liquid nitrogen and stored at �80�C for furtherpreparations.
Tissue preparation and laser microdissectionTissue preparation and lasermicrodissection (LMD)were
carried out as recently described (14). Briefly, 16 mm sec-tions of frozen tissue were cut using a cryostat (Leica). Afterstaining with cresyl violet according to the Ambion lasercapture microdissection (LCM) staining kit protocol, LMDwas conducted within 2 hours using LCM equipment(Molecular Machines & Industries or PALM). Normal livertissue wasmicrodissected at least 10 cm away from the host-tumor interface. The liver invasion front included the 10 celllayers of liver tissue adjacent to the tumor. Likewise, tissuefrom the tumor invasion front was obtained from the 10tumor cell layers next to the adjacent liver tissue. Tissue fromthe tumor center was microdissected at least 100 or morecell layers away from the invasion front.
Total RNA isolation and quantitative real-time PCRTotal RNA from cell lines and microdissected tissue was
extracted employingmiRNeasyMini Kit (Qiagen) followingthe manual’s instructions. RNA quality was evaluated byusing an Agilent 2100 Bioanalyzer. Total RNAwas reverselytranscribed using the miScript Reverse Transcription Kit(Qiagen). Five nanogram of the resulting cDNAwas furtherused for quantification by qPCR (SYBR Green PCR Kit,Qiagen) in a Roche Light Cycler (Roche DiagnosticsGmbH). Ready-specific primer pairs were purchased fromQiagen (Supplementary Table S1). Samples were normal-ized to 18s RNA and fold change of expression was calcu-lated according to the DDCt method as previously described(15).
ImmunohistochemistryImmunohistochemical staining for ZEB2 was carried out
as previously described (16). In summary, 1 mm sectionsfrom paraffin-embedded tissue blocks were placed on cap-illary gap microscope slides (SUPERFROST PLUS micro-scope slides,Menzel) and baked overnight at 37�C. Sectionswere deparaffinized by xylene, rehydrated in graded
Translational Relevance
In this article, we show evidence of overexpression ofzinc finger E-box–binding homeobox 2 (ZEB2) at theinvasion front in primary colorectal cancer and livermetastases. Furthermore, we show that ZEB2 regulatestumor invasion in vitro. Overexpression of ZEB2 at theinvasion front correlates significantly with tumor stageof primary colorectal cancer and is an independentprognostic marker for a shortened cancer-specific sur-vival. Therefore, overexpression of ZEB2 at the invasionfront may be a potential biomarker for tumor stagingand patient risk stratification.
ZEB2 Is a Prognostic Marker in Colorectal Cancer
www.aacrjournals.org Clin Cancer Res; 17(24) December 15, 2011 7655
Table 1. Clinical and pathologic parameters of patients with primary colorectal cancer and colorectal livermetastases and their correlation to the expression of ZEB2 in the tumor invasion front and tumor center
Total numberof cases
ZEB2 expression in TIF ZEB2 expression in Tc
Characteristics ZEB2Low ZEB2High P ZEB2Low ZEB2High P
Primary colorectal cancerAge 0.226 1.000< median ¼ 63 y 91 43 48 54 37�median ¼ 63 y 84 48 36 50 34Missing values 0 0 0 0 0
NOTE: P values were derived by using Fisher's exact test.Abbreviations: TIF, tumor invasion front; Tc, tumor control.aBRAF mutation analysis was only conducted for KRAS wild type.
Kahlert et al.
Clin Cancer Res; 17(24) December 15, 2011 Clinical Cancer Research7656
concentrations of ethanol and boiled in a microwave ovenfor 5 minutes [pH 6.0, 0.94 mL Antigen Unmasking Solu-tion (Vector Laboratories, Inc.)/100 mL distilled water] for3 times for heat-induced antigen retrieval. Endogenousperoxidase activity of the tissue was inhibited by treatmentwith3.0%hydrogenperoxidase solution inmethanol for 20minutes. Nonspecific-binding sites were blocked in 1mol/LPBS with 10% normal goat serum and an Avidin/BiotinBlocking Kit (Vector Laboratories). Primary antibodiesagainst ZEB2 (Rabbit polyclonal, Novus Biologicals,1:200) and IgG-negative control (Mouse IgG, BD Pharmin-gen, 1:200) were incubated at 4�C overnight. After washingwith PBS, sections were loaded with secondary antibodycoupled with peroxidase-conjugated polymers (EnVisionþSystem,DakoCytomationA/S) for 30minutes. Subsequent-ly, the primary antibodies were detected by using AECSubstrate Chromogen (DakoCytomation A/S) according tothe instructions of the manufacturer. The sections werecounterstained with hematoxylin, dehydrated in gradedconcentrations of ethanol andmounted. The staining inten-sity of each slide was separately scored for tumor invasionfront and tumor center by scanning the whole section atmedium (50�) and high magnification (200�) on a blindbasis by 2 independent researchers (C.K. and S.L.) and 1pathologist (C.M.) as absent: 0, weak: 1, medium: 2 andstrong: 3. A multihead microscope was used and consensuswas reached for each slide.
Microsatellite instability analysis, BRAF and KRASmutation analysisMicrosatellite instability analysis was carried out using
the markers BAT25, BAT26, and CAT25 as described previ-ously (17). BRAF and KRAS mutation analysis was carriedout using the following primers: sense 50-TCATAATGC-TTGCTCTGATAGGA-30 and antisense 50-GGCCAAAAATT-TAATCAGTGGA-30 for BRAF, sense 50-AAGGCCTGCT-GAAAATGACTG-30 and antisense 50-AGAATGGTCCTG-CACCAGTAA-30 for KRAS. Briefly, PCR products were puri-fied using the QIAquick PCR purification kit (Qiagen).Subsequently, sequencing reaction was carried out usingBigDye terminator kit (v1.1, Applied Biosystems) accordingto the manufacturer’s instructions. Sequencing productswere separated on an ABI 3100-automated sequencer(Applied Biosystems).
Cell culture and transfectionThe colon cancer cell line DLD-1 was purchased from the
Deutsche Sammlung von Mikroorganismen und Zellkultu-ren GmbH (ACC274, DSMZ). In August 2010, the cellline was retested and reauthenticated by analyzing cellsamples for 8 polymorphic short random repeats loci(DSMZ). DLD-1 cells were maintained in RPMI-1640(Sigma), supplemented with 10% (v/v) fetal calve serum,100 U/mL penicillin and 100 mg/mL streptomycin in ahumidified atmosphere of 5% CO2 at 37�C. ZEB2 siRNA[siZEB2 NM_014795_813 (sense): 30-GCACAACAACGA-GATTCTA-50 and siZEB2NM_014795_1503 (sense):30-GCACATCAGCAGCAAGAAA-50] and scrambled siRNA
[NM_053208_siRNA_control_1224 (sense): 30-GCUUAA-CCCGUAUUGCCUATT-50; both Invitrogen GmbH] weretransfected using Lipofectamine 2000 (Invitrogen GmbH)according to the manufacturer’s instruction.
Cell migration and cell invasion assayEight hours after transfection, the medium containing
Lipofectamine 2000 was discarded and DLD-1 cells weremaintained in serum-free RPMI–1640 medium overnight.After overnight starvation, ability of cell migration and cellinvasion was evaluated by the CytoSelect 24-Well CellMigration and Invasion Assay (8 mm pores, ColorimetricFormat, Cell Biolabs, Inc.) according to the manufacturer’sinstructions. For the Cell Migration Assay, DLD-1 cells wereincubated for 24 hours; for the Invasion Assay, DLD-1 cellswere incubated for 72 hours. Spectrophotometry for theCell Migration and Cell Invasion Assay was conducted on amicrotiter plate reader at 540nm. All assayswere carried out3 times in quadruplicates.
Statistical analysisThe software package SPSS, version 18.0 (SPSS), and the
statistical software environment R, version 2.13.0, wereused for all calculations. Wilcoxon-signed rank test wasemployed for determining a different regulation between2 different tissue compartments, respectively. Student t testwas conducted for analysis of the migration and invasionassays. Fisher’s exact test was applied to examine indepen-dence of ZEB2 expression and clinical and pathologicparameters. The Kaplan–Meier method was used to esti-mate cancer-specific survival curves, and differencesbetween survival curves were assessed with the log-ranktest. Multivariate survival analysis was carried out using Coxproportional hazards regression including the followingcovariates: gender, age (dichotomized at median age),tumor stage according to the UICC classification, dichoto-mized tumor grade (grade 1, 2 vs. 3, 4), type of resection(curative vs. noncurative), microsatellite stability, KRASmutation status, expression of ZEB2 in the tumor invasionfront and in the tumor center (dichotomized at medianexpression). The survival endpoint for univariate and mul-tivariate analysis was time from surgery until cancer-relateddeath, where death due to other reasons was counted ascensored event. The proportional hazards assumption wastested as proposed byGrambsch andTherneau (18). Resultswere considered significant at a P value less than 0.05.
Results
Expression analysis of EMT-associated genes andmiRNAs in the invasion front of colorectal livermetastases
We first evaluated the expression profile of 13 EMT-associated genes and miRNAs in 30 colorectal liver metas-tases. All liver metastases had been subjected to lasermicrodissection to obtain compartment-specific tissuefrom normal liver, liver invasion front, tumor invasionfront, and tumor center. The following genes were
ZEB2 Is a Prognostic Marker in Colorectal Cancer
www.aacrjournals.org Clin Cancer Res; 17(24) December 15, 2011 7657
included in the evaluation: the extracellular matrix-asso-ciated genes COL1A1, COL3A1, the mesenchymal mar-kers Vimentin, FSP1 and OB-Cadherin, the transcriptionfactors HIF1a, ZEB1 and ZEB2, EMT-repressing miRNAsmiR-141, miR-200a, miR-200b, miR-200c, and MMP3.Expression analysis revealed COL3A1, Vimentin, HIF1a,ZEB1, and ZEB2 to be significantly upregulated in theliver invasion front compared with normal liver andtumor center (Supplementary Fig. S1A–M, SupplementaryTable S2). Besides, there was a statistically significantincreased expression of ZEB2 in the tumor invasion frontcompared with the tumor center. HIF1a and ZEB1 were
found to be significantly upregulated in the tumor invasionfront compared with liver and tumor center. MMP3 dis-played also an overexpression at the tumor edge comparedwith tumor center and liver in20 clinical specimens, butwasnot detectable in 10 samples (Supplementary Fig. S1A–M,Supplementary Table S2).
Localization of ZEB2 in colorectal liver metastases andprimary cancer
HIF1a displays an increased expression in tumor cellsand proliferating fibroblasts at the invasion front(2, 19, 20). ZEB1 has been found to be upregulated in
Figure 1. Immunohistochemicalanalysis of ZEB2 in colorectal livermetastases (A–D) and primarycolorectal cancer (E–H). Figuresrepresent invasion front-specificoverexpression of ZEB2 in paraffin-embedded specimens. Originalmagnification �200 (left) andcorresponding areas (boxed areas)with lower magnification �80.
Kahlert et al.
Clin Cancer Res; 17(24) December 15, 2011 Clinical Cancer Research7658
dedifferentiated tumor cells of colorectal cancer at theinvasion front and in tumor-associated fibroblasts.These results of the upregulation of HIF1a and ZEB1 at
the invasion front in colorectal cancer are in good accor-dance with our findings of the expression analysis of EMT-associated genes. However, the increased expression ofZEB2 at the tumor edge of colorectal liver metastases hasnot yet been described. Therefore, we focused on this EMT-inducing transcription factor which displayed an upregula-tion at the liver invasion front and at the tumor invasionfront compared with pure liver and tumor center, respec-tively, based on the results of our qPCR expression analysis.To further localize the cellular origin of ZEB2, immunohis-tochemical analysis on matched paraffin-embedded colo-rectal livermetastases of the same clinical specimens, whichhad been subjected to laser microdissection, was done.Immunhistochemical analysis confirmed an overexpres-sion of ZEB2 at the invasion front in 18 of 30 samples. In10 clinical specimens there was a homogenous expressionof ZEB2 in the tumor center and the invasion front, in 1sample the tumor center displayed a stronger expressionthan the invasion front and 1 sample was not evaluable(Fig. 1A–D). Hepatocytes from the liver invasion front andfrom areas not adjacent to the tumor displayed a strongcytoplasmic expression of ZEB2. Immunohistochemicalanalysis of 175 primary colorectal cancer samples revealedan invasion front-specific overexpression of ZEB2 in 128specimens when compared with ZEB2 expression in thetumor center. In 95 of these 128 cases, the difference ingradeof intensity between tumor center and tumor invasionfront was 1, in 32 cases the difference amounted to 2 and in1 case the invasion front was strongly positive whereas thetumor center was negative. Thirty-nine specimens displayeda homogenous expression of ZEB2 in the tumor invasionfront and tumor center, 4 samples had a stronger expressionofZEB2 in the tumor center than tumor invasion front and6samples were not evaluable (Fig. 1 E–H). In all samples,ZEB2 was predominantly expressed in the cytoplasm,whereas nuclear staining was observed only in a few cells.Immunohistochemical analysis of ZEB2 in healthy controlcolon tissue adjacent to the colorectal tumor displayed
mainly no expression of ZEB2 or only a very weakexpression.
ZEB2 regulates migration and invasion of DLD-1 cellsin vitro
Due to the overexpression of ZEB2 at the invasion front invivo, we hypothesized that ZEB2 regulates migration orinvasion in colon cancer cells in vitro. ZEB2was antagonizedby siRNA in vitro in DLD-1 cells. Transfection efficiency wasevaluated by quantitative real-time PCR (qRT-PCR) andreached more than 90% (Fig. 2A). Twenty-four hours post-transfection, DLD-1 cells, which had been transfected withZEB2 siRNA, displayed an impaired migration capacity of25% compared with the cells having been transfected withscrambled siRNA (P ¼ 0.01; Fig. 2B). Likewise, 72-hourposttransfection, DLD-1 cells transfected with ZEB2 siRNAhad a reduced invasion capacity of 26% comparedwith DLD-1 cells transfected with scrambled siRNA(P ¼ 0.001; Fig. 2C).
Overexpression of ZEB2 at the invasion front is anindependent prognostic marker
The association of clinical and histopathologic para-meters with ZEB2 expression was evaluated separately forthe tumor invasion front and tumor center. To compare theexpression of ZEB2with clinical andpathologic parameters,samples were grouped as ZEB2Low (staining intensity �median immunohistochemical score ¼ 2) and asZEB2High (staining intensity >median immunohistochem-ical score ¼ 2). Fisher’s exact test revealed statisticallysignificant associations of ZEB2 expression in the tumorinvasion front with grading (P < 0.001) and tumor stageaccording to the UICC classification (P ¼ 0.014; Table 1).
Log-rank tests showed a highly significant impact oftumor stage (P < 0.001) and tumor grade (P < 0.001) oncancer-specific survival (Table 2). Furthermore, high expres-sion of ZEB2 at the tumor invasion front correlated signif-icantly with a worsening prognosis (P < 0.001; ZEB2High:estimated cancer-specific 5-year survival rate: 57.3%,ZEB2Low: estimated cancer-specific 5-year survival rate:76.2%; Fig. 3A, Table 2). Likewise high expression of ZEB2
1.2
1
0.8
0.6
0.4
0.2
0
Evaluation of transfection efficiencyA B C
Migration assay Invation assay
P = 0.018 P = 0.001
siZEB2 siControl siZEB2 siZEB2siControl siControl
Exp
ress
ion
fold
cha
nge
Per
cent
age
(%)
of m
igra
ting
DLD
-1 c
ells
140
120
100
80
60
40
20
0
140
120
100
80
60
40
20
0 Per
cent
age
(%)
of m
igra
ting
DLD
-1 c
ells
Figure 2. DLD-1 cells were transfected with ZEB2 siRNA or scrambled control siRNA. Transfection efficiency was evaluated by qRT-PCR and reached morethan 90% (A). DLD-1 cells were subjected to migration assays (B) and invasion assays (C). ZEB2 siRNA–transfected DLD-1 cells displayed animpairedmigration capacity of 25% (P¼ 0.018) and a reduced invasion capacity of 26% (P¼0.001) comparedwith control. All assayswere carried out 3 timesin quadruplicates.
ZEB2 Is a Prognostic Marker in Colorectal Cancer
www.aacrjournals.org Clin Cancer Res; 17(24) December 15, 2011 7659
in the tumor center was associatedwith grading (P¼ 0.036)and tumor stage according to the UICC classification (P ¼0.048; Table 1). Although there was a tendency for poorerprognosis in patients with high ZEB2 expression in thetumor center, this correlationwasnot statistically significant(Fig. 3B, Table 2); estimated cancer-specific 5-year survivalrate of 60.3% for ZEB2High and of 71.8% for ZEB2Low).
Multivariate analysis by Cox proportional hazards regres-sion was carried out to identify prognostic markers forcancer-specific survival. Expression of ZEB2 at the invasionfrontwas found tobe an independent prognosticmarker forcancer-specific survival (P ¼ 0.02) besides tumor stage
(P < 0.001) whereas ZEB2 expression at the tumor centerfailed to be significant (P ¼ 0.33; Table 3).
Discussion
The invasion front of colorectal cancer is characterizedby a dynamic process referred to as EMT (1, 2, 21). Duringthis process, tumor cells loose their epithelial cell–celljunctions and develop a mesenchymal-like phenotype.This process can be activated by cytokines, proteins of theextracellular matrix or hypoxia and is further mediated byseveral transcription factors. For a first overview, we
Table 2. Univariate analysis (log-rank test) of possibly prognostic factors in primary colorectal cancer forcancer-specific survival
started our study in 30 colorectal liver metastases byanalyzing the expression of 13 genes and miRNAs at theinvasion front, which are involved in the regulation ofEMT. Here, we were able to show an overexpression ofMMP3, HIF1a, ZEB1, and ZEB2 at the invasion front.These genes are involved in the repression of E-cadherin,which is a hallmark of EMT (11, 22–25). Brabletz andcolleagues have shown that membranous expression ofE-cadherin is lost at the invasion front of colorectalprimary cancer and liver metastases (1). In this context,our data support the assumption that EMT occurs pre-dominantly at the invading tumor edge. Interestingly,none of the EMT-repressing miRNAs of the miR-200
family (26–30) displayed a significantly different expres-sion between tumor center and tumor invasion front,though they were highly upregulated compared withnormal liver. Our results do not provide a more profoundmechanistic explanation for these observations. However,it could be conjectured that the upregulation of pro-EMTfactors tends to be more essential for tumor invasion thanthe inhibition of EMT-repressing molecules.
By immunohistochemistry, we confirmed the upregula-tion of ZEB2 at the tumor invasion front. ZEB2 was pre-dominantly expressed in the cytoplasm, whereas nuclearstaining was observed only in a few cells. These findings areconsistent with a recent report, comparing the expression ofZEB2 by 2 monoclonal antibodies in a small cohort ofpatients with colorectal cancer (31). Due to the overexpres-sion of ZEB2 at the invasion front, we postulated an inva-sion-promoting role of this transcription factor. In fact,downregulation of ZEB2 by siRNA resulted in a decreasedmigration and invasion capacity of DLD-1 cells in vitro.These data are in good accordance with a previous report, inwhich Vanderwalle and colleagues showed that doxycy-cline-induced overexpression of ZEB2 is associated with amore invasive type in DLD-1 cells (32).
One of our main findings is the significant correlationbetween overexpression of ZEB2 at tumor center and inva-sion front with the tumor stage according to the UICCclassification. Furthermore, multivariate analysis revealeda prognostic impact of invasion front-specific overexpres-sion of ZEB2 in primary colorectal cancer besides UICCstage, whereas the expression of ZEB2 in the tumor centerwas not significantly correlated to cancer-specific survivalwhen analyzed together with other risk factors. Althoughprevious reports have already described increased expres-sion of ZEB2 as a prognostic marker in bladder cancer (33),ovarian cancer (34), squamous cell carcinoma (35), andlung cancer (36), these studies have not distinguishedbetween the expression at the tumor edge and tumor center.However, our in vitrodata andother studies indicate stronglythat ZEB2 plays an important role in tumor migration and
Figure 3. Kaplan–Meier curvesdisplay tumor-specific overallsurvival in patients with primarycolorectal cancer (n ¼ 175). Solidline: negative/low expression ofZEB2; dashed line: high expressionof ZEB2. A, overexpression of ZEB2at the invasion front is significantlyassociated with a shortened tumor-specific survival (log-rank test, P �0.0001). B, overexpression of ZEB2in the tumor center (Tc) correlatedalso with a shortened survival, butfailed slightly to be statisticallysignificant (log-rank test, P ¼ 0.06).
Table 3. Multivariate analysis (Cox proportionalhazards regression model) of prognosticparameters for tumor-specific survival incolorectal cancer
Characteristics HR95% CI ofrelative risk P
Median age 3.10 1.52–6.32 0.002Gender 1.60 0.82–3.14 0.17Tumor grade(1, 2) vs. (3, 4)
invasion (27, 32). These abilities are requiredat the invasionfront to break down the basementmembrane and to extrav-asate into lymph and blood vessels in the adjacent tissue. Inthis context, our data suggest that overexpression of ZEB2 atthe invasion front is relevant for tumor cell disseminationand progression in primary colorectal cancer.
In summary, we report about an expression profile of apanel of EMT-associated genes in the tumor invasion frontof colorectal liver metastases. These data support the sig-nificant role of EMT in vivo at the tumor edge of colorectalcancer. Moreover, our study is the first to show that over-expression of ZEB2 correlates significantly with the tumorstage according to the UICC classification and tumor grad-ing. In our study, overexpression of ZEB2 at the invasionfront is associatedwithpoor prognosis in primary colorectalcancer besides known risk factors including age, tumorstage, and grading. These results suggest that ZEB2 mightbe a useful prognostic marker and should furthermore beinvestigated as a potential therapeutic target in patients withcolorectal cancer.
Disclosure of Potential Conflicts of Interest
No potential conflicts of interest were disclosed.
Acknowledgments
The authors thank Bettina Walter from the tissue bank of the NationalCenter for Tumor Disease (NCT) Heidelberg for technical help in tissuesections.
Grant Support
The Clinical Research Unit KFO 227 "Colorectal cancer: From primarytumor progression towardsmetastases" was funded by the German Researchfoundation (DFG).
The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.
Received October 21, 2010; revised October 13, 2011; accepted October14, 2011; published OnlineFirst October 31, 2011.
References1. Brabletz T, Jung A, Reu S, Porzner M, Hlubek F, Kunz-Schughart LA,
et al. Variable beta-catenin expression in colorectal cancers indicatestumor progression driven by the tumor environment. Proc Natl AcadSci U S A 2001;98:10356–61.
2. Spaderna S, Schmalhofer O, Hlubek F, Berx G, Eger A, Merkel S, et al.A transient, EMT-linked loss of basement membranes indicatesmetastasis and poor survival in colorectal cancer. Gastroenterology2006;131:830–40.
3. Kalluri R, Weinberg RA. The basics of epithelial-mesenchymal transi-tion. J Clin Invest 2009;119:1420–8.
4. Xue C, Plieth D, Venkov C, Xu C, Neilson EG. The gatekeeper effect ofepithelial-mesenchymal transition regulates the frequency of breastcancer metastasis. Cancer Res 2003;63:3386–94.
5. Strutz F, Zeisberg M, Ziyadeh FN, Yang CQ, Kalluri R, M€uller GA, et al.Role of basic fibroblast growth factor-2 in epithelial-mesenchymaltransformation. Kidney Int 2002;61:1714–28.
6. Copple BL. Hypoxia stimulates hepatocyte epithelial to mesenchymaltransition by hypoxia-inducible factor and transforming growth factor-beta-dependent mechanisms. Liver Int 2010;30:669–82.
7. Schmalhofer O, Brabletz S, Brabletz T. E-cadherin, beta-catenin, andZEB1 in malignant progression of cancer. Cancer Metastasis Rev2009;28:151–66.
8. Polyak K, Weinberg RA. Transitions between epithelial and mesen-chymal states: acquisition of malignant and stem cell traits. Nat RevCancer 2009;9:265–73.
9. Perez-Moreno MA, Locascio A, Rodrigo I, Dhondt G, Portillo F, NietoMA, et al. A new role for E12/E47 in the repression of E-cadherinexpression and epithelial-mesenchymal transitions. J Biol Chem2001;276:27424–31.
10. Comijn J, Berx G, Vermassen P, Verschueren K, van Grunsven L,Bruyneel E, et al. The two-handed E box binding zinc finger proteinSIP1 downregulates E-cadherin and induces invasion. Mol Cell2001;7:1267–78.
11. S�anchez-Till�o E, L�azaro A, Torrent R, Cuatrecasas M, Vaquero EC,Castells A, et al. ZEB1 represses E-cadherin and induces an EMT byrecruiting the SWI/SNF chromatin-remodeling protein BRG1. Onco-gene 2010;29:3490–500.
12. Batlle E, Sancho E, Francí C, Domínguez D, Monfar M, Baulida J, et al.The transcription factor snail is a repressor of E-cadherin gene expres-sion in epithelial tumour cells. Nat Cell Biol 2000;2:84–9.
13. Hajra KM, Chen DY, Fearon ER. The SLUG zinc-finger proteinrepresses E-cadherin in breast cancer. Cancer Res 2002;62:1613–8.
14. Kahlert C, Bandapalli OR, Schirmacher P, Weitz J, Brand K. Invasionfront-specific overexpression of tissue inhibitor ofmetalloproteinase-1in liver metastases from colorectal cancer. Anticancer Res 2008;28:1459–65.
15. PfafflMW.Anewmathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res 2001;29:e45.
16. Kahlert C, Weber H, Mogler C, Bergmann F, Schirmacher P, KenngottHG, et al. Increased expression of ALCAM/CD166 in pancreatic canceris an independent prognosticmarker for poor survival and early tumourrelapse. Br J Cancer 2009;101:457–64.
17. Findeisen P, Kloor M, Merx S, Sutter C, Woerner SM, Dostmann N,et al. T25 repeat in the 30 untranslated region of the CASP2 gene: asensitive and specific marker for microsatellite instability in colorectalcancer. Cancer Res 2005;65:8072–8.
18. Grambsch PM, Therneau TM, Fleming TR. Diagnostic plots to revealfunctional form for covariates in multiplicative intensity models. Bio-metrics 1995;51:1469–82.
19. Kuwai T, Kitadai Y, Tanaka S, Onogawa S, Matsutani N, Kaio E, et al.Expression of hypoxia-inducible factor-1alpha is associated withtumor vascularization in human colorectal carcinoma. Int J Cancer2003;105:176–81.
20. Sivridis E, Giatromanolaki A, Koukourakis MI. Proliferating fibroblastsat the invading tumour edge of colorectal adenocarcinomas areassociated with endogenous markers of hypoxia, acidity, and oxida-tive stress. J Clin Pathol 2005;58:1033–8.
21. Zlobec I, Lugli A. Invasive front of colorectal cancer: dynamic interfaceof pro-/anti-tumor factors. World J Gastroenterol 2009;15:5898–906.
22. Aigner K, Dampier B, Descovich L, Mikula M, Sultan A, Schreiber M,et al. The transcription factor ZEB1 (deltaEF1) promotes tumour celldedifferentiation by repressing master regulators of epithelial polarity.Oncogene 2007;26:6979–88.
23. Krishnamachary B, Zagzag D, Nagasawa H, Rainey K, Okuyama H,Baek JH, et al. Hypoxia-inducible factor-1-dependent repression ofE-cadherin in von Hippel-Lindau tumor suppressor-null renal cellcarcinoma mediated by TCF3, ZFHX1A, and ZFHX1B. Cancer Res2006;66:2725–31.
24. Radisky DC, Levy DD, Littlepage LE, Liu H, Nelson CM, Fata JE, et al.Rac1b and reactive oxygen species mediate MMP-3-induced EMTand genomic instability. Nature 2005;436:123–7.
25. Lochter A, Galosy S, Muschler J, Freedman N, Werb Z, Bissell MJ.Matrix metalloproteinase stromelysin-1 triggers a cascade of molec-ular alterations that leads to stable epithelial-to-mesenchymal
Kahlert et al.
Clin Cancer Res; 17(24) December 15, 2011 Clinical Cancer Research7662
conversion and a premalignant phenotype inmammary epithelial cells.J Cell Biol 1997;139:1861–72.
26. Bracken CP, Gregory PA, Kolesnikoff N,, Bert AG, Wang J, ShannonMF, et al. A double-negative feedback loop between ZEB1-SIP1 andthemicroRNA-200 family regulates epithelial-mesenchymal transition.Cancer Res 2008;68:7846–54.
27. HuM, Xia M, Chen X, Lin Z, Xu Y, Ma Y, et al. MicroRNA-141 regulatesSmad interacting protein 1 (SIP1) and inhibitsmigration and invasion ofcolorectal cancer cells. Dig Dis Sci 2010;55:2365–72.
28. Hurteau GJ, Carlson JA, Roos E, Brock GJ. Stable expression of miR-200c alone is sufficient to regulate TCF8 (ZEB1) and restore E-cadherinexpression. Cell Cycle 2009;8:2064–9.
29. Hurteau GJ, Carlson JA, Spivack SD, Brock GJ. Overexpression of themicroRNA hsa-miR-200c leads to reduced expression of transcriptionfactor 8 and increased expression of E-cadherin. Cancer Res 2007;67:7972–6.
30. Li Y, VandenBoom TG 2nd, Kong D,Wang Z, Ali S, Philip PA, et al. Up-regulation of miR-200 and let-7 by natural agents leads to the reversalof epithelial-to-mesenchymal transition in gemcitabine-resistant pan-creatic cancer cells. Cancer Res 2009;69:6704–12.
31. Oztas E, Avci ME, Ozcan A, Sayan AE, Tulchinsky E, Yagci T. Novelmonoclonal antibodies detect Smad-interacting protein 1 (SIP1) in the
cytoplasm of human cells from multiple tumor tissue arrays. Exp MolPathol 2010;89:182–9.
32. Vandewalle C, Comijn J, De Craene B, Vermassen P, Bruyneel E,Andersen H, et al. SIP1/ZEB2 induces EMT by repressing genes ofdifferent epithelial cell-cell junctions. Nucleic Acids Res 2005;33:6566–78.
33. Sayan AE, Griffiths TR, Pal R, Browne GJ, Ruddick A, Yagci T, et al.SIP1 protein protects cells from DNA damage-induced apoptosis andhas independent prognostic value in bladder cancer. Proc Natl AcadSci U S A 2009;106:14884–9.
34. Elloul S, Elstrand MB, Nesland JM, Trop�e CG, Kvalheim G, Goldberg I,et al. Snail, Slug, and Smad-interacting protein 1 as novel parametersof diseaseaggressiveness inmetastatic ovarian andbreast carcinoma.Cancer 2005;103:1631–43.
35. Maeda G, Chiba T, Okazaki M, Satoh T, Taya Y, Aoba T, et al.Expression of SIP1 in oral squamous cell carcinomas: implicationsfor E-cadherin expression and tumor progression. Int J Oncol 2005;27:1535–41.
36. Miura N, Yano T, Shoji F, Kawano D, Takenaka T, Ito K, et al. Clini-copathological significance of Sip1-associated epithelial mesenchy-mal transition in non-small cell lung cancer progression. AnticancerRes 2009;29:4099–106.
ZEB2 Is a Prognostic Marker in Colorectal Cancer
www.aacrjournals.org Clin Cancer Res; 17(24) December 15, 2011 7663
